Full bibliography

Background: Propionibacterium acnes (P. acnes) is the most common bacteria associated with infection after shoulder arthroplasty. These bacteria can be grown on culture of skin after standard preoperative skin preparation and antibiotics. The purpose of this study was to determine whether adding preoperative intravenous doxycycline reduces the prevalence of positive P. acnes cultures of skin and deep tissues at the time of prosthetic joint implantation during shoulder arthroplasty. Methods: This was a randomized controlled trial. An a priori power analysis determined that a sample size of 56 patients was necessary. Patients scheduled to undergo shoulder arthroplasty were randomized to receive either standard perioperative cefazolin or a combination of doxycycline and cefazolin. Tissue specimens for culture were then taken from the skin edge, and swabs of the superficial dermal tissue and glenohumeral joint were obtained. All cultures were maintained for 14 days to allow for P. acnes detection. Groups were compared to determine if the addition of doxycycline reduced the rate of culture positivity. Results: Fifty-six patients were enrolled and randomized. Twenty-one (38%) had ≥1 positive cultures for P. acnes, with no significant difference between the group treated with cefazolin alone (10 [37%] of 27 patients) and the combined doxycycline and cefazolin group (11 [38%] of 29 patients) (p = 0.99). The greatest numbers of culture-positive samples were obtained from the skin (30%), followed by dermal tissue (20%) and the glenohumeral joint (5%). Patients who had ≥1 positive cultures were younger than those who did not (mean age [and standard deviation], 64.9 ± 7.7 versus 69.4 ± 7.7 years; p = 0.041), had a greater tendency to be male (16 [76%] of 21 versus 17 [49%] of 35; p = 0.053), and had a lower Charlson Comorbidity Index (3.35 ± 1.3 versus 4.09 ± 1.4; p = 0.051). There were no significant differences between the culture-positive and culture-negative groups in terms of body mass index (BMI) (p = 0.446) or arthroplasty type, with positive cultures found for 8 of the 29 anatomic shoulder arthroplasty procedures compared with 13 of the 27 reverse shoulder arthroplasty procedures (p = 0.280). There were no doxycycline-related adverse events. Conclusions: In this randomized controlled trial, doxycycline did not significantly decrease P. acnes culture positivity of the skin, dermis, or glenohumeral joint of patients undergoing shoulder arthroplasty. The addition of prophylactic intravenous antibiotics to cover P. acnes may not be an effective method to reduce postoperative and periprosthetic shoulder joint infections.