Chronic hepatitis B (HBV) infection is widespread in sub-Saharan Africa, and the presence of cirrhosis is a common indication for antiviral treatment. The aspartate aminotransferase-to-platelet ratio index (APRI) is a low-cost biomarker for cirrhosis, and the World Health Organization has recommended an APRI cutoff value of ≥ 2.0 for diagnosis of cirrhosis in HBV patients, based mainly on data from Asia.
In this study, individual participant data from chronic HBV patients from eight sub-Saharan countries were used to determine the most appropriate diagnostic thresholds for APRI, using transient elastography as a gold standard and a Bayesian statistical approach. An APRI value ≥ 0.65 was optimal to rule-in cirrhosis (sensitivity 56.2%, specificity 90%), and an APRI value of < 0.36 was optimal to exclude cirrhosis (sensitivity 80.6%, specificity 64.3%). These thresholds appeared to be even more accurate in the subset of patients who had undergone liver biopsy. In contrast, the WHO APRI threshold value of 2.0 detected only 17% of patients with cirrhosis. The authors conclude that improved APRI thresholds that rule-in and rule-out cirrhosis may optimize treatment recommendations.
This study has immediate implications for chronic HBV treatment decisions in sub-Saharan Africa. While the results should be validated, the large sample size and multinational enrollment support immediate implementation of these new APRI threshold values, instead of the existing WHO recommendation. This is particularly true for patients who do not have access to transient elastography.
Johannessen A, et al. Systematic review and individual-patient data meta-analysis of non-invasive fibrosis markers for chronic hepatitis B in Africa. Nature Communications 2023:14,45.