Background & aims: Prednisolone is the first-line therapy for severe alcoholic hepatitis (AH). Patients with severe alcoholic hepatitis often develop severe infections that negatively impact short-term prognosis.
Methods: We performed this meta-analysis to assess the effect of corticosteroids on the occurrence of and mortality from infections in patients with severe alcoholic hepatitis. Randomized controlled trials examining the use of corticosteroids in severe alcoholic hepatitis and reporting data on infection rates and mortality were included. Random effects model was used to pool the data comparing arms with and without steroids for the occurrence of infection, 28-day mortality and cause specific mortality.
Results: Of 1062 patients (528 steroids treated) without infection at baseline from 12 studies, infection was reported in 213 (113 steroids treated) patients without differences comparing arms with and without steroids (OR: 0.98; CI: 0.49-1.94). However, frequency was higher for occurrence of fungal infections among steroid-treated patients (eight of 528 vs. one of 534; P = 0.02). Steroids provided mortality benefit at 28 days (OR: 0.55; CI: 0.34-0.90) mainly for liver failure-related death (OR: 0.46; CI: 0.24-0.87) without differences on mortality from infection (OR: 1.19; CI: 0.38-3.73) or gastrointestinal bleeding (OR: 0.90; CI: 0.43-1.87). Three of nine patients with fungal infections died, all in corticosteroid arm.
Conclusions: Corticosteroids do not increase occurrence of or mortality from bacterial infections in patients with severe alcoholic hepatitis. Further studies are needed to develop strategies of reducing the risk of fungal infection with use of steroids for patients with severe alcoholic hepatitis.
Keywords: alcoholic hepatitis; corticosteroids; infection; meta-analysis.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.