Challenging case of chronic Q fever endocarditis: usefulness of 18F-FDG PET/CT in the diagnosis and follow-up

Abstract

Diagnosis of infective endocarditis can be challenging for clinicians, especially when involving prosthetic valves. Recent data suggest that ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) could be a useful diagnostic tool in this setting. Here, we report a case of a patient with an aortic biological prosthesis who presented with a history of fever and fatigue. Echocardiograms were negative for vegetations. The ¹⁸F-FDG PET/CT revealed an infective process of the valve and serological tests were positive for chronic Coxiella burnetii infection. Specific treatment for chronic Q fever endocarditis was, therefore, started and the response was monitored using ¹⁸F-FDG PET/CT. This case highlights the challenges and pitfalls clinicians face when confronted with prosthetic valve endocarditis and the use of ¹⁸F-FDG PET/CT for diagnosis and follow-up.

Background

Infective endocarditis (IE) is a life-threatening disease. Despite the presence of modified Duke criteria (mDC) for clinical diagnosis and the improvement of the image quality of ultrasound diagnostics, IE diagnosis remains challenging.¹ Echocardiography is the imaging technique of choice: it can be either transthoracic (TTE) or transoesophageal (TOE), with a sensitivity of 70% and 96%, respectively.¹ Diagnosis of prosthetic valve endocarditis (PVE) is even more difficult because of the lower sensitivity of conventional imaging techniques (50% for TTE and 92% for TOE).^{1 2} The latest European Society of Cardiology (ESC) guidelines suggest the use of ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) for increasing the diagnostic accuracy in this setting.¹ Some authors propose the use of ¹⁸F-FDG PET/CT at the initial presentation of patients with suspected PVE to increase the diagnostic capability of the mDC.³ Here, we report a case of Coxiella burnetii PVE where PET/CT, coupled with serological tests, permitted guiding the correct diagnosis.

Case presentation

A 19-year-old man with an aortic biological prosthesis presented to the Emergency Department of Bambino Gesù Children’s Hospital in Rome with persistent fever and fatigue. The patient had been diagnosed at birth with an aortic bicuspid valve which progressively developed insufficiency; he, therefore, underwent valvular substitution with a biological prothesis at 13 years of age. He lived in a rural area and had contacts with farm animals. He had a history of recurrent fever over the previous 10 months and he had been experiencing daily sub-febrile temperatures for 1 month. On physical examination, he appeared unwell, but no signs of heart failure were noticed. He had painful papular lesions on the fingertips suggestive for Osler’s nodes. Blood cell count revealed mild anaemia and thrombocytopaenia with normal leukocytes; C reactive protein (CRP) was slightly elevated (3 mg/dL).

Thus, he was admitted for further diagnostic workup.

Investigations

With a clinical suspicion of IE, blood cultures were obtained and both TTE and TOE were performed, showing a focal thickening of the aortic valve without evidence of vegetations. Serological tests for Bartonella, Brucella and Mycoplasma were negative. A molecular panel for bacteria and fungi resulted negative on blood. One set of blood cultures resulted positive for Streptococcus parasanguinis. In accordance with the mDC, a ‘possible’ IE was diagnosed,⁴ and antibiogram-guided intravenous antibiotic therapy with vancomycin (750 mg every 12 hours) and gentamicin (150 mg/day) was commenced.¹ After 5 days of treatment, there was no improvement of the overall clinical state and fever. Thus, based on the susceptibility test of the isolated pathogen, vancomycin was changed to ceftriaxone (2 g/day) with prompt reduction of blood inflammatory markers (CRP 1.5 mg/dL after 2 days and negative after 7 days) and clinical improvement. Intravenous therapy with gentamicin was discontinued after 2 weeks, while ceftriaxone was administered for 3 weeks. The patient was discharged in good clinical condition. Oral antibiotic therapy with linezolid (500 mg two times a day) was continued at home for two more weeks.

Two months later, the patient was admitted again for subfebrile temperature. Since a relapse of the endocarditis was suspected, diagnostic workup was repeated. TTE and TOE did not indicate the presence of vegetations and blood cultures were negative. To further investigate a possible IE, PET/CT was performed. In order to minimise physiological ¹⁸F-FDG myocardial uptake, the patient was asked to observe a very low carbohydrate, high protein and high-fat diet the day before PET imaging and then to fast overnight on the day before imaging.⁵ The tracer uptake in the area of the prosthetic valve on PET/CT (figure 1, I) was compatible with an infective process. Suspecting a culture-negative IE and considering the rural area where the patient lived, serological tests for Coxiella burnetii were performed. Positive phase I Coxiella IgG with a titre of 1: 1024 were found, confirming the diagnosis of chronic Q fever with endocarditis according to guidelines.⁶

Figure 1.

Shows axial images of CT (A), 18F-FDG PET (B), non-attenuation-correction PET (C) and fused PET/CT images (D) performed at baseline (I), after 4 (II), 8 (III) and 16 (IV) months on treatment. In order to minimise physiological 18F-FDG myocardial uptake, the patient was asked to observe a very low carbohydrate, high protein and high-fat diet the day before each PET imaging and then to fast overnight on the day before imaging. At baseline, 18F-FDG PET images show increased radiopharmaceutical uptake around the aortic valve prosthesis, with a focal pattern on the posterior surface (red arrows; I: SUVmax 6.1). 18F-FDG uptake gradually decreases in subsequent PET/CT scans (red arrows; II: SUVmax 3.8; III: SUVmax 3.7) until it becomes not significant in the last imaging (red arrows; IV: SUVmax 2). 18F-FDG PET, 18F-fluorodeoxyglucose positron emission tomography. SUV, standardized uptake value.

Treatment

Treatment with doxycycline (100 mg two times a day) and hydroxychloroquine (200 mg three times a day) was commenced,⁶ with improvement of overall clinical conditions and subsiding of fever after 3 days. Treatment was continued for 18 months with no adverse effects.

Outcome and follow-up

The patient underwent follow-up with Coxiella serology and PET/CT. Coxiella serology was performed every 4 months. PET/CT was repeated after 4, 8 and 16 months of treatment (figure 1, II–IV). After 4 months, there was an increase of phase I IgG to 1:2048, but on the following determination titres dropped again to 1:1024. After 18 months of therapy, IgG phase I antigen titres for Coxiella remained 1:1024 but PET/CT showed a clear reduction of the metabolic uptake on the aortic valve; therefore, treatment was discontinued. The patient remained afebrile and in good clinical conditions after a 1-year follow-up. Echocardiograms remained negative throughout.

Discussion

The diagnosis of IE, especially on prosthetic valve, remains an important challenge for clinicians. The mDC offer guidance,⁴ but are not sufficiently sensitive in all settings, especially for early diagnosis of IE and for PVE.¹ On initial evaluation, the presented patient had four minor criteria (a predisposing heart condition, fever, Osler’s nodes and a single positive blood culture), therefore a ‘possible’ IE was diagnosed.⁴ Despite the TOE being negative, clinical suspicion remained high and antibiotic treatment was commenced. The clinical response and the reduction of inflammatory markers erroneously lead the clinicians to hypothesise that S. parasanguinis was the causative agent of the IE. Interestingly, none of the antibiotics administered are recommended for treating chronic Coxiella burnetii infection, but some in vitro studies have shown a partial bacteriostatic activity of ceftriaxone and linezolid against C. burnetii,^{7 8} which might explain the temporary clinical response.

Echocardiographic findings represent a major criterion for the diagnosis of IE. Both TTE and TOE are less accurate in prosthetic valves than in native valves. The introduction of ¹⁸F-FDG PET/CT among the imaging techniques for endocarditis diagnosis has resulted in a better diagnostic outcome for patients with prosthetic valves.⁹ The latest ESC guidelines suggest that, when endocarditis on prosthetic valve is suspected, abnormal metabolic activity around the site of implantation detected by ¹⁸F-FDG PET/CT should be considered as a major criterion.¹ A study conducted by de Camargo et al has shown that PET/CT has a high accuracy for the diagnosis of IE in cases of PVE (93%) when compared with the clinical strategy based on the mDC (68%). The sensitivity of PET/CT was lower in native valve endocarditis (NVE; 22%); this could be explained by a predominance of polymorphonuclear cells in PVE, while more extensive fibrotic tissues were noted in NVE.³ When employing ¹⁸F-FDG PET/CT, clinicians should be aware of the high physiological accumulation of FDG in the myocardium, which poses a major obstacle in recognising the abnormal FDG uptake. Therefore, an appropriate dietary preparation is needed to suppress myocardial-FDG uptake.^{5 18}F-FDG PET/CT is a high-resolution technique that enables precise localisation of septic emboli in the context of IE.³

C. burnetii is an intracellular bacterium causing a zoonotic infection called ‘Q fever’; it can present as an acute or chronic disease. C. burnetii endocarditis should be suspected in the presence of a culture negative endocarditis, especially in patients with a positive history for acute Q fever, occupational animal exposure or living in rural areas. However, a ‘certain’ diagnosis of Q fever endocarditis can be extremely difficult, because vegetative lesions are rarely visualised by echocardiography. PET/CT can provide useful information to support a clinical suspicion of endocarditis in this setting.¹⁰

According to guidelines, treatment for chronic Q fever endocarditis should be continued for at least 18 months. The treatment course should be further prolonged until phase I IgG decrease by a fourfold, even though guidelines report that a patient who has been treated appropriately for 18 months and has recovered from clinical symptoms, might not benefit from continued treatment despite high phase I IgG.⁶ Chieng et al described a patient with chronic Q fever endocarditis who underwent yearly follow-up with PET/CT. They reported a prolonged treatment of more than 5 years because of persistent high titres of phase I IgG, despite negative uptake on PET/CT. The authors conclude that more data were needed to optimise decision making in this setting.¹¹ Our patient did not achieve the target of serological recovery since phase I IgG remained 1:1024. Yet there is data suggesting that there is a poor correlation between clinical and serological response,¹² therefore, clinicians are in need of other means to guide treatment duration.

Recent data have shown that PET/CT could be used to monitor the response to treatment.¹⁰ In our case, PET/CT was planned every 4 months. Since after 8 months of treatment there was a clear improvement of ¹⁸F-FDG uptake around the valvular prothesis, the following PET/CT was scheduled after eight more months. The last PET/CT had no significant uptake, therefore, treatment was discontinued, despite persistently high titres of phase I Coxiella IgG.

Our case highlights the challenges clinicians face when confronted with PVE. PET/CT is a useful diagnostic tool in the setting of high suspicion when echocardiography is not conclusive. A modification of the DC including increased metabolic activity on the valve found through PET/CT should be considered. Further studies are needed to define ideal timing of PET/CT repetition in order to monitor the patient’s response to treatment.

Learning points.

• Diagnosis of infective endocarditis on prosthetic valve can be difficult since echocardiography has a lower sensitivity than in native valves.

• ¹⁸F-FDG PET/CT has a good accuracy in diagnosing endocarditis on prosthetic valves and should be employed when echocardiograms are negative but clinical suspicion is high.

• Chronic Q fever endocarditis is difficult to diagnose and requires a long course of antibiotic treatment, but the exact duration is difficult to define since serological response is highly variable.

• ¹⁸F-FDG PET/CT can be a useful tool in the diagnosis and follow-up of chronic Q fever endocarditis.

Ethics statements

Patient consent for publication

Obtained.