Background: Since the results obtained in controlled studies of antiviral treatment of chronic hepatitis by the hepatitis B virus are frequently insignificant the authors have undertaken a combined study of the different controlled trials with the meta-analysis technique.
Methods: The results of 7 controlled trials of treatment with adenosin arabinoside (AAR) and 18 published studies on therapy of a control group with interferon were studied. Meta-analysis was performed by two methods: Mantel expressing the results in odds ratio (OR) and Cochram-Dersimonian-Laird which consider the results in differences of risk (RD).
Results: The meta-analysis of AAR treatment shows a mean difference of response of 11% with no homogeneity among the different studies published. This lack of homogeneity may be due to the different selection criteria of the patients which determine very variable degrees of response. The controlled studies with interferon showed a mean difference of response of 20% with homogeneity of effects. No differences were observed in the degrees of response according to whether recombinant lymphoblastoid interferon was used or if the dose was of 5 or 10 megaunits per day. Non were differences observed in degrees of response with respect to the time in which the loss of viral DNA was evaluated.
Conclusions: Following analysis of the results it was concluded that interferon is clearly effective in the treatment of chronic hepatitis B and based on published studies adenosin arabinoside demonstrate less efficacy.