Background: Empirical, updated country-level estimates on the proportion of cirrhosis attributable to viral hepatitis are required. We estimated the prevalence of hepatitis B virus (HBV) and hepatitis C virus (HCV) infection in patients with cirrhosis at country, regional, and global levels as an approximation for the fractions of cirrhosis attributable to viral hepatitis.

Methods: In this systematic review, we searched MEDLINE, Embase, Web of Science, and Scielo between Jan 1, 1993, and Aug 1, 2021. Studies were eligible if they reported on the prevalence of both HBV and HCV infection in representative studies of at least 20 patients with cirrhosis. Studies were excluded if they used first-generation HCV assays or were from a selected population of patients with cirrhosis (eg, patients selected based on specific causes, veterans, injecting drug users). Two authors (CJA and CdM) selected and extracted aggregated data from the selected publications. Data were extracted for study recruitment period, age, sex, and cause of cirrhosis, among others. Data about heavy alcohol consumption and non-alcoholic fatty liver disease (NAFLD) were also extracted when available. Aggregated data from studies from key publications were requested from the authors of the original study if selection of patients was unclear or information on causes was missing. We estimated the country-specific prevalence of causes of cirrhosis by pooling study-level data from the same country using a random-effects model. Subsequently, we estimated the regional (WHO region and UN subregion) and global prevalence by weighting the country-specific prevalence by the number of new liver cancer cases that occurred in 2020, as estimated in GLOBOCAN. The study was registered with PROSPERO, CRD42020149323.

Findings: Our database searches identified 21 338 records, and a further nine records were identified by scanning references of key publications. After excluding duplicates and assessing full-text articles for eligibility, 520 publications from 86 countries or territories (and reporting on 1 376 503 patients with cirrhosis) were included in the systematic review. The prevalence of HBV infection was lower among patients with cirrhosis in Europe, the Americas, and Oceania (UN subregional prevalence ranges 3-14%) than in Africa and Asia (8-61%). HCV infection prevalence was heterogenous, even within regions (12-83%). The combined prevalence of HBV and HCV infection exceeded 50% in most Asian and African regions. Globally, among patients with cirrhosis, 42% had HBV infection and 21% had HCV infection. The contribution of heavy alcohol use was highest in Europe (country range 16-78%), the Americas (17-52%), and Oceania (15-37%) and lowest in Asia (0-41%). Data on NAFLD were limited.

Interpretation: HBV and HCV could account for almost two thirds of the global burden of cirrhosis. With the availability of effective interventions for the prevention or treatment of HBV and HCV, the data presented in this study will help to effectively allocate resources towards viral hepatitis elimination and to design interventions at the country level.

Funding: International Agency for Research on Cancer, World Health Organization.