Article Text
Abstract
Background Hepatitis B virus (HBV) is a major cause of liver cirrhosis and hepatocellular carcinoma worldwide. The greatest disease burden is seen in African countries, where liver-related mortality is rising, fewer than 1% are aware of their diagnosis and <0.2% have commenced antiviral treatment. Ascertainment of infection is a major problem in Malawi, where it is estimated that over 95% of people with chronic HBV infection are unaware of their diagnosis.
Methods The Chiwindi serosurvey, nested within a large cross-sectional study on long-term conditions, will test 5,000 adults (18+) resident within rural (Chilumba, Karonga District) and urban (Area 25, Lilongwe) health and demographic surveillance sites run by the Malawi Epidemiology and Intervention Research Unit (MEIRU). Testing started in in 2022 and is expected to end in 2023, and a continuous whole-population sample is being taken from both sites until recruitment targets are met. Monolisa ELISA tests (Bio-Rad, France) are being used to determine serum HBsAg and HBeAg levels. Results are shared with participants, and positive cases not currently taking tenofovir-containing ART for HIV are invited to clinic for treatment eligibility assessment, including the use of transient elastography (TE) and HBV DNA measurement (GeneXpert, Cepheid). Additional cases are being identified through contact tracing and other direct referral pathways.
Results From the first 3,148 tests in Karonga, 133 HBsAg-positive cases were diagnosed. Direct age-standardised seroprevalence was 4.1% (95% C.I. 3.4–4.9%). HBsAg prevalence was higher in men (5.3%; 95% C.I. 4.2–6.7) than women (3.4%; 95% C.I. 2.7–4.4%; p-value = 0.009). Seroprevalence (figure 1) was highest in men aged 30–39 (9.2%; 95% C.I. 6.2–13.4%) and lowest among women aged 18–29 (1.8%; 95% C.I. 1.0–3.3%). 8.0% (95% C.I. 5.1–12.1%) of people living with HIV tested positive compared to 4.1% (95% C.I. 3.4–5.0%; p-value = 0.004) of those without HIV. HBeAg prevalence was 1.0% (1/99). 25 additional participants were identified through contact tracing and direct referrals. Among 134 HBsAg-positive participants reviewed in clinic, 6.5% (95% C.I. 3.0–12.7%) had significant liver fibrosis (TE >7.9kPa), and 22.8% (95% C.I.15.0–33.0%) had HBV DNA greater than 2,000 copies/ml. Linkage with stored samples from 2013 showed that 33/34 (97.1%) of participants were previously HBsAg positive. 10/134 (7.5%) meet treatment thresholds based on EASL criteria. None of the cases identified through the serosurvey were previously aware of their diagnosis.
Conclusions There is a substantial burden of undiagnosed hepatitis B infection in Karonga. In this rural community study, at first evaluation, 7.5% require treatment. We plan to expand the serosurvey to Lilongwe and will evaluate the performance of different serum biomarkers in predicting liver fibrosis and cirrhosis.