Abiotrophia defectiva knee prosthesis infection: A case report

Abiotrophia organisms are nutritionally variant streptococci that form part of the commensal flora of the oral and intestinal mucosae [1]. The detection and identification of Abiotrophia organisms may be hampered by their inability to grow in standard media and their microscopic pleomorphism [1]. Abiotrophia organisms have been grouped into a unique genus separated from the Streptococcus genus based on genetic and phylogenetic analyses, which revealed a low relatedness to other Streptococcus organisms [2, 3]. Abiotrophia organisms cause bacteremia and an estimated 4-6% of all cases of streptococcal endocarditis [2, 4]. However, these organisms have rarely been implicated as the causative agents of osteoarticular infections. Indeed, only four cases of orthopedic infections caused by A. defectiva have been published. These cases include two instances of septic arthritis, one case of discitis and sacroiliitis and one case of total knee arthroplasty infection [5‐8] (Table 1). Here, we report on a fifth case of A. defectiva osteoarticular infection through the isolation of the organism and the direct detection of A. defectiva- specific DNA sequences through polymerase chain reaction (PCR) detection in the synovial fluid.

A 71-year-old man of Italian origin was admitted to our infectious orthopedic department presenting with chronic (two-year) left knee pain, swelling and decreasing ambulation. He had received a total knee prosthesis for left knee osteoarthritis four years prior. The prosthesis had been replaced twice for radiologically confirmed loosening, with the latest replacement being two years prior to admittance. Cultures of synovial fluid collected during both replacements remained sterile after inoculation of standard solid media. Our patient underwent a dental treatment without antibiotic prophylaxis three weeks before the consultation. Upon physical examination, our patient was found to be apyretic. His left knee was swollen and radiating heat, and a moderately sized effusion was present. His white blood cell count indicated 3.82 × 10³ polymorphonuclear cells/mL, his C-reactive protein (CRP) level was 32 mg/L and the erythrocyte sedimentation rate (ESR) was 66 mm (first hour). Our patient was referred to the orthopedic department for sample biopsy and arthroscopic lavage. The synovial fluid was directly inoculated into a set of Bactec Plus Aerobic/F and Bactec Lytic/10 Anaerobic/F bottles (BD Diagnostic Systems, Sparks, MD, USA) and collected in parallel into a sterile tube as part of an "arthritis kit", as previously described [9]. Microscopic examination of the synovial fluid indicated 32 × 10³ white blood cells/mL (98% neutrophil polymorphonuclear cells and 2% monocytes) and 4.77 × 10³ red blood cells/mL without crystals. No microorganisms were observed. Inoculation of the synovial fluid onto 5% sheep-blood agar showed no growth after a five-day incubation period at 37°C under a 5% carbon dioxide (CO₂) atmosphere. However, the aerobic broth bottle produced Gram-positive cocci after a five-day incubation period at 37°C under continuous automated monitoring for bacterial growth in the medium. Subculturing these organisms on 5% sheep-blood agar and chocolate agar (BioMérieux, Marcy l'Etoile, France) at 37°C under a 5% CO₂ atmosphere produced colonies. These colonies were identified as A. defectiva by 16S rDNA sequencing [10] (99.9% sequence similarity with the GenBank accession number AY879308); E test testing confirmed the isolate to be broadly susceptible with minimum inhibitory concentration of 0.01 mg/L for penicillin G and rifampicin, 0.016 mg/L for amoxicillin, ceftriaxone and vancomycin, 0.02 mg/L for imipenem, 0.1 mg/L for erythromycin and 0.2 mg/L for doxycycline [11]. PCR-based detection of bacterial DNA using universal 16S rDNA primers [10] detected A. defectiva DNA in the synovial fluid on the basis of sequence identity to the DNA recovered from the colonies. Three blood cultures incubated under an aerobic atmosphere and one blood culture incubated under an anaerobic atmosphere showed no growth after a five-day incubation period. A transthoracic echocardiography revealed no signs of endocarditis and a colonoscopy was unremarkable. The knee prosthesis was removed, and a temporary cement spacer containing 2 g vancomycin and 40 g cement was inserted [12]. Re-implantation of a total knee was delayed because of an intercurrent venal thrombosis without pulmonary embolism. Our patient received 100 mg/kg/day oral amoxicillin for nine months. The knee pain and swelling subsequently resolved, and the ESR decreased to 22 mm. The one-year follow-up examination indicated complete wound healing, with no evidence of a recurrence of infection, a white blood cell count within the normal range, a CRP value of 7.1 mg/L and an ESR of 17 mm.

In our patient, A. defectiva knee prosthesis infection was documented by the isolation and culture of the organism from a synovial fluid recovered under sterile procedures and the parallel detection of A. defectiva DNA from this specimen. Although only one joint fluid specimen was available for microbiological investigations, no other detection of A. defectiva had been made in our laboratory within the eight-week period preceding isolation of the A. defectiva strain here reported. Thus, laboratory contamination is unlikely.

The reported A. defectiva isolate was recovered after the direct inoculation of the synovial fluid into a culture broth as part of our protocol for the diagnosis of prosthetic joint infection using an "arthritis kit" procedure. The arthritis kit contains one Bactec Plus Aerobic/F and one Bactec Plus Anaerobic/F for culturing bacteria, one sterile tube for culturing mycoplasma, spirochetes and 16S rRNA-based PCR and one sterile tube for culturing and PCR-based detection of virus, in addition to the informed consent form and laboratory forms. The parallel inoculation of this same fluid specimen onto agar media failed to culture the organism. This illustrates the well-known fastidious nature of A. defectiva and indicates that the direct inoculation of the synovial fluid into broth is mandatory for the culture-based diagnosis of A. defectiva articular infections. The same holds true for the culture-based diagnosis of arthritis caused by Kingella kingae in children [13].

In this patient, the A. defectiva infection was documented shortly after a dental procedure without antibiotic prophylaxis. However, this dental procedure was most likely not the portal of entry for A. defectiva, as our patient had presented clinical signs of knee prosthesis infection long before the dental procedure. A prospective case-control study indicated that dental procedures were not risk factors for orthopedic prosthetic infection [14]. Indeed, antibiotic prophylaxis before dental interventions in patients with knee arthroplasty lacks evidence-based recommendations [15].

Our patient had a history of repeated early prosthesis loosening, a situation evocative of chronic infection of the prosthesis [16, 17]. Tentative documentations remained negative, but these diagnoses lacked direct inoculation of the synovial fluid into broth and molecular detection. Therefore, the loosening episodes of the two prostheses may have been due to a chronic A. defectiva infection.

The fastidious nature of A. defectiva may explain the rarity of A. defectiva orthopedic infections, and cases of Abiotrophia infection may be misdiagnosed as culture-negative infections. Accordingly, the role of Abiotrophia organisms in osteoarticular infections may be underestimated. A review of the literature indicates that only one case of an orthopedic prosthesis Abiotrophia infection has been published, and A. defectiva joint infection has been reported in only four cases (Table 1).

Written informed consent was obtained from the patient for publication of this case report. A copy of the written consent is available for review by the Editor-in-Chief of this journal.