Prosthetic Joint Infection Due to Cardiobacterium Hominis: Report of First Known Case and Review of C. Hominis Septic Arthritis Literature

Introduction

Case Presentation

A 57-year-old man with degenerative joint disease and a left total knee arthroplasty 17 months previously presents to his orthopedic surgeon with a 4–5-day history of the gradual onset of severe left knee pain and redness after performing 5–6 h of yard work the prior day. He went from normal activity one week previously to requiring a walker or wheelchair when he came for the orthopedic visit. He denies injury, tattooing, piercing, dental work, injection drug use or any other risk factor for bacteremia. His past medical history is pertinent for aortic valve replacement in 2003, mitral valve repair in 2006, left total knee arthroplasty in 2018 (17 months prior to this admission) and bovine aortic valve replacement in 2019 (12 months prior to this admission). Examination revealed a swollen, tender and mildly erythematous left knee with palpable joint effusion and a decreased range of motion. The left knee was aspirated and analysis revealed no crystals, negative Gram-stain smear, but a total white blood cell count of 10,771/μL with 95% neutrophils and 5% mononuclear cells. Pain was initially relieved by the aspiration but worsened over the next two days and patient was admitted to hospital; re-aspiration of the knee revealed a white blood cell count of 22,237/μL with 93% neutrophils and 7% mononuclear cells. The remainder of the physical examination was negative, including the absence of heart murmur and absence of any peripheral manifestations of endocarditis. On the second hospital day, the patient underwent operative drainage and debridement of the knee. Surgeons found 30 mL of bloody/purulent material in the knee joint. Patient had extensive irrigation and debridement and polyethylene insert exchange with primary closure. All knee hardware was well-seated and showed no evidence of instability or deep infection. Both blood cultures obtained 20 min apart on the day of admission revealed Cardiobacterium hominis on day 3 of incubation. Identification and susceptibility testing was confirmed by the reference laboratory, ARUP Labs (Associated Regional and University Pathologists, Inc.) in Salt Lake City, UT, USA. Both pre-operative knee aspirates and intra-operative fluid cultures remained sterile. Postoperative transesophageal echocardiography showed a normally seated and functioning bioprosthetic aortic valve and a normal native mitral valve with no evidence of vegetation on any structure. The C. hominis isolate was susceptible to ampicillin, ceftriaxone, ciprofloxacin, levofloxacin and meropenem. The patient was treated with 6.5 months total antimicrobial therapy: 6 weeks of parenteral ceftriaxone followed by 5 months of oral cefdanir. He remains well 6 months after completion of antimicrobial therapy.

Review of the Literature

Google Scholar™ and PubMed®searches were conducted using each combination of Cardiobacterium hominis/HACEK with septic arthritis, pyogenic arthritis and prosthetic joint infection. References from the search were reviewed and the reference list of each article was also reviewed for similar cases. We found only three case reports of Cardiobacterium hominis septic arthritis [13–15]. The patients are summarized in Table 1.

Discussion

This is the first reported case of C. hominis prosthetic knee infection, the second reported case of C. hominis septic arthritis of the knee joint and the fourth reported case of C. hominis joint infection when axial joints are included in the review [13–15]. In all four patients including ours, the definitive microbiological diagnosis was made by the isolation of C. hominis from multiple blood cultures. In only one patient, the patient with lumbar discitis, was a positive culture for C. hominis obtained from the joint itself [15]. This demonstrates the critical need for obtaining blood cultures in any patient with suspected pyogenic arthritis as blood cultures may be the only source for the isolation of the definitive microbiologic pathogen. In infections caused by fastidious bacteria such as Cardiobacterium, it may be very difficult to obtain a positive culture from the joint aspirate or surgical tissue specimen.

In two of the previously reported cases of pyogenic arthritis due to C. hominis, infective endocarditis was also diagnosed [13,15]. In our patient there was a high suspicion for endocarditis but TEE revealed no definitive evidence of endocarditis. Since our patient required 6 weeks of parenteral antibiotics because of the infected prosthetic joint, we believed that this 6-week course of therapy would be more than a sufficient length of treatment even if the bioprosthetic aortic valve were occultly infected.

Conclusion

C. hominis should be added to the list of organisms that can cause bacteremically spread prosthetic joint infection. Blood cultures may be the only positive diagnostic test and these may require 3–5 days of incubation to positivity rather than the usual 1–2 days of incubation required for other organisms such as staphylococci, streptococci and aerobic Gram-negative bacilli. Since two of the three previously described patients had coexistent infective endocarditis, a thorough search should also be made to identify any coexistent cardiovascular pathology [13,15]. There is no universally accepted length of oral suppressive therapy for Gram-negative peri-prosthetic joint infections. Microorganism characteristics and susceptibility patterns will play a key role in antimicrobial selection by the infectious disease physician [17].

Funding

This research received no external funding.

Conflicts of Interest

Declarations

This study is our original work, has not been previously published and is not under consideration for publication by any other journal.