Acute Coxiella burnetii Infection: A 10-Year Clinical Experience at a Tertiary Care Center in the United States

Abstract Background Identifying and treating patients with acute Q fever who are at an increased risk of progressing to persistent disease is crucial for preventing future complications. In this study, we share our decade-long clinical experience with acute Q fever, highlighting the challenges that clinicians encounter from making an initial diagnosis and performing risk stratification to determining the appropriate prophylaxis regimen and duration. Methods We retrieved records of adult Mayo Clinic patients (≥18 years) with positive Coxiella burnetii serology results between 1 January 2012 and 31 March 2022. Patients with Q fever anti–phase II immunoglobulin G ≥1:256 by indirect immunofluorescence were further analyzed. Results Thirty-one patients were included. Their median age was 58 years (IQR, 50–64), and the majority were men (84%). Acute hepatitis (29%), flu-like illness (25.8%), and pneumonia (16%) were the most common presentations. Thirteen patients (42%) received antibiotic prophylaxis to prevent disease progression, with significant variation in the indications and duration across physicians. The combination of doxycycline and hydroxychloroquine was the preferred regimen. Prophylaxis was administered for a median 333 days (IQR, 168–414). Four patients (13%) progressed to Q fever native valve infective endocarditis, with elevated anticardiolipin immunoglobulin G levels being the sole risk factor in 2 cases. The small sample size precluded drawing conclusions on the impact of prophylaxis in preventing disease progression. Conclusions Management of acute Q fever is complicated by the lack of comprehensive clinical guidelines leading to varied clinical practices. There is a critical need for randomized trials to establish robust evidence-based protocols for management.

Acute Q fever, a zoonotic disease caused by Coxiella burnetii, poses significant diagnostic challenges for physicians.Symptoms are present in half of individuals acutely infected [1,2] and are often nonspecific and self-limited.Progression to focal persistent disease occurs in <5% of cases following the primary acute infection [3,4].The 10-year average annual incidence of acute Q fever in the United States is an estimated 0.36 cases per million persons, almost doubling the annual incidence from 0.3 cases in 2008 to 0.5 cases in 2017 [5].While acute Q fever diagnosis remains rare overall, infections appear to be more prevalent in the West North Central states [5,6], representing the "Agricultural Heartland" of the United States, likely due to a higher probability of contact with animals and animal products [3,6].According to the Centers for Disease Control and Prevention (CDC), the highest incidence of Q fever infections in 2019 was reported in South Dakota (12.43 cases per million persons), followed by Iowa (5.71 cases) [7].
Despite its typically self-limiting nature, accurately identifying patients at increased risk of progressing to focal persistent disease is essential for establishing close clinical follow-up, guiding therapeutic strategies, and preventing future complications.However, this is complicated by the ambiguities surrounding the choice of diagnostic tests for patient risk stratification and the determination of specific host conditions that constitute a high risk for disease progression.The 2013 CDC guidelines [3] provide a framework for clinicians in the United States but fall short in offering detailed recommendations for risk factor assessment, leaving much to the clinical judgment of health care providers.Furthermore, the guidelines lack specific recommendations regarding the choice and duration of treatment for patients at high risk.This absence of high-quality data derived from randomized clinical trials can lead to variability in clinical practice and may affect patient outcomes.
In this study, we share our decade-long clinical experience with acute Q fever, highlighting the myriad challenges that clinicians face, from making an initial diagnosis and performing risk stratification to determining the appropriate prophylaxis regimen and duration.

Case Definition
We retrieved records of all Mayo Clinic patients seen at the Minnesota, Arizona, and Florida campuses with a positive C burnetii serologic test result reported between 1 January 2012 and 31 March 2022.
Patients aged ≥18 years who had a Q fever anti-phase II immunoglobulin G (IgG) ≥1:256 by indirect immunofluorescence were further analyzed [8].Patients with an anti-phase II IgG <1:256 were included only if they had a 4-fold rise in titers at follow-up within 3 to 6 weeks.The 2009 case definition and classification developed by the Council of State and Territorial Epidemiologists and adopted by the CDC were utilized to categorize patients into confirmed and probable cases of acute Q fever, as well as confirmed cases of focal persistent Q fever [3,9].Patients without clinical evidence of infection were excluded.
Throughout this study, "focal persistent disease" describes patients with an identifiable focus of C burnetii organ involvement (eg, osteoarticular, endovascular, or endocardial infection) in the setting of elevated phase I IgG titers (≥1:1024) and/or other consistent laboratory criteria [3,10,11].The traditional serologic definition of "chronic Q fever" describes the serologic pattern of phase I IgG ≥1:1024 irrespective of evidence of a persistent focus.

Laboratory Techniques
All patients were tested by the C burnetii immunofluorescence assay (Focus Diagnostics) to determine immunoglobulin M (IgM) and IgG antibody titers against phase I and II antigens.Briefly, patient serum undergoes 2-fold serial dilution starting at 1:16 and is overlayed onto inactivated C burnetii cells immobilized on a glass slide.Following incubation, slides are washed and incubated with fluorescein-labeled anti-human IgG or IgM conjugate.Following a final wash step, slides are examined for fluorescence.The highest serum dilution that reveals definite fluorescence is reported as the end point titer.
Molecular detection of C burnetii in tissue and serum was performed by real-time polymerase chain reaction (PCR), targeting the unique shikimate dehydrogenase gene (aroE) sequence specific to C burnetii, as previously described [12,13].

Study Definitions
In the absence of a universally accepted consensus on what constitutes "high risk" for disease progression, we used criteria that align with the current medical literature (Table 1).
These included the presence of valvopathy, vascular grafts, vascular aneurysms, pregnancy, elevated aCL IgG antibody level, and/or immunosuppression.For the purposes of this study, patients older than 40 years who lacked other high-risk factors were not considered to be at high risk for disease progression.
Other study definitions are outlined in the supplementary appendix.

Data Collection and Statistical Analysis
Data were collected pertaining to patient demographics, medical comorbidities, clinical presentation, laboratory evaluations, radiographic imaging, treatment, and outcomes.Patient records were reviewed up to the date of their last evaluation by the infectious diseases team to document their outcomes.Patient characteristics were summarized with frequency (percentage) for categorical variables and median (IQR) for continuous variables.All analyses were conducted with R version 4.2.2 (R Foundation for Statistical Computing).

Laboratory and Imaging Findings
At the time of initial evaluation, the median phase II IgG and IgM titers were 1:2048 (range, 1:16-1:32 768) and 1:512 (range, 1:16-1:3200), respectively (Table 3).Four patients had concomitant elevated phase I IgG (≥1:1024) despite the absence of clinical and radiologic evidence of focal persistent disease (supplementary appendix).Serum PCR testing for C burnetii was obtained in 17 patients at the time of presentation to clinical care, all of which returned negative.Plasma microbial cell-free DNA next-generation sequencing (Karius Test; Karius Inc) was obtained in 5 patients at the time of presentation, out of which 3 returned positive for C burnetii despite a negative result from serum C burnetii PCR.aCL IgG titers were measured in only 8 patients on initial presentation (26%), with 6 (75%) showing elevated levels.The median aCL IgG titer was 96.1 GPL (IQR, 68.0-125.3;reference range, ≤15 GPL).Concomitant positive serology results for Bartonella spp, Anaplasma phagocytophilum, Ehrlichia spp, and Borrelia burgdorferi were noted in few cases (Supplementary Table 2).
Overall, 25 patients (81%) underwent echocardiography at the time of initial presentation, including 17 transthoracic echocardiograms (TTEs) and 8 transesophageal echocardiograms (TEEs).Following initial TTE, TEE was performed in 3 patients for further characterization of valvular abnormalities (Table 3).Abnormal valvular findings posing a high risk for progression to Q fever endocarditis were noted in 5 of 17 (29%) TTEs and 6 of 11 (54%) TEEs as detailed in Supplementary Table 3.
Ten patients (32%) had a PET-CT scan (positron emission tomography-computed tomography) performed on initial presentation (median, 5.5 days following admission; Supplementary Table 3), out of which only 1 had phase I IgG >1:1024.The PET-CT results changed management in 1 patient only: following identification of deep 18F-fluorodeoxyglucoseavid lymphadenopathy above and below the diaphragm associated with splenic involvement, a lymph node biopsy was performed, which did not show evidence of lymphoma.

Clinical Management and Outcomes
The treatment regimen, criteria for antibiotic prophylaxis, and total duration of therapy for acute Q fever varied widely in our cohort.Doxycycline monotherapy was used in 21 patients (70%) at the time of initial diagnosis (Table 4).For those not requiring extended prophylaxis, the median duration of doxycycline administration was 14 days (IQR, [14][15][16][17][18][19][20][21]. None of the women in the cohort were pregnant.Fourteen patients were identified as being at high risk of progression to focal persistent Q fever per the study criteria, primarily due to the presence of valvulopathy (n = 9), elevated aCL IgG antibody levels (n = 6), and/or immunosuppression (n = 3).Of these, 10 received prolonged antibiotic prophylaxis (Figure 2).Three patients received prophylaxis because they were deemed to be at high risk per the treating physician, despite the absence of risk factors defined in this study (Supplementary Table 4).Generally, patients diagnosed with valvopathy were prescribed 12 months of prophylaxis.An increase in phase I IgG to ≥1:512 often prompted further investigations to exclude progression to focal persistent disease.Prophylaxis was not extended in patients who had serologic progression (phase I IgG ≥1:512) in the absence of symptoms or radiographic evidence of focal disease.For patients whose sole risk factor for progression was an elevated aCL IgG, prophylaxis was continued until their titers returned to normal levels (≤15 GPL).The most frequently adopted prophylactic regimen was the combination of doxycycline with hydroxychloroquine (n = 9), followed by doxycycline monotherapy (n = 3) and doxycycline with rifampin (n = 1).The median duration of prophylaxis was 333 days (IQR, 168-414; Supplementary Table 4).
Excluding 5 patients (16.1%) who were lost to follow-up after their initial diagnosis, the median duration of patient monitoring was 206 days (IQR, 63-517).These patients had a median 2 additional Q fever serologies beyond the initial diagnostic test (IQR, 2-4).At the time of the last assessment, 22 patients (71%) had clinical resolution of the infection while 4 (13%) progressed to focal persistent disease, with all developing native valve Q fever infective endocarditis (Table 4).Among those who progressed, 2 had elevated aCL IgG antibody titers as the only identifiable risk factor.The remaining 2 patients did not have aCL IgG antibody levels checked and had no other conventional risk factors for disease progression identified.At 1-year follow-up, 11 patients (30%) demonstrated serologic progression to chronic Q fever (phase I IgG ≥1:1024), 7 of whom did not have clinical evidence of focal persistent disease.Serologic progression in these 7 patients occurred after a median 179 days (IQR, 63-358) following initial diagnosis (Table 4).Prophylactic antibiotic administration was associated with decreased probability of serologic progression in our cohort (Figure 3).

DISCUSSION
We identified 31 patients with acute Q fever over 10 years at our institution.The majority of the cases were diagnosed in the summer season and were identified in patients residing in the Midwest and Arizona, respectively, likely representing the catchment area of our institution.Over the 10-year period of our cohort study, 4 patients (13%) progressed to focal persistent disease, specifically manifesting as native valve infective endocarditis.The decision to administer antibiotic prophylaxis for preventing progression to persistent disease varied significantly and was largely dependent on the treating physician's judgment.A combination of doxycycline and hydroxychloroquine was most commonly used for prophylaxis.Notably, elevated aCL IgG levels were the only identified risk factor for progression to Q fever endocarditis in 2 patients, despite both having negative TTEs.Due to the limited sample size, no conclusions could be drawn regarding the impact of antibiotic prophylaxis on reducing the risk of disease progression despite an apparent reduction in the probability of serologic progression.Some of our findings are consistent with prior reports in the literature describing patients with acute Q fever infection.Namely, the male:female ratio for cases reported between 2012 and 2022 was 5.2:1, with the majority having an identifiable exposure history to animal products.While differential occupational exposures to C burnetii could explain the higher prevalence of infection in men, some evidence suggests that sex hormones play a role in the predisposition of men to this pathogen [10,23].Moreover, most cases were diagnosed during summer.This may be related to changes in patterns of cattle husbandry, grazing, and birthing and to increased exposure of patients to C burnetii during the farming season [7,10,24].Consistent with published reports [6,10,11,25,26], acute hepatitis, flu-like illness, and pneumonia were the most common clinical syndromes in our cohort.These presentations, however, are nonspecific and could be misleading, potentially resulting in underdiagnosis and significant underestimation of the incidence of acute Q fever in hospitalized patients.Clinicians therefore should consider C burnetii infection in these clinical settings, particularly for patients with exposure risk factors (Table 2), regardless of residing in regions with a higher prevalence of reported cases [5][6][7].Clinicians should also be aware of the temporal delay between the onset of acute Q fever symptoms and seroconversion, the possible coinfection with Borrelia spp or  Anaplasma spp in endemic areas, the serologic cross-reactivity or coinfection with Bartonella spp [27,28] or E chaffeensis [29,30], and the low sensitivity of plasma C burnetii PCR during this window period [31].Microbial cell-free DNA next-generation sequencing has emerged as a promising tool that could enable the diagnosis of acute cases prior to seroconversion even when the PCR test result is negative [27].Yet, this test is costly and not widely accessible, and it requires further evaluation before it can reasonably be integrated into routine clinical practice for the evaluation of fever of unknown origin and atypical syndromes, such as those caused by acute C burnetii infection [27].
With the growing use of PET-CT imaging, lymphadenitis is increasingly being recognized as a manifestation of acute and persistent Q fever infection.It is now identified as a prelymphomatous stage, especially in cases of persistent disease [11,38,39], although progression to lymphoma has been noted in patients with acute Q fever lymphadenitis, albeit less frequently [38].While this observation may not justify PET-CT scans for all patients with acute Q fever, high aCL IgG and palpable superficial lymphadenopathy or identification of deep lymphadenopathy on computed tomography imaging should prompt further investigations, as well as close clinical and serologic monitoring.
Melenotte et al recently introduced criteria for diagnosing cases of possible and definite acute Q fever endocarditis [34].Recognizing and treating these patients is important, as current evidence suggests that they are at increased risk (up to 6-fold) for progression into persistent endocarditis [10,11,34,36].In our cohort, only 1 patient was diagnosed with definite acute Q fever endocarditis.However, this condition may have been underdiagnosed, as aCL IgG antibody levels were measured in only 8 patients (26%), which may have resulted in fewer TEEs being performed, especially in cases where TTE findings were negative or inconclusive [34].
Treatment of patients with symptomatic acute Q fever infection, ideally initiated within the first 3 days of symptom onset, has been shown to reduce the duration of illness, risk for hospitalization, and severe complications [3,10,40].In patients who are intolerant to doxycycline, acceptable alternatives include fluoroquinolones, minocycline, clarithromycin, and trimethoprimsulfamethoxazole [3,41,42].Yet, the regimen, duration, and clinical impact of antibiotic prophylaxis for patients with risk factors for disease progression are less well established.First and foremost, there is no consensus on defining "high risk" factors, leading to considerable variability across studies.Risk stratification often involves a combination of host factors and laboratory parameters, including aCL IgG positivity (Table 1).Second, the ideal prophylactic regimen and duration remain undefined [43], and the efficacy, safety, tolerability, and risk-benefit ratio of any suggested prophylaxis have yet to be evaluated in a randomized clinical trial.In a study by Million et al [16], a 12-month course of doxycycline and hydroxychloroquine was suggested in high-risk patients with acute Q fever to decrease the risk of Q fever endocarditis, noting that doxycycline monotherapy does not offer protection against endocarditis.Meanwhile, CDC guidelines lack specific recommendations for antibiotic prophylaxis in high-risk cases but advocate for regular clinical assessments and serologic monitoring to facilitate early disease progression management [3].Health care providers are therefore tasked with navigating the existing literature to conduct individualized risk stratification and discuss the risk-benefit of antibiotic prophylaxis with their patients.
Eight patients (26.7%) in our study were administered a combination of doxycycline and hydroxychloroquine at the time of diagnosis, which was continued as prophylaxis due to their categorization by the treating physician as being at increased risk for progression to focal persistent disease, while for 1 patient, hydroxychloroquine was added to doxycycline on day 14.Other prophylactic regimens included doxycycline monotherapy (n = 3) and a combination of doxycycline with rifampin (n = 1).It is worth pointing out that 1 patient with elevated aCL IgG levels (57.1 GPL) progressed to Q fever endocarditis while undergoing doxycycline monotherapy prophylaxis, whereas another patient (89.4 GPL) did not progress after 12 months with the same regimen.Both patients had no abnormalities on TTE.However, neither underwent TEE to evaluate for potential occult valvular changes that might not have been initially detected.
Three patients developed adverse events while receiving antibiotic prophylaxis, requiring interruption of therapy.These included gastrointestinal intolerance in a patient taking rifampin and doxycycline, severe phototoxicity due to doxycycline despite precautionary measures, and retinal toxicity in a patient prescribed hydroxychloroquine.The small sample size and     study design preclude drawing definitive conclusions about the efficacy of antibiotic prophylaxis in preventing progression to focal persistent disease.This study has several limitations.First, due to the retrospective design, the quality and availability of data are limited by the inherent accuracy of medical records.In addition, variables such as exposure history, which rely on self-reporting, are prone to reporting bias.Our team implemented all feasible measures to ensure data accuracy and reduce the risk of information bias.Second, the study is not powered enough to detect treatment effects due to the small sample size.Furthermore, this was a single-center study, and the generalizability of our findings may be limited, as they might not reflect the full spectrum of management and treatment outcomes across the United States.Additionally, the follow-up period was short for some patients, limiting further assessment of disease progression.Finally, our results and conclusions are partly contingent on the high-risk criteria used in this study.

CONCLUSIONS AND FUTURE DIRECTIONS
The diagnosis of acute Q fever presents significant challenges, compounded by the array of clinical presentations and extensive body of literature, as well as the lack of clear, comprehensive guidelines that govern the management of this disease.This has resulted in a variety of diagnostic and therapeutic approaches in clinical practice, often leaving clinicians with more questions than answers.Funding and conducting randomized clinical trials is crucial to address several key areas: • Validating and standardizing the factors considered high risk for progression to focal persistent disease • Standardizing screening protocols for the identification of high-risk patients • Establishing the optimal antibiotic regimen and duration and determining the subgroup of high-risk patients who may benefit from chemoprophylaxis (this has been emphasized for Q fever endovascular infections [44]) • Development of standardized evidence-based guidance for serologic follow-up tailored to the most common C burnetii strains responsible for human disease in the United States, highlighting key factors that can be used by clinicians to differentiate clinical vs serologic progression.
Concentrating on these key points may facilitate the development of updated and comprehensive clinical guidelines that offer more definitive guidance for clinicians.This can lead to reduced variability in clinical practice and enhanced outcomes for patients with acute Q fever infection.

Figure 1 .
Figure 1.Seasonal distribution of acute Q fever diagnoses.
Abbreviation: IgG, immunoglobulin G. a Defined as a patient with acute Q fever who, at any time during the evaluation, had a detectable phase I IgG titer ≥1:1024 but without clinical evidence of progression to focal persistent disease on subsequent follow-up.

Figure 2 .
Figure 2. Treatment outcomes and disease progression in patients with acute Q fever based on risk factors and antibiotic prophylaxis.

Figure 3 .
Figure 3. Kaplan-Meier curve of serologic progression in patients with acute Q fever stratified by antibiotic prophylaxis.IgG, immunoglobulin G.

Table 1 . Common Risk Factors for Progression to Focal Persistent Disease Reported in the Literature
This table does not represent an exhaustive list of studies.Its purpose is to highlight various factors deemed "high risk" for disease progression and the rationale for choosing specific risk factors in our study.
Abbreviations: IgG, immunoglobulin G; IgM, immunoglobulin M. a Elevated anticardiolipin IgG is associated with several complications: acute Q fever endocarditis, hemophagocytic syndrome, hepatitis, cholecystitis, thrombosis, meningitis, and persistent endocarditis.Acute Q fever endocarditis itself is a risk factor for persistent endocarditis.Hemophagocytic syndrome is a risk factor for lymphoma.b Older patients are more likely to have valvulopathies.c Renal insufficiency is associated with vasculopathy.