Remarkable spatial variation in the seroprevalence of Coxiella burnetii after a large Q fever epidemic

Roan Pijnacker, Johan Reimerink, Lidwien A. M. Smit, Arianne B. van Gageldonk-Lafeber, Jan-Paul Zock, Floor Borlée, Joris Yzermans, Dick J. J. Heederik, Catharina B. M. Maassen, Wim van der Hoek

Prior to the 2007-2010 Q fever epidemic in the Netherlands, the seroprevalence of antibodies against Coxiella burnetii in the general population was 1.5%, which is low compared to other countries. We aimed to determine the seroprevalence after the Q fever epidemic among people living in the affected area, compare the seroprevalence with the incidence of Q fever notifications during the 2007-2010 Q fever epidemic, and to identify farm exposures associated with having antibodies against C. burnetii. During the period March 2014-February 2015, residents aged 18-70 years from two provinces were invited by general practitioners to complete a questionnaire on their symptoms and personal characteristics and to submit a blood sample. We used the mandatory provincial database of livestock licences to calculate distance to farms/farm animals for each participant. To compare ELISA-positive participants for C. burnetii antibodies with those who were negative, we calculated prevalence ratios (PR) using binominal regression. We compared the C. burnetii seroprevalence in the period March 2014-February 2015 with the incidence of Q fever notifications during the 2007-2010 Q fever epidemic at municipal level by calculating the Spearman correlation coefficient. Of the 2296 participants (response rate: 34%), 6.1% (n = 139, 95% CI 5.1-7.1%) had C. burnetii antibodies (range in municipalities: 1.7-14.1%). C. burnetii seroprevalence was higher in individuals living within 1000 m of goat farms (PR 3.0; 95% CI 1.4-6.4) or within 1000 m of > 50 goats (PR 1.9; 95% CI 1.2-3.0). Seroprevalence increased with decreasing distance to the closest goat farm that was infected during the epidemic years (< 500 m, PR 9.5, 95% CI 2.8-32; 500-1000 m, PR 4.5, 95% CI 2.6-7.7; 1000-1500 m, PR 2.2, 95% CI 1.1-4.3, 1500-2000 m, PR 1.2, 95% CI 0.6-2.5; > 2000 reference group). There was no significant correlation between C. burnetii seroprevalence and Q fever incidence during the 2007-2010 epidemic (r s  = 0.42, p = 0.156). Results showed a remarkable spatial variation in C. burnetii seroprevalence in a relatively small livestock dense area. It confirms previous evidence that the Q fever epidemic was primarily the result of airborne C. burnetii transmission from Q fever affected goat farms.