Six clinical trials (82 patients) were included in the review: five (72 patients) assessed r-IFNa 2b and one (10 patients) assessed lamivudine.
Clinical response.
The overall estimates of proteinuria remission and sustained proteinuria remission (at least 6 months after completion of antiviral therapy) for trials of IFN only were 65.2% (95% CI: 52.7, 75.9) and 49.8% (95% CI: 33.3, 60.1), respectively. There was evidence of statistical heterogeneity in the latter data set (p=0.056).
Viral response.
HBeAg clearance was reached in 62.0% (95% CI: 50.5, 72.2) of patients, while HBsAg clearance was achieved in 16.4% (95% CI: 4.7, 44.1); there was evidence of statistical heterogeneity (p=0.005). For trials of IFN only, sustained HBeAg and sustained HBsAg were reached in 50.6% (95% CI: 31.7, 69.4) and 19.8% (95% CI: 5.9, 49.1) of patients, respectively; there was evidence of statistical heterogeneity (p=0.065 and p=0.015, respectively).
Tolerability.
The overall drop-out rate was 12.7% (95% CI: 6.4, 23.6). Six patients discontinued antiviral therapy due to myalgia (n=1) and the high cost of r-IFNa 2b (n=5).
The meta-regression analysis showed a significant association between HBeAg clearance and proteinuria remission after r-IFNa 2b. The effects were similar using random-effects or fixed-effect analyses.
Both the Begg and Mazumadar adjusted rank-correlation test and Egger regression asymmetry test showed a low risk for publication bias.