Exposure to Coxiella burnetii and risk of non-Hodgkin lymphoma: a retrospective population-based analysis in the Netherlands. Issue 5 (May 2018)

• Record Type: Journal Article
• Authors: van Roeden, Sonja E; van Houwelingen, Fedor; Donkers, Chiel M J; Hogewoning, Sander J; de Lange, Marit M A; van der Hoek, Wim; Kampschreur, Linda M; Bonten, Marc J M; Hoepelman, Andy I M; Bleeker-Rovers, Chantal P; Wever, Peter C; Oosterheert, Jan Jelrik
• Is Part Of: Lancet. Volume 5:Issue 5(2018)
• Page Start: e211
• Page End: e219
• Publication Date: 2018-05
• Subjects: Hematology -- Periodicals; Blood -- Diseases -- Periodicals; 616.15005
• DOI: 10.1016/S2352-3026(18)30038-3
• Languages: English
• ISSNs: 2352-3026

Abstract

Summary: Background: An association between Coxiella burnetii and non-Hodgkin lymphoma has been suggested. After a large Q fever epidemic in the Netherlands (2007–10), we postulated that the incidence of non-Hodgkin lymphoma would be increased during and after the epidemic in areas with a high endemicity of Q fever compared with those with low endemicity. Methods: We did a retrospective population-based analysis and calculated relative risks (RRs) of non-Hodgkin lymphoma during 1-year periods before, during, and after the Q fever epidemic, for areas with intermediate and high endemicity of Q fever compared with low endemic areas. We also calculated the RR of non-Hodgkin lymphoma in people with chronic Q fever compared with the general population. Findings: Between Jan 1, 2002, and Dec 31, 2013, 48 760 cases of non-Hodgkin lymphoma were diagnosed. The incidence of non-Hodgkin lymphoma ranged from 21·4 per 100 000 per year in 2002 to 26·7 per 100 000 per year in 2010. A significant association with non-Hodgkin lymphoma was noted in 2009 for areas with a high endemicity of Q fever compared with low endemic areas (RR 1·16, 95% CI 1·02–1·33; p=0·029); no further associations were noted in any other year or for areas with intermediate Q fever endemicity. Among 439 individuals with chronic Q fever, five developed non-Hodgkin lymphoma, yielding a crude absolute risk of 301·0 cases per 100 000 per year (RR 4·99, 95% CI 2·07–11·98; p=0·0003) compared with the general population in the Netherlands. Interpretation: These findings do not support the hypothesis that Q fever has a relevant causal role in the development of non-Hodgkin lymphoma. Several limitations, inherent to the design of this study, might lead to both underestimation and overestimation of the studied association. Funding: Foundation Q-support and Institut Mérieux.