RISK FACTORS PREDICTING SURVIVAL OF LIVER TRANSPLANTS

Abstract

P578 Aim: In a study population consisting 388 adult liver transplantations with 352 recipients risk factors predicting graft survival of liver transplants were studied. Methods: Predictive models were developed for analysing risk factors of liver transplantation outcome. Univariate testing of examined risk factors was performed using Person Chi-squared test for proportions in the case of nominal covariates and the Cochran-Armitage test for trend in the case of ordinal covariates and for time distribution a linear regression model was used. To study the joint effect of multiple factors on graft survival proportional hazard models were used to assess risk factors related to donor and recipient characteristics and on graft rejection exponential regression models were used. Recipient’s and donor’s age, gender compatibility, blood group and HLA compatibility, presensitisation to HLA antibodies (PRA), pretransplant cytotoxic crossmatch test, recipient’s cytomegalovirus status, cold ischemia and anhepatic time, total bleeding and blood transfusions were assessed related to graft survival time, time to acute rejection, graft loss to chronic rejection or chronic dysfunction. Results: Mean recipient age was 47yr, donor age 40 yr, main indications for liver transplantations were cholestatic liever disease in 146 patients, other cirrhosis 74, acute liver failure 74, tumour 26. PRA was positive in 18%, crossmatch in 19%, mean cold ischaemia time 6.3hr, anhepatic time 59min, mean number of blood transfusion 12 units. Frequency of acute rejection was 44%, within the first three months 38%, 7% had steroid resistant rejection and 10% experienced multiple rejections. In the univariate analysis, the significant predictors of graft survival were male gender, gender incompatibility if male recipient- female donor, blood transfusions over 20 units, long CIT, and liver tumour. Graft loss due to chronic graft dysfunction correlated to BDR mismatches. There was a tendency for more rejections with positive crossmatch (P=0.065) and for steroid resistant rejection with PRA positivity (p=0.055). In the multivariate analysis only gender compatibility, recipient blood group and number of blood transfusions were significant risk factors for graft survival. Risk ratio for male recipient and female donor was 1.68 compared with female recipient and male donor. Risk ratio for graft survival in blood group B was 1.75 and for blood group 0 1.68 comapred with A. In multivariate analysis immunological risk factors were not any more significant. Conclusions: In univarite analysis immunological risk factors were found to be significant but in multivariate analysis only gender mismatch was significant for graft survival but not for rejection.