Smoking and Hodgkin Lymphoma: A Meta-Analysis

INTRODUCTION: The etiology of Hodgkin lymphoma (HL) is largely unknown. However, certain associations have been noted, such as familial factors and infection with viruses. Smoking has been associated with the development of multiple malignancies and some studies have reported an association between HL and smoking but the relationship of tobacco use with lymphomas in largely unclear.

OBJECTIVE: To investigate the potential relationship between tobacco use and the development of HL using a meta-analysis methodology of retrospective, case-control studies.

METHODS: An extensive search was conducted using Pubmed/MEDLINE through July 2008. Case-control studies that reported odds ratio (OR) and 95% confidence intervals (CI) or allow for those values to be calculated were included in our analysis. Case reports, editorials, letters to the editor, review articles and prospective studies were excluded. The smoking status was then subdivided in three groups: never smokers, former smokers and current smokers. Meta-analyses were performed comparing the risks of former and current smokers against the risk of never smokers of developing HL. Fixed and random effects models were used to assess for heterogeneity.

RESULTS: Seven case-controls studies, accounting for 3201 cases and 15268 controls were included in the analysis. Most of the articles reported OR adjusted for age, sex and educational level. In former smokers, the OR was 0.73 (95% CI, 0.65 – 0.82) when compared to never smokers; no heterogeneity was detected. The current smoker group had an OR of 1.70 (95% CI, 1.36 – 2.13) when compared to the never smoker group; some heterogeneity was detected in this group (p < 0.003, I² = 69.6%).

CONCLUSIONS: Despite the heterogeneity observed in the analysis, the current smoker group seems to have a 70% increased risk of developing HL. Although a cause-effect linkage between tobacco use and HL is difficult to prove, further basic and translational research is necessary to clarify the potential etiological role of smoking in HL.