Pregnancy outcomes of Q fever: prospective follow-up study on Reunion island

Background Q fever has been associated with perinatal complications. We conducted a prospective follow-up study to assess both the incidence of adverse pregnancy outcomes (APOs) associated with Coxiella burnetii infection and the contribution of Q fever to APOs. Methods Between May 1 and October 31, 2013, within the regional perinatal health care centre of Saint Pierre, Reunion island, we investigated unexplained miscarriages, stillbirths, preterm births or small-for-gestational age children. Seropositivity for C. burnetii antibodies was defined using indirect immunofluorescence for a phase 2 IgG titre ≥1:64. Acute Q fever was defined for a high phase 2 IgG titre ≥1:256 (compatible with recent or active infection) or the detection of C. burnetii genome in miscarriage products and placentas. Incidence rate ratios (IRR) for Q fever related APOs (taken as a composite outcome or individually) were assessed using Poisson regression models for dichotomous outcomes controlling major confounders. Results Over a 6-month period, 179 pregnant women suspected or diagnosed with an APO were investigated for Q fever, of whom 118 met the definition for an APO. Of these, 19 were seropositive and 10 presented a profile indicative of an acute infection. For three women with an acute Q fever, the chronology between the onset of infection, the APO (2 miscarriages, 1 preterm birth) and the seroconversion suggested causality in the pathogenesis. The cumulative incidence of Q fever related APOs was estimated between 2.2‰ and 5.2‰, whether causality was required or not. Both C. burnetii exposure and acute Q fever were independently associated with APOs (IRR 1.55, 95% CI 1.31–1.84; IRR 1.47, 95% CI 1.15–1.89, respectively). Conclusions In the endemic context of Reunion island, acute Q fever may lead to APOs. To limit the burden of Q fever on reproduction, pregnant women should be kept away from farms and avoid direct contact with ruminants.