Fifty-three RCTs were analysed to determine the effectiveness of traditional Chinese medicines on chronic hepatitis B. Two hundred and thirty studies were used to describe the traditional Chinese medicines used. Overall study quality was poor. Of the RCTs, none reported the methods used for randomisation, allocation concealment or blinding. Sixteen RCTs had a Jadad score of 3 points; 37 had a score of 2 points.
Traditional Chinese medicine versus interferon (16 RCTs; n=1,918 patients): Traditional Chinese medicine had a significantly better effect on normalising serum alanine aminotransferase levels compared with interferon treatment alone (OR 2.42, 95% CI 1.51 to 3.89), although significant heterogeneity was detected (I2=58.7%). Traditional Chinese medicine and interferon were equivalent in reducing serum hepatitis B e antigen (OR 1.60, 95% CI 1.00 to 2.54) and clearing serum hepatitis B virus deoxyribonucleic acid (OR 1.31, 95% CI 0.87 to 1.98).
Traditional Chinese medicine plus interferon versus interferon alone (18 RCTs; n=1,738 patients): Traditional Chinese medicine plus interferon significantly reduced serum hepatitis B e antigen (OR 2.17, 95% CI 1.74 to 2.72) compared with interferon treatment alone, as well as improving clearance of serum hepatitis B virus DNA (OR 2.05, 95% CI 1.59 to 2.65) and improving normalization of serum alanine aminotransferase levels (OR 3.07, 95% CI 2.35 to 4.00). No significant heterogeneity was detected.
Traditional Chinese medicine versus lamivudine (six RCTs; n=723 patients): Traditional Chinese medicines were significantly more effective than lamivudine for normalising serum alanine aminotransferase levels (OR 1.96, 95% CI 1.15 to 3.32), although significant heterogeneity was detected (I2=60.8%). No differences between treatment groups were observed for the reduction in serum hepatitis B e antigen or the clearance of serum hepatitis B virus DNA.
Traditional Chinese medicine plus lamivudine versus lamivudine alone (14 RCTs; n=1,548 patients): Traditional Chinese medicine plus lamivudine significantly reduced serum hepatitis B e antigen (OR 2.54, 95% CI 1.95 to 3.32) compared with lamivudine alone, improved clearance of serum hepatitis B virus DNA (OR 3.20, 95% CI 2.09 to 4.92), and improved normalization of serum alanine aminotransferase levels (OR 3.40, 95% CI 2.45 to 4.70). No significant heterogeneity was detected.
The review presented data of the most commonly used herbs in the traditional Chinese medicine formulations. Four herbs were reported in detail: Astragalus (Huang Qi), Polygonum cuspidatum (Hu Zhang), Radix et rhizoma rhei (Da Huang), and Phyllanthus urinaris (Ye Xian Zhu).
Subgroup analyses of 16 higher quality RCTs (Jaded score=3) did not influence the results.
Publication bias was not detected.
One of the authors disclosed financial links with several pharmaceutical companies.