Glucocorticosteroids for people with alcoholic hepatitis (Cochrane review)

• Main Authors: Chavdar S Pavlov, Daria L Varganova, Giovanni Casazza, Emmanuel Tsochatzis, Dimitrinka Nikolova, Christian Gluud
• Format: Article
• Language: Russian
• Published: "Consilium Medicum" Publishing house 2019-08-01
• Series: Терапевтический архив
• Subjects: glucocorticosteroids; alcoholic hepatitis; randomized clinical trial; meta-analysis
• Online Access: https://ter-arkhiv.ru/0040-3660/article/viewFile/33614/pdf
• ISSN: 0040-3660; 2309-5342

Alcoholic hepatitis (AH) is a form of alcoholic liver disease. Glucocorticosteroids (GCS) are used as anti - inflammatory drugs for people with alcoholic hepatitis. Aim. To assess the benefits and harms of GCS in people with AH. Material and methods. We identified trials through electronic searches in Cochrane Hepato-Biliary's (CHB) Controlled Trials Register, CENTRAL, MEDLINE, Embase, LILACS, and Science Citation Index Expanded. We considered for inclusion randomised clinical trials (RCTs) assessing GCS versus placebo/no intervention in adult participants with AH. We allowed co - interventions in the trial groups if they were similar. We followed Cochrane methodology, CHB Group methodology using Review Manager 5 and Trial Sequential Analysis(TSA) to perform meta - analysis (M-A), assessed bias risk of the trials, certainty of evidence using GRADE. Results and discussion. Sixteen trials fulfilled the inclusion criteria. Fifteen trials provided data for analysis (927 participants received GCS, 934 - placebo/no intervention). The GCS were administered to adult participants at different stages of AH orally or parenterally for a median of 28 days. There was no evidence of effect of GCCs on our primary outcomes all - cause mortality up to 3 months following randomisation (RR 0.90, 95% CI 0.70-1.15; n=1861), on health - related quality of life (MD - 0.04 points; 95% CI -0.11-0.03; n=377; trial = 1) (EQ-5D-3L scale), on the occurrence of serious adverse events during treatment (RR 1.05, 95% CI 0.85-1.29; n=1861). We found no evidence of a difference between the intervention groups. The risk of bias was high in all the trials except one. The certainty of evidence was very low or low. One of the trials seems to be not industry - funded. Conclusion. We found no evidence of a difference between GCS and placebo or no intervention on all - cause mortality, health - related quality of life, and serious adverse events during treatment. We cannot exclude increases in adverse events and cannot rule out significant benefits and harms of GCSs. Future trials ought to report depersonalised individual participant data.