In order to describe the clinical features and the epidemiologic findings of 1,383 patients hospitalized in France for acute or chronic Q fever, we conducted a retrospective analysis based on 74,702 sera tested in our diagnostic center, National Reference Center and World Health Organization Collaborative Center for Rickettsial Diseases. The physicians in charge of all patients with evidence of acute Q fever (seroconversion and/or presence of IgM) or chronic Q fever (prolonged disease and/or IgG antibody titer to phase I of Coxiella burnetii > or = 800) were asked to complete a questionnaire, which was computerized. A total of 1,070 cases of acute Q fever was recorded. Males were more frequently diagnosed, and most cases were identified in the spring. Cases were observed more frequently in patients between the ages of 30 and 69 years. We classified patients according to the different clinical forms of acute Q fever, hepatitis (40%), pneumonia and hepatitis (20%), pneumonia (17%), isolated fever (17%), meningoencephalitis (1%), myocarditis (1%), pericarditis (1%), and meningitis (0.7%). We showed for the first time, to our knowledge, that different clinical forms of acute Q fever are associated with significantly different patient status. Hepatitis occurred in younger patients, pneumonia in older and more immunocompromised patients, and isolated fever was more common in female patients. Risk factors were not specifically associated with a clinical form except meningoencephalitis and contact with animals. The prognosis was usually good except for those with myocarditis or meningoencephalitis as 13 patients died who were significantly older than others. For chronic Q fever, antibody titers to C. burnetii phase I above 800 and IgA above 50 were predictive in 94% of cases. Among 313 patients with chronic Q fever, 259 had endocarditis, mainly patients with previous valvulopathy; 25 had an infection of vascular aneurysm or prosthesis. Patients with endocarditis or vascular infection were more frequently immunocompromised and older than those with acute Q fever. Fifteen women were infected during pregnancy; they were significantly more exposed to animals and gave birth to only 5 babies, only 2 with a normal birth weight. More rare manifestations observed were chronic hepatitis (8 cases), osteoarticular infection (7 cases), and chronic pericarditis (3 cases). Nineteen patients were observed who experienced first a documented acute infection, then, due to underlying conditions, a chronic infection. To our knowledge, we report the largest series of Q fever to date. Our results indicate that Q fever is a protean disease, grossly underestimated, with some of the clinical manifestations being only recently reported, such as Q fever during pregnancy, chronic vascular infection, osteomyelitis, pericarditis, and myocarditis. Our data confirm that chronic Q fever is mainly determined by host factors and demonstrate for the first time that host factors may also play a role in the clinical expression of acute Q fever.
