Predictors of heavy drinking after liver transplantation for alcoholic liver disease in Denmark (1990–2013): a nationwide study with competing risks analyses

Abstract

Objective. Heavy drinking following liver transplantation decreases survival. Little is known of predictors of heavy drinking, which should guide clinicians identifying patients at high risk of return to heavy drinking after transplantation. Material and methods. We calculated the cumulative incidence of heavy drinking among patients transplanted for alcoholic liver disease in Denmark 1990–2013. We then analyzed pre-transplant demographic and psychiatric characteristics as predictors of post-transplant heavy drinking. Information was obtained from medical records, from nationwide registries and by interview. Results. Among 156 liver-transplanted patients, the cumulative incidence of heavy drinking was 18%, 24% and 27% after 5, 10 and 15 years post-transplant. In univariate analyses of pre-transplant predictors of heavy drinking after transplantation, younger age (p < 0.001), being retired (p = 0.007), anxiety (p = 0.04), personality disorder (p = 0.05) and no lifetime diagnosis of alcohol dependence (p = 0.03) were associated with heavy drinking after transplantation. Smoking (p = 0.06) tended to be associated, whereas depression (p = 0.7) or being married was not (p = 0.7). In the multivariate analysis, only younger age (p = 0.03), being retired (p = 0.007) and no lifetime diagnosis of alcohol dependence (p = 0.003) remained significant predictors. Heavy drinking after transplantation decreased survival beyond 5 years post-transplant (p = 0.004). Conclusions.There is a high incidence of heavy drinking after liver transplantation for alcoholic cirrhosis in Denmark. Younger age, being retired and no lifetime diagnosis of alcohol dependence were predictors of heavy drinking after transplantation.

Key Words:

• alcohol recidivism
• alcoholic liver disease
• competing risk
• liver transplantation
• survival

Acknowledgments

We thank Joachim Knop for valuable comments. We thank The Research Fund of Copenhagen University Hospital, Rigshospitalet, who gave 1 year of financial support. We also thank Novartis who gave an unrestricted amount of DK 25,000.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.