Resolving the ACLF paradox: discharge hepatic reserve, not admission severity, predicts recovery

Objective Acute-on-chronic liver failure (ACLF) is associated with high short-term mortality. However, long-term outcomes among hospital survivors remain incompletely characterised. We evaluated postdischarge survival, readmissions and hepatic recompensation in patients hospitalised with acute decompensation (AD), with or without ACLF. Method This is a single-centre, retrospective, real-world cohort study with 387 patients with cirrhosis admitted with AD between 2018 and 2021; 69 had ACLF at or during index admission. Survivors were followed for 36 months. Competing risk analyses were used to assess mortality, liver transplantation and recompensation according to Baveno VII criteria. Results Among 330 patients discharged alive, cumulative 36-month survival was higher in ACLF cohort than in patients without ACLF (52.6% vs 40.7%, p=0.032) when liver transplantation is treated as competing risk. Model for end-stage liver disease-sodium (MELD-Na) at discharge independently predicted lower postdischarge survival (HR 1.08, 95% CI 1.04 to 1.11, p<0.001), whereas ACLF status did not. Recompensation at 36 months occurred more frequently in ACLF survivors when compared with patients without ACLF (21.1% vs 9.9%). In multivariable competing risk analysis, ACLF independently predicted recompensation (subdistribution HR (sHR) 2.23, 95% CI 1.10 to 4.52, p=0.027), while higher MELD-Na reduced its likelihood (sHR 0.91, 95% CI 0.86 to 0.97, p=0.006). Conclusion Among patients with cirrhosis discharged from hospital following AD, prior ACLF does not confer a worse long-term prognosis and is independently associated with higher rates of delayed recompensation. Hepatic reserve at the time of discharge, rather than admission severity, determines long-term outcome.