Background: The prognosis of autoimmune acute liver failure (ALF) without liver transplantation (LT) is poor worldwide. We subanalyzed infectious complications of autoimmune ALF using data of nationwide surveys between 2010 and 2015 retrospectively and tried to determine when to evaluate the efficacy of corticosteroid (CS) treatment or abandon it for LT based on objective data.
Methods: One hundred and forty-four patients with autoimmune ALF, comprising 79 ALF with coma ≤ I, 52 ALF with coma ≥ II and 13 late onset hepatic failure (LOHF), were analyzed.
Results: CS was administered to 140 (97%) patients. Thirty-seven (26%) patients had infectious complications. Patients with infection revealed more advanced disease type (p < 0.001) and poorer spontaneous survival (p < 0.001) than those without infection. Median (interquartile range) duration between diagnosis of ALF and onset of infection was 18.5 (11-36) days, and that between introduction of CS and onset of infection was 17 (10.5-36) days. Seventy-nine (55%) recovered without LT, 14 (10%) received LT and 51 (35%) died without LT. Dead or transplanted patients were older (p = 0.0057), and revealed more advanced liver failure (p < 0.001) and more occurrence of infection (p < 0.001).
Conclusions: A critical point for evaluating the efficacy of CS treatment and switching to LT is at most 2-week after diagnosis of ALF and introduction of CS. More important, we should accelerate the point and prepare for LT in cases of ALF with coma ≥ II and LOHF, and we should have performed LT by then at the latest in case of failure to improve.
Keywords: Acute liver failure; Autoimmune hepatitis; Corticosteroid; Infection; Liver transplantation.