[Intraoperative direct sonication for the diagnostic efficacy of periprosthetic joint infection]

Objective: To investigate the diagnostic efficacy of intraoperative direct sonication in periprosthetic joint infection (PJI). Methods: This is a retrospective case series study. The clinical data of 490 patients with PJI or aseptic prosthesis loosening after hip or knee arthroplasty admitted to the Department of Orthopaedics, First Affiliated Hospital of Xinjiang Medical University from January 2018 to December 2023 were retrospectively analyzed. There were 209 male cases, 281 female cases, aged (M(IQR)) 63 (19) years (range: 15 to 89 years). There were 328 cases of PJI (123 cases of traditional sonication, 205 cases of intraoperative direct sonication, and 204 cases had simultaneous intraoperative synovial fluid), and 162 patients of aseptic loosening (89 cases of traditional sonication, 73 cases of intraoperative direct sonication, and 96 cases had simultaneous intraoperative synovial fluid). The method of traditional sonication: the prosthesis components were placed in a sterile container, vortexed for 30 seconds, then placed in an ultrasound bath to sonication (frequency:(40±2) kHz, power density:(0.22±0.04) W/cm²) for 5 minutes, vortexed again for 30 seconds, and the resulting sonicate fluid was extracted for culture. The method of intraoperative direct sonication: during the operation, the surgical area, the prosthesis and the tissues around the prosthesis which were placed in a sterile container were respectively subjected to sonication for 5 minutes using a portable handheld ultrasonic cell disruptor device (frequency: 25 kHz, power density:(0.22±0.04) W/cm²), and the fluids before and after sonication were extracted for culture. The method of intraoperative synovial fluid: during the operation, the joint capsule was incised, and the synovial fluid was extracted under direct vision for culture. The sensitivity, specificity, positive and negative predictive values, Youden index, the effect of preoperative antimicrobial agents on culture results, and culture duration for different culture methods. Receiver operating characteristic (ROC) curves were plotted, and the diagnostic efficacy of these methods for PJI was compared using the χ² test, the Wilcoxon signed-rank test, and other appropriate statistical methods. Results: The sensitivity of intraoperative direct sonication was significantly higher than that of intraoperative synovial fluid culture (89.8%(184/205) vs.56.9%(116/204),χ²=44.457,P<0.01) and traditional sonication (89.8%(184/205) vs.66.7%(82/123), χ²=121.588,P<0.01). However, its specificity was lower compared to intraoperative synovial fluid (87.5%(64/73) vs. 99.0%(95/96), χ²=9.491, P=0.002). The culture duration for intraoperative direct sonication was shorter than for intraoperative synovial fluid 87.0 (41.8) hours vs. 112.5 (78.5) hours, Z=-5.121, P<0.01) and traditional sonication (87.0 (41.8) hours vs. 119.0 (67.5) hours, Z=-7.119, P<0.01). Gram-positive bacteria (predominantly Staphylococcus aureus and Staphylococcus epidermidis) were the most common isolates across all three culture methods. Furthermore, intraoperative direct sonication was more likely to detect polymicrobial infections compared to traditional sonication (38.6% vs. 2.4%, χ²=37.223, P<0.01) and intraoperative synovial fluid (38.6% vs. 0.9%, χ²=55.527, P<0.01). The ROC curve revealed that the area under the curve of intraoperative direct sonication,intraoperative synovial fluid culture and traditional sonication were 0.886 (95%CI:0.843 to 0.930), 0.779 (95%CI:0.743 to 0.815) and 0.788 (95%CI: 0.736 to 0.839). Conclusions: Compared with intraoperative synovial fluid and traditional sonication, intraoperative direct sonication has excellent sensitivity, the shortest appropriate culture duration for pathogenic bacteria, and it is more likely to diagnose PJI patients with polymicrobial infections, having better diagnostic efficacy in the diagnosis of PJI.