|Beitragstitel||Can a single positive sample of Propionibacterium acnes from revision arthroplasty be ignored?|
Propionibacterium acnes (P. acnes) is a skin commensal but also known to cause invasive infections such as periprosthetic joint infections (PJI). Extended cultivation times up to 10-14 days increase the likelihood for detection of P. acnes but also the chance of cultivating a contamination. Although according to the literature a single positive sample is typically judged as contamination, the surgeon feels uncertain whether longtime antibiotic treatment is favorable for an infection free outcome. The aim of this study was to evaluate the outcome of patients with a single positive sample for P. acnes in cases without previous signs of infection.
In a retrospective analysis from 2005 to 2014, patients with at least one positive sample for P. acnes in intraoperative tissue or sonication fluid in presence of an arthroplasty were identified. Only one positive of at least three taken specimens was considered a contamination whereas two or more positive samples were defined as an infection. Patients with more than one positive sample and outcome less than 2 years were excluded. Surgical, antibiotic treatment and outcome were evaluated. Relapse of infection was defined as growth of P. acnes in ≥ 2 samples during follow-up.
In total 55 patients with a positive sample for P. acnes were identified 27 cases (27 patients) were diagnosed as contamination (11 shoulders, 9 hips, 7 knees) with no previous sign of infection. All received surgery and 20 (74%) were also treated with antibiotics for at least 6 weeks. In the shoulder 1 patient (9%) had a two-stage, 8 a one-stage exchange (73%) and 18% a debridement with retention of the implant. 7 received antibiotics and 4 not. 3 total hips were exchanged two-staged (33%) and 6 one-staged (66%). 6 were treated with antibiotics and 3 without. 3 knee prostheses were temporarily replaced by a spacer (43%), 2 exchanged one-staged (29%) and 2 debrided with retention (29%). All knees received antibiosis for at least 6 weeks. In total 19 patients were administered antibiotics, leaving 7 contaminations (4 shoulders, 3 hips) without, which were analyzed regarding relapse. None of these, or the ones having received antibiotics, had later joint samples positive for P. acnes.
When correct intraoperative bacteriological sampling with at least 3 specimens is performed, a single positive sample of P. acnes during revision arthroplasty might not need to be treated with longtime antibiotics.