RESUMEN
The selection of liver transplantation (LT) candidates with alcohol-use disorder (AUD) is influenced by the risk of alcohol relapse (AR) after LT. We aimed to investigate the risk factors of AR after LT and its impact on graft and recipient outcomes. A retrospective study was conducted that included all consecutive patients with AUD undergoing LT from January 2004 to April 2016 (n = 309), excluding patients with alcoholic hepatitis. Odds ratios (ORs) and 95% confidence intervals (CIs) for AR were analyzed by multinomial logistic regression. Cox regression with time-dependent covariates was used to analyze patient survival and graft cirrhosis. There were 70 (23%) patients who presented AR (median follow-up, 68 months), most of them (n = 44, 63%) presenting heavy AR. The probability of heavy AR was 2.3%, 7.5%, 12%, and 29% at 1, 3, 5, and 10 years after LT, respectively. The independent risk factors for heavy AR included a High-Risk Alcoholism Relapse (HRAR) score ≥3 (OR, 2.39; 95% CI, 1.02-5.56; P = 0.04) and the duration of abstinence (months) before LT (OR, 0.81; 95% CI, 0.66-0.98; P = 0.03). In recipients with <6 months of abstinence before LT, the probability of heavy AR after LT was higher in patients with an HRAR score ≥3 than in those with an HRAR score <3 (20%, 36.7%, and 47% versus 6.8%, 12.4%, and 27% at 1, 3, and 5 years, respectively; log-rank 0.013). The risk of graft cirrhosis was increased in patients with heavy AR (hazard ratio, 3.44; 95% CI, 1.58-7.57; P = 0.002) compared with nonrelapsers, with no differences in patient survival. In conclusion, the HRAR score is helpful in identifying the risk of harmful AR after LT in candidates with <6 months of alcohol abstinence without alcoholic hepatitis. These patients could benefit from a longterm integrative patient-centered approach after LT until lifestyle changes are implemented.