| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
* From the Departments of Anesthesiology and
Surgery (Hepatobiliary Service), Centre Hospitalier de l’Université de Montréal (CHUM), Montréal, Québec; and
the Department of Anesthesiology and Pain Medicine, University of Alberta, Edmonton, Alberta, Canada.
Address correspondence to: Dr. Luc Massicotte, Hôpital St-Luc – CHUM, 1058, St-Denis, Montréal, Québec H2X 3J4, Canada. Phone: 514-890-8000, ext. 36581; Fax: 514-412-7310; E-mail: lmassicotte{at}hotmail.com
 |
Abstract |
|---|
 TOP Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Methods: A retrospective study of 206 consecutive liver transplantations was undertaken. Intraoperative transfusions of blood products were identified. Twenty-seven variables were studied using univariate and multivariate analyses to identify factors that were associated significantly with survival rate. For analysis of one-year survival, the cases were studied according to the transfused blood products. Patients were stratified according to the degree of RBC and plasma transfusion into four groups: more than four units of RBC, one to four units of RBC, plasma transfusion only, and no plasma or RBC transfusions.
Results: Patients received an average of 2.8 ± 3.5 units of RBC and 4.1 ± 4.1 units of plasma. Thirty-two percent of the patients did not receive any RBC transfusion and 19.4% did not receive any blood products. The one-year survival rate was 81.9% for all patients and 97.4% for patients without any transfusions. Of the 27 variables evaluated, only RBC and plasma transfusions were associated with significant decrease in the one-year survival rate, which was seen in the group who received only plasma (76.9%, P = 0.014) and the group who received more than four units of RBC (62.5%, P < 0.0001).
Conclusion: Although we cannot demonstrate causality, our analysis shows that our one-year survival rate following liver transplantation decreased significantly with the intraoperative transfusion of any amount of plasma or more than four units of RBC.
|
Introduction |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The St-Luc Hospital of the Centre Hospitalier de l’Université de Montréal (CHUM) performs more than 50 liver transplantations annually. The purpose of this study was to assess the relationship between one-year survival rate and intraoperative RBC or plasma transfusion.
|
Methods |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Surgery
Four surgeons were involved during the study period. Two of these surgeons were present during each transplant procedure. The different transplantation procedures were all similar. Neither veno-venous shunts nor any piggyback technique were utilized. All livers were from ABO-Rh compatible, cadaveric donors and there were no reduced-sized grafts.
Anesthesia
Twelve anesthesiologists participated in the liver transplant team. In all cases, vascular access and invasive and non-invasive monitoring techniques were identical. The anesthetics were comparable. Induction consisted of propofol, sufentanil or fentanyl as the narcotic of choice, and usually rocuronium as the muscle relaxant. Anesthesia was maintained with either isoflurane or desflurane in oxygen. There was no standardized protocol pertaining to blood product transfusion. All the transfused RBC and plasma were ABO-Rh compatible. The decision to transfuse was an individual one based on the measured hemoglobin (Hb) level and the evaluation of the ongoing blood loss. Aprotinin was given to every patient according to the Hammersmith protocol except in patients with Budd-Chiari syndrome. Isovolemic hemodilution was not used. A cell saver device was used in one patient. The main laboratory test used was arterial blood gas analysis, which also included Hb, potassium, and ionized calcium levels. The arterial blood gas analyses were done on an "as needed" basis dictated by the clinical situation. All these tests were done in the hospital’s central biochemical testing facilities.
Statistical analysis
Twenty-seven variables (Tables I
–III) were analyzed by univariate and multivariate methods. Patients were divided into four groups (more than four units of RBC, one to four units of RBC, plasma only, no blood products) to analyze the influence of RBC transfusion practice on patient survival based upon the same parameters used by the Stanford group.5 The influence of plasma transfusion on patient survival was analyzed by stratifying the patients into four plasma transfusion groups: no blood products, RBC only, one to four units of plasma, and more than four units of plasma.
|
|
|
Results |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
and II. No transplant patients were excluded from the study. Patient’s age ranged from 17 to 68 yr. Two hundred and six liver transplantations were performed on 193 patients. Ten patients had two liver transplantations and one patient required four transplantations. Four patients had a previous liver transplantation before the study period.
|
). A time-based review of laboratory data showed that intraoperative RBC transfusions were started when the average Hb level was 68.2 ± 12.8 g•L–1 and the final Hb value was 87.3 ± 13.9 g•L–1. Analysis of the pattern of transfusion practice revealed that 43 patients (20.9%) received more than four units of RBC, 97 patients (47.1%) received between one and four units of RBC, 26 patients (12.6%) received only plasma, and 40 patients (19.4%) received no blood products (Table I). Sixty-six patients (32.0 %) did not receive any RBC.
Survival
There were no intraoperative deaths. The overall one-year survival rate was 81.9%. Four of the deceased patients had received two liver transplantations each. Mortality was mostly due to multiple organ failure at a mean of 31 ± 24 days post-transplantation.
Univariate analyses (Table III) showed that three variables were associated significantly with the one-year survival rate: the number of RBC units transfused (P = 0.002), the number of plasma units transfused (P = 0.002), and the length of surgery (P = 0.023). The association between RBC or plasma transfusion and one-year survival rate was significant whether the independent variables were studied as categories (mean, median, or subgroups of RBC and plasma (low vs high) transfusion) or as continuous variables. These three variables were entered as continuous and categorical variables in a logistic regression model. Increased RBC or plasma transfusions were associated with decreased survival. However, the colinearity was too strong for these variables to be analyzed separately hence we created a new variable, low vs high transfusion. Table IV summarizes the logistic regression model. Although identified as a confounding variable, multivariate analysis showed that the length of surgery did not modify the relationship between the survival rate and RBC or plasma transfusion. All other study variables did not influence the one-year survival rate (Table III).
|
). One-year survival in patients receiving no blood products was 97.4%. In comparison, the one-year survival rate declined to 86.6% when one to four units of RBC were transfused, 76.9% with plasma transfusion only (P = 0.014), and a nadir of 62.5% when more than four units of RBC were transfused (P < 0.0001). Analysis of the pattern of plasma administration revealed that 40 patients received no blood products, 28 patients received RBC (mean 2.3 ± 1.1 units) but no plasma, 49 patients received between one and four units of plasma (mean 3.1 ± 0.9 units of plasma and 2.1 ± 1.6 units of RBC), and 89 patients received more than four units of plasma (mean 7.8 ± 4.3 units of plasma and 4.7 ± 4.3 units of RBC). The one-year survival rate was significantly decreased to 81.6% in patients receiving one to four units of plasma (P =0.038) and 75.6% in patients receiving more than four units of plasma (P = 0.002).
Table V shows the stratification of the one-year survival rate according to the RBC and plasma transfusion in order to determine if one of the two variables is more strongly associated with a decrease in survival. Survival was decreased significantly in the presence of plasma transfusion in patients receiving no RBC units. In contrast, survival is not decreased in a significant fashion in those patients who received RBC with or without plasma. Survival decreased to 60.0% in patients receiving greater than four units of both RBC and plasma.
|
Table VI shows the survival rate and the number of patients according to each diagnosis. The survival rate did not change after the adoption of leukoreduction of all allogenic blood products in June 1999 (80.0 % vs 83.8 %, P = NS).
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In a preliminary study, we determined that intraoperative plasma transfusion was the variable most closely associated with the need for RBC transfusion in liver transplantation surgery. Plasma was transfused to correct prolonged international normalized ratios (INRs). A great disparity was noted in the plasma transfusion rate between anesthesiologists, indicating a liberal plasma transfusion practice. For RBC transfusion the threshold did not differ significantly amongst anesthesiologists (67.6 ± 12.5 g•L–1 to 69.8 ± 10.7 g•L–1). In the present study, univariate analysis showed no decrease in one-year survival in the presence of an elevated initial INR (Table III).
In the Stanford study,5 patients receiving more than four units of RBC had a significant decrease in their one-year survival rate. Multivariate analysis, in their study, showed that gender and pre-transplant morbidity ultimately influenced the survival rate rather than RBC transfusion. The Stanford study and Ramos et al.9 did not report the amount of plasma transfused. Our univariate and multivariate analyses showed that only the number of transfused RBC units and the number of transfused plasma units were significantly associated with a decrease in the one-year survival rate. The starting Hb value, the starting platelet count, the cause of the liver failure, the starting creatinine value, the Pugh’s score, the MELD score, a history of previous high abdominal surgery, and previous liver transplantation were not associated with a decrease in the one-year survival rate. This correlates with the observation by Deschênes et al.16 that the severity of cirrhosis was a poor predictor of the six-months survival rate. We calculated that, to obtain a difference of 25% in the mean values of these variables, in the surviving patients, with an alpha error of 0.05 and a power (1-beta ) over 80%, we needed the number of deaths to be between three and 30, which was lower than the actual number (35 deaths) observed in our study.
Van de Watering et al.17 demonstrated a decreased mortality with transfusion of leukoreduced RBC during cardiac surgery. In our centre, allogenic blood products have been leukoreduced since June 14th, 1999. We did not detect an improvement of our survival rate after this date.
Our data seem to suggest that plasma transfusion may be more deleterious than RBC transfusion (Table V). For RBC transfusion, the number of administered units influenced the survival rate. The group receiving one to four units of RBC had a better one-year survival (86.6%) than those receiving more than four units of RBC (62.5%; P = 0.002) or those receiving plasma only (76.9%; P = 0.034). The group receiving only plasma received 1.3 units more plasma (mean 4.9 units) than the group receiving one to four units of RBC (mean 3.6 units) despite the fact that the former group may have been healthier on the basis of starting Hb value (Table I). The absolute presence, rather than the quantity, of plasma administered was significantly associated with a lower survival rate. This was seen most markedly in the patients receiving either no RBC or more than four units of RBC. If plasma has a deleterious effect, it could be that there is a harmful factor that is inconsistently present in plasma or whose effects are manifest inconsistently. It seems that transfusion of four or less units of RBC confers a protective effect in the presence of plasma transfusion. Furthermore, the total amount of blood products given during the whole hospitalization was also looked at, from one month prior the surgery till three months after. Patients that received few transfusions intraoperatively had the same profile in the perioperative period. The converse was also true. In the group that only received plasma, a mean of 9.2 units of plasma was transfused in the perioperative period. Patients who died in this group received even more plasma (mean 13.2 units) during their hospitalization.
Prior studies have shown that RBC transfusion is an independent predictor of morbidity.6–9,18–20 In the last three referenced studies,18–20 RBC or whole blood transfusions increased the incidence of multiple organ failure associated with trauma or major abdominal surgery. Mortality during coronary artery bypass grafting (CABG) also increased.19 However these studies did not differentiate between whole blood and RBC transfusions nor did they mention the amount of plasma transfused.
With transfusion of unwashed RBC some plasma is automatically transfused. If many units of RBC are transfused, eventually an appreciable quantity of plasma is also transfused. If the plasma is the culprit, this may explain a decrease in the one-year survival rate. When plasma units are transfused, a large amount of plasma is given and the survival rate decreases. This might be due to the debilitated and immunosuppressed status of the patients.
Moore et al.18 and Murphy et al.19 suggested that transfusion of allogenic blood could decrease immunity and lead to postoperative infections and multiple organ failure. Another hypothesis was banked blood could activate neutrophils (polymorphonuclear leukocytes) that would be part of the inflammatory cascade present in tissue damage, ischemia, or reperfusion. The authors did not specify which blood product was the culprit.
Vamvakas et al.21 could not demonstrate whether pneumonia or wound infections were increased with allogenic plasma transfusion during CABG. Allogenic plasma transfusion immunomodulation could not be proved in this study. Our patients who received plasma had a decreased one-year survival rate. It could be hypothesized that the correction of the coagulation dysfunction leads to vascular thrombosis at the graft level and hence shortens its survival; however, correction of the coagulation defects with transfusion of plasma was not measured.
Although our findings meet several criteria for causality (biological plausibility, the presence of temporal association, a dose-effect relationship, consistency, specificity, and strength of association),22 one should be cautious in concluding that there is a direct relationship between plasma or RBC transfusion and decreased one-year survival. Unfortunately, plasma or RBC transfusions are not the only variables that can affect the survival rate. In some other institutions, immunosuppressants like tacrolimus and cyclosporine are used separately and tacrolimus is associated with superior survival.14 In our cohort, drugs were associated in different ratios according to three different immunosuppressive research protocols; thus, it was possible for a given patient to have received as many as five different immunosuppressive medications (prednisone, azathioprine, tacrolimus, cyclosporine, and basiliximab). Therefore, no conclusions can be drawn from our data with respect to the relationship between a specific immunosuppressive program and the one-year survival rate.
The retrospective nature of this study has other inherent weaknesses. For example, there was no blood transfusion protocol in place. However, looking back raised the question of whether patients could be identified as having received blood products without absolute necessity. Another weakness was the relatively few patients in this study. Since this study, we have looked at the one-year survival rate of liver transplantations in the five-year period (January 1993 to December 1997) before this study. There were 183 liver transplantations and the one-year survival rate did decrease in the groups receiving only plasma or more than four units of RBC.
In conclusion, in our series of 206 consecutive liver transplantations done at the St Luc Hospital of the CHUM, patients received an average of 2.8 units of RBC and 4.1 units of plasma. Univariate and multivariate analyses of 27 variables showed that RBC and plasma transfusions were associated significantly with a decrease in the one-year survival rate following liver transplantation. One-year survival rate following liver transplantation decreased significantly in association with the intraoperative administration of any amount of plasma or the administration of more than four units of RBC. Plasma transfusion appeared to be more deleterious than RBC transfusion. A direct causal relationship between plasma or RBC transfusion and outcome could not be verified by our study. Despite this limitation, we feel that the result of this study is another call for prudent risk-benefit analysis in transfusion decision-making.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TOPAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
2 Carton EG, Plevak DJ, Kranner PW, Rettke SR, Geiger HJ, Coursin DB. Perioperative care of the liver transplant patient: part 2. Anesth Analg 1994; 78: 382–99.[Medline]
3 Schroeder RA, Johnson LB, Plotkin JS, Kno PC, Klein AS. Total blood transfusion and mortality after orthotopic liver transplantation (Letter). Anesthesiology 1999; 91: 329–30.
4 Steib A, Freys G, Lehmann C, Meyer C, Mahoudeau G. Intraoperative blood losses and transfusion requirements during adult liver transplantation remain difficult to predict. Can J Anesth 2001; 48: 1075–9.
5 Cacciarelli TV, Keeffe EB, Moore DH, et al. Effect of intraoperative blood transfusion on patient outcome in hepatic transplantation. Arch Surg 1999; 134: 25–9.
6 Goodnough LT, Brecher ME, Kanter H, AuBuchon JP. Transfusion medecine - blood transfusion - first of two parts. N Engl J Med 1999; 340: 438–47.
7 Goodnough LT, Brecher ME, Kanter H, AuBuchon JP. Transfusion medecine - blood conservation - second of two parts. N Engl J Med 1999; 340: 525–33.
8 Spahn DR, Casutt M. Eliminating blood transfusions. New aspects and perspectives. Anesthesiology 2000; 93: 242–55.[Medline]
9 Ramos E, Dalmau A, Sabate A, et al. Intraoperative red blood cell transfusion in liver transplantation: influence on patient outcome, prediction of requirements, and measures to reduce them. Liver Transpl 2003; 9: 1320–7.[Medline]
10 Houvenaeghel M, Lefevre P, Samson D, Dyen J, Limet L, Manelli JC. Autologous transfusion by perioperative salvage in orthotopic transplantation of the liver (French). Ann Fr Anesth Reanim 1989; 8: 326–33.[Medline]
11 Findlay JY, Rettle SR, Ereth MH, Plevak DJ, Krom RA, Kufner RP. Aprotinin reduces red blood cell transfusion in orthotopic liver transplantation: a prospective, randomized, double-blind study. Liver Transpl 2001; 7: 802–7.[Medline]
12 Edwards EB, Roberts JP, McBride MA, Schulak JA, Hunsicker LG. The effect of the volume of procedures at transplantation centers on mortality after liver transplantation. N Engl J Med 1999; 341: 2049–53.
13 Trotter JF, Wachs M, Evenson GT, Kam I. Adult-to-adult transplantation of the right hepatic lobe from a living donor. N Engl J Med 2002; 346: 1974–82.
14 O’Grady JG, Burroughs A, Hardy P, Elbourne D, Truesdale A; and the UK and Republic or Ireland Liver Transplant Study Group. Tacrolimus versus microemulsified ciclosporin in liver transplantation: the TMC randomised controlled trial. Lancet 2002; 360: 1119–25.[Medline]
15 Comparative statistics for Québec (December 31st) [cited 8 July 2004]. Available from URL: http://www.quebec-transplant.qc.ca/anglais/public_e.htm.
16 Deschênes M, Villeneuve JP, Dagenais M, et al. Lack of relationship between preoperative measures of the severity of cirrhosis and short-term survival after liver transplantation. Liver Transpl Surg 1997; 3: 532–7.[Medline]
17 van de Watering LM, Hermans J, Houbiers GA, et al. Beneficial effects of leukocyte depletion of transfused blood on postoperative complications in patients undergoing cardiac surgery. Circulation 1998; 97: 562–8.
18 Moore FA, Moore EE, Sauaia A. Blood transfusion. An independent risk factor for postinjury multiple organ failure. Arch Surg 1997; 132: 620–5.[Abstract]
19 Murphy PJ, Connery C, Hicks GL Jr, Blumberg N. Homologous blood transfusion as a risk factor for postoperative infection after coronary artery bypass graft operations. J Thorac Cardiovasc Surg 1992; 104: 1092–9.[Abstract]
20 Maetani S, Nishikawa T, Hirakawa A, Tobe T. Role of blood transfusion in organ system failure following major abdominal surgery. Ann Surg 1986; 203: 275–81.[Medline]
21 Vamvakas EC, Carven JH. Exposure to allogeneic plasma and risk of postoperative pneumonia and/or wound infection in coronary artery bypass graft surgery. Transfusion 2002; 42: 107–13.[Medline]
22 Hill AB. Principles of Medical Statistics, 9th ed. NewYork: Oxford University Press; 1971: 309.
This article has been cited by other articles:
 |
|
 |
|
  M. T. de Boer, M. C. Christensen, M. Asmussen, C. S. van der Hilst, H. G. D. Hendriks, M. J. H. Slooff, and R. J. Porte The Impact of Intraoperative Transfusion of Platelets and Red Blood Cells on Survival After Liver Transplantation Anesth. Analg., January 1, 2008; 106(1): 32 - 44. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Fabbroni and M. Bellamy Anaesthesia for hepatic transplantation CEACCP, October 1, 2006; 6(5): 171 - 175. [Full Text] [PDF]
|
|
|
|
 |
|
 L. Massicotte, S. Lenis, L. Thibeault, M.-P. Sassine, R. F. Seal, and A. Roy Reduction of blood product transfusions during liver transplantation Can J Anesth, May 1, 2005; 52(5): 545 - 546. [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
