Prevalence of hepatitis B virus infection in non-Hodgkin lymphoma: a systematic review and meta-analysis

Background and aim: A recent meta-analysis has demonstrated an association between hepatitis C virus and non-Hodgkin lymphoma (NHL). There is also evidence on the association between hepatitis B virus (HBV) and NHL. The aim of this study was to evaluate this evidence using a meta-analytic approach. Methods: We searched the MEDLINE database from 1962 to 2008 for case-control studies that have reported the association of HBV with NHL. We calculated the odds ratio (OR) and 95% confidence intervals (CI) to assess the prevalence of HBV infection and pooled the results using three different statistical models. Results: Our search yielded 12 studies with 11 studies (3262 NHL patients, 1 523 205 controls) evaluating HBV infection in NHL and one study (3888 HBV-infected individuals, 205 203 controls) that had investigated for NHL in HBV infection. The OR of detecting HBV infection in NHL when compared with the control population was 2.56 (95% CI, 2.24-2.92) by the fixed effects model; 2.61 (95% CI, 2.29-2.98) by the exact method and 2.67 (95% CI, 2.04-3.49) by the random effects model suggesting a high prevalence of HBV carrier state in lymphoma. There was evidence of statistical heterogeneity which disappeared after exclusion of retrospective studies on sensitivity analysis. Conclusions: The results of this study suggest a possible causal relation between HBV infection and NHL which needs to be confirmed by experimental and epidemiological studies. In countries where prevalence of HBV infection is 1% or more, it may be prudent to screen patients with NHL for occult HBV infection.