Perioperative Care of Patients with Inflammatory Bowel Disease: Focus on Nutritional Support

Table 3

Characteristics of primary studies investigating the use of perioperative PN in IBD patients.

Author and year	Patient population and interventions	Measured outcomes	Results	Conclusions	Study strengths	Study limitations
Rombeau et al. [52] 1982 retrospective cohort study	Patients with IBD undergoing abdominal surgery; group 1: 11 patients with 0–5 days preoperative PN versus group 2: 22 patients with ≤5 days of preoperative PN	Postoperative complications and length of hospital stay	(i) Group 2 (≤5 days preoperative PN) had significantly fewer postoperative complications (); no statistically significant difference in length of hospital stay; (ii) All patients with postoperative complications had either a preoperative serum albumin level <3.5 g/dl or a serum transferrin level < 150 mg/dl	Preoperative PN for at least 5 days is strongly recommended in patients with IBD who have severe protein deficiency	(i) Study groups comparable with respect to demographic data, diagnoses, and types of surgery; (ii) Preoperative nutritional status was comparable in the two groups	(i) Small sample size; (ii) Lack of PN composition data; (iii) Lack of control for diet in group 1 patients who did not receive PN
Lashner et al. [55] 1989 retrospective cohort study	103 patients with CD undergoing bowel resection (segmental small bowel resection, ileocecectomy, or segmental or total colectomy). Preoperative PN vs control groups compared between surgery types	Length of resected bowel and length of hospital stay	(i) Preoperative PN associated with reduced length of small bowel resection (20.4 ± 14.3 cm less bowel resected in those undergoing segmental small bowel resection and 11.2 ± 4.2 cm less in those undergoing ileocecectomy) that was independent of length of disease determined preoperatively; (ii) No difference in length of large bowel resected between the PN and non-PN groups; (iii) PN associated with a longer hospitalization (13.5 ± 2.6 days longer for the PN group)	Preoperative PN for CD patients is beneficial for those undergoing small bowel resection but was of little benefit for those undergoing colectomy	Controlling for confounding variables with the multivariate regression model did not change the results	(i) Lack of PN composition data; (ii) Lack of control for diet in non-PN groups; (iii) Patient selection bias (criteria for patient selection varied among attending physicians and no standardized criteria were established)
Yao et al. [53] 2005 nonrandomized controlled trial	32 severely malnourished CD patients (BMI < 15.0 kg/m^2) undergoing abdominal surgery; 16 patients received 1 week of preoperative PN and 3 weeks of postoperative PN versus 16 patients in the control group received intravenous fluids with an isocaloric diet “comparable in energy to PN”	Serum immunoglobulins, BMI, weight change, liver function, postoperative complications, and return to work	(i) No significant differences in postoperative complications between groups; (ii) BMI significantly increased in the PN group (from 13.9 ± 0.6 to 15.3 ± 0.7 kg/m^2, ) and did not change significantly in the control group (14.1 ± 0.7 to 14.5 ± 0.5 kg/m^2, ); (iii) Serum IgM levels in the PN group declined to normal value (139 ± 41 to 105 ± 29 mg/dl, ), whereas no statistically significant change is seen in control group’s IgM levels (from 133 ± 16 to 129 ± 13 mg/dl, ); (iv) More patients in the study group returned to work than in the control group	Perioperative PN possibly ameliorates the humoral immunity, reverses malnutrition, and facilitates rehabilitation	(i) Novel study measures able to show objective improvement in nutritional status and humoral immunity with preoperative PN; (ii) Equal number of patients in the study vs control groups (important in studies with small number of patients)	(i) Small sample size; (ii) Limited description of results; (iii) Methods for controlling for diet were not specified
Grivceva Stardelova et al. [26] 2008 retrospective/prospective cohort study	90 patients with IBD undergoing abdominal surgery; 29 patients (16 CD, 13 UC) received an average duration of 19.1 ± 18.5 days of preoperative PN and 61 patients (50 CD, 11 UC) did not receive PN	BMI, disease activity index (CDAI/AI), laboratory indices, and length of hospital stay	(i) CDAI scores decreased in the PN group more than in the control group and BMI increased more in the PN group than in the control group; however, neither of these findings was statistically significant (no values provided); (ii) No statistically significant difference in length of hospital stay	Patients in the PN group had more severe baseline disease (lower BMIs and higher CDAI scores of statistical significance). Although improvements in BMI and CDAI are not statistically significant, PN aided improvement in clinical status to the level of the healthier controls.; Authors estimate that duration of effective PN should be at least 7–14 days	(i) First study that investigated the effect of PN on disease activity scores	(i) Relatively small sample size; (ii) Lack of PN composition data; (iii) Lack of control for diet in the non-PN group; (iv) Inconsistent reporting of statistical significances and vague explanations of the significant differences in laboratory indices
Jacobson [33] 2012 nonrandomized controlled trial	120 patients with moderate to severe CD undergoing intestinal resection. 15 patients received preoperative PN versus 105 patients not given PN preoperatively	Early (within 30 days) postoperative complications	(i) A significant reduction in early postoperative complications () was seen in the PN group (0 patients with complications) versus the control group (29 patients with complications); (ii) Results suggested that preoperative PN increased body weight, serum albumin, and serum triiodothyronine, which reflected an improved nutritional state as well as a lowered risk of postoperative complications.; (iii) PN also decreased serum haptoglobin and white blood cell count which suggested decreased inflammatory activity	Patients with moderate to severe CD undergoing intestinal resection should be treated with at least 18 days of preoperative PN to lower the risk of early surgical complications	(i) PN solution composition provided; (ii) Lab markers for nutrition and inflammation were measured, and authors were able to show that improvements in these measures were associated with preoperative PN use	(i) Small sample size; (ii) Lack of control for nutrient intake in the control group; (iii) Average CDAI for control subjects was not calculated; (iv) Sevenfold matching based on baseline characteristics; (v) Controls were historical rather than contemporaneous; (vi) PN compositions varied widely
Salinas et al. [54] 2012 retrospective cohort study	235 UC patients undergoing abdominal surgery. 56 patients received preoperative PN and 179 did not	Early (within 30 days) postoperative complications	(i) PN was associated with more total postoperative complications (50% vs 35.2% in the control group, ); however, when line infections were excluded, total complications were comparable (37.5% vs 35.2%, )	Routine preoperative PN for the general population of UC patients undergoing surgeries is not indicated	(i) Larger sample size compared to other studies; (ii) Total and individual postoperative complications were compared between groups	(i) Statistically significant disparities in baseline nutritional status and disease severity between groups; (ii) Lack of control for diet in the non-PN group; (iii) Lack of PN duration and composition
Ayoub et al. [20] 2018 retrospective cohort study	144 CD patients undergoing major abdominal surgery; 55 received preoperative PN and 89 did not	30-day postoperative complications	(i) No statistical differences were seen in infectious complications (18.2% in the PN group, 12.3% in the non-PN group, ) or noninfectious complications (14.5% in the PN group, 16.8% in the non-PN group, ); (ii) Patients receiving preoperative PN for >60 days had significantly lower odds of developing noninfectious complications (OR 0.07, 95% CI 0.01–0.80, ) with no associated increase in infectious complications; (iii) Weight loss of >10% in the past 6 months was a significant predictor of postoperative complications	In a subset of malnourished CD patients, PN is safe and allows comparable postoperative outcomes to controls	(i) Largest study to date on preoperative PN in CD patients; (ii) Detailed description and comparison of baseline patient characteristics, including disease and surgical characteristics between groups; (iii) Authors provided a detailed description of rationale for initiation of PN	(i) Lack of a validated preoperative nutritional assessment for all patients

PN: parenteral nutrition; IBD: inflammatory bowel disease; CD: Crohn’s disease; BMI: body mass index; UC: ulcerative colitis; CDAI: Crohn’s disease activity index.