Journal Logo

BRIEF COMMUNICATIONS: Clinical Transplantation

Red blood cell alloantibodies and liver transplantation in Chinese patients

Au, W. Y.1 4; Liu, C. L.2; Lo, C. M.2; Fan, S. T.2; Lam, M. F.1; Lam, C. K.3

Author Information
doi: 10.1097/01.TP.0000065741.31599.9C
  • Free


Studies have shown the presence of red blood cell (RBC) alloantibodies in 8% to 14% of white recipients of liver transplants (1). It is a significant cause of perioperative hemolysis and jaundice. A higher degree of blood group homogeneity in Chinese recipients results in a lower rate of alloimmunization even in heavily transfused populations (2). The spectrum of RBC alloantibodies and indications for transplants are also different. We investigate the significance of pretransplant RBC alloantibodies in adult Chinese recipients of liver transplants.

Consecutive adult transplants from 1995 to 2002 were evaluated. A pretransplantation RBC alloantibody screen was performed using a commercial two-cell panel (Biopool International, Denver, CO) and an in-house group O cell expressing Miltenberger determinants GP.Mui (Mi), present in 6.3% of the Chinese population (3). The test was read at immediate spin after 30 min at 37°C in saline and then with polyspecific antiglobulin (Ortho Diagnostic, Raritan, NJ) with microscopy confirmation of negative results. A direct antiglobulin test was performed as auto-control. A total of 84 consecutive renal and 350 marrow transplant recipients in the same period were similarly screened.

A total of 200 liver transplants performed in 192 patients (median age 46 years, range 17–68, 29% >50; 137 men, 55 women) were evaluated. Hepatitis B virus and hepatocellular carcinoma were present in 141 (73.4%) and 37 (19.2%) cases, respectively. Emergency transplantation was performed in 74 cases (37%), and 84 organs (42%) were cadaveric. Preoperative intensive care (n=60) and renal failure (n=29) were frequent. None of these factors were significant predictors of organ or patient survival on univariate Kaplan Meier analysis (data not shown). A total of 17 patients (8.8%) had alloantibodies detected (Table 1). They included 10 men and 7 women with a median age of 47 years (range 24–63 years). The most common antibodies were anti-Mi (n=5), anti-E (n=4), anti-e (n=2), and Lewis-a (n=2). Six patients had multiple RBC alloantibodies (41%), and 7 of 25 RBC alloantibodies (28%) could not be specified. The incidences of antibodies in marrow and kidney recipients were 1.6% and 11.9%, respectively. The incidence of RBC alloantibodies in solid-organ recipients is significantly higher than in the background hospital population (3.7%) (4), but the spectrum is comparable.

Table 1:
Table 1. Clinical characteristics and outcome in 17 liver recipients with red blood cell alloantibodies

Three transplants were performed across the ABO barrier (involving pretransplant plasmapheresis and ABO-mediated hemolysis) and were excluded from further analysis. Donors and recipients were ABO identical in 110 cases and compatible in 87 cases. There was no significant difference in age (t test), gender, and ABO blood group (χ2 test) between liver recipients with and without RBC alloantibodies (Table 2), and the pretransplant hemoglobin and bilirubin levels were also comparable (t test). However, there was increased transfusion in RBC alloantibody-positive cases and higher postoperative bilirubin levels during the second week at comparable levels of hemoglobin. Depending on transplantation urgency, blood demand, and RBC alloantibody specificity, 7 to 86 antigen-positive units were issued in five cases, including only two survivors. At a median follow-up of 31 months, 12 patients are alive without any retransplant. Repeat screening at a median of 6 months showed the disappearance of RBC alloantibodies in only one case (orthotopic liver transplantation no. 167). Univariate Kaplan Meier analysis showed a reduced median survival of organs and patients in cases with detectable pretransplant RBC alloantibodies. However, the significance was lost in the Cox model multivariate analysis with seven other preoperative and perioperative variables known to affect transplant outcome (Table 2).

Table 2:
Table 2. Statistical comparison between liver recipients with and without red blood cell alloantibodies

A standard type and screen precede most major surgeries. Despite blood salvage and other measures, the transfusion requirements in liver transplantation are high, and meticulous preparation before the operation is necessary (5). Furthermore, antibody and hemolysis investigations after the transplant may be difficult as the result of interdonor compatibility, development of autoimmune (6) and alloimmune (7) antibodies, and nonimmune-mediated hemolysis. Little has been published on RBC alloantibody problems in liver transplantation. The Pittsburgh group reported 31 cases of clinically significant RBC alloantibodies in 616 transplants (5%) in 496 adults (6.3%), which is comparable to our experience. The most common RBC alloantibodies were anti-K, anti-E, anti-D, and anti-Jka. Multiple RBC alloantibodies were present in 39.5% of cases (1). The incidence was higher in female patients with primary biliary cirrhosis and lower in liver tumor cases (1). The series also included significant numbers of retransplants (22%) and ABO-mismatched grafts (6%) and is probably ethnically diverse.

Our study in Chinese patients illustrated a number of interesting points. First, the increased incidence of RBC alloantibodies in solid-organ transplant recipients was probably the result of blood-product exposure. The increased passage of gut antigens and reduced liver clearance of antibodies may also contribute to alloimmunization in cirrhotic patients with hypergammaglobulinemia. The low incidence in heavily transfused marrow recipients is attributable to chemotherapy and immunosuppression. The high transfusion requirement in liver transplant recipients also means that alloimmunization problems become more significant. In marrow transplants, this is less of an issue because the blood group is completely altered in the course of the procedure. Second, we illustrated that the spectrum of antibodies follows strict ethnic patterns. Because 99.73% and 99.94% of Chinese patients are Rh-D+ve and K-ve, the corresponding antibodies were undetectable in our screening. Conversely, Mi determinants were present in 2% of the recipients and may have mediated severe hemolytic reactions; these must be screened with proper cell panels in liver recipients of Chinese descent. Unlike other published reports, our study showed that RBC alloantibody problems may persist despite posttransplant immunosuppression. Third, we illustrated that the presence of RBC alloantibodies was associated with significant delayed hemolysis, especially during the second week, and increased blood requirements. Part of this can be attributed to the transfusion of antigen-positive units because of logistic reasons. To reduce postoperative delayed hemolysis, it may be reasonable (if possible and foreseeable) to reserve most compatible units for the end of the operation (5). Increased transfusion requirements and high bilirubin are known adverse factors for patient and graft survival (8). Because the presence of RBC alloantibodies, incompatible transfusion, total transfusion need, delayed hemolysis, and survival are all interrelated, the increased transplant risk associated with the presence of RBC alloantibodies and the transfusion of antigen-positive blood is difficult to assess. However, unlike cold agglutinins, which do not seem to affect transplant results (9), our small retrospective series indicates that RBC alloantibodies might affect the outcome. The survival difference in our data is mainly the result of four early postoperative deaths in the RBC alloantibody positive group, and the salvage of seven RBC alloantibody negative cases by early and late regrafts. Although hemolysis and incompatible transfusion were not the direct cause of death in any case, their contribution to a higher incidence of early death cannot be overlooked. It is also notable from the Pittsburgh experience that four of eight patients requiring the use of less compatible units died early (1). And although RBC alloantibodies were not a factor affecting the retransplant decision, the stabilization and support of RBC alloantibody cases may be clinically easier. It must be strongly emphasized, however, that the number of cases and events were both small, and the preoperative, perioperative, and postoperative variables were heterogeneous. These render a conclusive statistical analysis difficult. Thus, survival difference was lost on multivariate analysis. Nevertheless, we may conclude that an efficient RBC alloantibody screening and supply of fully compatible RBCs may improve operation and survival results and simplify the investigation of hemolysis and jaundice after liver transplantation.


1. Ramsey G, Cornell FW, Hahn LF, et al. Red cell antibody problems in 1000 liver transplants. Transfusion 1989; 29 ( 5): 396–400.
2. Ho HK, Ha SY, Lam CK, et al. Alloimmunization in Hong Kong southern Chinese transfusion-dependent thalassemia patients. Blood 2001; 97 ( 12): 3999–4000.
3. Lin CK, Mak KH, Cheng G, et al. Serologic characteristics and clinical significance of Miltenberger antibodies among Chinese patients in Hong Kong. Vox Sang 1998; 74 ( 1): 59–60.
4. Wong KF, Tse KT, Lee AW, et al. Is antenatal antibody screening worthwhile in Chinese? Br J Haematol 1997; 97 ( 4): 917–919.
5. Steib A, Freys G, Lehmann C, et al. Intraoperative blood losses and transfusion requirements during adult liver transplantation remain difficult to predict. Can J Anaesth 2001; 48 ( 11): 1075–1079.
6. Au WY, Lo CM, Hawkins BR, et al. Evans’ syndrome complicating chronic graft versus host disease after cadaveric liver transplantation. Transplantation 2001; 72 ( 3): 527–528.
7. Au WY, Lo CM, Fan ST, et al. Life-threatening abo-mediated hemolysis after cadaveric orthotopic liver transplantation. Transplantation 2002; 74 ( 2): 285–286.
8. Cacciarelli TV, Keeffe EB, Moore DH, et al. Effect of intraoperative blood transfusion on patient outcome in hepatic transplantation. Arch Surg 1999; 134 ( 1): 25–29.
9. Marsman WA, Moore SB, Kondo M, et al. Do cold agglutinins have any impact on the outcome of liver transplantation? Transplantation 1996; 62 ( 11): 1673–1676.
© 2003 Lippincott Williams & Wilkins, Inc.