- About this Journal ·
- Aims and Scope ·
- Article Processing Charges ·
- Author Guidelines ·
- Bibliographic Information ·
- Citations to this Journal ·
- Contact Information ·
- Editorial Board ·
- Editorial Workflow ·
- Free eTOC Alerts ·
- Publication Ethics ·
- Recently Accepted Articles ·
- Reviewers Acknowledgment ·
- Submit a Manuscript ·
- Subscription Information ·
- Table of Contents
Volume 2012 (2012), Article ID 298345, 7 pages
Sexual Dimorphism in Hematocrit Response Following Red Blood Cell Transfusion of Critically Ill Surgical Patients
1Department of Surgery, Denver Health Medical Center, University of Colorado Health Science Center, 777 Bannock Street MC0206, Denver, CO 80206, USA
2Departments of Research and Information Services, Bonfils Blood Center, Denver, CO 80206, USA
Received 15 December 2011; Accepted 17 January 2012
Academic Editors: H. Schmetzer and C. Tecchio
Copyright © 2012 Fredric M. Pieracci et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The change in hematocrit (ΔHct) following packed red blood cell (pRBCs) transfusion is a clinically relevant measurement of transfusion efficacy that is influenced by post-transfusion hemolysis. Sexual dimorphism has been observed in critical illness and may be related to gender-specific differences in immune response. We investigated the relationship between both donor and recipient gender and ΔHct in an analysis of all pRBCs transfusions in our surgical intensive care unit (2006–2009). The relationship between both donor and recipient gender and ΔHct (% points) was assessed using both univariate and multivariable analysis. A total of 575 units of pRBCs were given to 342 patients; 289 (49.9%) donors were male. By univariate analysis, ΔHct was significantly greater for female as compared to male recipients (3.81% versus 2.82%, resp., ). No association was observed between donor gender and ΔHct, which was 3.02% following receipt of female blood versus 3.23% following receipt of male blood (). By multivariable analysis, recipient gender remained associated significantly with ΔHct (). In conclusion, recipient gender is independently associated with ΔHct following pRBCs transfusion. This association does not appear related to either demographic or anthropomorphic factors, raising the possibility of gender-related differences in recipient immune response to transfusion.
Transfusion of critically ill patients with packed red blood cells (pRBCs) is commonplace; nearly all anemic intensive care unit (ICU) patients with a length of stay (LOS) ≥ 7 days receive at least one pRBCs transfusion [1, 2]. Despite the widespread use of pRBCs transfusions as a treatment of ICU anemia, there are a paucity of data demonstrating their efficacy in terms of improvement of oxygen delivery, oxygen consumption, and ultimately patient outcome [3–6]. By contrast, research over the last several decades has elucidated the detrimental effects of blood product transfusion, particularly with respect to immune suppression, infection, inflammation, and organ failure [5, 7–9]. Such phenomena are believed to involve both transfusion and host response components. In the case of the former, proinflammatory components within the transfusion, such as leukocytes, cytokines, and antibodies, inflict damage directly upon recipient organ systems (including the immune system). In the case of the later, the recipient immune system reacts to antigens within the transfused blood, such as proinflammatory cytokines, lipid mediators, and the erythrocytes themselves, which may in turn cause collateral damage to the host [10, 11].
Research on the impact of gender on ICU outcomes has received recent attention and yielded conflicting results. Several studies have reported sexual dimorphism with respect to both overall and infection-related mortality among critically ill patient cohorts ranging from severe trauma [12, 13] to postcardiac arrest . Demonstration of sexual dimorphism in both risk and survival of severe sepsis [15, 16] has led investigators to invoke gender-related differences in immune response as a potential etiology. Specifically, laboratory investigations have demonstrated an immunosuppressive effect of androgens [17–19], and gender-related genetic polymorphisms have been observed with respect to human proinflammatory proteins .
Evidence of a possible immune-related sexual dimorphism with respect to outcomes following blood product transfusion has been reported by our group and others. Specifically, transfusion-related acute lung injury appears both more frequent and severe following transfusion of female blood as compared to male blood [21, 22]. Furthermore, we observed both decreased survival and more aggressive tumor behavior for male mice, as compared to female mice, inoculated with pancreatic cancer cells following transfusion of female blood [23, 24]. Such differences may involve alloimmunization [22, 25] or even microchimerism  of multiparous female donors.
We have investigated the change in hematocrit (ΔHct) following pRBCs transfusion as a clinically relevant potential surrogate marker of posttransfusion hemolysis. Specifically, based upon the associations between pRBCs storage time, accumulation of inflammatory mediators [27–29], and decreased erythrocyte longevity , we demonstrated a significant inverse relationship between pRBCs storage time and ΔHct following transfusion . The objective of the current investigation was to determine the relationship between both donor and recipient gender and ΔHct following pRBCs transfusion of critically ill surgical patients. Based upon the aforementioned observations, we hypothesized that both female donor gender and female recipient gender would be associated with a decreased ΔHct.
2. Material and Methods
This study was a cross-sectional analysis of all leukocyte-reduced, pRBCs transfusions administered to critically ill patients from June 2006 through June 2009 at a 20-bed academic surgical ICU. Only single-unit transfusions that occurred while the patient was in the ICU were included; multiple unit transfusions as well as transfusions administered at other locations (e.g., operating room and angiography suite) were excluded.
The independent variables of interest were both donor and recipient gender. For the purposes of this analysis, gender refers to sex chromosomal status: XX for female and XY for male. The outcome variable was the ΔHct following pRBCs transfusion (% points, continuous). ΔHct was calculated as the difference between the first posttransfusion value (Hctpost) and the last pretransfusion value (Hctpre). The time from Hctpre determination to the blood transfusion (Tpre, hours), as well as the time from the blood transfusion to Hctpost determination (Tpost, hours) were also abstracted. In order to access for a potential effect of fluid equilibration over time, we also abstracted the second hematocrit determination following pRBCs transfusion (Hctpost2, % points, continuous). This data point was used to calculate ΔHct2 (% points, continuous), defined as the difference between Hctpost2 and Hctpre. Transfusion of additional blood products (e.g., fresh frozen plasma, platelets, and cryoprecipitate) along with the pRBCs transfusion was also recorded.
The following covariates were hypothesized a priori to potentially confound the relationship between gender and ΔHct: recipient age (years), recipient admission diagnosis (trauma versus other), the indication for transfusion (hemorrhage versus ICU anemia, the former being identified by the terms “bleeding” or “hemorrhage” used in the medical record within 48 hours of the transfusion), recipient admission weight (kg), recipient admission body mass index (BMI, kg/m2), recipient 24 hour fluid balance (L), time from surgical ICU admission to transfusion (days), Hctpre (%), and the age of transfused blood. Recipient ICU LOS (days) as well as mortality was also abstracted.
All statistical analyses were performed using SAS version 9.2 (SAS Inc., Carey, NC, USA). The alpha error level was set at 0.05, with being considered significant statistically, unless indicated otherwise. Continuous data are expressed as mean, range; categorical data are expressed as number, %. Correlations between predictor variables and ΔHct were assessed using Pearson’s correlation coefficient (). The means of two continuous variables were compared using the student’s t-test. The means of >2 groups were compared using ANOVA (F statistic); individual group differences were then compared using Tukey’s test in the case of overall significance. It the case of multiple comparisons Bonferroni-corrected -values were used and calculated as 0.05/k, where k is equal to the number of groups analyzed. Differences in proportions of categorical variables were compared using the chi-square test, except when expected cell counts were <10, in which case Fischer’s exact test was used. Multivariable linear regression (sum of squares) was used to access for the independent contribution of gender to the variability in ΔHct. Variables correlated with either recipient gender or ΔHct (or both) at the level by univariate analysis were entered into the model using a forward selection method. Overall model significance was calculated using the statistic, with the significance of individual variables calculated using the t statistic. This study was approved by the Colorado Multiple Institutional Review Board (Protocol no.10-0028).
A total of 1,011 units of pRBCs were administered during the study period. Of these, 594 (56.9%) units were given to 342 patients as single-unit transfusions. Complete data were available for 575 (96.8%) units, which served as the final sample size. Sample demographics are shown in Table 1. The vast majority of pRBCs transfusions were given for ICU anemia (91.4%), well into the recipient’s ICU course (mean time from ICU admission to transfusion 13.7 days), and without additional blood products (95.1%). The mean Hctpre was 21.5 (range 8.2–32). The mean pRBCs storage time was 29.5 days, and 94.3% of pRBCs units were ≥14 days old. Donor gender was distributed equally among male (49.9%) and female (50.1%) donors.
Correlations between independent variables and ΔHct are listed in Table 2. Recipient female gender (, ) and Hctpre (, ) were correlated most strongly with ΔHct. Additional significant correlations [in order of decreasing magnitude ()] were as follows: increasing recipient admission weight was associated with a decreased ΔHct (, ), transfusion for ICU anemia as opposed to hemorrhage was associated with a decreased ΔHct (, ), increasing age was associated with an increased ΔHct (, ), increasing 24 hour fluid balance was associated with a decreased ΔHct (, ), and increasing Tpre was associated with increased ΔHct (, ). Admission diagnosis, recipient admission BMI, time from ICU admission to transfusion, Tpost, the age of blood, donor gender, and transfusion of additional blood products with the pRBCs unit were not correlated significantly with ΔHct.
Table 3 lists sample characteristics stratified by recipient gender. As compared to female recipients, male recipients were significantly younger (), more likely to have an admission diagnosis of trauma (), more likely to be transfused for an indication of ICU anemia (), had a higher BMI (), and received older blood ().
By univariate analysis, ΔHct was significantly greater for female recipients, as compared to male recipients (3.81% versus 2.82%, resp., ). However, donor gender did not impact ΔHct, which was 3.02% following receipt of female blood versus 3.23% following receipt of male blood (). In order to assess for any possible effect of fluid equilibration over time, these analyses were repeated using ΔHct2 instead of ΔHct. The mean ΔHct2 was 2.97% (range −12.3, 15.8) and the mean Tpost2 was 14.1 hours (range 1–92). ΔHct2 remained significantly higher for female as compared to male recipients (3.71% versus 2.63%, resp., ). Furthermore, donor gender was not associated with ΔHct2 (2.91% following receipt of female blood versus 3.03% following receipt of male blood, ).
The relationship between both donor and recipient gender and ΔHct was explored further by stratifying the sample into the 4 donor-recipient gender pairs listed in Table 4: Male donor male recipient (MM), female donor male recipient (FM), male donor female recipient (MF), and female donor female recipient (FF). ΔHct for each of these pairs is listed in Table 4 and depicted in Figure 1. Although significant differences in ΔHct existed between these 4 groups (ANOVA , , ), analysis of differences between individual pairs revealed only those with disparate recipient genders to be significant. By contrast, ΔHct was not significantly different when comparing pairs with disparate donor genders and identical recipient genders. Differences in ΔHct as well as adjusted -values for individual pair comparisons are listed in Figure 1.
Results from multivariable logistic regression analysis are shown in Table 5. The following variables were added to the model in addition to recipient gender based on their association with either recipient gender or ΔHct (or both) by univariate analysis: recipient age, admission diagnosis, indication for transfusion, Hctpre, recipient admission weight, Tpre, age of blood, and recipient 24 hour fluid balance. Of note, recipient BMI was not included in the model based on a high degree of colinearity with recipient weight (, ). After controlling for these covariates, recipient gender remained significantly associated with ΔHct (t = 4.32, ) (model DF-9, F-12.57, ). In additional to recipient gender, patient age, Hctpre, recipient admission weight, and Tpre were also associated independently with ΔHct.
In this cross-sectional analysis of anemic, critically ill surgical patients, we observed an independent association between recipient gender and ΔHct following pRBCs transfusion. By contrast, a relationship between donor gender and ΔHct was not observed. Specifically, ΔHct was higher for female, as compared to male recipients, irrespective of donor gender.
Demonstration of sexual dimorphism with respect to recipient gender in this retrospective analysis raises the issues of both confounding and causality. In terms of the former, several covariates may confound the relationship between recipient gender and ΔHct. Chronologic factors, such as Tpre and Tpost, would appear to influence ΔHct but were not associated strongly with either recipient gender or ΔHct in this analysis, nor did they significantly alter the observed relationship between these two variables in multivariable analysis. Furthermore, because pRBCs transfusion in effect entails adding a relatively fixed volume and concentration of erythrocytes to a variable recipient volume of distribution and Hct, variables such as Hctpre, recipient weight, 24 hour fluid balance, and recipient BMI are of relevance. Indeed, each of these variables was correlated with ΔHct but did not explain the association between recipient gender and ΔHct. Finally, male and female recipients may have differed with respect to the condition of the donor erythrocytes themselves. Specifically, we recently observed that older blood is less effective in raising the Hct . This possibility was addressed by abstracting the age of the transfused pRBCs unit. However, although females tended to receive newer blood than males (27.7 days versus 30.4 days, ), this relatively small degree of difference in storage time was not found to significantly alter ΔHct in our previous work . Furthermore, adding the age of blood to the regression model in this analysis did not eliminate the observed association between recipient gender and ΔHct. Thus, to the best of our ability to control for such confounding variables, female recipients were noted to raise their Hct by a greater amount than their male counterparts by a mechanism independent of either chronologic or dilutional phenomena.
The immunomodulatory properties of pRBCs transfusions involve both donor and recipient qualities. Previous work has implicated transfusion of female blood in both antibody-mediated acute lung injury  and cancer progression [23, 24] and invoked alloimmunization of multiparous women as one possible etiology. Similarly, androgens have been shown to possess immunosuppressive properties . In this study, however, we did not observe a relationship between donor gender and ΔHct. This finding suggests that any gender-related difference in passenger immunologic components (e.g., pro-inflammatory cytokines, antibodies) did not appreciably result in hemolysis of recipient (or transfused) erythrocytes as reflected by the first two posttransfusion hematocrits. By contrast, recipient gender appeared to drive the sexual dimorphism in ΔHct, with female recipients demonstrating a significantly greater ΔHct as compared to male recipients.
We hypothesized originally a smaller ΔHct for female, as compared to male recipients based upon possible immune or sex hormone-mediated posttransfusion hemolysis. However, the opposite effect was observed. One possible explanation for this finding involves immunologic tolerance via microchimerism. Microchimerism entails proliferation of a relatively small (<5%) population of cells (e.g., leukocytes) from a distinct zygote and as a result of mixing of blood between the two organisms. Examples of circumstances leading to microchimerism include solid organ transplantation, pregnancy, and blood transfusion itself. One hypothesis for the observed findings thus involves a lesser degree of immune-mediated hemolysis of transfused pRBCs in female, as compared to male recipients, as a result of microchimerism-mediated immune tolerance. Although testing of this hypothesis is beyond the scope of our dataset, microchimerism and its relationship to immune tolerance represent an important development in the field of transfusion that may help to elucidate the sexual dimorphism observed in various outcomes of critically ill patients. Beyond microchimerism, immunologically relevant gender differences in sex hormone expression may contribute to the observed sexual dimorphism in ΔHct. Continued research is necessary to elucidate and potentially manipulate the observed gender-related disparity in the hopes of ultimately decreasing both patient exposure to and morbidity of allogeneic blood product transfusion.
Out study is limited by a retrospective design. We were unable to capture additional variables that may have influenced ΔHct irrespective of recipient gender such as donor and recipient sex hormone concentrations, menopausal status, and obstetric and transfusion histories. Markers of hemolysis, such as haptoglobin, bilirubin, lactate dehydrogenase, and iron, as well as the presence of hemoglobinuria, were not measured routinely. Demonstration of an association between these markers and ΔHct would provide further evidence for the hemolysis hypothesis and is the focus of a current prospective study in our surgical ICU. Study strengths include a large and relatively uniform sample of critically ill surgical patients, rigorous collection of chronologic and demographic variables, and a low percentage of missing data.
Among this cohort of critically ill surgical patients, we observed a significant association between recipient gender and ΔHct following pRBCs transfusion, such that female recipients raised their hematocrit more than male recipients. This relationship was independent of either chronologic or anthropomorphic factors, differences in the age of transfused blood, and the gender of the donor. No relationship was observed between donor gender and ΔHct in any analysis. Additional factors that were associated independently with ΔHct included recipient age, recipient admission weight, Hctpre, and Tpre. The etiology of the observed sexual dimorphism is unknown but may involve gender-related differences in immune-mediated hemolysis of transfused erythrocytes; additional research at our institution is focused on exploring this hypothesis.
- J. L. Vincent, J. F. Baron, K. Reinhart, et al., “Anemia and blood transfusion in critically ill patients,” Journal of the American Medical Association, vol. 288, no. 12, pp. 1499–1507, 2002.
- H. L. Corwin, A. Gettinger, R. G. Pearl et al., “The CRIT study: anemia and blood transfusion in the critically ill—current clinical practice in the United States,” Critical Care Medicine, vol. 32, no. 1, pp. 39–52, 2004.
- L. M. Napolitano and H. L. Corwin, “Efficacy of red blood cell transfusion in the critically ill,” Critical Care Medicine, vol. 20, no. 2, pp. 255–268, 2004.
- L. M. Napolitano, S. Kurek, F. A. Luchette et al., “Clinical practice guideline: red blood cell transfusion in adult trauma and critical care,” Critical Care Medicine, vol. 37, no. 12, pp. 3124–3157, 2009.
- L. M. Napolitano, “Current status of blood component therapy in surgical critical care,” Current Opinion in Critical Care, vol. 10, no. 5, pp. 311–317, 2004.
- M. J. Englesbe, S. J. Pelletier, K. M. Diehl et al., “Transfusions in surgical patients,” Journal of the American College of Surgeons, vol. 200, no. 2, pp. 249–254, 2005.
- C. C. Silliman, E. E. Moore, J. L. Johnson, R. J. Gonzalez, and W. L. Biffl, “Transfusion of the injured patient: proceed with caution,” Shock, vol. 21, no. 4, pp. 291–299, 2004.
- M. A. Blajchman, S. Dzik, E. C. Vamvakas, J. Sweeney, and E. L. Snyder, “Clinical and molecular basis of transfusion-induced immunomodulation: summary of the proceedings of a state-of-the-art conference,” Transfusion Medicine Reviews, vol. 15, no. 2, pp. 108–135, 2001.
- M. A. Croce, E. A. Tolley, J. A. Claridge et al., “Transfusions result in pulmonary morbidity and death after a moderate degree of injury,” Journal of Trauma, vol. 59, no. 1, pp. 19–23, 2005.
- E. C. Vamvakas and M. A. Blajchman, “Transfusion-related immunomodulation (TRIM): an update,” Blood Reviews, vol. 21, no. 6, pp. 327–348, 2007.
- C. C. Silliman, N. F. Voelkel, and J. D. Allard, “Plasma and lipids from stored packed red blood cells cause acute lung injury in an animal model,” The Journal of Clinical Investigation, vol. 101, no. 7, pp. 1458–1467, 1998.
- M. A. Croce, T. C. Fabian, A. K. Malhotra, T. K. Bee, and P. R. Miller, “Does gender difference influence outcome?” Journal of Trauma, vol. 53, no. 5, pp. 889–894, 2002.
- G. Mostafa, T. Huynh, R. F. Sing et al., “Gender-related outcomes in trauma,” Journal of Trauma, vol. 53, no. 3, pp. 430–435, 2002.
- J. G. Wigginton, P. E. Pepe, J. P. Bedolla, L. A. DeTamble, and J. M. Atkins, “Sex-related differences in the presentation and outcome of out-of-hospital cardiopulmonary arrest: a multiyear, prospective, population-based study,” Critical Care Medicine, vol. 30, no. 4, supplement, pp. S131–S136, 2002.
- S. R. Eachempati, L. Hydo, and P. S. Barie, “Gender-based differences in outcome in patients with sepsis,” Archives of Surgery, vol. 134, no. 12, pp. 1342–1347, 1999.
- A. Combes, C. E. Luyt, J. L. Trouillet, A. Nieszkowska, and J. Chastre, “Gender impact on the outcomes of critically ill patients with nosocomial infections,” Critical Care Medicine, vol. 37, no. 9, pp. 2506–2511, 2009.
- M. W. Wichmann, R. Zellweger, C. M. DeMaso, A. Ayala, and I. H. Chaudry, “Mechanism of immunosuppression in males following trauma-hemorrhage: critical role of testosterone,” Archives of Surgery, vol. 131, no. 11, pp. 1186–1192, 1996.
- J. R. Roubinian, N. Talal, and J. S. Greenspan, “Delayed androgen treatment prolongs survival in murine lupus,” Journal of Clinical Investigation, vol. 63, no. 5, pp. 902–911, 1979.
- M. K. Angele, A. Ayala, B. A. Monfils, W. G. Cioffi, K. I. Bland, and I. H. Chaudry, “Testosterone and/or low estradiol: normally required but harmful immunologically for males after trauma-hemorrhage,” Journal of Trauma, vol. 44, no. 1, pp. 78–85, 1998.
- J. A. Hubacek, F. Stüber, D. Fröhlich et al., “Gene variants of the bactericidal/permeability increasing protein and lipopolysaccharide binding protein in sepsis patients: gender-specific genetic predisposition to sepsis,” Critical Care Medicine, vol. 29, no. 3, pp. 557–561, 2001.
- M. Palfi, S. Berg, J. Ernerudh, and G. Berlin, “A randomized controlled trial of transfusion-related acute lung injury: is plasma from multiparous blood donors dangerous?” Transfusion, vol. 41, no. 3, pp. 317–322, 2001.
- O. Gajic, M. Yilmaz, R. Iscimen et al., “Transfusion from male-only versus female donors in critically ill recipients of high plasma volume components,” Critical Care Medicine, vol. 35, no. 7, pp. 1645–1648, 2007.
- D. D. Benson, M. R. Kelher, X. Meng et al., “Gender-specific transfusion affects tumor-associated neutrophil: macrophage ratios in murine pancreatic adenocarcinoma,” Journal of Gastrointestinal Surgery, vol. 14, no. 10, pp. 1560–1565, 2010.
- P. K. Moore, D. Benson, M. Kehler et al., “The plasma fraction of stored erythrocytes augments pancreatic cancer metastasis in male versus female mice,” Journal of Surgical Research, vol. 164, no. 1, pp. 23–27, 2010.
- A. F. Eder, R. Herron, A. Strupp et al., “Transfusion-related acute lung injury surveillance (2003–2005) and the potential impact of the selective use of plasma from male donors in the American red cross,” Transfusion, vol. 47, no. 4, pp. 599–607, 2007.
- G. H. Utter, W. F. Reed, T. H. Lee, et al., “Transfusion-associated microchimerism,” Vox Sanguinis, vol. 93, no. 3, pp. 188–195, 2007.
- A. Tinmouth and I. Chin-Yee, “The clinical consequences of the red cell storage lesion,” Transfusion Medicine Reviews, vol. 15, no. 2, pp. 91–107, 2001.
- I. Chin-Yee, M. Keeney, L. Knieger, G. Dietz, and G. Moses, “Supernatant from stored red cells activates neutrophils,” Transfusion Medicine, vol. 8, no. 1, pp. 49–56, 1998.
- A. Shanwell, M. Kristiansson, M. Remberger, and O. Ringdén, “Generation of cytokines in red cell concentrates during storage is prevented by prestorage white cell reduction,” Transfusion, vol. 37, no. 7, pp. 678–684, 1997.
- A. Tinmouth, D. Fergusson, I. C. Yee, and P. C. Hébert, “Clinical consequences of red cell storage in the critically ill,” Transfusion, vol. 46, no. 11, pp. 2014–2027, 2006.
- F. M. Pieracci, E. E. Moore, T. Chin, et al., “The age of transfused blood predicts hematocrit response among critically ill surgical patients,” The American Journal of Surgery. In press.