From 2007 to 2009, the Netherlands experienced a major Q fever epidemic, with higher hospitalization rates than the 2–5% reported in the literature for acute Q fever pneumonia and hepatitis. We describe epidemiological and clinical features of hospitalized acute Q fever patients and compared patients presenting with Q fever pneumonia with patients admitted for other forms of community-acquired pneumonia (CAP). We also examined whether proximity to infected ruminant farms was a risk factor for hospitalization.
A retrospective cohort study was conducted for all patients diagnosed and hospitalized with acute Q fever between 2007 and 2009 in one general hospital situated in the high incidence area in the south of the Netherlands. Pneumonia severity scores (PSI and CURB-65) of acute Q fever pneumonia patients (defined as infiltrate on a chest x-ray) were compared with data from CAP patients. Hepatitis was defined as a >twofold the reference value for alanine aminotransferase and for bilirubin.
Among the 183 hospitalized acute Q fever patients, 86.0% had pneumonia. Elevated liver enzymes (alanine aminotransferase) were found in 32.3% of patients, although hepatitis was not observed in any of them. The most frequent clinical signs upon presentation were fever, cough and dyspnoea. The median duration of admission was five days. Acute Q fever pneumonia patients were younger, had less co-morbidity, and lower PSI and CURB-65 scores than other CAP patients. Anecdotal information from attending physicians suggests that some patients were admitted because of severe subjective dyspnoea, which was not included in the scoring systems. Proximity to an infected ruminant farm was not associated with hospitalization.
Citation: Wielders CCH, Wuister AMH, de Visser VL, de Jager-Leclercq MG, Groot CAR, Dijkstra F, et al. (2014) Characteristics of Hospitalized Acute Q Fever Patients during a Large Epidemic, The Netherlands. PLoS ONE 9(3): e91764. https://doi.org/10.1371/journal.pone.0091764
Editor: Ulrike Gertrud Munderloh, University of Minnesota, United States of America
Received: December 3, 2013; Accepted: February 13, 2014; Published: March 10, 2014
Copyright: © 2014 Wielders et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Netherlands Organisation for Health Research and Development (ZonMw)(http://www.zonmw.nl/), grant number 205520006. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Q fever is a zoonosis caused by the intracellular bacterium Coxiella burnetii, which can be transmitted from animals (mainly goats and sheep) to humans . The incubation period ranges from two to three weeks, and the clinical presentation of acute Q fever is diverse: from flu-like illness to pneumonia or hepatitis , , . Geographical differences are reported in clinical presentation. Patients in the Netherlands ,  and the Basque region of Spain  experience more pneumonia, while data from France , ,  and southern Spain  show more cases of hepatitis. Clinical features of acute Q fever include: fever, headache, fatigue, cough, night sweating, dyspnoea, chest pain, nausea, vomiting, diarrhoea, and joint and muscle pain , , , . A C. burnetii infection can also remain asymptomatic. Both symptomatic and asymptomatic C. burnetii infections can evolve into chronic Q fever . Chronic Q fever is difficult to diagnose and involves serological testing as well as a thorough clinical evaluation .
From 2007 to 2009, the Netherlands experienced a major acute Q fever epidemic with over 3,500 notified cases, which mainly occurred in the southern part of the country . Q fever-infected dairy goat farms were established as the source of the epidemic in the Netherlands and living close to an infected farm was identified as a strong risk factor for acquiring acute Q fever . The percentage of notified acute Q fever patients being hospitalized was 46.1% in 2007, decreasing to 21.1% in 2008 and 19.5% in 2009, with an overall average admission rate in these three years of 21.6% . Previous studies from France reported a hospitalization rate of only 2–5% for acute Q fever pneumonia and hepatitis , . There are no explanations for the high admission rate in the Netherlands, though it has been suggested that a high dose of C. burnetii bacteria or prolonged continuous exposure to the bacterium might lead to a more severe clinical presentation , , .
The objectives of this study were: (1) to describe the clinical and epidemiological features of patients hospitalized for acute Q fever in the Netherlands in the period 2007–2009; (2) to compare the clinical presentation of hospital-admitted patients with acute Q fever pneumonia with patients with other forms of community-acquired pneumonia (CAP); (3) to compare the proximity to an infected ruminant farm as a measure of infection dose between hospitalized and non-hospitalized patients.
The medical ethical committee of Bernhoven Hospital (‘Comissie Ethiek’) approved this study. Patient information was anonymized and de-identified prior to analysis. According to Dutch legislation, written consent from each individual patient was not required because of the retrospective nature of this study and the use of anonymized information.
Study design and setting
A retrospective cohort study was conducted with patients hospitalized for acute Q fever from 2007 to 2009 in Bernhoven Hospital (locations in Oss and Veghel, Noord-Brabant, currently located in Uden), which is located at the centre of the Dutch Q fever epidemic area. Data were collected from the clinical patient files and the laboratory information system.
All patients living in the hospital's catchment area who were hospitalized with acute Q fever at Bernhoven Hospital from January 2007 to December 2009 were included in the study. Acute Q fever-related hospitalization was defined as hospital admission of one or more nights at Bernhoven Hospital, either within one month before or after microbiological diagnosis of acute Q fever, or within one month before or after the onset of symptoms retrospectively attributed to acute Q fever based on microbiological diagnostics, which were carried out after the patient was discharged. Relevant medical staff were consulted when it was uncertain whether the patient was hospitalized for the acute Q fever infection or for another concurrent medical condition. When patients were admitted for Q fever-related reasons on two or more occasions within the defined period, clinical information was collected for the first admission only.
Exclusion criteria were (1) non-Q fever-related hospitalizations of patients who had an acute Q fever infection; (2) presence of antibodies against C. burnetii found during hospitalization but no IgG phase II antibodies found during follow-up; and (3) presence of proven chronic Q fever at the time of hospitalization, defined as a positive polymerase chain reaction (PCR) result in combination with an IgG phase I titre≥1∶1,024 measured by immunofluorescence assay (IFA; Focus Diagnostics, Inc., Cypress, CA, USA).
Microbiological diagnosis of acute Q fever and other infections
For the microbiological diagnosis of acute Q fever we used four categories, based on PCR results as described previously  and on serology (Table 1). Standard microbiological assays including culture, serology, PCR and urinary antigen tests were performed for detection of other respiratory pathogens (e.g., Streptococcus pneumoniae, Mycoplasma pneumoniae, Legionella pneumophila). All microbiological tests for Bernhoven Hospital were conducted by the Department of Medical Microbiology and Infection Control of the Jeroen Bosch Hospital in ’s-Hertogenbosch.
Clinical data and follow-up
Patient demographic characteristics, clinical symptoms, medical history, laboratory and radiological findings, additional microbiological tests, and treatment details were retrospectively collected from the clinical patient files and the laboratory information system (January–June 2013). Data on clinical features and laboratory findings was at first contact at the time of admission and also included information on medical history. Data on follow-up, such as mortality, development of chronic Q fever, and other long-term effects of the disease, were collected for two years after the date of admission.
Additionally, the CURB-65 score (Confusion, blood Urea nitrogen, Respiratory rate, Blood pressure, age≥65)  and the Pneumonia Severity Index (PSI)  were determined. Unavailable features of the scores (e.g., respiratory rate or pH) were counted as being normal. The CURB-65 score ranges from 0 to 5, while the PSI is divided into classes I to V. CURB-65 score≥2 and PSI≥IV were defined as severe pneumonia requiring hospitalization. Hepatitis due to acute Q fever was defined as a >twofold increase in the reference value for alanine aminotransferase (ALT; >90 U/L for males, >70 U/L for females) in combination with a >twofold increase in bilirubin (>34 µmol/L). Relative bradycardia was defined as body temperature >38.9°C and heart rate <120 beats/min without the use of beta-blocker medication, pacemaker-induced rhythms, or arrhythmias . Adequate antibiotic treatment was defined as doxycycline (200 mg/day), moxifloxacin (400 mg/day) or ciprofloxacin (1,000 mg/day per oral dose) . The definition of chronic Q fever was made according to the classification of the Dutch Q fever Consensus Group, which uses an IgG phase I titre≥1∶1,024 measured IFA as cut-off .
An analysis was performed to compare the clinical presentation and pneumonia severity of acute Q fever pneumonia patients (defined as an infiltrate observed on a chest x-ray) with non-Q fever-related pneumonia patients, by using data from a study performed from November 2007 to January 2010 in the neighbouring Jeroen Bosch Hospital, ’s-Hertogenbosch (which is 19 km from the Bernhoven Hospital) . This control group consisted of patients attending the emergency department with CAP, defined as an acute symptomatic infection of the lower respiratory tract which developed outside the hospital or nursing home, whereby a new infiltrate is demonstrated on a chest x-ray . Patients diagnosed with acute Q fever and patients that were not hospitalized after visiting the emergency department were excluded from this database. A separate analysis was performed for pneumonia patients with microbiologically proven bacterial aetiology.
The second analysis was performed to study the role of the extent of environmental exposure, by measuring the distance in metres (m) between the nearest infected farm and the patients' home address (postal code). Both hospitalized and non-hospitalized acute Q fever patients diagnosed between 2007 and 2009 and living in the catchment area of Bernhoven Hospital were included. Distances were categorised according to Schimmer et al. : 0–2,000 m, >2,000–5,000 m, and >5,000 m. Small ruminant farms that experienced abortion waves (defined as >5% abortions of all pregnant animals) caused by C. burnetii and bulk tank milk positive farms were considered infected (data provided by the Animal Health Service and the Food and Consumer Product Safety Authority).
Descriptive characteristics, medical history, symptoms, radiologic findings, laboratory tests results, treatment, and follow-up were investigated by calculating relative frequencies, and median and interquartile ranges (IQRs). Prevalence of underlying disease and health status of admitted patients were compared with prevalence data of the general population in the same region (Municipal Health Service, Hart voor Brabant), and of the entire country based on data from Statistics Netherlands, the National Public Health Compass, and the National Cancer Registry , , .
For the analysis of pneumonia and other CAP patients, relative frequencies, and median and IQRs were calculated. Chi-square tests, Chi-square tests for trend, and Mann-Whitney U tests were used to test for differences between acute Q fever pneumonia and either CAP patients or other bacterial pneumonia patients. In these tests, a p-value <0.05 was considered statistically significant. Odds ratios (ORs) and 95% confidence intervals (95% CI) were also calculated. For the environmental exposure analysis, median distances for hospitalized and non-hospitalized patients from their home address to the closest infected small ruminant farm were calculated and the Mann-Whitney U test was used to test for statistical significance. Logistic regression was performed for the analysis of the environmental exposure with the distance category >5,000 m used as a reference. Data were analysed using IBM SPSS Statistics version 19.0.0 (SPSS Inc.).
Characteristics of hospitalized patients
From 2007 to 2009, 1,728 Q fever patients living in Bernhoven Hospital's catchment area were diagnosed with Q fever. Within one month before or after diagnosis and/or onset of symptoms, 208 of these 1,728 (12.0%) acute Q fever patients were hospitalized. These rates in the hospital's catchment area differed largely during the epidemic years: 40/79 (50.6%) in 2007, 81/684 (11.8%) in 2008, and 87/965 (9.0%) in 2009. Twenty-five patients were excluded: 22 because of a non Q fever-related reason for hospitalization despite having acute Q fever, two because of undetectable IgG phase II antibodies during follow-up, and one because of hospitalization with a proven chronic Q fever infection, leading to a hospital admission rate of 10.6% (183/1,728). The 183 hospitalized acute Q fever patients comprised of 114 males (62.3%) with a median age of 54 years at admission (range 4–86; two patients <18 years) and a median length of admission of 5 days (range 1–53) (Table 2). Hospitalized patients were significantly older than the non-hospitalized patients living in Bernhoven Hospital's catchment area (55 vs. 48 years; p<0.001), and no significant difference was found for the gender distribution. All patients except one were admitted following a visit or referral by the general practitioner (GP) to the emergency department.
Microbiological diagnosis of acute Q fever and other infections
Table 1 presents the number of patients that fulfilled each of the categories of our case definition. Additional microbiological tests and their results are presented in Table 3. A sputum sample was obtained in 42 (23.0%) of patients. Nine patients had no leucocytes (21.4%), 22 had few leucocytes (52.4%), a moderate number of leucocytes was found in eight (19.0%) patients, and only three (7.1%) patients had numerous leucocytes in their sputum sample. One patient had pneumonia infection caused by Coxiella burnetii and Legionella pneumophila. Detectable Mycoplasma pneumoniae antibodies (≥1∶40) were found in 22 patients (17.9% of patients in whom the specific test was performed), although only two patients met our local definition of a M. pneumoniae infection (antibody titre≥1∶320).
Clinical data and follow-up
Fever was the most reported symptom (80.9%) among hospitalized acute Q fever patients, followed by cough (50.8%) and dyspnoea (47.5%) (Table 4). Pneumonia with unilateral or bilateral infiltration on a chest x-ray was observed in 154/179 (86.0%) patients with an x-ray available (Table 5). Moderate to severe pneumonia based on a CURB-65 score≥2 was seen in 34 (22.1%) acute Q fever pneumonia patients, while PSI≥IV was found in 27 (17.5%). A total of 16 (10.4%) patients had high PSI and CURB-65 scores, and combining both scores resulted in 45 (29.2%) patients with a moderate to severe pneumonia infection. Anecdotal information from attending physicians suggests that some patients were admitted because of severe subjective dyspnoea. Adequate antibiotic treatment was started during or after hospitalization in 84.7% of the patients (Table 5). Relative bradycardia was observed in 35/36 patients with a body temperature >38.9°C in the absence of the use of beta-blocker medication, pacemaker-induced rhythms, or arrhythmias.
Almost all patients had elevated C-reactive protein (CRP) levels (>10 mg/L); median 182 mg/L (range 0–672; IQR 107–247). Upon admission, trombocytopenia and trombocytosis were present in 3.3% and 7.7% of cases, respectively, but no consistent pattern in resolution of these abnormalities was observed. Liver chemistry tests at admission were available for 167/183 (91.3%) patients, and abnormal values (i.e., ALT; males >45 U/L, females>35 U/L) were seen in 32.3% (Table 5). Bilirubin was elevated in 25.6% of the admitted patients. None of the patients met our definition of hepatitis. No liver biopsies were performed. Hyponatraemia (<135 mmol/L) was found in 50.9%; the majority had mild hyponatraemia (77/89; 86.5%), and severe hyponatraemia was not observed. A lumbar puncture was performed in five patients: three were suspected of meningitis (none confirmed), and two of Guillain-Barré syndrome (both confirmed). Ten (5.5%) patients needed to be re-hospitalized within six weeks after they had been initially discharged.
Heart failure (11.5%) and chronic obstructive pulmonary disease (COPD) (n = 19; 10.4%) were the most common co-morbidities in hospitalized acute Q fever cases (Table 2). Pregnancy or infection with HIV/AIDS was not observed. One underlying condition was reported for 38 (20.8%) patients, two or more co-morbidities for 37 (20.2%). Eighty-two patients (47.1%) were smokers, which is a higher percentage than observed in the general Dutch population (26.3%) .
Eleven (6.0%) patients, all with underlying illness, died within two years after hospital admission due to different causes, of whom two (1.1%) died in the intensive care unit during their hospital admission (one with disseminated malignancy and one with Staphylococcus aureus sepsis) (Table 5). In one deceased patient, the proven chronic Q fever infection might have contributed to the death, the other ten had other causes of death. The median number of weeks between the date of admission and death was 23 weeks (range 1–97) and the median age at death was 76 years (range 63–86). Deceased patients were significantly older than patients still alive two years after hospitalization (p<0.001).
One or more sequelae following acute Q fever were reported for 81 of 153 (52.9%) patients with available information. The most common complaints were fatigue (n = 55; 67.9% of patients with sequelae), dyspnoea (n = 11; 13.6%), and decreased physical fitness (n = 7; 8.6%). Other complaints recorded for no more than four patients were: fever, chronic cough, neurological sequelae, headaches, thoracic pain, depression, and sweating. Seventeen patients (9.2%) developed chronic Q fever during follow-up (four proven, two probable, and eleven possible chronic Q fever cases - Table 4).
In a study on CAP patients conducted by van Gageldonk-Lafeber et al. , 339 CAP patients were included. We excluded 48 patients because of they had C. burnetii pneumonia and another 37 patients who were not hospitalized. The features of the remaining 254 hospitalized CAP patients, the bacterial aetiology pneumonia subgroup (n = 104), and the 154 hospitalized Q fever patients with an infiltrate on their chest x-ray are listed in Table 6. Hospitalized Q fever-related pneumonia patients were significantly younger than hospitalized CAP patients (median age at admission 56 vs. 68 years, respectively, p<0.001). The median PSI score was significantly lower (p<0.001) in hospitalized acute Q fever patients (median 58, range 0–142) than for the overall group of CAP patients (median 88, range 21–184) and the subgroup of hospitalized CAP patients diagnosed with another bacterial infection (median 86, range 27–184). The odds of being hospitalized with mild pneumonia (PSI I–III) was significantly higher for Q fever patients than for other CAP patients (OR 3.69, 95% CI 2.26–6.04), and also compared to CAP patients diagnosed with another bacterial infection (OR 3.22, 95% CI 1.79–5.81). When using the CURB-65 score, similar results are found: mild pneumonia (CURB-65 0–1) in the hospitalized Q fever group compared with CAP patients (OR 4.99, 95% CI 3.14–7.94) and CAP patients diagnosed with another bacterial infection (OR 5.49, 95% CI 3.12–9.67).
The median distance from the home address of hospitalized acute Q fever patients to the closest infected small ruminant farm was 4,330 m (IQR 1,990–5,375 m), while non-hospitalized acute Q fever patients (n = 1,545) lived at a median distance of 3,555 m (IQR 1,841–5,237 m; p = 0.169). Logistic regression showed no significant increase in risk for hospitalization for people living closer (0–2,000 m or >2,000–5,000 m) to an infected goat or dairy sheep farm than people living further away (i.e., ≥5,000 m).
Remarkable differences in clinical presentation were observed between hospitalized C. burnetii pneumonia patients and patients hospitalized for pneumonia with another aetiology. In general, hospitalized acute Q fever patients were younger with lower pneumonia severity scores and less co-morbidity, and the duration of their admission was shorter. This might be explained by the severe subjective symptoms of pneumonia, mainly dyspnoea, witnessed by treating physicians in hospitalized Q fever patients (personal communications). As subjective dyspnoea is not included in the pneumonia severity scoring systems, this may be a reason for hospital admission, despite the low severity score. Also the discrepancy between low pneumonia severity scores and abnormalities on the chest x-ray has been associated with hospitalization . While in the Netherlands, the large majority of CAP patients (97%) are treated by their GPs as outpatients , all but one of the hospitalized acute Q fever patients were admitted through the hospital emergency department. This suggests that self-referral or referral to the emergency department by the GP of otherwise relatively healthy patients might also be related to the high hospitalization rate.
In 2007, the epidemic was confined mainly to Bernhoven Hospital's catchment area and we presume the high admission rate to be explained in part by unfamiliarity with acute Q fever at the beginning of the epidemic, and retrospective active case finding among hospitalized pneumonia patients. In 2008 and 2009, when the epidemic was spreading, admission rates in Bernhoven Hospital were lower than reported nationwide for notified cases , although still higher than the 2–5% found in France , . However, from other reported outbreaks elsewhere, large variations in hospitalization rates have been reported, for example 8% and 25% in Germany , , 37% in Serbia , and 60% in the UK . It has been suggested that a high dose of C. burnetii bacteria or a prolonged continuous exposure to it might lead to a more severe clinical presentation , , , which was not confirmed by our analysis of proximity to an infected farm. Higher awareness and familiarity with acute Q fever in Bernhoven Hospital and adequate public health information may play a role in Dutch regional hospitalization rate differences, which is currently under investigation. In addition, later in the epidemic diagnostic tests for Q fever were done more frequently and more rapidly requested , , while diagnostic facilities improved through the introduction of PCR .
We found moderately elevated liver chemistry tests (ALT or bilirubin) in hospitalized acute Q fever patients, but no cases of hepatitis, although liver biopsies were not performed. Nevertheless, elevated liver enzymes are a common feature of acute Q fever infections , , . The case definition of hepatitis differs between studies . In addition, the majority of the patients with elevated liver chemistry tests had pneumonia. In contrast to our findings, hepatitis seems to be the major clinical presentation in France , ,  and southern Spain ; while in the Spanish Basque region pneumonia predominates . The reasons for this geographical variation may include differences in the route of the infection (aerosol or ingestion) , , host factors , differences in C. burnetii strains , and the infectious dose level , .
Furthermore, hyponatraemia was more common in our study population (50.9%) than described previously for CAP and acute Q fever patients (27.9% and 28.2%, respectively) , . Hyponatraemia is the most common electrolyte imbalance observed in hospitalized patients, and has been associated with prolonged hospital stay ,  and increased in hospital mortality in hospitalizations for all causes , and for pneumonia hospitalizations . However, other studies reviewing pneumonia patients reported no difference in hospital mortality , .
While the clinical presentation of hospitalized acute Q fever patients was severe enough to be hospitalized, the median hospital stay of just five days suggests that the duration of symptoms was generally short. Typically, they were relatively young, male, had little co-morbidity, and were smokers, which is in accordance with previous studies , , , , . Low CURB-65 scores in acute Q fever patients have also been reported before . Reported clinical symptoms were mostly in line with the literature , , . Relative bradycardia was also observed, which has been described previously , , , . We observed two cases of Guillain-Barré syndrome – which is a rare, but previously described complication of acute Q fever , , . Additional microbiological testing revealed the presence of antibodies against M. pneumoniae in 22 patients, although only two met our definition, which required an antibody titre≥1∶320. Overlapping positive M. pneumoniae and C. burnetii serologic results have been described before . Adequate antibiotic treatment was started during or after hospitalization in 84.7% of patients, which is higher than what was found nationwide among patients treated by their GP or hospital physician in 2007 or 2008, when 60.3% and 72.2% received adequate treatment, respectively . Presumably, this reflects higher awareness of, and familiarity with, acute Q fever in Bernhoven Hospital. An acute Q fever-related mortality rate of 1.1% was found, which is in accordance with previous findings , .
The diagnosis of chronic Q fever relies on high IgG phase I titres in combination with a clinical evaluation and detection of C. burnetii DNA . Seventeen (9.2%) patients developed a serological profile indicative of chronic Q fever, though the majority only had high IgG phase I titres (≥1∶1,024) and no risk factors for a chronic Q fever infection, indicating a possible chronic Q fever infection according to the Dutch Q fever Consensus Group . Only four (2.2%) patients had a proven chronic Q fever infection which is similar to overall conversion rates reported in literature (0–5%), though comparison with data from literature is hampered by the use of different case definitions .
Though exposure dose is established as an individual risk factor for infection with C. burnetii , distance to the nearest positive Q fever ruminant farm as a measure of exposure dose was not associated with hospitalization. This corresponds with studies that found a high prevalence of antibodies against C. burnetii among the occupationally exposed, such as goat and sheep farmers and veterinarians , , with relatively low levels of clinical illness suggesting that the antibodies are protective against symptomatic Q fever .
The strength of this study is that we were able to describe a large cohort of consecutively admitted patients in one hospital in one specific epidemic. Most of the collected data were complete and follow-up data were available for the majority of patients.
A limitation of this study is the absence of a pre-defined control group. In the database of the non-hospitalized C. burnetii infected patients, information on most clinical characteristics were absent, as the patients were treated by their GP. In order to compare the clinical characteristics, we used CAP patients from a historical study group . Yet, all these patients were diagnosed within the same region with the same protocols for diagnostic procedures and treatment, and within the same time period. Secondly, not all items of the pneumonia severity scores were available in the medical records. This means that the observed scores sometimes actually might have been higher. However, the low CURB-65 scores we found in acute Q fever patients have been reported before .
In conclusion, hospitalized Dutch acute Q fever patients mostly presented with fever and pneumonia, while hepatitis was not observed. Patients with acute Q fever pneumonia were hospitalized despite young age, low PSI and CURB-65 scores, and limited co-morbidity, presumably because of severe subjective symptoms in disease presentation. Proximity to an infected ruminant farm, reflecting level of exposure to C. burnetii, does not influence admission rates.
We would like to thank Jan Jelrik Oosterheert (University Medical Centre Utrecht) for his comments on a previous version of this manuscript.
Conceived and designed the experiments: CCHW VLdV MGdJL PMS. Performed the experiments: CCHW AMHW VLdV MGdJL CARG ABvGL JPGvL PCW PMS. Analyzed the data: CCHW AMHW. Contributed reagents/materials/analysis tools: MGdJL CARG FD ABvGL JPGvL PCW WvdH PMS. Wrote the paper: CCHW. Revised the manuscript critically for important intellectual content and approved the final document: CCHW AMHW VLdV MGdJL CARG FD ABvGL JPGvL PCW WvdH PMS.
- 1. Parker NR, Barralet JH, Bell AM (2006) Q fever. Lancet 367: 679–688.
- 2. Maurin M, Raoult D (1999) Q fever. Clin Microbiol Rev 12: 518–553.
- 3. Raoult D, Marrie TJ, Mege JL (2005) Natural history and pathophysiology of Q fever. Lancet Infect Dis 5: 219–226.
- 4. Dijkstra F, van der Hoek W, Wijers N, Schimmer B, Rietveld A, et al. (2012) The 2007–2010 Q fever epidemic in the Netherlands: characteristics of notified acute Q fever patients and the association with dairy goat farming. FEMS Immunol Med Microbiol 64: 3–12.
- 5. van der Hoek W, Morroy G, Renders NHM, Wever PC, Hermans MHA, et al. (2012) Epidemic Q fever in humans in the Netherlands. Adv Exp Med Biol 984: 329–364.
- 6. Sobradillo V, Ansola P, Baranda F, Corral C (1989) Q fever pneumonia: a review of 164 community-acquired cases in the Basque country. Eur Respir J 2: 263–266.
- 7. Raoult D, Tissot-Dupont H, Foucault C, Gouvernet J, Fournier PE, et al. (2000) Q fever 1985–1998. Clinical and epidemiologic features of 1,383 infections. Medicine (Baltimore) 79: 109–123.
- 8. Tissot Dupont H, Raoult D, Brouqui P, Janbon F, Peyramond D, et al. (1992) Epidemiologic features and clinical presentation of acute Q fever in hospitalized patients: 323 French cases. Am J Med 93: 427–434.
- 9. Alarcon A, Villanueva JL, Viciana P, Lopez-Cortes L, Torronteras R, et al. (2003) Q fever: epidemiology, clinical features and prognosis. A study from 1983 to 1999 in the South of Spain. J Infect 47: 110–116.
- 10. Caron F, Meurice JC, Ingrand P, Bourgoin A, Masson P, et al. (1998) Acute Q fever pneumonia: a review of 80 hospitalized patients. Chest 114: 808–813.
- 11. Kampschreur LM, Dekker S, Hagenaars JCJP, Lestrade PJ, Renders NHM, et al. (2012) Identification of risk factors for chronic Q fever, the Netherlands. Emerg Infect Dis 18: 563–570.
- 12. Wegdam-Blans MCA, Kampschreur LM, Delsing CE, Bleeker-Rovers CP, Sprong T, et al. (2012) Chronic Q fever: review of the literature and a proposal of new diagnostic criteria. J Infect 64: 247–259.
- 13. van der Hoek W, Dijkstra F, Schimmer B, Schneeberger PM, Vellema P, et al.. (2010) Q fever in the Netherlands: an update on the epidemiology and control measures. Euro Surveill 15: pii = 19520. Available: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19520.
- 14. Schimmer B, ter Schegget R, Wegdam M, Zuchner L, de Bruin A, et al. (2010) The use of a geographic information system to identify a dairy goat farm as the most likely source of an urban Q-fever outbreak. BMC Infect Dis 10: 69.
- 15. Tigertt WD, Benenson AS, Gochenour WS (1961) Airborne Q fever. Bacteriol Rev 25: 285–293.
- 16. Gilsdorf A, Kroh C, Grimm S, Jensen E, Wagner-Wiening C, et al. (2008) Large Q fever outbreak due to sheep farming near residential areas, Germany, 2005. Epidemiol Infect 136: 1084–1087.
- 17. Oyston PCF, Davies C (2011) Q fever: the neglected biothreat agent. J Med Microbiol 60: 9–21.
- 18. Wielders CCH, Wijnbergen PCA, Renders NHM, Schellekens JJA, Schneeberger PM, et al. (2013) High Coxiella burnetii DNA load in serum during acute Q fever is associated with progression to a serologic profile indicative of chronic Q fever. J Clin Microbiol 51: 3192–3198.
- 19. Lim WS, van der Eerden MM, Laing R, Boersma WG, Karalus N, et al. (2003) Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax 58: 377–382.
- 20. Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, et al. (1997) A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 336: 243–250.
- 21. Cunha BA (2000) The diagnostic significance of relative bradycardia in infectious disease. Clin Microbiol Infect 6: 633–634.
- 22. Dijkstra F, Riphagen-Dalhuisen J, Wijers N, Hak E, van der Sande MAB, et al. (2011) Antibiotic therapy for acute Q fever in The Netherlands in 2007 and 2008 and its relation to hospitalization. Epidemiol Infect 139: 1332–1341.
- 23. van Gageldonk-Lafeber AB, Wever PC, van der Lubben M, de Jager CPC, Meijer A, et al. (2013) The aetiology of community-acquired pneumonia and implications for patient management. Neth J Med 71: 418–425.
- 24. Wiersinga WJ, Bonten MJ, Boersma WG, Jonkers RE, Aleva RM, et al. (2012) SWAB/NVALT (Dutch Working Party on Antibiotic Policy and Dutch Association of Chest Physicians) guidelines on the management of community-acquired pneumonia in adults. Neth J Med 70: 90–101.
- 25. Comprehensive Cancer Centre the Netherlands (2011) Netherlands Cancer Registry. Comprehensive Cancer Centre the Netherlands (IKNL). Available: http://www.cijfersoverkanker.nl/Accessed 2013 July 30.
- 26. Gommer AM, Poos MJJC (2011) Heart failure: prevalence, incidence and mortality by age and sex [in Dutch: Hartfalen: prevalentie, incidentie en sterfte naar leeftijd en geslacht]. In: Compass NPH, editor. Towards better health The Dutch 2010 Public Health Status and Forecasts Report [Volksgezondheid Toekomst Verkenning 2010].Bilthoven: National Institute for Public Health and the Environment (RIVM). Available: http://www.nationaalkompas.nl/gezondheid-en-ziekte/ziekten-en-aandoeningen/hartvaatstelsel/hartfalen/cijfers-hartfalen-prevalentie-incidentie-en-sterfte-uit-de-vtv-2010/. Accessed 2013 July 30.
- 27. Statistics Netherlands (2013) Health, lifestyle and medical contacts; regions, 2008/2011 [in Dutch: Gezondheid, leefstijl en medische contacten; regio's, 2008/2011]. The Hague/Heerlen: Statistics Netherlands (CBS), Statline. Available: http://statline.cbs.nl/statweb/. Accessed 2013 July 30.
- 28. de Wit NC, de Jager CPC, Meekelenkamp JCE, Schoorl M, van Gageldonk-Lafeber AB, et al. (2009) Markers of infection in inpatients and outpatients with acute Q-fever. Clin Chem Lab Med 47: 1407–1409.
- 29. Snijders BEP, van der Hoek W, Stirbu I, van der Sande MAB, van Gageldonk-Lafeber AB (2013) General practitioners' contribution to the management of community acquired pneumonia in the Netherlands: a retrospective analysis of primary care, hospital, and national mortality databases with individual data linkage. Prim Care Respir J 22: 400–405.
- 30. Porten K, Rissland J, Tigges A, Broll S, Hopp W, et al. (2006) A super-spreading ewe infects hundreds with Q fever at a farmers' market in Germany. BMC Infect Dis 6: 147.
- 31. Medić S, Nitzan Kaluski D, Šeguljev Z, Obrenović J, Rudan P, et al.. (2012) Q fever outbreak in the village of Noćaj, Srem County, Vojvodina Province, Serbia, January to February 2012. Euro Surveill 17: pii = 20143. Available: http://www. eurosurveillance.org/ViewArticle.aspx?ArticleId = 20143.
- 32. Smith DL, Ayres JG, Blair I, Burge PS, Carpenter MJ, et al. (1993) A large Q fever outbreak in the West Midlands: clinical aspects. Respir Med 87: 509–516.
- 33. Angelakis E, Raoult D (2010) Q fever. Vet Microbiol 140: 297–309.
- 34. Marrie TJ, Stein A, Janigan D, Raoult D (1996) Route of infection determines the clinical manifestations of acute Q fever. J Infect Dis 173: 484–487.
- 35. Nair V, Niederman MS, Masani N, Fishbane S (2007) Hyponatremia in community-acquired pneumonia. Am J Nephrol 27: 184–190.
- 36. Zilberberg MD, Exuzides A, Spalding J, Foreman A, Jones AG, et al. (2008) Hyponatremia and hospital outcomes among patients with pneumonia: a retrospective cohort study. BMC Pulm Med 8: 16.
- 37. Wald R, Jaber BL, Price LL, Upadhyay A, Madias NE (2010) Impact of hospital-associated hyponatremia on selected outcomes. Arch Intern Med 170: 294–302.
- 38. Waikar SS, Mount DB, Curhan GC (2009) Mortality after hospitalization with mild, moderate, and severe hyponatremia. Am J Med 122: 857–865.
- 39. Ergas D, Keysari A, Edelstein V, Sthoeger ZM (2006) Acute Q fever in Israel: clinical and laboratory study of 100 hospitalized patients. Isr Med Assoc J 8: 337–341.
- 40. Karagiannis I, Schimmer B, van Lier A, Timen A, Schneeberger P, et al. (2009) Investigation of a Q fever outbreak in a rural area of The Netherlands. Epidemiol Infect 137: 1283–1294.
- 41. Marrie TJ (1990) Q fever - a review. Can Vet J 31: 555–563.
- 42. Lai CH, Huang CK, Chin C, Chung HC, Huang WS, et al. (2008) Acute Q fever: an emerging and endemic disease in southern Taiwan. Scand J Infect Dis 40: 105–110.
- 43. Bernard E, Carles M, Laffont C, Durant J, Dellamonica P (1994) Guillain-Barré syndrome associated with acute Q fever. Eur J Clin Microbiol Infect Dis 13: 658–659.
- 44. Bernit E, Pouget J, Janbon F, Dutronc H, Martinez P, et al. (2002) Neurological involvement in acute Q fever: a report of 29 cases and review of the literature. Arch Intern Med 162: 693–700.
- 45. Vardi M, Petersil N, Keysary A, Rzotkiewicz S, Laor A, et al. (2011) Immunological arousal during acute Q fever infection. Eur J Clin Microbiol Infect Dis 30: 1527–1530.
- 46. Fournier PE, Marrie TJ, Raoult D (1998) Diagnosis of Q fever. J Clin Microbiol 36: 1823–1834.
- 47. Kampschreur LM, Wegdam-Blans MCA, Thijsen SFT, Groot CAR, Schneeberger PM, et al. (2010) Acute Q fever related in-hospital mortality in the Netherlands. Neth J Med 68: 408–413.
- 48. Wielders CCH, Morroy G, Wever PC, Coutinho RA, Schneeberger PM, et al. (2013) Strategies for early detection of chronic Q-fever: a systematic review. Eur J Clin Invest 43: 616–639.
- 49. van den Brom R, Schimmer B, Schneeberger PM, Swart WA, van der Hoek W, et al. (2013) Seroepidemiological survey for Coxiella burnetii antibodies and associated risk factors in Dutch livestock veterinarians. PLoS One 8: e54021.