Inpatient Q Fever Frequency Is on the Rise

Background Q fever is a zoonotic bacterial infection caused by Coxiella burnetii that is reportable in the USA. This infection is often asymptomatic; acute infection usually manifests as a self-limited febrile illness, hepatitis, or pneumonia. Chronic infection (usually infective endocarditis) often affects patients with valvulopathy or immunosuppression. Herein, we study the inpatient frequency of Q fever in the United States. Methods We used a nationwide inpatient sample (NIS) for our retrospective cohort study to include hospitalizations with a diagnosis of Q fever between 2010 and 2019. Survey procedures were applied to accommodate for complex sampling design of NIS. Chi-square and least-square means were used for categorical and continuous variables, respectively. Jonckheere–Terpstra test was used to study the trends over the years. SAS 9.4 was used for data mining and analysis. Results A total of 2,685 hospitalizations with a diagnosis of Q fever were included, among which 451 (17%) cases had a concurrent diagnosis of infective endocarditis. The mean age of patients was 58 years, and less than a third was female. Our analysis demonstrated that infective endocarditis was the most common cardiac complication associated with Q fever and was associated with increased inpatient mortality (p value <0.001). There is a trend of an increase in cases of inpatient Q fever with or without endocarditis over the years (p value <0.05). Q fever cases were more common across the Pacific and the South Atlantic divisions. Conclusion Physicians should be aware of an increasing trend of hospitalized patients with Q fever and the significant association with infective endocarditis. Further studies are needed.


Introduction
Coxiella burnetii is an obligate intracellular, Gram-negative bacterium that causes a zoonosis called Q fever, which was frst described in Australia in 1937.Q fever is usually diagnosed by serology [1], and it can present as an acute or chronic infection.Te most common clinical manifestation of acute infection is a nonspecifc febrile illness that might occur in conjunction with pneumonia or hepatitis.Chronic Q fever is rare, occurs in <5% of persons with acute infection [2], and includes endocarditis, chronic hepatitis, chronic vascular infections, osteomyelitis, osteoarthritis, and chronic pulmonary infections.Myocarditis and pericarditis have been reported although rare with this illness.Overall, this zoonotic disease is spread worldwide, and it is rare in the United States.It is classifed as a potential bioterrorism agent by the Centers of Disease Control and Prevention (CDC), and it was designated as a nationally reportable disease in the United States in 1999.In 2008, the Q fever defnition was changed to allow for the reporting of chronic and acute Q fever separately.Te number of Q fever reported to CDC in 2000 was 19 cases.Tis number increased to 212 in 2019 (178 cases of acute Q fever and 34 cases of chronic Q fever) [3,4].Herein, we study the frequency of Q fever in hospital settings in the United States between 2010 and 2019.

Methodology and Material
Te Department of Health and Human Services in the States has established a federal agency for research and quality improvement, the Agency for Healthcare Research and Quality (AHRQ).Te latter has sponsored healthcare cost and the utilization project (HCUP) which includes multiple inpatient and outpatient databases.Tese databases do not include physician ofce visits and pharmacy or laboratory or radiology information.Te project includes collaborative eforts between several states, the federal government, and the health industry.Inpatient databases include state inpatient databases (SID) which contain encounter abstracts of admissions from participating states and form the cornerstone for the National Inpatient Sample (NIS) which is a 20% stratifed sample of all discharges from United States community hospitals that allow researchers to generate national and regional estimates of inpatient visits.For our research purposes, we utilized NIS databases for the years 2010-2019.We did not include the years after 2019 as the admission rate could have been afected by the coronavirus disease of 2019 (COVID-19) pandemic.We utilized the ICD-9 and the ICD-10 codes for Q fever ("0830" and "A78," respectively) to defne the cases for our project.
Tere was a change in ICD codes in 2015.Te reports in the frst three quarters of the year 2015 used ICD-9.Te third quarter was reported using ICD-10 codes (see Supplementary Tables 1-2 for case defnitions).
For NIS data after 2011, hospital ID was renamed to hospital NIS (HOSPID to HOSP_NIS).Also, the hospital division variable (HOSP_DIVISION) was not available before 2012.Instead, we were able to drive the value for hospital division from the hospital state variable (HOSPST) in the hospital NIS data fle.In addition, in the year 2012, there was a change in the name of PL_NCHS2006 to PL_NCHS [5].
We included all encounters that contain a diagnosis of Q fever.Subpopulation analysis (domain analysis) was performed as per HCUP guidelines [6].
Utilizing the "SURVEY" procedure in Statistical Analysis Software (SAS) version 9.4, we obtained national estimates adjusting for complex data design.We compared the basic demographic information including gender, age, race, and socioeconomic status between cases that had a diagnosis of infective endocarditis and cases without this diagnosis, and we did not include other baseline variables that did not reach a statistical signifcance (diabetes and chronic liver disease).Comparison between diferent variables was evaluated using the appropriate test (Chi-square for categorical variables and least-square means for continuous variables).Logistic regression was used to study factors afecting inpatient mortality.Jonckheere-Terpstra test was used to study the trends over the years.Te signifcance of 0.05 was used in all statistical analyses.A geometric map was carried out using R version 4.2.1 (Figure 1). Figure 2 was created using JMP ® Pro 17.0.0.

Results
Tis cohort included 2685 inpatient visits with a diagnosis of Q fever, of which 451 visits had a diagnosis of endocarditis.Te mean age of hospitalized patients with a diagnosis of Q fever in 2010-2019 was around 58 years.Te majority were Caucasian (63%) or Hispanic (15%).Only 29% were female.Compared to patients without endocarditis, cases of Q fever associated with infective endocarditis were older (mean age of 62 years versus 57 years, p value <0.001).Tey had a higher percentage of males (76% versus 70%, p value 0.1).Also, they were more likely to have heart failure, pulmonary hypertension, and valvular heart disease (p value <0.001) in comparison to patients without a concurrent diagnosis of infective endocarditis (Table 1).
Te main fnding of this study is the increase in inpatient frequency of cases with Q fever especially since 2013.Figure 3 shows an upward trend in the number of discharged cases with a diagnosis of Q fever with infective endocarditis (dotted red line) and without discharge diagnosis of endocarditis (solid blue line) across the decade from 2010 to 2019.Te frequency of Q fever per 100,000 hospital discharges each year across the years from 2010 to 2019 has increased as well (see Figure 1).Tis upward trend is primarily driven by an increase in frequency in East North Central, West North Central, South Atlantic, West South Central, and Pacifc divisions (see Figure 2).Figure 2 shows the density of hospital discharges with a Q fever from 2010 to 2019 across each United States division.It indicates that Q fever cases were more frequent across the Pacifc and the South Atlantic divisions.
Table 2 presents patients' outcomes during hospital stay.Most of the cases were admitted to hospitals in large metropolitan areas (more than 46% of the cases).Te length of the stay was on average 12 days.Inpatient mortality overall was high (4%) in comparison to average inpatient mortality of 2% [7].Patients with Q fever and infective endocarditis had higher inpatient mortality (7% versus 3%, p value 0.01).Most cases were discharged home although a higher percentage of people with endocarditis was discharged home with home healthcare or to a facility.

Discussion
In this study, we report an upward trend of inpatient Q fever frequency from 0.4 per 100,000 discharges in 2010 to 1.6 per 100,000 discharges in 2019.Although Q fever frequency and incidence are historically underestimated [8], our fndings correlate with the increased number of annually reported cases to the Centers for Disease Control and Prevention  (CDC) from 135 reported cases in 2010 to 212 cases in 2019 [9].Q fever is a rare but potentially fatal disease.Its occurrence is usually sporadic although outbreak of hundreds of cases has been reported [10].C. burnetii is very resistant to physical stresses, including heat and desiccation, and can survive in the environment for months to years.It is transmitted mainly via aerosols, and the most common animal reservoirs are cattle, sheep, and goats.Hence, selective vaccination is the best strategy for prevention in humans as elimination is difcult [11].
Before 1989, only 234 cases had been reported in the literature [12].Te incidence of Q fever, although still rare, has doubled in the US from 2008 to 2017 [13].Our study shows that inpatient frequency is also trending up between 2010 and 2019.A Mayo Clinic study, conducted from 1980 to 2005, declared the absence of reliable reporting for Q fever endocarditis [14].Cases and outbreaks of Q fever have increased in France from 1985 to 2009 [15].Some patients (5-19%) with Q fever can develop IE [8,[16][17][18].Q fever endocarditis was almost unknown in the United States before 1979 [19].Positive anticardiolipin antibodies (>22 immunoglobulin G-type phospholipid units (GPLU)) were independently associated with acute Q fever endocarditis (odds ratio (OR): 2.7, 95% confdence interval [1.3-5.5], and p value �0.004) [20].Frequency of complications with chronic Q fever is up to 34%, and they are adverse prognosticators [21].Valvulopathy is a risk factor especially aortic and mitral regurgitation [22].Development of arterial fstula has been associated with high mortality rates [23].A phase I IgG antibody titer >1: 800 or a single positive blood culture for C. burnetii serves as a major criterion for infective endocarditis [22].A review of Q fever cases submitted between 1999 and 2015 suggests that phase I and phase II titers comparisons may need to be reexamined as some cases were not classifable by the criteria of the Council for State and Territorial Epidemiologists (CSTE) [22].
In our study, a concurrent diagnosis of unspecifed endocarditis was considered related to Q fever as there is no specifc ICD-9 or ICD-10 code for Q fever-related endocarditis.

Canadian Journal of Infectious Diseases and Medical Microbiology
A small study suggests that acute Q fever-related myocarditis is rare (0.6%-3%) [24,25].In our study of 2,685 discharges that had Q fever, only thirty cases had either a concurrent diagnosis of myocarditis or pericarditis.
Although our study was not designed to study Q fever mortality, we noted that patients with Q fever and endocarditis had higher inpatient mortality compared to cases without concurrent diagnosis of endocarditis (7% vs. 3%, p value 0.01).Patients with acute Q fever should be carefully assessed for the risk of progression to chronic Q fever, and this includes a review of possible immunosuppression, pregnancy, and vascular and valvular heart disease.If they have any of the above risk factors, they should be monitored with a series of serologic and physical exams for at least 2 years [3].
Patients with chronic Q fever should be screened with an echocardiogram (preferably transesophageal) to look for endocarditis especially if they have a preexistent valvular disease.Positron emission tomographic (PET) scanning could be proposed for all patients with suspected persistent focused infection to rapidly diagnose vascular and lymphatic infections associated with death and lymphoma, respectively [26].
A similar national study in Spain revealed that Q fever is associated with high-cost long hospitalization.Also, patients with severe clinical presentations had higher mortality stressing the importance of early suspicion and detection [27].Our study shows that cases with endocarditis had higher average charges ($151,378 vs. $127,307, p value <0.001).
Due to the retrospective design, the study has several limitations.First, there is no specifc ICD code for acuity of Q fever infection or for Q fever complications such as endocarditis.Also, other complications or confounders may not have been captured in the study design.We cannot presume that the increased frequency of Q fever discharge diagnosis correlates with increased prevalence and/or incidence of the disease, and this trend is possibly related to increased awareness and testing for Q fever.Moreover, this trend could be a physician's response to CDC recommendations in 2023 [28].Further studies are needed.Also, the NIS database provides abstracts at the encounter-level rather than patient-level.Tis means that infected patients with multiple admissions could be counted multiple times.Furthermore, the NIS data may not be necessarily an unbiased sample of Q fever patients.Also, it is hard to say if the primary reason for admission was related to Q fever.In addition, coding errors might happen which may afect results.
In conclusion, clinicians and public health experts should pay attention to the possible increasing frequency of Q fever especially across the Pacifc and South Atlantic divisions.Clinicians should screen for endocarditis, especially in patients with valvular/vascular disease and/or immunosuppression.Early detection of Q fever and treatment is associated with better outcomes.

Figure 3 :
Figure 3: Inpatient Q fever trend with and without IE.
LOS: length of stay.SD: standard deviation.