Background: Proprionibacterium acnes (P. acnes) is a skin commensal which is often interpreted as a contaminant when found in cultures of surgical specimens. However, recent reports suggest that P. acnes can be identified as the causative micro-organism of infection. Furthermore P. acnes infections occur more often after shoulder surgery than after surgery of the lower extremities or spine. The aim of this study was to identify how frequent P. acnes was responsible for infection after orthopaedic surgery of the shoulder, lower extremity and spine in a single centre.

Patients and Methods: Inclusion criteria were the occurrence of infection after surgery of the shoulder, lower extremity (hip and knee), or spine in a 100-bed orthopaedic hospital. The inclusion period was between January 2000 and May 2008. Infection was defined when two or more cultures were positive with the same microorganism in the presence of clinical signs and symptoms. The first goal was to identify the incidence of infection due to P. acnes amongst all infections. The secondary outcome was the incidence of infection versus contamination amongst all cases with positive cultures for P. acnes. Both outcomes were compared for surgery of the shoulder, lower extremity and spine.

Results: A total of 3703 surgeries of the shoulder were performed, compared to 19906 lower extremity- and 5687 spine surgeries. The incidence of infection after surgery of the shoulder was 1.4% (52 cases; prosthesis [n=23], fractures [n=5], soft tissue surgery [n=16] and others [n=8]). After surgery of the lower extremity and spine this was 2.8% (548 cases) and 3.1% (177 cases), respectively. The incidence of infection due to P. acnes after shoulder surgery (23%) was significantly greater then after surgery of the lower extremity (1.3%; p < 0.001) or spine (0%; p < 0.001). Furthermore, in cases where P. acnes was cultured after surgery of the shoulder it was more often identified as the causative pathogen of infection than when P. acnes was cultured after surgery of the lower extremity or spine (71% vs 22%; p < 0.05 and 71% vs 0%; p < 0.01).

Conclusion: The low virulent P. acnes can cause orthopaedic surgical infections and should not be regarded a priori as a contaminant in cultures. This is especially true for shoulder surgery, where P. acnes infections occur frequently and significantly more often than in surgery of other joints.