RESUMEN
PURPOSE: To investigate the rate of bacterial contamination of semitendinosus (ST) tendons during graft harvest in anterior cruciate ligament reconstruction (ACLR), in order to precisely specify the underlying pathogens and obtain data on their susceptibility to potential antibiotics. METHODS: In a prospective study, a total of 59 consecutive patients undergoing primary ACLR were recruited from one centre. No patient had history of previous surgery to the knee or showed clinical signs of infection. Four tissue samples of harvested ST tendons for anterior cruciate ligament (ACL) autografts (case group; ST) were examined for evidence of bacterial colonisation and compared to four tissue samples of the native ACL as negative controls (control group; ACL). Three of the respective samples were subjected to cultural microbiological examination and one to 16S rRNA-PCR. Antibiotic susceptibility testing was performed for each pathogen that was identified. RESULTS: A total of 342 samples were analysed by culture. Significantly more patients showed a positive culture of the ST (33.9%; n = 20/59) compared to 3.4% of patients (n = 2/59) with positive culturing of the ACL (p < 0.0001). Including 16S rRNA-PCR, in a total of 42.4% (25/59) of patients, bacteria were detected in at least one ST sample either by PCR and/or culture. All species found (n = 33) belong to the typical skin flora with Staphylococcus epidermidis (39.4%; n = 13/33) being the most common species, followed by Staphylococcus capitis (24.2%; n = 8/33). All tested isolates (n = 29) were susceptible to vancomycin (29/29, 100%), 69% (n = 20/29) to oxacillin and 65.5% (n = 19/29) to clindamycin. CONCLUSION: ST autografts for ACLR were commonly contaminated with skin commensal bacteria during harvest. One-third of the isolates showed resistance to typical perioperative intravenous antibiotics, whereas all isolates were sensitive to vancomycin. Therefore, routine prophylactic decontamination of all hamstring autografts before implantation should be recommended, preferably with topical vancomycin. LEVEL OF EVIDENCE: Level III.