CAG Paper Session—IBD Microbiota, Friday February 26, 08 h00–09 h30


Determinants of Intestinal Permeability in Healthy First Degree Relatives of Crohn’s Disease Patients
W. Turpin,1 D. Kevans,2 K. Shestopaloff,3 M. Smith,1 D. Guttman,1 M. Silverberg,2 W. Xu,3 A. Paterson,4 and K. Croitoru2

1University of Toronto, Toronto, ON, Canada
2Mount Sinai Hospital, Toronto, ON, Canada
3Dalla Lana School of Public Health, Toronto, ON, Canada
4The Hospital for Sick Children Toronto, Toronto, ON, Canada

Background. Increased intestinal permeability (IP) has been observed in a number of autoimmune diseases. Our recent study has demonstrated that the host genetic and intestinal microbial composition has a limited influence on IP while smoking status and age as two important factors contributing to IP.

Aims. To investigate if demographic factors, environmental factors or bacterial functions are associated with intestinal permeability.

Methods. IP was measured with high-pressure liquid chromatography by timed urine collection after ingestion of an oral load of two saccharide probes, lactulose and mannitol. For each subject, the lactulose-mannitol ratio (LacMan ratio) was calculated as the fractional excretion of lactulose divided by that of mannitol. Bacterial DNA extracted from the stool of 1098 healthy subject was sequenced for the V4 hypervariable regions of the 16S rRNA using the Illumina MiSeq platform. The function of the fecal microbial communities was then imputed using PICRUSt V0.1 after a rarefaction step to 30,000 sequences per sample. The PICRUSt pre-calculated table of gene counts based on OTUs was used to identify the gene counts in the organisms present in the stool samples. The Kyoto Encyclopedia of Genes and Genomes (KEGG) and clusters of orthologous groups (COG) databases were used to identify gene families. Association was performed using a linear regression controlling for age, gender and smoking status. Bacterial functions with a mean count <10 were excluded.

Results. A total of 65 demographic and environmental factors were analyzed. We found that individuals currently living with a dog had higher IP (). However this association was temporary as dog exposure within younger age classes but not currently exposed shows no evidence of association. Living with other types of animals aside from dogs did not show an association with IP. Among 3,773 KEGG and 3,618 COG functions, we found several nominal associations with IP, the most significant being involved in tyrosine metabolism and degradation of aromatic compounds (K01826), possibly involved in tellurite resistance (COG3615), and DNA uptake process and recombination (COG4469) ().

Conclusions. Multivariate analysis controlling for major contributing factors to IP allowed us to identify that individuals currently living with a dog had increased IP. In addition, while the specific microbial taxa do not appear to be associated with IP, microbial community functions are likely contributing to IP in healthy humans. These results indicate the importance of environmental influences on IP.

Submitted on behalf of GEM Project research team.

Funding Agencies: CAG, CIHR

CAG Paper Session—CAG/CCC Student Prize Paper Presentations, Friday February 26, 10 h00–11 h30

CAG Student Prize


Integrin 11 Is Controlled by the MYC Oncogenic Factor and Confers Pro-Proliferative and Pro-Migratory Advantage to Colorectal Cancer Cells
S. Boudjadi, G. Bernatchez, B. Senicourt, M. Beauséjour, P. Vachon, J. Carrier, and J. Beaulieu

Université de Sherbrooke, Sherbrooke, QC, Canada

Background. Colorectal cancer (CRC) is a multi-step process that involves successive mutation, epigenetic alteration and gene dysregulation. Integrins are a family of heterodimeric glycoproteins involved in bidirectional cell signaling and participate in the regulation of cell shape, adhesion, migration, differentiation, gene transcription, survival and proliferation. The integrin α1 subunit is known to be involved in RAS/ERK proliferative pathway activation and plays an important role in mammary carcinoma cell proliferation and migration. In the small intestine, α1 is present in the crypt proliferative compartment and absent in the villus. In mouse models, the α1β1 integrin supports breast cancer cell motility and, together with the Kras oncogenic factor, potentiates tumor growth. Very little is known about α1β1 function in CRC.

Aims. As we have recently shown that α1 is present in 65% of CRC (Boudjadi et al., 2013) and that its expression is controlled by the MYC oncogenic factor and that they correlate in 72.3% of colon adenocarcinomas (Boudjadi et al., Oncogene 2015) we postulated that integrin α1β1 has a pro-tumoral contribution in CRC related to α1 function.

Methods. α1β1 function was studied in HT29, T84 and SW480 CRC cell lines using shRNA silencing targeting α1 (shα1) compared to an shRNA control (shCtrl). Cell proliferation was assessed by cell count and BrdU incorporation. Migration was tested by the scratch test assay. For the survival test, cells were kept in suspension without serum for 24 hours on poly-2-hydroxyethyl methacrylate (polyHEMA)-coated dishes and were then lysed and subjected to caspase3 activity measurement and cleaved PARP expression. To test tumorigenic capacity, shα1 and shCtrl HT29 cells were injected into the dorsal subcutaneous tissue of female CD1 nu/nu mice. The tumor volume was assessed by external measurement. After resection, α1 knockdown was confirmed at the mRNA and protein levels.

Results. In HT29, T84 and SW480 cells, α1 mRNA silencing resulted in reduced cell growth and proliferation compared to the control. Caspase3 activity measurement and PARP cleaved expression in HT29 and T84 cells showed that resistance to anoikis was altered in shα1 cells compared to shCtrl. Wound healing was delayed in sh-α1 HT29 and T84 cells compared to shCtrl. Moreover, tumor development in xenografts was reduced in HT29 shα1 cells.

Conclusions. Our results show that α1β1 is involved in tumor cell proliferation, survival and migration. This finding suggests that α1β1 is involved in colorectal cancer progression. (Supported by the CIHR).

Funding Agencies: CIHR

CAG Student Prize


Role of the Phosphatase DUSP6 in the Control of Intestinal Tumorigenesis and Inflammation
K. Beaudry, A. Montagne, M. Langlois, S. Cagnol, and N. Rivard

Université de Sherbrooke, Sherbrooke, QC, Canada

Aims. The RAS/Mitogen-activated protein kinase pathway (MAPK) is an evolutionarily conserved kinase module that links extracellular signals to the machinery that controls fundamental cellular processes such as growth, proliferation, differentiation, migration and apoptosis. The phosphatase DUSP6 controls this pathway in the cytoplasm by dephosphorylating and inactivating ERK1/2 MAP kinases. To determine the role of this phosphatase in the maintenance of intestinal homeostasis, we characterized the intestinal epithelial phenotype of Dusp6 knock-out (KO) mice under normal, oncogenic and pro-inflammatory conditions.

Methods. Control (Dusp6+/+), Dusp6+/− and Dusp6−/− mice were sacrificed for histology, immunofluorescence [ML1] [KB2], immunohistochemistry, Western blot, and quantitative polymerase chain reaction analysis.

Results. Our results show that loss of DUSP6 does not alter intestinal architecture nor crypt cell proliferation (Ki67 staining). Additionally, no significant difference was observed in the number of Goblet cells (Alcian blue coloration), Paneth cells (lysozyme immunofluorescence), enteroendocrine cells (Chromogranin A staining) and enterocytes (sucrase-isomaltase expression). We tested the progression of inflammation in Dusp6 KO mice in the acute DSS-colitis model. Our data demonstrate that Dusp6 KO mice are protected from colitis, compared to wild-type mice, as determined by measurement of weight loss and histologic scoring. To analyze the potential involvement of DUSP6 in intestinal tumorigenesis, we crossed Dusp6 mutant mice with mice. Notably, a major effect on intestinal tumor initiation is observed in ; Dusp6/ mice compared to mice. We finally knocked out DUSP6 in colorectal cancer (CRC) cells (HT29) using Crispr-Cas9 technology. The deficiency in DUSP6 in CRC cells enhanced ERK1/2 activation levels and promoted anchorage-independent growth in soft agar.

Conclusions. These results demonstrate that the phosphatase DUSP6, by controlling ERK1/2 activation, regulates colonic inflammatory response and protects the intestinal epithelium against oncogenic stress.

Funding Agencies: CIHR

CAG Student Prize


Fecal Immunochemical Testing and Fecal Calprotectin Predict Mucosal Healing in Inflammatory Bowel Disease: A Prospective Study
C. Ma, R. Lumb, R. Foshaug, T. Dang, S. Verma, V. Huang, K. Kroeker, K. Wong, L. Dieleman, B. Halloran, and R. Fedorak

University of Alberta, Edmonton, AB, Canada

Background. Achieving mucosal healing (MH) in patients with Crohn’s disease (CD) and ulcerative colitis (UC) is associated with improved long-term outcomes but direct endoscopic assessment for MH is costly and invasive. Non-invasive biomarkers such as fecal calprotectin (FCP) and fecal immunochemical test (FIT) are potential alternatives for assessing disease activity.

Aims. To evaluate the accuracy of FCP and quantitative FIT for predicting endoscopic MH.

Methods. A prospective cross-sectional cohort study was performed in adult (≥18 years) IBD outpatients presenting for routine colonoscopy. Patients provided a first morning stool sample for FCP and FIT within 48 hours of colonoscopy. Patients on anticoagulation were excluded. MH was defined by (a) Simple Endoscopic Score for CD (SES-CD) of 0 or 1; (b) Rutgeerts score of i0 or i1; or (c) UC Endoscopic Index of Severity (UCEIS) score of 0 or 1. Receiver operating characteristic (ROC) curves were plotted for FCP, FIT, and additive combination FCP + FIT for MH.

Results. Eighty patients (40 CD, 40 UC) were enrolled. Patient characteristics are summarized in Table 1. Disease extent was predominantly ileal in CD (50%) and pancolonic in UC (60%). 23 patients (29%) were on biologic therapy and 50 patients (63%) had endoscopic MH. FCP < 150 μg/g had a sensitivity of 0.97 for detecting MH and an area under the curve (AUC) of 0.75 (95% CI: 0.63–0.86). In comparison, FIT < 50 ng/mL was less sensitive (0.70) but more specific for MH (AUC 0.79 (95% CI: 0.69–0.90)). When used in additive combination (FCP + FIT), performance characteristics were only modestly improved (Figure 1). Combined FCP + FIT score <375 had a sensitivity of 0.80 for MH. FIT was more sensitive for predicting MH in UC compared to CD.

Conclusions. FCP and FIT are sensitive non-invasive methods for predicting MH in IBD patients. They may be used to rule out active disease as an alternative to endoscopic evaluation, especially in UC.

Funding Agencies: The Centre of Excellence for Gastrointestinal Inflammation and Immunity Research

CCC Student Prize


The Interaction between NOD2 and Smoking Is Specific to the 1007fs SNP of the NOD2 Gene in Crohn’s Disease: A Systematic Review and Meta-Analysis
E. Kuenzig,1 B. Eksteen,1 H. Barkema,1 C. Seow,1 C. Barnabe,1 M. Silverberg,2 P. Lakatos,3 R. Panaccione,1 S. Ghosh,1 and G. Kaplan1

1University of Calgary, Calgary, AB, Canada
2Mount Sinai Hospital, Toronto, ON, Canada
3Semmelweis University, Budapest, Hungary

Background. NOD2 variants and cigarette smoking are both commonly implicated risk factors for Crohn’s disease (CD). The three most commonly studied single nucleotide polymorphisms (SNP) of the NOD2 gene are 1007fs, G908R, and R072W. However, only the 1007fs SNP has been confirmed as the susceptibility NOD2 gene for Crohn’s disease in a genome-wide meta-analysis. Prior studies examining the interaction between NOD2 variants and cigarette smoking have reported heterogeneous findings. Because many of these studies were underpowered, the 1007fs, G908R, and R072W SNPs were often pooled together rather than evaluated individually.

Aims. We will examine if some of the heterogeneity observed between studies is explained by SNP-specific NOD2-smoking interactions.

Methods. We searched MEDLINE and EMBASE for studies that provided data on both NOD2 and cigarette smoking among patients with CD. Authors were contacted if the interaction was not reported or when the 1007fs, G908R, and R072W variants were combined. Pooled odds ratios (OR) and 95% confidence intervals (CIs) were calculated using random effects models to estimate the NOD2-smoking interaction. Smoking status was defined as ever or never. We compared the odds of ever smoking among carriers of a NOD2 mutation to those without a NOD2 mutation. All analyses were a priori conducted separately for the 1007fs, G908R, and R702W variants. Heterogeneity was assessed using the and Cochran statistic. Publication bias was assessed using the Begg and Mazumdar adjusted rank correlation test.

Results. Eighteen studies provided SNP-specific NOD2-smoking interaction data. A significant interaction between the 1007fs SNP and smoking (OR 0.70, 95% CI 0.60 to 0.82) was observed (Figure 2). Neither the G908R variant (OR 0.93, 95% CI 0.79 to 1.10) nor the R072W variant (OR 0.89, 95% CI 0.75 to 1.06) were found to have a significant interaction with smoking. Statistically significant heterogeneity and publication bias were not observed for the pooled analyses of 1007fs, G908R, or R702W.

Conclusions. Only the 1007fs NOD2 variant interacts with cigarette smoking in CD. Individuals with CD who have a 1007fs NOD2 mutation are less likely to smoke prior to their diagnosis. Future gene-environment studies in CD should be designed and powered to evaluate SNP-specific mutations.

Funding Agencies: CIHR, Alberta Innovates-Health Solutions

CCC Student Prize


Intravenous Immunglobulin-Induced Regulatory Macrophages Produce IL-10 and May Be Useful to Treat Inflammatory Bowel Disease
L. Kozicky, S. Menzies, and L. Sly

UBC, Vancouver, BC, Canada

Background. Macrophages are key mediators of inflammation, initiating and perpetuating the innate immune response. However, macrophages can be skewed to a regulatory phenotype (Mregs), which plays an equally important role in turning off the inflammatory response. Intravenous Immunoglobulin (IVIG) is a blood product composed of pooled polyclonal immunoglobulins from more than 1000 donors. Our laboratory has reported that IVIG can skew macrophages to Mregs, which produce high amounts of the anti-inflammatory cytokine, IL-10, in response to inflammatory stimuli, like lipopolysaccharide (LPS). High dose IVIG is used to treat some autoimmune and inflammatory diseases. It may work, in part, by skewing macrophages to a regulatory phenotype and may be useful to treat intestinal inflammation, like that, which characterizes IBD.

Aims. My hypothesis is that Mregs can reduce intestinal inflammation, by producing IL-10 in response to pro-inflammatory stimuli. To address this hypothesis, I propose three specific aims:

Aim 1. To determine whether IVIG-induced Mregs can block innate immune-driven inflammation in vitro by producing IL-10.

Aim 2. To determine whether adoptive transfer of IL-10-producing Mregs can reduce intestinal inflammation in vivo in a mouse model.

Aim 3. To determine whether IVIG can reduce intestinal inflammation by skewing macrophages to an Mreg phenotype in vivo in a mouse model.

Methods. Macrophages were derived from mouse bone marrow aspirates and primed with IVIG to skew them to an Mreg phenotype. The ability of Mregs to reduce innate immune-mediated inflammation and its dependence on IL-10 was assessed in vitro in co-culture experiments. The ability of IL-10-producing Mregs to reduce intestinal inflammation in vivo was assessed during dextran sodium sulfate (DSS)-induced colitis. The ability of IVIG to reduce intestinal inflammation in vivo by skewing macrophages to an Mreg phenotype was assessed during DSS-induced colitis.

Results. Mregs suppressed pro-inflammatory cytokine production from LPS-stimulated macrophages in an IL-10-dependent manner. Adoptive transfer of IL-10-producing Mregs and IVIG treatment reduced clinical disease activity and histopathological features of intestinal inflammation in mice during DSS-induced colitis.

Conclusions. Mregs have potent anti-inflammatory activity that can be used to reduce intestinal inflammation in vivo. Adoptive transfer of in vitro-derived Mregs or skewing macrophages to an Mreg phenotype with IVIG in situ may provide novel immunotherapeutic strategies to treat intestinal inflammation in people with IBD. Future studies include assessing whether IVIG skews macrophages to an Mreg phenotype in patients receiving IVIG to treat autoimmune disease.

Funding Agencies: CCC

CAG Paper Session—CAG Selected Clinical Presentations, Friday February 26, 10 h00–11 h30


Trends in Incidence of Pediatric Inflammatory Bowel Disease in Canada: Population-Based Estimates from the Canadian Gastro-Intestinal Epidemiology Consortium (CANGIEC)
E. Benchimol,1 C. Bernstein,2 A. Bitton,3 M. Carroll,4 W. El-Matary,2 A. Otley,5 H. Singh,2 G. Nguyen,6 A. Griffiths,6 D. Mack,1 N. Mojaverian,1 M. Vutcovici,3 Y. Cui,5 Z. Nugent,2 D. Tanyingoh,7 and G. Kaplan7

1University of Ottawa, Ottawa, ON, Canada
2University of Manitoba, Winnipeg, MB, Canada
3McGill University, Montreal, QC, Canada
4University of Alberta, Edmonton, AB, Canada
5Dalhousie University, Halifax, NS, Canada
6University of Toronto, Toronto, ON, Canada
7University of Calgary, Calgary, AB, Canada

Background. The incidence of pediatric inflammatory bowel disease (PIBD) is increasing worldwide, and Canada has amongst the highest rates. Provincial population-based health administrative data can be used to determine national Canadian disease rates and compare regional trends in epidemiology.

Aims. To determine the incidence of PIBD in Canada, and assess trends over time.

Methods. We used validated algorithms to identify children <16 years diagnosed with IBD from administrative data in 5 provinces: Alberta (AB) 1999–2008, Manitoba (MB) 1999–2010, Nova Scotia (NS) 2000–2008, Ontario (ON) 1999–2010, Quebec (QC) 1999–2008. Age- and sex-adjusted incidence was calculated with 95% confidence intervals (CI) by gamma distribution. Statistical trends over time were determined using Poisson regression analyses and reported as annual percentage change. Incidence and annual percentage change were pooled and meta-analayzed across provinces using random-effects models.

Results. A total of 5204 cases of PIBD were newly diagnosed (3456 CD, 1438 UC). The pooled incidence of PIBD in Canada was 9.8 (95% CI 9.2–10.4) per 100,000 children. Incidence was similar amongst provinces, but higher in NS (Figure 3). Meta-analysis of time trends revealed a non-significant rise in incidence for IBD (+2.0%/y, 95% CI −0.7 to +4.7%), CD (+1.6%/y, 95% CI −1.1 to +4.4%), and UC (+1.6%/y, 95% CI −3.7 to +7.0%). The only age subgroup with a significant increased incidence was children 0–5 y (+7.2%/y, 95% CI +2.8–11.5%). Incidence in all children increased significantly in Ontario (IBD: +5.8%/y, 95% CI +4.7–6.9%; CD: +4.8%/y, 95% CI +3.3–6.2%; UC: +6.2%/y, 95% CI +4.3–8.2%) and Quebec (CD only: +4.3%/y, 95% CI +2.3–6.2%). Incidence increased in Ontario for children in all age groups. In addition, CD increased for adolescents aged 14–15.9 y in QC (+5.7%/y, 95% CI +2.5–8.9%) but decreased in NS (−9.8%/y, 95% CI −18.6 to −0.02%).

Conclusions. Canada has amongst the highest incidence of PIBD in the world. Incidence was similar amongst the provinces studied, but highest in Nova Scotia. While meta-analysis demonstrated a non-significant increased incidence overall, the rate rose rapidly and significantly in the youngest children (aged 0–5 y).

Funding Agencies: CCC, Ontario Early Researcher Award, CIHR/CHILD Foundation Canadian Children IBD Network


Steroid-Free Remission among Canadian Pediatric Inflammatory Bowel Disease Patients
P. Church,1 T. Walters,1 E. Benchimol,2 K. Jacobson,3 W. El-Matary,4 C. Deslandres,5 H. Huynh,6 M. Carroll,6 E. Wine,6 D. Mack,2 J. Van Limbergen,7 A. Otley,7 and A. Griffiths1

1The Hospital for Sick Children, Toronto, ON, Canada
2University of Ottawa/CHEO, Ottawa, ON, Canada
3BC Children’s Hospital, Vancouver, BC, Canada
4University of Manitoba, Winnipeg, MB, Canada
5Hôpital Sainte-Justine, Montréal, QC, Canada
6Stollery Children’s Hospital, Edmonton, AB, Canada
7IWK Health Centre, Halifax, NS, Canada

Background. Achieving durable remission without ongoing corticosteroid use is a measure of quality IBD care.

Aims. To ascertain rates of corticosteroid free clinical remission (SFR) and normal linear growth among children with established ulcerative colitis (UC) and Crohn’s disease (CD) at C.H.I.L.D. Foundation/CIHR Canadian Children IBD Network sites.

Methods. Over 6 months, prospective data were collected on consecutive clinic patients (<18 yrs) with diagnosed IBD ≥12 months. Physicians recorded demographics; type of IBD; date of diagnosis; medications; PCDAI/PUCAI; Physician Global Assessment (PGA) of disease activity and clinical symptom pattern; appraisal of linear growth in prior 12 months. Chi-square and Kruskal-Wallis tests were used as appropriate.

Results. 713 patients (CD: 62%; UC: 31%; IBD-U: 7%) were reviewed at 8 sites (6 provinces). Median disease duration was 39 months (IQR 23–62). Median ages were 15.1 and 14.1 years for CD and UC respectively (). Based on PGA, 72% of CD and 78% of UC patients had inactive disease. PCDAI and PUCAI scores were <10 in 80% of CD and 84% of UC patients. 8% of CD versus 4% of UC patients had height velocity <−2SD for age. Assessment of disease activity over the preceding 6 months (continuously quiescent or minimally active in 84%) were similar for CD and UC. In the preceding 6 months, 61% of CD and 65% of UC patients were in SFR (defined as continuous absence of symptoms, PCDAI/PUCAI < 10, normal growth, with no systemic steroid use). CD and UC patients required different therapies to achieve similar rates of SFR (Table 2).

Conclusions. Canadian pediatric gastroenterologists minimize steroid use after the first year of diagnosis with significant use of anti-TNF therapy and immunomodulators. 5-ASA/sulfasalazine use in CD and UC appears to follow evidence-based guidelines. Rates of SFR are similar for CD and UC.

Funding Agencies: CIHR, C.H.I.L.D. Foundation


A Randomized Comparison of High Definition Colonoscopy Alone with High Definition Dye Spraying Chromoendoscopy and Electronic Virtual Chromoendoscopy Using Iscan for Detection of Colonic Dysplastic Lesions during IBD Surveillance
M. Iacucci,1 M. Fort Gasia,1 R. Panaccione,1 A. Oluseyi,1 S. Urbanski,2 M. Parham,2 and S. Ghosh1

1Gastroenterology -University of Calgary, Calgary, AB, Canada
2Department of pathology-University of Calgary, Calgary, AB, Canada

Background. Dye chromoendoscopy (DCE) is currently considered the preferred endoscopic technique for IBD surveillance colonoscopy. However, the resolution of high definition (HD) and virtual chromoendoscopy (VCE) colonoscopy has increased considerably and therefore further studies are needed to determine the optimal endoscopic technique for detection of dysplastic lesions (DL).

Aims. Randomized trial to compare three different techniques for surveillance colonoscopy to detect colonic DL in IBD patients: (HD), (DCE) and (VCE) using iSCAN.

Methods. A randomized study (NCT02098798) was conducted to determine the detection rates of DL with HD alone, DCE or EVC in patients with long standing colitis (8 years from diagnosis, including both UC and CD). Consecutive patients with inactive disease were enrolled in 1 : 1 : 1 ratio into three arms of the study. Colonoscopy was performed using a Pentax EPKi processor and HD video colonoscope (EC-3490Fi; Pentax Tokyo). Endoscopic colonic lesions were classified by the Paris classification as polypoid/non-polypoid and Kudo pit pattern. The lesions were histologically categorized by the modified Vienna classification as dysplasia (ALM and DALM), sessile serrated adenomas (SSAs), adenoma-like polyps (ALP) and hyperplastic polyps (HP). Chi square test was calculated for comparison between the three arms. Sensitivity, Specificity, PPV and NPV were calculated for each arm of the study.

Results. 200 patients (108 male, median age 48 years, range 20–77 years) were assessed by HD (, 35%), VCE (, 32.5%) or DCE (, 32.5%). Twenty-eight SSAs were found in sixteen patients (20.9%); forty-three ALPs were found in thirty-eight patients (32%); six dysplastic lesions were found in five patients (4.5%). Detection rates for ALPs favored the HD group () and when comparing between dysplastic and non-dysplastic lesions, the detection rate favored the HD group (). The three techniques had similar sensitivity and specificity in detecting DL. HD had a sensitivity of 92.5%, specificity of 78.6%, PPV 92.5% and NPV 78.6%. DCE had a sensitivity of 90.9%, specificity of 84.5%, PPV 90.9%, NPV 89.5% and VCE had a sensitivity of 91.7%, specificity of 73.3%, PPV 84.6% and NPV 84.6%.

Conclusions. Our results indicate that DCE, does not yield higher detection rates for colonic DL than either HD or VCE. In fact, the majority of DL were detected in the HD group.

Funding Agencies: None


A Virtual Reality Curriculum in Non-Technical Skills Improves Performance in Colonoscopy: A Randomized Trial
S. Grover,1 M. Scaffidi,1 B. Chana,1 K. Gupta,1 M. Zasowski,1 O. Zarghom,1 C. Dargavel,1 T. Alomani,1 A. Kamani,1 S. Sharma,1 and C. Walsh2

1University of Toronto, Toronto, ON, Canada
2Hospital for Sick Children and The Wilson Centre, Toronto, ON, Canada

Background. Non-technical skills (NTS) are cognitive, social and personal resource skills that complement technical skills and contribute to safe and efficient task performance. Six core NTS are relevant to endoscopy: teamwork, communication, situational awareness, decision making, leadership and professionalism. The need for NTS competence is acknowledged by gastroenterology organizations such as CAG and ASGE but there is minimal evidence supporting the effectiveness of curricular NTS training.

Aims. To assess the effectiveness of a simulation-based curriculum in NTS on novice endoscopists’ performance of simulated colonoscopy.

Methods. 20 novice endoscopists were randomized to 2 groups. The conventional training group received 6 hours of interactive small-group didactic sessions on colonoscopy theory and 6 hours of simulation-based training (SBT) that started on bench-top simulators (low fidelity) and progressed to virtual reality (VR) simulators (high fidelity). Hours 5 and 6 of SBT were integrated scenarios wherein participants interacted with a standardized patient and nurse while performing a VR simulated colonoscopy. The NTS group also received the same didactic sessions with hour 6 focusing on NTS, and 6 hours of SBT that progressed from low- to high-fidelity simulators, including integrated scenarios. Prior to each integrated scenario, participants reviewed a checklist of relevant core NTS concepts. Participants were assessed at baseline, immediately after training, and 4–6 weeks post-training. The primary outcome was NTS performance during an integrated scenario test, measured by OSANTS, an assessment tool for NTS in surgery modified for endoscopy. Secondary outcomes were attitudes towards NTS measured by TEAMSTEPPS, a validated questionnaire of NTS perception; and global performance and communication during integrated scenarios respectively assessed using ISGRF and ISCRF, two previously validated rating scales.

Results. The NTS group outperformed the conventional training group on the integrated scenarios immediately after training and 4–6 weeks after training, in terms of NTS-specific performance (), global performance () and communication (). The NTS group regarded NTS more positively as compared to the conventional training group ().

Conclusions. A colonoscopy simulation-based curriculum focused on NTS improved NTS performance, communication and global performance during simulated colonoscopy encounters, and attitudes regarding NTS. Further research should evaluate the impact of a NTS curriculum on clinical colonoscopy performance.

Funding Agencies: None


Cost-Effectiveness of Hemospray in Patients With Non Variceal upper Gastrointestinal Bleeding
A. Barkun, V. Adam, Y. Chen, Y. Lu, and M. Martel

McGill University, Montreal, QC, Canada

Background. Hemospray (TC-325) is an endoscopic hemostatic powder that achieves hemostasis through adherence to actively bleeding biological surfaces.

Aims. Compare the cost-effectiveness of traditional endoscopic hemostatic therapies (except epinephine injection alone) and Hemospray in different combinations.

Methods. A decision tree of patients with active Non Variceal Upper Gastrointestinal Bleeding (NVUGIB) assessed four possible treatment strategies: traditional therapy alone (T), Hemospray alone (H), traditional therapy completed by Hemospray if needed (T + H), or Hemospray completed by traditional therapy if needed (H + T). Using published probabilities, effectiveness was the likelihood of avoiding rebleeding over 30-day. Costs in 2014US$ were based on the US National Inpatient Sample. Physician and procedure fees were obtained from the American Medical Association and recent publications. A third-party payer perspective was adopted. Sensitivity and subgroup analyses were performed.

Results. For all patients, T + H is more efficacious and less expensive than all other approaches, with 97% of patients eventually avoiding rebleeding at an average cost per patient of US$9,150. The second most cost-effective approach is H + T, 5.57% less effective and costing on average US$635 more per patient. Sensitivity analyses show that T + H followed by a strategy of H + T remain more cost-effective than H or T alone when varying all probability assumptions across plausible, a priori determined ranges. Variations in physician, procedural fees and in the price of Hemospray do not change the final selection of preferred strategy. Varying four assumptions of disease-specific lengths of stay make T less costly than T + H (T + H still more effective). Subgroup analyses showed that patients with non-ulcer lesions at low risk of delayed rebleeding, the Hemospray first approach (H + T) was most effective at low incremental cost ($341 per patient) compared to T + H; the T + H strategy was most effective with varying costs differences relative to the different strategies for all other subgroups. It is more cost-effective to deliver T + H or H + T at the same endoscopy session than at a second-look endoscopy. A strategy of H alone appears ineffective in peptic ulcer bleeding.

Conclusions. Hemospray improves the effectiveness of traditional hemostasis, while being less costly in most patient populations presenting with NVUGIB; a Hemospray first approach may be the most cost-effective for non-ulcer bleeding lesions at low risk of delayed hemorrhage.

Funding Agencies: None


New Oral Anticoagulants and Gastrointestinal Hemorrhage: A Systematic Review and Meta-Analysis
A. Dorreen,1 C. Miller,2 M. Martel,3 and A. Barkun2

1Dalhousie University, Halifax, NS, Canada
2McGill University, The Montreal General Hospital, GI Division, Montreal, QC, Canada
3McGill University Health Center, Montreal, QC, Canada

Background. Several new oral anticoagulants (NOACs) have been approved for clinical use or are in advanced-phase clinical trials, yet evidence regarding associated risk of gastrointestinal hemorrhage (GIB) is limited.

Aims. To determine the risk of GIB associated with NOACs as compared to conventional anticoagulation therapy.

Methods. An initial search for randomized controlled trials comparing NOACs to conventional anticoagulation therapy was performed using the EMBASE, Medline, Cochrane and ISI Web of knowledge databases from inception through March 2015. NOACs already approved or in active development were included. Trials assessing NOACs for the treatment of acute coronary syndrome and other unapproved indications were excluded. Two independent reviewers analyzed abstracts and reviewed manuscript content. Data from relevant papers, including baseline characteristics, indication for and duration of NOAC and number, severity and location of GIB events were compiled. A meta-analysis was conducted with results reported as odds ratios (OR) with 95% confidence intervals (CI). The primary outcome was major GIB. Secondary outcomes included clinically-relevant non-major (CRNM), upper and lower GIB. A subgroup analysis of individual NOACs was performed. Heterogeneity and publication bias were assessed.

Results. An initial search yielded 1654 papers, following review 36 trials were included that assessed dabigatran, rivaroxaban, apixaban, edoxaban and betrixaban. A total of 145,639 patients were randomized. There was no difference in major GIB between NOACs and conventional anticoagulation (OR 0.98, 95% CI: 0.80–1.22). No difference was observed for CRNM GIB (OR 0.92, 95% CI: 0.63–1.34), upper GIB (OR 0.76, 95% CI: 0.37–1.56) or lower GIB (OR 0.86, 95% CI: 0.66–1.13). Subgroup analysis revealed an increased odds of major GIB with dabigatran (OR 1.27, 95% CI: 1.04–1.55) and rivaroxaban (OR 1.40, 95% CI: 1.15–1.70) when compared to conventional anticoagulation.

Conclusions. No difference was found between NOACs and conventional anticoagulation regarding odds of major GIB. Subgroup analysis, however, indicates that dabigatran and rivaroxaban are significantly associated with a 27% and 40% relative increase in odds of major GIB, respectively.

Authors’ Contribution. C. Miller & A. Dorreen are co-first authors.

Funding Agencies: None

CAG Paper Session—Cellular Reprogramming in GI Diseases, Friday February 26, 12 h30–14 h30


The Gut Microbiota-Dependent Metabolite, TMAO, Protects against Colitis: A Role for the Inhibtion of Apoptosis?
A. Al Rajabi, A. Wang, and D. McKay

University of Calgary, Calgary, AB, Canada

Background. Some patients with IBD exhibit signs of endoplasmic reticulum (ER; i.e., unfolded protein response) stress in their intestinal gut epithelium. Also, mutations in multiple genes that encode proteins associated with the ER stress response have been identified as risk factors for IBD development. Unresolved ER stress can lead to apoptosis, which in turn may cause a transient increase in gut permeability, and loss of barrier function is often a hallmark of IBD. Trimethylamine-N-oxide (TMAO), a gut microbiota-dependent metabolite of dietary choline, is a low molecular weight chaperone that can facilitate proper protein folding, thereby, attenuating ER stress.

Aims. To test if TMAO can ameliorate colitis development by means of preventing ER stress and ER stress-induced apoptosis.

Methods. In vivo - CD1 male mice (8–10 weeks old) were given free access to 3% (w/v) dextran sodium sulfate (DSS) in drinking water for 5 days followed by 3 days of normal drinking water ± TMAO (days 0–7; 35 mg/day; oral gavage) and necropsy performed on day 8. Controls received normal drinking water only and both negative and positive (i.e., DSS) controls received water via gavage. Daily body weight was recorded. Colitis was assessed by disease activity score (DAS), colon length and MPO activity. In vitro - Caco-2 (human colon-derived epithelial) cells were treated with tunicamycin (10 μg/mL), an ER stressor, ± TMAO and apoptosis was assessed by immunoblotting for cleaved caspase-3 and cleaved PARP.

Results. In vivo - In 2 separate experiments (-7 mice/group), TMAO-treated mice displayed significant protection against DSS-induced colitis as gauged by colon length, tissue MPO levels and the cumulative DAS. In vitro - as expected tunicamycin induced apoptosis in Caco-2 cells at 20–48 h post-treatment incubation and levels of cleaved caspase-3 and PARP were significantly reduced in the TMAO (50 mM) co-treated epithelia.

Conclusions. The gut microbiota can be both cause and “cure” of intestinal inflammation. Our data indicate that the microbial metabolite, TMAO, inhibits DSS-induced colitis/disease in mice (possibly via inhibition of ER stress-induced apoptosis). Thus, TMAO could be a novel therapeutic in IBD and may be of particular value in individuals whose disease is characterized by ER stress.

Funding Agencies: CIHR, Eyes High Postdoctoral Fellowship


Effects of Milk Lipid Globule Membrane on Post-Natal Intestinal Development
G. Bhinder, R. Dyer, S. Innis, and B. Vallance

University of British Columbia, Vancouver, BC, Canada

Background. At birth, there is a drastic change in nutrient acquisition in a switch from placental nutrition transfer to the introduction of food. Although breastmilk is the ideal nutrient source during the first 6 months after birth it is often not available in great enough quantity, if at all, to the developing infant, resulting in a need for formula feeding. As achieving appropriate growth during this early time of development is essential, it is of utmost importance that the composition of formula reflects that of breastmilk as closely as possible. Interestingly, the lipid fraction of breastmilk, composed of a triglyceride core surrounded by a unique triple membrane structure-the Milk Lipid Globule Membrane (MLGM), represents the main source of energy for the newborn. Surprisingly, most available formulas do not contain this MLGM component, but rather derive their lipids from vegetable sources. To date, the ability of MLGM to effect development, specifically at the intestinal surface, has not been extensively explored.

Aims. To examine the effects of Milk Lipid Globule Membrane supplementation in formula on post-natal intestinal development.

Methods. The rat pup-in-a-cup model was utilized to examine the effects of MLGM (1.2 or 6 mg/mL) supplementation on early intestinal development. In brief, rat pups underwent gastronomy at postnatal day (PD) 5 and were supplemented with formula containing soy (control) or MLGM + soy until PD15, at which point they were euthanized. Ileal and distal colonic samples were collected for histological assessment and immunohistochemistry (IHC), while stool samples were collected for assessment of commensal microbes.

Results. MLGM supplementation in soy formula resulted in a dose dependent increase in colonic crypt depth at PD15, with the 6 mg/mL MLGM + soy group displaying similar crypt depth to mother-reared pups. IHC analysis of colonic intestinal mucus (Muc2+), epithelial cell proliferation (Ki-67+), and enterocyte numbers (CA-1+) revealed similar positive staining at PD15 between mother-reared and 6 mg/mL MLGM + soy supplemented pups compared to soy formula alone, while enteroendocrine (5-HT+) numbers were similar between all three groups. No overt differences in commensal microbes were found between MLGM + soy and soy formula alone supplemented rat pups.

Conclusions. Milk Lipid Globule Membrane supplementation in formula accelerates early intestinal development, resulting in similar colonic proliferation and mucus production to that observed in mother-reared pups. This accelerated development may be protective against early enteric infections, suggesting that MLGM supplementation may be beneficial in neonates who do not have access to breastmilk, particularly pre-term infants who are especially vulnerable to infections.

Funding Agencies: CIHR

CAG Paper Session—Colonic Motility, Friday February 26, 15 h00–16 h30


Mesotrypsin Evokes PAR2 Dependent Excitabilty of Nociceptive Dorsal Root Ganglia (DRG) Neurons
C. Lopez Lopez,1 J. Jaramillo Polanco,1 C. Rolland-Fourcade,2 N. Vergnolle,2 and S. Vanner1

1Queen’s University, Kingston, ON, Canada
2INSERM UMR-1043, Toulouse, France

Background. Mesotrypsin protein and mRNA levels in colonic epithelium are increased in Irritable bowel syndrome (IBS) patients. Our previous studies have shown that proteases in supernatants from IBS-D patients elicited a marked increase in nociceptive dorsal root ganglia neuronal excitability by activating PAR2 but it is unknown if mesotrypsin could play a role in this action.

Aims. This study examined whether mesotrypsin increased the excitability of nociceptive DRG neurons and whether PAR2 signaling was involved.

Methods. Nociceptive mouse DRG neurons (capacitance < 30 pF) were incubated with mesotrypsin (10 nM, 20 min) or trypsin (10 nM, 10 min), to provide a comparator to the known actions of other proteases. In order to evaluate if the protease effect in cell excitability was mediated by PAR2, neurons were incubated with the PAR2 specific antagonist GB83 (10 μM, 30 min) prior to the incubation with mesotrypsin or trypsin. The excitability of neurons was measured by perforated patch clamp recordings by recording changes in the rheobase and action potential discharge at twice rheobase.

Results. Mesotrypsin evoked increased excitability (rheobase decreased 29%, , and the action potential number at twice rheobase increased 29%, ) compared to controls. The effect of trypsin on neuronal excitability was similar to that observed with mesotrypsin (rheobase decreased by 26%, , and the action potential number at twice rheobase increased by 41%, ) compared to control, as previously reported (Valdez Morales et al., 2013). In a separate series of experiments, the effects of mesotrypsin and trypsin were blocked by the PAR2 antagonist GB83. Mesotrypsin or trypsin plus GB83 treated cells showed no difference versus control cells, whereas both protease alone caused increased excitability (mesotrypsin rheobase 29% lower than mesotrypsin plus GB83, , trypsin rheobase 24% lower than trypsin plus GB83, ).

Conclusions. These data suggest that epithelial derived mesotrypsin induces hyperexcitability of mouse DRG neurons in a PAR2 dependent fashion. Taken together with our previous studies demonstrating a PAR2 dependent role for proteases in IBS pain signaling, this data suggests that mesotrypsin could be one of the important mediators.

Funding Agencies: CIHR

CAG Paper Session—Epigenetics of Gastrointestinal Cancer, Saturday February 27, 08 h00–10 h00


VacA-Disrupted Autophagy Promotes Accumulation of Helicobacter pylori Cytotoxin Associated Gene A during Chronic Infection
M. Abdullah,1 L. Greenfield,1 M. Capurro,1 C. O’Brien,2 and N. Jones1

1Cell Biology Program, Research Institute, The Hospital for Sick Children, Toronto, ON, Canada
2Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada

Background. Infection with Helicobacter pylori is the most important risk factor for the development of gastric cancer. The cytotoxin associated gene A (CagA) and vacuolating cytotoxin (VacA) are major virulence determinants of H. pylori. The mechanisms regulating these virulence factors in host cells are important for understanding pathogenesis. We have shown that prolonged VacA exposure disrupts the autophagy pathway. Recent studies suggest that CagA, which is considered an oncoprotein, can be degraded by autophagy. Furthermore, alterations in autophagy can impact the ubiquitin-proteasome system (UPS), another mechanism responsible for degrading cellular proteins. We hypothesized that during chronic H. pylori infection, VacA-disrupted autophagy promotes CagA accumulation leading to enhanced downstream oncogenic CagA signaling.

Aims. Here we investigated the mechanisms by which autophagy modulates CagA in host cells. The objectives were to (1) determine if VacA-disrupted autophagy increases intracellular CagA levels, and (2) determine if disrupted autophagy impacts proteasomal degradation of CagA.

Methods. Human gastric epithelial (AGS) cells and human gastric organoids were infected with the vacA isogenic mutant H. pylori and co-cultured with or without VacA+ or VacA− cultured conditioned media supernatant (CCMS) for up to 48 hours using a gentamycin assay. Wild-type (WT) and autophagy deficient (atg5/) mouse embryonic fibroblasts (MEF) were infected with the vacA isogenic mutant H. pylori to determine the effects of autophagy on intracellular CagA in the absence of VacA. Western blotting was performed to assess CagA levels. Bacterial viability assays were performed to control for differences in survival over time and normalize CagA levels.

Results. When vacA H. pylori infected AGS cells were co-cultured with VacA+ CCMS, there was a significant increase in normalized CagA levels. These findings were recapitulated in human gastric organoids. Atg5/ MEFs infected with vacA H. pylori showed increased CagA levels compared to WT MEFs. Of note, CagA levels decreased over time even in the atg5/ MEFs. AGS cells infected with vacA H. pylori had a 4-fold increase in CagA levels when treated with the proteasome inhibitor, MG132, compared to vehicle control. However, proteasome inhibition of VacA+ CCMS treated cells did not increase the levels of CagA.

Conclusions. Taken together, our findings reveal that VacA− disrupted autophagy may modulate the UPS, which both lead to an accumulation of CagA.

Funding Agencies: CIHR


The Role of SHP-1 as a Tumor Suppressor Gene in Intestinal Epithelium
C. Leblanc, G. Coulombe, and N. Rivard

Université de Sherbrooke, Sherbrooke, QC, Canada

Background. SHP-1, a src homology 2 (SH2) domain containing protein tyrosine phosphatase, functions as a negative regulator of signaling downstream of cytokine receptors, receptor tyrosine kinases and receptor complexes of the immune system. Additionally, SHP-1 has been proposed to be a tumor suppressor gene for several cancers. Of note, we demonstrated that this phosphatase negatively regulates the nuclear transcriptional function of β-catenin in intestinal epithelial cells in culture (Simoneau, Cell Signalling 2011).

Aims. These studies suggest that SHP-1 might exert a tumor suppressive action in the intestinal epithelium.

Methods. Colorectal cancer (CRC) samples paired with their margins were used to analyse relative expression of PTPN6. CRC cell lines were infected with lentivirus coding for a shRNA against SHP-1. Mice with a specific deletion of SHP-1 in intestinal epithelial cells (IECs) were generated using the cre-loxP system. These mice were then breeded with mice.

Results. SHP-1 gene expression was investigated by quantitative analysis in paired samples of CRC (resection margins and primary tumors). Importantly, relative amounts of SHP-1 transcripts are effectively found to be significantly reduced in these colorectal tumors compared to corresponding normal specimens. SHP-1 silencing in human CRC cells (HCT116, HT29) enhances BrdU incorporation in comparison to control cells, suggesting an increased growth rate for these cells. Additionally, SHP-1-depleted CRC cells form significantly more colonies in soft agar. Conversely, ectopic expression of wild-type SHP-1 inhibits the capacity of CRC cells to grow under anchorage-independent conditions. In mice, conditional deletion of SHP-1 in intestinal epithelium leads to an intestinalomegaly associated with an increase in crypt depth and cell proliferation. A marked increased expression of β-catenin protein and Akt phosphorylation is also observed in IECs from mutant mice. While loss of epithelial SHP-1 is not sufficient by itself to initiate tumorigenesis in mice, it severely enhances intestinal tumor load in mice.

Conclusions. These results reveal an anti-proliferative function for SHP-1 in IECs and suggest that this phosphatase functions as modifier gene in colorectal neoplasia.

Funding Agencies: NSERC

CASL Paper Session 1—Saturday February 27, 08 h30–10 h00

CASL Student Prize


Prevalence and Factors Associated with Nonalcoholic Fatty Liver Disease as Diagnosed by Transient Elastography with Controlled Attenuation Parameter in HIV Mono-Infected Patients
E. Vuille-Lessard, L. Lennox, C. Pexos, B. Lebouche, M. Klein, and G. Sebastiani

McGill University Health Center, Montreal, QC, Canada

Aims. Nonalcoholic fatty liver disease (NAFLD) is the most frequent liver disease in Canada and may significantly contribute to mortality among HIV-infected persons. Due to the invasiveness of liver biopsy, data on NAFLD in HIV mono-infected patients are scarce. We investigated prevalence and predictors of NAFLD and liver fibrosis in a large cohort of HIV mono-infected patients, without coinfection with hepatitis B or C, by transient elastography (TE)/controlled attenuation parameter (CAP).

Methods. This was a prospective cohort study at McGill University Health Centre, which included 310 consecutive HIV mono-infected persons (mean age 49.9 years, 77% men). Patients with significant alcohol intake or coinfection with hepatitis B or C were excluded. Any grade NAFLD (>10% of hepatocytes), significant NAFLD (>30%) and severe NAFLD (>60%) were defined as CAP > 232, CAP > 260 and CAP > 292 dB/m, respectively. Significant liver fibrosis and cirrhosis were defined as TE measurement >8 kPa and >13 kPa, respectively. Predictors of NAFLD and liver fibrosis were determined by multivariate logistic regression models.

Results. CAP identified any grade NAFLD, significant NAFLD and severe NAFLD in 55.3%, 33.7% and 16.3% of cases, respectively. Significant liver fibrosis and cirrhosis were found in 11% and 2.3% of cases, respectively. Multivariate analysis results are reported in Table 4. A model combining the identified predictors for significant NAFLD (overweight, protease inhibitors and elevated ALT) showed that presence of at least two predictors had 100% sensitivity to rule in significant NAFLD.

Conclusions. NAFLD diagnosed by TE with CAP is frequent in HIV mono-infected persons, particularly in those with overweight, elevated ALT and exposed to protease inhibitors as antiretrovirals. Of note, significant NAFLD is a predictor of significant liver fibrosis. Longitudinal studies are needed to evaluate the impact of interventions, such as weight loss, aimed at reducing morbidity and mortality due to liver disease in this population.

Funding Agencies: CIHR Canadian HIV Trials Network; unrestricted research funds from ViiV; unrestricted research funds from Merck, study number IIS#51841


Minimal Hepatic Encephalopathy Renders the Brain Susceptible to Hypotension-Induced Neuronal Cell Loss In BDL Rats
M. Clément,1 C. Bosoi,1 M. Tremblay,1 C. Bemeur,2 and C. Rose1

1CRCHUM, Montréal, QC, Canada
2Université de Montréal, Montreal, QC, Canada

Background. Hepatic encephalopathy (HE) is a major neuropsychiatric complication caused by liver disease characterized by cognitive and motor dysfunction. Historically, HE has always been considered to be a reversible metabolic disorder and has therefore been expected to completely resolve following liver transplantation (LT). However, persisting neurological complications remain a common problem affecting as many as 47% of LT recipients. LT is a major surgical procedure accompanied by intraoperative stress, including blood loss and hypotension.

Aims. We hypothesize, in the setting of minimal HE (MHE), the compromised brain becomes susceptible to hypotensive insults, resulting in cell injury and death.

Methods. Six-week bile-duct ligated (BDL) rats with MHE and respective controls (SHAM) were used. Blood is withdrawn from the femoral artery (inducing hypovolemia) until a mean arterial pressure of 30, 60 and 90 mmHg (hypotension) and maintained for 120 minutes. Cerebral blood flow (BCF) was assessed by injecting fluorescent microspheres through the brachial artery. Upon sacrifice, brains were extracted for apoptotic analysis (western blot) and neuronal cell count (immunohistochemistry). In a separate group, BDL rats were treated for MHE with ornithine phenylacetate (OP; OCR-002), administered orally (1 g/kg) for 3 weeks.

Results. Both BDL rats and SHAM-operated controls without hypotension did not display any cell injury or neuronal loss. However, BDL rats following hypotension (30 and 60 mmHg) demonstrated a significant decrease in neuronal cell count in the frontal cortex (using NeuN + DAPI and Cresyl Violet) compared to hypotensive SHAM-operated controls. In addition, neuronal loss was associated with an increased in cleaved caspase-3, suggesting apoptotic cell death. CBF decreased in BDL rats compared to SHAM and correlated with degree of hypotension insult. BDL rats treated with OP resulted in a decrease in blood ammonia and improvement in behaviour and did not lead to neuronal cell death following hypotension.

Conclusions. These findings strongly suggest that cirrhotic patients with MHE are more susceptible to hypotension-induced neuronal cell loss. Moreover, these results suggest a patient with HE (even MHE), with a “frail brain”, will fare worse during liver transplantation and consequently result in poor neurological outcome. Combination of MHE and hypotension may account for the persisting neurological complications observed in a number of cirrhotic patients following LT. Therefore, MHE, should not to be ignored and merits to be treated in order to reduce the risk of neurological complications occurring post-LT.

Funding Agencies: CIHR


Risk Factors and Otcomes of Non-Skin Cancers after Liver Transplantation for Primary Sclerosing Cholangitis
M. Mouchli, S. Singh, J. Talwalkar, E. Loftus, C. Rosen, J. Heimbach, R. Wiesner, K. Watt, and J. Poterucha

Mayo Clinic, Rochester, MN, USA

Background. PSC is associated with significantly increased risk of cancer (Ca) and related mortality. It is unclear how liver transplantation (LT) for PSC modifies this risk.

Aims. To determine the cumulative incidence of and risk factors for Ca and long-term cancer-related mortality in patients (pts) with PSC undergoing LT.

Methods. We identified all pts who underwent LT for advanced stage PSC for non-cholangiocarcinoma indications at Mayo Clinic between 1984–2012 with follow-up through 2/2015. Information on Ca incidence and Ca-related mortality in pts with PSC were extracted. Non-melanoma skin Ca were not included in the analysis. The 1-, 5-, 10- and 20-yr cumulative risks of Ca were estimated using Kaplan-Meier curves. Risk factors were assessed using multivariate Cox proportional hazard analysis.

Results. Two hundred ninety-three pts (mean age,  yrs; 63.3% males, 2.4% smoking at time of LT). Over a median follow-up of 11.5 yrs (6.4–18.6), 60 pts (20.5%) developed 70 non-skin Ca (8 pts developed 2 Ca and 1 pt developed 3 Ca). The most commonly observed Ca were: 48 solid-organ Ca (11 renal, 11 colorectal, 7 prostate, 6 pancreatic, 5 breast, 3 ovarian/endometrial/vulvar, and 1 hepatocellular carcinoma), and 4 patients with recurrent PSC developed de novo cholangiocarcinoma. 22 hematological malignancies occurred (18 PTLD, 2 Hodgkin’s disease, and 2 myelodysplastic syndrome). The 1-, 5-, 10- and 20-yr cumulative incidences of Ca were 1.7%, 6.8%, 14.0% and 17.7%, respectively. Median survival of pts who developed Ca was reduced compared to PSC pts without Ca (9.6 versus 22.1 years, ). On multivariate Cox proportional hazard analysis, mycophenolate mofetil use, tacrolimus-based immunosuppression, and male sex were associated with increased risk of non-skin Ca.

Conclusions. The 10-year cumulative risk of Ca after LT for advanced stage PSC was 14.0% with decrease in overall survival. Mycophenolate mofetil use, tacrolimus-based immunosuppression, and male sex were associated with increased non-skin Ca risk.

Funding Agencies: None


Cirrhotic Patients with Sarcopenia and Sarcopenic-Obesity Have an Increased Risk of Hyperammonemia and Hepatic Encephalopathy
A. Montano-Loza,1 A. Duarte-Rojo,2 R. Bhanji,1 and C. Rose3

1University of Alberta, Edmonton, AB, Canada
2University of Arkansas for Medical Sciences, Little Rock, AR, USA
3Hôpital St-Luc, CRCHUM, Université de Montréal, Montreal, QC, Canada

Background. Muscle mass functions as an alternative site of ammonia detoxification in patients with cirrhosis.

Aims. In this study we aimed to investigate if sarcopenia, myosteatosis, and obesity are associated with hyperammonemia and hepatic encephalopathy (HE) in patients with cirrhosis.

Methods. A total of 204 cirrhotic patients were studied. Muscularity assessment was analyzed using CT scans at the level of the 3rd lumbar vertebral body. Sarcopenia was defined using the skeletal muscle index and myosteatosis according to the muscle attenuation index. Overweight-obesity was defined as a body mass index ≥25 kg/m2. Sarcopenic-obesity was defined as concomitant sarcopenia and overweight-obesity. HE was evaluated clinically (West-Heaven criteria) and defined as absent in patients not using specific treatment (i.e., lactulose, rifaximin) and with no prior episodes of HE in the preceding year. Ammonia blood levels were also performed (nl. 0–35 μmol/L) at the time of the muscularity assessment.

Results. Mean age was years and 141 were males (69%). Sarcopenia was noted in 96 patients (47%), 137 had myosteatosis (67%), 136 were overweight-obese (67%), and 53 (28%) had sarcopenic-obesity. Patients with sarcopenia ( versus μmol/L, ), and sarcopenic-obesity ( versus μmol/L, ) had higher levels of ammonia (Figure 5). Levels of ammonia were not different among patients with myosteatosis ( versus μmol/L, ), and overweight-obesity ( versus μmol/L, ). Patients with sarcopenia (84 versus 63%, ) and sarcopenic-obesity (93 versus 65%, ) had higher frequency of hyperammonemia. Patients with myosteatosis had a trend (), and overweight-obesity was not associated with hyperammonemia (). Lastly, patients with sarcopenia (42 versus 26%, ), and sarcopenic obesity (47 versus 30%, ) had higher frequency of HE. Sarcopenia and sarcopenic-obesity increased the risk of hyperammonemia (OR 3.2, , and OR 7.0, ). Also, sarcopenia increased the risk of HE (OR 2.0, , and OR 2.1, ).

Conclusions. Cirrhotic patients with sarcopenia and sarcopenic-obesity have higher ammonia levels and risk for HE. Muscle mass plays a protective role for hyperammonemia and therapeutic strategies to avoid muscle depletion might decrease the risk of HE in cirrhosis.

Funding Agencies: This study has been funded with a Clinical Research Award from the American College of Gastroenterology Institute 2011.


Reduced Hepatic PGC-1 Leads to Oxidative Stress and Worsened Nafld Progression
A. Besse-Patin and J. Estall

Institut de Recherches Cliniques de Montréal, Montreal, QC, Canada

Background. Non-alcoholic fatty liver disease (NAFLD) is a better predictor of type 2 diabetes than anthropometric parameters, yet diagnosis is difficult and mechanisms underlying this condition are not fully understood. Inefficient mitochondrial fatty acid oxidation and increased reactive oxygen species (ROS) production link mitochondrial health to hepatic insulin resistance and NAFLD. PGC-1α is a transcriptional co-activator shown to regulate mitochondrial function and inflammation. In patients with NAFLD, hepatic PGC-1α expression is decreased, correlating with increased liver fat levels and insulin resistance.

Aims. A causative role or mechanistic link between reduced PGC-1α, hepatic insulin resistance and NAFLD progression has not been investigated. We hypothesized that low hepatic PGC-1α potentiates NAFLD in vivo by increasing ROS to cause insulin resistance.

Methods. Wild Type (WT), Liver Heterozygote (LH), and liver-specific knock out (LKO) mice (males and females) were fed either a chow diet or high fat diet supplemented with fructose (30%) for 25 weeks.

Results. We show that reduced hepatic PGC-1α did not alter hepatic inflammatory pathways or lipid content in chow-fed mice. However, when combined with a western diet, a 50% reduction of hepatic PGC-1α (LH) increased hepatic lipids, liver inflammation and oxidative damage. Interestingly, effects were sex-dependent (pathology was more pronounced in female mice) and unexpectedly, mice with a complete knock-out of PGC-1α in liver were similar to wild-type. To understand the mechanistic link between low PGC-1α and liver pathology, we analyzed ROS production (ROS-sensitive CM-H2DCFDA) and ROS-induced oxidative damage in vitro and found it significantly increased in LH hepatocytes, whereas there were no differences in LKO hepatocytes, agreeing with in vivo data. However, oxidative damage was restored when the related PGC-1b was also deleted in LKO hepatocytes, suggesting compensation by other family members in a total KO. Consistent with a causative link between increased ROS and insulin sensitivity, we show in primary hepatocytes that reduction of insulin-induced Akt activation following PGC-1α depletion was not dose-dependent. LH primary hepatocytes exhibited decreased phospho-Akt in response to insulin, while complete loss of PGC-1α lead to significantly increased phosphorylation of Akt compared to WT controls.

Conclusions. This work demonstrates that reduced liver PGC-1α can worsen fatty liver disease progression towards steatohepatitis when exacerbated by environmental factors such as diet. These data also suggests that PGC-1-dependent ROS production may be a significant contributing factor to hepatic insulin resistance in NAFLD.

Funding Agencies: CIHR


Post-Transplant Cholestasis within 1-Year Predicts PSC Recurrence
S. Wasilenko, E. Lytvyak, A. Montano-Loza, and A. Mason

University of Alberta, Edmonton, AB, Canada

Background. Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease affecting both the intrahepatic and extrahepatic biliary tree of which liver transplant is the only effective cure. PSC recurrence (rPSC) after liver transplant significantly affects long-term graft survival and occurs in 6–59% of transplanted patients. Numerous risk factors for recurrence have been proposed however findings are not reproducible by independent groups. We addressed the hypothesis that rPSC has similar dynamic changes in LFTs within the first year following liver transplant, as seen in patients with viral hepatitis, and that LFT changes may identify patients more likely to develop disease recurrence.

Aims. To determine if the development of cholestasis in the first 12 months after transplant subsequently predicts remote rPSC.

Methods. PSC patients who underwent liver transplant at the University of Alberta Hospital from 1991 to 2012 were included. All data was obtained from electronic medical records. Diagnosis of recurrence was defined on the basis of cholangiography and/or histological findings consistent with rPSC. Cholestasis was evaluated at 3, 6, 9, and 12 months after liver transplant. Severe cholestasis was defined as bilirubin ≥ 100 μmol/L and/or alkaline phosphatase (ALP) ≥ 3XULN. Mild cholestasis was defined as those without severe cholestasis and (i) ALP ≥ 2XULN or (ii) abnormal ALP ≥ 1-2XULN and a bilirubin value from 20 to 100 μmol/L. Recurrence free survival was compared between patients diagnosed with rPSC and those without rPSC.

Results. Seventy two patients were included. Fifty-eight (81%) were male. Mean age at transplant was 42 years (8 to 66 years). rPSC occurred in 18/71 (25%) patients. Mean time to recurrence was 77 months (9 to 172 months). rPSC rates were 9% and 28% at 5 and 10 years respectively. rPSC developed significantly earlier in patients with severe cholestasis at 3 months compared to all other patients without cholestasis (mean versus months Log Rank ). Development of mild cholestasis was associated with earlier rPSC than those without cholestasis at 9 months (mean versus Log Rank ) and at 12 months (mean versus Log Rank ). Overall, the hazard ratio for rPSC was 4.8 (95% CI 1.3–17.0, ) in patients with severe cholestasis at 3 months. Hazard ratios for mild cholestasis at 9 and 12 months was 4.9 (95% CI 1.0–22.9, ) and 4.8 (95% CI 1.8–12.8, ) respectively.

Conclusions. Our preliminary results indicate post-transplant cholestasis within the first 12 months following liver transplant is associated with rPSC. Our results mimic observations of other infectious disease recurrence following liver transplantation.

Funding Agencies: None

CAG Paper Session—Innate Mucosal Immunology, Sunday February 28, 08 h30–10 h30


A Multicenter, Double-Blind, Placebo-Controlled Ph3 Study of Ustekinumab, a Human Monoclonal Antibody to IL-12/23P40, In Patients with Moderately-Severely Active Crohn’s Disease Who Are Naïve or Not Refractory to Anti-TNF: Uniti-2
B. Feagan,1 C. Gasink,2 Y. Lang,2 J. Friedman,2 J. Johanns,2 L. Gao,2 B. Sands,3 S. Hanauer,4 P. Rutgeerts,5 S. Targan,6 S. Ghosh,7 W. de Villiers,8 J. Colombel,3 Z. Tulassay,9 U. Seidler,10 and W. Sandborn11

1Robarts Clinical Trials Inc, London, ON, Canada
2Janssen R & D, LLC, Spring House, PA, USA
3Mt Sinai Med Ctr, New York, NY, USA
4Northwestern U, Chicago, IL, USA
5U Hosp Gasthuisberg, Leuven, Belgium
6Cedars-Sinai Med Ctr, LA, CA, USA
7U of Calgary, Calgary, AB, Canada
8U of Cape Town, Cape Town, South Africa
9Semmelweis U, Budapest, Hungary
10Hannover Med School, Hannover, Germany
11UCSD, La Jolla, CA, USA

Background. In the Ph 2b CERTIFI study, a single IV UST induction dose was effective & safe in CD pts previously failing anti-TNFs (Sandborn W. J. et al., N Engl J Med 2012; 367:1519–1528), but efficacy in pts only failing conventional therapy is unknown.

Aims. We evaluated 2 IV UST induction dose-regimens in a CD population not refractory to anti-TNFs.

Methods. Pts with moderate-severely active CD (CDAI220–450) who failed conventional therapy but were not refractory to anti-TNFs were randomized to a single dose of IV PBO, UST 130 mg, or weight-based tiered UST dosing ~6 mg/kg. Primary endpoint was clinical response at Wk 6 (reduction in CDAI score of >100 pts). At Wk 8, pts transitioned to IM-UNITI maintenance study or had safety follow-up through Wk 20.

Results. Of 628 pts randomized, median disease duration was 6.4 yrs; baseline (BL) mean CDAI was 303; 39%  & 35% were receiving steroids & immunomodulators, respectively at BL; 69% were naïve to anti-TNFs. At Wk 6, 55.5% &  51.7% in ~6 mg/kg & 130 mg UST grps were in clinical response versus 28.7% PBO (). At Wk8, 40.2%  & 30.6% of pts in ~6 mg/kg & 130 mg UST grps were in clinical remission versus 19.6% PBO (). Both UST doses showed significant improvements versus PBO in CDAI, IBDQ, CRP, & fLac & fCal. Proportions of AEs, SAEs, & infections (incl serious infections) were similar in UST & PBO grps. No malignancies, deaths, opportunistic infections or TB occurred in UST-treated pts.

Conclusions. IV UST induced clinical response & remission in pts with moderate-severe CD not previously failing anti-TNFs & was well-tolerated through induction.

Funding Agencies: Janssen Research and Development, LLC


Small Intestinal Bacteria Determine Gluten Metabolism And Immunogenicity
A. Caminero Fernandez,1 H . Galipeau,1 J. McCarville,1 C. Johnston,1 S. Bernier,1 A. Russell,2 J. Jury,1 J. Casqueiro Blanco,3 J. Tye-Din,2 M. Surette,1 N. Magarvey,1 and E. Verdu1

1McMaster University, Hamilton, ON, Canada
2The Walter and Eliza Hall Institute of Medical Research, PARKVILLE, VIC, Australia
3Universidad de Leon, LEON, Spain

Background. About 30% of the population is genetically susceptible to develop celiac disease (CD), but only 4-5% of these will develop intestinal atrophy upon ingestion of gluten. Additional environmental factors, related to alterations in gut microbiota, have been suggested to modulate CD risk. The underlying mechanisms are unknown.

Aims. Our aim was to investigate whether human small intestinal bacteria participate in gluten metabolism and CD pathogenesis using gnotobiotic mouse models.

Methods. Germ free C57BL/6 mice (/group) were di-colonized with Lactobacillus rhamnosus and L. fermentum (Lactobacillus spp) from duodenal aspirates of non-celiac subjects and mono-colonized with Pseudomonas aeruginosa X46.1 (Psa), or di-colonized with Staphylococcus warneri X18.3 and S. epidermidis X18.1 (Staphylococcus spp) isolated all of them from duodenal aspirates of celiac subjects. Mice were also di-colonized with Lactobacillus spp and Psa. Germ-free and altered Schaedler flora (ASF)-colonized mice were used as controls. One week after colonization, 8 out 13 mice received a gliadin gavage (7 mg/mouse). Small intestinal content was collected after 2 h to measure gluten amount by ELISA kit G12 and hydrolytic activities by incubation of gliadin with intestinal washes. Gluten peptide (33-mer) hydrolysis by bacteria was analyzed after incubations using a LC/MS/MS in vitro. After sequencing of peptides produced, immunogenicity was tested using peripheral blood mononuclear cells (PBMCs) induced in vivo in CD patients after oral gluten challenge.

Results. ASF colonization decreased gluten content (<2,500 ng/mL) in the small intestine compared to germ-free mice, that exhibited a range of values reaching a maximum of 12,000 ng/mL. Mono-colonization with Psa also decreased gluten content. Notably, Psa hydrolysis of 33-mer gluten peptide led to the generation of multiple peptides with retained immunogenicity, while Lactobacillus spp hydrolyzed Psa-modified gluten peptides, reducing their immunogenicity. An elastase-like protease from Psa was identified as responsible for the production of immunogenic peptides.

Conclusions. Gluten hydrolysis in the SI results from a combination of digestive and bacterial protease activity. Pathobionts and commensals determine end products of gluten hydrolysis and their antigenicity.

Funding Agencies: CAG, CIHR

CASL Paper Session 2—Sunday February 28, 09 h00–10 h30

CASL Student Prize


Characterization of Hepatitis B Virus (HBV) Lymphotropism and Immune Status in Chronic Hepatitis B (CHB) Pregnant Carriers
D. Wong, S. Gao, S. Joshi, T. Matwiy, G. Samadi Kochaksaraei, G. Bindra, G. van Marle, S. Martin, E. Castillo, and C. Coffin

University of Calgary, Calgary, AB, Canada

Background. Our previous studies have shown that mild CHB disease flares are common in pregnancy in association with diverse viral quasispecies in plasma, including minor variants at positions associated with immune escape. HBV can infect lymphoid cells (i.e., peripheral blood mononuclear cells, PBMC), and the risk of HBV vertical transmission in untreated highly viremic mothers despite immunoprophylaxis has been linked to the transplacental passage of HBV-infected PBMC.

Aims. In pregnant CHB carriers, to characterize HBV genome and replicative intermediates in PBMC, along with assessment of cytokine/chemokine changes during pregnancy.

Methods. In 32 CHB pregnant patients analyzed to date (median age 30.5 y, 55% Asian [16/29], 34% African [10/29], 11% Caucasian [3/29], 3 unspecified; 10% genotype A [2/21], 43% B [9/21], 19% C [4/21], 14% D [3/21], 14% E [3/21], 12 unspecified; 22% HBeAg pos [7/32]), median HBV DNA at baseline versus post-partum in HBeAg pos (including 8/32 who received NA treatment) = versus  IU/mL, and in HBeAg neg = versus  IU/mL, median qHBsAg concentration in pregnant versus postpartum HBeAg pos = versus  IU/mL and HBeAg neg = versus  IU/mL, median ALT pregnant versus post-partum = 19 versus 25.5 U/L (range = 6–111 U/L). Total DNA or RNA was isolated from DNase/Trypsin treated PBMC from 15/32, and HBV DNA and HBV messenger (m)RNA was detected using HBV core gene specific primers by direct/nested PCR and/or reverse transcriptase (RT)-PCR. 32 patients (including 5 matched pregnant/post-partum) were tested for serum levels of cytokines/chemokines (i.e., CXCL-10, IL-12, MIP-1β, IFN γ, & CCL2 pg/mL) in parallel with healthy pregnant controls by ELISA (R&D Systems).

Results. Compared to healthy controls (our data and published literature), all CHB cases tested in pregnancy showed increase in 4/5 serum cytokines measured except in CXCL10. A significant difference in mean IL-12 levels in pregnancy versus post-partum ( versus  pg/mL, ) was noted along with 2-fold increase in ALT from baseline. Additionally, HBV genomes, including mRNA was detectable in 74% (14/19) of the tested PBMC samples.

Conclusions. In CHB pregnant carriers, HBV genomes including replicative indicative mRNA can be detected in circulating immune cells (i.e., PBMC). Compared to healthy pregnant controls, 4/5 cytokines tested were elevated and IL-12 levels significantly increased post-partum in association with ALT flares. Further analysis of maternal TH1/TH2 cytokine profile and assessment of HBV variants in PBMC, along with infant follow-up is warranted.

Funding Agencies: CIHR


End-of-Life Healthcare Costs and Utilization among Patients with End-Stage-Liver-Disease in Ontario: A Population-Based Study
E. Kelly,1 S. Murthy,1 F. Wong,2 T. Shaw-Stiffel,1 L. Scully,1 M. Chalifoux,3 P. Tanuseputro,3 and P. James1

1Department of Medicine, The Ottawa Hospital, Ottawa, ON, Canada
29N/983 Toronto General Hospital, Toronto, ON, Canada
3OHRI, Ottawa, ON, Canada

Background. Healthcare for patients with end stage liver disease (ESLD) is often initiated in response to acute deterioration from disease progression. Despite guarded prognoses, many ESLD patients continue to receive expensive therapeutic interventions at the end-of-life (EOL).

Aims. The aims of this study were to evaluate EOL costs and utilization in patients with ESLD as compared to non-ESLD patients within the province of Ontario.

Methods. Using the Ontario Health Administrative Database, we performed population-based retrospective cohort study was conducted of all decedents within the province of Ontario, between April 1, 2010 and March 31, 2013. Patients with ESLD were defined using international classification of disease codes (ICD-9) for cirrhosis and decompensation (variceal bleeding, encephalopathy, ascites, hepatorenal syndrome and peritonitis). Patients with ESLD were compared to non-ESLD patients on direct health care costs in the last year of life, including hospitalizations, outpatient services and long-term care. Multivariate modelling was performed to compare the total costs in the final 90 days of life, adjusting for individual demographic (age and sex), co-morbidity burden.

Results. The study cohort consisted of 264,754 decedents, of which ESLD patients comprised 2.1% (,575). Direct health care expenditure for patients with ESLD increased more in the last 90 days of life compared to non-ESLD patients. The mean cost in the last 90 days of life was $35,008 for patients with ESLD and $21401 for patients without ESLD (), and this was predominantly related to increased acute care utilization (ESLD = $30,667 (88% of total costs) versus non-ESLD = $15,162 (71%)). ESLD patients had higher rates of acute care utilization in the last and 90 days of life (19 days versus 6 days, ), longer hospitalization stays (additional 4.5 days (95% CI 4.2–4.9 days ()), and increased odds of dying in an institutional (OR 1.9 (95% CI 1.8–2.0; )). ESLD patients with peritonitis or hepatorenal syndrome incurred the highest mean total costs in the last 90 days of life ($43,196 and $38,584, resp.).

Conclusions. EOL care in ESLD patients is associated with substantially higher costs than in the general population, predominantly from increased acute care utilization including hospitalization. Whether optimizing outpatient services for patients and/or offering timely palliative services for patients with ESLD can lead to more cost-effective EOL care warrants further evaluation.

Funding Agencies: None


Protective Immunity upon HCV Reinfection Is Associated with Selection of Memory CD8 T Cell Clonotypes with the Highest Functional Avidity
M. Boisvert,1 M. Abdel-Hakeem,1 J. Bruneau,2 H. Soudeyns,3 and N. Shoukry1

1Centre de recherche du CHUM, Montreal, QC, Canada
2Centre hospitalier de l’Université de Montréal, Montreal, QC, Canada
3Département de microbiologie, infectiologie et immunologie, Université de Montréal, Montréal, QC, Canada., Montreal, QC, Canada

Background. While 25% of individuals with acute HCV infection can clear the infection spontaneously, long-term protection is elusive and reinfections remain common among people who inject drugs (PWID). Importantly, some individuals fail to clear subsequent infections despite a pre-existing HCV-specific memory immune response. We have previously shown that protection from viral persistence upon HCV reinfection correlates with expansion of HCV specific T cells with increased breadth. Viral persistence was associated with limited expansion of virus specific T cells.

Aims. We hypothesized that protective immune memory response is associated with selection of CD8 T cell clonotypes with the highest functional avidity and a polyfunctional profile.

Methods. We FACS sorted HCV-specific CD8 T cells to perform longitudinal T cell receptor (TCR) repertoire analysis as well as to generate CD8 T cell clones from patients that spontaneously resolved (SR) both infections (SR/SR) or that became chronically infected (CI) during reinfection (SR/CI).

Results. Our results showed that, upon reinfection, HCV specific CD8 T cells are recruited from the memory pool. The T cell repertoire was narrower (fewer number of dominant clonotypes) prior to reinfection in SR/SR group, compared to SR/CI group and became even more focused upon reinfection in SR/SR group. Individual HCV-1073 specific CD8 T cell clones generated from one SR/SR and one SR/CI individual exhibited comparable TCR avidity irrespective of the infection outcome. Clones established from SR/SR patient had a higher functional avidity and polyfunctionality index than clones established from SR/CI patient.

Conclusions. In conclusion, our results suggest that protective immune response upon HCV reinfection was associated with focusing of the HCV-specific CD8 T cell repertoire recruited from the memory pool whereby clonotypes with the highest functional avidity and a polyfunctional profile were selected.

Funding Agencies: CIHR, FRQS, ALF


Developing a Prognostic Model for Significant Liver Fibrosis in Hiv-Hepatitis C (HCV) Co-Infected Individuals from the Canadian Co-Infection Cohort Study
N. Moqueet,1 C. Kanagaratham,2 D. Radzioch,2 S. Saeed,3 R. Platt,1 and M. Klein4

1Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, QC, Canada
2Department of Medicine and Department of Human Genetics, McGill University, Montreal, QC, Canada
3Epidemiology, Biostatistics, and Occupational Health, McGill University, MONTREAL, QC, Canada
4Division of Infectious Diseases and Chronic Viral Illness Service, McGill University Health Centre, Montreal, QC, Canada

Background. Liver fibrosis, which can lead to fatal liver failure, advances faster in HIV-Hepatitis C (HCV) co-infection due to higher inflammation. Immune and genetic markers could provide a non-invasive prognostic tool to target HCV therapy to those most at risk.

Aims. Does measuring pro-fibrogenic markers allow better prediction of significant liver fibrosis than clinical risk factors alone?

Methods. A prospective case-cohort study was nested in the Canadian Co-infection Cohort (). From the eligible population (), a random subcohort () and all cases (AST-to-platelet ratio index (APRI) ≥ 1.5) were drawn. Pro-fibrotic markers (IL8, MIP1α  &  β, MCP1, TNFα, RANTES, sICAM1, sVCAM1, CXCL9, CXCL11, TGFβ1, hsCRP, sCD14) were measured from first available visit in the subcohort and cases. Genetic marker at Interferon Lambda (IFNL) rs8099917 was available for majority of the individuals in the study sample. We used Cox proportional hazards with Barlow weights for analysis. Discrimination and calibration were compared between Model 1 (clinical factors only) and Model 2 (Model 1 plus selected markers) for predicting 3-year risk of liver fibrosis. Discrimination was estimated with weighted Harrell’s C index; calibration with the Hosmer-Lemeshow statistic and Gronnesby-Borgan test. Models were internally validated with bootstrapping.

Results. 113 individuals developed significant liver fibrosis over 1300 years of risk for an event rate of 8.63 per 100 person-years (95% CI: 7.08, 10.60 per 100 py). Model 1 included gender, current alcohol use, HIV viral load, baseline APRI, HCV genotype, and age as a restricted cubic spline with 3 knots. Model 1 was nested in Model 2, which also included IFNL rs8099917 genotype and 5 immune markers: IL-8, sICAM-1, RANTES, hsCRP, and sCD14. The C indexes for model 1 versus model 2 were 0.720 (95% CI 0.649, 0.791) and 0.756 (95% CI 0.688, 0.825) respectively. Both models were well-calibrated.

Conclusions. Including markers at IFNL rs8099917, IL-8, sICAM-1, RANTES, hs-CRP, and sCD14 enabled better prediction of the 3-year risk of significant liver fibrosis over clinical risk factors alone. While the improvement in discrimination was small, the model with the immune markers fit better. To assess whether this improvement justifies the additional cost of measuring these markers in the face of highly expensive HCV treatment requires further cost-benefit analyses.

Funding Agencies: CIHR, Canadian Network on Hepatitis C (CanHepC, formerly NCRTP-Hep C)


Ornithine Phenylacetate Attenuates Loss of Muscle Mass and Improves Hepatic Encephalopathy in Bile-Duct Ligated Rats
C. Bosoi,1 M. Oliveira,2 M. Clément,2 M. Tremblay,2 G. Ten Have,3 N. Deutz,3 and C. Rose2

1IRCM, Montréal, QC, Canada
2CRCHUM, Montréal, QC, Canada
3Texas A&M University, College Station, TX, USA

Background. Chronic liver disease (CLD) induces numerous complications including muscle mass loss and hepatic encephalopathy (HE) which negatively impact the clinical outcome. Furthermore, muscle mass wasting and HE have been shown to lead to poor prognosis following liver transplantation. Hyperammonemia is considered the central component in the pathogenesis of HE, however recent studies have suggested ammonia to be toxic to other organs besides the brain, such as the muscle.

Aims. The aim of this study was to investigate the effect of ammonia on muscle mass in rats treated with an oral formulation of ornithine phenylacetate (OP; OCR-002).

Methods. Bile-duct ligated (BDL) rats were divided into 4 experimental groups; (1) Sham; (2) BDL; (3) Sham + OP; (4) BDL + OP. OP was administered orally by gavage (1 g/kg) daily for 5 weeks starting 1 week after surgery. Two days before sacrifice, locomotor activity (day/night) was assessed in infrared beam cages for 24 h. The day of the sacrifice, body weight, fat and lean mass (EchoMRI) were measured, followed by i.p. injection of a stable isotopes tracers cocktail (Phe/Gly) in order to asses fractional synthesis of protein (FSR). At sacrifice, samples were collected to measure blood ammonia (commercial kit), cerebral edema (specific gravity method) and muscle FSR.

Results. At 6-weeks, BDL rats demonstrated a 4-fold increase in blood ammonia versus Sham-operated controls. This increase was reduced by 40% in OP-treated BDL rats. Body weight decreased in BDL rats compared to sham-operated controls (360.2 g ± 13.6 versus 476.8 g ± 10.4; ) and significantly increased following OP-treatment (429.6 g ± 117.9; versus BDL). This was due to a higher gain of lean mass in OP-treated BDL rats compared to BDL rats (303.1 g ± 10.7 in BDL + OP versus 264.4 g ± 10.5 in BDL, ). This was accompanied by increased muscle FSR in OP-treated BDL rats. Fat mass remained unchanged between treated and untreated BDL groups. OP treatment also normalized brain water content in BDL rats. Locomotor activity in BDL rats was reduced compared with sham-operated controls but no significant change was found between BDL + OP and SHAM + OP.

Conclusions. This is the first study demonstrating the efficient ammonia-lowering effect of an oral formulation of OP. Moreover, OP long-term treatment is a safe, non-antibiotic alternative with protective effects on the development of cirrhosis complications such as HE and muscle mass loss in rats with CLD. Whether the effect of OP on muscle mass loss attenuation is a result of lowering blood ammonia or directly improves muscle metabolism remains to be established.

Funding Agencies: CIHR


Protein-Calorie Malnutrition Is Prevalent amoung Cirrhotic Patients Awaiting Liver Transplant as Measured by Direct Estimates of Protein and Calorie Intake as Well as Both Subjective and Objective Tools
K. Marr,1 A. Shaheen,1 L. Lam,2 M. Stapleton,1 K. Burak,3 and M. Raman1

1University of Calgary, Calgary, AB, Canada
2Alberta Health Services, Calgary, AB, Canada
3Univ Calgary, Calgary, AB, Canada

Background. Malnutrition is an important predictor of morbidity and mortality among cirrhotic patients.

Aims. Our objectives were to assess protein-calorie malnutrition (PCM) in cirrhotic pre-liver transplant patients and to study the correlation between subjective global assessment (SGA) and other objective measures of malnutrition.

Methods. We recruited pre-liver transplant adult patients at our center between October 2012 and September 2015. Nutrition status was assessed via the SGA. PCM was assessed by comparing recommended to actual protein and calorie intake. SGA was correlated with body mass index (BMI), dry BMI, handgrip strength (HGS) by calibrated dynometer, and mid-arm circumference (MAC). We used non parametric statistical Methods in our analysis.

Results. Seventy patients were included in this study. The majority were males (, 66%) with a median age of 58 years (IQR: 50–61). Moderate to severe malnutrition was prevalent in our cohort (SGA-A: (21.4%), SGA-B: (42.9%) and SGA-C: (35.7%). There was a significant difference in the recommended calories consumed between SGA groups (A 99% versus C 72%, ). A similar trend was observed for the recommended protein consumed (A 85%, C 62%; ). SGA correlated with BMI (, ; ), Dry BMI (, ; ), and MAC ( cm,  cm; ). HGS was significant according to gender. There was a significant difference in male HGS between SGA ( versus PSI, ), while in females the HGS trended towards a difference ( versus PSI, ). HGS and MAC were strongly correlated (Spearman correlation 0.49, ).

Conclusions. Cirrhotic patients have significant protein-calorie malnutrition. Multiple malnutrition tools including dry BMI, HGS and MAC were precisely able to assess malnutrition.

Funding Agencies: Abbott and Baxter

CAG Paper Session—New Technologies, Sunday February 28, 15 h30–17 h00


MIR-142-3P and Vitamin D-Mediated Regulation of Autophagy: Linking Environmental Gene Interactions in IBD
F. Dang,1 A. Hsieh,1 I. Verpalen,1 R. Murchie,1 L. Greenfield,1 M. Abdullah,1 D. Philpott,2 and N. Jones1

1The Hospital for Sick Children, Toronto, ON, Canada
2University of Toronto, Toronto, ON, Canada

Not published at author’s request.

Funding Agencies: CAG, CCC, CIHR, NASPGHAN

Poster Session 1, Saturday, February 27, 18 h30–20 h00

Chronic Liver Disease Including Alcoholic, Cholestatic, and Metabolic Disease


Sitagliptin for the Treatment of Non-Alcoholic Steatohepatitis in Patients with Type 2 Diabetes
N. Malhotra,1 T. Joy,1 C. McKenzie,1 and M. Beaton2

1Western University, London, ON, Canada
2London Health Sciences Centre, London, ON, Canada

Background. The prevalence of non-alcoholic fatty liver disease (NAFLD) is increasing worldwide. This is likely due to the rising numbers of those with impaired insulin sensitivity, dyslipidemia and obesity. NAFLD is best characterized based on histologic changes with non-alcoholic steatohepatitis (NASH) showing the presence of hepatocyte damage, inflammation and possible fibrosis. Pharmacotherapy has been a growing area of interest to treat NAFLD, specifically through modifying underlying risk factors. In patients with type two diabetes mellitus (DM2), oral hypoglycemic agents such as sitagliptin have proven to be effective. As a highly selective dipeptidyl peptidase-4 (DPP-4) inhibitor, it has proven to decrease HbA1C levels while being weight neutral.

Aims. To determine improvement in liver disease with sitagliptin therapy among patients with DM2 and NASH.

Methods. A randomized double-blinded, placebo-controlled pilot study of sitagliptin therapy (100 mg/day) in patients with biopsy proven non-alcoholic fatty liver disease and type two diabetes mellitus. After baseline evaluation, repeat liver biopsy, anthropometric and biochemical measurements were performed 6 months following treatment. Primary outcome was improvement in liver histology, assessed using the non-alcoholic fatty liver disease activity score (NAS) and change in hepatic steatosis measurement using MRI Iterative Decomposition of water and fat with Echo Asymmetry and Least-squares estimation (IDEAL). Secondary outcomes included improvement in the individual components of the NAS and liver fibrosis.

Results. Twelve patients completed follow up. There was no significant reductionin NAS (0.20, ) 95% CI (−1.62, 2.02) or MRI IDEAL (2.0, ) 95% CI (−7.3, 11.2) in those treated with sitagliptin compared to placebo. There was a non-significant improvement in hepatocyte ballooning, but no improvement in lobular inflammation (0.60, ) 95% CI (−0.13, 1.33), steatosis (0.00, ) 95% CI (−1.08, 1.08) or fibrosis (0.40, ) 95% CI (−0.98, 1.78).

Conclusions. Use of sitagliptin therapy in non-alcoholic fatty liver disease patients with DM2 did not lead to a significant improvement in liver histology or hepatic fat measurement on MRI. The small number of patients as well as the relatively short follow up duration of study may have an effect on potential clinical significance.

Funding Agencies: PSI—Physicians Services Inc. Foundation


A Quality Assurance Audit of the Atlantic Liver Transplant Program
A. Dorreen,1 S. Gruchy,1 M. Laryea,2 C. Guimont,2 M. MacNeil,2 and K. Peltekian2

1Dalhousie University, Montreal, QC, Canada
2Queen Elizabeth II Health Sciences Centre, Halifax, NS, Canada

Background. Variable outcomes for liver transplantation have been reported since its introduction as a therapeutic option for patients with end-stage liver disease. In Atlantic Canada, the Multi-Organ Transplant Program in Halifax, Nova Scotia serves as the primary referral center for liver transplantation. Regular evaluation of our program remains a main process by which quality improvement is made. Here we present a 10-year audit of the Atlantic Liver Transplant Program (ALTP).

Aims. We conducted a quality assurance audit of the ALTP in order to characterize patients accessing transplant services and their outcomes post transplant. The information will be used to promote development and provide feedback to all health care professionals involved in our transplant program.

Methods. The ALTP database was accessed to identify all patients that were referred for transplantation between 2005 and 2015. Basic demographic information was extracted, including information on the number of patients referred and transplanted, gender, age, Model for End-Stage Liver Disease (MELD), and reason for transplantation. Basic descriptive statistics and qualitative analyses were performed. Approval from the research ethics board was sought.

Results. Since 2005, a total of 264 transplants were performed on 248 patients, of which 37.1% were female () and 62.8% male (). During this time period, 42 patients who were on transplant waitlist were withdrawn, 26 of which were removed for disease stability (61.9%). The mean age at transplant was for females and for males. The mean MELD was and for males and females respectively. The main indication for transplant in women was primary biliary cirrhosis (PBC) (23.5%, ) and HCV in men (20.0%, ). 16 patients underwent repeat transplantation for varying indications, representing 6.1% of all transplants. The mean time spent on the waitlist was days. The overall survival rate post transplant was 71.4% (); 71 patients died post transplant with a mean survival of days.

Conclusions. These data highlight that the patient population being referred to the ATLP is variable. Primary indication for transplantation varies according to gender, with PBC and HCV related cirrhosis being the main indication in females and males respectively. Survival post transplant was 71.4% and mean wait list time was days. Further development is needed to identify factors that predict poor outcomes post transplant. The data presented here will be used for program development and education.

Funding Agencies: None

Clinical Practice

Poster of Distinction


Correlating Fit Results with Neoplastic Findings in the BC Colon Screening Program
N. Shahidi,1 L. Gentile,2 G. Lovedeep,2 J. Hamm,2 M. Colleen,2 R. Enns,1 and J. Telford1

1St. Paul’s Hospital, Division of Gastroenterology, Department of Medicine, University of British Columbia, Vancouver, BC, Canada
2British Columbia Cancer Agency, Vancouver, BC, Canada

Background. Fecal immunochemical test (FIT) performance depends on the test cut-off chosen with the Canadian Partnership Against Cancer (CPAC) recommending FIT positive predictive value (PPV) for neoplasia be ≥50%. Currently, there is no data assessing FIT performance at different cut-offs in a large average risk Canadian population.

Aims. Evaluate FIT performance for detecting neoplasia.

Methods. Data was obtained from a prospectively collected central database maintained at the British Columbia Cancer Agency (BCCA) (Vancouver, Canada) with consecutive participants of the BC Colon Screening Program (CSP) included. A single quantitative FIT with a cut-off of ≥50 ng hemoglobin/mL buffer solution was used. Participants with a positive FIT were referred for colonoscopy (CSPY) and were classified by the highest risk pathology. High-risk polyps (HRPs) were defined as adenomas or sessile serrated adenomas/polyps (SSA/Ps) ≥10 mm, adenomas ≥20% villous, adenomas with high-grade dysplasia, SSA/Ps with dysplasia, and traditional serrated adenomas. Cancer, HRPs and multiple (≥3) polyps were considered high-risk while 1-2 tubular adenomas or SSA/Ps <10 mm were considered low risk. This study was approved by the BCCA Research Ethics Board.

Results. From Nov-2013 to Dec-2014 32,129 participants had a positive FIT (positivity rate: 13.6%). Of those 20,000 (62.2%) underwent CSPY within the CSP, with 1679 (5.2%) pending CSPY at time of analysis. Pathology results were available for 13,497 (67.5%). The PPV of CRC and HRPs increased with increasing FIT cut-off alongside an increase in CSPYs saved; however, this also resulted in an increase in significant lesions missed (Table 7).

Conclusions. This is the first Canadian study evaluating the PPV of different FIT cut-offs in a large screening population. As the FIT cut-off rises PPV for CRC and HRPs increases alongside CSPYs saved but at the cost of missed lesions. The current cut-off of 50 ng/mL meets CPAC recommendations.

Funding Agencies: None


Endoscopists Based Performance Report Cards Improve Adenoma Detection Rates in Screening Colonoscopies in High Risk Patients
A. Liu,1 M. Sey,1 S. Asfaha,2 C. Vinden,3 L. Stitt,1 and B. Yan3

1University of Western Ontario, London, ON, Canada
2Columbia University, New York, NY, USA
3London Health Sciences Centre, London, ON, Canada

Aims. High quality colonoscopy is a necessary component of an effective colon cancer screening program. The adenoma detection rate (ADR) is considered one of the most important determinants of quality. The aim of this study is to determine the impact of endoscopists based performance report cards on ADR in patients undergoing colonoscopy for positive fecal occult blood tests or family history of colon cancer.

Methods. As a quality improvement initiative starting in 2012, annual report cards were issued to each endoscopist comparing their performance to the overall group mean. The histology of all polyps retrieved was manually confirmed from pathology reports. High risk adenomas were defined as those >1 cm, villous component, or serrated adenoma. The ADR at baseline, one year (Year 1), and two years (Year 2) after report card implementation were compared. Secondary outcomes include polyp detection rate (PDR) and high risk ADR. Comparison between the three years was performed with a chi-squared test followed by pairwise comparisons using a Bonferroni correction. Effect modification by endoscopist specialty and baseline ADR were assessed using interaction terms. All endoscopists included fulfilled Cancer Care Ontario colonoscopist standards and performed >200 colonoscopies yearly.

Results. A total of 3,115 screening colonoscopies performed by 17 endoscopists at a single center were included. The overall ADR, PDR, and high risk ADR was 38.2% (95% CI, 36.5–39.9%), 48.2% (95% CI, 46.5–50.0%), and 14.3% (95% CI, 13.1–15.6%), respectively. The overall ADR for gastroenterologists was 43.0%, general surgeons 31.8%, and hepatologists 24.0% (). The ADR increased from 34.5% at baseline to 39.4% at Year 1 (), which was sustained in Year 2 with an ADR of 41.7% (). From baseline to Year 2, the PDR improved from 45.0% to 51.8% () and high risk ADR from 11.7% to 17.2% (). There was no evidence of effect modification by specialty or baseline ADR although there was a trend towards greater benefit among those with low and mid-ranged baseline ADR.

Conclusions. Significant sustained improvements were seen in the ADR in high risk patients undergoing screening colonoscopy following the introduction of performance report cards. Physician performance reporting should be included in colonoscopy quality assurance and improvement initiatives.

Funding Agencies: None


The Impact of Warmed Carbon Dioxide Insufflation during Colonoscopy on Polyp Detection: A Randomized Double-Blind Controlled Trial
J. Green,1 A. Patel,2 and L. Hookey1

1Queen’s University, Kingston, ON, Canada
2Queen’s University, Mississauga, ON, Canada

Background. Colonoscopy is used for detection of neoplastic polyps, but significant miss rates are reported. Methods to reduce spasm of the colon have been investigated to increase adenoma detection rates by allowing better inspection of colonic folds. Room temperature carbon dioxide (CO2) insufflation has been demonstrated to be as efficacious as water immersion for both decreasing patient discomfort and achieving similar adenoma detection rates. These studies, however, utilized un-warmed CO2, which can produce spasms when released from high-pressure storage tanks. Warmed water instillation has been shown to reduce colon spasm; therefore, administration of warmed CO2 during colonoscopy may improve polyp detection.

Aims. To determine whether colonoscopy using warmed CO2 insufflation achieves greater detection of polyps per patient compared to room air insufflation.

Methods. This was a prospective, single centre, double-blinded, randomized control trial using warm CO2 versus room air insufflation. Patients undergoing colonoscopy for screening and surveillance indications were included and randomized to receive either room temperature room air or warmed CO2 (37 degrees Celsius). The primary outcome was polyp detection rate. A pre-specified power calculation determined that 444 enrolled patients would allow for detection of 50% increase in polyp detection rate, with alpha 5% and beta 20%. Secondary outcomes included adenoma detection rates and advanced lesion detection rates.

Results. The study was stopped after 222 patients had been recruited, as an interim analysis determined that continuation would be futile. Data was available for 202 participants. The room air and warmed CO2 groups consisted of 106 and 96 participants, respectively. The groups were similar in age (), gender (), indication for examination (), and bowel preparation score (). Sixty-five percent of participants in the room air group had polyps (), compared with 59% of participants in the warmed CO2 group () (). Adenomas were detected in 51 and 44 participants in the room air and warmed CO2 groups, respectively (). There was no difference between groups in number of adenomas detected ().

Conclusions. Warmed carbon dioxide insufflation did not improve polyp or adenoma detection rates when compared with room air insufflation. One potential reason is that CO2 does not exert a significant effect on colonic motility. Alternatively, there may have been a loss of temperature of the CO2 as it travelled from the insufflator to the tip of the endoscope, thereby reducing its potential effect. At this time, warmed CO2 cannot be recommended as a method for increasing polyp or adenoma detection rates.

Funding Agencies: None


Psychosocial Factors as Predictors of Sexual Functioning in Inflammatory Bowel Disease
D. Tripp, H. Yurgan, M. Ropeleski, C. Nickel, A. Muere, S. Vanner, and M. Beyak

Queen’s University, Kington, ON, Canada

Background. Inflammatory Bowel Disease (IBD), consisting of Crohn’s Disease (CD) and Ulcerative Colitis (UC), is a painful chronic gastrointestinal disease characterized by inflammation. The impact of IBD on sexual function is poorly understood. More specifically, differences in sexual function between CD and UC have not been examined.

Aims. The initial objective was to examine differences in sexual function between CD and UC. The second objective was to examine predictors of sexual dysfunction.

Methods. 302 patients participated with 94 excluded due to incomplete measures of sexual function. Patients, both male () and female (), were recruited from tertiary care clinics at Hotel Dieu Hospital in Kingston, Ontario, and completed a questionnaire including demographic, pain catastrophizing, social support, disability, depressive symptoms, and sexual function measures. Sexual function was assessed with the Golombok-Rust Inventory of Sexual Satisfaction (GRISS). The GRISS has subscales of impotence and premature ejaculation in males, anorgasmia and vaginismus in females, nonsensuality, lack of communication, avoidance and dissatisfaction in both males and females. Higher scores indicate more sexual dysfunction. Differences in GRISS subscales between CD () and UC () were examined using -tests with Bonferroni corrections. A hierarchical regression with a biospsychosocial framework was used to examine predictors of total sexual functioning. Biological and demographic (age, gender, disability) variables were included in the first step, followed by psychological variables (pain catastrophizing, depressive symptoms) in step 2 and social variables (social support) in step 3.

Results. The overall sample had low to moderate total sexual dysfunction (M = 4.87, SD = 2.59). CD patients (M = 3.94, SD = 2.42) scored significantly higher on avoidance than UC patients (M = 3.01, SD = 2.17), , . CD and UC did not significantly differ on any other subscale of the GRISS. The regression was significant, , . In step 2 of the regression, depressive symptoms (, ), age (, ) and gender (, ) were significant predictors of total sexual function. However, when perceived social support (, ) was added in step 3, depressive symptoms were no longer a predictor of sexual function. Therefore, older age, being female, and less social support predict greater sexual dysfunction in this IBD sample.

Conclusions. CD and UC did not differ in overall sexual function. However, CD patients may avoid sexual activity more than UC patients. Furthermore, a greater perception of social support may allow for better sexual functioning.

Funding Agencies: CCC


International Multicentre Study Comparing Risk Scoring Systems for Patients Presenting with upper Gastrointestinal Bleeding: Findings of the Canadian Cohort
A. AlNasser, H. Gregor, M. Chu, B. Yan, J. Gregor, A. Rahman, and M. Sey

London Health Sciences Centre, London, ON, Canada

Background. Upper gastrointestinal bleeding (UGIB) is common and causes significant morbidity and mortality. Epidemiological data regarding UGIB in Canada is scarce and the last cohort study examining this was conducted nearly 15 years ago.

Aims. The objective of this study is to provide an update on the epidemiology of UGIB in Canada based on the Canadian Cohort of an international multicentre study.

Methods. A prospective open cohort study was conducted at two tertiary care hospitals in London, Ontario over 12 months.

Patients admitted to hospital with a primary diagnosis of UGIB were invited to participate. Baseline demographics, presenting symptoms, comorbidities, medication usage, vital signs, and Clinical Rockall Score (CRS) were recorded at the time of admission. Patient outcomes were evaluated during their hospitalization, including blood transfusion requirement, rebleeding, length of stay, esophagogastroduodenoscopy (EGD) finding and therapy, need for embolization or surgery, and 30-day mortality.

Results. Ninety-nine patients were recruited during the study (mean (SD) age was 65.6 (17), 39% females). The most common presentation was melena (73%). Twenty-two percent had established coronary heart disease and 59% were on an antiplatelet agent or anti-coagulation. Hypotension was the presenting symptom in 8% of patients and 12% had an initial hemoglobin ≤60 g/L. The mean (SD) CRS was 2.9 (1.7) and 35% has a low risk score <3. EGD was performed within 24 hours in 67% of patients and the most common finding was esophagitis/gastritis/duodenitis (23%), normal (22%), peptic ulcer disease (20%), and varices (16%). Endoscopic therapy was performed in 26% of cases and the most common modality used was combination injection and cautery/hemostatic clip (7%) or esophageal variceal banding (11%). Compared to patients with CRS ≥3, patients with low risk CRS <3 had a trend toward fewer blood transfusions (mean of 1.9 versus 1.5 units), less rebleeding (10% versus 6%), fewer embolization or surgery (2% versus 1%), and a shorter length of stay (9.8 versus 6.8 days), although none reached statistical significance. A CRS ≥ 3 was strongly associated with 30 day mortality (6% versus 0%, ).

Conclusions. This is an update on the epidemiology of UGIB in Canada. A CRS ≥ 3 is strongly associated with 30-day mortality.

Funding Agencies: None


Colonoscopy Report Completeness Improves at St. Paul’s Hospital following the Implementation of a Dictation Template
N. Harris, J. Yonge, C. Galorport, M. Suzuki, B. Bressler, E. Lam, J. Telford, and R. Enns

St. Paul’s Hospital, Vancouver, BC, Canada

Background. The completeness of a colonoscopy dictation report is a quality indicator for endoscopic practice. Several guidelines outlining the key elements of a colonoscopy report were used to develop a dictation template at St. Paul’s Hospital in 2013.

Aims. To assess if colonoscopy report completeness at St. Paul’s Hospital has improved with the implementation of a dictation template.

Methods. Literature review findings regarding current dictation guidelines, such as those released by the CAG and ASGE, were compared to dictation template recommendations in use at St. Paul’s Hospital. The colonoscopy reports of five physicians were reviewed at two time points, before (2008) and after (2014) the introduction of the dictation template. 150 charts were reviewed per doctor for each time period. The presence of variables will be assessed and percent completion of reports in 2008 and 2014 were compared. Cecal visualization rate, polyp detection rate and Spearman correlation between the percent completion of reports and report length were calculated. The study was approved by the UBC Ethics Board.

Results. The overall completeness of reports increased from 64% to 85% with the implementation of the dictation report template. Most item completion rates remained stable or increased; however, inclusion of a clinical preamble/indication(s) for procedure and reporting of patient comfort both decreased. Reporting of patient comfort, co-morbid conditions and current medications remained low at both time points.

Conclusions. Dictation report completeness increased from 2008 to 2014 following the implementation of a colonoscopy dictation template. Most items not included in the St. Paul’s Hospital dictation template were consistently under-reported and modifications will be made to further improve report completeness.

Funding Agencies: None


Impact of Types of Questions Asked on Gastroenterology Econsultation Outcomes
S. Canning, N. Saloojee, A. Afkham, C. Liddy, and E. Keely

University Of Ottawa, Ottawa, ON, Canada

Background. Wait times in Canada to see a gastroenterologist continue to exceed the recommended targets of 2 weeks to 2 months for most indications. eConsult services facilitate primary care providers (PCPs) ability to communicate directly with specialists for advice. It can also reduce the need for patients to wait for face-to-face consultations with specialists. Since 2010, the Champlain BASE (Building Access to Specialist Advise) eConsult service has permitted PCPs to submit patient specific clinical questions to specialists via a secure web service.

Aims. To describe the types of Gastroenterology questions asked through a unique eConsult service, and assess the impact on referrals for face-to-face consultations.

Methods. Gastroenterology cases submitted to the Champlain BASE eConsult service between April 2014 and January 2015 were categorized for Gastroenterology-content using a modification of the International Classification for Primary Care (ICPC-2) taxonomy. The type of question (e.g., diagnosis or management) was classified using a validated taxonomy. Other data included the time for specialist to complete the eConsult, the perceived value of the eConsult by the PCP and the need for a face-to-face referral following the eConsult.

Results. Of the 121 Gastroenterology eConsults, 33% were liver related, 23% were GI symptom related (abdominal pain, gastroesophageal reflux disease, diarrhea, and constipation), and 13% were related to specific luminal diseases (irritable bowel syndrome, coeliac disease and inflammatory bowel disease). Of the 40 eConsults related to hepatology, 47% were questions regarding abnormal liver function testing. This was also the most common area of questioning overall (16%). Overall 51% of eConsults were related to diagnosis, 30% to management, 9% to drug treatments and 7% to procedures. It took the specialist <15 minutes to complete the eConsult in 67% of cases. The service was perceived as highly beneficial to providers and patients in 97% of cases. In 47% of submitted cases, a traditional referral was originally contemplated by the PCP but was now avoided and 1% resulted in a new referral that was not originally contemplated by the PCP. In the 24% in whom a referral was still needed, the PCP indicated that a more effective face-to-face consultation would occur.

Conclusions. The eConsult service provided timely, highly regarded advice from gastroenterologists directly to PCPs and often eliminated the need for a face-to-face consultation. With limited resources and access to gastroenterologists across Canada, eConsults provide a means to assist PCPs. Unnecessary referrals are avoided, thus reducing wait times for more urgent referrals. We plan to use the types of questions asked to inform planning of future CPD events for PCPs.

Funding Agencies: CIHR, Ministry of Health and Long-term Care, The Ottawa Hospital Academic Medical Organization Innovation Fund, eHealth Ontario, The Ottawa Hospital Department of Medicine and Bruyere Research Institute


Is the Canadian Version of the Global Rating Scale a Valid and Reliable Tool for Measuring Quality Process in Endoscopy Units?
S. Carpentier, N. Sharara, M. Sewitch, S. El Ouali, M. Martel, and A. Barkun

McGill University, Montreal, QC, Canada

Background. Quality in endoscopy is critical to ensure improved patient outcomes in colorectal cancer screening. The Global Rating Scale (GRS) was developed in the UK to provide metrics for quality in endoscopy and, although never formally validated, has been associated with improved patient outcomes nationally.

Aims. To psychometrically test the adapted Canadian version of the Global Rating scale (GRS-C), in view of its deployment by the Canadian Association of Gastroenterology throughout the country to participating endoscopy units.

Methods. The GRS-C was assessed at 3 institutions by endoscopy unit physicians, endoscopy nurses, and administrative personnel. The psychometric properties evaluated included validity, reliability, and responsiveness to change. For face validity, a group comprising staff not familiar with the GRS was assembled. Each of the groups responded to the questions of the GRS-C that span 12 items. Content validity was assessed by comparing GRS-C questions to national quality indicators in endoscopy, while construct validity was determined by ssociating questions of the GRS-C with those of its original UK GRS counterpart. The GRS-C was completed at the 3 sites both at time 0 and 2 weeks later, with no intervening change to processes, as well as 6 months later after practice changes had been implemented as a result of patient satisfaction questionnaire responses, to respectively assess test-retest reliability and responsiveness to change. Descriptive and inferential data analyses were completed, including kappa values for agreement and paired assessments when comparing question responses.

Results. Face validity was demonstrated as the majority of participants were able to accurately identify the overarching theme each item was intended to measure. For content validity, 18 of 23 key quality indicators (78%, 95% CI: 56–93%) determined by an expert consensus group were addressed in the GRS-C. Statements not included related to educational programs and monitoring of competency. When comparing GRS-C and GRS-UK ratings for all 3 sites, concordance ranged from 75–100% across all three sites, while Kappa agreement levels on test-retest reliability ranged from 0.65 to 0.83. Following a series of process change initiatives, responsiveness to change in 6-month post-implementation scores were statistically higher () in two endoscopy units.

Conclusions. The GRS-C appears to exhibit satisfactory psychometric properties that can be used as part of a national quality initiative aimed at improving processes in endoscopy units. Linking GRS scores to actual patient outcomes is required to ensure that GRS-C implementation helps to achieve improved patient care.

Funding Agencies: None


Gastroenterologists Concerns and Expectations towards Personalised Medicine
S. Veilleux,1 M. Villeneuve,1 N. Lachapelle,1 R. Kohen,2 L. Vachon,3 B. White Guay,4 J. Rioux,4 and A. Bitton5

1Université Laval, Quebec, QC, Canada
2McGill University & Montreal General Hospital, Montreal, QC, Canada
3iGenoMed Consortium, Montreal, QC, Canada
4Université de Montréal, Montreal, QC, Canada
5McGill University, Montreal, QC, Canada

Background. Canadian physicians generally recognise the benefits of personalised medicine (PM) and using tools that predict disease course or response of treatment to inform therapeutic decisions. The implementation of PM is inconsistent across specialities and across Canada. This irregularity results in a decrease in treatment efficiency, an increase in healthcare systems costs and a diminished return on investment for firms involved in the development of predictive tests.

Aims. The aim of this study is to identify factors that affect the decision to use a predictive test that would assist gastroenterologists in their clinical decision-making for the personalised treatment of Inflammatory Bowel Disease (IBD).

Methods. Data was collected through six focus groups across Canada. Meetings were held in Vancouver, Saskatoon, Toronto (2), Quebec and Hamilton. A total of 28 gastroenterologists contributed to the study. Participants were asked about their current tools for treatment decision-making, their perception of present predictive tests and their expectations towards characteristics and implementation conditions of a new predictive test concerning IBD. Meetings were transcribed and encoded using QDA Miner software: 61 codes were initially produced and regrouped into 8 categories and 21 concepts were analysed.

Results. Four major concerns were raised by physicians about predictive testing are issues of accessibility (availability and cost), delays in turnaround time for the reception of results, doubts on the reliability of the test, and lack of proper training in the field of pharmacogenomics. Accordingly, physicians state three expectations: upcoming predictive tests have to show clear reliability qualities, significant benefits for the patient, and being accessible in a timely manner. Finally, the fundamental condition to the implementation of such a test refers to gastroenterologists active involvement in both the clinical trials leading to the test’ endorsement by health agencies and training conferences.

Conclusions. Personalised medicine is spreading rapidly and new tools are being developed continuously to get the best out of the available treatments for patients and other biologics to come. Therefore, physicians are opened to predictive testing to help them in their decision-making process. However, they want to be involved in the development, regulatory approbation as well as the diffusion of such technology.

Funding Agencies: CCC, CIHR, Genome Quebec, Genome Canada


Developing an Assessment Tool for Non-Technical Skills in Endoscopy: A Qualitative Study
M. Scaffidi,1 B. Chana,1 F. Kassam,1 U. Farrukh,1 S. Brar,1 W. Lee,1 S. Scaffidi,1 A. Rasheed,1 G. Hui,1 C. Walsh,2 and S. Grover1

1University of Toronto, Toronto, ON, Canada
2Hospital for Sick Children and The Wilson Centre, Toronto, ON, Canada

Background. Non-technical skills (NTS) in endoscopy are regarded by gastroenterology-focused organizations such as CAG and ASGE as core competencies. In other procedural specialities, such as surgery and anaesthesiology, instruments have been developed to assess NTS. To date, no assessment tool for NTS in endoscopy has been developed.

Aims. To determine the necessary NTS relevant to endoscopy and develop an assessment tool for these skills.

Methods. To identify existing assessment tools of NTS in other disciplines, a comprehensive literature search was conducted. Six tools were identified as relevant to NTS assessment in endoscopy. Gastroenterologists and experienced endoscopic registered nurses (RNs) from an endoscopic unit at an academic hospital were invited to participate in 3 two-hour focus groups to discuss the role of NTS in endoscopy and to identify areas of assessment of NTS. The first focus group was with gastroenterologists, and the second was with RNs. The final focus group combined participants from the previous sessions to reach concordance on the salient themes and identify exemplars of behaviours in each theme. Each group was led by a single moderator with prior experience in conducting focus groups. All focus groups were recorded and transcribed. Transcripts from the first two sessions were analyzed via thematic network analysis using NVIVO software. These findings were presented during the final focus group.

Results. 40 staff (18 physicians; 22 RNs) participated in the focus groups. After initial coding of the transcripts, there were 60 and 45 individual themes derived from the gastroenterologist and RN focus groups, respectively; 29 themes overlapped between the two groups. With refinement of these themes, six dimensions of NTS were found to encapsulate those relevant to endoscopy: teamwork, communication, situational awareness, decision making, leadership and professionalism. Behaviours commensurate with these dimensions identified from the focus groups were adopted into a scoring tool that ranked performance of each domain on a Likert-type scale of 1 to 5, representing a performance of poor to excellent, respectively.

Conclusions. We developed an instrument for the assessment of NTS in endoscopy on the basis of themes and behaviours raised from three focus groups, and arranged in a framework similar to NTS assessment tools in other procedural disciplines. Further research is required to test the feasibility, validity and reliability of the tool in the endoscopic setting.

Funding Agencies: None


The Role of Colonoscopy within 24 Hours of Presentation for Acute Lower Gastrointestinal Bleeding (ALGIB)—A Systematic Review
I. Roshan-Afshar,1 M. Seyed Sadr,1 L. Strate,2 M. Martel,1 and A. Barkun1

1McGill University, Montreal, QC, Canada
2University of Washington, Seattle, WA, USA

Background. ALGIB is a common and potentially lethal disorder that often requires hospitalization. The role and benefits of EARLY colonoscopy in the management of ALGIB remain controversial.

Aims. The present study is a meta-analysis of published studies assessing the role of early colonoscopy (within 24 hours of presentation) in the management of ALGIB.

Methods. Systematic searches were completed querying MEDLINE, EMBASE, CENTRAL and ISI Web of knowledge until March 2015 using the terms related to gastrointestinal hemorrhage and early (emergency, early) colonoscopy. We included randomized controlled trials (RCTs) and observational studies assessing the role of early colonoscopy in patients with ALGIB. The primary outcome was 30 day-rebleeding. Descriptive statistics were generated as were risk ratios and weighted mean differences to characterize the utility of early colonoscopy.

Results. Amongst 824 citations, 10 observational studies (1 abstracts) and 2 RCTs included an early colonoscopy population (total ,807 (including in the 2 RCT intervention early colonoscopy arms), mean age range 52–78 yrs, 48.8% females, 71.1% hemodynamic instability, units of PRBC in first 24 hrs). Of the two RCT control groups (colonoscopy after 24 hours, ; mean age 61.5 yrs, 37.2% females, 68.0% hemodynamic instability, mean units of PRBC in first 24 hrs), one included randomization to angioembolization. Endoscopic findings classified as probable or definitive cause of bleeding amongst all early colonoscopies (excluding two studies that solely assessed diverticular bleeding) were diverticula (41.5%), ulcers (7.7%), angiodysplasia (5.1%), colitis (4.1%), cancer (2.1%), and other findings (24.1%). No cause was found in 15.4% of patients. 34.7% of patients had endoscopic therapy when data were available (injection 15.3%, ligation 12.0%, thermal 4.0% or combination therapy 3.3%); other approaches included surgery (10.6%) or radiological therapy (15.5%). The rebleeding rate after 24 hours was (28/171) 16.4%, and the overall rebleeding rate (54/285) 18.9%, while all cause of mortality was (5/256) 2.0%; one study reported a mean ICU stay of 1.8 days (50 patients). Amongst the 2 RCTs there were no significant attributable differences after 24-hour rebleeding rates compared to controls RR = 0.99 (0.55; 1.65) or mortality RR = 0.50 (0.09; 2.66).

Conclusions. Early colonoscopy with appropriate endoscopic hemostasis can be carried out in patients with ALGIB within the first 24 hours of presentation, with diverticula being the most common etiology. Observational data are disparate, but based on RCT evidence, early colonoscopy identifies more patients with a definitive cause of bleeding but does not result in decreased rebleeding or mortality.

Funding Agencies: None


Wait Time for Colonoscopy Is Reduced by Email Communication
A. Liao,1 T. Jeyalingam,1 and F. Saibil2

1University of Toronto, Toronto, ON, Canada
2Sunnybrook Health Sciences Centre, Toronto, ON, Canada

Background. Timely access to specialist care is a challenge in Canada. The CAG recommends colonoscopy be completed within 2 months for patients with a positive FOBT, but only 55% of such patients receive colonoscopy within 2 months. Delays in time-sensitive diagnosis can negatively impact outcome. Traditionally, patients see their endoscopist prior to colonoscopy (Visit 1) for a medical assessment and explanation of the procedure. While some endoscopists forgo this visit and first meet patients immediately prior to the procedure, this practice may negatively impact outcomes by increasing the risk of medical errors and not allowing for full informed consent. We suggest that, for some patients requiring colonoscopy, Visit 1 can be replaced by email communication to obtain a medical history and explain the procedure and prep.

Aims. The goals of this study are to evaluate: (1) the impact of email communication on wait-time intervals from referral to colonoscopy; and (2) patient satisfaction with this method, compared to a standard office visit prior to colonoscopy.

Methods. We retrospectively reviewed 109 new patients referred for colonoscopy to a single gastroenterology practice between January 2013 and June 2015. 13 patients for whom the wait time from referral to colonoscopy was greater than 180 days were excluded, as this was beyond the maximum offered wait time for this particular practice and was often related to patient-initiated delay. The remaining 96 patients were divided into two cohorts based on whether or not email communication was established prior to colonoscopy. For patients in the email cohort (), a standardized email history had been obtained, and both prep instructions and procedure risks were emailed. Age, gender, reason for referral (symptomatic/FOBT+, positive family history, or routine screening), number of patients in which Visit 1 was eliminated, time from initial referral to colonoscopy, and patient satisfaction (determined by a standardized Likert scale questionnaire) were calculated and compared between cohorts.

Results. The average age and proportion of females were 44.6 years and 69.8% in the email cohort, and 56.5 years and 71.7% in controls. Visit 1 was eliminated in 67.4% of patients in the email cohort. Mean wait time from referral to colonoscopy was 26 days shorter in the email cohort as compared to controls (72.5 days versus 98.5 days, ). This effect was most pronounced in symptomatic patients; wait time was 35 days shorter in email patients within this subgroup (60.3 days versus 94.8 days, ). No significant difference was observed in patient satisfaction scores.

Conclusions. Use of email in colonoscopy scheduling decreases wait time by roughly one month without compromising patient satisfaction. Larger, prospective/randomized studies are required to confirm these findings.

Funding Agencies: None


Are We Choosing Wisely? An Analysis of Institutional Adherence to Colonoscopy Surveillance Schedules
M. Sheridan,1 A. Dorreen,1 S. Gruchy,2 J. Jones,3 and S. Williams3

1Dalhousie University, Bedford, NS, Canada
2QEII Health Sciences Centre, Halifax, NS, Canada
3Queen Elizabeth II Health Sciences Centre, Halifax, NS, Canada

Background. Colorectal Cancer (CRC) is the second most common malignancy in males and third most common in females. Colonoscopy remains the gold standard for screening, but is a limited resource, with local wait time in excess of 2 years. Shortening the interval between screening colonoscopies exposes patients to unnecessary testing and risk, and also further limits access to colonoscopy. The factors used to determine the screening interval include the patient’s family and personal history of polyps, number, size and location of polyps, their complete removal, quality of bowel preparation, and pathology of polyps. All these factors except pathology findings are available at the time of colonoscopy. It is our hypothesis that the information needed to make a recommendation is available at the time of colonoscopy for the majority of patients. If this is proven correct it would provide a strong argument to make this a standard component of each colonoscopy report.

Aims. The aim is to determine whether local colonoscopists are adhering to Canadian colon cancer screening guidelines when making screening interval recommendations.

Methods. Using Clinical Outcomes Reporting Initiative software, all colonoscopies completed from January to December 2014 at the QEII Health Sciences Centre for CRC Screening for patients with average risk, family history and previous polyp/cancer were analyzed. Data on the findings and details of the procedure were recorded (number and size of polyps, location of polyps, completeness of polyp removal, quality of bowel preparation, cecal intubation rate, operator characteristics, and recommended screening interval).

Results. 175 colonoscopy reports were analyzed. A screening interval recommendation was made at the time of colonoscopy in 150 cases (86%). Of the remaining 25 cases in which a recommendation was not made, sufficient information was available at time of colonoscopy in 44% (). If no recommendation was made at time of colonoscopy, 28% () made a follow-up recommendation within 6 months. The correct screening interval was recommended in 83% () of cases, which varied between gastroenterologists (92%, ) and surgeons (80%, ).

Conclusions. Further education is required to ensure all colonoscopists are aware of and comfortable following the current screening guidelines. Making a follow up recommendation at the time of colonoscopy is important to ensure appropriate follow up. Including this as a required field in colonoscopy reporting may help ensure that appropriate and timely recommendations are being made.

Funding Agencies: None


Does It Work in the Real World? Effectiveness Study of Bowel Preparation for Colonoscopy—A Comparison of the Real World and Several Randomized Controlled Trials
C. Wang, R. Yang, S. Vanner, and L. Hookey

Queen’s University, Kingston, ON, Canada

Background. The importance of bowel preparation quality is well recognized for high quality colonoscopies. There have been many different bowel preparations protocols tested over the years in randomized controlled trials (RCT’s). However, RCT’s are completed in an ideal setting. In reality, factors such as compliance, education, and patient motivation can significantly affect the effectiveness of the interventions. There have been no effectiveness studies for bowel preparations which measure the real world impact of interventions.

Aims. In this study, we compared the quality of bowel preparations of real-world patients undergoing colonoscopies to those enrolled in several RCT’s.

Methods. The type of bowel preparation, age, sex, average number of bowel movements per day, comorbidities, indication for colonoscopy, the Ottawa Bowel Preparation Scale (OBPS), and colonoscopy completion rate were collected from five prospective randomized controlled trials (RCT) aimed at assessing colon cleansing using various preparation regimens, and compared with data from two real world diary studies. A total of 1372 patients who underwent colonoscopy from these 7 studies were analysed. The different bowel preparation types included a polyethylene glycol preparation (PEG) or sodium picosulfate plus magnesium citrate preparation (P/MC) taken in a traditional or split-dose regimen.

Results. There were no significant differences in sex and average number of bowel movements per day between the RCT and real world patients. However, RCT patients tended to be younger than real world patients (), this is seen when a prospective study assessing specifically patients over 70 was excluded. 6.0% of RCT patients undergoing colonoscopy were diagnosed diabetes mellitus (DM) and 39.8% of real world patients were diagnosed with DM, COPD, and/or kidney disease. PEG traditional and split-dose preparations do not differ significantly in OBPS between RCT and real world patients (, 0.2 resp.). However, both the traditional and split-dose Pico-Salix preparations are associated with significantly better OBPS in RCT patients when compared to real world patients ( for both). Pooled RCT results showed that split-dose P/MC regimen produced better preparation than traditional dose P/MC (); this was not seen in the real world population ().

Conclusions. We have shown that real world patients have higher OBPS than RCT patients. This suggests that the motivation and education around these regimens may play a very important role in their success.

Funding Agencies: None


A Proposal for Optimizing Patient Selection for Urgent Open Access upper Endoscopy in a Large Central Access Model of Acute Gastroenterology Care
K. Rioux,1 M. Swain,1 P. Belletrutti,1 K. Novak,1 B. Kathol,2 G. Heather,2 and R. Kabani1

1Division of Gastroenterology, University of Calgary, Calgary, AB, Canada
2Alberta Health Services, Calgary, AB, Canada

Background. Many centres in Canada have adopted a central access model and open access endoscopy to improve consistency and efficiency in providing appropriate and timely GI services. Despite guidelines for the appropriate use of upper endoscopy, it remains difficult to determine from referring history which patients require the most urgent attention for common indications such as dysphagia.

Aims. Our aim was to perform a quality assurance audit of the Urgent+ Endoscopy Pathway at GI Central Access and Triage, Foothills Medical Centre, Calgary.

Methods. All patients triaged to urgent consultation and same-day upper endoscopy from January 1 to July 31, 2015 were identified. Anonymized patient records were obtained to determine referral indication, patient characteristics, and endoscopic findings.

Results. 202 patients were seen during the seven-month study period. 112 (55%) patients were referred for dysphagia with severe or progressive symptoms, 42 (21%) for suspected upper GI bleeding, and 39 (19%) for abnormal imaging. The median wait time from date of referral to date of consultation and endoscopy was 4.4 weeks. Abnormalities were identified in 79 (39%) patients, including cancer in 10 (5%) patients; 61% of patients had completely normal upper endoscopies. Of 112 patients with dysphagia, 41 (37%) had abnormal findings, the most common of which was Schatzki ring in 15 patients (13%); 4 patients (3.6%) had gastro-esophageal malignancy. Of 42 patients with suspected upper GI bleeding, 6 (14%) had gastro-duodenal erosions, and 6 (14%) had esophagitis; only one patient had a high-risk peptic ulcer requiring endoscopic therapy. Of 39 patients referred for abnormal imaging, 6 patients (15%) had malignancy.

Conclusions. Despite triage criteria and GI physician review of these referrals, a significant percentage of patients triaged to our Urgent+ Endoscopy Pathway have normal endoscopies or findings irrelevant to the referring indication. In patients with dysphagia in particular, lack of clinical detail or inaccuracies in referring history make it difficult to discern which patients are most likely to have significant pathology. At time of direct access endoscopy booking, we propose that a nurse-led, check-list driven, patient-reported confirmation of onset, timeline, progression, and impact of dysphagia will allow better prediction of important endoscopic findings, including malignancy, to best utilize our most urgent portal of access to outpatient upper endoscopy.

Funding Agencies: Division of Gastroenterology, Department of Medicine, University of Calgary


Weak Correlation between Left and Right Adenoma Detection: A Single Regional Endoscopic Colorectal Cancer Screening Centre Experience
P. James,1 M. Hejaji,1 M. Alhegagi,1 A. Rostom,2 C. Dube,3 S. Murthy,4 and A. Chatterjee1

1The University of Ottawa, Ottawa, ON, Canada
2University of Ottawa, Ottawa, ON, Canada
3the ottawa hospital, Ottawa, ON, Canada
4The Ottawa Hospital, Ottawa, ON, Canada

Background. Although overall and right adenoma detection (RAD) has been widely emphasized in recent studies, left-sided adenoma detection (LAD) may also be an important tool for measuring colorectal cancer (CRC) screening quality.

Aims. To determine if LAD is is associated with RAD and evaluate factors that may be associated with LAD and RAD.

Methods. This is a retrospective cohort study of patients who underwent a screening colonoscopy between May 2009 and December 2011 with a noted family history of CRC or positive fecal occult blood test (FOBT). Data regarding patient demographics (age and sex), procedure details (indication, number and location of polyps identified, bowel preparation and completeness) and polyp histology were captured. The main outcomes examined indclude adenoma detection rate (ADR), LAD rate (LADR), RAD rate (RADR), mean number of adenomas detected per positive colonoscopy (MADPC), left-sided MADPC (LMADPC) and right-sided MADPC (RMADPC).

Results. 2,178 patients and 14 endoscopists were included in the analysis. The median patient age was 59 years and 42% were male. 35% of procedures were performed for family history of CRC and 65% were performed for abnormal FOBT. ADR was 24%, RADR 12%, LADR 14%, MADPC 1.75, RMADPC 0.81 and LMADPC 0.83. Endoscopist RADR was found to be moderately associated with LADR (Spearman rank correlation [S] = 0.60, ; see Figure 6). Two endoscopists had RADRs above the median and LADRs below the median. One endoscopist had RADR below the median and LADR above the median. Three endoscopists had both RADRs and LADRs above the median and their ADRs were greater than 25% (ADR range 27 to 43%). LMADPC was poorly associated with RMADPC (S = −0.37, ). In multivariate analyses, factors associated with RAD and LAD include patient age, patient sex, procedure indication and endoscopist experience. RAD, but not LAD, was associated with endoscopist colonoscopy volume.

Conclusions. LAD is poorly associated with RAD. LAD may offer another opportunity to improve quality by providing detailed feedback for endoscopists involved in colon cancer screening.

Funding Agencies: The University of Ottawa Department of Medicine


Do Adjuvants Add to the Efficacy of Polyethylene Glycol-Based Bowel Preparations? A Meta-Analysis of Randomized Controlled Trials
S. Restellini,1 O. Kherad,1 M. Martel,2 C. Ménard,3 and A. Barkun2

1University of Geneva, Geneva, Switzerland
2McGill University, Montreal, QC, Canada
3Université de Sherbrooke, Sherbrooke, QC, Canada

Background. Polyethylene glycol (PEG) bowel preparations are safe and effective but require the consumption of large volumes of fluid, with relative low associated adherence. Alternatively, some data suggest adjuvants may enhance bowel cleansing quality and patient acceptance.

Aims. We performed a meta-analysis to determine the efficacy, willingness-to-repeat, and procedural outcomes of adding any type of adjuvant to a PEG bowel preparation, given as split-dose and non-split, in high (>3 L) or low-volume (<2 L) regimens.

Methods. We performed systematic searches of MEDLINE, EMBASE, Scopus, CENTRAL and ISI Web of knowledge until March 2015 for published randomized trials that assessed any regimen of PEG with adjuvant versus PEG without adjuvant. We excluded studies that included pediatric, hospitalized, or IBD patients, or trials in which the control group also received an adjuvant. Adjuvants were categorised as osmotic laxatives, irritant laxatives, antifoam products or other (prokinetics). The primary outcome was efficacy of bowel cleansing. Secondary outcomes included patients’ willingness to repeat the procedure, polyp and adenoma detection rates.

Results. Of 2813 citations, 31 trials () fulfilled the inclusion criteria. PEG low-dose preparations with an adjuvant were not inferior to PEG high-dose OR = 1.03; (0.79–1.34); 20 studies. PEG high-dose preparations plus an adjuvant resulted in a significantly greater proportion of patients with adequate preparations (OR = 1.96 (1.32–2.94), 9 studies). Adjuvant combined to PEG low-dose did not enhance bowel cleansing compared to PEG low-dose alone (OR = 0.68 (0.43–1.09), 2 trials) Results were similar when analyses were restricted to split-dose comparisons. To our knowledge, no study assessed split PEG low-dose versus split PEG low-dose with adjuvants. Willingness-to-repeat was significantly greater with the use of PEG low-dose with adjuvants compared to PEG high-dose preparations (OR = 3.70 (2.0–6.67); 12 studies), but was lower for PEG high-dose with adjuvants versus PEG high-dose (OR = 0.63 (0.42–0.94)). Results were similar for split-dose comparisons. No differences were noted in polyp or adenoma detection rates.

Conclusions. Efficacy of bowel cleansing for PEG low-dose with the addition of an adjuvant was not inferior to PEG high-dose, and yielded a higher proportion of patient willingness to repeat the preparation. PEG high-dose was more efficient with an adjuvant compared to same dosage without adjuvant but was less tolerated. No differences were noted in polyp or adenoma detection rate. Additional research is required to further characterize the impact of adjuvants in PEG low-volume, especially with split-dosing regimens.

Funding Agencies: None


Split-Dose versus Same-Day Bowel Preparations for Colonoscopy: A Meta-Analysis
C. Ménard,1 A. Barkun,2 M. Martel,2 O. Kherad,3 and S. Restellini3

1Université de Sherbrooke, Sherbrooke, QC, Canada
2McGill University, Montreal, QC, Canada
3University of Geneva, Geneva, Switzerland

Background. A variety of bowel preparation types and administration schedules are available. Contemporary regimens include polyethylene glycol (PEG), sodium phosphate (NaP), picosulfate (PICO) and oral sulfate solution (OSS).

Aims. To compare efficacy, willingness to repeat the preparation, polyps and adenoma detection rates and side effects of split-dose versus same-day preparations amongst all contemporary regimens and subgroups comparing PEG high-dose (≥3 L) and PEG low-dose (<3 L).

Methods. Systematic searches were completed querying MEDLINE, EMBASE, Scopus, CENTRAL and ISI Web of knowledge from January 1980 to September 2015. All fully published randomized controlled trials with colon preparation for colonoscopy were included. Populations including pediatric, sole inpatients or sole IBD patients were excluded. The primary outcome measure was the efficacy of colon cleansing (excellent or good). Secondary outcomes included willingness to repeat, polyp and adenoma detection rates and side effects. A meta-analysis was conducted with results reported as odd-ratios (OR) with 95% confidence intervals. Heterogeneity and publication bias were assessed and quantified

Results. From an initial 2580 citations, 11 trials fulfilled the inclusion criteria ( ITT). Same-day administration does not provide a significant benefit in bowel cleansing efficacy in comparison to split-dose regimens, regardless of preparation type, volume or use of adjuvants (OR = 1.19 (0.81; 1.75)). When performing sensitivity analysis, the exclusion of Cesaro et al. revealed a significant benefit in the efficacy of split-dose administration (OR = 1.47 (1.13; 1.91)). Willingness to repeat does not differ between the 2 groups (RC = 0.87 (0.38; 2.01)). Split-dose administration causes significantly more cramping, abdominal pain or bloating (OR = 1.50 (1.10; 2.05)). Polyp or adenoma detection rates are not different between the 2 administration regimens. 5 studies were included ( ITT) in subgroup analysis of PEG split high-dose versus PEG same-day low-dose. There was no significant difference between the 2 regimens (OR = 1.07 (0.64; 1.78)). Only one study (125 patients) permitted the comparison of PEG split low-dose versus PEG same-day low-dose and showed no difference in efficacy (OR = 0.91 (0.34; 2.47)).

Conclusions. There is no significant difference in efficacy between same-day and split-dose administration of bowel preparations. However, there is great variability in type and administration schedule of adjuvants in available same-day preparations arms, which may have an effect on the generalizability of the present Results. Further studies should focus on more rigorous comparisons of the different administration schedules.

Funding Agencies: None


Randomized Prospective Study: Impact of the Patient Education Website on the Quality of Outpatient Bowel Preparation for Colonoscopy: Intrim Data
M. Suzuki,1 M. Bardi,2 O. Takach,2 C. Galorport,1 J. Yonge,1 N. Harris,1 E. Lam,1 J. Telford,1 G. Rosenfeld,1 H. Ko,1 and R. Enns1

1St Paul’s Hospital, Vancouver, BC, Canada
2UBC, Vancouver, BC, Canada

Background. Low level bowel cleanliness that occurs in 25% of patients, hinders polyp detection rates and limits colonoscopy effectiveness. Those with inadequate preparation have incomplete examinations, fewer polyps detected, more repeat colonoscopies and higher resource utilization. Aside from pharmacological and timing of purgative factors to optimize bowel preparation, non-pharmacological factors that influence patient compliance (i.e., patient education) in the preparation phase can significantly improve bowel preparation quality.

Aims. To assess if interactive, individualized web based instruction lead to improved colonoscopy preparation through enhanced patient compliance, satisfactin and tolerability of preparation.

Methods. A randomized, prospective, single blinded trial initiated at St. Paul’s hospital in Vancouver, B.C. Inclusion criteria: age > 19, planned outpatient colonoscopy, and willingness/ability to participate by reading the online English material supplied by sending the subject a specific domain (which contains the educational platform of information for their colonoscopy). Exclusion criteria: None. Consecutive patients enrolled into the study (target of 450 participants). Data Collected: demographics, cancellations, bowel preparation cleanliness scores as per Boston bowel preparation quality (BBPS) and Ottawa bowel preparation score. Primary end points: percentage of patients that achieve an excellent BBPS following web-based instructions versus paper instructions. Assessment of patient satisfaction, preparation tolerability and patient activation score through post colonoscopy follow-up surveys.

Results. As of October 2015, 285 subjects have been recruited. 127 are male; mean age 57 years (range 20–81). 142 were assigned to Group A (paper based) and 143 to Group B (web based). A Fisher’s exact test showed a significant difference in the proportion of subjects achieving an excellent BBPS score ≥8 (Group A = 57% (81/142), Group B = 71% (102/143) ). There was no significant difference in patient reported satisfaction (), helpfulness () or clarity of instructions ().

Conclusions. Interim analysis detected a significant difference in patients achieving excellent bowel preparation scores between interactive individualized web based instructions versus written instructions. This study will continue to recruit until a target sample size of 450 to determine if our other primary end points will achieve significance. We plan to complete recruitment by Dec 2015. If preliminary data is supported by final results; the use of this platform will be encouraged to maximize ideal patient preparations for colonoscopy.

Funding Agencies: None


Colonscopy Quality Assurance and Maintanance of Cometency among Pediatric Gastroenterology Ftass Members—A Pilot Project
C. Barker and M. Alaifan

University of British Colubmia, Vancouver, BC, Canada

Background. Colonoscopy quality indicators in addition to maintenance of competency skills are relatively well established in the adult literature, however it is much less so in pediatric gastroenterology. One of the suggested quality assurance measures which is relevant for both adult and pediatric patients would be cecal intubation rate, which it has been suggested should be ≥90% as per ASGE guidelines.

Aims. The purpose of this study was to evaluate the cecal and terminal ileal (TI) intubation rates at our tertiary care pediatric centre. The aim is evulate the centre quialty of colonoscopies compared to the adult standards.

Methods. A retrospective chart review study was performed on all pediatric patients (age 16 months–18 year old) who underwent colonoscopies at our single centre performed between January 2013 to July 2014 (18 months period). Patients scheduled for sigmoidoscopy were excluded. The endoscopy reports were reviewed to ascertain whether the cecum and TI were reached as well as quality of bowel prep and any other stated reasons for reasons of failure. Clinical charts were reviewed to obtain indication for colonoscopy.

Results. A total of 288 colonoscopies were performed by 5 gastroenterologists during the 18 month period. The number of colonoscopies per staff ranged from 36–70 procedures. The numbers of year in practice ranged from (3–25 years). The overall cecal intubation rate was 98.3% (range 97.1%–100%). TI intubation rate was lower at 84.4% (range 66.7%–90%). The main stated reason for inability to enter cecum/TI was technical difficulty and poor bowel prep. No complications were encountered in those procedures.

Conclusions. Despite relatively low volumes, cecal intubation rates are very good exceeding some suggested standards. TI intubation rates were lower and it was noted there was a higher degree of variability. Multi centre evaluation over a longer time period and collaboration should take place to establish relevant parameters for quality assurance in pediatric endoscopy.

Funding Agencies: None


The Canadian Pediatric Survey of Access to Gastroenterology (PSAGE) 2015—Preliminary Report
J. Yap,1 P. Sathya,2 A. Noble,3 M. Boland,4 H. Brill,5 and W. El-Matary6

1University of Alberta, Edmonton, AB, Canada
2Memorial University, St John, NF, Canada
3Dalhousie University, Halifax, NS, Canada
4University of Ottawa, Ottawa, ON, Canada
5McMaster University, Hamilton, ON, Canada
6University of Manitoba, Winnipeg, MB, Canada

Background. The Pediatric Survey of Access to GastroEnterology (PSAGE) program was designed to provide cross-sectional data of current wait times related to non-urgent indications for digestive health care in children. Twenty-one non-urgent pediatric-specific gastrointestinal indications were developed through consensus by the PSAGE steering committee including proposed benchmark wait-times.

Aims. To described the physican demographics, practices and current wait times for Canadian children accessing non-urgent digestive health care.

Methods. Canadian pediatric and adult gastroenterologists known to care for children were approached to complete an online or hard copy survey between the 13th to the 24th of April 2015. The survey consisted of a one-time physician demographics questionnaire. Participating physicians were requested to record seven data points per patient for 5 consecutive new patient consults and 5 consecutive new endoscopic procedures. These included the number of days off school and the second character of the patient’s postal code.

Results. Thirty one percent () of physicians approached participated in the survey, entering data for 241 patients. Ninety six percent of surveyed physicians practiced within a teaching hospital; 83.3% were in full-time practice. Surprisingly, 20% of physicians were limiting new patient referrals. The majority of physicians (86.6%) were “not at all” to “somewhat satisfied” with current wait times. Adolescence (46.1%) was the most common patient age group, followed by the 6–10 year olds (25.7%). The patient referrals from family doctors were broadly distributed between provinces (25−72.7%) with the majority of referrals from urban centres. The most common indications for referral were celiac disease confirmation (43%), chronic abdominal pain (36%) and sub-acute rectal bleeding (21%). Less than 10% of patients had indicated greater than 5 days of missed school. All physicians indicated the presence of an anaesthetist during endoscopic procedures.

Conclusions. The new PSAGE survey has provided useful insight into the current demographics, practices and perceived wait times concern of pediatric gastroenterologists looking after children with gastrointestinal problems. More detailed analysis of wait times relative to the proposed benchmarks is required.

Funding Agencies: CAG


The 5-Ht4 Receptor Agonist Prucalopride Induced a Variety of Human Colonic Pressure Waves Assessed by High Resolution Manometry
W. Chen, J. Chen, and J. Huizinga

McMaster University, Hamilton, ON, Canada

Background. The study of human colon motility is a challenge since the colon may be inactive for long periods of time, however, to understand motor dysfunction in chronic constipation, manometry is essential. It is therefore important to evaluate the use of various stimuli for assessment of colon motor function.

Aims. The aim of this study was to investigate the effect of oral prucalopride on human colon intraluminal pressure activity, within the constraints of a 6 hour colon function test.

Methods. We examined the colonic intraluminal pressure patterns in 23 subjects using high resolution colonic manometry (HRCM) with 36 solid-state sensors, 1 cm apart. 13 of the subjects had chronic constipation, 10 were either healthy or had non-constipation IBS. Two mg of prucalopride, a 5-HT4 agonist, was given orally to all subjects.

Results. Oral prucalopride is rapidly absorbed and maximum bioavailability is reached 1–3 hours after administration (Winter et al. JPGN 57(2013)197). Prucalopride elicited an excitatory effect in 18/23 subjects, 11/13 of the constipated patients and 7/10 of the non-constipated. The induced activities were simultaneous pressure waves, propagating pressure waves including High Amplitude Propagating Pressure Waves (HAPWs, also called HAPCs) and isolated pressure transients. An increase in occurrence was seen in 3/23 subjects for HAPWs, 13/23 subjects for isolated pressure transients and simultaneous pressure waves, and 2/23 for propagating pressure waves. An increase in amplitude was seen in 9/23 subjects for simultaneous pressure waves and 3/23 subjects for propagating pressure waves. A biphasic effect was observed where prucalopride elicited an initial excitation within 5 minutes of administration and/or a second phase of excitation after 10–60 minutes of administration. We propose that the first phase is elicited by a gastrocolonic reflex involving enterochromaffin cells in the stomach and vagal colonic excitation. We propose that the second phase is elicited by the active drug in systemic circulation acting on the enteric nervous system.

Conclusions. In conclusion, prucalopride has a prokinetic acute effect, likely through a 5-HT4 transduction-pathway. The colon function test shows the sensitivity of the patient to the activation of the 5-HT4 pathway to elicit simultaneous pressure waves and isolated pressure transients. To evaluate a potential defect or absence of the 5HT4 pathway, higher doses of prucalopride may have to be given.

Funding Agencies: National Natural Science Foundation of China


Bacterial Presence on Flexible Endoscopes versus Time Since Disinfection
K. Mallette,1 P. Pieroni,2 D. Thomson,2 L. Struthers,2 and S. Dhalla2

1University of Manitoba, Winnipeg, MB, Canada
2Brandon Regional Health Centre, Brandon, MB, Canada

Background. Flexible endoscopes are extremely valuable in the diagnosis and management of gastrointestinal disease. Recently, there have been documented cases of transmission of antibiotic resistant microbes in the United States via endoscopy. Previous guidelines suggested that endoscopes be reprocessed prior to use; however, a study conducted at our institution demonstrated that endoscopes could be stored up to 7 days prior to use when maintained in a ventilated, dust free cabinet.

Aims. The objective of this study was to validate the prior study conducted at our institution, correlating the length of time that an endoscope was hung in a cabinet and how much bacteria was cultured prior to use.

Methods. Prospectively, we cultured specimens from 19 gastroscopes, 24 colonoscopes and 5 side viewing duodenoscopes during the period of 2011 to 2015. Two scopes were evaluated weekly on a rotational basis, with a total of 327 evaluations. However, only 164 results had complete data denoting date of cleansing, number of days stored and culture Results. Endoscope hang time was calculated by counting the number of days between disinfection and microbiological evaluation.

Results. All positive culture results were within the acceptable range for potable water (less than 200 cfu/mL). The highest count was 80 cfu/mL, which was cultured from colonoscope CHD4 after a hang time of 1 day. The highest count for ERCP scopes was 10 cfu at both 2 and 7 days, and the highest count for gastroscopes was 50 cfu/mL after 1 day. The majority of cultures, irrespective of hang time, were negative for bacterial growth (Figure 7). There was no significant difference in the number of bacteria cultured after 1 day compared to 7 days when all scopes were combined. There was no statistical difference observed in bacterial cultures after 1 day compared to subsequent days for gastroscopes, colonoscopes, or ERCP scopes.

Conclusions. There does not appear to be a correlation between the length of hang time and the results of the bacterial culture. However, in this study, the sample sizes were small and thus a difference may not have been detected. This data further supports the previous study conducted at our hospital. Endoscopes do not need to be reprocessed prior to use if they are decontaminated according to the manufacturer’s instructions after use, hung in a ventilated, dust-free cabinet prior to use, and reused within a period of 7 days.

Funding Agencies: None


Carbon Dioxide versus Room Air Insufflation in Colonoscopy: A Retrospective Study
K. Mallette,1 D. Thomson,2 and S. Dhalla2

1University of Manitoba, Winnipeg, MB, Canada
2Brandon Regional Health Centre, Brandon, MB, Canada

Background. An increased awareness for colon cancer screening has resulted in an increased demand for colonoscopy in North America. Carbon dioxide (CO2) or room air (RA) can be utilized for insufflation, however, RA is the default method employed by every manufacturer of endoscopes. The use of CO2 as a method of insufflation is being increasingly employed due to its ability for rapid absorption and respiratory expiration; which leads to less abdominal distention and hopefully less discomfort and pain.

Aims. The purpose of this study was to compare CO2 to RA insufflation during colonoscopy to assess for differences in sedation utilized, oxygen required, endoscopy time, recovery time and post-procedure pain.

Methods. A total of 445 consecutive patients who underwent only colonoscopy between November 28, 2014 and May 29, 2015 utilizing CO2 or RA at our regional referral site were selected. Patients were excluded due to previous large bowel resection, scheduled endoscopic mucosal resection, and incomplete colonoscopy. Patient demographic data included age, gender and BMI. Intra-procedure measurements tabulated included amount of sedation, O2 requirement and endoscopy time. Post-procedure, recovery time (defined as time from admission to the post-procedure area to the time of discharge) and presence of pain, were noted. Two sample unpaired -tests were utilized to analyze the aforementioned data and pain was further evaluated for differences between genders, and at three 15 minute time intervals.

Results. Patient demographics, O2 requirement and endoscopy time were not statistically different at the 95% confidence level. The amount of midazolam utilized was significantly higher in the CO2 group (3.17 mg versus 3.67 mg, ), while amount of fentanyl was significantly higher in the RA group (80.2 mcg versus 64.5 mcg, ); largely due to physician preference. Patient recovery time was not different between the two groups (59 mins (RA) versus 61 mins (CO2), -value 0.225) and was not statistically longer for more invasive procedures compared to biopsy (RA: -value 0.571) (CO2: -value 0.138). Patients in the CO2 group experienced less pain overall (15% versus 36%), at admission to recovery (14% versus 31%), and 15 minutes after admission to recovery (4% versus 13%). Both males and females saw decreased levels of pain overall (males: 17% versus 40%) (females: 13% versus 32%) and at admission to recovery (males: 16% versus 34%) (females: 12% versus 29%). Males also experienced less pain at 15 minutes of recovery (4% versus 17%) (Figure 8).

Conclusions. Insufflation utilizing CO2 did not significantly alter O2 requirement, endoscopy time or recovery time. However, CO2 significantly reduced the percentage of patients experiencing post-colonoscopy pain and decreased the frequency of pain in both male and female patients.

Funding Agencies: None


Endoscopy Utilization and Outcome for the GI Nurse Navigator Pathway: A Quality Improvement Project for Chronic Dyspepsia, Heartburn & Irritable Bowel Syndrome
K. Milne,1 B. Kathol,2 M. Swain,3 C. Johnstone,2 J. Kwan,4 W. Schoombee,4 C. Andrews,1 and K. Novak1

1University of Calgary, Calgary, AB, Canada
2Alberta Health Services, Calgary Zone, Calgary, AB, Canada
3Univ Calgary, Calgary, AB, Canada
4Calgary Foothills Primary Care Network, Calgary, AB, Canada

Background. The Gastrointestinal Nurse Navigator (NN) pathway is a collaborative strategy developed by the Division of Gastroenterology (GI) and the Calgary Foothills Primary Care Network (PCN), aimed to provide comprehensive care to patients through nurse-lead medical education as well as nutrition and behaviour health support for patients with non-urgent GI concerns. Since 2012, referrals for dyspepsia, gastroesophageal reflux disease (GERD) and irritable bowel syndrome (IBS) were selected, with nurse-lead telephone assessment, direct referral to endoscopy for red flags, and group multidisciplinary medical education session with GI consultation.

Aims. To evaluate endoscopy usage and diagnostic outcome in the NN pathway.

Methods. This is an ethics approved, single center, prospective observational study, including 443 patients from July 2012 to December 2014. Demographics, endoscopic indication and diagnostic outcome were evaluated.

Results. Of the 443 patients, 198 had dyspepsia, 211 GERD, and 34 had IBS. 251 (56%) Underwent endoscopy, with 7 patients (1.6%) having simultaneous referrals to other gastroenterologists and endoscopy performed privately outside of the pathway. Gastroscopy was the most commonly performed procedure (193/251, 77%), followed by colonoscopy (48/251, 19%) the remainder were sigmoidoscopy (10/251 4%). More females than males (48% versus 45%) underwent endoscopy, and the average age of patients who underwent endoscopy was higher at 48 versus 46 yrs (). Of those patients who underwent endoscopy, 15 studies (5.6%) revealed diagnoses changing medical management (H. Pylori, adenomas, inflammatory bowel disease (IBD) and Barrett’s esophagus). Those most likely to have these diagnoses had an average age of 52. There were no cancers diagnosed and IBD was mild.

Conclusions. The NN pathway is safe, with low morbidity given minimal significant pathology identified with no malignancies. The identification of patients for entry into this pathway is appropriate and furthermore, many may not have required endoscopy at all. Future strategies should aim at conservative therapy, focused on lifestyle and medical management within primary care.

Funding Agencies: None


Effect of Feedback of Conoloscopy Patient Comfort Scores on Endoscopist Behaviour
B. Chan,1 A. Hussey,2 N. Rubinger,1 and L. Hookey1

1Queen’s University, Kingston, ON, Canada
2Hotel Dieu Hospital, Kingston, ON, Canada

Background. There have been many studies evaluating the type, amount, and risks of sedation during colonoscopy, as well as patient comfort during the procedure. For many experts, the optimal colonoscopy is a complete exam with the least amount of sedation while maintaining patient comfort.

Aims. Using a Canadian, validated patient comfort scoring system, this study aimed to determine whether the introduction of a patient comfort score would affect the amount of sedation used by endoscopists.

Methods. The Nurse Assisted Patient Comfort Score (NAPCOMS) was used to assess patient comfort during colonoscopy and was added as routine procedural documentation. The study was conducted over two phases; this report focuses on Phase Two. Phase One consisted of endoscopist blinded and endoscopist aware NAPCOMS collection. In Phase Two, data was collected over a five month period and scores fed back to individual endoscopists on a monthly basis.

Results. We previously presented Phase One data, which showed no significant differences in sedative use or NAPCOMS.

In Phase Two, we documented 13 endoscopists and 932 cases over the course of five months. There were nine gastroenterologists (773 cases) and four general surgeons (159 cases). Analysis of group data between individual months showed significant differences in midazolam () and fentanyl use (). The amount of fentanyl declined, while there was no trend in midazolam use. Total NAPCOMS did not show any significant differences, but subgroup analysis showed a decline in pain score (). When month one was compared to all other months in the study, there were no significant differences.

Data compared between gastroenterologists (GI) and general surgeons (GS) showed no significant differences in NAPCOMS. Gastroenterologists used significantly less midazolam () but more fentanyl (). In addition, GI utilized more position changes () but there was no difference in procedure duration or use of abdominal pressure.

Conclusions. Phase One showed no significant differences in the primary endpoint. One possibility was that monitoring was unobtrusive and not recognized by the endoscopist. Phase Two was designed to test this theory and showed a significant increase in fentanyl use, with a decline in pain score but not total NAPCOMS. The correlation of increased fentanyl and decreased pain score shows that a quality control measure can affect physician activity.

Additionally, our data showed a difference between GI and GS, with GI using significantly more position changes. While this did not change NAPCOMS scores, there were differences in the amount of sedation used. These changes do not allow us to draw inferences between the two specialties but provide interesting insight into the preferences of endoscopists at this centre.

Funding Agencies: None


Correlation of the St. Paul’s Endoscopy Comfort Scale with Post-Procedure Pain Recollection following upper Endoscopy
J. Yonge,1 N. Harris,1 C. Galorport,1 O. Takach,1 I. Tavakoli,2 M. Suzuki,1 J. Amar,1 S. Whittaker,1 H. Ko,1 G. Rosenfeld,1 A. Ramji,1 B. Bressler,1 E. Lam,1 J. Telford,1 and R. Enns1

1St. Paul’s Hospital, Vancouver, BC, Canada
2University of British Columbia, Vancouver, BC, Canada

Background. Patient comfort during endoscopy is an important measure of endoscopic quality and is associated with improved patient satisfaction and compliance with future procedures. We modified the original St. Paul’s Endoscopy Comfort Score (SPECS) for colonoscopy to validate it for use in outpatients undergoing upper endoscopy.

Aims. To determine whether there is any correlation between SPECS for upper endoscopy and self-reported patient comfort and satisfaction.

Methods. 300 outpatients undergoing upper endoscopy at St. Paul’s Hospital were prospectively enrolled between May and August 2015. Inclusion criteria: Age ≥ 19 years and outpatient upper endoscopy. Exclusion criteria: unable to speak and understand English, undergoing both colonoscopy and upper endoscopy at the same time, and unwilling or unable to complete the questionnaire. The SPECS and Gloucester Score (GS) were completed independently by the physician, nurse, and research assistant (RA). The RA also completed the Non-Verbal Pain Assessment Tool (NPAT) and Nurse Assessed Patient Comfort Score (NAPCOMS). Patient demographics were collected. Patients completed a patient satisfaction questionnaire and Visual Analogue Scale (VAS) that assessed the patient’s pain and satisfaction. Cohen’s kappa coefficient and Fleiss’ kappa was used to assess the inter-rater reliability and agreement. This study was approved by the IRB.

Results. Mean age was 56 (range 19 to 88) and 160 subjects were male. VAS was used as the gold standard and compared to the other scales. VAS has slight agreement to all scales (no scale was significantly better when compared to VAS). SPECS and NAPCOMS had the highest κ scores. Patients were asked to rate their pain intensity (none, mild, moderate, and severe) during procedure. All scales showed slight agreement. SPECS and GS had the highest κ scores.

Conclusions. There was no significant difference in the correlation between any of the scales and the patients self-reported pain as a linear scale (VAS) or when organized categorically. SPECS was previously used as a tool to measure patient comfort in colonoscopy and could use further adjustment to be meaningfully applied to upper endoscopy.

Funding Agencies: None


Adequacy of Documentation of Follow-Up Plans for Patients Undergoing Inpatient Colonoscopy
C. Parker,1 M. Brahmania,1 M. Kowgier,1 S. Sharma,1 T. Alomani,1 G. Malhi,1 A. Gulamhusein,1 N. Bollegala,1 M. Cino,2 A. Weizman,3 M. Bernstein,4 and E. Irvine5

1University of Toronto, Toronto, ON, Canada
2Toronto Western Hospital, Toronto, ON, Canada
3Mount Sinai Hospital, Toronto, ON, Canada
4Sunnybrook Health Sciences Centre, Toronto, ON, Canada
5St. Michael’s Hospital, Toronto, ON, Canada

Background. The transition of care from the inpatient to outpatient setting can be fragmented and may contribute to poor patient outcomes. Lack of appropriate follow-up for patients undergoing inpatient colonoscopy who are found to have colonic polyps may put the patient at risk for developing interval colon cancer. This may be related to inadequate documentation upon hospital discharge.

Aims. To assess the adequacy of documentation for appropriate follow-up among those with colonic polyps found during inpatient colonoscopy.

Methods. A retrospective chart review was performed on patients who had colonic polyps found during inpatient colonoscopy during a one year period at St. Michael’s Hospital, Toronto, Canada. Discharge summaries were reviewed for adequate documentation of follow-up plans including the need for follow-up, time interval for follow-up, if required, and the contact information of the follow-up provider. Descriptive statistics were used to calculate the proportion of patients who had adequate documentation of follow-up plans upon discharge.

Results. 45 patients were included in the final analysis. All patients had a completed discharge summary. The need for follow-up was found in 46.7%, and the interval for follow-up in 24.4% of the discharge summaries. Contact information for the follow-up consultant was present in 17.8% summaries. 31 patients had one or more tubular adenoma (with or without high grade dysplasia) or tubulovillous adenoma. Of these 31 patients, 48.4% had the need for follow-up in their discharge summary, 22.6% had the interval of follow-up and 38.7% had the contact information of the follow-up provider. 27% patients had polyps that were not removed or retrieved at colonoscopy. Of these 12 patients, 50% had the need for follow-up in their discharge summary, 25% had the interval of follow-up recommended and 25% had the name of the consultant they were to follow-up with.

Conclusions. Adequate documentation of the need for follow-up was lacking in most discharge summaries of inpatients found to have colonic polyps during colonoscopy. The problem was magnified further in patients with adenomas or with polyps that were either not removed or not retrieved. This report highlights the importance of developing new initiatives to improve communication among healthcare providers at the time of discharge to ensure appropriate follow-up after inpatient colonoscopy.

Funding Agencies: None


Endoscopic Evaluation of Graft-versus-Host Disease: Retrospective Review from a Tertiary Centre
S. Ip, V. Marquez, D. Schaeffer, and F. Donnellan

University of British Columbia, Vancouver, BC, Canada

Background. Graft-versus-host disease (GVHD) is a complication of hematopoietic stem cell transplantation (HSCT) that frequently affects the gastrointestinal (GI) tract. The diagnosis requires pathologic confirmation from endoscopic biopsies; however, the ideal location of these biopsies has not been clearly established.

Aims. To determine the best sites for obtaining biopsies in evaluating GI GVHD.

Methods. All cases of biopsy-proven GI GVHD (GVHD+) were obtained from a pathology database over a two-year period at a tertiary centre (). Demographic, clinical, and endoscopic data were extracted. For comparison, a randomized sample of GVHD negative cases (GVHD−) was obtained (). Sensitivities for the diagnosis of GVHD at different sites of both the upper GI tract and colon were determined.

Results. Diarrhea was the most common symptom in both the GVHD+ and GVHD− groups. In the GVHD− group, they were commonly investigated with an esophagastrodudenoscopy (EGD) (60% versus 22% in the GVHD+ group, ) while a colonoscopy (CLN) was commonly performed in the GVHD+ group (33% versus 12%, ). Non-specific erythema was more often found in the GVHD+ group (). Among the GVHD+ patients, for EGDs, the sensitivity was highest for duodenal biopsies at 89%. There was only one case in which GVHD was not detected by duodenal biopsy but found on a gastric biopsy. For FS and CLN, the sensitivities among all sites were similar (85% agreement, kappa 0.58, ). There were no cases in which GVHD was diagnosed in the right-side of the colon without a positive biopsy in the left-side of the colon. The grade of GVHD appeared to have no effect on sensitivities.

Conclusions. In this cohort of GI GVHD patients, duodenum biopsies seem to produce the highest yield for diagnosing GVHD with a sensitivity of 89% when compared to other sites of the upper GI tract. Sensitivities were similar among all sites on lower endoscopies, suggesting that a FS is sufficient for diagnosing GVHD in suspected patients with diarrhea. As shown in this cohort, CLNs may be overly utilized and unnecessary in the investigation for GVHD.

Funding Agencies: None


Single Center Experience in the Use of Device Assisted Enteroscopy: A Retrospective Study
A. Benmassaoud, M. Sasson, C. Soulellis, and T. Bessissow

McGill University Health Center, Montreal, QC, Canada

Background. Over the last 15 years, the endoscopic evaluation of the small bowel has gone through a major revolution with the development of device-assisted enteroscopy (DAE), including single and double balloon enteroscopy. Since then, it has been used for diagnostic and therapeutic purposes in various clinical situations such as obscure gastrointestinal bleeding (OGIB), Crohn’s disease (CD) and small bowel tumors.

Aims. The main objective of this study was to evaluate the diagnostic and therapeutic yield of DAE in the evaluation and treatment of small bowel diseases using our database.

Methods. This was a single center retrospective cohort study from the McGill University Health Center. Adult patients who had a DAE between January 2010 and July 2015 were included. Patients were identified using a prospectively maintained database. Patients were excluded if data related to the enteroscopy was missing. Electronic and paper medical records were extensively reviewed. Demographic and clinical data was collected. A descriptive analysis of the recorded data was performed.

Results. 246 device-assisted enteroscopies were available for analysis. In our cohort, patients’ median age was 64 years old (IQR 47–75), and were inpatients in 9% of cases. The three most common causes of referral were OGIB in 65%, CD in 9% and gastrointestinal malignancy or polyp in 8% of cases. DAE was anterograde in 92% and retrograde in 8% of cases. 58% of patients had a previous gastroscopy or colonoscopy, 17% had prior video capsule evaluation, and 17% had prior DAE. About 49% of patients had a CT scan before DAE and 40% had no previous imaging done. Sedation consisted mainly of a combination of Midazolam and Fentanyl in 96% of cases with average doses of 3.3 mg ± 1.6 mg and 93.2 mcg ± 39.1 mcg respectively. General anesthesia was required in 6 cases. Approximately 54% of entroscopies had positive findings. Amongst them, the three most common findings were an arteriovenous malformations, an ulcer or erosion and the presence of polyps or stricture in 43%, 26%, and 9% of cases respectively. A therapeutic intervention was deemed necessary in 34% of all cases, or in 62% of cases with a positive finding.

When compared to all comers, patients with a pre-endoscopic diagnosis of OGIB trended towards being more likely to have a positive finding (65% versus 54%, OR = 1.55, ) and were more likely to have treatment applied (52% versus 34%, OR = 2.13, ).

Conclusions. Our study showed that the most common indication for the use of DAE was OGIB. Patients with a pre-endoscopic diagnosis of OGIB trended towards being more likely to have a positive finding and have treatment applied. Further studies are underway to validate these findings.

Funding Agencies: None


Quality Improvement of Endoscopic Procedure Dictation Reports at St. Paul’s Hospital: A Comparison of Transcriptions From 2008 and 2014
J. Yonge, N. Harris, C. Galorport, M. Suzuki, E. Nap-Hill, J. Amar, B. Bressler, E. Lam, J. Telford, and R. Enns

St. Paul’s Hospital, Vancouver, BC, Canada

Background. Following an esophagogastroduodenoscopy (EGD), the physician dictates their findings which are later transcribed and kept in patient’s medical records. A dictation template for upper endoscopy was first introduced to St. Paul’s Hospital approximately 3 years ago. The template was created based on the recommendations made by the ASGE and was revised by the endoscopists at St. Paul’s. To evaluate the completeness of dictation reports, key points have been outlined based on the available literature and the current EGD reporting guidelines available at St. Paul’s.

Aims. The purpose of this study is to assess and compare the quality and completeness of EGD procedure reports from 2008 and 2014 for physicians currently working at St. Paul’s Hospital to determine if key quality elements of documentation were more consistently included following institution of a dictation template.

Methods. This was a retrospective chart review of dictation reports completed by gastroenterologists at St. Paul’s Hospital from 2008 and 2014. 150 charts were reviewed for each doctor in each year. Data was collected from a comprehensive EMR system that included demographics, patient history, procedure report details (appropriate quality indicators as outlined by ASGE), and length of procedure. This study was approved by the IRB.

Results. The overall completeness for all gastroenterologists improved from 71.53% in 2008 to 76.82% in 2014 (). Most variables remained consistent or increased; however, reporting of comorbidities, medications, complications, and patient comfort remained low at both time periods.

Conclusions. The use of the dictation template has improved documentation of quality parameters from 2008 to 2014. The variables that are frequently missed from dictation reports have been identified and educational maneuvers as well as other adjustments to include these variables in future procedure reports can be targeted at these items.

Funding Agencies: None


Inter-Observer Reliability of the St. Paul’s Endoscopy Comfort Scale (SPECS) for upper Endoscopy
N. Harris,1 J. Yonge,1 C. Galorport,1 I. Tavakoli,2 O. Takach,1 M. Suzuki,1 J. Amar,1 S. Whittaker,1 H. Ko,1 G. Rosenfeld,1 A. Ramji,1 B. Bressler,1 E. Lam,1 J. Telford,1 and R. Enns1

1St. Paul’s Hospital, Vancouver, BC, Canada
2University of British Columbia, Vancouver, BC, Canada

Background. Patient comfort is a key quality indicator for gastroscopy and is associated with enhanced patient satisfaction and improved compliance with future procedures. The St. Paul’s Endoscopy Comfort Scale (SPECS) has been previously validated to evaluate patient comfort during colonoscopy; however, it has not yet been assessed for use during upper endoscopy. This scale has three categories namely Vocalization, Positioning/Body Language and Patient Anxiety/Emotion. Each category is scored from 0–3 based on the frequency and severity of specific indicators, yeilding a total score ranging from 0–9. Another tool used to assess patient comfort during endoscopy is the Gloucester scale (GS), which is a five point global rating system.

Aims. To modify SPECS for Colonoscopy for use during upper endoscopy, to compare the inter-observer reliability of SPECS for Upper Endoscopy to the GS, and to determine if the use of sedation during upper endoscopy affects the SPECS score.

Methods. 300 outpatients undergoing upper endoscopy at St. Paul’s Hospital were enrolled (May 2015–August 2015). Inclusion criteria: outpatients ≥19 years scheduled for upper endoscopy. Exclusion criteria: individuals undergoing additional procedures during the same visit. SPECS and GS were completed independently by the physician, nurse and research assistant. To avoid behavioral bias, patients were debriefed of study aims post-procedure. Kappa statistical claculations were performed and patient demographics and sedation were documented. The study was approved by the UBC Ethics Board.

Results. Mean age of participants was 56.7 years and 46.7% were female. Sedation was used for 89.0% of cases; the mean Midazolam dose was 3.3 mg (SD 1.6) and the mean Fentanyl dose was 51.4 mcg (SD 29.7). The mean SPECS score, calculated using the average of the three observers, was similar for non-sedated and for sedated patients: 1.3 (SD 1.3) and 1.5 (SD 1.6), respectively.

Conclusions. SPECS for Upper Endoscopy demonstrated slightly higher inter-observer reliability than the GS, but the results were not statistically significant. Although not statistically significant, the category demonstrating the weakest inter-rater reliability was Positioning/Body Language and this may be due to the subjectivity in assessing the gag reflex. Sedation had no notable effect on the SPECS score.

Funding Agencies: None


Incidence of Venous Thromboembolism in Gastrointestinal Bleeding
C. Sheasgreen, M. Almakadi, and G. Leontiadis

McMaster University, Hamilton, ON, Canada

Background. Venous thromboembolism (VTE) is a common complication of hospital admission. For patients admitted with gastrointestinal bleeding (GIB), confusion can arise as to whether it is in the patient’s best interest to use pharmacological prophylaxis against VTE.

Aims. This was a pilot study to assess the use of VTE prophylaxis and the incidence of VTE in patients admitted to hospital with GIB.

Methods. Hospital charts of adult patients admitted for GIB from 2009–2011 at one centre in Ontario were reviewed. Charts were pulled in aliquots of 50 sequentially admitted patients. Those with previously diagnosed VTE, risk factors for VTE (malignancy, active inflammatory bowel disease, hypercoaguable state, thrombophilia, or myeloproliferative disorder), or hospital stay less than 24 hours were excluded. Patients were classified as having “confirmed" GI bleeding or “probable" GI bleeding based on reported history and physical exam. Criteria for being classified as “confirmed" included having hematochezia, melena, hematemesis, or coffee ground emesis observed by a physician or documented GIB on endoscopy at time of admission. Hospital records were reviewed for the presence of mechanical foci for thrombus formation (e.g., central venous catheters or inferior vena cava filters), smoking and alcohol use, admission to hospital within the previous 6 months, use of pharmacological prophylaxis for VTE while in hospital, death, and incidence of VTE within 6 months from index admission.

Results. 250 patient charts were reviewed. After exclusions, 125 patients were included in the analysis. 69 patients were “confirmed" GIB and 56 were “probable." 7 (10.1%) of the confirmed cases were given VTE prophylaxis whereas 11 (19.6%) of the probable cases received the same. There were 2 VTE events; a pulmonary embolism in “Patient A" and a right internal jugular vein thrombus in “Patient B," both of whom were confirmed GIB patients. Patient A had a history of cigarette and alcohol use and was not given pharmacological VTE prophylaxis. Patient B had a right central venous catheter and was given pharmacological VTE prophylaxis. 4 patients died, 2 of whom had been given VTE prophylaxis. Niether of the 2 patients with VTE died.

Conclusions. These data suggest that patients in whom the diagnosis of GIB is clinically obvious are less likely to receive pharmacological VTE prophylaxis and that this may translate into an increased risk for VTE events. VTE does not appear to increase the occurrence of death in GIB. A larger review encompassing more events will help deliniate these relationships further.

Funding Agencies: None


Endoscopic Ultrasound in Nova Scotia, a Quality Assurance Study
A. Alghamdi

Dalhousie university, Halifax, NS, Canada

Background. Endoscopic ultrasound (EUS) is technique that utilizes endoscopic technology with an ultrasound transducer at the tip to allow visualization of submucosal lesions, and structures surrounding the gastrointestinal tract. Newer technology has allowed real-time fine needle aspiration (FNA) to be performed under EUS guidance. It has proven to be a highly sensitive tool for diagnosing lesions in and adjacent to the gastrointestinal tract.

Aims. Since the single most important function of EUS is in its ability to obtain tissue via FNA, our primary outcome measure will be yield of FNA for the various indications. Secondary outcome measures will include the referral base, indications, waiting time and complications of EUS in Nova Scotia. This quality assurance study will help in improving the EUS program in our province.

Methods. It is an observational, retrospective cohort study of all the men and women who had undergone EUS in Nova Scotia, in the CDHA, throughout the calendar year of 2013. Subjects of this research consist of 114 patients. Patient files will be analyzed to determine the reason for referral to EUS, the complications if any, and the waiting time for an EUS appointment in the out patients sittings. Results of EUS with or without FNA will be charted as well as the diagnosis obtained via cytological analysis.

Results. The most common reasons for referral to EUS were for evaluation of pancreatic mass/cyst (44 patients, 39%), and assessment of sub-mucosal lesions (26 patients, 22.8%). Other indications were lymph node FNA (mostly mediastinal), Dilated CBD, pancreatic cancer screening, chronic unexplained pancreatitis. Rectal EUS were performed in 4 patients; in which 3 of them referred for fecal incontinence and 1 had para-anal mass for FNA.

A total of 49 FNA’s were performed by EUS for different indications; most of them were from a pancreatic mass/cyst, Lymph node and submucosal lesions; 69, 10 and 8 percent respectively. 82% of total FNAs results were conclusive, either positive or negative; among the FNA obtained from a pancreatic mass 85% were conclusive, while FNAs from Lymph node and submucosal lesions were conclusive in 60 and 50 percent respectively. The most common abnormal FNA results from the pancreas were pancreatic adenocarcinoma (46%) and mucinous neoplasia (30.7%). Other results included pancreatic lymphoma, metastatic malignancy from lymph node FNA, lung cancer and anal cancer.

4 patients developed complications post EUS, 2 (1.7%) had pancreatitis and 2 (1.7%) had mild bleeding.

Conclusions. EUS can be used for variety of indications, most commonly to further characterize a pancreatic lesion, with the ability of obtaining a tissue diagnosis through FNA with good diagnostic yield that guided patients management. It is a minimally invasive procedure with low complication rate.

Funding Agencies: None


A Retrospective Analysis of the Long-Term Outcomes of Patients with Hiatus Hernias, Cameron Erosions, and Iron Deficiency Anemia
E. Nap-Hill, C. Galorport, and R. Enns

St. Paul’s Hospital, Vancouver, BC, Canada

Background. Mechanical trauma of hiatus hernias (HH) can cause linear gastric erosions on the crest of the mucosal folds near the diaphragm called Cameron Erosions (CE). CE are an uncommon source of upper GI bleeding, and are usually identified by an upper endoscopy. The erosions bleed intermittently, become less prominent, then reappear, at times making the diagnosis difficult. There is no standard treatment regimen to remedy these lesions however most are managed with acid suppression. Long-term follow-up is uncommon in the literature.

Aims. This study aims to assess the long-term outcomes of patients with CE in regards to therapy required: oral/parental Fe, acid suppression, blood transfusion and/or surgery, and subsequently to determine risk factors for failure of conservative therapy.

Methods. A retrospective chart review of patients with HH and CE treated for iron deficiency anemia between January 2005 and June 2015 was performed. Data collected includes demographics, endoscopy dates, indication, findings, history of blood and iron transfusions, co-morbidities, history of GI surgery, medications, blood work, surgery, and hospitalizations. All patients had undergone upper and lower endoscopic examinations. Additionally, all patients referred for surgery underwent a capsule endoscopy to exclude small bowel etiology of bleeding.

Results. Preliminary data on 21 of 60 patients with CE is presented. 21 patients were identified with both CE, and iron deficiency anemia; followed for an average of 27 months. 67% of patients had their anemia successfully treated, while 33% had persisting anemia after five years. 43% of patients were referred with known CE as the likely cause of their anemia, while the other 57% were referred for obscure or occult bleeding, and later found to have CE. 34% of patients who were successfully treated, were treated conservatively with proton pump inhibitors (PPI), iron supplements, and occasional blood transfusions (33% required a mean of 2.3 units of blood over the course of their follow up). 33% did not respond to conservative therapy, and were successfully treated with laparoscopic repair of their HH.

Conclusions. Conservative treatment of PPIs, Fe supplements, and occasional blood transfusions has shown to be effective in treating most patients, however, many require ongoing monitoring and correction of anemia. In those that did not respond to conservative treatment, surgical correction of HH resolved anemia in all patients who underwent repair. Further study will determine risk factors for requiring more aggressive (i.e., surgical) therapy.

Funding Agencies: None


Tidying up a Waitlist: A Quality Assurance Project
M. Sheridan,1 A. Dorreen,2 and S. Williams1

1Dalhousie University, Bedford, NS, Canada
2Dalhousie University, Halifax, NS, Canada

Background. Access to Gastroenterology services is limited in Atlantic Canada, resulting in long waitlists for outpatient consultation. At our institution, the current wait time for non-urgent referrals exceeds three years. Periodic review of the referrals on such a long wait list is important to assess whether these patients still require consultation and to determine if their triage category is still appropriate. This report provides an update on a three-year waitlist validation pilot project carried out in our division.

Aims. The aim of this project is to evaluate non-urgent referrals from the 2013 calendar year to identify those referrals that no longer need to be seen or can be reassigned to a direct-to-endoscopy waitlist.

Methods. All non-urgent outpatient referrals with a date of referral in 2013 were identified. A letter was sent to the referring physician of each patient in the fall of 2014 asking if the referral was still necessary and if there had been a change in the patient’s condition. If no response was provided, this letter was sent a second time. All responses were evaluated by a physician on the triage committee. Referrals that were deemed no longer necessary were removed from the waitlist. Referrals that were still deemed to be necessary were then re-triaged and, if appropriate, redirected to a direct-to-endoscopy waitlist.

Results. 404 referrals were evaluated, 66% () of patients were women and 34% () were men. The majority of referrals were sent by family physicians (91%, ). The most common indications for referral were: reflux/dyspepsia 36% (), abdominal pain 17% (), colon cancer screening 14% (), diarrhea 12% (), celiac disease 4% (), stable IBD 3% () and constipation 3% (). The response rate was 89% (). 65% () of patients were deemed by the referring physician to still need to be seen. Of these, 29% () were felt to be appropriate for a direct to procedure referral and only 2% () were re-triaged to a semi-urgent waitlist.

Conclusions. A significant number of non-urgent referrals no longer require consultation after one year on a waitlist. Of those that do, very few need to be upgraded to a more urgent triage criteria. A significant proportion of these patients are appropriate for direct-to-procedure consultations. Periodic review of referrals with prolonged wait times can result in significant shortening of waitlists.

Funding Agencies: None


The Use of High Volume Simethicone to Improve Visualization Quality during Small Bowel Video Capsule Endoscopy: A Pilot Study
D. Segal,1 B. Yan,2 N. Chande,3 T. Ponich,2 J. Gregor,4 and M. Sey2

1Western university, London, ON, Canada
2London Health Sciences Centre, London, ON, Canada
3The University of Western Ontario, London, ON, Canada
4Los Alamos National Laboratory, London, ON, Canada

Background. Poor bowel preparation affects up to one third of capsule endoscopy studies. Simethicone has been studied although its benefit has been inconsistent, possibly due to an inadequate volume being used.

Aims. The goal of this study is to compare standard volume with high volume simethicone for small bowel preparation during capsule endoscopy.

Methods. A double blind randomized clinical trial was conducted among outpatients undergoing capsule endoscopy. Patients were randomized to either 200 mL (standard volume) or 750 mL (high volume) of simethicone (1.5 mg/mL) 30 minutes prior to capsule ingestion. All patients received 2 L of PegLyte the night before the procedure and started fasting at midnight. Visualization quality (0–3) was assessed by a previously validated scale composed of the mean of the visualized mucosa (0–3) and degree of obstruction (0–3) scores.

Results. At the time of interim analysis, 20 patients had been randomized (10 standard volume and 10 high volume). The mean (SD) age was 64.1 (17.7) and 60% were females. The most common indication was obscure occult GI bleeding (50%). Compared to standard volume, the high volume group had higher visualization quality score (2.32 versus 2.45), visualized mucosa score (2.59 versus 2.67), and degree of obstruction score (2.18 versus 2.22) although this did not reach statistical significance given the interim analysis. This trend was seen in the proximal half, distal half, and when the entire small intestine was compared. There were no adverse events in either group.

Conclusions. In this interim analysis, a strong and consistent trend was seen in favour of high volume simethicone over standard volume simethicone for improved visualization quality during capsule endoscopy.

Funding Agencies: None


Gastroenterology Curriculum in Canadian Medical Schools
T. Dang,1 C. Wong,2 and L. Bistritz1

1University of Alberta, Edmonton, AB, Canada
210240 Kingsway Ave., Edmonton, AB, Canada

Background. Gastroenterology is a diverse subspecialty that covers a wide array of topics ranging from functional abdominal pain to gastrointestinal (GI) malignancies. The pre-clinical GI curriculum is often the only formal training that medical students receive prior to becoming residents. Recently, Canadian medical schools have shifted from didactic approaches to teaching to more interactive methods. Despite this change in teaching methodology and the diversity of topics in GI, there is no national consensus or awareness on content, learning objectives, or instructional methods for the GI curriculum at other Canadian institutions; this lack of consensus and awareness results in variable background knowledge for new residents and lack of guidance for curriculum development.

Aims. (1) Elucidate gastroenterology topics taught at the pre-clinical level.

(2) Determine instructional and assessment methods used at Canadian medical schools.

Methods. A survey of GI teaching topics, teaching methods, and assessment tools was developed by two educational content experts involved in organizing the GI curriculum at the University of Alberta. This survey was piloted internally and externally to gastroenterologists involved in organizing the GI curriculum at other institutions. The final questionnaire was sent to all the GI pre-clinical curriculum coordinators at all 17 Canadian medical schools in the October 2014. After receiving the responses from the different schools, a curriculum map of the GI topics from the different institutions was constructed and remaining results regarding curriculum content, teaching methods, and assessment tools were compiled.

Results. A curriculum map of GI topics was constructed from the responses gathered from 10 of the 17 Canadian medical schools and showed a heterogenous curriculum across the country. Topics often not covered included pediatric GI diseases, surgery/trauma, food allergies/intolerances, nutritional support and obesity. The curriculum was taught primarily by gastroenterologists and surgeons. Didactic teaching and small group teaching were the most common teaching methods employed. When broken down by topics, Liver Diseases used the most diverse teaching methods including small groups, online modules, and self-directed learning and the least amount of didactic teaching. Final summative exams and interim quizzes were the primary assessment tools used for evaluation.

Conclusions. This is the first study examining the GI curriculum at a pre-clinical level. Certain topics, such pediatric GI diseases, surgery/trauma and nutrition were not as well represented in responding programs. The data from this study can be used to reform existing curriculum or as a guide in future curriculum design.

Funding Agencies: None


Initial Experience with Small Histological Cores Obtained via a New Eus-Guided Fine Needle Biopsy System
T. Ishikawa, R. Mohamed, P. Kumar, and P. Belletrutti

University of Calgary, Calgary, AB, Canada

Aims. As EUS-guided tissue acquisition techniques evolve, there is increasing interest in obtaining histological samples to improve diagnostic accuracy. Several new “core" FNA needles are in development and one such needle system was tested in our tertiary care center.

Methods. A retrospective review of consecutive patients undergoing EUS-guided tissue sampling of solid lesions using the Beacon Sharkcore© fine needle system. Three experienced endosonographers (>250 EUS cases per year) performed the procedures. Selection of needle gauge and number of passes were left to the discretion of the endoscopist but at least one pass was to be submitted in formalin for histological processing.

Results. Twenty-seven patients underwent 30 EUS-guided fine needle biopsy (FNB) procedures from June-Sept 2015. The specific lesions targeted were pancreas masses (16), lymph nodes (6), submucosal masses (7) and pancreas parenchyma (1). A 25 g needle was used in the majority of cases (66.7%). In 22 cases, both cytology and histology specimens were sent from the same tissue target. Of the 30 FNB specimens, 28 (93.3%) were adequate for histological examination and 25 (89.2%) of those were diagnostic specimens (includes 3 atypical/suspicious for adenocarcinoma). Two of the non-diagnostic FNB samples were proven to be adenocarcinoma after surgical resection. Of the 22 FNA cytology specimens, 20 (90.9%) were diagnostic (including 6 atypical/suspicious for adenocarcinoma). From our own historical data, based on 267 solid lesions in 244 patients (Jan 2013–May 2015), our FNA cytology diagnostic yield is 91.3%. Of the 22 cases with both cytology and histology sent, the majority yielded the same diagnosis (77.3%), but in 4 cases (18.2%) the FNB was diagnostic whereas the FNA was not; in 1 case (4.5%) only the FNA was diagnostic (FNB specimen was insufficient for analysis). There were no complications reported after FNB.

Conclusions. In this initial experience with a new EUS-guided FNB system we found that, even with a 25 g needle, obtaining small cores to submit for histological analysis is safe, technically feasible and usually provides an evaluable sample. In some cases only the FNB specimen yielded a diagnosis.

Funding Agencies: None


Warm Carbon-Dioxide Insufflators Fail to Deliver Target Temperatures during Colonoscopies—An Ex-Vivo Study
F. Jowhari,1 K. Robertson,2 and L. Hookey2

1Queen’s University, Kingston, ON, Canada
2Hotel Dieu Hospital, Kingston, ON, Canada

Background. With the recent shift from air to carbon dioxide (CO2) for insufflation during adult colonoscopies, one manufacturer is now marketing a warm CO2 insufflator as a potential means of reducing pain & increasing tolerability during colonoscopies. While previous studies have shown some benefit with using warm water irrigation during colonoscopies, no studies exist assessing outcomes with warm CO2 insufflation. For this to even have potential for similar effects, the warm CO2 insufflator would first need to deliver the desired temperature of gas to the distal end of the colonoscope.

Aims. To assess whether warm CO2 insufflators deliver target temperatures to the distal end of the colonoscope, in a simulated environment replicating close to core body temperatures.

Methods. Three CO2 insufflators manufactured by Olympus® (Olympus UCR), Medivators® stratus (EGA-501, with the heating option) & Bracco (EZEM-CO2 effecient®) were chosen for this study. Using two adult colonoscopes (Olympus® (CF-H180DL) & Pentax (EC-3890Li)) with their lights on, the air button was constantly depressed & temperatures were recorded at each insufflator end & distal colonoscope end for 10 min in increments of 1 min (assuming an average cecal intubation time of ~10 min). Experiments were performed both at room temperature, and with the scope immersed in a warm water bath maintained at 34°C, as well with heat on & off for Medivators stratus. Mean temperatures were then compared at 0, 5 & 10 minutes using a one-way ANOVA, with the level of significance established at .

Results. The insufflator end temperatures between the heater on & off groups were similar at time 0 min (); but a difference was detected at 5 min () & 10 min (). In spite of this, no difference was seen in the scope tip temperatures between the heater on & off groups at 0 min (), 5 min () or 10 min (). With the heater on, temperatures at the scope tip & the insufflator end were similar at 0 min (), but did show statistically significant difference at 5 min () & 10 min (). The addition of a warm water bath maintained at 34°C made no difference to scope tip temperatures at 0 min (), 5 min () & 10 min ().

Conclusions. Our data suggests that although they warm the gas at the insufflator end, a new model of heated CO2 insufflators make no difference to delivered temperatures at the distal colonoscope tip. For reasons unclear, they fail to deliver target temperatures to the distal colonoscope end both at room temperature & in a heated body simulating a real colonoscopy. One possibility is the dissipation of heat as heated CO2 passes through the length of the colonoscope umbilicus; however, further studies are needed to demonstrate this conclusively.

Funding Agencies: None


A Phase 1b/2a Randomized Open-Label Study Measuring Chyme Concentrations of Intravenously Administered Ceftriaxone in the Presence of the Oral Beta-Lactamase SYN-004
M. Rufiange,1 S. Boily,1 E. Sicard,1 M. Lefebvre,1 T. Roberts,2 H. Whalen,2 O. Coughlin,2 J. Kokai-Kun,2 and J. Sliman2

1Algorithme Pharma, Laval, QC, Canada
2Synthetic Biologics Inc., Rockville, MD, USA

Background. SYN-004 is an oral recombinant β-lactamase developed by Synthetic Biologics, USA, and intended to degradeβ-lactam antibiotics excreted into the gut, thereby mitigating their compromising effects on the gut microbiome, and preventing opportunistic hospital-acquired, bacterial infections such as C. difficile.

Aims. The purpose of this study was to evaluate the ability of SYN-004 to degrade ceftriaxone secreted into the small intestine after IV administration without affecting antibiotic pharmacokinetics (PK) in the bloodstream.

Methods. Nine otherwise healthy subjects with functioning ileostomies, aged 18–80 years, were enrolled at Algorithme Pharma Inc., a Phase I Clinical Research Organization based in Montreal, Canada. In the 1st treatment period, all subjects received an infusion of ceftriaxone and, in the 2nd period, 2 single oral doses of SYN-004 (75 or 150 mg) twice in 1 day (morning and early afternoon), with a single IV dose of 1 g ceftriaxone 30 min following the first dose of SYN-004. Subjects were confined in the clinical unit for 24 hrs, and medical monitoring lasted 1 week after the end of the second period. PK sampling was up to 8.5 hrs post-dose, both in plasma and chyme.

Results. The uniqueness of this study design resides in the special population that are patients with ileostomies. The presence of the ileostomy permits access to the chyme, the semifluid mass expelled in the ostomy bag. This constitutes a direct measurement of the degradation of ceftriaxone in the small intestine. Preliminary results show that the 75 and 150 mg single doses of SYN-004 as well as the single dose of ceftriaxone were very well tolerated. AEs were mostly mild to moderate in intensity, and all patients had recovered from their AEs at the end of their study participation. The most frequently observed AE was headache.

Conclusions. The clinical portion of the unique Phase 1b/2a study was successfully completed. The methodology used in this study of sequential sampling and analyzing the chyme should prove to be a powerful approach to allow the direct measurement of PK/PD for oral medications whose primary mechanism of action occurs in the gut. Chyme concentration analyses for ceftriaxone and the plasma PK analyses for systemic ceftriaxone are ongoing at the time of writing this abstract. SYN-004 is currently being investigated in a multicenter, placebo-controlled Phase 2b study in hospitalized patients being treated with IV ceftriaxone for lower respiratory infections. The primary outcome of the study will be prevention of C. difficile infection, with a secondary outcome of prevention of antibiotic-associated diarrhea.

Funding Agencies: None


Results of the 2014 Equity and Gender Survey of the Canadian Association of Gastroenterology
S. Perera,1 L. Bistritz,2 and M. Beaton1

1London Health Sciences Centre, Western University, London, ON, Canada
2Royal Alexandra Hospital, University of Alberta, Edmonton, AB, Canada

Background. Achieving equity remains an ongoing challenge within the field of gastroenterology.

Aims. The Equity & Gender Committee of the Canadian Association of Gastroenterology (CAG) developed a survey to identify issues pertaining to equity and gender faced by its membership, as well as to determine potential areas of action that would be of most benefit.

Methods. In 2014 a survey was emailed out to all members of the CAG.

Results. A total 111 members (52% female and 48% male) responded to the study, which was a response rate of about 10%. The majority (75%) of respondents were between 26–45 years of age, and 55% were in their first 10 years of practice. Field of specialization varied with 51% in clinical adult or pediatric gastroenterology practices, 20% were basic scientists and 20% were residents or fellows. Sixty five percent worked in academic settings, while 13% worked in the community. Commitments outside of the workplace included a spouse or partner for 81% of respondents, with 52% having children under 18 years of age. Furthermore, 45% of members surveyed stated that they cared for an aging relative or had another significant area of non-work related responsibility. 70% of the respondents surveyed stated that they were either satisfied or very satisfied with their career path. The majority (77%) did not feel that age, gender, ethnicity or marital status had affected the advancement of their careers. However, 58% felt that gender and equity challenges exist within gastroenterology. To ascertain where CAG might assist in addressing the issue of equity, members were also asked to use a five-point Likert scale to rate the importance of several areas of interest. Of those surveyed, 87% ranked work life balance as important or very important, while 70% felt physician wellbeing and leadership skills were important or very important. Other areas that were highlighted were negotiation skills and academic promotion. Of potential areas for CAG involvement, mentoring and networking were ranked as important or very important by 55% and 56% respectively of the members surveyed.

Conclusions. This survey highlights that gender and equity challenges continue to exist within gastroenterology. Furthermore, this study revealed that work life balance, physician wellbeing and negotiation skills are areas of importance to many CAG members. The results of this survey also underscored that creating mentoring and networking opportunities were two potential areas that would be of benefit to the CAG membership.

Funding Agencies: CAG


Treatmen of a Refractory Anastomotic Stricture Post-Low Anterior Resection Using a Fully Covered Enteral Stent: Case Report
S. Alshankiti, R. Sultanian, and H. Wang

University of Alberta, Edmonton, AB, Canada

Background. The complication of anastomotic stricturing post-low anterior resection (LAR) has been reported to occur in 5.8% to 20% of cases. Through-the-scope and over-the-wire dilation techniques are both effective and safe for treatment of benign colorectal anastomotic strictures, but often these strictures are refractory to conventional dilation and require surgical revision. Endoscopic stenting of LAR strictures is challenging due to the close proximity to the anal canal.

Aims. We report a case of an anastomotic stricture 4 cm proximal to the anal verge that occurred post-LAR for rectal cancer that was managed successfully by inserting a fully covered colonic stent.

Methods. Case Presentation: A 57 years old gentleman underwent a screening colonoscopy after having a positive fecal immunochemical test (FIT) in January 2014. He was found to have a rectal adenocarcinoma and was referred for surgical resection. He underwent low anterior resection with primary anastomosis and temporary diverting loop ileostomy. His post-operative period was complicated by an anastamotic leak that was managed in a conservative fashion with antibiotic treatment and a drain.

A flexible sigmoidoscopy exam 4 months after his surgery demonstrated an extremely tight stricture 4 cm proximal to the anal verge at the site of the rectal anastomosis. Over a six month period, the patient underwent a total of 9 endoscopic dilations via CRE balloon without sustained effect achieved. In addition to the balloon dilations, the stricture proved to be refractory to Triamcinolone injection and needle-knife stricture incision. The patient then underwent successful placement of Hanaro fully-covered 6 cm stent across the stricture. No immediate or delayed complications occurred.

Results. Three months later, the stent was endoscopically removed without difficulty and the anastomosis was widely patent with no significant stricture remaining. After an additional 3 months, a repeat flexible sigmoidoscopy again showed a fully patent anastomosis. Soon after this, the patient went on to have an uneventful reversal of his loop ileostomy and has been asymptomatic since.

Conclusions. This is the first published case report about using fully covered stent to successfully manage a refractory anastomotic stricture after LAR treatment of a rectal cancer. Further studies are needed to determine optimal strategies to treat distal rectal strictures with endoscopic stenting.

Funding Agencies: None


Academic Outputs and Utility of Grit Course Abstract Presentations: The UBC Experience
F. AlShatti,1 E. Yoshida,2 and V. Marquez3

1UBC, Vancouver, BC, Canada
2University of BC, Vancouver, BC, Canada
3University of British Columbia, Vancouver, BC, Canada

Background. The Gastroenterology Residents-in-Training (GRIT) Course is held in conjunction with the annual Canadian Digestive Disease Week. Its predecessor was the Post Graduate Course in Gastroenterology. The format of the GRIT Course, and its predecessor, requires Gastroenterology trainees to submit an abstract, and if accepted, they are then allowed to attend the meeting. At UBC, it is strongly recommended that trainees submit to the meeting. The academic utility of the experience to the trainee and the outcome of the submitted abstracts, however, remains unknown.

Aims. To assess the utility of the GRIT course from a UBC academic perspective by reviewing the outcomes (including publication and presentation at international meetings) of the projects submitted and to determine the value of the process to the trainees.

Methods. A list of former Gastroenterology trainees was obtained from the UBC database. A questionnaire composed of 11 multiple choice questions was sent to all former and current trainees.

Results. 88.8% of fellows responded (32 of 36). 43.75% are currently working in Academic Centers, 37.5% are in the Community, and 18.75% are still in training (that may be extra to core GI training). The abstract was a case report (33.3%), a clinical research (61.9%), or a basic science project (4.8%). 43.75% were presented at international meetings. 68.75% were published (only one was a non-peer review paper). The reasons for not publishing were: “Too busy and not enough time given during my training" (22.2%), “the abstract was appropriate for the GRIT/CDDW meeting, I did not feel that it was strong enough to be published in a journal" (44.5%) “the abstract reported work that was part of a greater research project and I was not significantly involved in the overall project" (33.3%). 21.8% received awards for their projects in GRIT either at the GRIT or at UBC trainee research days. 68.3% thought the GRIT experience was worthwhile, although one responder thought it was irrelevant.

Conclusions. We can conclude that more the two third of the projects submitted to GRIT were published, although less than half were presented internationally. The main reason for not publishing was that the abstract was not felt strong enough to be published. Most responders thought that the GRIT experience was worthwhile.

Funding Agencies: None


An Atypical Intra-Abdominal Mass in a 28 Year Old Crohns Patient on Longterm Azatioprine and Infliximab
G. Eustace1 and J. Marshall2

1McMaster University, Oakville, ON, Canada
2McMaster University Medical Centre, Hamilton, ON, Canada

Aims. This report presents the case of a young man with longstanding Crohn’s disease, presenting to the hospital with a new atypical intra-abdominal mass of unknown etiology. With his azathioprine use in mind, lymphoma or other malignancy was considered along side an inflammatory mass related to his poorly controlled IBD. The atypical features of his mass and the diagnostic work up, as well as a framework for investigating similar clinical problems in the future will be discussed.

Methods. The patient was diagnosed with terminal ileal Crohns disease in 2011 and managed on azathioprine monotherapy. Infliximab was added in early 2015 after worsening symptoms and evidence of penetrating disease on an MR enterography. He then presented to the Juravinski Hospital, a large tertiary care center in Hamilton, ON on August 12th 2015 with concerns of multiple intra-abdominal abscesses visualized on an outpatient ultrasound. CRP was grossly elevated at 197 mg/L but bowel symptoms were unremarkable. The patient also complained of ongoing lower back pain.

Results. Intravenous antibiotics were initiated. A CT scan reported an infiltrative soft tissue mass, extending off of the small bowel into the mesenteric leaves and encasing the SMA, transverse duodenum, and pancreatic head. Associated necrotic adenopathy yielded differential diagnoses of malignancy, sclerosing retractile mesenteritis and IBD-associated fibrosis. After discussions with interventional radiology, percutaneous biopsy was deemed not to be possible. An endoscopic ultrasound guided biopsy was performed, and FNA identifiefd only benign glandular cells with evidence of chronic inflammation. Serial monitoring of the patient’s mass is ongoing.

Conclusions. This case illustrates an atypical mass in a young man around which there was some diagnostic uncertainty. Although only 36 cases of thiopurine-associated hepatosplenic T cell lymphoma have been described in IBD patients 1, our patient’s young age and gender raised this concern. More commonly, treatment of IBD with azathioprine carries a four-fold increase risk of lymphoma based on a 2005 review by Kandiel et al. 2Finally, the diagnosis of sclerosing mesenteritis was raised, a condition that may affect up to 0.6% of the population based on a recent review 3. The key in this case was communication with our radiologists along with quick access to EUS guided FNA. While our patient’s mass was thankfully benign, his case can provide a framework for workup of similar patients in the future.

Funding Agencies: None


The Snare Project: Closing the Loop on Synoptic Endoscopic Reporting and Skills Assessment
P. Rossos,1 T. Xenodemetropoulos,2 K. Lane,3 and D. Armstrong2

1University of Toronto, Toronto, ON, Canada
2McMaster University, Hamilton, ON, Canada
3Techna Institute, Toronto, ON, Canada

Background. Structured reporting improves the completeness and timeliness of procedure reports to ensure effective communication and data capture. Adoption barriers to endoscopy EMRs include costs, workflow, lack of optimized content and inability to incorporate clinical best practice. Traditional mechanisms of endoscopic skill assessment are inherently biased and do not support objective comparative analysis. Peer-comparator practice audits have demonstrated a basis for evaluating variation while providing opportunities to improve clinical practice.

Aims. The Structured Notes Auditing and Reporting in Endoscopy Project combines synoptic point of care clinical reporting through a recently developed pan-Canadian data model with the Practice Audit in Gastroenterology (PAGE) program. The combined data model will be supported through CAG to serve as the first national initiative to combine standardized reporting, quality indicators and endoscopist practice and performance feedback.

Methods. Endoscopists were engaged nationally to achieve consensus on clinical content, terminology, performance, data quality indicators and patient outcomes to develop a Pan-Canadian data model. A National Endoscopist Working Group representing academic, community, adult and pediatric practice reviewed and incorporated elements and indicators from CAG Consensus Guidelines, Clinical Outcomes Research Initiative (CORI), UK Global Rating Scale (GRS), Colonoscopy Reporting and Data System (CO-RADS) and Minimal Standard Terminology (MST). Data elements were defined as either essential or optional. Participants were advised to consider use generically across technology platforms.

PAGE is a mobile IT based evaluation instrument developed for the CAG Quality Program in Endoscopy. It is a well-accepted, simple mechanism of administering peer-comparator practice audit to Gastroenterologists in independent practice. An initial SNARE pilot will combine the national trainee RPAGE program with the pan-Canadian data model at the University Health Network in Toronto to evaluate the feasibility of a scalable approach towards a nationally administered program based on CAG and industry partnerships.

Results. For presentation and discussion:

(1)Pan-Canadian data models for colonoscopy and upper GI endoscopy; lessons learned for establishing other clinical procedures.(2)RPAGE instruments for comprehensive professionalism and performance evaluation with anonymized, peer-comparator functions.(3)SNARE pilot Results.

Conclusions. The integration of synoptic reporting and practice audit represents a unique Canadian opportunity to support clinical and industry collaboration to achieve both endoscopic data capture and objective performance feedback.

Funding Agencies: CAG, Canada Health Infoway


Upper Gastrointestinal Bleeding due to Gastric Stromal Tumor-One of the Forgotten Differentials
S. Bharadwaj,1 M. Alzahrani,1 R. Alkhiari,1 R. Al-Dabbagh,2 T. Gohel,1 and R. Spaziani2

1McMaster Univeristy, Hamilton, ON, Canada
2McMaster University, Stoney Creek, ON, Canada

Background. Gastro-intestinal stromal tumours are the most common mesenchymal tumours of the gastro-intestinal tract. This case report highlights the importance of GIST in patients with no known risk factors for gastrointestinal bleeding.

Aims. This case report highlights the importance of GIST in patients with no known risk factors for gastrointestinal bleeding.

Methods. Case report and literature review.

Results. 54 year old female with past medical history of iron deficiency anemia and menorrhagia for which she underwent dilatation and curettage came with chief complaint of melena for 2 days. No known risk factors of gastrointestinal bleeding was elicited in history except for 1 dose of oral naproxen given prior to the procedure. Subsequently, also had a syncopal episode. On physical examination, was orthostatic and hypotensive. Rectal examination was evident for melena. Laboratory investigations showed a drop in hemoglobin from baseline of 114 to 83 g/L and also elevated BUN. After initial resuscitation with IV fluids and pantoprazole drip, EGD done showed an ulcerated sessile polyp about 5 cm in diameter at the gastric body. The suspicion of GIST tumor was confirmed by a CAT scan of the abdomen. A biopsy was not obtained due to friable nature of the polyp.

Conclusions. Gastro-intestinal stromal tumours (GIST) are the most common mesenchymal tumours of the gastro-intestinal tract (GI). They account for approximately 0.1 to 3% of all GI neoplasms. In patients with no known risk factors for gastrointestinal bleeding, GIST should be suspected as one of the etiologies.

Funding Agencies: None

Cytokines and Intracellular Signals

Poster of Distinction


PAR2 Activation Inhibits Epithelial Wound Healing by Affecting E-Cadherin Expression and Lamellipodia Formation
E. Trusevych,1 P. Beck,1 and W. MacNaughton2

1University of Calgary, Calgary, AB, Canada
2Univ Calgary, Calgary, AB, Canada

Background. Protease-activated receptors (PARs) and their activating enzymes have been postulated to play a role in IBD pathogenesis. While previous studies have shown that PAR2 is highly expressed on intestinal epithelial cells and its activating enzymes are increased in IBD, the specific roles of PAR2 in disease initiation and progression remain unclear. However, PAR2 activation has both pro-proliferative and pro-migratory effects, and could be involved with restoration of the epithelial barrier following injury.

Aims. We tested the hypothesis that activation of PAR2 could increase the rate of epithelial wound healing.

Methods. Using colonic epithelial Caco2 cells, PAR2 was activated with the selective activating peptide 2f-LIGRLO (2fLI 0.5 μM–10 μM). For wound healing experiments, circular wounds were made in cell monolayers with a pipette tip and monitored with live-cell imaging. Proliferation was measured using an EdU assay. For immunofluorescence, wide-field images of E-cadherin and F-actin were taken at 20x and stitched together to capture the entire wound border and surrounding cells. To visualize actin dynamics, cells were transfected with a LifeAct plasmid to GFP-tag actin. Live cell videos were captured on a spinning disk confocal over 24 hr.

Results. Contrary to our hypothesis, PAR2 activation with 2fLI significantly inhibited the rate of wound closure over 48 hr (% wound closure) compared to control (%). In confluent monolayers, 2fLI was able to significantly increase proliferation of Caco2 cells (% EdU+ cells) compared to control (). However, 2fLI did not affect proliferation in wound-edge cells. Since both adherens junction and actin dynamics can affect epithelial migration, we next imaged the entire border of a wound stained for E-cadherin and F-actin. In control cells, there was a distinct loss of E-cadherin in cells surrounding the wound edge that was not seen in PAR2-activated cells. 2fLI treatment also resulted in a prominent actin cable surrounding the wound and prevented leader cell formation. Using LifeAct to visualize actin dynamics, 2fLI-treated cells could form filipodia projections but lacked lamellipodia. The actin cable appeared to prevent cells from migrating to close the wound.

Conclusions. We uncovered a novel effect of PAR2 activation, where 2fLI was able to inhibit wound closure in Caco2 cells. Although PAR2 activation had no effect on proliferation in wound-edge cells, it significantly slowed the rate of wound healing by inhibiting cell migration. PAR2 activation may be preventing the internalization of E-cadherin, which could prevent leader cell formation and sheet migration, in addition to inhibiting lamellipodia formation. Future directions include determining the role of RhoGTPases following PAR2 activation.

Funding Agencies: CCC, Alberta IBD Consortium, Alberta Cancer Foundation


ATP-Induced Inflammasome Activation Increases Bacterial Clearance through ROS Production
M. Bording-Jorgensen, M. Alipour, G. Danesh, and E. Wine

University of Alberta, Edmonton, AB, Canada

Background. The proinflammatory cytokine interleukin (IL)-1β is released from macrophages and monocytes through a class of protein complexes called inflammasomes. Nod-like receptor protein-3 (NLRP3) inflammasomes have been linked to various inflammatory conditions such as inflammatory bowel diseases (IBD). Conditions associated with inflammasomes are typically characterized by an overabundance of IL-1β with the exception of IBD, where its dysregulation leads to an IL-1β reduction. ATP has been shown to be protective against Escherichia coli and Staphylococcus aureus infections. The mouse pathogen Citrobacter rodentium, a common mouse model pathogen for enteropathogenic E. coli, is used to understand the dynamic relationship between pathogens, the inflammasome, and the epithelial barrier. We have previously shown that NLRP3−/− mice given exogenous IL-1β had improved ability to clear C. rodentium infections. Our hypothesis was that ATP-induced inflammasome activation increases macrophages ability to eliminate C. rodentium, possibly through ROS activation.

Aims. To determine whether ATP-induced inflammasome activation decreases intracellular bacterial survival and define mechanisms involved.

Methods. Gentamicin protection assay with J774A.1 cell line macrophages was used to determine the rate of phagocytosis and bacterial killing. ATP (2.5 mM; NLRP3 activator) was utilized to stimulate endogenous IL-1β production; Apocynin, Diphenyleniodonium (DPI), and N-acetyl cysteine were used as ROS inhibitors. IL-1β was measured using an ELISA on supernatants and cell lysates. ROS production was measured using DCFDA.

Results. Activation of the inflammasome, using extracellular ATP, significantly increased the ability of J774A.1 macrophages to kill C. rodentium and this was associated with an increase in ROS production. ROS inhibition, using NAC and Apocynin, resulted in a reduction of microbial death and ROS production while DPI did not. Cytokine analysis showed that the secretion of IL-1β was not different between treatments.

Conclusions. Inflammasome activation appears to play a critical role in the clearance of pathogens, mediated by ROS activation, whether through direct pathogen elimination or localizing the immune response. In relation to IBD, this dysregulation of the inflammasome may contribute to an increase in host susceptibility to pathogens. Studying the role of IL-1β on macrophage activity during inflammation will lead to a better understanding of inflammatory diseases.

Funding Agencies: None


Investigating the Underlying Mechanisms of Aquaporin 3 Involvement in Intestinal Epithelial Cell Proliferation
E. Morgan,1 M. Peplowski,1 and W. MacNaughton2

1University of Calgary, Calgary, AB, Canada
2Univ Calgary, Calgary, AB, Canada

Background. Aquaporin 3 (AQP3) is an aquaglyceroporin that is permeable to water and small solutes, such as glycerol and urea, and is known to be associated with cell proliferation, cell death and migration. In inflammatory bowel disease (IBD), water balance is disturbed resulting in impaired absorption and secretion, as well as barrier dysfunction. While aquaporins are expressed throughout the gastrointestinal tract, little is known about the physiological regulation of AQPs in intestinal epithelial cells.

Aims. We aim to better understand the functional importance of AQP3 in intestinal epithelial cells. We hypothesize that reduced AQP3 expression will result in reduced proliferative capacity, greater susceptibility to induced cellular stress, and decreased cell survival.

Methods. A stable AQP3 knockdown in HT29 human adenocarcinoma cells was developed using short hairpin RNA (shRNA). Cell proliferation was assessed in AQP3 knockdown cells, non-targetting scrambled shRNA control cells as well as untransfected HT29 cells over 72 hours under 10% FBS conditions. Fluorescence microscopy with EdU staining was also used to confirm active cell proliferation. The level of apoptosis was determined by detection of cleaved caspase-3 and cleaved poly (ADP-ribose) polymerase (PARP) using western blot. Cells were treated with 40 ng/mL IFNγ and 10 ng/mL TNFα and cleaved caspase-3 and cleaved PARP products were assessed over 24 hours. Necrosis was assessed by lactate dehydrogenase (LDH) assay over 72 hours. Flow cytometry was used to assess cell cycle progression using propidium iodide nucleid acid binding dye, which was quantified using FlowJo software.

Results. At 72 hours, AQP3 knockdown cell clones exhibit reduced proliferation by 47–61% under 10% FBS conditions. AQP3 knockdown cells also exhibit significantly reduced EdU incorporation, confirming decreased active DNA synthesis. Western blot did not show differences in apoptosis markers at baseline levels or following cytokine-induced apoptosis, nor were there evelated levels of LDH indicating that AQP3 knockdown cells are not undergoing significantly increased levels of cell death. However, AQP3 knockdown cells display significant differences in progression through the cell cycle with increased accumulation of cells in the G2 phase.

Conclusions. We have shown that AQP3 promotes proliferation in intestinal epithelial cells which appears to be due to altered progression through the cell cycle, and not due to increased rates of cell death. Our data improve the understanding of the functional role and importance of AQP3 in intestinal epithelial cell homeostasis and could lead to improvements in managing water balance, as well as improved treatments for IBD.

Funding Agencies: None


CD4+ T Cell Derived TNF Is Elevated in Patients with Diarrhea-Predominant but Not Constipation-Predominant Irritable Bowel Syndrome
Y. Nasser, C. Petes, C. Simmers, K. Gee, and S. Vanner

Queen’s University, Kingston, ON, Canada

Background. Systemic immune activation with sustained, low-grade inflammation is thought to underlie the pathogenesis of irritable bowel syndrome (IBS), but these findings have been inconsistent. We hypothesized that this inconsistency is due to the heterogeneity of IBS and the variability of immune expression measured in serum.

Aims. Our primary aim was to assess whether CD4+ T-cell derived pro-inflammatory cytokines were elevated and if this was confined to subsets of IBS patients. To gain insights into potential mechanisms, our secondary aim was to assess whether this immune activation correlated with the severity of psychological scores.

Methods. IBS patients () or healthy volunteers () were recruited from the outpatient gastroenterology clinic at the Hotel Dieu Hospital. CD4+ T-cells were isolated from blood and incubated with media or phytohaemagglutinin (PHA) at 5 μg/mL for 24 hrs. Supernatants were analyzed for the production of TNFα, IL-6 and IL-10 by ELISA. Subjects also completed validated psychological, symptom severity (IBS-SSS) and quality of life (QOL) questionnaires.

Results. PHA-stimulated cytokines were unchanged in IBS patients when compared to controls. When patients were analyzed by IBS subgroup, a significant increase in PHA-stimulated TNFα was seen in IBS-D but not IBS-C patients when compared to controls ( pg/mL, Controls;  pg/mL, IBS-D;  pg/mL, IBS-C. , Kruskal-Wallis Test; Control versus IBS-D and IBS-D versus IBS-C). IBS-SSS were in the moderate severity range ( IBS; Controls, ), yet these patients still exhibited significantly worse QOL ( IBS; Controls, ), increased anxiety ( IBS; Controls, ), depression ( IBS; Controls, ), and somatization ( IBS; Controls, ) scores when compared to controls. However, no differences in symptom severity or psychological scores were noted between IBS-C and IBS-D patients. CD4+ derived TNFα was not correlated with psychological or symptom severity scores.

Conclusions. IBS-D but not IBS-C patients have increased CD4+ T-cell derived TNFα when compared to controls, suggesting there is immune activation in IBS-D patients. This may suggest different underlying mechanisms in IBS-D compared to IBS-C. The cytokine elevation was not correlated with psychological scores however, suggesting that these parameters may be not mechanistically linked.

Funding Agencies: CAG, CCC, CIHR, YN is the recipient of a SEAMO (South Eastern Ontario Academic Medical Association) clinical fellowship. CP is supported by a Dr. Robert John Wilson Graduate Fellowship.


The Pekin Duck Programmed Death Ligand-2: CDNA Cloning, Genomic Structure, Molecular Characterization and MRNA Expression Analysis
Q. Yao,1 K. Fischer,2 L. Tyrrell,1 and K. Gutfreund3

1University of Alberta, Edmonton, AB, Canada
2Li Ka Shing Institute of Virology, Edmonton, AB, Canada
3Department of Medicine, Edmonton, AB, Canada

Background. Programmed death ligand 2 (PD-L2) plays an important role in the attenuation of adaptive immune responses in higher vertebrates.

Aims. Here we describe the identification of the Pekin duck PD-L2 orthologue (duPD-L2) and its gene structure.

Methods. The duPD-L2 cDNA was obtained by RT-PCR on RNA from splenocytes and primers based on duck genomic sequences. Nucleotide sequences of the 5′ and 3′ ends of the duPD-L2 ORF were determined by RACE. A homology model of duPD-L2 was obtained using the structure of mouse PD-L2 as a template. An eukaryotic expression vector was generated for expression of a C-terminally-His-tagged PD-L2 protein and culture supernatants of transfected 293T cells were assessed by immunoblot using an anti-his antibody.

Results. The duPD-L2 cDNA encodes a 321 amino acid protein that has an amino acid identity of 76% and 35% with chicken and human PD-L2, respectively. Mapping of the duPD-L2 cDNA with duck genomic sequences revealed an exonic structure of its coding sequence similar to those of other vertebrates. Homology modeling of the duPD-L2 extracellular domain was compatible with the extracellular domain structure of mouse PD-L2. Residues known to be important for receptor binding of mouse PD-L2 were mostly conserved in duPD-L2. DuPD-L2 mRNA was constitutively expressed in most tissues examined with highest expression levels in lung, spleen, bursa, cloaca, cecal tonsil and very low levels of expression in muscle, kidney and brain.

Conclusions. Our observations demonstrate evolutionary conservation of the exonic structure of its coding sequence, the extracellular domain structure and residues implicated in receptor binding but the role of the longer cytoplasmic tail in avian PD-L2 proteins remain to be determined.

Funding Agencies: CIHR

Gastro Intestinal Oncology


Functional Impact of Colorectal Cancer-Associated Mutations in the Tyrosine Phosphatase SHP-2
J. Gagné Sansfaçon,1 M. Langlois,1 C. Qu,2 and N. Rivard3

1Université de Sherbrooke, Sherbrooke, QC, Canada
2Aflac Cancer and Blood Disorders Center, Atlanta, GA, USA
3University of Sherbrooke, Sherbrooke, QC, Canada

Background. Gain-of-function mutations of PTPN11 gene (E76K, E76G, D61Y) were associated with pediatric leukemias (>30% of juvenile myelomonocytic leukemias) and certain solid carcinomas including colorectal cancer (CRC) (Bentires-Alj, Cancer Res 2004). In vitro, these specific mutations activate SHP-2 phosphatase activity and enhance its binding to signalling partners resulting in sustained activation of downstream effectors especially the RAS/MAPK pathway (Matozaki, Cancer Sci 2009). Dysregulation of this pathway is a common event in CRC. Indeed, activating mutations in KRAS or BRAF genes are found in up to 60% of colorectal tumors and are acquired at the very early premalignant stage.

Aims. The aim of this study was to investigate the functional impact of CRC-associated mutations in the tyrosine phosphatase Shp-2.

Methods. To determine the pathogenic effect of somatic mutation on E76 residue in intestinal epithelial cells (IECs) in vivo, we generated Ptpn11E76Kneo/+/Villin-Cre+ mice ( mice) by crossing Ptpn11E76Kneo/+ mice with Villin-Cre mice.

We also crossed our model with ApcMin/+ mice, heterozygous for Apc truncation mutation frequently found in human sporadic CRC, and which spontaneously develop adenomas in the intestine (; ).

Results. Our results show that mutant mice exhibited similar body weight to control mice. However, colon and small intestine length and weight were significantly increased one month after birth and thereafter. Furthermore, Ki67 immunostaining revealed that there was a significant increase in the number of proliferating cells in mutant mice in comparison to control mice. Surprisingly, decreased number of Paneth cells was observed in mice while Goblet cells were expanded. Additionally, reduced expression of total and active β-catenin protein was found in IECs while MERK/ERK signalling was markedly activated. We then analyze the potential involvement of the mutant in intestinal tumorigenesis. Notably, a major effect on intestinal tumor initiation was observed in ; mice compared to control littermate. The multiplicity of polyps with expression was indeed increased in the small and large intestine. Increased ERK and NFkB activities were observed in polyp extracts from ; mice.

Conclusions. Therefore, these results demonstrate that CRC-associated SHP-2 mutations promote IEC proliferation and tumorigenesis probably through activation of RAS/MAPK signalling pathway.

Funding Agencies: Cancer Research Society; FRQS


Exploring the Nature of Common Biological Roles between NCOR1 and Its Newly Identified Protein Interactor CHD8 in Colorectal Cancer Cells
A. Loiselle, S. St-Jean, F. Boisvert, and F. Boudreau

Université de Sherbrooke, Sherbrooke, QC, Canada

Background. NCOR1 (nuclear receptor corepressor 1) is a transcriptional corepressor that interacts with nuclear receptors. Using a quantitative protocol of SILAC (Stable Isotope Labeling In Cell Culture) immunoprecipitations combined with mass spectrometry, we previously identified CHD8 (Chromodomain helicase DNA binding 8) protein as a new interaction partner of NCOR1. However, the nature of CHD8 biological function in colorectal cancer (CRC) cell lines is currently unclearNCOR1 (nuclear receptor corepressor 1) is a transcriptional corepressor that interacts with nuclear receptors. Using a quantitative protocol of SILAC (Stable Isotope Labeling In Cell Culture) immunoprecipitations combined with mass spectrometry, we previously identified CHD8 (Chromodomain helicase DNA binding 8) protein as a new interaction partner of NCOR1. However, the nature of CHD8 biological function in colorectal cancer (CRC) cell lines is currently unclear.

Aims. To investigate the biological roles for CHD8 in CRC cells and to monitor biological complementary role(s) when compared to NCOR1 functions.

Methods. Caco-2/15 and HT-29 cells lines were depleted in NCOR1 or CHD8 by RNAi. Cell proliferation was measured by cell counting and senescence detected by measuring senescence associated secretory phenotype, SABeta-galactosidase assays and DNA damage. Tumour growth properties were assessed by xenografts in immunodeficient mice.

Results. CHD8 physical interaction with NCOR1 was validated by co-immunoprecipitations in various CRC cell lines. NCOR1 depletion in both Caco-2/15 and HT-29 led to a strong reduction in cell proliferation and induction of a senescence phenotype. In parallel, CHD8 depletion in these cells led to a partial reduction in cell proliferation without leading to the senescence phenotype. A similar observation was made in the context of xenografts obtained in immunodeficient mice with an intermediate decrease of tumour growth for HT-29 CHD8 depleted cells when compared to NCOR1 depleted cells. Finally, a closer analysis of the CHD8 protein expression profile among different cell lines identified a shorter CHD8 isoform in normal and CRC cells,() and the classical longer form predominantly expressed in CRC cells (). Depletion of both short and long CHD8 forms in Caco-2/15 and HT-29 cells led to drastic growth arrest as observed for NCOR1 depletion assays.

Conclusions. CHD8 and NCOR1 are crucial regulators of cell proliferation in CRC cells. Functional characterization of the biological relevance of both CHD8 isoforms in comparison to NCOR1 is currently ongoing to better understand the functional roles of these interactions.

Funding Agencies: CIHR, NSERC


HNF4A Plays Role in DNA Repair in Colorectal Cancer
S. Wilson, J. Babeu, F. Boudreau, and F. Boisvert

Université de Sherbrooke, Sherbrooke, QC, Canada

Not published at author’s request.

Funding Agencies: CAG, CIHR


Regulation of P2Y6 Receptor Expression by P53
C. Molle, M. Placet, G. Arguin, and F. Gendron

Université de Sherbrooke, Sherbrooke, QC, Canada

Background. Extracellular UDP selectively activates the G protein-coupled P2Y6 receptor (P2Y6R). In fact, extracellular nucleotides, such as UDP, are found in high concentration in the vicinity of colorectal tumours. However, the role of these nucleotides and associated receptor system is not fully understood. Previous results showed that P2RY6 gene and protein expression is upregulated in cancerous intestinal epithelial cells (cIECs) harbouring P53 mutations. The P2Y6R isoform 1 is coded by different RNA variants that share an overlapping promoter region. More recently, we identified a new promoter region for the P2RY6 gene that appears to code for P2Y6R isoform 2 for which the function is still unknown. In this context, the hypothesis is that P2RY6 gene expression is regulated by epigenetic modifications caused by P53 mutations in cIECs, which could lead to the differential regulation of P2Y6R isoforms expression.

Aims. The aims are to: (1) determine if P53 or its R273H mutant isoform (), a common mutant found in colorectal cancer, could regulate P2RY6 gene expression and (2) characterize the epigenetic marks associated with P2RY6 expression in cIECs.

Methods. We determined P2Y6R expression levels by qPCR in cIECs and receptor activity using intracellular calcium mobilization assays. We targeted two promoter regions on the 11q13.4-13.5 locus (NC_000011.10: ) that are coding for isoform 1 (R1 and R4: ) and isoform 2 (R4: ). We cloned these regions in the pGL4.10 luciferase-expressing vector and cotransfected these constructions with recombinant wild-type P53 () or in HCT 116 cells prior to luciferase assays as previously described. The epigenetic modifications affecting P2RY6 expression were determined by ChIP assays to measure the level of trimethylated lysine27 of histone H3 (H3K27me3) associated with the proximal promoter region.

Results. P2Y6R expression is upregulated in IECs harbouring a mutated TP53 gene as compared to those having the wild-type isoform. Luciferase assays showed that P53 activates both R1 and R4 promoter regions, whereas the mutant could only induce the transcription of the R4 region. Hence, by increasing the quantity of , we reduce the capacity of to induced R1 transcription, which is dominant promoter region coding for P2Y6R isoform 1. This result suggests a form of competition between P53 and . Finally, qPCR analyses confirmed that expression increased the transcription of P2RY6 gene.

Conclusions. The mutant seems associated to increase expression of P2Y6R isoform 2 for which the function is not yet identified but that are most likely linked to stimulation of cell proliferation and resistance to apoptosis.

Funding Agencies: CIHR


Hedgehog Pathway and the Primary Cilium in Colorectal Cancer Cells
B. Senicourt, S. Boudjadi, N. Basora, J. Carrier, and J. Beaulieu

Université de Sherbrooke, Sherbrooke, QC, Canada

Background. The Hedgehog pathway is involved in the maintenance of numerous cell types both during development and in the adult. Often deregulated in cancers, its involvement in colorectal cancer has come into view during the last few years, although its role remains poorly defined. In most tissues, the Hedgehog pathway is highly connected to the primary cilium, an organelle not expressed in the normal colonic epithelium which recruits the functional components and regulates the pathway.

Aims. Since the intestinal epithelium is known to be a non-ciliated tissue, the HH pathway as related to the PC has not been explored. We investigated the presence of PC in colon cancer tumors and cell lines. We used cellular models to look at HH activation, focusing on the final effector Gli1. The link between PC and the HH pathway was shown by the recruitment of the Smo receptor in the PC.

Methods. We looked for PC in a subset of 63 tumors from a Tissue Micro Array and in 4 colorectal cell lines by immunofluorescence using two well-known markers, acetylated α-tubulin and polyglutamylated tubulin. 3D deconvoluted pictures were obtained to characterize the shape and size of PC. Using cellular models we investigated HH pathway expression by qPCR. We assessed the functional link between HH pathway and PC through localization of the Smo receptor in the PC using immunofluorescence.

Results. We observed the presence of the primary cilium in the epithelium of primary colorectal tumors at all stages but not in their normal counterparts. Using human colorectal cancer cell lines we found a clear correlation between the presence of the primary cilium and the expression of the final Hedgehog effector, GLI1, and provide evidence of a functional link between the two by demonstrating the recruitment of the SMO receptor to the primary cilium membrane.

Conclusions. We conclude that the primary cilium directly participates in the Hedgehog pathway in colorectal cancer cells.

Funding Agencies: CAG


Prevalence of Colorectal Polyps in Liver Transplant Patients
A. Ma, A. Therrien, and M. Bouin

Centre Hospitalier de l’Université de Montréal, Montréal, QC, Canada

Background. Liver transplantation (LT) is a risk factor for the development of neoplasm, the third cause of mortality at one year following LT. Colorectal cancer (CRC) is particularly lethal in post transplant patients, with a survival of 30% at five years (compared to 63% in the general population). However, the risk of CRC and the risk of developing colorectal polyps post LT have not been prospectively studied.

Aims. The main objective of our study was to determine the prevalence of colorectal polyps post LT.

Methods. We undertook a prospective study of all LT recipients between January 1st, 2007 and December 31st, 2009 at our tertiary center. Were included patients who underwent a screening colonoscopy within 10 years before and years after LT. The demographic, medical and endoscopic data were extracted from charts. Patient subgroups were formed and compared based on the presence or not of polyps.

Results. Of 98 liver transplant recipients with a pre-LT colonoscopy, 50 patients underwent post-LT colonoscopy screening and were included (mean age years; 40% female). Twenty-four of the excluded patients had died before their follow-up colonoscopy.

In the pre-LT setting, colorectal polyps were found in 40% of patients. The most common histological findings were hyperplastic polyps (40%) and adenomas (25%). The histology results were unavailable in the rest of cases.

In the post-LT setting, the colonoscopy was performed on average at years and polyps were found in 28% of patients. The histological findings were adenomas in 42.8% and hyperplasia in 28.6% of cases (not significantly different from the pre-LT findings). There was one case of neuroendocrine tumor but none of CRC. The histology results were unavailable in the rest of cases.

Furthermore, the presence of pre-LT polyps was not predictive of developing post-LT polyps (25% of patients with pre-LT polyps versus 30% without developed post-LT polyps; ).

Patients with post-LT adenomas were not significantly older than those without (56.3 years versus 53.4%; ). There was a trend towards greater risk of developing post-LT polyps in men (36.7% versus 15% in women; ).

Conclusions. In this prospective study, the prevalence of colorectal polyps at years post-LT is high (28%). Polyps were adenomatous in almost half of the cases. A trend exists between the male gender and the risk of developing polyps. This study suggests that post-LT colonoscopy surveillance may be of benefit for CRC prevention.

Funding Agencies: None


Improved Sample Quality Obtained by EUS-Guided Sink Compared to FNA for Foregut Subepithelial Lesions
M. Boulos, D. Wang, P. James, T. Moyana, and A. Chatterjee

University of Ottawa, Ottawa, ON, Canada

Background. Gastric subepithelial lesions (SELs) can be divided into three major categories, namely smooth muscle tumors (leiomyomas and leiomyosarcomas), neurogenic tumors (schwannomas and neurofibromas) and gastrointestinal stromal tumors (GIST). GIST are the most common type of foregut SEL and carry an important malignant potential. Small SELs (<2 cm) have been notoriously difficult to sample endoscopically. Endoscopic ultrasound (EUS)-guided single incision needle knife (SINK) biopsy has become increasingly used for deep tissue sampling of foregut SELs, however there exists limited evidence to suggest that this results in superior specimen acquisition.

Aims. We sought to review our experience regarding the difference in sample quality of SELs obtained by EUS-guided SINK compared to EUS-guided fine needle aspiration (FNA).

Methods. We performed a retrospective chart review of EUS-guided SINK cases performed at The Ottawa Hospital for the evaluation of foregut SELs. These samples were compared to consecutive EUS-guided FNA samples obtained over a similar time period. Two pathologists reviewed the specimens blindly and independently. The quality of each sample was determined based on a 5-point scale, where poor = 1, adequate = 2, good = 3, very good = 4 and excellent = 5.

Results. 13 patients with foregut SELs were sampled by SINK and these were compared to 26 consecutive EUS-guided FNA samples. 12 out of the 13 (92%) SINK cases were reported to be of excellent quality (5/5) whereas one case was of adequate quality (2/5). The median FNA quality score was 3 with an interquartile range of 2–5, which was found to be significantly inferior to SINK (). 8 SINK cases (62%) were reported to have a cellularity of ≥5 000. Only 4 EUS-guided FNA specimens (15%) were reported to have a cellularity of ≥5 000.

Conclusions. The sample quality of subepithelial lesions obtained by EUS-guided SINK may be superior to EUS-guided FNA.

Funding Agencies: None


Topical Hemostatic Spray for the Management of Malignancy-Related Gastrointestinal Bleeding: A Systematic Review and Meta-Analysis
M. Sandhu, P. James, and S. Piscopo

The University of Ottawa, Ottawa, ON, Canada

Background. Hemostatic powder spray agents (HPSAs) have been shown to be effective for gastrointestinal haemorrhage (GH), however their role as first-line agents is limited. Conventional endoscopic methods often fail to achieve hemostasis in cases of malignancy-related GH due to lesion location, lesion distribution and altered tissue responses secondary to the malignant process, anticoagulation and/or chemoradiation treatment. The ability of HPSAs to treat large surface areas without touching tissue render them ideal for the management of malignancy-related GH, however their role in this setting remains unclear.

Aims. To review the literature on the efficacy of HPSAs in malignancy-related GH.

Methods. We performed a systematic search of EMBASE and MEDLINE through June 2015 for studies reporting the use of HPSAs for malignancy-related GH. Duplicate articles and case reports were excluded. The primary outcome was hemostasis at 72 hours post-treatment. A pooled estimate was calculated using random effects models. The methodological quality of the included studies was assessed using the Newcastle-Ottawa scale.

Results. Of the 1,704 citations identified, a full-text review was performed on 89 and 8 were included in the meta-analysis (44 patients). Four different HPSAs were identified: Hemospray®, cyanoacrylate spray, Costasis®, and Endoclot®. The most commonly used spray in these patients was Hemospray (5 studies). Five studies included less than 5 patients. Nine studies scored 7 out of 9 and one study scored 6 out of 9 by using the Newcastle-Ottawa Quality Assessment Scale. Immediate hemostasis was achieved in all cases. Meta-analysis showed that treatment with HPSAs resulted in hemostasis for up to 72 hours in 90% of cases (95% confidence interval 0.67–0.99).

Conclusions. The limited evidence to date suggests that topical hemostatic sprays are effective in the setting of malignancy-related GH. Larger prospective studies are required.

Funding Agencies: None


A Comparison of Fecal Immunochemical Testing to Guaiac-Based Fecal Occult Blood Testing for the Detection of Colorectal Adenomas and Cancer
L. Du,1 C. Teshima,2 B. Moysey,1 A. Morse,1 and R. Sultanian1

1University of Alberta, Edmonton, AB, Canada
2University of Toronto, Toronto, ON, Canada

Background. Programmatic colorectal cancer (CRC) screening was initiated in Edmonton in 2011 as the SCOPE program, offering colonoscopy to patients with above average CRC risk, including positive guaiac-based fecal occult blood tests (gFOBT), as well as a personal or family history of CRC. The newly developed fecal immunochemical test (FIT) has demonstrated improved sensitivity without the loss of specificity for detection of advanced adenomas and CRC. In November 2013, FIT replaced gFOBT as the first-line CRC screening test in Alberta. The goal of our study was to assess the impact of the transition from gFOBT to FIT on a screening colonoscopy program.

Aims. To compare the detection rates of polyps, adenomas, advanced adenomas, and CRC of gFOBT to those of FIT in patients selected for colonoscopy through a region-wide CRC screening program.

Methods. A retrospective cohort analysis was performed using a prospectively maintained database of all patients, aged 50–74, who underwent colonoscopy in the SCOPE program between January 1, 2013 and December 31, 2014 as a result of a positive gFOBT or a positive FIT. Patients with morbid obesity, significant co-morbidities, or overt gastrointestinal symptoms were excluded. Colonoscopy was offered to patients with at least one out of three samples positive using the guaiac-based Hemoccult II® SENSA in 2013 or with a positive FIT, defined as ≥75 micrograms/gram of stool, using the Polymedco OC FIT-CHEK® in 2014. All procedures were performed during dedicated SCOPE endoscopy time and by endoscopists certified by the program.

Results. 633 patients underwent colonoscopy due to a positive gFOBT in 2013, and 2137 patients underwent colonoscopy due to a positive FIT in 2014, values that represent a substantial increase in the number of patients referred for colonoscopy due to a positive screening test between the two years. Patients who were FIT-positive had significantly higher polyp detection rates (PDR) (71% versus 63%; ) and adenoma detection rates (ADR) (62% versus 48%; ) in comparison to gFOBT-positive patients. CRC detection was significantly higher in gFOBT-positive patients in comparison to patients who were FIT-positive (5.6% versus 3.1%; ).

Conclusions. The conversion of the programmatic CRC screening in Edmonton from gFOBT to FIT-based selection of patients resulted in significantly increased CRC screening rates, increased referral numbers, as well as higher PDR and ADR. FIT resulted in a lower CRC detection rate percentage although the total number of cases detected was higher than the FOBT-positive group due to increased FIT utilization.

Funding Agencies: None

Hepatobiliary Neoplasia


A Diagnostic Dilemma: A Case of Cholestatic Jaundice due to Al-Amyloidosis
R. Al-Dabbagh, S. Bharadwaj, S. Patterson, and M. Puglia

McMaster University, Hamilton, ON, Canada

Background. Amyloidosis is a rare, infiltrative condition associated with extracellular deposition of fibrils that can lead to end organ dysfunction. Most often, patients with primary amyloidosis present with cardiac or renal involvement. If the liver is involved, it usually as asymptomatic hepatomegaly. Furthermore, serious liver dysfunction, with initial presentation of cholestatic jaundice is very rare, accounting for less than 5% of amyloidosis.

Aims. We present a case of cholestatic jaundice due to amyloidosis with unclear concurrent multiple myeloma.

Methods. A full chart review of the case was undertaken, including assessment of radiographic, biochemical and biopsy Results. A subsequent literature review of the topic was also conducted.

Results. A 69 year old male initially presented with a 3-4 month history of right upper quadrant abdominal pain. He also reported reduced oral intake and an associated weight loss of 25 pounds. However, he denied fevers, night sweats, rashes, and review of systems was otherwise unremarkable. Physical examination was prominent for scleral icterus, right upper quadrant tenderness, nonpulsatile hepatomegaly, and peripheral edema. Laboratory investigations revealed hemoglobin of 121 g/L (MCV 96.0 fL), creatinine of 103 μmol/L, total bilirubin of 82 μmol/L (conjugated 59.6 μmol/L), albumin of 21 g/L, gamma-glutamyl transpeptidase of 1773 U/L, alkaline phosphatase of 692 U/L, alanine transaminase 45 U/L, aspartate transaminase of 97 U/L, and INR of 1.1. Additionally, abdominal ultrasonography revealed a liver span of 20 cm, with diffuse fatty infiltration, spleen of 12 cm in size, and normal caliber and patency of the portal vein and common bile duct. A subsequent CAT scan of the chest, abdomen and pelvis, and MRCP were also unremarkable. His hospital course was complicated by worsening laboratory abnormalities, including worsening hyperbilirubinemia (conjugated 247 μmol/L), INR (1.8), acute kidney injury (creatinine 314 μmol/L), and nephrotic range proteinuria. Due to suspicion of amyloidosis in the setting of multi-organ failure, serum electrophoresis was done which revealed free kappa of 645.46 mg/L and free lambda of 38.21 mg/L. Finally, liver biopsy was performed, showing severe amyloidosis occupying the sinusoids, spaces of Disse, portal connective tissue and walls of vessels, with compression of hepatocytes. Congo red staining showed green birefringence. He was started on dexamethasone, but further chemotherapy had been withheld until further characterization can be made of possible concurrent multiple myeloma.

Conclusions. Cholestatic jaundice is common, but is rarely the initial presentation of amyloidosis. If initial investigations rule out any obvious etiology, suspicion for infiltrative diseases, such as amyloidosis, should be raised.

Funding Agencies: None


Should Anticoagulation Be Offered in Patients with PVT in the Setting of HCC?
T. Mahmoudi, A. Kayal, R. Carvalho, and A. Weiss

UBC, Vancouver, BC, Canada

Background. Portal vein thrombosis (PVT) is a seen in about 20–44% of patients with hepatocellular carcinoma (HCC). To our knowledge, no other study has looked at the need for anticoagulation in patients with HCC and PVT.

Aims. The aim of this study is to investigate the natural history and progression of portal vein thrombosis in patients with hepatocellular carcinoma with or without anticoagulation therapy.

Methods. Using the British Columbia Cancer Agency database, a cohort of 54 patients who were diagnosed with both conditions were evaluated retrospectively. Nine patients were excluded secondary to lack of follow up. HCC and PVT diagnosis and follow up was made with contrast enhanced CT or MRI. Most patients received a single or a combination of the following treatments: transarterial chemoembolization, radiofrequency ablation or surgical resection. Thirty five (78%) patients received systemic therapy with Sorafenib.

Results. Thirty eight patients were males and mean age was 62.8. Liver disease etiology was HCV in 19 (42%), HBV in 18 (40%), ETOH in 5 (11%) and hemochromatosis in 1 (2%). Results: Average survival after HCC diagnosis was 28 months and 15 months after PVT diagnosis. Among the 45 patients evaluated, 8 patients received anticoagulation while 39 did not. PVT progression occurred in 19 (49%) of the non anticoagulated group, and 4 (67%) of the anticoagulated group. Right portal vein involvement was seen in 18 (40%) patients with progression in 67% of the time, Left PVT in 13 (28%) with a progression in 7(54%), and main PVT 6 (13%) with a progression in (67%). In 1 case, PVT progressed from the main PVT to Superior mesenteric vein (SMV) and from the LPV to SMV in 2 other cases. No symptoms directly related to PVT development were reported.

Conclusions. The possible anticoagulation related complications need to be considered before attempting therapy in patients with HCC and PVT. Despite the small number of patients included in this study, this review shows that PVT progression in patients with HCC and the absence of clinical complications is similar in both anticoagulated and non anticoagulated groups. Thus, the usefulness of anticoagulation in this patient population needs to be further studied.

Funding Agencies: None

Immunology and Inflammatory Bowel Disease


Litterature Review of the Economic Impact of Treatments for Inflammatory Bowel Diseases (The Igenomed Consortium)
J. Lachaine,1 A. Bitton,2 A. Miron,1 and D. Nait Ladjemil1

1University of Montreal, Montreal, QC, Canada
2McGill University, Montreal, QC, Canada

Background. The last decade witnessed great advances in the treatment of inflammatory bowel diseases (IBD) with the introduction of biologic therapies. Several economic evaluations have been run to evaluate these treatments.

Aims. The goal of this study was to analyse the existing evidences and key parameters included in IBD cost-effectiveness studies.

Methods. A systematic literature review was conducted to identify economic evaluations of IBD therapy. Electronic databases (Embase and Medline) were used to identify full economic evaluations published from 2004 to 2015. Cross-references of selected articles and gray literature search were also performed to find additional publications. The health outcomes, costs, incremental cost-effectiveness (ICERs) and cost-utility ratios (ICURs) were analysed.

Results. The literature review allowed identifying 3,631 potentially relevant studies. Titles and abstracts screening allowed the selection of 53 articles. After assessment of those articles, 36 were found relevant for the review. Four other studies were added from gray literature. Different treatments were evaluated including biologics (53%), mesalamine (28%), biologics and immunosuppressants combination (5%) and immunosuppressants alone (3%). Infliximab was the most common biologic treatment evaluated (65% of all biologics). In the cost-utility analyses (CUA) (88%), 35% had utility scores derived from IBD severity scores. The remaining studies used direct and indirect utility measurement methods, including EQ-5D (43%), standard gamble (33%), time trade off (25%) and visual analog scale (8%). Markov modeling, decision tree or a combination of both were used in 38%, 38% and 5% of the studies respectively. All studies included drug acquisition costs, 50% included treatment administration costs, 65% included hospitalization costs and 45% included surgical costs. In CUA, the main outcome measures were ICURs ranging from dominant (less costly and more effective) to USD2,757,857/QALY (CAD2,955,814/QALY). More specifically, treatment under investigation was dominant in 34% of the analyses. ICURs were bellow CAD50,000/QALY in 57% of cases and bellow a threshold of CAD100,000/QALY in 71% of cases.

Conclusions. Several economic evaluations especially involving biologics were conducted in the past decade. This study showed that there is some homogeneity in the selection of key parameters, such as the use of Markov modeling and decision tree in model development, use of EQ-5D utility measurements and costs included.

Funding Agencies: Genome Canada


Hepatocellular Carcinoma in a Patient with Crohn’s Disease on Azathioprine
V. Heron,1 K. Fortinsky,2 G. Spiegle,2 N. Hilzenrat,1 and A. Szilagyi3

1McGill University, Montréal, QC, Canada
2University of Toronto, Toronto, ON, Canada
3Jewish General Hospital, Montreal, QC, Canada

Background. Hepatocellular carcinoma (HCC) rarely occurs in patients without underlying liver disease. While Crohn’s disease (CD) has been linked to certain forms of liver disease, HCC in these patients is rare.

Aims. We report the 12th case of HCC in a non-cirrhotic patient with CD and discuss the possible role of azathioprine.

Methods. Case report.

Results. A 61-year-old woman with CD was found to have elevated liver enzymes on routine blood work. Past medical history includes type 2 diabetes, dyslipidemia, and a family history of hemochromatosis in her father. She has a remote smoking history and no alcohol or drug use. Her CD was controlled with azathioprine for the past 8 years. She had previously been treated with 5-ASA, infliximab and multiple bowel resections.

Blood work revealed a mild asymptomatic transaminitis with normal liver function (lab results in Table 16). A liver mass was identified on both abdominal ultrasound and CT scan (Figure 12). Abdominal MRI confirmed a 3 cm lesion in segment 7. Liver biopsy showed well-differentiated HCC. Biopsies from non-neoplastic liver showed 30% macrovesicular steatosis without steatohepatitis, and minimal iron overload (grade 0-1/4).

The patient underwent tumor resection. Surgical pathology revealed no malignant cells despite initial biopsy showing HCC. The pathologist believes the tumor may have infarcted post biopsy. Azathioprine was stopped prior to surgery. There is no evidence of recurrence on imaging at 5-month follow-up.

This is the 12th reported case of HCC in a patient with CD in the absence of cirrhosis. Azathioprine was reported in 9 of the previous cases and has been proposed as a potential trigger given its known association with malignancies and its hepatotoxicity, especially in the setting of toxic levels of 6-MMP. Azathioprine may impair the immune system’s ability to correct dysplasia caused by inflammation related to CD.

This is the 3rd reported case of biopsy proven HCC with complete histologic spontaneous regression. This may be explained by the withdrawal of azathioprine or by the liver biopsy as operative trauma, including biopsy, has been linked to spontaneous regression.

Conclusions. Though further studies are needed, the relationship between azathioprine and HCC remains concerning. Imaging may be considered in CD patients with abnormal liver enzymes, especially those on azathioprine with elevated drug levels.

Funding Agencies: None

Intestinal Disorders

Poster of Distinction


Giardia Dduodenalis-Induced Goblet Cell Mucin Depletion Is Cysteine Protease-Dependent
C. Amat, J. Motta, K. Chadee, and A. Buret

University of Calgary, Calgary, AB, Canada

Background. Giardia duodenalis (syn. G. lamblia, G. intestinalis) is a cosmopolitan diarrheagenic parasite of the small intestine. Giardiasis leads to post-infectious irritable bowel syndrome (PI-IBS) and extra-intestinal complications via mechanisms that remain unclear. The mucus layers of the small and large intestines are protective against enteric infections, but we have hypothesized that Giardia may disrupt this barrier by two mechanisms: (i) by direct degradation of mucin proteins and (ii) by causing mucin depletion via hypersecretion in goblet cells. These effects could lead to both acute and chronic disease.

Aims. The aims of this study are to characterize how the enteropathogen Giardia duodenalis interacts with host mucus and assess its effects on the primary constituent of the mucus layer, mucin-2 (MUC2).

Methods. C57BL/6 wild-type (WT) mice and Muc2−/− (KO) mice were gavaged with Giardia trophozoites (Assemblage B, strain GS/M). Mice were weighed daily and samples collected on day 7, at the peak of infection. Trophozoites were counted in the small intestine, and the small intestine and colon were processed for histological staining. The liver and spleen were collected aseptically and incubated on Columbia blood agar (aerobically and anaerobically) to assess bacterial translocation. In vitro, secreted products obtained from Giardia trophozoites (Assemblage A, strain NF) were co-incubated with purified human mucin. Co-incubation products were evaluated for MUC2 protein concentration by western blotting. Giardia trophozoites were co-incubated with human colonic goblet cells LS174T in the presence and absence of E-64, a broad spectrum irreversible cysteine protease inhibitor. Staining intensity of MUC2 was normalized to cell count and quantified by immunofluorescence compared to control.

Results. Infected KO mice had higher parasitic loads and failed to gain weight compared to WT mice. Goblet cell mucin was depleted in the small intestine and colon of infected WT mice. Infected mice showed increased bacterial translocation of aerobic and anaerobic species into the liver and spleen. Staining intensity of MUC2 was reduced inside the human goblet cells in vitro after exposure to Giardia trophozoites. These effects were inhibited by a cysteine protease inhibitor. In addition, Giardia’s secreted products degraded human purified MUC2 in vitro.

Conclusions. Mucus protects the host against parasite accumulation and Giardia-induced weight loss. Giardia disrupts the mucus barrier by degrading MUC2 and causing hypersecretion leading to mucin depletion in the small intestine and colon. These effects are associated with increased bacterial translocation and may contribute to both acute and chronic disease. Mucin depletion is inhibited by a cysteine protease inhibitor, indicating a potential mechanism for hypersecretion.

Funding Agencies: NSERC CREATE Host-Parasite Interactions


Temporal Deletion of IEC-Specific Hdac1 and Hdac2 Modifies Gut Homeostasis
A. Gonneaud, N. Turgeon, C. Jones, F. Boudreau, and C. Asselin

Université de Sherbrooke, Sherbrooke, QC, Canada

Background. Hdac1 and Hdac2 deacetylase activities regulate gene expression by controlling the acetylation levels of epigenetic histone marks and of many regulatory proteins. We have observed that villinCre-mediated intestinal epithelial cell (IEC)-specific Hdac1 and Hdac2 deletion from day E15 alters cell differentiation and proliferation, resulting in chronic inflammation. In addition, different levels of Hdac1 and Hdac2 expression lead to incremental modifications in IEC determination and intestinal homeostasis. These effects could be in part due to compensatory mechanisms established during murine development after day E15.

Aims. Considering the fact that the intestinal environment is crucial to maintain gut homeostasis, we hypothesized that IEC-specific inducible deletion of Hdac1 and Hdac2 in adult mice could result in different alterations on intestinal homeostasis. We also speculated that Hdac1 and Hdac2 would alter intrinsic IEC physiology in enteroid cultures.

Methods. To determine the effect of IEC-specific Hdac1 and Hdac2 deletion in adult mice, floxed Hdac1 and Hdac2 mice were crossed with villin mice to obtain Hdac1−/−/Hdac2−/− mice after Tamoxifen injection of two-month-old mice. Small intestinal sections were stained with hematoxylin and eosin, and with Periodic Acid Schiff or lysozyme antibodies to detect respectively goblet and Paneth cells. PCNA antibodies were used to assess cell proliferation. IEC-intrinsic effect of Hdac1 and Hdac2 was determined in enteroid cultures. Jejunal villin Hdac1 and Hdac2 crypt enteroid cultures were treated with hydroxytamoxifen to induce gene deletion. Enteroid structure was observed by microscopy.

Results. Loss of Hdac1 and Hdac2 in the adult intestinal epithelium led to (1) jejunal tissue architecture defects and (2) goblet cell number decreases. In contrast to dual villin-Cre Hdac1 and Hdac2 deleted mice, villin mice did not display significant differences in proliferation as well as in Paneth cell numbers. Inducible deletion of both Hdac1 and Hdac2 resulted in the formation of enteroids with spheroid structures, suggesting alterations in cell growth and/or determination.

Conclusions. IEC-specific Hdac1 and Hdac2 regulate differently IEC and may alter IEC responses to the mucosal environment, leading to disturbances of intestinal homeostasis. HDAC inhibitors, which reduce intestinal inflammation severity and symptoms in mice, may well target IEC in addition to immune cells, thus controlling inflammatory responses.

Funding Agencies: CCC, CIHR


Anti-Tissue Transglutaminase Normalization Post Diagnosis in Children with Celiac Disease
D. Migliarese Isaac,1 S. Rajani,1 and J. Turner2

1Department of Pediatrics, University of Alberta, Edmonton, AB, Canada
2Department of Pediatric Gastroenterology & Nutrition, University of Alberta, Edmonton, AB, Canada

Background. Celiac disease (CD) is a common autoimmune enteropathy to gluten, leading to intestinal inflammation, villous atrophy, and malabsorption. Screening for CD screening involves anti-tissue transglutaminase (atTG) IgA levels, followed by intestinal biopsy for confirmation. A gluten-free diet (GFD) is required to alleviate symptoms, normalize atTG, and heal the intestinal mucosa in CD patients. Monitoring in CD includes following atTG titers post diagnosis. Variability exists in the literature regarding time to atTG normalization, with no studies examining the trend and predictors of atTG over time in the pediatric CD population.

Aims. We aimed to evaluate time to normalization of atTG in children post CD diagnosis, and examine factors impacting this time.

Methods. A retrospective chart review was completed to evaluate the rate of atTG normalization in pediatric CD patients attending the Stollery Pediatric Celiac Clinic from 2007 to 2014. Clinical predictors that may impact time to resolution were examined, including: initial atTG, GFD compliance (GFDC), age at diagnosis, gender, ethnicity, presenting symptoms, Marsh score at diagnosis, medical comorbidities, and family history of CD. Multivariate binary logistic regression was utilized to determine independent predictors of atTG normalization. Cox hazard regression analysis was then undertaken to determine predictors of time to normalization.

Results. 339 patients met the inclusion criteria. Mean age was 9 years at diagnosis (range 1–17 years), with 64% females. Patients were followed for 6 months to 6 years. 83% of patients normalized atTG levels within the study time period. Mean time to normalization was 484 days for all patients (range 30 to 2084 days), and 460 days for GFD compliant patients (range 81 to 1809 days). Type 1 diabetes melllitus (T1DM) and South Asian ethnicity were independent predictors of failure to normalize atTG (OR = 0.23, ; OR = 0.41, resp.), with T1DM patients being 4 times less likely to normalize. Conversely GFDC was a positive predictor of atTG normalization (OR = 7.0, ). Cox hazard regression demonstrated T1DM (HR = 0.5, ) and South Asian ethnicity (HR = 0.67, ) were predictors of longer time to atTG normalization. Patients with T1DM normalized atTG levels on average 240 days longer than those without.

Conclusions. There is a wide variation of rate and time to atTG normalization in children with CD. GFDC is the strongest predictor of early normalization. Patients with T1DM and South Asians are less likely to normalize atTG levels, and have longer time to normalization. This highlights the need for closer attention and education for these high-risk populations.

Funding Agencies: Women and Children’s Health Research Institute Resident Trainee Research Grant


The Effect of Sham Feeding on Small Bowel Transit Time in Patients Undergoing Capsule Endoscopy: An Interim Analysis
G. Ou,1 D. Prichard,2 C. Galorport,1 and R. Enns1

1University of BC, Vancouver, BC, Canada
2Mayo Clinic, Rochester, MN, USA

Background. Capsule endoscopy (CE) does not completely visualize the small bowel approximately 16.5% of the time due to limited battery life. We hypothesize that bacon-chewing is a potent stimulus of cephalic response, which may reduce capsule transit times and improve complete examination rate (CER).

Aims. To determine if chewing bacon improves CE transit times and CER.

Methods. A prospective, single-blinded, randomized controlled trial is in process at St. Paul’s Hospital in Vancouver, BC. Inclusion: outpatient CE, age ≥ 19. Exclusion criteria: inpatient, use of motility-enhancing or slowing drugs, active bowel obstruction, bowel resection, dysphagia, diabetes with end-organ damage, untreated hypo-/hyperthyroidism, and endoscopic placement of capsule. Given Imaging PillCam SB3 (Yoqneam, Israel) were used.

Randomization took place via concealed allocation per randomization sequence generated prior to study initiation. All subjects underwent bowel preparation (2 L) the day before, and fasted ≥2 h prior to the procedure. Clear fluids and normal diet were allowed 2 h and 4 h post-capsule ingestion, respectively. Immediately after swallowing the capsule, subjects in the bacon group were asked to chew 10 pieces of bacon, each over 30 seconds at one-minute intervals. This process was repeated an hour after ingesting the capsule. Gastric transit time (GTT), small bowel transit time (SBTT), and CER were compared between the groups.

This study was approved by institutional research ethics board and registered on (NCT02353208).

Results. From 01/2015 to 08/2015, 46 CE’s were included in the study, 42 of which were for GI bleeding, 2 for Crohn’s disease, 1 for abnormal CT finding, and 1 for recurrent intussusception. Female comprised 61% () of the cohort. Mean age was years. CE did not pass the pylorus in 2 patients (bacon group) during the recording, but did subsequently pass through the bowel; these were excluded from further analysis. Otherwise, there were no significant differences in GTT, SBTT, and CER.

Conclusions. The interim analysis does not demonstrate any change in capsule transit/completion with chewing bacon. The study will continue recruitment until the goal sample size of 122.

Funding Agencies: None


Mycophenolate Mofetil-Induced Immunosuppression Is Associated with a Wasting Phenotype, Altered Expression of Colonic Inflammatory Mediators and Reductioned Intestinal Microbial Diversity—Dissecting the Mechanisms of Gastrointestinal Dysfunction Triggered by Anti-Rejection Drugs
S. Pereira, A. Moffat, L. Alston, K. Rioux, P. Beck, M. Workentine, S. Greenway, and S. Hirota

University of Calgary, Calgary, AB, Canada

Background. Mycophenolate mofetil (MMF) is used post-transplant to prevent allograft rejection. In some cases, MMF is associated with GI complications and a wasting phenotype that necessitates treatment cessation.

Aims. To explore the mechanisms responsible for these issues, we developed and characterized a mouse model of MMF-induced immunosuppression.

Methods. Medicated feed (0.563% MMF) was administered to C57/Bl6 mice ad libitum for 9 days. In some experiments, mice were then removed from MMF treatment, returned to control chow and then sacrificed 7 days later. Intestinal tissues were removed for histological assessment, TUNEL staining (to assess apoptosis) and inflammatory mediator expression (via Luminex). Fecal pellets were collected throughout the course of the experiment and microbial composition assessed.

Results. Within 9 days of MMF treatment, we observed significant reductions in body weight, colon length, cecum weight, spleen weight and hematocrit. Upon cessation of MMF treatment, these parameters reversed to normal but splenomegaly developed. No differences in TUNEL staining were observed. Colonic tissue isolated at day 9 of treatment exhibited increased expression of inflammatory mediators including G-CSF, IFNγ, IL-6, IL-7, IL-10, IL-13, KC, LIF, MCP-1, MIP-1B, MIP-2, TNFα and VEGF. In contrast, IL-4 and IL-5 were significantly reduced. Characterization of the fecal microbiome revealed progressive alterations of bacterial populations correlating with duration of drug exposure with significant changes in composition accompanied by loss of diversity. Furthermore, phyla analysis revealed an expansion of Proteobacteria and Firmicutes, and a contraction of Bacteroidetes, in MMF-treated mice.

Conclusions. MMF treatment recapitulates human clinical symptoms in the mouse. The observed symptoms are associated with alterations in the intestinal immune status and reductions in intestinal microbial diversity. We will continue to characterize this model to better understand the mechanisms responsible for MMF-related complications. Furthermore, we intend to assess microbe-targeting interventions in future studies.

Funding Agencies: Canadian Foundation for Innovation; The Dr. Lloyd Sutherland Investigatorship in IBD/GI Research


Prediction of Esophageal and Gastric Histology by Macroscopic Diagnosis during Upper Endoscopy in Pediatric Celiac Disease
E. Boschee,1 J. Yap,1 and J. Turner2

1University of Alberta, Edmonton, AB, Canada
2Stollery Children’s Hospital, Edmonton, AB, Canada

Background. Celiac disease (CD) is the most common autoimmune enteropathy in children. Recent guidelines support diagnosis without biopsy in select pediatric patients, yet concerns exist over the potential for missing alternate tissue diagnoses. The frequency of endoscopic and histological abnormalities in intestinal sites other than duodenum in pediatric CD has yet to be studied.

Aims. The aim of the study was to determine the sensitivity of macroscopic appearance for predicting histology diagnosis at sites other than the duodenum. It was hypothesized that normal esophageal and gastric histology can be predicted by normal endoscopic appearance, obviating the need for routine biopsy from these sites.

Methods. Endoscopic and biopsy findings in patients diagnosed with CD at Stollery Children’s Hospital from 2010–2012 were retrospectively reviewed. The primary outcome was the diagnostic performance of endoscopic findings in predicting normal esophageal and gastric histology. A secondary outcome was the prevalence of other esophageal and gastric diagnoses.

Results. A total of 140 patients were included (61.4% female). The mean age at biopsy was 9.1 years, and the mean aTTG was 393.9. Endoscopic appearance was reported as normal in the esophagus and stomach in 84.8% and 87.7% of patients, respectively. Abnormal endoscopic esophageal diagnoses included eosinophilic esophagitis (EE) (5.8%), esophagitis (5.1%), glycogenic acanthosis (1.4%) and non-specific abnormalities (2.9%). Endoscopic gastric abnormalities were gastritis (6.5%), pancreatic rest (0.7%) and non-specific abnormalities (5.1%). Esophageal and gastric biopsies were taken in 54.3% and 77.9% of patients, respectively. Histology was normal in 77.6% of esophageal and 87.2% of stomach specimens. Abnormal esophageal histology included EE (10.5%), esophagitis (9.2%), glycogenic acanthosis (1.3%) and non-specific abnormalities (1.3%). Gastritis was reported in 12.8% of gastric specimens.

The sensitivity and specificity of normal macroscopic diagnosis for predicting normal esophageal histology was 86.2% and 61.1%, and for normal gastric histology was 88.3% and 26.7%. The positive predictive value of endoscopic diagnosis for predicting normal histology was 87.7% in the esophagus and 88.3% in the stomach.

Conclusions. This study suggests that, in the absence of macroscopic abnormalities, routine esophageal and gastric biopsy during endoscopy for pediatric CD is not needed. Endoscopic diagnosis is sufficiently sensitive to predict normal histology. These findings have cost and time saving applications for current clinical practice.

Funding Agencies: Women and Children’s Health Research Institute (WCHRI)


Inpatient Capsule Endoscopy Guides the Need for Device Assisted Enteroscopy in Patients with Suspected Small Bowel Bleeding
M. Alsahafi, P. Cramer, N. Chatur, and F. Donnellan

Division of Gastroenterology, Vancouver General Hospital, University of British Columbia, Vancouver, BC, Canada

Background. The availability of capsule endoscopy (CE) and device assisted enteroscopy has remarkably changed the management of patients with suspected small bowel bleeding. Device assisted enteroscopy is invasive, time consuming and associated with high health care cost. CE has several advantages including the non-invasive nature and the high diagnostic yield and as such it could be utilized to guide further intervention.

Aims. To assess the utility of inpatient CE to guide management and selection for device assisted enteroscopy in patients with suspected small bowel bleeding.

Methods. We retrospectively reviewed the hospital database for all patients requiring an inpatient CE between November 2011 and September 2015. Only Patients with suspected small bowel bleeding were included for analysis. Patients’ characteristics, decision and outcome after CE were analyzed including the diagnostic and therapeutic yield of device assisted enteroscopy.

Results. A total of 44 patients were included in the analysis. The mean age was 62. 72% of patients were male. The indication was overt bleeding in 41 and occult bleeding in 3 patients. The small bowel completion rate for CE was 75% () with retention rate of 2.2% (). Diagnostic yield was 63% (). Among patients who had small bowel pathology (), vascular lesions were the most common 59.3% (), followed by ulcer/erosion 22.2% (), mass lesion 3.7% () and blood of unknown origin 14.81% (). Of all patients with no identified small bowel pathology on CE, no patient required device assisted enteroscopy. Of all patients with positive small bowel pathology on CE (), 12 patients underwent device assisted enteroscopy: 11 single balloon enteroscopy and 1 retrograde enteroscopy with through the scope balloon catheter system (NaviAid, SMART Medical Systems). Among patients who underwent device assisted enteroscopy, the diagnostic yield was 66% (), all of which received therapeutic intervention: 7 had thermal coagulation and 1 had polypectomy.

Conclusions. Inpatient CE is a useful tool to guide the need for device assisted enteroscopy with a small bowel completion rate comparable to outpatient CE.

Funding Agencies: None


Constipation as the Primary Presentation of Celiac Disease
A. Sarkhy, Z. Zaidi, A. Saeed, and A. Assiri

Department of Pediatrics, King Khalid University Hospital, King Saud University, Riyadh, Saudi Arabia

Background. Celiac disease (CD) is a multifactorial, autoimmune disorder that occurs in genetically susceptible individuals. While typical CD presents with chronic diarrhea, steatorrhea, abdominal distention and failure to thrive, atypical CD may has subtle presentation and can be easily missed or overlooked. Constipation is one of the atypical presentations of CD and thus selected patients may need celiac screening for early diagnosis of the disease.

Aims. The aim of this study is to describe the frequency and clinical characteristics of children with CD patients who presented primarily with constipation.

Methods. This is a retrospective study conducted between the periods of January 2013 to June 2014, at King Khalid University Hospital, Riyadh, Saudi Arabia. We included children less than 18 years of age with confirmed diagnosis of CD based on serology and small bowel biopsies. We collected data about the clinical characteristics of the patients, details of the bowel movements, anthropometric measurements, treatment and the outcomes.

Results. 100 cases of confirmed CD were found during the study period with a mean age of 7 years (range; 9 months to 18 years). There were 68 males (68%). Fifty four percent presented with classical symptom, while 46% presented with atypical presentations. Constipation was found in 15 patients (15%). The mean duration of the constipation was 7 months (±3 months SD) before the diagnosis. Severe constipation (bowel opening once a week) was observed in 8 (53.3%) and rest had mild to moderate constipation. Mild to moderate abdominal pain and distension were observed in the severe constipated children. None of the children had a satisfactory response to laxatives therapy prior to the diagnosis of CD; however, all of them responded very well when they follow the strict GFD with total resolution of the symptoms.

Conclusions. CD should be kept in mind during the work up for patient with intractable constipation. Early diagnosis and early introduction of GFD improves the bowel habits among those patients.

Funding Agencies: None


Balloon-Assisted Enteroscopy Is an Emerging Tool in the Management and Maintenance of Crohn’s Disease
M. Reeson,1 S. Zepeda-Gomez,2 R. Fedorak,1 K. Kroeker,1 and B. Halloran1

1University of Alberta, Edmonton, AB, Canada
2University of Alberta, Edmonton, AB, Canada

Background. Seventy percent of CD patients suffer from small bowel involvement. Balloon-Assisted Enteroscopy (BAE) allows for improved visualization and for therapeutic intervention.

Aims. This study aims to assess the safety and efficacy of balloon-assisted enteroscopy in the management and maintenance of small bowel Crohn’s disease.

Methods. A retrospective chart review was undertaken to examine all patients with small bowel CD who underwent BAE over the last four years. The data collection included patient demographics, disease characteristics, surgical history, procedure characteristics, and stricture dilation data were all recorded.

Results. 68 BAEs were performed on 42 patients (23 male). The mean age at BAE was  yrs. 58 (85.3%) of procedures were on patients with known Crohn’s disease. 35 (83.3%) of patients had ileal CD, 4 (9.5%) had jejunal. 26 (61.9%) of patients had at least one previous bowel resection. Stricturing was the most common complication with 28 (66.7%) showing stenosis. 17 patients (12 male) underwent serial (>1) BAE procedures. 21 patients underwent 29 operations involving balloon-assisted stricture dilation. 66 strictures were dilated in total. Of all BAE procedures, 2 (2.9%) were met with complications; 1 patient developing pancreatitis, the other with minor bleeding. There were no perforations or deaths.

Conclusions. Balloon assisted enteroscopy and balloon-assisted stricture dilation are safe and effective procedures for visualization and treatment of small bowel Crohn’s disease and are likely under-utilized as an endoscopic tools. The combination of therapeutic intervention with direct and extensive mucosal visualization makes BAE the optimal modality for diagnosis and treatment of small bowel CD.

Funding Agencies: Self Funded (CEGIIR)


Magnetic Resonance Enterography in the Assessment of Post-Operative Crohn’s Disease
M. Scaffidi, K. Lim, P. Vlachou, G. Mnatzakanian, O. Zarghom, E. Irvine, E. Colak, and S. Grover

University of Toronto, Toronto, ON, Canada

Background. In Crohn’s disease (CD) post-ileocolic resection, the ileal component of the Rutgeerts score at colonoscopy prognosticates recurrent disease at the anastomosis, but is limited by potential inability to cannulate the neoterminal ileum due to stricturing, which occurs in >30% of cases. Magnetic resonance enterography (MRE) is a preferred imaging modality for small bowel CD, but its operating characteristics for predicting recurrent ileal disease post-ileocolic resection – for which post-surgical and inflammatory findings may be conflated – have not been clearly identified.

Aims. To determine the association of MRE findings of neoterminal ileal CD activity with Rutgeerts score at ileocolonoscopy.

Methods. This retrospective study enrolled patients with prior ileocolic resection for CD with MRE and colonoscopy performed within 90 days at a single academic centre. Exclusion criteria were poor bowel preparation or inability to reach the ileocolic anastomosis at the time of colonoscopy, or change in therapy between MRE and colonoscopy. Clinical information was collected to calculate Crohn’s disease activity index (CDAI). MRE studies were interpreted by two experienced abdominal radiologists blinded to clinical and endoscopic information. The primary outcome was association between Rutgeerts score and the following MRE findings: bowel wall thickness, mural edema, enhancement pattern, ulceration, stenosis, prestenotic dilatation, mesenteric hyperemia and mesenteric lymphadenopathy. A secondary outcome was association between MRE findings and CDAI score. The association between Rutgeerts and CDAI scores to MRE findings were determined using the exact Wilcoxon rank-sum test and Spearman correlation for categorical and continuous predictors, respectively.

Results. 25 patients who met the inclusion criteria were included in the analysis. MRE findings that had a significant positive correlation () with Rutgeerts score are: mural thickness (); mural edema (); ulceration (); mesenteric edema (); and mesenteric hyperemia (). We found no statistically significant association between mesenteric lymphadenopathy, luminal narrowing or fistulizing disease with the Rutgeerts score. There was a significant, positive correlation between mural thickness (), mesenteric edema () and reduced peristalsis () with CDAI.

Conclusions. In patients with ileocolic resection for CD, radiological characteristics from MRE correlate with known endoscopic-prognostic and disease activity scoring systems. Further research should codify these findings into an MRE rating tool to prognosticate recurrent CD post-ileocolic resection.

Funding Agencies: CAG


Successful Eradication of Recurrent Clostridium Difficile Infection (rCDI) of Small Bowel with Frozen Encapsulated Fecal Microbiota Transplantation (FMT) in a Patient with Crohn’s Disease and Ileostomy
J. Zhu, B. Roach, and D. Kao

University of Alberta, Edmonton, AB, Canada

Aims. We report a case of ileal-pouch Crohn’s disease with rCDI of small bowel who failed vancomycin treatment, and is successfully treated with frozen FMT from upper GI tract without any adverse events.

Methods. Case report.

Results. A 31 year-old male underwent subtotal colectomy and ileostomy in Sept 2013 for ulcerative colitis as he did not respond to infliximab. His post operative course was complicated by high grade small bowel obstruction, requiring multiple hospital admissions, and subsequently found to have Crohn’s involving neoterminal ileum. His maintenance therapy consisted of infliximab at 10 mg/kg q 4 weeks and methotrexate 25 mg SQ weekly. At baseline, he empties his ileostomy bag 4-5 times per day, each time about 250 cc of mushy stools. He developed his first episode of CDE in Jan 2014, during one of these post operative admissions. His stool C. difficile toxin was positive with no other enteric pathogens or alternative diagnosis identified. Ileoscopy revealed only mild patchy mucosal inflammation. He was treated with oral metronidazole 1 g daily for 10 days with symptom resolution. Unfortunately, his symptoms recurred within 2 weeks of discontinuing metronidazole. A repeat C diff toxin was again positive, and he responded well to a course of metronidazole. His symptoms recurred again within 2 weeks of discontinuing metronidazole, associated with positive C diff toxin again. He was then treated with a long tapered course of vancomycin, again with symptom resolution. Unfortunately, his diarrhea recurred shortly after discontinuing vancomycin. In total, he had 6 episodes of recurrent CDE between Jan 2014 and March 2015. He was referred to the Edmonton FMT Program for consideration of FMT in May 2015. He received encapsulated FMT, consisted of 30 capsules daily for 3 days, from one of the universal stool donors registered with the program. The patient reported having more formed stools in his ileostomy within the first week post FMT, and by week 3 his bowel habit had returned to baseline. He had no adverse events from FMT or rCDI during the follow-up period from May to Aug 2015.

There are few literatures on successful treatment of small bowel rCDI using frozen encapsulated FMT. Not only do IBD patients have an increased risk of developing CDI, but they can also develop CDI in the small bowel and ileal pouch-anal anastomosis (IPAA) following colectomy. Post operative mechanical complications, male gender and serum immunoglobulin G1 deficiency have been identified as risk factors for recurrent pouch CDI.

Conclusions. Frozen encapsulated FMT appeared to be a safe and effective therapeutic alternative for patients with small bowel rCDI, and warrants further investigation.

Funding Agencies: None


Efficacy and Safety of Over-the-Scope Clip (OTSC) in the Endoscopic Closure of Fistula and Perforation in the Gastrointstinal Tract: A Case Series
A. Yau, E. Lam, R. Enns, and F. Donnellan

University of British Columbia, Vancouver, BC, Canada

Background. Over-the-scope clip (OTSC) (Ovesco Endoscopy GmbH, Tübingen, Germany) is a novel device utilized in the management of fistula, perforation, dehiscence, and bleeding in the gastrointestinal tract via tissue approximation and compression.

Aims. To determine the efficacy and safety of OTSC in the endoscopic closure of fistula and perforation in the gastrointestinal tract.

Methods. A retrospective chart review was performed.

Results. Seven patients (mean age 62.9 years; 3 women (42.9%)) were treated with OTSC from 10/13 to 03/15 in an outpatient (42.9%) or inpatient (57.1%) setting and on an elective (14.3%), semi-elective (42.9%), or urgent (42.9%) basis.

The gastrointestinal diagnosis and treatment were nausea/vomiting with fistulizing percutaneous endoscopic gastrostomy tube (), duodenal ulcer perforation with failed Graham omental patch (), gastric cancer with total gastrectomy and leaking esophagojejunal anastomosis (), transverse colon cancer with left hemicolectomy and fistulizing primary anastomosis (), and rectosigmoid cancer with low anterior resection and leaking primary anastomosis ().

The OTSC was utilized in the endoscopic closure of gastrocutaneous fistula (), duodenal ulcer perforation (), jejunocutaneous fistula (), colocutaneous fistula (), and rectocutaneous fistula (). The defect size ranged from 2 to 10 mm. Technical success with defect closure was achieved completely in 62.5% (5/8 clips) and partially in 25.0% (2/8 clips). There were no complications related to OTSC application.

Additional interventions were hemoclips (), argon plasma coagulation (), sclerotherapy with histoacryl and lipiodol (), and hyperbaric oxygen (). Clinical success was achieved in 71.4% (). One patient required surgical resection of fistula for definitive management. Another patient died of persistent bleeding from anastomotic site.

Conclusions. The endoscopic application of OTSC appeared to be safe. The rates of technical success and long-term clinical success were satisfactory. Future prospective studies should compare the relative efficacy of OTSC to other endoscopic modalities in an effort to determine the most optimal indications and to maximize clinical outcomes.

Funding Agencies: None


Q Fever in a Patient with Crohn’s Disease on Adalimumab and Methotrexate
M. Alkhattabi,1 S. Hosseini-Moghaddam,1 A. AlNasser,1 and M. Beaton2

1Western University, London, ON, Canada
2London Health Sciences Centre, London, ON, Canada

Background. Q fever has been rarely reported in patients with inflammatory bowel disease (IBD) on immunosuppressive therapy.

Aims. To present a confirmed case of Q fever in a gentleman with Crohn’s disease (CD) and review the literature. The patient presented with fever of unknown origin who despite a lack of direct contact with zoonotic vectors, after an extensive evaluation he was eventually diagnosed and treated successfully for Q fever.

Methods. Case report and literature review.

Results. A 53-year-old automotive mechanic with a 30 year history of CD in remission with combination Adalimumab and Methotrexate since 2006. He was well until 2 weeks prior to his presentation when he developed a persistent fever and drenching night sweats. Over this period, he experienced a 5 lb weight loss but denied any symptoms suggestive of a flare of his underlying CD. His systemic review and physical examination were otherwise unremarkable. Initial investigations demonstrated a normal white blood cell count but significantly elevated CRP (121 mg/L). He was admitted to hospital and following acquisition of blood, stool and urine cultures, started on broad spectrum antibiotics. All cultures were negative and further evaluation demonstrated positive antinuclear antibody and rheumatoid factor, but negative viral, histoplasmosis and blastomycosis serologies. Imaging studies were unremarkable. WBC scan were negative. Gastroscopy and colonoscopy were normal. The infectious disease service was involved and requested Q fever serology which confirmed recent infection. He was started on a 10 day course of oral Doxycycline (200 mg every 24 hours) with resolution of his fever by day 3. Ongoing follow up with ID as an outpatient was arranged with serial monitoring of Q fever. Without discontinuation of treatment for CD, he continued treatment for Q fever. The process of improvement was not complicated by any significant event.

After obtaining further history, the patient was likely exposed through servicing vehicles used to transport sheep’s. Only one previous case of acute hepatitis due to Q fever in an IBD patient on chronic treatment with steroids has been reported.

Conclusions. To the best of our knowledge, this is the first reported case of acute Q fever in a known case of CD receiving Adalimumab and metothrexate. In spite of simultaneous immunosuppressive therapy, the patient did not develop any organ involvement which was reported in previous case report. This case report shows management of acute Q fever is successful despite continuing immunosuppression with biologic therapy.

Funding Agencies: None


The Effect of Longer Battery Life on Small Bowel Capsule Endoscopy: A Single Centre Experience
G. Ou, C. Galorport, and R. Enns

University of BC, Vancouver, BC, Canada

Background. Small bowel capsule endoscopy (CE) does not completely visualize the small bowel approximately 16.5% of the time due to limited battery life. Patients with incomplete and negative CE may require additional investigation to clear the rest of the small bowel. Previous study suggests that complete examination rate (CER) may be improved by simply extending the battery life and recording time.

Aims. To determine the CER and diagnostic yield of CE with longer battery life.

Methods. A prospective observational study is currently recruiting at St. Paul’s Hospital (Vancouver, BC). Inclusion criteria: age ≥ 19. Exclusion criteria: dysphagia, endoscopic placement of CE, active bowel obstruction, and concomitant enrollment in another study receiving motility-altering treatment.

All subjects underwent bowel preparation with 2L PEG3350 with electrolytes the day before, and fasted ≥ 2 h prior to the procedure. They were allowed to drink clear fluids and resume normal diet 2 h and 4 h post-capsule ingestion, respectively. Given Imaging PillCam SB3 (Yoqneam, Israel) were used.

Gastric transit time is defined as time of first duodenal image minus time of first gastric image. Small bowel transit time is defined as time of first cecal image minus first duodenal image; or time of last image recorded minus first duodenal image if CE did not reach cecum. CE is considered complete if colonic mucosa is demonstrated on recording.

This study was approved by institutional research ethics board and registered on (NCT02382705).

Results. From 02/2015 to 09/2015, 57 subjects, consisting of 29 (50.9%) male, were enrolled in the study. Mean age was years. The most frequent indication was GI bleeding (, 71.9%), followed by IBD (, 14.0%), polyposis syndrome/malignancy (, 8.8%), and history of obstruction (, 5.2%). Seven (12.3%) CE’s were incomplete, 5 of which were due to retention in the stomach for the duration of the recording. Twenty-four (42.1%) subjects had prior history of bowel resection proximal to the ileocecal valve. All 5 patients with gastric retention had previous bowel surgeries. Among the remaining 33 subjects, mean GTT was  min and SBTT was  min. Overall mean recording time was  min Endoscopic findings were identified in 20 (35.1%) CE’s.

Conclusions. CE with extended battery life appears to be a simple way to ensure high CER.

Funding Agencies: None


Control of Gut Homeostasis by Membrane Trafficking Regulators
S. Jean

Université de Sherbrooke, Sherbrooke, QC, Canada

Background. Approximately 233 000 Canadians suffer from Crohn’s disease or ulcerative colitis, which are collectively named inflammatory bowel diseases (IBD). It is now well established that IBD are multifactorial diseases occurring in individuals with specific genetic predispositions. The impact of the host genome on IBD susceptibility and progression or the host interaction with the gut microbiota is not well defined. Genome wide association studies (GWAS) have identified multiple genes linked to IBD, illuminating some of the genetic aspect of IBD. However, the effects of these variants and how they predispose individuals to IBD are for most of them unknown. The lack of a direct an easily amenable in vivo system to study gene functions in the context of intestinal cell biology, and intestinal interaction with microbes has precluded large scale studies of GWAS identified variants. Drosophila melanogaster represents an ideal system in which to perform such studies since its intestinal biology is generally conserved with humans, that multiple intestinal human pathogens can cause related pathologies in flies, that a large array of tools are available for loss and gain of function studies and that it represents an economical research model. In predisposed individuals, IBD development is triggered by poorly understood environmental and microbial factors. The combination of these factors eventually leads to chronic immune response in the gut, resulting in inflammation and high stress responses in intestinal cells. The membrane-mediated process of autophagy defined as the lysosomal degradation and recycling of cytoplasmic components, is an important stress response mechanism. Importantly, multiple genes regulating specific aspects of autophagy and membrane trafficking were identified as IBD risk variants in GWAS studies. Importantly, membrane and autophagy regulators were identified in Drosophila as important regulators of gut stem cell renewal.

Aims. To characterize the role of membrane trafficking regulators in autophagy and to understand their importance on gut homeostasis.

Methods. Here, I will describe the establishment of new tools to probe and understand the role of membrane regulators in IBD.

Results. I will show how Drosophila can be used as an accelerated screening tool for investigating the biological relevance of membrane trafficking IBD-related risk variants.

Conclusions. Drosophila represents a great and affordable model organism to identify and describe genes involved in gut homeostasis. Results obtains in fly will thus dictate our rationale in developing translational strategies in the context of conditional knockout mouse models as well as human patients.

Funding Agencies: CIHR, Startup funds, Université de Sherbrooke and Centre de Recherche du CHUS

Microbiology and Parasite-Host Interactions

Poster of Distinction


A TH17-Neutrophil Axis Involved in the Containment of Commensal Microbiota
K. Flannigan,1 V. Ngo,1 D. Geem,1 A. Harusato,1 S. Hirota,2 and T. Denning1

1Georgia State University, Atlanta, GA, USA
2University of Calgary, Calgary, AB, Canada

Background. At mucosal surfaces certain bacteria, including segmented filamentous bacteria (SFB) in mice, elicit the differentiation of T helper 17 (Th17) cells. While Th17 cells can potentiate IBD, they play an important role in barrier function by containing the microbiota. How Th17 cells achieve this is incompletely understood but involves, in part, the cytokine IL-22 produced in response to IL-23.

Aims. Recent evidence suggests neutrophils can also produce IL-22 to promote intestinal barrier protection, therefore we hypothesized that neutrophils may be involved in an axis with Th17 cells that can contain mucosal bacteria, specifically SFB, to protect mucosal surfaces.

Methods. SFB-free (SFB−) mice were obtained from Jackson and SFB+ mice were obtained from Taconic. Feces from SFB+ mice were mixed into drinking water and given to mice for 7 days. Small intestinal sections were digested in collagenase to isolate lamina propria cells for flow cytomertry. Monocloncal antibodies were used to block IL-17A or deplete CD4+ T cells and neutrophils (αLy6G). SFB loads were detected via qPCR performed on fecal DNA. Ileal explants were taken using biopsy needles and stimulated with IL-23.

Results. After colonizing SFB− mice with SFB-containing microbiota, we observed an influx of neutrophils into the ileum correlating with the load of SFB over the course of 7 days. Neutrophil numbers also correlated with IL-17A and IL-22 expression. Based on studies suggesting a role for IL-17A in neutrophil recruitment we injected SFB- mice with αIL-17A antibody. 7 days post-SFB colonization αIL-17A-injected mice had lower levels of neutrophils in the ileum compared to isotype treated mice. This same effect was observed after αCD4 injection in mice with SFB-containing microbiota, suggesting CD4+ T cells as a source of IL-17A. We next utilized αLy6G to deplete neutrophils and found that after 7 days of colonization there was a significant expansion in SFB loads compared to isotype treated mice. The expansion of SFB was also seen after αIL-17A treatment or CD4 depletion. Both αLy6G and αIL-17A treated mice also had higher levels of Th17 cells in the ileum. Finally, incubation of ileal explants with IL-23 promoted robust IL-22 production along with an increase in the expression of the antimicrobial peptides SA100A8, SA100A9, and RegIIIα; all effects that were lost after neutrophil depletion.

Conclusions. After introduction of SFB, intestinal CD4+ T cells produce IL-17A to recruit neutrophils. Upon entering the intestine neutrophils are exposed to IL-23 prompting IL-22 production, which induces expression of antimicrobial peptides to help control SFB. Without proper neutrophil responses, SFB levels expand, increasing Th17 cell number in the intestine that may potentiate disease.

Funding Agencies: NIH, Snyder Institute for Chronic Diseases, Cumming School of Medicine, None

Poster of Distinction


Adult Nod2−/− Mice Show an Altered Microbial Resilience and an Increased Fecal IGA Response following Antibiotic Perturbation of the Microbiota
A. Goethel,1 W. Turpin,1 S. Rouquier,1 G. Zanello,1 D. Philpott,1 and K. Croitoru2

1University of Toronto, Toronto, ON, Canada
2Mount Sinai Hospital, Toronto, ON, Canada

Background. Inflammatory bowel diseases (IBD) are multifactorial diseases, involving genetic mutations, environmental triggers and alterations of the gut microbiota. The strongest genetic association is with nucleotide-binding oligomerization domain-containing protein 2 (NOD2), a pattern recognition receptor that recognizes a component of the bacterial cell wall. During adulthood, environmental perturbations, such as antibiotics, induce transient shifts in the microbiota composition.

Aims. We sought to determine whether Nod2 played a role in antibiotic-induced dysbiosis and/or resilience of the microbiota. Moreover, we investigated whether the post-antibiotic microbiota in a Nod2-deficient mouse could lead to an altered immune response.

Methods. Adult WT and Nod2/ mice received control water or amoxicillin (200 mg/L) ad libitum in drinking water for 7 days, followed by control water for 4 weeks. Fecal samples were collected weekly to monitor changes in the microbial community structure using 16S sequencing. Fecal IgA was used to assess the mucosal immune response to the microbiota over time. On day 35, small intestinal mucosal damage was induced by acute polyclonal T cell activation following i.p. anti-CD3 injection.

Results. Bacterial load, measured using targeted qPCR for the beta-subunit of bacterial RNA polymerase (rpoB), was significantly reduced at day 7 in antibiotic-treated mice; coinciding with a significant shift the in the microbial community structure of both WT and Nod2/ mice. Antibiotic treatment resulted in a significant reduction of Firmicutes and a bloom of Proteobacteria at day 7. By day 21 (two weeks after removal of antibiotics), WT mice had returned to their pre-treatment community structure, however Nod2/ mice retained a significantly different microbiota from pre-treatment, highlighted by a significant reduction in α-diversity. Fecal IgA was significantly enhanced at day 7 in antibiotic-treated Nod2/ mice; returned to normal at day 21. Water-treated Nod2/ mice had a significantly enhanced IL-17 response following i.p. anti-CD3 injection, which was reduced by antibiotic treatment.

Conclusions. Our data suggests that Nod2 plays a significant role in resilience of the gut microbial community structure following perturbation. Moreover, antibiotic alteration of murine adult microbiota in Nod2/ mice leads to an altered immune response to the commensal gut microbiota.

Funding Agencies: CAG, CIHR, OGS and Mount Sinai Hospital

Poster of Distinction


A Prospective, Dual Center, Randomized Trial Comparing Colonoscopy versus Capsule Delivered Fecal Microbiota Transplantation (FMT) in the Management of Recurrent Clostridium Difficile Infection (RCDI)
D. Kao,1 B. Roach,1 N. Hotte,1 M. Silva,2 K. Madsen,1 P. Beck,2 and T. Louie2

1University of Alberta, Edmonton, AB, Canada
2University of Calgary, Calgary, AB, Canada

Background. FMT is highly effective in the treatment of RCDI. Various delivery mechanisms have described, resulting in success rate of 80–90%. However, the ideal route of delivering FMT has not yet been determined.

Aims. To compare the efficacy, safety, patient preference, and cost between colonoscopy versus encapsuled FMT.

Methods. This prospective, dual center trial randomized patients with RCDI in Edmonton and Calgary, Alberta, to FMT by oral capsule or colonoscopy at 1 : 1 ratio. Inclusion criteria were: (1) age > 18 and (2) at least 3 episodes of CDI. Exclusion criteria included: (1) complicated CDI; (2) chronic diarrhea; (3) chemotherapy/radiation; (4) dysphagia; (5) ileus or small bowel obstruction; (6) colostomy or ileostomy. Seven universal stool donors registered with our FMT programs provided stool materials. The primary objective was to compare the cure rate of RCDI by FMT delivered by capsules versus colonoscopy. The secondary objectives included: (1) safety of FMT by each delivery modality; (2) patient preference and satisfaction; (3) cost difference. There was a significant increase in quality of life after FMT. The cost of capsule was significantly lower than colonoscopy FMT.

Results. A total of 43 patients had been randomized to date, with 22 in capsule group and 21 in colonoscopy group. Patient baseline characteristics were shown in Table 19. The cure rate was 100% in colonoscopy and 92% in capsule group following one treatment. There were no fevers, infections attributable to FMT, or colonic perforation. One patient had mild nausea and vomiting from sedation following colonoscopy. One patient had nausea and vomiting following capsule ingestion. Only 7% of patients found the idea of FMT by capsules unpleasant, compared to 24% by colonoscopy.

Conclusions. FMT by either colonoscpy or capsules appeared to be similar in efficacy. Significant improvement in quality of life was seen in both groups of patients. No significant adverse events had occurred in any patients in either group. More patients preferred FMT by capsules to colonoscopy. Capsule administered FMT was less expensive and required less resources compared to colonsocopy.

Funding Agencies: CAG, Alberta Health Services

Poster of Distinction


Independent Predictors of Failure of Fecal Microbiota Transplantation (FMT) for Recurrent or Refractory Clostridium Difficile Infection
J. Dimitry, A. Keshteli, and D. Kao

Department of Medicine, University of Alberta, Edmonton, AB, Canada

Background. Fecal microbiota transplantation (FMT) is a safe and effective alternative therapy for treatment of refractory or recurrent C. diff infection (RCDI), with an overall success rate of 80–90% with one treatment. However, it is unknown which patients will require more than one FMT or who will not respond to FMT.

Aims. Our goal was to identify clinical predictors of treatment failure following a single FMT in recurrent or refractory disease.

Methods. This retrospective study included 136 patients who received FMT for refractory or recurrent CDI at the University of Alberta hospital. Patient baseline characteristics and clinical datasets were abstracted from in-patient charts and electronic medical records.

Results. Of the 136 patients, 106 (77.9%) were cured following one FMT. Univariate analysis identified 7 factors, shown in Table 20, associated with failure of 1 FMT. Multivariate analysis revealed that inpatient status (OR 7.4; 95% CI 2.2–24.6, ), immunosuppression (OR 4.5; 95% CI 1.3–16.2, ), and severe or complicated CDI (OR 5.2; 95% CI 1.2–21.6, ) were factors most independently associated with failure following 1 FMT. All 9 refractory patients failed the 1st FMT. Of the 30 patients (22.1%) who failed the first FMT, 25 received a second and 16 were cured.

Conclusions. Inpatient status, immunosuppression, and severe or complicated disease are independent predictors of treatment failure following 1 FMT in RCDI patients. Clinicians can predict patients most at risk of treatment failure and adjust treatment and discharge planning accordingly.

Funding Agencies: CAG

Poster of Distinction


Giardia muris Protects against Citrobacter rodentium Infection: Effects on the Inflammosome and Gut Antimicrobial Peptides
A. Manko,1 J. Motta,1 J. Cotton,1 A. Oyeyemi,1 B. Vallance,2 J. Wallace,3 and A. Buret1

1University of Calgary, Calgary, AB, Canada
2BC Children’s Hospital, Vancouver, BC, Canada
3Antibe Therapeutics Inc., Toronto, ON, Canada

Background. Our knowledge of polymicrobial gastrointestinal infections effects on host innate immune responses remains incomplete. The diarrheagenic protozoan parasite Giardia sp. alters gut microbiota and dampens host inflammatory responses. The production of intestinal epithelial anti-microbial peptides (AMPs) and activation of canonical and non-canonical inflammasomes play essential roles in host defense against various enteropathogens. Citrobacter rodentium infection in rodents is a model of Enteropathogenic Escherichia coli (EPEC) and Enterohemorrhagic E. coli (EHEC) in humans. In a study perfomed in Tanzania, Giardia-infected children seemed to be protected from diarrhea via unknown mechanisms.

Aims. The aim of this study was to determine if Giardia muris infection protects against C. rodentium colitis by enhancing AMPs production and/or activation of the inflammasome, and its key components caspase-1 and caspase-11.

Methods. Male C57BL/6 mice were co-infected with green fluorescent protein (GFP)-labeled Citrobacter rodentium, and Giardia muris trophozoites. Weight gain, fecal blood, histopathology, and colonic C. rodentium colonization were recorded. GFP-Citrobacter rodentium, β-Defensin-2 and TFF-3 AMPs were visualized by immunofluorescence. Colonic caspase-1 and -11, and TFF3 and β-Defensin-2 protein levels were assessed by Western blot.

Results. At 14 days post-infection, mice infected with C. rodentium had an inflamed colon, lost weight and passed fecal blood. Co-infection with G. muris significantly inhibited these effects. In Giardia co-infected animals, mucosal adhesion and translocation of C. rodentium were reduced. Caspase-1 and -11 protein levels were increased in the co-infected group compared to animals given C. rodentium alone, suggesting Giardia activates canonical and noncanonical inflammasome pathways. In the colon of mice co-infected with Giardia, concentration and expression of β-Defensin-2 and TFF3 proteins were increased.

Conclusions. The findings indicate that Giardia upregulates AMP production and inflammasome pathways, and protects against C. rodentium colitis. The findings shed new light on how microbial-microbial interactions may protect children against bacterial enteritis in zones endemic for giardiasis.

Funding Agencies: CIHR

Poster of Distinction


Afuabc Is a Phosphorylated Carbohydrate Transporter That Promotes Intestinal Colonization and Transmission of Citrobacter rodentium
S. Crowley,1 B. Sit,2 K. Bhullar,1 C. Lai,2 C. Tang,2 Y. Hooda,2 C. Calmettes,2 H. Khambati,3 C. Ma,1 J. Brumell,4 A. Schryvers,3 T. Moraes,2 and B. Vallance5

1University of British Columbia, Vancouver, BC, Canada
2University of Toronto, Toronto, ON, Canada
3University of Calgary, Calgary, AB, Canada
4Hospital for Sick Children, Toronto, ON, Canada
5BC Children’s Hospital, Vancouver, BC, Canada

Background. Attaching and effacing (A/E) pathogens cause severe diarrheal disease in their respective hosts. In order to colonize the mucosal surfaces of intestinal epithelial cells, these bacterial pathogens, such as enterohemorrhagic Escherichia coli (EHEC) and its murine-specific relative Citrobacter rodentium, must scavenge extracellular nutrients required for virulence from the metabolically competitive environment of the intestinal lumen.

Aims. Here, we investigated the function of the bacterial AfuABC transport system and its role in C. rodentium pathogenesis.

Methods. This study employed a variety of biochemical and microbiological techniques as well as a three animal infection models to elucidate the role of AfuABC.

Results. By crystallographic and biochemical approaches, we demonstrate that AfuABC is a cyclic hexose/heptose-phosphate transporter with high specificity and selectivity for glucose-6-phosphate, fructose-6-phosphate and sedoheptulose-7-phosphate. AfuABC is present in both EHEC and C. rodentium, and located adjacent to the putative pathogenicity island OI-20 which regulates carbohydrate detection and acquisition. A competitive index assay, simultaneous infection of C. rodentium wildtype and ΔafuA to measure comparative fitness, revealed ΔafuA was significantly impaired in both shed stool (6 days post infection, d.p.i.) and colonic tissues (10 d.p.i.). Interestingly, C. rodentium ΔafuA did not display any significant difference from wildtype in either in vitro and in vivo single infections, type three effector secretion, bacterial tissue localization or colonic pathology. However in a transmission model, measuring C. rodentium transmission to naïve mice via shed stool, the ability of ΔafuA to transmit and establish colonic disease was significantly decreased compared to wildtype. Sugar-phosphates were detected by targeted tandem liquid chromatography/mass spectrometry in both normal and infected intestinal mucus and stool samples, indicating these nutrients are available for enteric bacteria to metabolize during infection.

Conclusions. These results indicate that AfuABC-dependent uptake of sugar-phosphates impacts the ability of C. rodentium to transmit to new hosts and uncovers a previously unrecognized role for these metabolites as important contributors to successful pathogenesis.

Funding Agencies: CAG, CIHR

Poster of Distinction


Human Microbiota Modulates Intraepithelial Lymphocyte Numbers and Phenotype in the Mouse Small Intestine: Implications for Chronic Intestinal Inflammation
J. McCarville,1 M. Meisel,2 H. Galipeau,1 Y. Sanz,3 B. Jabri,2 and E. Verdu1

1McMaster University, Hamilton, ON, Canada
2University of Chicago, Chicago, IL, USA
3Microbial Ecology, Nutrition & Health Research Group, Institute of Agrochemistry and Food Technology, National Research Council (IATA-CSIC), Valencia, Spain

Background. Intraepithelial lymphocytes (IELs), αβTCR and γδTCR, play a critical role in mucosal barrier maintenance and as demonstrated by some gnotobiotic studies, their populations are modulated by commensal colonization. Previous research has demonstrated that IELs have increased cytolytic activity in Crohn’s patients, are responsible for infection-associated damage in the small intestine, and are critical for the development of villous atrophy in celiac disease (CeD). The factors that lead to increased numbers or cytotoxic transformation of IELs described in chronic intestinal inflammatory disorders such as Crohn’s and CeD are unclear. Interestingly both disorders are associated with alterations in the gut microbiota and with presence of pathobionts in the intestine.

Aims. To investigate whether the microbiota from CeD patients, influence IEL numbers and phenotype in the small intestine of gnotobiotic mice.

Methods. We transferred human fecal bacteria from patients with active CeD (collected anaerobically at diagnosis) or healthy individuals into germ-free (GF) C57BL/6 mice.

Results. Colonization with microbiota from a HLA-DQ8 heterozygous donor (CeD2), markedly increased IEL numbers within small intestine villi tips compared to colonization with microbiota from the other donors. The phenotype of IELs and inflammatory status of the small intestine was also affected by CeD2 microbiota, with increased proportions of CD3+CD8+αβTCR+IELs, and increased small intestinal IFN-γ expression. Sequencing of small intestinal contents in CeD2 recipient mice revealed a higher composition of Proteobacteria, specifically Parasutterella. Further investigation by colonization of MyD88−/−; Ticam1−/− mice demonstrated that IEL responses to CeD2 colonization were dependent on these signaling adaptors.

Conclusions. These results suggest that the composition of the gut microbiota can alter IEL numbers and phenotype in the small intestine through a MyD88/TICAM1 pathway, leading to IFN-γ induction. In a subgroup of patients with genetic predisposition, presence of pathobionts may promote cytotoxic transformation of IELs and contribute to disease pathogenesis. This mechanism has important implications for CeD, and IBD.

Funding Agencies: CIHR, Canadian Celiac Association (CCA)

Poster of Distinction


Antibiotics Administration Alters the Fecal Microrna Signature of Healthy Subjects
A. Taibi,1 S. Girard,2 T. Tompkins,2 and E. Comelli1

1University of Toronto, Toronto, ON, Canada
2Lallemand Health Solutions, Montréal, QC, Canada

Background. Fecal microRNA (miRNA), likely originating from intestinal cells, are an emerging biomarker of gut health. In humans, selected fecal miRNA were found to be associated with different stages of colorectal cancer. In animals, we and others found that the presence of the gut microbiota is associated with a distinctive intestinal miRNA signature. Moreover, the gut microbiota also modifies the intestinal miRNA response to pathogenic infection. It is likely that alteration of the gut microbiota composition, such as via antibiotics, will elicit a miRNA response in the host intestine which can be monitored in the feces.

Aims. To investigate the effect of antibiotic administration on the fecal miRNA signature.

Methods. Eleven healthy subjects received 875 mg of amoxicillin and 125 mg of clavulanic acid twice a day for 7 days. Total RNA was extracted from their feces, collected before and after antibiotic treatment, and used to profile the expression of 829 miRNA with the nCounter human version 3 miRNA expression assay (NanoString Technologies). Data were analyzed with nSolver 2.5; statistics and hierarchical clustering were performed using R. Data were validated by quantitative PCR.

Results. 700 miRNA were detected in the feces at baseline. Of these, 48 were differentially expressed in response to antibiotics (), 7 being under-expressed and 41 being over-expressed (0.2 to 4.3 fold change). Samples obtained before and after antibiotic treatment clustered separately based on the expression profile of these 48 miRNA. MiR-378b, whose expression was decreased by antibiotics, is a member of the miR-378 family previously found to depend on the gut microbiota.

Conclusions. This study shows that antibiotic treatment, at a dosage previously shown to change the composition of the microbiota, alters the fecal miRNA signature. It is likely that this results from microbial dysbiosis. This study may help understand the role of the microbiota in diseases that are accompanied by altered miRNA expression and microbiota composition, including inflammatory bowel disease and colorectal cancer.

Funding Agencies: NSERC, Lallemand Health Solutions

Poster of Distinction


The Serotonergic Endocrine- Gut Microbiota Axis in Experimental Colitis
E. Denou,1 J. Ghia,2 H. Wang,1 J. Kim,1 M. Shajib,1 M. Shah,1 M. Surette,1 S. Collins,1 and W. Khan1

1Farncombe Family Digestive Health Research Institute, McMaster University, Hamilton, ON, Canada
2University of Manitoba, Winnipeg, MB, Canada

Background. Serotonin, also known as 5-hydroxytryptamine (5-HT), is predominantly secreted in the gut by enterochromaffin (EC) cells. Tryptophan hydroxylase 1 (Tph1) catalyzes the synthesis of 5-HT in EC cells, and we have recently shown that in Tph1−/− mice which have significantly reduced 5-HT in the gut exhibit reduced severity of colitis. Conversely, replenishing gut 5-HT increased the severity of colitis. Due to strategic location of EC cells in gut mucosa, we hypothesize that 5-HT contributes to gut homeostasis by influencing its microbial composition.

Aims. To elucidate the role of 5-HT in regulation of gut microbiota in the context of experimental colitis.

Methods. We used PCR amplification of the v3 region of the 16S rDNA gene sequences in combination with deep Illumina sequencing to assess the composition of fecal microbiota in Tph1−/− and heterozygous Tph1+/− mice housed in the same cage. In vitro growth curves of 13 isolated commensals in presence of various concentrations of 5-HT was measured by optical density at 650 nm. Adoptive transfer was done using cecal microbiota from Tph1+/− mice gavaged to Tph1−/− mice and vice-versa beginning 2 days before colitis induced by 5% DSS. The extent of DSS-colitis was assessed by investigating macroscopically and histological damage scores, myeloperoxidase (MPO) activity, C-reactive protein (CRP) and cytokines levels.

Results. We observed the following differences in microbial composition between Tph1−/− and Tph1+/− mice. Microbiota of Tph1−/− mice is characterized by prevalence of Desulfovibrio and lower relative abundance of Parabacteroides, Erysipelotrichaceae and Turicibacter compared to Tph1+/− microbiota. In a species-dependant manner, 5-HT at low concentration (below 0.01 mg/L) stimulates in vitro bacterial growth but carries out a bacteriostatic effect at higher concentration (from 1 mg/L). Transfer of microbiota from Tph1+/− to Tph1−/− mice prior to DSS colitis significantly increased the histological score, the levels of CRP, MPO, IL-1β, IL-6, and IL-17. In contrast Tph1+/− mice that received microbiota from Tph1−/− mice showed a significant decrease in these inflammatory markers.

Conclusions. These results demonstrate that 5-HT influences the microbial composition of the gut and alters susceptibility to DSS-colitis. These data generate novel information on the interaction between 5-HT and microbiota in relation to pathogenesis in colitis and may lead to new therapeutic strategies for gut inflammatory disorders.

Funding Agencies: CCC, CIHR


MUC2 Mucin and Microbiota Play Distinct Roles in Conferring Innate Host Defense and Susceptibility to Colonic Injury
A. Leon-Coria,1 K. Chadee,1 F. Moreau,1 and M. Surette2

1University of Calgary, Calgary, AB, Canada
2McMaster University, Hamilton, ON, Canada

Background. The human colon has approximately 1014 microorganisms that form the microbiota which plays a key role in digestive health and in the development of a balanced immune system. The intestinal epithelium is covered with bilayer MUC2 mucin that acts as a protective barrier and is constantly exposed to commensals and pathogenic microorganisms. Both dysbiosis (alteration in microbiota composition) and disruption of the mucus layers are associated with gastrointestinal pathological conditions such as IBS, IBD and colorectal cancer. However, the distinct contribution of the microbiota and the mucus barrier in the pathogenesis of colitis is not well understood.

Aims. (1) To determine whether shifts in microbiota can alter the susceptibility, onset, progression or recovery of colitis in Muc2+/+ and Muc2/ mice. (2) To quantify the distinct roles of an intact Muc2 barrier and the microbiota in susceptibility to colonic injury.

Methods. To minimize variations in microbiota only Muc2+/+ and Muc2/ littermates were used. To quantify the role of the microbiota in disease pathogenesis, animals were treated with a cocktail of antibiotic (Ab) to eliminate indigenous bacteria as revealed by llumina sequencing DNA analysis, and then fecal transplanted with littermate stool and susceptibility to dextran sulphate sodium (DSS) quantified. To determine early innate responses, untreated and Ab-treated Muc2+/+ and Muc2/ animals were infected with the colonic parasite Entamoeba histolytica (Eh) in closed colonic loops. Chemokines, cytokines and other markers were measured by qPCR and luminex. Intestinal permeability was assessed using FITC-dextran.

Results. DNA analysis showed that Muc2+/+ and Muc2/ mice shared similar phyla distribution regardless of their distinct phenotype. Surprisingly, Muc2+/+ mice that received Muc2/ microbiota were highly susceptible to DSS-induced colitis with increased colonic permeability, ulcerated lesions and loss of crypts architecture, increased inflammatory cellular infiltrate and mortality as compared to Muc2+/+ receiving their own microbiota. Muc2/ mice showed no increase/decrease in disease susceptibility receiving their own or Muc2+/+ microbiota. In colonic loops, Eh in Ab-treated mice showed significant increase in watery secretions, luminal pro-inflammatory cytokines (IFN-γ and TNF-α) and chemokines (MCP-1, MIG, MIP-1a, MIP-1b and MIP-2) as compared to untreated controls.

Conclusions. Muc2/ microbiota alone could confer increased susceptibility to DSS-induced colitis in Muc+/+ mice with an intact Muc2 layer. Indigenous microbiota in Muc2/ animals play an important role in mediating host protection against Eh-induced pro-inflammatory responses in the absence of an intact Muc2 layer.

Funding Agencies: CCC


Human Microbiota Induces Barrier Homeostasis in Germ-Free Mice
C. Hayes,1 J. Jury,1 J. Dong,2 A. Naidoo,1 J. Libertucci,1 D. Bowdish,1 and E. Verdu1

1McMaster University, Hamilton, ON, Canada
2McMaster University, Burlington, ON, Canada

Background. The intestinal barrier involves various immune and permeability functions that are regulated throughout life by cytokines, diet, drugs, pathogens and their toxins. Commensal colonization induces antimicrobial peptide and IgA secretion in the gut; however, the kinetics for induction of functional aspects of the barrier that are key to maintaining intestinal homeostasis remain unclear.

Aims. Our aim was to investigate the sequence of immune and functional barrier changes induced by colonization of germ-free mice with human bacteria.

Methods. Adult male and female germ-free (GF) C57BL/6 mice were colonized by oral gavage with fecal slurry derived from a healthy human. At days 1, 7 and 21 following colonization, translocation of live bacteria to the spleen was assessed by culture, and NFκB-SEAP reporter assays were performed to evaluate the presence and immunostimulatory capacity of serum lipopolysaccharide (LPS) and muramyl dipeptide (MDP). At the same time points, paracellular and transcellular permeability of the proximal colon were evaluated by Ussing chambers with 51Cr-EDTA and HRP probes, respectively, and RNA expression of intestinal barrier genes were quantified using a custom designed NanoString Elements panel. GF and conventionally raised specific pathogen free (SPF) mice were used as controls.

Results. At day 1 post-colonization, <104 cfu/mL were detected in the spleen, and NFκB activation by serum LPS and MDP was higher following colonization compared to GF mice. Paracellular permeability was increased to a level comparable to SPF mice by day 7 of colonization and sustained by day 21. No significant changes in transcellular permeability were observed post-colonization. RNA sequencing identified a number of expression changes of barrier functional, innate immune and neuronal signaling proteins; most notably, Pparg and Gpbar1 were decreased only at day 1 post-colonization, while Nod2 was increased at day 7 post-colonization compared to GF mice.

Conclusions. Upon colonization, a rapid increase in systemic exposure to bacteria and their products occurs, which may be attributable to diminished expression of Nod2 and barrier fortifying proteins Pparg and Gpbar1 at day 1 post-colonization. Induction of paracellular permeability, comparable to that seen in conventional SPF mice, is observed by day 7 post-colonization, once initial systemic translocation is controlled. These findings suggest immune detection precedes functional barrier changes characteristic of a colonized status.

Funding Agencies: CCC


IL-22 Inhibits IL-10 And IL-25 Production and Attenuates Helminth-Inducedsuppression of Colitis in Mice
J. Reyes,1 F. Lopes,1 M. Fernando,1 G. Leung,1 C. Matisz,1 A. Wang,1 and D. McKay2

1University of Calgary, Calgary, AB, Canada
2Uni. Calgary, Calgary, AB, Canada

Background. Awareness of the precise immunological basis of host-parasite interactions has the potential to reveal new approaches to cure and/or treat inflammatory disease. We have shown that mice infected with the tapeworm Hymenolepis diminuta (HD) developed significantly less severe dinitrobenzene sulphonic acid (DNBS)-induced colitis, that was accompanied by a TH2 response, presumably driven by IL-25, and was mediated, in part by IL-10.

Aims. In ongoing analyses of this helminth-mouse model system we assessed the effect of IL-22, a cytokine associated with anti-microbial responses in the gut, in HD-induced protection of DNBS-induced colitis.

Methods. Colitis was induced in wild-type (WT) and IL-22 knock-out (KO) deficient mice by DNBS (5 mg, ir) +/− HD infection (5 cysticercoids), and disease assessed 72 h later by colon length, weight loss, a cumulative disease activity score and histopathology. Cytokinelevels were assessed by q-PCR and ELISA. HD were cultured onthe small intestine murine epithelial cell line, IEC-4, and IL-25 was measured.

Results. IL-22KO HD-infected mice displayed a slightly delayed, but enhanced TH2 response as gauged by MLN and gut cytokines, and increased IL-10 and IL-25 production. In vitro studies showed that HD directly evoked IL-25 from gut epithelia and this was reduced by addition of recombinant IL-22. The IL-22 KO mice had less DNBS-induced colitis compared to WT mice, and when infected with HD and challenged with DNBS, IL-22 KO mice had negligible disease. Finally, anti-IL-25 antibody treatment of IL-22 KO (+/− HD) resulted in more severe DNBS-induced colitis.

Conclusions. IL-22 is identified as an endogenous inhibitor of helminth-elicited TH2 immunity in the gut (i.e. reciprocal IL-22-IL-25 signalling), which reduces the effectiveness of HD-induced protection against colitis. We speculate that helminth therapy for inflammatory disease, under specific conditions, could be enhanced by inhibition of IL-22.

Funding Agencies: CIHR, NSERC


Culturing Techniques for Primary Intestinal Epithelial Cell Monolayers and Their Use in the Study of Enteric Infection and Immunity
M. Stahl1 and B. Vallance2

1University of British Columbia, Vancouver, BC, Canada
2BC Children’s Hospital, Vancouver, BC, Canada

Background. The tool of choice for studying the cellular mechanisms of intestinal epithelial cells in vitro has always been immortalized cell lines. These cells grow quickly and reliably, but can be problematic as they may not accurately represent the intestinal epithelium, and they cannot represent the genetic and phenotypic diversity found amongst human patients or animal models. More recently, the cultivation of intestinal organoids (enteroids) has become a popular tool for cultivating primary intestinal epithelial cells from either mouse or human origin. This technique centers on the ability to replicate the stem cell niche and induce intestinal stem cells to proliferate and differentiate in vitro. The resulting enteroids can be passaged almost indefinitely, and develop into a structure similar to the intestinal epithelium. Unfortunately, their structure and growth requirements make them complicated to use effectively for certain applications.

Aims. We aim to employ a novel method for cultivating primary intestinal epithelial cells, derived from enteroid cultures, as a means of studying host-pathogen interations and innate immune responses of the intestinal epithelium.

Methods. By dissociating the cells of an enteroid culture and plating them onto a matrigel-coated culture well, coverslip or hanging well insert, we can induce the cells to adhere and proliferate to form a confluent monolayer, similar to conventional cell cultures. We have begun using these primary colonic epithelial cell monolayers as a novel model for studying the interactions between enteric pathogens and intestinal epithelial cells. We infected cells with the common intestinal pathogen Campylobacter jejuni and have been comparing its interactions with primary mouse colonic epithelial cell monolayers against its interactions with conventional intestinal cell lines.

Results. There are both similarities and differences in patterns of cell adhesion and invasion between primary cells and conventional cell cultures, as well as the cytokine responses from the cells. Importantly, we have found that C. jejuni does not attack an epithelial monolayer evenly, but rather invades in large numbers through any breach in the cell juntions, while leaving an intact monolayer largely unscathed. Furthermore, when comparing monolayers derived from wild-type or mice deficient in Single Ig IL-1-related receptor (Sigirr/), we found a substantial decrease in C. jejuni cell invasion, suggesting a role for Toll-like receptors in cell resistance to C. jejuni invasion.

Conclusions. We can successfully induce primary intestinal epithelial cells to grow into a confluent monolayer, and our current work with C. jejuni suggeats that they may be a useful tool in identifying novel pathogen-epithelial cell interactions.

Funding Agencies: CIHR, MSFHR


Development of a Caenorhabditis Elegans Model to Study Mechanisms of Microbiota Mediated Protection against Clostridium difficile
T. Gerbaba,1 A. Moeez,2 I. Chin-Sang,2 and E. Petrof1

1Gastrointestinal Diseases Research Unit, Department of Medicine, Queen’s University, Kingston, ON, Canada
2Department of Biological Sciences, Queen’s University, Kingston, ON, Canada

Background. Clostridium difficile, a toxin producing bacterium, is the leading cause of antibiotic-associated diarrhea and pseudomembranous colitis. By destroying the diversity of the microbiota, antibiotic treatment creates an environment conducive for C. difficile proliferation and toxin production, rendering the host susceptible to C. difficile infection. Although the importance of a healthy microbiota for defense against C. difficile infection is well established, the keystone protective species (and mechanisms of protection) remain poorly understood.

Aims. To develop a simple and high-throughput applicable model to study mechanisms of microbiota-mediated protection and to discover novel microbial therapeutics against C. difficile toxins.

Methods. Age synchronized L1 stage worms were grown to L4 stage by feeding E. coli OP50 in the presence or absence of antibiotics (Carbenicilin). L4 stage worms were treated with C. difficile toxin (TcdB) in a 96 well plate and worm survival was monitored. The effect of toxin treatment on the cytoskeletal structure of the intestine was determined by FITC-phalloidin staining. Toxin effects on gut barrier/integrity was determined by FITC-dextran and propidium iodide (PI) assays.

Results. Pre-exposure of worms to E. coli OP50 with antibiotics rendered the worms susceptible to TcdB, whereas worms not exposed to antibiotic remained resistant (; log rank test, ). FITC-phalloidin staining of TcdB treated worms revealed destabilization of intestinal cytoskeletal structures and distension of the intestinal lumen. TcdB treatment also resulted in intestinal barrier breach as demonstrated by FITC-dextran assay. FITC-dextran was restricted to the gut lumen and was not seen in the pseudocoelom of controls, whereas TcdB treated worms showed leakage of FITC-dextran in to the pseudocoelom. PI remained restricted to the intestinal lumen in controls, whereas treatment with TcdB resulted in PI entry and intracellular staining of the intestinal cells. Co-treatment of worms with TcdB in the presence of secreted/excreted products of a consortium of 33 healthy human gut microbial isolates protected worms against the effect of TcdB on intestinal barrier function, integrity of intestinal cell plasma membrane, and survival, in antibiotic pretreated worms.

Conclusions. The developed model system would be useful to study mechanisms of microbiota-mediated protection and to discover novel microbial therapeutics against C. difficile.

Funding Agencies: John Alexander Stewart Fellowship; Women’s Giving Circle


Human Catestatin Treatment Results in Gut Microbiota Dysbiosis in Mice
M. Rabbi,1 P. Munyaka,1 N. Eissa,1 A. Khafipour,1 E. Khafipour,2 and J. Ghia1

1University of Manitoba, Winnipeg, MB, Canada
2Animal Science, University of Manitoba, Winnipeg, MB, Canada

Background. An increased ratio of Firmicutes/Bacteroidetes has been reported in inflammatory bowel diseases (IBD) or inflammatory bowel syndrome (IBS). Gut bacterial homeostasis can be regulated by epithelial antimicrobial peptides, such as Chromogranin-A (CgA)-derived peptides, released by enterochromaffin cells. Catestatin (CTS) derived from the C-terminal portion of CgA possess antimicrobial properties in vitro, but its effect on gut microbiota is unknown. We hypothesized that CTS injection will alter the trasnient gut microbiota and those colonized in the coloninc tissue.

Aims. To evaluate the in vivo impact of a CTS treatment on gut microbial composition and functional changes using a naive mouse model.

Methods. CTS (human CgA352-372, 1.5 mg/kg/day) or normal saline were administered intrarectally to C57BL/6 male mice (7 weeks of age) for 6 days and then sacrificed. DNA was extracted from feces and colonic mucosa associated microbiota (MAM) samples and the V3-V4 region of bacterial 16S rRNA was amplified and subjected to Illumina sequencing. Alpha-diversity was calculated using Chao 1, while beta-diversity was determined using weighted and unweighted UniFrac distances in QIIME. Differences at the genus level were determined using partial least squares discriminant analysis (PLS-DA), and phylogenetic investigation of communities by reconstruction of unobserved states (PICRUSt) was used to determine the functional capacity of bacterial community.

Results. CTS treatment did not change the bacterial alpha-diversity of microbiota in fecal and MAM samples, however, significantly altered the beta-diversity between the groups. CTS-treated mice demonstrated a lower relative abundance of Firmicutes () associated to a higher abundance of Bacteroidetes () in their feces, however, no significant changes at the phylum level were observed in the CTS-treated MAM samples. At lower taxonomic levels, in both fecal and MAM samples, the PLS-DA analysis revealed an association between specific taxa and the treatment. In particular but not limited, Bifidobacterium, Bacteroides, Prevotella and Parabacteroides were positively associated to the CTS-treated group. Differences in microbial functional pathways were detected at fecal and MAM samples, including but not limited to nitrogen, nicotinate, nicotinamide metabolism and cell division and ribosome biogenesis.

Conclusions. This study supports the hypothesis that CTS treatment alters the gut microbiota composition under physiological conditions. This might open new venues for CTS as a new line of antibiotic or as a therapeutic agent to restore the dysbiosis observed during the development of intestinal disorders.

Funding Agencies: CCC, CIHR


Acute 2,4 Dinitrobenzensulfonic Acid (DNBS)-Induced Colitis Results in Gut Microbiota Dysbiosis in Mice
A. Khafipour,1 P. Munyaka,1 M. Rabbi,1 N. Eissa,1 E. Khafipour,2 and J. Ghia1

1University of Manitoba, Winnipeg, MB, Canada
2Animal Science, University of Manitoba, Winnipeg, MB, Canada

Background. The most widely used experimental model of Crohn’s disease (CD) is the 2,4 dinitrobenzenesulfonic acid (DNBS)-induced colitis, which is developed by intrarectal (i.r.) administration of DNBS. Until recently, much attention has been given to the role of mucosal response to DNBS treatment, and only few studies investigated colonic microbiota in relation to DNBS. We hypothesized that i.r. treatment can modify both transient gut microbiota and those colonized in the colon tissue.

Aims. The aim of the study was to determine the microbial composition and functional changes in mice treated with DNBS.

Methods. C57Bl/6 mice received a solution of DNBS (4 mg/kg, i.r.) plus ethanol (30%) and control mice only PBS. Three days later confirmation of inflammation was evaluated clinically and by analysis of colonic tissue cytokine levels. From colon tissue and fecal samples, the V3-V4 region of bacterial 16S rRNA was amplified and subjected to Illumina sequencing. Alpha- and beta-diversities were calculated in QIIME and subjected to PROC MIXED analyses of SAS and PERMANOVA, respectively. Differences at the genus level were determined using partial least squares discriminant analysis. Functional capacity of bacterial community was predicted using open source software PICRUSt. Differences in inflammatory markers were tested using one-way ANOVA followed by multiple comparison post hoc analysis with significance at .

Results. Disease severity and IL-1β and IL-6 cytokines were increased in DNBS-treated versus vehicle mice. DNBS treatment reduced bacterial alpha-diversity and significantly altered the beta-diversity of microbiota. DNBS treatment also significantly reduced the abundance of several bacterial taxa including but not limited to lachnospiraceae and christensenaleace of Firmicutes phylum, Comomonadaceace from Proteobacteria phylum and increased the abundances of Bacteroidaceae and Prevotellaceae from Bacterioidetes phylum and Enterobacteriaceae of Proteobacteria phylum. Predicted functional capacity of bacterial community in both fecal and colon tissue samples was altered by DNBS treatment, including but not limited to lipopolysaccharide biosynthesis, aminobenzoated degradation, lysine degradation, and fatty acid metabolism.

Conclusions. Colitis in DNBS model was accompanied with the disruption of gut microbiota. While colitis development is driven by interplay between mucosal epithelial damage, an inflammatory response, and a dysbiotic microbiota, it is however, not clear if DNBS directly induces dysbiosis, or if microbiota dysbiosis occurs as a result of mucosal damage and inflammatory response.

Funding Agencies: CCC, CIHR


NOD2−/− Mice Show a Unique Response in Microbiota and Mucosal T Cell Responses after Neonatal Exposure to Antibiotics
A. Goethel,1 W. Turpin,1 S. Rouquier,1 G. Zanello,1 D. Philpott,1 and K. Croitoru2

1University of Toronto, Toronto, ON, Canada
2Mount Sinai Hospital, Toronto, ON, Canada

Background. Antibiotic exposure during early life is associated with development of auto-inflammatory diseases, including asthma, multiple sclerosis and Crohn’s disease, although the mechanism remains unclear. Exposure to antibiotics and the resultant changes in gut microbiota during early life may lead to disruption of normal mucosal immune development, as this is strongly influenced by commensal microbes.

Aims. We hypothesized that antibiotic disruption of the microbiota during early life would have a prolonged impact on both gut microbiota community structure and T cell function within the intestinal tract, resulting in defective immune tolerance to the gut microbiota in a genetically susceptible host (Nod2/ mice), leading to increased susceptibility to colitis.

Methods. Heterozygous Nod2 (+/−) breeders were used to generate litters of Nod2 +/+ (WT), +/− (HET) and −/− (KO) littermates, who all received the same foundational maternal microbiota. From birth, WT and Nod2/ littermates received amoxicillin (200 mg/L) in the drinking water until weaning. Amoxicillin transfers through the dam’s breast milk to the pups until day 14, when the pups begin to drink the water directly. Fecal samples collected at weaning were analyzed by targeted quantitative PCR (qPCR) and 16S ribosomal RNA sequencing for microbiota composition. At 8 weeks of age, mucosal lymphocyte populations of the small and large bowel were analyzed via flow cytometry.

Results. Neonatal amoxicillin treatment resulted in a significant shift of the gut microbial community structure, irrespective of Nod2 genotype or sex. The antibiotic-driven shift was associated with a significant reduction of Bifidobacterium and Lactobacillus. Neonatal antibiotics resulted in a significant reduction in serum IgA levels at 8 weeks of age. Phenotypes of small intestinal lamina propria lymphocyte (LPL) populations were not different in water-treated WT and Nod2/ littermates. However, antibiotic-treated Nod2/ littermates showed an increase in Foxp3+ and RORγt+ T cells.

Conclusions. Together, this suggests that neonatal antibiotic perturbation of the microbiota development alters Nod2 signalling in microbe-driven immune responses.

Funding Agencies: CAG, CIHR, OGS and Mount Sinai Hospital


Colonic Abnormalities in Manitoban Children with Helicobacter pylori Gastritis
U. Banik,1 C. Stefanovici,2 J. Griffin,1 and W. El-Matary3

1Section of Pediatric Gastroenterology, University of Manitoba, Winnipeg, MB, Canada
2Department of Pathology, University of Manitoba, Winnipeg, MB, Canada
3University of Manitoba, Winnipeg, MB, Canada

Background. Helicobacter pylori (H. pylori) is a pathogenic bacterium that can cause chronic gastritis. In Canada, Aboriginal population and first generation immigrants are most at risk for acquiring the organism. Association between H. pylori and colonic pathology is under-investigated.

Aims. To examine the prevalence and nature of colonic changes in children diagnosed with H. pylori gastritis in Manitoba.

Methods. A comprehensive retrospective chart review for all children (≤17 years) diagnosed with H. pylori gastritis from January 1996 to May 2015 at the Children’s Hospital, Winnipeg, Manitoba, was conducted. The medical records of children with H. pylori gastritis who had colonoscopy were thoroughly examined. Patients’ demographics, symptoms, laboratory findings, indications for colonoscopy and colonoscopic findings were documented.

Results. During the study period, 231 patients were found to have H. pylori gastritis. The mean age at diagnosis was years, 108 (46.6%) were girls. Of the 231 patients, 37 (16%) patients were found to have colonoscopy performed. 22 (59%) out of a total 37 children who had colonoscopy had significant endoscopic and histopathological findings on colonoscopy including polyposis and colitis. Males with colonic changes were diagnosed earlier compared to those without. No significant difference in colonic changes was found among children from different ethnic Backgrounds.

Conclusions. A significant number of pediatric patients with H. pylori gastritis in Manitoba had colonic pathologies. Aboriginal population and first generation immigrants are not at increased risk for having colonic changes in conjunction with H. pylori. Our study highlights a possible relationship between H. pylori gastritis and colonic changes in children and warrants properly planned prospective studies to confirm our results.

Funding Agencies: None


Heliobacter pylori among Obese and Non-Obese Patients in Najran, Saudi Arabia: A Controlled Cross-Sectional Study
A. Al-Zubaidi,1 A. Alzobydi,1 H. Alghamdy,1 A. Alakwa,2 K. Aman,3 R. Kurian,1 S. Kassim,4 and A. Alshahrani5

1King khalid Hospital, Najran, Saudi Arabia
2National gaurd hospital, Ahsa, Saudi Arabia
3Aden University, Aden, Yemen
4Taibah university, Al-Madinah Al-Munawwarah, Saudi Arabia
5Najran University, Najran, Saudi Arabia

Background. The association between Helicobacter pylori infection (HPI) and obesity is contested.

Aims. We aimed to enumerate and stratify by gender obese and non-obese patients with HPI.

Methods. This controlled cross-sectional study took place in the Department of Endoscopy of a central hospital that received all referrals in the Najran region of Saudi Arabia. A total of 340 obese Saudi patients (body mass index = 48.04) who had undergone diagnostic upper endoscopy before sleeve gastrectomy, were compared with 340 age and gender matched normal weight (body mass index = 23.13) control patients who had undergone diagnostic upper endoscopy for other reasons. Data collected included diagnosis of HPI based on histology of 2 biopsy specimens from gastric antrum and items related to socio-demographic characteristics of patients. Descriptive and bivariate data analysis was performed.

Results. Mean patients’ age was years, and 55% were males. Obese patients presented with significantly more HPI than non-obese patients (66% versus 50%, ). Both male and female obese patients presented with significantly HPI than non-obese patients (65% versus 45%, ; 69% versus 54%, , resp.). There was no statistically significant difference between males and females with respect to HPI among obese patients (64% versus 69%, ) and among non-obese the differences was marginal (44% versus 52%, ).

Conclusions. In this study, the significance of Helicobacter pylori infection in obese Saudi male and female patients who had undergone endoscopy compared with non-obese patients was demonstrated.

Funding Agencies: None

Pancreatico-Biliary Disease


Indications for Biliary Metal Stent Use and Stent Type Preference in a Large Prospective Multi-Centre Canadian Registry
G. Sandha,1 A. Roy,2 T. Maniere,3 J. Telford,4 N. Arya,5 R. Mohamed,6 J. Ellsmere,7 P. Kortan,8 A. Singh,9 and R. Bailey10

1University of Alberta, Edmonton, AB, Canada
2Centre Hospitalier de l’Université de Montréal-Saint-Luc Hospital, Montreal, QC, Canada
3Hopital Charles Le Moyne, Sherbrooke, QC, Canada
4Providence Health, St. Paul’s Hospital, Vancouver, BC, Canada
5Oakville-Trafalgar Memorial Hospital, Oakville, ON, Canada
6University of Calgary, Calgary, AB, Canada
7Queen Elizabeth II Health Sciences Center, Halifax, NS, Canada
8St. Michael’s Hospital, Toronto, ON, Canada
9Victoria General Hospital, Victoria, BC, Canada
10Royal Alexandra Hospital, Edmonton, AB, Canada

Background. The clinical indications for using biliary self-expanding metal stents (SEMS) and the preferences for type of SEMS in a real-life setting are not well documented.

Aims. The goal of the study is to document practice patterns pertaining to indications for biliary SEMS placement and stent type selection per standard of practice across Canadian hospitals.

Methods. This is a prospective, multi-centre and open-label study at 10 centres across Canada. Consecutive data was collected for a commonly used biliary SEMS in uncovered (UC), partially covered (PC) and fully covered (FC) versions (WallFlex®, Boston Scientific Corporation, Marlborough, MA, USA).

Results. To date 150 patients have been enrolled at 10 centres. Indications were palliation of malignant biliary obstruction (Group A) in 93/150 (62%), pre-operative drainage of malignant biliary obstruction with or without neoadjuvant therapy (Group B) in 21/150 (14%), treatment of a benign biliary stricture (Group C) in 21/150 (14%) and other indications (Group D) in 15/150 (10%). Group D indications were: EUS-guided cystgastrostomy for pancreatic fluid collections (7), EUS-guided cholecystoduodenostomy for recurring cholecystitis in non-surgical patients (2), refractory CBD stone disease with impacted stone in non-surgical candidate (2), intraductal papillary mucinous neoplasm (1), indeterminate biliary stricture (1), prevention of post-sphincterotomy bleeding (1) and extrinsic biliary compression (1). Stent type preferences are detailed in Table 21. Adverse events were seen in 2/150 (1.3%) and included acute pancreatitis (1) in Group C and recurrent cholangitis due to food impaction (1) in Group A, both in a FC stent.

Conclusions. For palliation of malignant biliary obstruction, UC stents seem to be preferred twice as commonly as FC stents. For benign biliary stricture treatment, FC stents are uniformly selected. PC stents are rarely used. No significant differences in adverse events are seen between UC and FC stents.

Funding Agencies: This study is sponsored by Boston Scientific Corporation, Marlborough, MA, USA


Eus Increases the Identification of Unresectable Disease among Adults with Pancreatic Adenocarcinoma: A Meta-Analysis
P. James,1 Z. Meng,1 M. Zhang,2 P. Belletrutti,2 R. Mohamed,2 W. Ghali,2 D. Roberts,3 G. Martel,4 and S. Heitman2

1Department of Medicine, University of Ottawa, Ottawa, ON, Canada
2Department of Medicine and Community Health Sciences, University of Calgary, Calgary, AB, Canada
3Department of Surgery, University of Calgary, Calgary, AB, Canada
4Department of Surgery, University of Ottawa, Ottawa, ON, Canada

Background. Although preoperative endoscopic ultrasound (EUS) is increasingly being used among adults with pancreatic adenocarcinoma, it remains unclear if this procedure influences surgical management.

Aims. This systematic review sought to determine if preoperative EUS evaluation is associated with the identification of unresectable disease among adults with pancreatic adenocarcinoma.

Methods. We searched MEDLINE, EMBASE, bibliographies of included articles and conference proceedings for studies reporting original data regarding surgical management among patients with pancreatic adenocarcinoma. Our main outcome was the incremental benefit of EUS for the identification of unresectable disease (). The pooled was calculated using random effects models. Heterogeneity was explored using stratified meta-analysis and meta-regression.

Results. Among the 4,505 citations identified, we included 9 cohort studies that examined the identification of unresectable disease (,030). Random effects meta-analysis suggested that EUS alone identified unresectable disease in 17% of patients (95% confidence interval (CI), 11–25%) who appeared to have resectable disease on CT scan. This number increased in studies where EUS was the only modality used after CT scan (24%, 95% CI 16–39%), and decreased when extensive imaging (CT, abdominal ultrasound and MRI) were performed before EUS (5%, 95% CI 3–11%). Among those studies that considered portal or mesenteric vein invasion as potentially resectable, EUS alone was able to identify unresectable disease in 15% of patients (95% CI 9–27%) after a CT scan was performed.

Conclusions. Existing evidence suggests that EUS evaluation is associated with increased identification of unresectable disease among adults with pancreatic adenocarcinoma.

Funding Agencies: CAG, CIHR, Alberta Innovates: Health Solutions


Repeat Endoscopic Ultrasound-Guided Fine Needle Aspiration in Patients with Suspected Pancreatic Cancer: Diagnostic Yield and Associated Change in Access to Appropriate Care
D. Stanger,1 R. Enns,2 J. Telford,1 and E. Lam3

1University of British Columbia, Vancouver, BC, Canada
2St. Paul, Vancouver, BC, Canada
3Univ. of California, Vancouver, BC, Canada

Background. Pancreatic malignancy carries a poor prognosis. In order to direct appropriate therapy towards a potential pancreatic malignancy, a tissue diagnosis is often required.

Endoscopic ultrasound (EUS) is a contemporary technique that uses both endoscopy and ultrasound (US) to image the pancreas. Studies have demonstrated EUS is more sensitive than both transabdominal US and CT scan in detecting pancreatic solid masses.

Despite the fact that EUS-FNA has demonstrated instrumental diagnostic utility, recent literature has determined inconclusive diagnosis at initial EUS-FNA.

To obtain appropriate preoperative chemotherapy, definitive Class 5 cytology is required. It is important to clarify how we obtain appropriate diagnosis on those patients whose initial diagnosis is not easily obtained with the first EUS-FNA.


Primary Objective. To determine the cumulative yield of repeat EUS-FNA in patients with suspected pancreatic cancer.

Secondary Objectives. To determine how a second biopsy influences access to next steps in management.

To determine complications associated with a second EUS-FNA.

Methods. A retrospective cohort study was conducted evaluating the yield of repeat EUS-FNA in determining a cytological diagnosis in patients who had undergone an initial but inconclusive EUS-FNA for pancreatic lesions suspicious for malignancy.

Patients were selected from a database at one tertiary university-based referral centre for pancreatico-biliary disorders during the period of 2007–2014.

All patients included in the study underwent repeat EUS-FNA for investigation of a pancreatic mass.

Results. 45 repeat EUS-FNA procedures were performed.

Of the 45 repeat procedures, the cumulative yield for diagnosing pancreatic adenocarcioma was 17.8%.

22% of patients who underwent repeat EUS-FNA received treatment including either surgery, radiotherapy or chemotherapy. Not all of these patients had definitive cytological diagnoses of pancreatic adenocarcinoma.

The complication rate associated with repeat EUS-FNA was 6.7%. All complications were mild and not life-threatening.

The mean time from referral to 1st FNA, second FNA and inter-FNAs were 24.8 days, 67.8 days and 43.0 days, respectively.

Conclusions. Pancreatic biopsies often require a repeat procedure for diagnostic certainty. The cumulative yield of repeat EUS-FNA for diagnosing pancreatic adenocarcioma was found to be 17.8% (8/45). Repeat EUS-FNA facilitated access to care as 22% of patients who underwent repeat FNA were treated with either chemotherapy, radiotherapy or surgery. Complications from a second procedure were relatively common but were not severe.

Funding Agencies: None


Comparison of a Novel Side Port Needle with a Conventional End Port Needle in Eus-Guided FNA of Solid Lesions
J. Tan,1 B. Tessier-Cloutier,2 D. Schaeffer,1 and F. Donnellan1

1Vancouver General Hospital, Vancouver, BC, Canada
2University of British Columbia, Vancouver, BC, Canada

Background. Endoscopic ultrasound (EUS) guided fine needle aspiration (FNA) plays an integral role in the diagnosis of pancreas and stomach solid lesions. The conventional needle used for EUS guided FNA has a single port at the end of the needle. Diagnostic accuracies rarely exceed 90% resulting in need for repeat EUS. The development of an additional novel side port needle with its “cheese-like" grating effect during movement of the needle in addition to the end needle port carries a theoretical advantage of increasing cellular acquisition thereby improving diagnostic yield.

Aims. To compare the sample cellularity obtained with both EUS guided FNA needles.

Methods. This randomized prospective study was carried out at Vancouver General Hospital Canada where competent adults undergoing EUS for FNA of a lesion seen on imaging were recruited. All consented patients were randomly allocated to either the conventional end port needle (Boston 22 G) or the novel side port needle (Olympus 22 G). Two consecutive passes were made using the allocated needle with tissue acquired sent into separate cytology containers labeled A and B. This FNA process was repeated using the other study needle for the same patient on the same day labeled C and D. FNA technique was standardized for all passes ie. Suction only. Pathologists were blinded to the needle used. Scoring for cytology cellularity were as follows: 0 for 0%, 1 for <5%, 2 for 5–30%, 3 for 30–60% and 4 for >60%. Scoring of core tissue acquired were defined as 0 for no cell cluster >0.5 mm, 1 for cell cluster 0.5 mm to 1 mm and 2 for cluster >1 mm.

Results. 20 patients were recruited between May 2014 and June 2015. 18 were referred for pancreatic mass with remaining 2 patients for gastric mass. The median age was 64.5 years with 12 males and 8 females. A total of 40 FNA cytology specimens were collected and included in the final analysis. (Table 22).

There was no difference in the total cytology score (92 versus 88; ) or the cytology score per specimen (2.3 versus 2.2; ) between the Boston and Olympus needles respectively.

There was a greater number of core specimens obtained with the Olympus needle (47.5% versus 37.5%; ).

Conclusions. Our study demonstrates that a novel side port needle has no advantage over a conventional end port needle in terms of sample cellularity obtained with EUS guided FNA of solid lesions.

Funding Agencies: This study was kindly supported by a grant from Pancreas Centre BC


Clinical, Endoscopic, and Cytopathologic Determinants of Non-Diagnostic Endoscopic Ultrasound Guided Fine Needle Aspiration in Solid Pancreatic Masses
M. Alfawaz, M. Sey, A. AlNasser, N. Hussain, M. Weir, M. Joseph, and B. Yan

Western University, London, ON, Canada

Background. Endoscopic ultrasound guided fine needle aspiration (EUS-FNA) diagnostic yield in solid pancreatic masses should be approximately 75% based on previously published studies. A recent quality improvement study of our EUS-FNA results for pancreatic masses revealed non-diangnostic results in 47% of cases, despite having rapid on-site evaluation (ROSE). This study was completed to determine reasons for low diagnostic EUS FNA in solid pancreatic masses.

Aims. The aim of this study was to determine the clinical, procedural, and cytopathologic features that predict a non-diagnostic EUS-FNA for a pancreatic mass.

Methods. Retrospective chart review of all EUS-FNA cases performed for pancreatic masses between January 2010 and Dec 31, 2014. Predictors of a non-diagnostic EUS-FNA including patient related risk factors for pancreatic cancer, imaging characteristics, tumor marker, EUS-FNA procedural factors, and ROSE evaluations were recorded. Cases were considered diagnostic if their cytopathology were reported as either (1) positive for a malignancy, or (2) negative for a malignancy in the setting of sufficient cellularity. Cases were deemed non-diagnostic if cytopathology were reported as: (1) suspicious for malignancy, (2) atypical, (3) indeterminate, or (4) insufficient. Potential predictors of non-diagnostic EUS-FNA were assessed using univariate and multivariate logistic regression modeling.

Results. A total of 254 pancreatic masses were included in this study. One hundred sixty were in the head of the pancreas, 61 in the body, and 15 in the tail, 1 in the uncinate, and 8 not reported.

Of these lesions, 103 were diagnostic and 142 non-diagnostic. No significant patient clinical factors predicted non-diagnostic FNA. The only statistically significant determinant for non-diagnostic FNA was mass location in the head of the pancreas.

On multivariate analysis, the odds ratio for a non-diagnostic specimen in the head compared to elsewhere in the pancreas else was 2.6 ().

EUS procedural factors (including needle size, number of passes, year of procedure, physician and trainee involvement) did not affect probability of diagnostic specimen. Non-diagnostic samples were not associated with any particular cytopathologist.

Conclusions. Lesions in the head of the pancreas were associated with a higher non-diagnostic EUS FNA rate compared to lesions elsewhere in the pancreas. The reason for this requires further study. Efforts to optimise sampling and interpretation of pancreatic head lesions should be a focus of quality improvement programs in centers with low diagnostic rates in EUS FNA.

Funding Agencies: None


A Prospective, Randomized Trial on the Timing of Rectal Indomethacin Administration for the Prevention of Post-ERCP Pancreatitis: A Preliminary Report
S. Kenshil, P. D’Souza, R. Sultanian, G. Sandha, V. Bain, R. Hundal, M. Laghari, A. Montano-Loza, and S. Zepeda-Gomez

University of Alberta, Edmonton, AB, Canada

Background. Rectal administration of indomethacin has been shown to reduce the risk of post-ERCP pancreatitis. Most of the studies have been done with administration of indomethacin immediately after the procedure. There have been no comparative studies about peri-procedural timing of indomethacin administration.

Aims. To compare the efficacy of pre-ERCP versus post-ERCP administration of rectal indomethacin in preventing the occurrence of post-ERCP pancreatitis.

Methods. All patients undergoing ERCP were randomized to receive 100 mg of rectal indomethacin either before (Group 1) or after (Group 2) the procedure. Patients with previous sphincterotomy, hypersensitivity to NSAIDs, rectal disease, renal failure, coagulopathy, active peptic ulcer disease, pregnancy, breastfeeding or acute pancreatitis were not eligible to be included. Lipase levels were collected two to four hours post-procedure. Patients were also evaluated clinically and contacted 48 hours later to determine if they had symptoms suggestive of pancreatitis. Patients who met high-risk criteria for development of post-ERCP pancreatitis were analyzed as a subgroup.

Results. A total of 165 patients have been included in the study (99 females (60%), mean age 52 (range 20–91) years). Eighty patients were included in group 1 and 85 in group 2. The most common indication for ERCP was choledocolithiasis (73.9%). Technical success was achieved in 95.2% of patients (157/165). Post-ERCP pancreatitis occurred in 7.3% (12/165) of all cases; 7/80 in group 1 (8.75%) versus 5/85 (5.88%) in group 2 (). Sixty-two patients (37.6%) were classified as high-risk for post-ERCP pancreatitis. A sub analysis performed revealed that pancreatitis occurred in 9 patients categorized as high-risk in comparison to 3 in patients categorized as non-high risk (). Amongst the high-risk patients, 5/21 of group 1 and 4/34 of group 2 developed pancreatitis (). There were no cases of severe pancreatitis.

Conclusions. Our preliminary data suggests that the timing of indomethacin administration does not influence the rate of post-ERCP pancreatitis.

Funding Agencies: None


Optimizing the Diagnostic Yield of EUS-FNA for Solid Pancreatic Lesions: A Single-Centre Quality Assurance Study
M. Abunassar,1 A. Chatterjee,1 B. Dube,2 C. Marginean,3 G. Martel,4 S. Murthy,1 A. Rostom,1 C. Dube,1 and P. James1

1The Ottawa Hospital, Department of Medicine, Division of Gastroenterology, Ottawa, ON, Canada
2University of Ottawa/OHRI, Ottawa, ON, Canada
3The Ottawa Hospital, Department of Pathology, Ottawa, ON, Canada
4The Ottawa Hospital, HPB Surgery, Ottawa, ON, Canada

Background. Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) is a safe and effective procedure for the investigation of pancreatic masses. Improving EUS-FNA diagnostic yield will reduce the necessity for repeat procedures, thereby reducing risk to patients and resource use.

Aims. To examine factors associated with EUS-FNA diagnostic yield at our centre.

Methods. We performed a retrospective chart review of EUS-FNA procedures performed for the sampling of solid pancreatic lesions between September 1st 2009 to August 31st 2015 at The Ottawa Hospital. Rapid on-site evaluation (ROSE) for EUS-FNA was introduced in September 2010. Data regarding patient demographics (age and sex), lesion location, procedure details (endoscopist, FNA needle gauge, suction technique, number of passes) and the reviewing pathologist were collected. In addition to descriptive statistics, univariate and multivariable analyses were performed to determine factors associated with diagnostic yield.

Results. 350 EUS-FNAs for solid pancreatic lesions were examined by chart review. 288 (82%) of the procedures involved ROSE. The median patient age was 66 (interquartile range (IQR) 57–76) years and 56% were female. The overall EUS-FNA diagnostic yield was 81%. The diagnostic yield by the following factors were observed: patient sex (male 78%, female 84%), endoscopist (A 81% versus B 82%), lesion location (head 84%, body 78%, tail 74%), needle gauge (g) (19 g 67%, 22 g 82%, 25 g 80%), and number of FNA passes performed (one 50%, two 70%, three 84%, four 79%, five 80%, six 82%). The diagnostic yield with ROSE was 81% compared to 75% without ROSE. 11 pathologists were involved in the EUS cytopathology review, with a wide range in the number of cases reviewed by each pathologist (from 1 to 68 cases) and in their diagnostic yield (from 67% to 93%). No single factor was found to be significantly () associated with diagnostic yield in univariate or multivariate analyses.

Limitations. This was a retrospective study. Not all EUS-FNA cases have been captured to date.

Conclusions. Although our overall diagnostic yield is comparable to what is reported in the literature, there is an opportunity for improvement. Multidisciplinary FNA Cytopathology rounds have begun at The Ottawa Hospital with an aim to optimize at optimizing specimen acquisition, processing and evaluation.

Funding Agencies: The Ottawa Hospital Department of Medicine Patient Safety and Quality Research Grant


Single-Operator Cholangioscopy Is More Cost-Effective Than Bile Duct Exploration for Management of Difficult Common Bile Duct Stones after Failed Conventional Ercp
J. Sandha, R. Ennis-Davis, and G. Sandha

University of Alberta, Edmonton, AB, Canada

Background. Common bile duct (CBD) stones are currently managed by ERCP with sphincterotomy and stone extraction with a balloon catheter or basket. However, some stones are difficult to extract by conventional means. These include multiple, large, impacted, or faceted stones, or those proximal to a stricture. Cholangioscopy with electro-hydraulic lithotripsy (EHL) is a modality to treat these difficult stones.

Aims. The aim of this study was to describe the clinical efficacy of a single-operator cholangioscopy system (SpyGlass, Boston Scientific Corporation, Marlborough, MA, USA) for difficult stones and discuss possible cost savings by avoiding surgical intervention.

Methods. We performed a retrospective chart review of all patients referred for SpyGlass cholangioscopy with EHL for difficult stones. Clinical efficacy was defined as the successful clearance of the CBD of all stones. The total cost (based on Alberta Health Services reimbursement codes) was calculated by adding all costs associated with cholangioscopy, including any subsequent procedures, surgery, hospital stay, or treatment of any complications. This cost was compared with the projected cost of open and laparoscopic CBD exploration (OCBDE, LCBDE).

Results. A total of 51 patients with difficult CBD stones (35 female, median age years (range 30–88 years)) with 108 prior ERCPs (average 2.1/patient) were referred. They underwent 58 SpyGlass cholangioscopy ERCPs and 7 additional ERCPs (average 1.3/patient). The average procedure time was 67 minutes (range 24–124 minutes). The CBD was successfully cleared in 47/51 patients (93% clinical efficacy). Minor complications were seen in 7 patients (14%). These included mild EHL-induced CBD wall trauma in 4 patients, wire-induced cystic duct stump leak in 1 patient, mild post-ERCP pancreatitis in 1 patient and mild mucosal tearing at gastro-esophageal junction during extraction of a plastic stent in 1 patient. The average cost of all procedures was $4550. This compares to a projected cost of $7766 and $6175 for OCBDE and LCBDE, respectively. The average cost saving per patient using SpyGlass cholangioscopy instead of OCBDE or LCBDE was $3216 and $1625, respectively. The published rate of complications (bile leak, hemorrhage, and abscess) for CBDE is 3.2%. Treatment of these complications added $4977, $5216, and $3701 to the cost of CBDE.

Conclusions. Single-operator SpyGlass cholangioscopy with EHL is highly effective for the treatment of difficult CBD stones. By adopting this modality as primary treatment for these difficult stones, significant cost savings may be realised by avoiding surgical intervention.

Funding Agencies: None


Double-Baloon Endoscopic Retrograde Cholangiopancreatography in Patients with Surgically Altered Anatomy: A Single Center Experience
J. Nilsson, A. Montano-Loza, and S. Zepeda-Gomez

University of Alberta, Edmonton, AB, Canada

Background. Balloon assisted enteroscopy has improved our ability to perform endoscopic retrograde cholangiopancreatography (ERCP) in patients with surgically altered anatomy. We reviewed the experience with double-balloon ERCP (DBE-ERCP) in patients with altered anatomy and suspicion of biliary obstruction in a tertiary center.

Aims. To assess procedure indications, rates of success and procedural related complications with DBE-ERCP.

Methods. Retrospective analysis of all patients who underwent DBE-ERCP at the University of Alberta hospital between August 2011 and September 2015.

Results. A total of 57 DBE-ERCPs were performed in 28 patients (16 males) with a mean age of years (range: 20–81) using a short-type double balloon enteroscope. Twenty-seven patients had a Roux-en-Y reconstruction (25 hepatico-jejunostomies) and one patient had a prior Billroth-II gastro-jejunostomy. There were 19 patients that had previous liver transplantation (9 cadaveric, 10 living donor).

Mean time from surgery to the first DBE-ERCP was significantly lower in liver transplant patients compared to other surgeries ( versus days, ()). There was a trend to earlier DBE-ERCP in living related versus cadaveric transplants ( versus ()).

The main indications for procedures were suspicion of stricture at the hepatico-jejunostomy ( (44%)), recurrent cholangitis ( (37%)) and stent retrieval ( (14%)). Therapeutic maneuvers included: stricture dilation (), extraction of stones (), stent placement () and stent retrieval (). The hepatico-jejunostomy or major papilla was reached in 46 of 57 procedures (81%). Bile duct cannulation was successful in 40 of 46 procedures (87%).

The mean number of procedures per patient was (range: 1–7 procedures). The number of procedures was higher in those with liver transplantation compared to other surgeries (mean: versus ()). There were two patients with mild cholangitis that resolved with intravenous antibiotic therapy.

Fourteen patients required stenting and dilation of the hepatico-jejunostomy. No subsequent intervention was required in ten of these patients after a mean of (range 1–7) procedures. In 4 patients, subsequent percutaneous drainage (PTC) was required for failure of endoscopic therapy, mean time to PTC was 136 days ±104 (30–274).

Conclusions. DBE-ERCP allows for successful therapy in patients with surgically altered anatomy of the upper-GI tract. Our single center study suggests this is a safe, and effective first line option at managing post-surgical biliary obstruction/strictures, however more than one session is generally required to achieve good outcomes.

Funding Agencies: None


EUS-Guided Cystgastrostomy Using a Self-Expanding Metal Stent Is a More Cost-Effective Strategy for the Treatment of Pancreatic Fluid Collections
S. Kenshil,1 C. Teshima,2 P. D’Souza,1 and G. Sandha1

1University of Alberta, Edmonton, AB, Canada
2St. Michael’s Hospital, Toronto, ON, Canada

Background. EUS-guided cystgastrostomy (EUS-CG) has become the mainstay for endoscopic treatment of symptomatic pancreatic fluid collections (PFCs). The traditional approach (Approach I) of inserting multiple double-pigtail plastic stents with sequential dilation of the tract with or without naso-cystic (NC) irrigation is successful but requires repeated endoscopic intervention. The clinical utility of newer approaches utilizing a fully covered expandable metal stent (FCEMS, Approach II) is less well understood.

Aims. Our aim is to compare the clinical efficacy and cost impact of these approaches in the management of PFCs.

Methods. We performed a retrospective chart review of patients that underwent EUS-CG for PFCs. Clinical efficacy was defined as symptomatic improvement and cyst resolution on cross-sectional imaging. Follow up was until stent removal. The total cost of each procedure was calculated based on Alberta Health Services reimbursement codes for each EUS-CG (including anesthesia and radiology costs), all subsequent procedures and hospital length of stay (HLOS) until cyst resolution was confirmed. The clinical efficacy and average cost of the two approaches was then compared.

Results. Between 11/2010 and 10/2015, 10 patients underwent Approach I and 5 patients underwent Approach II. Cyst resolution was documented in all cases with a mean follow up of 304 days in Approach I (range 86–628 days) and 64 days in Approach II (range 3–124 days). Patients in Approach I had a mean of 3.7 endoscopic interventions (range 1–8) and 11.9 days of HLOS (range 0–33 days) compared with 1 endoscopic intervention and 2.4 days (range 0–4 days) in Approach II, respectively. Complications were seen in 5/10 patients (50%) in Approach I (4 patients with pus developing in the cyst space after the index procedure requiring prolonged hospital stay and repeated NC irrigation and antibiotic treatment. One of these patients and another patient had bleeding from the CG site after the index procedure requiring blood transfusion) compared with 1/5 patients (20%) in Approach II (ER visit 6 days after the index procedure with pain and increased c-reactive protein but managed with outpatient antibiotic therapy). In Approach I, 3/10 patients received hydrogen peroxide flushing compared with none in Approach II. The average cost of Approach I was $18,413 compared with $6,151 for Approach II.

Conclusions. Despite the increased upfront cost of metal stents, the use of FCEMS for EUS-CG in the treatment of PFCs results in significant overall cost savings by reducing the need for multiple endoscopic re-interventions and drainage procedures.

Funding Agencies: None


EUS-Guided Cholecystenterostomy Is Effective in the Management of High-Risk Cases of Acute Cholecystitis
S. Kenshil and G. Sandha

University of Alberta, Edmonton, AB, Canada

Background. Acute cholecystitis is a common clinical entity and optimal management is open or laparoscopic cholecystectomy. For high-risk patients that are not surgical candidates, conservative management includes antibiotics and percutaneous cholecystostomy (PC). EUS-guided cholecystenterostomy (EUS-CE) is an emerging alternative to PC.

Aims. We describe two high-risk patients with acute cholecystitis deemed not to be surgical candidates managed with EUS-CE.

Methods. Using a linear array echoendoscope (Olympus America, Center Valley, PA) and fluoroscopic guidance, a trans-duodenal puncture of the gallbladder was performed with a 19-gauge FNA needle (Slimline Expect®, Boston Scientific, Marlborough, MA). A 0.035-inch guidewire (Jagwire®, Boston Scientific, Marlborough, MA) was then coiled within the gallbladder following which a fistula was created using a 10 French (Fr) cystotome (Cook Medical, Bloomington, IN). A fully covered 10 mm wide × 40 mm long metal biliary stent (Wallflex, Boston Scientific, Marlborough, MA) was then inserted to create the CE. After that a 7 Fr, 4 cm long double-pigtail plastic biliary stent (Cook Medical, Bloomington, IN) was positioned within the metal stent so as to minimize any risk of migration.

Results. EUS-CE was performed on 2 patients for conservative management of their acute cholecystitis. Both patients were deemed non-surgical candidates by the surgical team. Patient 1, a 58 year-old man with decompensated cirrhosis, was admitted to the ICU with recurring bacterial peritonitis thought to be secondary to translocation of bacteria from his gallbladder as a result of his repeated episodes of calculus cholecystitis. Patient 2, an 83 year-old man with multiple comorbidities, including a recent acute coronary syndrome, had multiple admissions secondary to repeat episodes of acute cholecystitis. As a result of his repeat courses of antibiotics, he developed refractory Clostridium difficile colitis. Both patients underwent successful EUS-CE with drainage of the gallbladder through the stents documented endoscopically. Both patients were subsequently discharged home from the hospital. Neither patient has required repeat admission for acute cholecystitis or repeat interventions as a result of the EUS-CE.

Conclusions. EUS-CE is an effective treatment for acute cholecystitis in high-risk patients that are not fit for surgery. It is less invasive and has better patient acceptance than PC. We advocate an increasing role for EUS-CE in the non-surgical management of acute cholecystitis.

Funding Agencies: None


Diagnostic Yield of Endoscopic Ultrasound Guided Fine Needle Aspiration versus Fine Needle Biopsy for Solid Lesions
A. Kayal,1 C. Chan,2 M. Alsahafi,1 A. Weiss,1 M. Byrne,1 D. Schaeffer,3 and F. Donnellan1

1Division of Gastroenterology, Vancouver General Hospital, University of British Columbia, Vancouver, BC, Canada
2University of British Columbia, Vancouver, BC, Canada
3Department of Anatomical Pathology, Vancouver General Hospital, University of British Columbia, Vancouver, BC, Canada

Background. Endoscopic ultrasound guided fine needle aspiration (EUS-FNA) is the standard technique for obtaining tissue samples. The Sharkcore Needle (Covidien) is a new fine biopsy needle (FNB) for obtaining core tissue at time of EUS.

Aims. To compare the diagnostic yield of a conventional EUS FNA needle with a new EUS FNB needle for solid lesions in close proximity to the upper GI tract.

Methods. A retrospective study of patients who underwent EUS for tissue acquisition of solid lesions using both a conventional FNA needle (Boston Scientific) and a novel FNB needle (Sharkcore/Covidien) in the same session between February and June 2015. Two passes were made with the FNA needle using a standard EUS technique (no stylet, with suction). Two passes were also made with the FNB needle using a slow pull technique on the first pass and suction on the second pass. All were examined by a GI pathologist for neoplasia, diagnostic or non-diagnostic. Diagnostic yield was calculated based on a confirmed diagnosis by EUS sampling or surgically resected specimen or a presumed diagnosis by radiological imaging and overall clinical picture.

Results. 21 patients were included in the study. Mean age was 58.2 and 8 were male (38%). 11 (52.4%) had a pancreatic mass while the rest included both gastric and duodenal subepithelial tumors, and mediastinal and intra-abdominal lesions.

Using the FNA method, in 18 out of 21 (85%) a diagnosis was made compared to 15 out of 21 (71.4%) using FNB technique. This was not statistically significant with a value of 0.45 based on Fischer’s exact test. Combining both methods 19 out of 21 (90.5%) had a diagnostic sample.

Conclusions. EUS-FNB does not appear to increase the diagnostic yield compared to EUS-FNA. However, combining both techniques may increase this yield.

Funding Agencies: None


Drug Associated Acute Pancreatitis Secondary to Apixaban: A Case Report
H. Azhari,1 C. Ma,2 and P. Kumar1

1University of Calgary, Calgary, AB, Canada
2University of Alberta, Edmonton, AB, Canada

Background. A wide variety of drugs are known to cause acute pancreatitis. Apixaban is a relatively new anticoagulant approved by Health Canada and the Federal Drug Administration (FDA) in 2012 for the treatment and prevention of venous thromboembolic events and prevention of thromboembolism in non-valvular atrial fibrillation (AF). A variety of adverse reactions have been reported with apixaban, with bleeding being the most significant.

Aims. Here, we present a case of recurrent acute pancreatitis secondary to treatment with apixaban, highlighting the importance of considering newer drugs as possible etiologies of acute pancreatitis. To the best of our knowledge, this is the first reported case of apixaban-associated pancreatitis.

Methods. An 87-year-old female with a known history of atrial fibrillation, hypertension, previous cholecystectomy, and gout presented with a one-day history of severe epigastric abdominal pain, nausea, and vomiting. She had been started on apixaban three weeks prior for persistent non-valvular atrial fibrillation: the decision for anticoagulation was made by her family physician in consideration of her moderate-to-high risk for AF-related thromboembolism. Examination showed a tender epigastrium and periumbilicus but without peritonitis, and investigations demonstrated leukocytosis (12.8 × 109/L) and an elevated lipase (>3000 U/L). She was admitted for conservative management including analgesia and intravenous hydration. Her apixaban was temporarily suspended and her acute pancreatitis resolved rapidly. The patient was discharged home but she restarted apixaban after discharge, and 17 days after resuming apixaban, she developed recurrent symptoms of acute pancreatitis and required re-hospitalization.

Results. Extensive investigations were performed to exclude other diagnostic possibilities. An incidental pancreas divisum was seen on magnetic resonance cholangiopancreatography (MRCP), but no other drugs, metabolic, or autoimmune causes were identified. Endoscopic ultrasound was also performed and did not show any choledocholithiasis or biliary sludge. It was therefore strongly suspected that apixaban might be the offending agent. The patient was discharged after permanently discontinuing apixaban, and has not had any recurrence of symptoms since.

Conclusions. When evaluated using the World Health Organization Uppsala Monitoring Centre (WHO-UMC) Causality Assessment for drug-related adverse effects, this case represents at minimum a probable to likely case of drug-induced pancreatitis given the plausible time correlation, response to drug withdrawal and re-challenge, and exclusion of other diagnostic etiologies.

Funding Agencies: None


Main-Duct Intraductal Papillary Mucinous Neoplasm of the Pancreas Associated with Spontaneous Pancreaticoduodenal and Pancreaticogastric Fistulas
R. Almeida, C. Dargavel, and J. Mosko

University of Toronto, Toronto, ON, Canada

Background. Intraductal papillary mucinous neoplasms (IPMNs) of the pancreas are slow growing neoplasms arising from the epithelial lining of the pancreatic duct system. IPMNs represent a spectrum ranging from benign to invasive carcinoma. IPMNs complicated by the development of fistulas, however, are rare.

Aims. To describe a case of a main-duct (MD) IPMN associated with spontaneous pancreaticoduodenal and pancreaticogastric fistulas.

Methods. Case report and literature review.

Results. A 90-year-old woman with a prior history of a distal pancreaticojejunostomy for a pancreatic ductal carcinoma in situ 18 years ago, presented with cholangitis. An endoscopic retrograde cholangiopancreatography (ERCP) at the initial institution was unsuccessful due to altered anatomy. She was then transferred for percutaneous transhepatic cholangiography drain placement, which achieved biliary drainage. An esophagogastroduodenoscopy undertaken prior to a repeat ERCP showed a large gastric lesion with central ulceration along the greater curvature of the proximal body, with mucinous extrusion from the center and further drainage emanating from the second part of the duodenum obscuring visualization of the papilla (Figure 14). A multiphase CT of the pancreas showed an abnormal pancreaticobiliary system with a complex loculated cystic lesion in the pancreatic bed, approximately 13 × 6 × 16 cm in size, compressing the stomach and communicating to the greater curvature of the stomach and the superior wall of the third part of the duodenum (Figure 15). Histological examination of the gastric biopsies showed superficial villous architecture and gastric foveolar type epithelium with intestinal metaplasia and low grade dysplasia (Figure 16). This constellation of endoscopic, radiographic and histologic features was suggestive of malignant transformation of a MD-IPMN with spontaneous fistulization to the stomach and the duodenum.

The IPMN-associated fistulization to adjacent viscera has an incidence rate of 1.9%–6.6%. The mechanistic basis is hypothesized to include mechanical pressure, tumor penetration and pancreatic enzyme related autodigestion. While predominantly associated with malignancy, fistulization has also been reported in benign IPMNs. IPMN fistulas commonly involve the duodenum, followed by the stomach, CBD and colon. Whilst CT and MRI imaging characterize and diagnose IPMN fistulas, definitive diagnosis depends on histopathology. Accurate prognostic data on IPMN fistulas is unknown, however, scant literature suggests a 5-year survival rate of 43% after resection.

Conclusions. The rare complication of fistula formation in IPMN preferentially involves the duodenum, and usually occurs in the setting of malignant transformation. While uncommon, IPMN fistulization should be considered in the setting of cholangitis.

Funding Agencies: None


Use of Rectal Indomethacin for Post-ERCP Pancreatitis Prevention: A Quality Assurance Study
S. Rolland, S. Shanmuganathan, A. Chatterjee, S. Grégoire, H. Dhaliwal, and P. James

University of Ottawa, Ottawa, ON, Canada

Background. The evidence to date suggests that rectal indomethacin should be provided for post-ERCP pancreatitis (PEP) prevention for all high risk cases. This also benefits patients at average risk of PEP as well.

Aims. The aim of this quality assurance study is to determine the current use of rectal indomethacin for PEP prevention in our centre as well as its association with risk of PEP.

Methods. This is a retrospective chart review study for all ERCP cases performed at our institution from January to March 2015. Data regarding patient demographics and clinical status, procedure indication, interventions performed and use of indomethacin for PEP prevention was collected.

Results. Data from 41 ERCP cases where a sphincterotomy was performed was collected. The median patient age was 71 years and 54% were female. 24% of cases included the use of indomethacin for PEP prevention. Among cases that involved females under 50 years or patients with a history of pancreatitis, 11% received rectal indomethacin. Among the cases considered, two patients were seen in hospital for PEP (risk 5%) and no other complications were identified.

Conclusions. Rectal Indomethacin for PEP prevention is underused in our centre, especially among higher risk patients. However, the overall risk of hospitalization for PEP remained low. This is a retrospective chart study with a very small sample of patients. However, these results will be used to develop an algorithm aimed at identifying patients at elevated risk of PEP and facilitating increased use of rectal indomethacin for PEP prevention.

Funding Agencies: None


A Rare Nidus for Biliary Stone Formation
R. Battat,1 M. Drapeau,2 B. Faulques,2 and J. Wyse1

1McGill University, Montreal, QC, Canada
2Université de Montreal, Montreal, QC, Canada

Background. Early complications of Laproscopic Cholecystectomy (LC) include haemorrhage, perforation of the gallbladder, common bile duct (CBD) injury and iatrogenic bowel and vascular injuries1. Late complications involve intra-abdominal bile leakage, sub-hepatic abscesses, retained bile duct stones, post-cholecystectomy syndrome and bile duct stricture2 Surgical clips placed on the cystic duct and arteries avoid cystic duct leakage and arterial bleeding, but allows the rare late LC complication of post-cholecystectomy clip migration (PCCM) with gallstone formation. While rare, consequences of this complication, such as ascending cholangitis, can be life threatening.

Aims. We describe a 54-year-old Caucasian female patient with Crohn’s disease presenting with abdominal pain attributable to post cholecystectomy clip migration with choledocholithiasis.

Methods. NA

Results. A 54-year-old woman presented with one episode of vomiting, a one month history of anorexia, and postprandial right sided and epigastric abdominal pain. Her past medical history includes Crohn’s disease diagnosed in 1976, requiring total colectomy and end ileostomy in 1977 and a small bowel resection for structuring in 1980. A cholecystectomy for biliary pancreatitis was performed in 2004. Physical exam revealed a comfortable patient with normal vital signs and tenderness to deep palpation in the right upper quadrant. Laboratory investigations revealed a total bilirubin of 25.6 μmol/L; aspartate aminotransferase 73 IU/L; alanine aminotransferase 174 IU/L; gamma-glutamyl transferase 310 IU/L; alkaline phosphatase: 243 IU/L; amylase: 91 IU/L; lipase: 87 IU/L; CRP: 87.5 mg/dL, and a white blood cell count of 8.3 × 109/L. Computed tomography scan demonstrated a metallic object within the CBD with dense material organised around it. The CBD was dilated to 2.3 cm with intra-hepatic biliary duct dilation. The patient was diagnosed with subacute CBD obstruction from choledocholithiasis with gall stone formation around a surgical clip nidus. Endoscopic retrograde cholangiopancreatography (ERCP) with sphincterotomy was performed and the CBD stone was extracted and all symptoms and laboratory abnormalities resolved (Figure 17).

Conclusions. Up to 80 cases of post-cholecystectomy and post LC clip migration with biliary stone formation have been reported in the literature. Most cases occur in female patients with a median age of 60 years old. The primary indications for cholecystectomy in these patients were acute or chronic cholecystitis or biliary pancreatitis. The median time between the cholecystectomy and the development of symptoms and clip migration with gallstone formation was 26 months post-cholecystectomy. Most were successfully treated by ERCP. No explanation or risk factors have been validated to clarify how the clips migrated in the common bile duct.

Funding Agencies: None


Genetic Analysis of Pancreatic Cyst Fluid for GNAS and KRAS Mutations to Aid Pancreatic Cyst Classification
L. Ayres,1 J. Ho,2 B. Sheffield,2 C. Carter,2 A. Weiss,1 M. Byrne,1 D. Schaeffer,2 and F. Donnellan1

1Pancreas Centre British Columbia and Division of Gastroenterology, Vancouver General Hospital, Vancouver, BC, Canada
2Anatomical Pathology, Vancouver General Hospital, Vancouver, BC, Canada

Background. Endoscopic ultrasound (EUS) guided fine needle aspiration (FNA) is routinely used to evaluate pancreatic cysts. Cytology and biochemistry are limited in their ability to fully characterize cystic lesions. Recently, analysis of cyst fluid for the presence of genetic mutations has become available. KRAS and GNAS mutations have been shown to correlate with intraductal pancreatic mucinous neoplasia (IPMN).

Aims. To assess the correlation of KRAS and GNAS mutations with cytological diagnosis and surgical pathology where available.

Methods. Eleven consecutive patients attending a tertiary referral centre for assessment of pancreatic cysts had cyst fluid genetic analysis for KRAS and GNAS mutations where there was a sufficient quantity of fluid so as not to compromise routine cytology and biochemistry.

Results. Of 11 patients, 4 had wild type (WT) KRAS/GNAS with corresponding cytological diagnoses of pseudocyst, benign columnar cells and liposarcoma (one cytological specimen was non-diagnostic). Four patients had a mutation (3 KRAS mutations alone, 1 dual mutations) with corresponding cytological/surgical diagnoses of IMPN and pancreatic adenocarcinoma. In three cases it was not possible to amplify the DNA.

Where both cytology and genetic analysis was available there was 100% concordance of WT and mutations of KRAS and/or GNAS with non-mucinous lesions and mucinous/malignant lesions respectively.

Conclusions. Genetic analysis of pancreatic cyst fluid is helpful in differentiating non-mucinous and mucinous lesions.

Funding Agencies: Pancreas Centre British Columbia


Performance of EUS-Guided Fine Needle Biopsy of Solid Pancreatic Masses
L. Ayres,1 E. Lam,2 and J. Telford3

1St. Paul’s Hospital, Vancouver, BC, Canada
2Univ. of California, Vancouver, BC, Canada
3University of British Columbia, Vancouver, BC, Canada

Background. Fine needle biopsy (FNB) has been developed to procure histological samples of solid lesions during endoscopic ultrasound (EUS).

Aims. To assess the accuracy of a new FNB needle in solid pancreatic masses.

Methods. A retrospective review of 45 consecutive adult patients with solid pancreatic masses who underwent EUS-guided FNB by two experienced endosonographers at St. Paul’s Hospital using the 25 gauge Sharkcore needle was performed. Diagnostic accuracy of FNB was compared to gold standard surgical pathology (where available) or six month follow up. If histology was suspicious for malignancy but further sampling was required to prove malignancy, this was categorized as a false negative, however if further sampling was not judged necessary by the interdisciplinary team this was categorized as a true positive.

Results. 50 procedures were carried out on 45 patients. Diagnosis was adenocarcinoma (69%), lymphoma (8.8%), post operative inflammatory (4.4%) reactive/inflammatory change (4.4%) and chronic pancreatitis, neuroendocrine tumour, sarcoma, metastatic melanoma, glandular atypia, normal pancreas all (2.2%). The site of the tumor was pancreatic head in (67%), uncinate process (6.7%), genu (6.7%), body (11%) and tail (4.4%). There were no technical failures of the needle. The median number of passes was 2 (range 2–4). Sufficient tissue was obtained in 100%. Performance characteristics per procedure were: diagnostic accuracy 90%, sensitivity 88.4% and specificity 100%. Significant complications occurred in 4%: bleeding and pancreatitis .

Conclusions. Our initial experience using the Sharkcore FNB for solid pancreatic masses demonstrates a high diagnostic yield and an acceptable complication rate.

Funding Agencies: None

Pediatric Liver Disease

Poster of Distinction


An Evaluation of the Role of Transient Elastography in Assessing Pediatric Cystic Fibrosis Associated Liver Disease in Children with Cystic Fibrosis
S. Lam,1 H. Machida,2 R. Myers,1 C. Ortiz-Neira,1 S. Martin,2 J. Yap,3 and J. deBruyn1

1University of Calgary, Calgary, AB, Canada
2Alberta Children’s Hospital, Calgary, AB, Canada
3University of Alberta, Edmonton, AB, Canada

Background. Cystic fibrosis associated liver disease (CFLD) and its complications are increasingly recognized as the highest non-pulmonary cause of death in children with CF. The gold standard of liver biopsy for diagnosis of CFLD has limitations, including invasiveness, association with morbidity, and poor practicality for screening in children. Early ultrasonographic (US) changes may be subtle and subject to inter-observer variability.

Aims. The primary objective was to evaluate the diagnostic properties of Transient Elastography (TE) using FibroScan in children with CF for detection of CFLD, as defined by EuroCare Criteria. The secondary objective was to identify factors associated with the presence of CFLD.

Methods. Children from the Southern Alberta cystic fibrosis clinic at the Alberta Children’s Hospital underwent liver stiffness measurements (LSM) by TE. Sensitivity, specificity, and receiver operator characteristic (ROC) curve of TE were calculated and compared to EuroCare criteria for diagnosis of CFLD (≥2 of the following: persistent abnormal liver biochemistry over 12 months, hepatosplenomegaly, or US abnormalities). Age, anthropometrics, hepatosplenomegaly, genotype, lung and pancreatic function, history of small bowel bacteria overgrowth and meconium ileus, severity of liver disease on US with validated scoring systems, and past medications were examined to determine any correlation with the presence of CFLD.

Results. Forty-one of 130 patients in the CF clinic completed the study. The median age was 8.5 years, (interquartile range (IQR) 5–12 years) with 56% females. The prevalence of CFLD was 9.7% (). The TE failure rate was 7.3%. (); An 18 month and 20 month old child were uncooperative, a 6 year old with autism spectrum disorder did not complete testing due to anxiety). Children with CFLD had significantly higher median LSM 13.6 kPa (IQR 5.7–27.8 kPa) compared to those without CFLD 4.6 kPa (IQR 3.2–5.1 kPa) (). When a cut-off value of ≥5.3 kPa was used, the sensitivity, specificity, positive and negative predictive values were 100% (95% CI 39–100%), 87% (95% CI 71–95%), 44% (95% CI 26–64%), 100%. A ROC curve for detecting CFLD with this cut off was 0.93 (95% CI 0.87–0.98). No examined factors showed association with CFLD.

Conclusions. TE is well tolerated and successful in the majority of children with CF. TE has a role as a useful non-invasive test to screen and diagnose CFLD in children with CF.

Funding Agencies: Alberta Children’s Hospital Research Institution Small Research Grant


Role of Transient Elastrography in Assessment of Cystic Fibrosis-Associated Liver Disease
J. Woolfson, S. Raveendran, M. Chilvers, R. Schreiber, and O. Guttman

BC Children’s Hospital, Vancouver, BC, Canada

Background. Cystic Fibrosis-associated liver disease (CFLD) occurs in 30% of patients and is the 3rd most common cause of mortality in CF patients. Diagnosis is challenging as specific tests for detection of fibrosis in pediatric CFLD have not been developed and existing investigations do not correlate well with presence or severity of disease. Liver biopsy is rarely indicated because of the patchy nature of the disease. Transient Elastrography (TE) is a rapid non-invasive method for assessing liver fibrosis. Studies suggest it may be a valuable tool in pediatric patients, though its role in detecting CFLD has only begun to be explored. AST : platelet ratio index (APRI) has been validated as a surrogate marker of hepatic fibrosis in chronic liver diseases.

Aims. The purpose of this study was to assess the utility of TE and to determine the role of APRI and standard biochemistry in identifying liver fibrosis in CF patients.

Methods. Patients 2–18 years old were recruited from the British Columbia Children’s Hospital CF clinic. Charts were reviewed for demographic and clinical data including bloodwork and abdominal imaging. Each patient underwent TE by a single trained operator. Patients were determined to have CFLD using standard criteria based on hepatic biochemistry, imaging and clinical examination. Where the original basis for CFLD diagnosis was unclear from chart review, patients maintained on ursodiol were included in the CFLD group.

Results. 55 patients were included in the study (50.9% male, mean age 11.6 (range 5.1–17.5) years). 49% were homozygous for ΔF508 gene, 36.3% were heterozygous, 7.3% had other mutations and 7.3% were genotype unknown. 22 patients had a diagnosis of CFLD (40%) and 20 of these were on ursodiol (90.9%). Two patients had ultrasound findings of cirrhosis and one had portal hypertension. Of the 22 CFLD patients, 45.5% were male (), 59% were homozygous for ΔF508 () and 90.9% were pancreatic insufficient (). All mean liver enzymes were higher in the CFLD group, significantly ALT () and ALP (). Mean TE values were significantly higher in the CFLD group (5.92, range 3.9–16.5) versus no liver disease (4.54, range 2.1–7.2; ). APRI was higher in the CFLD group (0.396 versus 0.324, ). Linear regression showed a positive association between TE value and APRI (Slope 0.058; CI 0.038–0.79; R2 = 0.386).

Conclusions. CFLD is one of the leading causes of morbidity in CF, but limitations of existing tests hamper diagnosis and monitoring. In this study, TE values were significantly higher in CFLD patients and correlate with APRI values, suggesting that TE may have clinical applications for identifying and following patients with this condition. Further research is needed at a larger scale to determine TE cutoff values for diagnosing CFLD.

Funding Agencies: None


Liver Transplant in an Infant Presenting with Hepatic Failure Secondary to Severe Pyruvate Kinase Deficiency
M. Chartier,1 M. Paganelli,2 N. Ahmed,3 and F. Alvarez2

1CHU Ste-Justine, Montreal, QC, Canada
2CHU-Sainte Justine, Montreal, QC, Canada
3McGill University Health Centre, Montreal, QC, Canada

Background. Pyruvate kinase deficiency (PKD) is the most common cause of congenital non-spherocytic chronic hemolytic anemia and results from an erythrocyte enzyme defects. Patients with pyruvate kinase deficiency can have a broad spectrum of clinical manifestations, ranging from mild asymptomatic anemia to severe and transfusion dependent anemia. Most patients normally present with some degree of hemolysis, hyperbilirubinemia, anemia and splenomegaly. Only few reports have documented associated severe progressing liver failure.

Aims. To describe the case of an infant with severe pyruvate kinase deficiency leading to liver failure and requiring liver transplantation.

Methods. We retrospectively reviewed the medical chart of our patient with pyruvate kinase deficiency and liver failure. All articles about such a rare complication of pyruvate kinase deficiency published in the English literature from 1962 o October 2015 were reviewed.

Results. Our patient presented with severe hemolytic anemia and cholestasis at birth, requiring double exchange transfusion and repeated transfusions thereafter. He subsequently developed progressive cirrhosis, portal hypertension, ascites and liver failure requiring prolonged hospitalization and biweekly paracentesis. Two liver biopsies done more than one month apart showed progressive liver fibrosis. Despite extensive investigations, the only identified etiology for cholestasis and liver failure was compound heterozygous mutations for PKD and single heterozygous mutation for ABCB4, the latter being a likely benign variant. The patient was transplanted at 6 months of age and underwent a splenectomy during the same intervention. To the best of our knowledge, only three cases of severe hepatic failure secondary to PKD have been reported but this is the first to have successfully undergone liver transplant.

Conclusions. The hepatic failure in patients with severe pyruvate kinase deficiency is most likely multifactorial, involving prenatal hemolysis with subsequent bile ducts obstruction, minimal inflammation secondary to iron overload and extramedullary hematopoiesis, but the most likely explanation is that genetic mutations of PKLR in our patient affect both the expression of PK-R (in erythrocytes) and PK-L (in hepatocytes) with an inappropriate compensation of PKM2, leading to severe and fatal enzymatic defect.

Funding Agencies: None


Acute Deterioration in Metabolic Control in a Child Post Liver Transplant for Maple Syrup Urine Disease
P. Kawada, S. Jain, A. Chan, and J. Yap

University of Alberta, Edmonton, AB, Canada

Background. In Maple Syrup Urine Disease (MSUD) the inherited deficiency is in the gene that codes for the branched-chain α-keto acid dehydrogenase (BCKDH) complex. The classical form of MSUD is the severest. BCKDH degrades branched-chain amino acids (BCAA), therefore deficiency leads to accumulation of leucine, isoleucine and valine along with allo-isoleucine which is diagnostic. Treatment is dietary protein restriction although acute decompensation, with high leucine, ketosis and metabolic acidosis with neurological symptoms, can still occur. Liver transplantation supplements BCKDH activity and stabilizes metabolic control. Acute deterioration following liver transplantation has only previously been described once.

Aims. As more patients are being transplanted for MSUD, it is important to recognize that with significant intercurrent illness, there is potential for metabolic decompensation; particularly in those with previously severe MSUD.

Methods. Case report.

Results. A 4 year old boy 3 years post liver transplant for severe, classical MSUD presented acutely with ataxia and profoundly elevated BCAA levels. Prior to admission, the post transplant course had been stable with normal allograft function and no acute metabolic decompensation. Post-transplant neurodevelopment was normal, apart from mild fine motor delay. The child had fever, diarrhea and vomiting 2 days prior to admission. At presentation to the emergency room, the child was lethargic, drowsy, dysarthric and ataxic. An impressive rise in the BCAA was seen (Table 23). He was rehydrated with two normal saline boluses and maintained on IV fluids which included 5% dextrose with improvement in symptoms. No additional metabolic treatment was required and he was discharged home the next day. There was no residual neurological sequelae.

Discussion. Post liver transplant, classical MSUD patients achieve satisfactory metabolic control allowing liberalization of dietary protein restriction. There has only been a single report of acute metabolic decompensation post transplant in a child who developed transient leucinosis in the context of gastroenteritis with severe dehydration. It is postulated that catabolic illness with dehydration transiently affects clearance of BCAA by the liver graft.

Conclusions. Classical MSUD liver recipients who are significantly unwell, especially if dehydrated, should have early BCAA monitoring and a sick day protocol that include high dextrose solutions.

Funding Agencies: None

Viral Hepatitis

Poster of Distinction


Adherence to Response-Guided Peg-Interferon and Ribavirin for People Who Inject Drugs with HCV Genotype 2/3 Infection: The Activate Study
E. Cunningham,1 B. Hajarizadeh,1 O. Dalgard,2 B. Conway,3 G. Foster,4 P. Bruggmann,5 M. Backmund,6 G. Robaeys,7 T. Swan,8 J. Amin,1 P. Marks,1 S. Quiene,1 M. Weltman,9 D. Shaw,10 A. Dunlop,11 M. Hellard,12 J. Bruneau,13 C. Staehelin,14 G. Dore,1 and J. Grebely1

1The Kirby Institute, UNSW, Randwick, NSW, Australia
2Akershus University Hospital, Oslo, Norway
3Vancouver Infectious Diseases Center, Vancouver, BC, Canada
4The Liver Unit, Queen Marys University of London, London, UK
5Arud Centres for Addiction Medicine, Zurich, Switzerland
6Ludwig Maximilians-University Munich, Munich, Germany
7Department of Gastroenterology and Hepatology, Ziekenhuis Oost Limburg, Genk, Belgium
8Treatment Action Group, New York, NY, USA
9Nepean Hospital, Sydney, NSW, Australia
10Royal Adelaide Hospital, Adelaide, NSW, Australia
11School of Medicine and Public Health, University of Newcastle, Newcastle, NSW, Australia
12Burnet Institute, Melbourne, VIC, Australia
13Universite de Montreal, Montreal, QC, Canada
14Division of Infectious Diseases, University Hospital and University of Bern, Bern, Switzerland

Background. Adherence to HCV therapy impacts SVR, but data is limited among people who inject drugs.

Aims. This study assessed PEG-IFN adherence and associated factors.

Methods. Participants with HCV G2/3 who had recently injected drugs (last 12 weeks) or receiving opioid substitution therapy were recruited (2012–14) and received directly observed PEG-IFN and self-administered RBV. Participants with an RVR received 12 weeks (shortened duration) and those without RVR received 24 weeks (standard duration) therapy. The primary endpoint was SVR12 and 80/80 PEG-IFN adherence.

Results. Overall, 93 initiated HCV treatment (mean age 42; 82% men; 87% G3; 55% injected drugs in the last month; 70% on OST at baseline). Sixty-five percent () received shortened treatment, while 29% () received standard treatment. Six participants discontinued prior to week 4. 80/80 PEG-IFN adherence was 81% (), 6% missed ≥1 dose (on-treatment adherence >99%) and 27% () discontinued early (virological failure (), lost to follow up/unwillingness () and side effects ()). Treatment completion was higher in those receiving 12 versus 24 weeks of therapy (97% versus 52%, , Figure 19). Injecting drugs in the last month did not impact 80/80 adherence (81% versus 80%, ). Treatment duration of 12 weeks (those with RVR) was the only factor associated with ≥80/80 adherence (versus 24 weeks; AOR 40.6, 4.9–338.0). SVR was 63% (82%—12 weeks; 37%—24 weeks), and was associated with ≥80/80 adherence (79% versus 0%, ).

Conclusions. High adherence to therapy was observed, irrespective of recent injecting drug use. Sub-optimal exposure was driven by early treatment discontinuation, not missed doses during therapy. Treatment completion and adherence were higher in people receiving 12 weeks of therapy compared to 24 weeks.

Funding Agencies: Merck & Co.


Ledipasvir/Sofosbuvir (LDV/SOF) for 8 Weeks in Genotype 1 (GT1) Treatment-Naïve (Tn) Non-Cirrhotic (NC) Patients with HCV Viral Load (VL) <6 Million IU/mL (6 m); A Comparative Analysis of the Phase-3 Ion-3 Efficacy Data to Real World Effectiveness (RWE)
N. Tsai,1 M. Curry,2 P. Buggisch,3 S. Milligan,4 D. Mumm,5 M. Natha,6 E. Eggleton,6 B. Kreter,6 D. Brainard,6 and K. Kowdley7

1Queens Medical Center, Honolulu, HI, USA
2Beth Israel, Boston, MA, USA
3ifi-Institute for Interdisciplinary Medicine, Hamburg, Germany
4TRIO Health Analytics, La Jolla, CA, USA
5Gilead Sciences Canada, Mississauga, ON, Canada
6Gilead Sciences Inc., Foster City, CA, USA
7Swedish Medical, Seattle, WA, USA

Aims. The optimal duration of therapy to achieve SVR depends on multiple factors. Patients treated with LDV/SOF with 8, 12 or 24 weeks achieved SVR12 from 94–100% in the ION Phase 3 studies. Duration of therapy is based on treatment history, cirrhosis status and baseline VL. In a post-hoc analysis of the ION-3 (TN, NC patients) 8 week data, a VL < 6 M was shown to be the best predictor of SVR.

Methods. Diverse RWE LDV/SOF data is emerging from single-center (Buggisch) and multicenter (GECCO) retrospective chart reviews to large multicenter prospective cohorts (TARGET, TRIO). In this analysis, the Phase-3 ION-3 data is compared with several real-world cohorts. Patient demographics, characteristics and SVR12 data has been collated and compared.

Results. The ION-3 post-hoc analysis reported 123 patients who were TN, NC and VL < 6 M and treated with 8 weeks of LDV/SOF. Mean age was 52, 22% black, 72% GT1a; the SVR12 was 97%. In TARGET; 87% of GT1, TN, NC patients had a baseline VL < 6 M; 35% of these received an 8-week regimen. Preliminary SVR4 in 59 patients is 97%. In the TRIO cohort, 8-week therapy was initiated for 37% of patients with baseline VL < 6 M, Mean age was 57, 70% GT1a, 20% black and SVR12 data in 181 patients is 97%. Buggisch et al. shows 100% SVR12 (). Mean age is 50, 52% GT1a and 88.1% had comorbid conditions. The GECCO cohort also includes patients with baseline HCV VL > 6 M, advanced fibrosis and HIV/HCV co-infection; SVR4 is 100% (). Low rates of Adverse Events (AEs), relapse rates and discontinuations were seen in all 4 cohorts. Complete SVR12 data is expected in these cohorts by the time of presentation.

Conclusions. LDV/SOF for 8 weeks in the appropriate patient population yielded high SVR rates in ION-3. Analysis of real world effectiveness data from several diverse & heterogeneous cohorts from the US & EU show SVR outcomes that were consistent with the Phase-3 ION-3 results and supports the use of 8 weeks LDV/SOF in treatment-naive, non-cirrhotic GT1 patients with a baseline HCV VL <6 million IU/mL.

Funding Agencies: Gilead Sciences Inc.


Long Term Efficacy and Safety of Tenofovir DF (TDF) in Chronic Hepatitis B Patients (CHB) with Documented Lamivudine Resistance (LAM-R): 5 Year Results from a Randomized, Controlled Trial
S. Fung,1 H. Hann,2 M. Elkashab,3 M. Khan,4 T. Berg,5 M. Fabri,6 A. Horban,7 I. Sporea,8 and J. Flaherty9

1Toronto General Hospital, Toronto, ON, Canada
22Thomas Jefferson University, Philadelphia, PA, USA
3Toronto Liver Centre, Toronto, ON, Canada
4Gilead Sciences Canada, Inc, Mississauga, ON, Canada
5University Hospital Leipzig, Leipzig, Germany
6Clinic for Infectious Diseases, Novi Sad, Serbia
7Medical University of Warsaw, Warsaw, Poland
8Spitalul Clinic Judetean de Urgenta, Timisoara, Romania
9Gilead Sciences, Inc, Foster City, CA, USA

Background. In CHB patients with LAM-resistance (LAM-R), TDF has shown efficacy comparable to FTC/TDF and no detectable TDF resistance at 2 years (Gastroenterology 2014; 146:980-88).

Aims. The final 5 year efficacy and safety results from this trial are presented.

Methods. CHB patients on LAM with HBV DNA >3 log10 IU/mL and with documented LAM-R (INNO-LiPA Multi-DR, v3) were randomized (1 : 1) to TDF or FTC/TDF and followed in a blinded fashion for 5 years.

Results. Two hundred eighty patients were randomized; 239 (85%) completed 5 years of treatment. At baseline, mean age was 47 years, most were male (75%) and non-Asian (66%); 53% were HBeAg-negative, and HBV genotype distribution (A®D) was 22%, 13%, 19%, and 43%, respectively. Mean (SD) HBV DNA was 5.7 (1.9) log10 IU/mL, and 42% had ALT ≤ ULN at baseline. At Year 5, virologic, serologic, and biochemical responses were similar among groups, and remained stable from Year 2 to 5 (Table 24). Nine patients (4-TDF, 5-FTC/TDF) discontinued due to an adverse event, including increased serum creatinine in 1 patient. Hepatocellular carcinoma was reported in 4 (1.4%) patients. Confirmed renal safety endpoints (both groups combined) over 5 years were: CrCL <50 mL/min in 19 (6.8%) patients (12 requiring dose modification), increases in serum creatinine of 0.3 and 0.5 mg/dL from baseline in 21 (7.5%) and 2 (0.7%) patients, respectively, and serum phosphorus <2 mg/dL in 3 (1.1%) patients. For both groups combined, mean declines in BMD (g/cm2) from baseline for hip and spine BMD, respectively, were 1.7% and 1.5% at Year 2, and 2.5%, and 1% at Year 5. Seven patients experienced fracture (all except 1 were trauma-related). No TDF resistance was detected through 5 years of treatment by population sequencing.

Conclusions. In LAM-R patients with CHB treated for 5 years with TDF, a high rate of HBV DNA suppression was achieved and maintained with no detectable TDF resistance. There is no apparent advantage of combination FTC/TDF in this population. Renal events associated with TDF occurred in up to 7.5% of patients, and average losses in bone mineral density of 1–2.5% were observed.

Funding Agencies: Gilead Sciences, Inc.


On-Treatment HCV RNA Decline in Pre- and Post-Liver Transplant Patients with Different Degrees of Fibrosis and Cirrhosis: A Combined Analysis of the Solar Trials
T. Welzel,1 R. Reddy,2 M. Manns,3 S. Flamm,4 D. Mutimer,5 E. Gane,6 R. Hyland,7 J. McHutchison,7 D. Mumm,8 D. Samuel,9 M. Charlton,10 X. Forns,11 G. Everson,12 S. Zeuzem,1 and N. Afdhal13

1Johann Wolfgang Goethe University Medical Center, Frankfurt, Germany
2University of Pennsylvania, Philidelphia, PA, USA
3Hannover Medical School, Hannover, Germany
4Northwestern University Feinberg School of Medicine, Chicago, IL, USA
5Queen Elizabeth Hospital, Birmingham, UK
6University of Auckland, Auckland, New Zealand
7Gilead Sciences Inc., Foster City, CA, USA
8Gilead, Mississauga, ON, Canada
9Universite Paris-Sud, Villejuif, France
10Intermountain Medical Center, Murray, UT, USA
11Liver Unit, IDIBAPS and CIBEREHD, Barcelona, Spain
12University of Colorado, Denver, CO, USA
13Beth Israel Deconess Medical Cnter, Boston, MA, USA

Aims. In the SOLAR-1 and SOLAR-2 studies, ledipasvir/sofosbuvir (LDV/SOF) + ribavirin (RBV) for 12 or 24 weeks resulted in high SVR rates in genotype (GT) 1 or 4 HCV-infected patients with decompensated cirrhosis or who were liver transplant recipients. In this large combined post hoc analysis, we investigate whether on-treatment HCV RNA response varied by patient population and/or was predictive of treatment outcome.

Methods. Data from the identically designed SOLAR-1 & 2 studies were combined. Six groups of GT 1 or 4 patients were randomized to receive 12 or 24 wks of LDV/SOF + RBV treatment: patients without transplant and either CPT B cirrhosis, or CPT C cirrhosis; or patients who have undergone transplantation (post-OLT) and who were either without cirrhosis (F0 to F3), CPT A cirrhosis, CPT B cirrhosis, or CPT C cirrhosis. For analysis of early viral kinetics, the 12 and 24 wk treatment durations were combined. Serum HCV RNA was quantified using Roche CAP/CTM v2.0 with a lower limit of quantitation (LLOQ) of 15 IU/mL.

Results. Patients with advanced liver disease had SVR12 rates of 86–89%. Patients who were post-OLT had SVR rates that ranged from 96–98% in those with F0-F3 fibrosis to 60–75% of patients with severe hepatic impairment. Rapid HCV RNA declines were observed in all treatment groups. The majority of subjects achieved HCV RNA <LLOQ at WK 4. The percentages of patients who were <LLOQ or <LLOQ Target Not Detected (TND) at Wks 1, 2, 4, and 6 will be presented. Among baseline variables, high viral load and the presence of liver cirrhosis were associated with slower on treatment HCV RNA decline. An analysis of baseline variables, on-treatment HCV RNA and SVR12 will be presented.

Conclusions. Early viral decline during treatment is rapid in patients with compensated and decompensated HCV before and after liver transplantation. The proportion who achieved HCV RNA <LLOQ at Wk2 was highest in the patients without cirrhosis, similar among patients with CTP A and B cirrhosis, and lowest in those with CPT C cirrhosis.

Funding Agencies: Gilead Sciences Inc.


Examination of Genetic Variation within the Hepatitis C Virus NS3 Helicase Using Deep Sequencing
C. Ablenas,1 N. Bensoussan,1 M. Powdrill,2 L. Mayrthaler,3 J. Pezacki,2 and M. Gotte4

1McGill, Montreal, QC, Canada
2University of Ottawa, Ottawa, ON, Canada
3Ludwig-Maximilians-Universität, Munich, Germany
4University of Alberta, Edmonton, AB, Canada

Background. Hepatitis C virus (HCV) subgenomic replicons are extremely useful for the study of viral replication. Replication enhancing mutations (REMs), also referred to as cell culture adaptive mutations, have been identified that result in efficient RNA amplification in the highly permissive Huh-7 hepatoma cell line. Within the HCV replicon REMs cluster to NS5a, as well as to the amino terminus of the NS3 helicase (NS3h), and at two positions in NS4b. In the context of NS3h, the majority of REMs map to domain 1, with few REMs in domains 2 and 3. Domains 1 and 2 are highly structurally related, containing the ATP binding site and the majority of the contacts with the RNA substrate, while domain 3 is primarily structural.

Aims. To use a deep sequencing approach to investigate genetic variation in NS3h using the replicon system. We compared genetic variation in domain 1, which contains the majority of NS3h REMs reported in the literature, to that in domain 3 where few REMs have been reported.

Methods. Deep sequencing of amplicons covering the different domains of NS3h was performed to detect minor variants arising during replication. Briefly, RNA was extracted from Huh7 cells harbouring the 1b replicon and reverse transcribed into cDNA. PCR amplicons were then amplified using barcoded PCR primers with 454 adaptor sequences and subjected to pyrosequencing on a GS Junior sequencer (Roche). The raw reads were processed, and minor variants calculated using Roche amplicon variant software. Select mutants were introduced back into the replicon by site-directed mutagenesis, and the replication capacity was measured using a luciferase reporter.

Results. The error rate calculated from variants sequenced in a region of domain 1 in NS3h was approximately 2-fold higher than that for a region in domain 3. This suggests a higher tolerance for variation in domain 1 as compared to domain 3, consistent with the greater number of REMs previously reported in domain 1. In agreement with previous reports from our group, the majority of the variants consisted of transitions due to the high bias by the NS5b polymerase to form G/U or U/G mismatches, with only a small number of transversions. At the level of the individual variants, we noted five mutations in domain 1 and one mutation in domain 3 that were detected repeatedly over 2 to 3 independent RNA transfections. When the replication capacities of select mutants were tested, four mutants were found to replicate ~4–9 fold higher than WT.

Conclusions. We found that genetic variation was better tolerated in domain 1 as compared to the more structural domain 3. In addition, certain nucleotide positions were identified as hot spots for specific mutations.

Funding Agencies: CIHR, CanHepC, and FRQS


RNA Pull-Down Strategies to Investigate the Roles of Microrna-122-Associated Complexes in Hepatitis C Virus Infection
A. Bernier and S. Sagan

McGill University, Montreal, QC, Canada

Background. Approximately 200 million individuals worldwide are infected by hepatitis C virus (HCV), including more than 268 000 Canadians. MicroRNA-122 (miR-122) is a highly abundant liver-specific microRNA shown to interact at two “tandem" microRNA-binding sites in the 5′ end of the HCV genome. This unusual interaction promotes HCV RNA accumulation in both HCV-infected cells and the livers of infected patients. Mutation, truncation, or exchange of the 3′ terminal ribonucleotides of miR-122 for deoxynucleotides reduces HCV RNA accumulation. However, these nucleotides are not required for canonical miRNA activities. This suggests that sequences in the 3′ tail of miR-122 may mediate important interactions with viral or cellular factors involved in HCV RNA accumulation.

Aims. We hypothesize that miR-122 forms a distinct complex with host and/or viral proteins that together mediate HCV RNA accumulation. Hence, we aim at identifying and characterizing host and viral factors associated with non-canonical miR-122 complexes in HCV-infected cells to understand miRNA-mediated viral RNA accumulation and identify novel antiviral targets.

Methods. Alkyne-tagged miR-122 molecules are transfected into HCV RNA-harboring Huh-7 cells. Following miR-122 biotinylation by a click reaction, miR-122 ribonucleoprotein complexes from naïve and HCV-infected cells are isolated by streptavidin affinity purification. MiR-122-associated proteins are then analyzed by SDS-PAGE and liquid chromatography tandem mass spectrometry. Comparison of miR-122 complexes from naïve and miR-122 site 1 or 2 mutant HCV-infected cells will allow the identification of proteins acting specifically at each site of the HCV genome. Proteins interacting with the 5′ end of HCV in the absence of miR-122 will also be identified following pull-down of a mutant HCV RNA containing a BoxB stem-loop using biotinylated-λN peptide.

Results. We demonstrate that alkyne-tagged miR-122 molecules are functional in mediating HCV RNA accumulation in Huh-7 cells. We show that the click reaction is stable under physiological conditions and permits efficient labeling and affinity purification of miR-122 molecules in cell lysates. Western blot of affinity purified miR-122 complexes show enrichment in the RNA-induced silencing complex (RISC) protein Argonaute 2. Additionally, we show that a mutant HCV RNA with a BoxB loop in the place of stem-loop I is replication competent.

Conclusions. We expect that the results will provide insight into a novel microRNA “capping complex” as well as a non-canonical “microRNA enhancing complex”. We anticipate that we will identify novel host-virus interactions important for viral replication that will provide new targets for therapeutic intervention.

Funding Agencies: CIHR, CanHepC; FRSQ


Two Serologically Distict Forms of Occult Hepadnaviral Infection Accompanied by Hepatocellular Carcinoma Development Identified in the Woodchuck Model of Hepatitis B
T. Michalak, P. Mulrooney-Cousins, and N. Churchill

Memorial University, St. John’s, NF, Canada

Background. Woodchuck hepatitis virus (HBV) is molecularly and pathogenically closely related to hepatitis B virus (HBV). Both viruses cause similar liver pathology, where acute hepatitis (AH) can progress to chronic hepatitis (CH) and hepatocellular carcinoma (HCC). They also establish persistent, asymptomatic (occult) infections.

Aims. To determine molecular and immunological characteristics, and pathological consequences of experimentally induced occult hepadnaviral persistence in WHV-infected woodchucks. To recognize if different amounts of infectious WHV establish distinct forms of asymptomatic WHV carriage.

Methods. Woodchucks were followed for life after i.v. injection with 10 to 1010 DNase-digestion protected virions and examined for WHV DNA in serial serum, PBMC and liver biopsies, and for virus replication intermediates in PBMC and liver by PCR/nucleic acid hybridization-based assays. Serum WHV surface antigen (WHsAg), antibodies to WHV core (anti-WHc) and WHsAg (anti-WHs), liver histology, WHV-specific T cell responses, and infectivity of persisting WHV were assessed.

Results. Animals injected with >103 virions developed serum WHsAg/anti-WHc-positive AH. In the majority (~90%) of animals, AH resolved, serum WHsAg seemingly completely cleared within 6–18 weeks post-infection (p.i.), anti-WHs developed, while anti-WHc and low levels of WHV DNA persisted for life in serum (≤100–200 copies/mL), PBMC and liver (~0.1–10 copies/μg DNA) with evidence of WHV replication in the liver and the immune system. Intermittent minimal to moderate liver inflammation protracted during lifespan and HCC developed in ~20% of the animals. This form was termed secondary occult infection or SOI. The remaining ~10% of animals with AH progressed to CH and ~80% of them developed HCC. Woodchuck injected with 10 or 100 virions established serum WHsAg/anti-WHc-negative infection in which WHV replication was initially restricted to the lymphatic system. This primary occult infection (POI) with time (~3 year p.i.) spread to the liver and HCC developed in ~20% of animals in 5 years p.i. The virus persisting during both SOI and POI caused serum WHsAg-positive hepatitis advancing to HCC in some cases when concentrated and administered to virus-naïve woodchucks. SOI coincided with WHV-specific B and T cell immune responses and protection against reinfection, but POI was accompanied by virus-specific T cell but not B cell response and by a lack of immune protection against challenge with liver pathogenic doses of WHV (>103 virions). WHV DNA integrated into the liver and the immune system in both SOI and POI.

Conclusions. The amount of WHV determines whether anti-WHc/anti-WHs-positive SOI or totally seronegative POI is established. These two forms of occult WHV infection.

Funding Agencies: CIHR


Distribution of Hepatitis C Risk Factors and HCV Treatment Outcomes among Central Canadian Aboriginals
P. Parmar,1 D. Corsi,2 and C. Cooper3

1Faculty of Medicine, University of Ottawa, Ottawa, ON, Canada
2Ottawa Hospital Research Institute, Ottawa, ON, Canada
3Division of Infectious Diseases, Department of Medicine, Univesity of Ottawa, Ottawa, ON, Canada

Background. Aboriginal Canadians face higher levels of lifestyle risk factors for hepatitis C exposure including drug use and incarceration.

Aims. We examined multiple risk factors clustering among Aboriginals.

Methods. The Ottawa Hospital Viral Hepatitis Clinic Cohort (January 2000–August 2013) was evaluated. Demographic data, HCV infection risk factors, and HCV treatment outcomes were assessed. Markers of socioeconomic status were based on area-level indicators linked to postal code.

Results. 55 (2.8%) Aboriginal and 1923 (97.2%) non-Aboriginals were evaluated. Aboriginals were younger (45.6 versus 49.6 years, ). The distribution of gender (63.6% versus 68.3% male), HIV co-infection (9.1% versus 8.1%), genotype 1 infection (68.5% versus 65.4%), advanced fibrosis stage (29.2% versus 28.0% F2+), and SVR rate (56.3% versus 58.9%) was similar between Aboriginal and non-Aboriginals (). Aboriginal status was associated with a higher number of HCV risk factors, (mean 4.2 risk factors versus 3.1, ) with an odds ratio of 2.5 (CI 1.4–4.4) for having at least 4 risk factors. This was not explained after adjustment for markers of socioeconomic status including income, social deprivation, and poor housing. Multivariable logistic regression suggested that SVR was unrelated to Aboriginal status (). Aboriginal patients interrupted therapy more often due to lost-to-follow-up, lesser adherence and substance abuse (25.0% versus 4.6%) and serious adverse events (25.0% versus 21.3%, ).

Conclusions. Aboriginal Canadians have higher levels of HCV risk factors; even when adjusting for area-level socioeconomic status markers. Despite facing greater barriers to care, SVR rates were comparable with non-Aboriginals.

Funding Agencies: CIHR, OHTN


Direct Acting Antiviral Treatment Uptake in a Canadian HIV/Hepatitis C Co-Infected Population
S. Saeed, E. Strumpf, and M. Klein

McGill University, Montreal, QC, Canada

Background. In Canada approximately 14,000 people are co-infected with HIV and Hepatitis C (HCV). Co-infection accelerates progression to end-stage liver disease and is now one of the leading causes of death in this population. To reduce the clinical and health system burdens of advanced liver disease, co-infected individuals need to be treated and cured of HCV. Fortunately highly effective and well tolerated direct acting antivirals (DAAs) are available in Canada.

Aims. The aim of this study was to describe DAA treatment uptake in a co-infected population.

Methods. Using data from the Canadian Co-Infection Cohort (CCC) Study, we investigated second-generation DAA treatment initiations until database closure (early 2015). The CCC is a prospective longitudinal cohort of 1498 HIV/HCV co-infected individuals from 18 centers, representing ~23% the total co-infected population in Canada. Socio-demographic, clinical and behavioural information is collected via self-administered questionnaires/chart review bi-annually.

Results. Overall, 39 people in 2010; 35 (2011); 37 (2012); 37 (2013); 59 (2014); 15 (until Feb 2015) initiated HCV treatment. A total 66 people initiated DAAs. Sofosbuvir/ledipasvir were the most frequently used DAA combinations accounting for 18 (27%) of treatments, followed by 17 (26%) sofosbuvir/ribavirin +/− peg-interferon (peg-IFN), 7 (11%) simeprevir/sofosbuvir, 1 (2%) simeprevir/ribavirin/(peg-IFN) the remaining 23 (35%) treatments, were accessed through clinical trials. Table 25 compares demographic, clinical and behavioural characteristics of CCC participants who initiated second generation DAA therapy either by standard of care or through clinical trials and those eligible for treatment but who did not receive DAAs.

Conclusions. Health Canada approved simeprevir/sofosbuvir in late 2013, followed by ledipasvir, however provinces have been slow and restrictive at rolling out DAA treatment. Early signs do show that the absolute number of HCV treatments have increased 59% between 2013 & 2014, in the co-infected population. Sub-population under represented in this initial treatment uptake wave were; women, people of Aboriginal decent and people who inject drugs. Steps should be taken to increase the availability of treatment in these vulnerable populations.

Funding Agencies: CIHR, Can-Hep C


Ultrasound Predicts Steatosis in Patients with Chronic Hepatitis B
E. Kelly,1 R. Hudock,2 V. Feldstein,2 and M. Peters2

1Department of Medicine, The Ottawa Hospital, Ottawa, ON, Canada
2UCSF, San Francisco, CA, USA

Background. Inflammation and fibrosis may impair the ability of ultrasound to identify steatosis in patients with chronic hepatitis B (CHB).

Aims. We determined the accuracy of ultrasound (US) in grading steatosis in patients with CHB compared to liver biopsy, and examined clinical factors associated with steatosis.

Methods. This was a single-center, retrospective study of all non-transplanted CHB patients undergoing US and same day liver biopsies from 2004–2014. Steatosis was graded by ultrasound as 0: none, 1: mild, 2: moderate or 3: severe. Liver histology graded steatosis (0: <5%, 1: <33%, 2: <66%, 3: ≥66%), steatohepatitis, and staged fibrosis. Significant steatosis was defined as grade 2/3 for both biopsy and ultrasound. Obesity was defined as a body mass index (BMI) of >30 in non-Asians and >27 in Asians (Asia Pacific obesity classification). Clinical variables within 6 months of liver biopsy were collected and their association with steatosis analyzed by univariate logistic regression.

Results. We studied 109 patients with CHB with a median (IQR) age of 45 (37–54) and BMI of 25.3 (22.0–27.7). Patients were predominantly Asian (83%, ), male (62%, ), and HBeAg negative (62%, ). 27% of patients () were obese, 9% () had diabetes mellitus, 23% () hypertension, and 31% () hyperlipidemia and 21 (19%) met the definition for metabolic syndrome. 44% () of patients had any steatosis on liver biopsy; 8% () had significant steatosis. The absence of steatosis on US had excellent specificity in ruling out biopsy steatosis (92%). Steatosis on US accurately predicted presence of significant steatosis (grade ≥ 2) on liver biopsy (sensitivity 89%, specificity 94%, ). US imaging suggesting any amount of steatosis poorly predicted the presence of steatosis on liver biopsy (sensitivity 60%), but improved to 84% in obese individuals (). Predictors of biopsy steatosis on univariate analysis included diabetes (), hypertension (), hypercholesterolemia (), and BMI ().

Conclusions. Metabolic risk factors (diabetes, hypertension, hyperlipidemia and obesity) were common in our cohort and highly associated with steatosis. The absence of steatosis on ultrasound effectively ruled out steatosis on liver biopsy. Ultrasound may over-estimate mild steatosis in CHB, but had excellent sensitivity and specificity in identifying patients with significant steatosis on biopsy. Ultrasound accuracy in predicting any steatosis improved in obese individuals, including Asian Americans with lower BMI cutoffs. Abdominal ultrasound can be used to predict clinically important steatosis on liver biopsy in CHB patients which may assist counselling on lifestyle modifications and aggressive management of metabolic risk factors.

Funding Agencies: CIHR, None


The Effect of PTEN on HCV Infection
Q. Wu and Q. Liu

University of Saskatchewan, Saskatoon, SK, Canada

Background. Hepatitis C virus (HCV) infection causes serious global public health problems. There are more than 130 million chronic HCV patients worldwide. Hepatocellular carcinoma (HCC) is the most deadly clinical consequence of HCV infection. Phosphatase and tensin homolog deleted on chromosome 10 (PTEN) can suppress PI3K-AKT pathway, one of the most critical cancer-promoting pathways. PTEN is frequently mutated or deleted in tumors including HCC. However, the role of PTEN in HCV replication and pathogenesis is not well characterized. PTEN protein contains an N-terminal PIP2 (phosphatidylinositol-4,5-bisphosphate)-binding motif, a phosphatase domain, a C2 domain, a C-terminal tail containing two PEST (proline, glutamic acid, serine, threonine) sequences and a PDZ (PSD-95/DLG/ZO-1)-binding interaction motif at the end. Three naturally occurring mutations on the phosphatase domain disrupt PTEN’s phosphatase activity: C124S mutation abrogates both lipid and protein phosphatase activity; G129E mutation abrogates lipid phosphatase only; and Y138L mutation abrogates protein phosphatase only.

Aims. To determine the effect of PTEN on HCV infection and the underlying molecular mechanisms.

Methods. We characterized HCV infection after PTEN overexpression or knocking down. We also determined whether PTEN interacts with HCV viral proteins as a mechanism for its effect on HCV infection.

Results. TEN negatively regulated HCV viral entry by using HCV genotype 2a pseudo-particles. We also observed that PTEN Y138L (protein phosphatase deficient) but not C124S (lipid and protein phosphatase deficient) nor G129E (lipid phosphatase deficient) inhibited HCV viral entry. Knocking down PTEN significantly enhanced viral replication; consistently, PTEN overexpression significantly inhibited HCV replication and secretion. Interestingly, PTEN C124S and Y138L could no longer inhibit HCV replication and secretion. We also observed that neither knocking down nor overexpressing PTEN affected HCV RNA translation. In co-immunoprecipitation and pull-down assays, we showed that HCV core protein interacted with PTEN. HCV core aa. R50 was required for the interaction. PTEN could no longer inhibit HCV genomic replication carrying core R50A mutation.

Conclusions. PTEN regulates HCV viral entry, replication, and secretion, but not translation. The lipid phosphatase activity of PTEN is required for inhibiting HCV entry. The protein phosphatase activity of PTEN is required for inhibiting HCV replication and secretion. HCV core interacts with PTEN, which contributes to PTEN’s effect on HCV replication. Our study may help justify further development of PTEN as a new drug target for HCV therapy.

Funding Agencies: CIHR, CanHepC


Estimation of Fibrosis Progression Rates for Chronic Hepatitis C: A Systematic Review and Meta-Analysis Update
A. Erman,1 T. Hansen,2 J. Bielecki,3 M. Krahn,3 and R. Thein4

1Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON, Canada
2Department of Medicine, University of Toronto, Toronto, ON, Canada
3Toronto Health Economics and Technology Assessment Collaborative (THETA), University of Toronto, Toronto, ON, Canada
4Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada

Background. Chronic Hepatitis C viral infection (HCV) when left untreated is a leading cause of cirrhosis, liver failure, cancer and transplantation, making it a major medical and economic burden. Given the recent availability of highly effective but costly antivirals, accurate estimation of HCV-disease progression is essential for evaluating the cost effectiveness of treatment and determining treatment prioritization.

Aims. The purpose of this study was to obtain the most up-to-date stage-specific and stage-constant liver fibrosis progression rates (FPR) in individuals with chronic HCV infection though an updated systematic review and meta-analysis.

Methods. Literature search was conducted using MEDLINE, EMBASE and PubMed databases covering a period of January 1990 to August 2014 and supplemented by reference and citation searches. In general, the review included published English and non-English peer-reviewed prognostic studies which examine liver fibrosis progression in HCV-infected individuals. Publication bias was assessed by Funnel plots and Egger’s test for asymmetry. Stage-constant FPRs were estimated for each study via the indirect method using the fibrosis score distribution and the estimated duration of infection reported in each study. Stage-specific FPRs (, , , ) were estimated using the Markov Maximum Likelihood estimation (MMLE) method developed by Yi et al1. Random and fixed effects meta-analyses were used to obtain pooled stage constant and stage-specific FPR estimates.

Results. Overall, the updated systematic review included a total of 152 reports of HCV-infected individuals (,982). The pooled stage-constant FPR estimates derived through the indirect method were 0.086 (95% CI, 0.085–0.086) and 0.102 (95% CI, 0.098–0.0106) METAVIR units per year for the fixed and random effects models respectively. The stage-specific FPRs based on the random effects model were : 0.111 (95% CI, 0.101–0.122); : 0.087 (95% CI, 0.078–0.096); : 0.121 (95% CI, 0.110–0.132); : 0.115 (95% CI, 0.105–0.127).

Conclusions. The current study provides the most recent/updated estimates of both stage-constant and stage-specific liver disease progression rates associated with chronic HCV infections through an updated meta-analysis and systematic review. These results are consistent with the original study but suggest a slightly slower disease progression for stage-specific FPRs.

Funding Agencies: CanHepC


Estimation of Transient Elastography Based Liver Stiffness Progression Rates for Chronic Hepatitis C: A Systematic Review and Meta-Analysis
A. Erman,1 A. Sathya,2 J. Bielecki,3 J. Feld,4 J. Hoch,5 R. Thein,6 and M. Krahn3

1Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON, Canada
2Department of Medicine, Queens University, Kingston, ON, Canada
3Toronto Health Economics and Technology Assessment Collaborative (THETA), University of Toronto, Toronto, ON, Canada
4University Health Network University of Toronto, Toronto, ON, Canada
5Institute of Health Policy, Management, and Evaluation, Toronto, ON, Canada
6Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada

Background. Chronic hepatitis C (CHC) is a leading cause of liver fibrosis, cirrhosis, cancer and transplantation. Although liver biopsy is the gold standard for determining the degree of liver disease, limitations with regards to its invasive nature and diagnostic accuracy due to sampling errors and intra- and interobserver variability have led to the development of non-invasive methods; among which transient elastography (TE) is the most common. TE uses a measure of liver stiffness (LSM) as a surrogate for fibrosis. The accurate diagnosis of liver fibrosis is essential for decision-making in CHC. Given its increased utilization in clinical practice, there is still a need for accurate prediction of non-invasive diagnosis of fibrosis.

Aims. The aim of the study is to estimate TE-based liver stiffness disease progression rates (LSPR) in treatment naïve CHC patients through a systematic review and meta-analysis.

Methods. Literature search was performed using MEDLINE, EMBASE, CochraneCENTRAL trials as well as clinical trial registries. The search covered January 1990 to February 2015 with no language limit. Studies were included if they were full-length original studies of over 20 treatment naïve CHC patients undergoing TE-based evaluation. Studies were excluded if LSPRs could not be calculated due to missing data (i.e., LSM, duration of infection). LSPRs were obtained for each study through either a direct method using difference in serial LSMs and time interval between them (ΔLSM/Δtime) or an indirect method using a single LSM and the duration of infection (DOI) assuming a baseline of 5.33 kPa for healthy liver (ΔLSM/DOI). Heterogeneity was evaluated using statistic. Pooled direct and indirect LSPRs were estimated through a fixed and random effects meta-analysis.

Results. The review identified 31 reports for indirect and 8 for direct estimates. Based on random-effects model indirect LSPR was 0.141 kPa/year (95% CI. 0.115–0.166) and the direct LSPR was 0.137 kPa/year (95% CI. 0.169–0.205). LSPRs were generally higher for HIV/HCV coinfected versus monoinfected cohorts (Table 26).

Conclusions. Overall these data are generally consistent with liver biopsy studies but suggest slower progression. Estimating disease progression through non-invasive Methods may allow for alternative ways to model CHC and overcome some limitations of biopsies.

Funding Agencies: CanHepC


Utility of Quantitative Hepatitis B Surface Antigen (QHBSAG) Compared to HBV DNA Testing for Predicting Maternal Viremia Associated with Mother to Child Transmission (MTCT) of HBV in a Multiethnic Cohort of Pregnant Chronic Hepatitis B (CHB) Carriers in Canada
G. Samadi Kochaksaraei,1 S. Congly,1 T. Matwiy,1 E. Castillo,1 S. Martin,1 C. Coffin,1 and C. L. Charlton2

1University of Calgary, Calgary, AB, Canada
2Provincial Laboratory for Public Health (Prov Lab), Edmonton, AB, Canada

Background. HBV MTCT despite immunoprophylaxis is linked to high maternal viremia, hence nucleos/tide analog (NA) therapy in highly viremic mothers (HBV DNA ≥ 7 log IU mL−1) is recommended to reduce maternal HBV DNA levels to below the threshold associated with vaccine failure. Quantitative HBsAg (qHBsAg) is a new test proposed for management of CHB, but there is limited data in pregnancy.

Aims. To determine the utility of qHBsAg as a surrogate marker of HBV DNA.

Methods. CHB pregnant patients were recruited from a hepatology outpatient practice or an obstetrics internal medicine clinic. Demographics and laboratory data, HBV DNA and qHBsAg were assessed in the second-third trimester. Statistical analysis was performed by Spearman’s rank correlation and student’s -test.

Results. 99 women with 103 pregnancies, median age 32 (IQR 29–35), 65% Asian, 23% African, and 12% other (Hispanic, Caucasian) were enrolled. Overall, 23% (23/99) were HBeAg (+), median ALT was 21 U/L (IQR 14–30.25), median HBV DNA and qHBsAg was 2.79 log IU mL−1 (IQR 1.9–3.95) and 3.52 (IQR 2.87–4.12), respectively. There was a statistically significant difference in qHBsAg in HBeAg positive (+) versus HBeAg negative (−) patients (4.38 log IU mL−1 (IQR 3.61–4.88) versus 3.33 log IU mL−1 (IQR 2.76–3.78), ) and HBV DNA (7.76 log IU mL−1 (IQR 2.83–8.29) in HBeAg (+) versus 2.56 log IU mL−1 (IQR 1.81–3.19) in HBeAg (−), ). In HBeAg (+) patients, a significant correlation between qHBsAg titer and HBV DNA level (, ) was noted while there was no significant correlation between qHBsAg titer and HBV DNA level in HBeAg (−) group (, ). In receiver operating characteristic (ROC) analysis, the optimal qHBsAg cut-off values for predicting HBV DNA level associated with immunoprophylaxis failure (i.e., HBV DNA ≥ 7 log IU mL−1) was ≥4.33 log IU mL−1 (accuracy 98.6%, sensitivity of 94.7% and specificity of 94.4% (95% CI, 97%–100%, ). Based on the direct cost per test for qHBsAg ($28) and HBV DNA ($152); qHBsAg may be a more cost-effective test for predicting high maternal viremia.

Conclusions. Serum qHBsAg positively correlates with HBV DNA in HBeAg (+) CHB pregnant patients and may be a more cost-effective test in assessing maternal viremia, and the need for NA therapy to prevent HBV immunoprophylaxis failure, especially in resource-limited settings.

Funding Agencies: None


Treatment of Mixed Cryoglobulinemic Vasculitis with Direct Acting HCV Therapy
J. Emery,1 M. Kuczynski,2 D. La,3 S. Almarzooqi,3 and J. Feld2

1University of Toronto, Toronto, ON, Canada
2University Health Network University of Toronto, Toronto, ON, Canada
3UHN, Toronto, ON, Canada

Background. Mixed cryoglobulinemia (MC) is a lymphoproliferative disorder with a strong association to HCV infection. Manifestations of MC range from asymptomatic to life threating with HCV eradication leading to significant improvements in morbidity. Traditionally, clearance of HCV has required a combination of PEGinterferon and ribavirin which achieves sustained virological responses in 36–64% of patients. Importantly, remission of MCV symptoms is seen in over 80% of those achieving SVR. However, expectations of SVR rates and side effects profiles in the primary treatment of HCV have rapidly changed in the era of novel direct acting antivirals (DAA). Dramatic impacts on SVR rates have been reported (over 90%) and replicated but little has been published on their efficacy in the subpopulation with MCV.

Aims. To investigate the efficacy and safety of DAA in the treatment of Mixed Cryoglobulinemia.

Methods. Patients with immunological evidence of HCV related mixed cryoglobulinemia and prior treatment with direct acting antivirals were identified at tertiary care medical centre. Treatment response was evaluated based on clinical, immunological and virological outcomes at treatment cessation and at 12 weeks post treatment. Treatment side-effects, use of rescue therapy and decompensating events were recorded to confirm safety.

Results. Seventeen symptomatic and fifty non-symptomatic patients were reviewed. To date, SVR12 was achieved in ten (92%) symptomatic and twenty nine (93.5%) asymptomatic patients. At SVR12 full immunological response was achieved in four (40%) symptomatic and nineteen (59%) asymptomatic patients with five (33%) patients achieving full clinical response. One patient (14%) on PEG-IFN based regimens and three (44%) patients on interferon-free regimens had full clinical response rates. Full immunological response rates were seen in four (40%) patients on PEG-IFN and nineteen (60%) on IFN free regimens.

All fifty seven (100%) patients were able to complete therapy. Two (3%) patients had direct therapy related side effects (significant ribavirin related anemia) with four (6%) and five (7%) patients requiring hospitalization for decompensation or vasculitis.

Conclusions. Direct acting antivirals are efficacious in achieving sustained virological responses in symptomatic and asymptomatic patients with cryoglobulinemia. Immunological and clinical response rates in patients achieving SVR12 are suboptimal compared to previous reports, which may reflect shorter treatment courses or lower use of interferon. Longer follow up of our cohort is required to make adequate conclusions about clinical efficacy. Overall, use of DAA’s in patients with cryoglobulinemia is well tolerated in symptomatic patients.

Funding Agencies: None


National Hepatitis C Education Learning Needs Assessment
H. Shah,1 K. Seto,2 C. Klassen,3 D. Emokpare,4 B. Conway,5 M. Kelley,6 and E. Yoshida2

1Canadian Association for the Study of Liver, Edmonton, AB, Canada
2Canadian Liver Foundation, Markham, ON, Canada
3Canadian Association of Hepatology Nurses, Vancouver, ON, Canada
4Family Practice, Regina, SK, Canada
5Vancouver Infectious Disease Clinic, Vancouver, BC, Canada
6Memorial University, St. John’s, NF, Canada

Background. Hepatitis C virus (HCV) is highly prevalent in Canada, with estimates of between 250,000 to 400,000 infected individuals. Recently, new therapeutic paradigms have increased the number of infected persons eligible for treatment. However, without adequate education of healthcare professionals (HCPs), necessary treatments may not be delivered to the infected population. To inform educational activities that will grow capacity for HCV care in Canada, a learning needs assessment is necessary.

Aims. To assess the current knowledge of healthcare professionals (HCPs) in the area of hepatitis C management, including screening and treatment using new antivirals, as well as the educational needs of HCPs, at a national level.

Methods. A survey was distributed by facsimile and email to more than 3500 individuals in the fields of hepatology, gastroenterology, nursing, other medical specialties and primary care physicians (PCPs). The survey consisted of 29 questions collecting participant and practice demographics, access to resources, screening habits, communication, knowledge of new treatments, and educational preferences. Partially completed surveys were accepted, with analysis based on the number of respondents for each question. Standard descriptive statistical methods were used.

Results. 163 participants answered all or part of the survey. The largest proportion were nurses with 69 individuals identifying as a nurse and 6 as a nurse practitioner. Others included hepatologists (20%), PCPs (15%), gastroenterologists (12%) and other specialists (7%). Most providers reported seeing patients with cirrhosis (80%) and past IDU was the main risk factor (60%). All hepatologists prescribed HCV therapy compared to 8% of PCPs. There was large variability in access to fibroscan and allied health by physician group. All provider groups were comfortable counselling HCV patients but hepatologists had the highest level of comfort with prescribing therapy across all severity of patients. Hepatologists reported higher perceived knowledge levels about new therapies. All participants welcomed association-sponsored medical education.

Conclusions. There is significant engagement amongst all providers in Hepatitis C. Many different provider types are involved in aspects of Hepatitis C care. To ensure growing capacity for HCV care in Canada, PCPs and nurses should be provided targeted education and resources to screen and treat alongside other specialists.

Funding Agencies: Canadian Liver Foundation


A National Assessment on Knowledge and Attitudes of 644 Primary Care Providers to HCV Infection and Therapy in the Era of All-Oral Therapy
H. Shah1 and A. Ramji2

1University of Toronto, Toronto, ON, Canada
2Gastrointestinal Research Institute, Vancouver, BC, Canada

Background. It is estimated that 250,000 or 0.8% of the Canadian population are infected with chronic hepatitis C virus (HCV), the leading cause of economic loss and health burden due to any infectious disease. HCV is underdiagnosed, despite being the only curable chronic viral infection. There is a lack of data on the knowledge gaps and attitudes of primary care physicians (PCPs) in the era of all-oral and highly effective HCV therapy as well as systematic barriers to care. Past research in the interferon-era demonstrated wide ranges of competency and attitudes towards therapy. A survey just prior to widely available all-oral therapies demonstrated that 60% of PCPs were not very confident or only somewhat confident screening patients for HCV and 21% felt confident in initiating treatment.1 It is unclear what impact better therapies have on these knowledge levels and attitudes.

Aims. To evaluate the knowledge of, attitudes and barriers to care in HCV by Canadian PCPs.

Methods. A survey invitation (without incentive) was sent to Canadian PCP’s via email and fax from an established database. Responses were analyzed using standard descriptive statistical methods.

Results. This is the largest Canadian survey of HCV in PCPs (). 69% of PCPs were in a group practice setting, 21% in a solo practice and 10% at a walk-in clinic. Every Canadian province was represented. Only 33.4% of PCPs indicated they had a good understanding or felt confident in screening/testing for HCV. 66.9% of PCPs knew that HCV was curable and 69.8% were aware of treatment options. The highest perceived barriers in relation to testing for HCV (PCPs could choose more than one answer) included lack of guidelines (33.2%), uncertainty on which patients to test (27.6%) and uncertainty on which lab tests to order (22.2%), while many responded that there were no barriers (42.6%). People born between 1945 and 1975 are considered at high risk of HCV, but only 33.9% of PCPs said they would screen this group. 48.6% noted specialist referral wait time was a barrier. Other barriers to referrals included unawareness of specialists treating HCV in their region (22.0%) and uncertainty in appropriate HCV-related laboratory tests prior to referral (25.0%). 27.5% of PCPs were interested in treating patients with HCV.

Conclusions. Despite advances in HCV therapy, primary care knowledge gaps in screening, referrals and management remain large. An educational program focused for PCPs has been developed based on the results of this survey.

Funding Agencies: Unrestricted grant from Gilead Sciences Inc.


HCV Treatment of HIV-HCV Co-Infected PWID at a Tertiary Clinic
S. Hakobyan, S. Vafadary, T. Raycraft, S. Sharma, and B. Conway

Vancouver Infectious Diseases Centre, Vancouver, BC, Canada

Background. There is an over-representation of people who inject drugs (PWID) among HIV-HCV co-infected adults. Recent data indicate that HCV treatment regimens are equally effective in the setting of HIV co-infection. However, within co-infected PWID populations, the feasibility (and success rates) of such therapy has not yet been clearly established outside of clinical trials. The aim of this analysis was to address this gap in knowledge.

Aims. Our goal was to identify, recruit, and retain HCV-infected PWID in care. As such, we established a multi-disciplinary program targeted at maintaining a lasting relationship with such individuals. We hypothesized that this approach would facilitate a long-term relationship with this vulnerable population.

Methods. This program includes facilitated access to specialty medical care, access to support services, comprehensive management of social needs, and addiction treatment. We have conducted a retrospective analysis of all HIV co-infected patients treated for HCV infection at our centre. This analysis correlates the likelihood of achieving sustained virologic response (SVR) following HCV treatment with a range of baseline demographic and clinical variables, including housing and active drug use.

Results. Of 522 HIV-infected individuals regularly seen at our centre, 247 (47.3%) were co-infected with HCV. Among the latter, 167 (67.6%) were PWID and 77 (31.2%) have completed HCV treatment (72 interferon-based, 5 all-oral regimens), 46 (59.7%) of which had genotype 1 infection. The mean age of treated patients was 51, 70 (90.9%) were male, 24 (31.2%) were on opiate substitution, 73 (94.8%) were on HIV treatment (62/73 with full virologic suppression), 21 (27.3%) were homeless, and 33 (42.9%) attended weekly HCV support groups. The SVR rate was 46.8% (36/77), 3/5 (60%) on all-oral regimens, 21/46 (45.7%) with genotype 1 infection, and 3/3 (100%) for patients with genotype 1 on all-oral regimens. Success rates were higher in subjects on methadone at 16/24 (66.7%), and no lower in those who were homeless 11/21 (52.4%) or active PWID 26/54 (48.1%).

Conclusions. PWID with HIV co-infection can be successfully treated for HCV infection within multi-disciplinary programs such as ours. Such programs will serve as an important tool to address the HCV epidemics in vulnerable populations often considered as “core transmitters" of HCV and HIV infection, especially as highly effective all-oral regimens become the standard of care.

Funding Agencies: None


HCV Re-Infection in High-Risk People Who Inject Drugs
B. Conway, S. Vafadary, T. Raycraft, F. Zahedieh, S. Sharma, and S. Hakobyan

Vancouver Infectious Diseases Centre, Vancouver, BC, Canada

Background. People who inject drugs (PWID) constitute the majority of cases of HCV infection in Canada. Although a number of strategies have been developed to engage them in care, reluctance to implement them relates at least in part to concerns about re-infection following successful HCV treatment. We have examined this issue in a prospective longitudinal cohort to establish whether this concern is confirmed in clinical practice.

Aims. The aim of this study was to assess the risk of HCV re-infection in a high-risk PWID population.

Methods. Within a multidisciplinary program to engage and treat PWID, we have documented 45 cases of HCV therapy having resulted in a sustained virologic response (SVR) in which patients continued to engage in high-risk behaviour for HCV acquisition after SVR was achieved. These individuals have been followed prospectively to document recurrent viremia, with the performance of HCV RNA testing every 6 months, more frequently if elevated ALT or symptoms of acute hepatitis were noted. The endpoint of this analysis is a positive HCV RNA test following the clear establishment of an SVR.

Results. Among the 666 HCV positive patients, with a mean age of 52.8 years, there were 51 (7.6%) females. Of these patients, 86 (12.9%) were co-infected with HIV, 419 (62.9%) had genotype 1, and 616 (92.5%) were previously treatment naïve. In a mean of 5.95 person-years of follow-up/subject, 4 cases of re-infection were noted (1.49/100 person-years) with all being HIV co-infected patients and 3 being genotype 1. The only factor associated with an increased risk of HCV re-infection was use of stimulants.

Conclusions. In our cohort, PWID successfully treated for HCV infection experience re-infection at a lower rate than previously encountered in uninfected at-risk individuals, and this negative outcome is often associated with stimulant use. The risk of HCV re-infection in individuals receiving care in multidisciplinary programs such as ours has probably been overestimated.

Funding Agencies: None


Evaluation of Treatment of HCV Infection in People Who Inject Drugs
A. Alimohammadi, S. Sharma, S. Hakobyan, and B. Conway

Vancouver Infectious Diseases Centre, Vancouver, BC, Canada

Background. Approximately 70% of HCV infected individuals in Canada are people who inject drugs (PWID). However, many healthcare providers require PWID to be drug-free for 6–12 months before commencing HCV treatment, fearing high reinfection rates. The aim of this study was to illustrate that HCV treatment could be successful in PWID in the right circumstances, without requiring a mandated period of abstinence.

Aims. The aim of this study was to assess the effectiveness of HCV treatment among active PWID populations.

Methods. A retrospective observational study was conducted in active PWID (currently injecting recreational drugs) receiving HCV therapy between 2011 and 2015 at a multi-disciplinary inner city clinic, favoring engagement and retention in care of the target population. Data regarding HCV treatment, HIV co-infection status, as well as demographic and social variables was collected. The primary endpoint was a sustained virologic response (SVR) with respect to HCV infection.

Results. We treated 40 eligible subjects (34 male) with a median age of 53 years, 24 (60%) genotype1a/b, 10 (25%) genotype 3, 33 (83%) previously treatment naïve, 11 (27.5%) co-infected with HIV. With respect to illicit drug use, there were 25 (63%) using heroin, 28 (70%) using cocaine, 9 (22.5%) using other stimulants and 23 (58%) on opiate substitution therapy. With respect to HCV therapy, 25 (63%) received IFN-based and 15 (37%) all-oral regimens. In total, 31 (78%) subjects achieved SVR, 17 (68%) and 14 (93%) on IFN-based and all-oral regimens ( favoring all-oral regimens). Within the study population, 7 (64%) with HIV co-infection, 18 (75%) with genotype 1, 9 (90%) with genotype 3, 21 (84%) on heroin, 21 (75%) on cocaine and 7 (78%) using other stimulants achieved SVR. Three (8%) discontinued due to toxicity and 4 (10%) experience a virologic Relapse. There were no cases of recurrent viremia with a mean follow-up period of 600 days.

Conclusions. Active PWID can be effectively treated for HCV infection with high SVR rates, especially with all-oral regimens. With structured post-treatment follow-up, rates of recurrent viremia can be minimized, enhancing the feasibility of programs to increase treatment uptake in high-risk populations of “core transmitters” of HCV infection.

Funding Agencies: None


Differential Expression Profile of Circulating Micrornas Mir-24 and Mir-223 in Hepatitis C-Infected Patients Who Achieve a Treatment-Based Viral Cure
A. Hyrina,1 P. Steven,2 M. Krajden,3 E. Tam,4 and F. Jean1

1University of British Columbia, Vancouver, BC, Canada
2QIAGEN, Manchester, UK
3BC Centre for Disease Control, Vancouver, BC, Canada
4LAIR Centre, Vancouver, BC, Canada

Background. Hepatitis C virus (HCV) hijacks host lipid metabolic pathways as part of its replication cycle; this likely plays a role in viral pathogenesis. MicroRNAs (miRNAs) are small non-coding RNAs that silence gene expression by binding to mRNA transcripts and act as regulators of cellular processes. However, the function of miRNAs in lipid homeostasis during HCV infection and after viral clearance and their association with progressive liver disease are poorly understood. To help clarify the role of circulating miRNAs known to be involved in hepatocellular lipid metabolism, we determined circulating levels of three candidate miRNAs (miR-122, miR-24, and miR-223) from the plasma of HCV-infected patients undergoing interferon-based treatment.

Aims. In this study, we investigated how circulating levels of miR-122, miR-24 and miR-223 change during chronic HCV infection and after viral cure.

Methods. Circulating levels of miR-122, miR-24, and miR-223 from 94 patients with HCV were measured using qRT-PCR assays at multiple time-points before, during, and after interferon therapy. Liver fibrosis was scored by biopsy and/or FibroScan® before treatment. Serum HCV RNA levels and routine blood parameters were measured at each sample collection.

Results. We demonstrated that circulating miR-122 decreased after HCV viral clearance, correlating with the normalization of liver-specific enzymes (AST, , ; ALT, , ) and with a linear relationship to the liver injury APRI score, while miR-24 and miR-223 levels significantly increased after viral clearance (). In contrast, circulating levels of miR-24 and miR-223 in patients who underwent an HCV treatment-based relapse remained unchanged, while miR-122 levels demonstrated a statistically significant increase in abundance (). Quantitative correlation in amounts of circulating miR-24 and miR-223 was also detected (, ) in relation to viral clearance but not in miR-122.

Conclusions. Our findings provide the first experimental evidence of upregulation of circulating miR-24 and miR-223 in plasma in HCV-infected patients who achieve an interferon-based virological cure. These results also reveal that miRNAs known to act as regulators of lipid metabolism may be correlated with interferon-based therapeutic outcomes in patients with HCV infection, suggesting that dynamic changes in the lipidome following viral cure as assessed by miRNA expression may be correlated with the risk of progressive liver disease.

Funding Agencies: Research Collaboration Agreement with QIAGEN Sciences LLC (No 14-0810 to F. Jean) and the National CIHR Research Training in Hepatitis C to A. Hyrina


Hepatitis B Reactivation Prophylaxis for Patients Undergoing Chemotherapy for Lymphoma in Canada: Current Practice in Hematology/Oncology
G. Ou,1 K. Savage,1 L. Shepherd,2 J. Connors,1 and E. Yoshida1

1University of BC, Vancouver, BC, Canada
2Queen’s University, Kingston, ON, Canada

Background. Patients receiving cytotoxic chemotherapy have an increased risk of hepatitis B virus (HBV) reactivation and related hepatitis, which are associated with significant morbidity and mortality. Previous studies in the United States have demonstrated low rates of HBV screening and reactivation prophylaxis among patients undergoing chemotherapy.

Aims. To determine the current practice pattern of Canadian hematologists/oncologists in regards to screening for HBV infection and consideration of HBV reactivation prophylaxis for patients undergoing chemotherapy for lymphoma.

Methods. We conducted a survey in May 2015. Members of Canadian Hematology Society () and NCIC Clinical Trials Group () were invited by email to participate in an online, 9-multiple choice survey. Those with concomitant membership in both organizations received duplicate invitations.

Results. In total, there were 69 participants. 64/67 (96%) participants reported routine screening for HBV infection prior to chemotherapy. For the remaining participants, two physicians only screen patients with established risk factors for HBV; and another physician confined screening to patients with risk factors for HBV undergoing rituximab therapy. 64/67 (96%) participants routinely prescribe antiviral prophylaxis and/or consult another specialist for patients with positive HBV surface antigen (HBsAg) but no evidence of hepatic inflammation. However, only 51/66 (77%) participants routinely prescribe antiviral prophylaxis and/or consult another specialist for patients with negative HBsAg but positive anti-HBV core antibody (anti-HBc); two would prescribe prophylaxis if HBV DNA is also positive; and one would prescribe prophylaxis if rituximab is used in this setting.

Conclusions. Canadian hematologists/oncologists are screening and offering HBV prophylaxis to most of the patients at risk of HBV reactivation during chemotherapy. Future efforts should be directed at ensuring that all at-risk patients, including those with positive anti-HBc/negative HBsAg, receive appropriate prophylaxis.

Funding Agencies: None


Access to New Hepatitis C Therapy: Do the Sickest Patients Receive Treatment? Practice Audit
J. Kiberd,1 G. Hirsch,2 C. Burgess,2 J. Igoe,3 M. Laryea,4 K. Peltekian,4 and M. McLeod4

1Dalhousie University & Queen Elizabeth II Health Sciences Centre, Halifax, NS, Canada
2Queen Elizabeth II Health Sciences Center, Halifax, NS, Canada
3Dalhousie, Halifax, NS, Canada
4Dalhousie University, Halifax, NS, Canada

Background. Success rates of second generation DAA therapy for HCV have climbed to 95% with very few adverse effects. The course of treatment has been halved with most achieving viral eradication in just 12 weeks. However, this new oral regimen is costly. In Nova Scotia, a 12-week regimen of oral therapy for HCV is covered only in presence of fibrosis.

Aims. To determine if the patients who did not have third party coverage had worse liver disease compared to those who did have access through third party coverage. To determine the reasons that HCV infected patients are not on oral therapy in our center.

Methods. Retrospective chart review was performed on all patients who had documented HCV infection and had at least one-year follow-up between Apr 2014-15 in our clinic. New treatment status, sex, age, platelets, AST, ALT, INR, bilirubin, and creatinine were recorded. MELD, FIB4, and APRI scores were calculated and used to assess liver disease severity. Reasons for not having access to new therapy were “no third party coverage,” “alcohol/opioid dependence,” “considered palliative” or “Other.”

Results. 454 patients were identified with HCV infection. Fifty-one were removed because they did not have lab work or did not have coverage information resulting in 403 patients included in this audit. Sixty-two percent were male with a mean age of 55.8 years (95% CI: 54.8, 56.8). Cirrhotic patients made up 42% (168) of the sample as defined by low platelet count. Only 37% (148) of patients received new therapy. Median MELD score was 6 (IQR: 6, 8), median APRI was 0.75(IQR: 0.38, 1.62), and median FIB4 score was 2.06(IQR: 1.26, 4.28). High MELD score was predictive of new treatment (OR: 4.47, 95% CI: 2.02, 9.86). Both APRI and FIB4 scores were also predictive of new treatment (). Not having third party coverage was associated with lower MELD () but not APRI and FIB4 scores. Of the 244 patients who were not on new oral therapy, 188 (77%) were due to lack of third party payer. 12 (5%) were deemed to have substance abuse and were not considered for treatment; while 12 (5%) were considered palliative due to competing health problems.

Conclusions. Patients who did not receive new DAAs had comparable liver disease severity to those who received drug. Patients who do have third party coverage had faster access to treatment as compassionate treatment was given to those only in the presence of severe liver disease.

Funding Agencies: None


Laboratory Assessment of 1207 Hepatitis B Surface Antigen Positive Patients in Calgary, Alberta in 2014
K. Lau,1 A. Shaheen,1 T. Ricento,2 M. Swain,1 S. Lee,1 K. Burak,1 and C. Coffin1

1University of Calgary, Calgary, AB, Canada
2Alberta Health Services, Calgary, AB, Canada

Background. Addressing chronic hepatitis B (CHB) infection is a priority for the Canadian health care system. Despite awareness of its prevalence and impact on morbidity and mortality, there is no data about how health care providers in Canada comply with current guidelines.

Aims. (1) In a 1-year period, we aim to characterize prevalence of Hepatitis B virus (HBV) infection and clinical status of all identified CHB carriers within a large urban Canadian centre (Calgary, Alberta).

(2) Assess if required laboratory investigations were appropriately done for identified HBV surface antigen (HBsAg) positive.

Methods. Based on Calgary Lab services administrative data, we identified all adult patients with positive HBV surface antigen within the Calgary Zone (Alberta Health Services), between January 1st and December 31st 2014, under an approved University of Calgary ethics protocol. All related demographic and relevant laboratory information on this cohort was extracted within 6 months of a positive HBsAg test. Non-parametric statistical methods were used for analyses.

Results. We identified 1207 patients with positive HBsAg in the Calgary area (48% female (582/1207), median age 44 (IQR 36–55), 4% patients (22/1207) hepatitis C virus co-infected). With a population of 1.2 million (municipal census) in 2014, the prevalence of CHB is 1%. In 51% (610/1207) with completed HBV E antigen (HBeAg) test, only 12% (75/610) were HBeAg positive. Overall 98% (1184/1207) had single alanine aminotransferase (ALT) testing within 6 months with median ALT 37 (IQR 25–55). 9.6% (114/1207) were found to have elevated ALT at single time-point, and 62% (71/114) with HBV DNA testing completed showed a median viral load of 2.8 log IU/mL (1.2–5.8). Among patients with normal ALT, 57% (608/1070) had HBV DNA testing and 89% (542/608) had low-level viremia (median HBV DNA 2.4 log IU/mL (1.2–3.3)). 59% (67/114) of patients with high ALT also had detectable HBV DNA and 39% (26/67) had HBV DNA levels that could potentially meet treatment criteria (73% male (19/26), median age 45 (IQR: 31–56)). Available referral/follow-up data show that overall 23% (277/1207) had seen a hepatologist within the previous 3 years.

Conclusions. Our results shows a higher prevalence (1%) of CHB in Calgary than the estimates for Canada (0.76%). More importantly, only a quarter (23%) of positive HBsAg patients are directed to specialists, suggesting that majority of patients lack appropriate care, surveillance, or treatment of HBV. Improvements in the current public health screening of HBV and subsequent referral to a hepatologist are necessary.

Funding Agencies: CIHR


HCV Infection Causes Multiple Forms of Programmed Cell Death in Infected and Neighbouring Uninfected Cells
H. Kofahi, N. Taylor, M. Grant, and R. Russell

Memorial University, St. John’s, NF, Canada

Background. Chronic infection with HCV increases the risk of developing cirrhosis and hepatocellular carcinoma. Induction of programmed cell death (PCD) in the infected liver plays a role in the pathogenesis and studying it will help to understand the mechanism of development of these liver diseases.

Aims. To study the effect of HCV infection on inducing different forms of PCD in infected and neighbouring cells, and to analyze the mechanism by which it is induced.

Methods. Huh-7.5 cells were infected with the JFH1T strain of HCV. Cell viability was measured by MTT assay. DNA fragmentation was tested by staining with propidium iodide (PI) and detecting hypodiploid cells. Apoptosis was tested by detecting cPARP-positive cells and by inhibiting caspase-3. Bystander apoptosis was detected by co-culturing infected Huh-7.5 cells and S29 cells, then detecting cPARP-positive cells in the S29 cell population. Co-culturing infected Huh-7.5 and S29 cells in a transwell plate was used to test the role of cell-to-cell contact in the induction of bystander apoptosis. Induction of pyroptosis was tested by measuring the LDH activity and by staining them with the caspase-1-specific FAM-YVAD-FMK-FLICA. The induction of pyroptosis was confirmed by testing the effect of inhibiting caspase-1 on DNA fragmentation. Bystander pyroptosis was tested by staining a co-culture of Huh-7.5 and S29 cells with FAM-YVAD-FMK-FLICA.

Results. Infection reduced the viability of Huh-7.5 cells and induced DNA fragmentation. HCV infection increased the number of cPARP-positive cells. Inhibiting caspase-3 resulted in a significant decrease in DNA fragmentation, indicating that HCV infection induces apoptosis. cPARP-positive cells were found in both Huh-7.5 and S29 cell populations demonstrating the induction of bystander apoptosis. We could not detect any increase in the number cPARP-positive S29 cells in a transwell co-culture, indicating that direct cell-to-cell interaction is required for the induction of bystander apoptosis. Virus infection increased the activity of LDH in the supernatant of infected cells. The number of active-caspase-1-positive cells increased significantly following infection. Inhibition of caspase-1 resulted in a significant reduction in the number of hypodiploid cells, confirming the induction of pyroptosis in the infected population. A significantly higher number of S29 cells stained positive for active-caspase-1 when co-cultured with infected Huh-7.5 cells, providing evidence for the induction of bystander pyroptosis.

Conclusions. HCV infection induces two forms of PCD: apoptosis and pyroptosis directly and indirectly (bystander). Bystander apoptosis was found to be contact dependent.

Funding Agencies: CIHR


The Community Pop-Up Clinic as a Tool of Engagement for Vulnerable Populations with HCV and HIV Infections
B. Conway, S. Vafadary, S. Sharma, F. Zahedieh, J. Shravah, T. Raycraft, and S. Hakobyan

Vancouver Infectious Diseases Centre, Vancouver, BC, Canada

Background. The Downtown East Side Vancouver (DTES) is known for a high prevalence of HCV and HIV infection. Despite available services, significant numbers of patients remain undiagnosed or unengaged in care. There is a need to develop innovative structures to address this issue and understand the level of knowledge about infection and the interest to seek care.

Aims. The aim of the study was to evaluate the effectiveness of community pop-up clinics (CPCs) at DTES sites and assess the willingness of participants to receive care.

Methods. Participants were evaluated at community pop-up clinics (CPCs) held at DTES sites (including InSite, the only supervised injection facility in North America). HCV and HIV point-of-care testing was offered. Participants also completed targeted questionnaires to collect demographic information, knowledge about HCV infection, and desire to receive care. A $10 incentive was offered for participation.

Results. Since January 2014, 1,850 individuals (mean age 46.5, 82.0% male) were tested, with 631 (34.1%) infected with HCV including 57 (3.1%) co-infected with HIV. Of 840 PWID, 435 (51.8%) were infected with HCV, and 32 (1.7%) co-infected with HIV A total of 136 and 15 were not previously aware of being infected with HCV and HIV respectively. Participants identified HCV transmission as occurring through casual contact (14.1%), unprotected sex (36.9%), sharing needles (45.6%), sharing injection equipment (36.3%), or blood transfusion (42.2%). Only 37.1% were aware of curable treatment being available for HCV infection, and 53.7% would consider treatment for it where it was offered.

Conclusions. Despite the widespread availability of HIV and HCV service