Polymicrobial Infections Among Patients with Vascular Q Fever, France, 2004–2020

We report 5 cases of vascular Q fever complicated by polymicrobial superinfection in patients who had no risk factors for acute Q fever. Q fever was diagnosed by serologic and molecular assays for Coxiella burnetii. We confirmed additional infections using conventional graft cultures.

We report 5 cases of vascular Q fever complicated by polymicrobial superinfection in patients who had no risk factors for acute Q fever. Q fever was diagnosed by serologic and molecular assays for Coxiella burnetii. We confi rmed additional infections using conventional graft cultures.
the French National Reference Center for species confirmation (Tables 1, 2). Four patients tested positive by PCR on vascular or graft samples, whereas patient 5 tested positive by PCR on a vertebral biopsy (Table 1). We conducted PCR on serum samples from 2 patients; the samples tested negative for C. burnetii. None of the patients had risk factors for acute Q fever, such as contact with animals, consumption of raw milk, or tick bites. Three of the patients lived in the countryside of the Nouvelle-Aquitaine region. We isolated 2-10 additional microorganisms using conventional graft cultures, identifying concurrent bacteremia in 3 patients (patients 2, 3, and 4) ( Table 1). All isolated microorganisms were common commensals of the oral or gut microflora. Only patient 3 had a fungal co-infection (Candida albicans).
Patients 3 and 5 had been treated for previous episodes of vascular Q fever; their infections relapsed after the end of treatment. Patient 3 had been treated with hydroxychloroquine and doxycycline for 2 years for a Q fever aortic graft infection. The infected graft was not removed, and the infection relapsed 2 months after the end of treatment. Patient 5 had Q fever aortitis and spondylodiscitis 4 years before this episode. He had been treated with hydroxychloroquine and doxycycline for 18 months and had received an aortic graft implantation. However, the infection in the aortic graft relapsed 2 years after the end of treatment.
In total, 3 patients died of vascular graft and endograft infections; another died of a different cause. Patients 1, 3, and 4 were treated with hydroxychloroquine and doxycycline. Only patients 1 and 3 completed the 18-month therapy; the other patients died before or during treatment (Table 1). All patients also had a 6-week course of antimicrobial therapy for the other identified microorganisms.

Conclusions
We identified 5 cases of vascular Q fever complicated by polymicrobial superinfection in patients with no documented risk factors for acute Q fever. Q fever was diagnosed by serologic and molecular assays. We isolated additional microorganisms from clinical samples from 5 of 16 patients with vascular fever, suggesting that co-infections might be more common than previously thought. Furthermore, 4 of the cases complicated by polymicrobial superinfection were diagnosed during the past 5 years, suggesting that this condition might be emergent.
All patients had an aortoduodenal fistula or intimate contact between the aortic graft and the duodenum. The role of C. burnetii in vascular fistulas is well-described, especially in aortoenteric fistulas but also in aortobronchial, aortocaval, and arteriocutaneous fistulas (2,(11)(12)(13). Death rates among patients with chronic Q fever complicated by arterial fistula are higher than among those without fistulas (2). Aortoenteric fistulas arise from infection and inflammation of the aortic wall or the perigraft tissues created by C. burnetii infection, which erodes the adjacent digestive tract. Diagnostic delays might contribute to fistula development; therefore, earlier detection of vascular Q fever might reduce the incidence of these complications.
We found that 31% of patients with vascular Q fever in this study also had an aortoenteric fistula, a concurrent condition that might have contributed to superinfection. This rate is higher than that suggested by previously published studies on vascular Q fever and aortoenteric fistulas (2). We might have found a higher rate because we have conducted systematic Q fever screening in every patient with vascular infection since 2019.

1962
Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 27, No. 7, July 2021 In conclusion, we report a small case series of vascular Q fever complicated by polymicrobial superinfection. Our findings support systematic screening for C. burnetii in patients with vascular infections, especially when an arterial fistula is suspected or confirmed. We believe these screenings should be conducted even when more common microorganisms are isolated by culture. The screening should not be limited to patients with risk factors for acute Q fever. The patients in this study had low C. burnetti phase I IgG titers (none >6,400) and all had vascular Q fever confirmed by molecular diagnosis. Low phase I IgG titers have been described in acute Q fever endocarditis (10,14), suggesting that some of our patients might have had acute rather than chronic vascular Q fever. Therefore, physicians should conduct PCR selective for C. burnetii on vascular grafts or arterial biopsies when patients with a vascular infection have a phase I IgG titer >100. However, this low cutoff might impair specificity and positive predictive value (15) and should be further investigated. We highlight that vascular Q fever requires a specific and prolonged therapy, including surgery, to prevent relapse and other complications. We emphasize the need for systematic C. burnetii screening in patients with vascular infections, even when cultures test positive for other microbes.