Incidence of Propionibacterium acnes infection in orthopedic and trauma surgery

Introduction

Propionibacterium acnes has been associated with late, smoldering and healthcare-associated infections of the shoulder and spine.

Objectives

The epidemiology of P. acnes with respect of other orthopedic locations and patient populations remains largely unknown.

Methods

Retrospective, single-center, descriptive and case-control studies of adult patients hospitalized for orthopedic infections from 2004-2014. We used only intraoperative microbiological samples and first clinical infection episodes. Cefuroxime (or vancomycine) was used for perioperative prophylaxis. Microbiological samples were incubated for a median of 5 days.

Results

P. acnes was isolated intraoperatively in only 37/2740 (1.35%) surgical procedures. A total of 22/37 infections were monomicrobial. Overall, 665 surgical procedures (24%) involved hardware/osteosynthesis material. P. acnes was more frequently identified during procedures in the presence compared with the absence (24/665 vs. 13/2075; p<0.01) of hardware/foreign material. P. acnes was frequently associated with other skin commensals (12/291 vs. 25/2134; p<0.01) and involved the lumbar and shoulder regions. The proportion of P. acnes among all pathogens in the spine and shoulder were 8% and 6%, respectively. In contrast, P. acnes was almost never identified (3/1021 vs. 334/1719; p<0.01) among immune-suppressed patients, in foot infections, septic bursitis, native bone and joint infections, soft tissue abscesses, prosthetic joints, and tibia nails. By multivariate analysis adjusting for case-mix, the lumbar region (odds ratio 7.4, 95% CI 1.2-46.3), the shoulder (OR 9.9, 1.6-60.1) and the presence of hardware (OR 8.2, 2.4-28.4) were significantly associated with P. acnes infection; while sex, age, immune-suppression and the administration of antibiotic therapy prior to intraoperative sampling were not.

Conclusion

In our institution, P. acnes is very rarely associated with clinical orthopedic infections. It is almost never responsible for infection below the lumbar spine level. P. acnes infections are associated with less inflammatory response than other infections. P. acnes is particularly associated with plate and spondylodesis infections.

Disclosure of interest

None declared.