Incidence of Culture Positive Propionibacterium acnes in Shoulder Arthroscopy

Abstract

Objectives:Accurate detection and diagnosis of Propionibacterium acnes infection after shoulder surgery is often challenging. Aspiration and tissue cultures yield both false negative and false positive cultures and determination of infection is not always straightforward. The purpose of our study was to measure P. acnes positive cultures from patients undergoing arthroscopic surgery, with no previous surgical history on that shoulder. Our hypothesis was that P. acnes would not be found in patients who had not previously had shoulder surgery.Methods:Seven samples were collected from 57 consecutive patients undergoing shoulder arthroscopy, without any previous shoulder surgery. An air swab was taken as a control for this study. Two skin swabs (one after skin preparation and one prior to skin closure), aspiration of joint fluid, and three samples of debrided tissue were obtained for culture. All samples were placed on three different aerobic plates and held for 48 hours. Samples were also plated on anaerobic plates and placed into a Thioglycollate broth and held for 28 days. All bacteria cultures were identified. Isolated P. acnes were tested for biotype with a Microscan Rapid Anaerobic ID and then streaked on brucella agar/ ANA CDC agar to measure hemolysis. The biotype isolated was also tested with Epsilometer tests.Results:Fifty-seven patients underwent arthroscopic shoulder surgery for subacromial decompression (22.8%), rotator cuff repair (57.9%), labrum or instability repair (19.3%). The mean age of the patients was 51 years (range 17-81). A total of 81 samples (21.8%) were positive for P. acnes in the timeframe of zero to fourteen days (Table 1). All positive cultures were non-hemolytic. Diagnosis, history of corticosteroid injection, or antibiotic received had no significant bearing on culture results. Positive skin cultures for P. acnes increased from 10.5 % before incision to 31.9 % at closure. This was even more pronounced in males as skin cultures increased from 31.3% before incision to 62.96% at closure. Fifteen (26.3%) patients had more than three cultures positive. None of the patients in this study have had signs or symptoms to suggest clinical infection of P. acnes.Conclusion:There was high rate (21.8 %) of positive culture for P. acnes culture positives in a group of patients that we would not expect to find bacteria. When samples are held for 28 days, this percent increased to 25.3%. This either represents false positives via contamination, inadequate skin preparation or may represent colonization as no patients have clinical symptoms of infection. These findings make the interpretation of positive P. acnes culture even more challenging when managing a patient with shoulder pain after surgery. The consequence of treating a false positive is not benign. There was also a remarkable increase in positive rate from culture obtained from the skin at the beginning of surgery (10.5%) as compared to at the end (31.9%). Future studies should examine alternate surgical skin preparation.