Value of mNGS in sonication fluid for the diagnosis of periprosthetic joint infection

OBJECTIVE: To evaluate the effectiveness of metagenomics next-generation sequencing (mNGS) for the detection of pathogenic microorganisms in periprosthetic joint infection (PJI) using the sonication fluid from removed prosthesis. METHODS: In this prospective diagnostic cohort study, 44 patients who underwent revision arthroplasty in our hospital from December 2016 to December 2018 were screened. Seven cases were excluded due to incomplete clinical data, insufficient synovial fluid or failure of sequencing. According to the PJI diagnostic criteria recommended by the Musculoskeletal Infection Society (MSIS), the patients were defined as PJI or aseptic failure (AF). Conventional culture, sonication fluid culture and mNGS were performed, in order to assess the value of mNGS using sonication fluid for the diagnosis of PJI, and the mNGS results were analyzed and compared with the conventional and sonication fluid culture. RESULTS: Among the 37 patients, 24 were diagnosed with PJI (64.86%), while 13 were diagnosed with aseptic failure. Among the 24 patients diagnosed with PJI, 15 cases (62.5%), 17 cases (70.8%) and 24 cases (100%) yielded positive results in conventional culture, sonication fluid culture and mNGS, respectively. In addition, mNGS detected the same pathogenic microorganisms in 16 out of the 17 (94.12%) culture-positive (conventional + sonication fluid) PJI cases. In the only one discrepancy case, Enterococcus faecalis was identified in the cultures, while Enterobacter cloacae was detected by mNGS. In the AF group, the results of the conventional culture were all negative. Nevertheless, Staphylococcus epidermidis was detected in the sonication fluid culture and mNGS in one case. The diagnostic sensitivity of mNGS for PJI was 100%, which was significantly higher than 70.83% (P = 0.039) of the sonication fluid culture and 62.5% (P = 0.021) of the conventional culture. The diagnostic specificity of mNGS for PJI was 92.31%, which was not significantly different (P > 0.05) from those of the conventional culture (100%) and sonication fluid culture (92.31%). CONCLUSION: We demonstrated that mNGS using sonication fluid can improve the detection rate of pathogenic microorganisms and provide valuable information for the diagnosis of PJI. In addition, mNGS can effectively identify pathogenic microorganisms in culture-negative PJIs cases, especially for the cases who have been treated with antibiotics before sample acquisition or have fastidious microorganisms. Therefore, this method can potentially help to guide the clinical use of antibiotics.