Polymicrobial Q Fever and Enterococcal Aortic Prosthetic Valve Endocarditis with Aortic Root Abscess

Introduction

Polymicrobial endocarditis is uncommon, reported in 1% of cases of infective endocarditis (Murdoch et al. 2009). It occurs primarily and with increasing frequency in intravenous drug users, predominately on the tricuspid valve. Other risk factors include cardiac structural abnormalities, prosthetic heart valves, and central venous catheters (Sousa et al. 2012). Involved organisms previously reported include Staphylococcus aureus, Streptococcus pneumoniae, Pseudomonas aeruginosa, Candida spp., and others (Valencia Ortega et al. 1997, Mehrzad et al. 2013) with a mortality rate of over 30% (Saravolatz et al. 1978).

Case

We report a case of polymicrobial endocarditis caused by Enterococcus faecalis and Coxiella burnetii in a 64-year-old male with a prosthetic aortic valve. The patient had congenital aortic stenosis, with valvuloplasty in 1966, and aortic valve replacement (AVR) to a mechanical valve in 1989. Other diseases included hypertension, hyperlipidemia, benign prostatic hypertrophy, and iron deficiency anemia. He lived in a city in central Israel and worked as a pastry cook. He was exposed to peacocks raised by his next door neighbor and did not travel outside Israel in the last few years before admission. The patient denied using illicit drugs.

He was admitted to the hospital because of daily fever of up to 39–40°C degrees for about 1 month, weight loss of 6 kg, and night sweats. Physical examination on his admission was described as normal. Blood tests demonstrated mild anemia and thrombocytopenia, similar to previous exams. Renal function and liver enzymes were normal. Six blood culture bottles taken on his admission were positive for E. faecalis, which was ampicillin sensitive (minimum inhibitory concentration [MIC] −0.5) and highly gentamicin resistant (MIC >1024). No vegetations were demonstrated on transthoracic echocardiogram (TTE); however, a transesophageal echocardiogram (TEE) demonstrated an aortic root abscess with mild aortic paravalvular leak. Due to the high-level resistance to gentamicin, the patient was treated with ampicillin and ceftriaxone.

He was transferred to our tertiary center for AVR 9 days after the first hospitalization. Blood cultures turned negative 10 days after his first hospitalization. AVR was performed 4 days after his admission to our hospital. During the operation, an aortic root abscess was drained around the left coronary cusp and the valve was replaced. Cultures from the operation grew E. faecalis.

During his hospitalization, a serology for chronic Q fever returned positive (immunoglobulin G [IgG] phase 1 >6400 and IgM positive). This serology was taken before hospitalization by the family physician. PCR from the resected aortic valve was also positive for Q fever. Doxycycline and hydroxychloroquine were added to the treatment regimen.

Under this treatment, the patient improved clinically. He was discharged and returned to normal physical activity, without signs of heart failure. Toaday, approximately 6 months after the operation, he is still taking doxycycline and hydrochloroquine. Repeated Q fever serology has remained IgG phase I >12,800.

Discussion

Periannular abscess develops in ∼20% of patients with aortic valve endocarditis, more commonly associated with prosthetic valve infection (Anguera et al. 2005). In a study of 311 patients with aortic valve endocarditis, 2% had polymicrobial endocarditis. Among periannular abscess cases in this study, S. aureus was the causative organism in 18% and Enterococcus spp. in 12% (Anguera et al. 2005).

Q fever is a bacterial zoonosis distributed worldwide and caused by C. burnetii. In Israel the average annual incidence of Q fever is 0.6 cases per 100,000 persons, and several outbreaks have been reported in the past, including an urban outbreak (Amitai et al. 2010). Endocarditis is the most severe and potentially fatal form of chronic Q fever, with an overall 10-year mortality of 27% (Million et al. 2010).

Enterococcal infective endocarditis (IE) has increased significantly over the last decade and recently was reported to represent ∼20% of endocarditis cases (Slipczuk et al. 2013). Rates of polymicrobial endocarditis involving Enterococcus spp. are unknown. In a prospective series reviewing 500 enterococcal endocarditis cases, no polymicrobial infections were described (Chirouze et al. 2013). However, cases of combined infections with S. aureus and Streptococcus bovis are documented (Bishara et al. 1999, Ferreiros et al. 2006, Allocca et al. 2007).

There are only two case reports of concomitant enterococcal and Q fever endocarditis in the literature. One describes a case of mitral prosthetic valve endocarditis in a patient from Spain. This patient was treated with daptomycin and ampicillin for E. faecalis, and doxycycline plus hydroxychloroquine for Q fever, along with valve replacement surgery. The patient survived (Mora-Rillo et al. 2013). The other case describes a hemodialysis patient with diseased mitral and tricuspid valves, treated with tricuspid replacement, vancomycin, and gentamicin for E. faecalis and doxycycline plus hydrochloroquine for Q fever. The patient was cured (Rovery et al. 2009).

Three other cases of polymicrobial endocarditis involving C. burnetii were reported from The Netherlands, two of them co-infected with Streptococcus viridans and one with S. aureus (Kampschreur et al. 2011). Another case of dual C. burnetii–Staphylococcus aureus endocarditis was reported in a patient with congenital heart disease from Switzerland (Kaech et al. 2008). In a prospective cohort including 26 definite C. burnetii endocarditis cases from France (Raoult et al. 2005), two cases were co-infections, one with Streptococcus viridans and the other with Streptococcus gallolyticus. Another case of co-infection with S. gallolyticus was also reported from France (Rovery et al. 2009).

Overall, including our case, 10 cases of dual Q fever endocarditis have been reported in the literature. In five of them, no valvular disease was reported. In most of these cases, the diagnosis of Q fever was made incidentally, either by serology or PCR taken as part of a routine investigation of endocarditis or fever of unknown origin (before the diagnosis of endocarditis). Serology was confirmative for chronic Q fever in all cases (Table 1).

Q fever endocarditis is an indolent disease with nonspecific symptoms. The diagnosis of dual Q fever endocarditis might be missed when another pathogen compatible with endocarditis is isolated in blood or valve cultures. This is troubling, considering the grave prognosis of untreated chronic Q fever described above. For these reasons, it was previously suggested that either C. burnetii serologic analysis or PCR testing of heart valve tissue be performed in patients with infective endocarditis who have a pre-existing valvulopathy and live in an endemic region, regardless of whether another etiological agent has been identified or not (Raoult et al. 2005). Wegdam-Blans et al. described screening strategy after a large outbreak in The Netherlands. They identified 763 patients at high risk for chronic disease that were tested for Q fever, and found 52 patients (7%) with a positive serology, among them 10 patients (19%) that were diagnosed with chronic Q fever. Therefore, they suggest a targeted screening after an outbreak. (Wegdam-Blans et al. 2013).

Considering five of 10 patients without previous valvulopathy reported, further study should evaluate whether serology for Q fever should be considered in any patient with endocarditis living in an endemic area.