Liver transplant outcomes in HIV-infected patients

Abstract

We read with interest the article entitled ‘Liver transplant outcomes in HIV-infected patients: a systematic review and meta-analysis with a synthetic cohort’ by Cooper et al.[1]. This article is very interesting because it describes systematically 15 cohort studies and 49 case series with individual patient data. It is now generally accepted that liver transplantation in HIV-infected patients is feasible; in HIV/hepatitis C virus (HCV) co-infected patients, we and others have demonstrated that survival at 1 year after liver transplantation ranges from 58% to 89% [2,3]; when we compared these rates with a series of mono-infected patients, survival was lower in the co-infected group, the 2-year and 5-year patient survival rates being 73% vs. 91% and 51% vs. 81%, respectively, (P = 0.004). Moreover, concerning HIV/hepatitis B virus (HBV) co-infected patients, survival was excellent, the rate reaching 100% survival at 5 years [4]. In the study by Cooper et al. [1], some points need to be detailed or explained and discussed. The constitution of a synthetic cohort from grouped case data is an original and interesting method. However, the character of ‘grouped cases’ with a lack of details concerning the indications of liver transplantation (hepatocellular carcinoma or end-stage liver cirrhosis) and virologcial characteristics (HCV viral load at the time of liver transplantation, HCV genotype) is a concern. More generally, provision of such information is particularly important for overcoming the problem of patient heterogeneity. Concerning HIV/HCV co-infected patients, the authors considered that HCV was a negative predictor of patient survival in the unadjusted model [0.23, 95% confidence interval (CI) 0.07–0.79],l but not when it was adjusted by other key predictors of patient survival (0.54, 95% CI 0.17–1.80). This result is surprising; the authors need to provide details on which other predictor factors they used. The authors found that a Model End-Stage Liver Disease (MELD) score of at least 14 is a prognostic factor for survival in nonadjusted analyses. Interpretation of this cut-off point is difficult because it pools different subgroups of patients, such as those with hepatocellular carcinoma and a low MELD score and those with complicated cirrhosis with a high MELD score. In our cohort of HIV/HCV co-infected patients, we demonstrated that the MELD score as a continuous variable was an independent factor for survival under multivariate analysis [2]. The authors found a significant difference in survival as a function of the year of publication (</>2001). This point is interesting; we agree that modifications to antiretroviral regimens may influence postliver transplantation survival [5]; other important factors such as registration on the waiting list of patients with less severe liver disease, or a more systematic introduction of anti-HCV therapy after liver transplantation, also need to be mentioned. The authors found a percentage of fibrosing cholestatic hepatitis of 12% at 5 years after liver transplantation. This very severe complication has been reported in 5% HBV or HCV mono-infected patients. We recently demonstrated that this type of complication occurred in 11 (19%) of 59 HIV/HCV co-infected patients transplanted for end-stage liver disease. Clearly, it is necessary to prevent this high recurrence rate through the early detection of predictive factors and the early initiation of anti-HCV therapy [6]. In conclusion, this study is important because it demonstrates that survival after liver transplantation in HIV-infected patients is similar to that of HIV-negative patients. However, other large studies are needed to individualize prognosis factors after liver transplantation in a well characterized subgroup of HIV-infected patients. Acknowledgements Conflicts of interest None declared.