Antiviral prophylaxis for preventing reactivation of hepatitis B virus in rheumatic patients: a systematic review and meta-analysis.

To estimate the risk of reactivation of hepatitis B virus (HBV) and evaluate the effectiveness of antiviral prophylaxis (AVP) in patients with different status of HBV infection undergoing antirheumatic therapies. We searched Cochrane Library, Medline, and EMBASE for randomized controlled trials (RCTs), quasi-RCTs, non-RCTs, cohort studies, or case series studies examining reactivation of HBV in patients undergoing antirheumatic therapy with or without AVP. We estimated the HBV reactivation rate (HRR) and its 95% confidence interval (CI) among different patient groups (indirect comparison). We also calculated rate ratio (RR), rate difference (RD) with their 95% CIs, and the number needed to treat (NNT) of AVP (direct comparison). Fifty-three case series studies with 2162 patients were included. The RD of AVP was - 0.13 (95% CI - 0.21 to - 0.05) for all patients, - 0.16 (95% CI - 0.26 to - 0.06) for rheumatic patients with chronic HBV infection, but not statistically significant for patients with other status of HBV infection. Lamivudine (RD - 0.10, 95% CI - 0.25 to 0.05) was less effective than other prophylactic antiviral drugs (RD - 0.31, 95% CI - 0.52 to - 0.11). The HHR varied from 55 to 5% by HBV status and treatment. There is limited evidence that AVP was effective for preventing reactivation of HBV in patients undergoing antirheumatic therapy. The effectiveness varies by patient HBV status and antiviral regimens. Rheumatic HBV carriers may be more beneficial from AVP, and lamivudine may be inferior to other AVP regimens. Findings in this study warrant further investigation in rigorous RCTs.