Molecular Tools for Guiding Therapy in Patients With Staphylococcal Bone and Joint Infections: A Diagnostic Test Accuracy Meta-analysis

Abstract

Background: Timely detection of causative pathogens and their antimicrobial resistance are essential for guiding targeted therapies in bone and joint infections (BJI) patients. We performed a systematic review and meta-analysis to assess the diagnostic value of testing osteoarticular samples with the nucleic acid amplification tests (NAAT) for effective staphylococcal strain identification and the administration of appropriately targeted antimicrobial agents in BJI patients. Methods: Five databases, including PubMed, Embase, Scopus, Web of Science, and the Cochrane Library, were searched for related publications from inception to July 24, 2021. Studies comparing the diagnostic accuracy of NAAT to a microbiological culture reference standard of osteoarticular specimens were eligible. Pooled summary values of sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR) of NAAT compared to the microbiological culture reference standard were calculated using bivariate random-effects meta-analyses. Results: From 906 citations, 11 studies were included. Eleven studies comprising 13 datasets (n = 1047) evaluated NAAT accuracy for methicillin-sensitive Staphylococcus aureus (MSSA) identification, while seven studies comprising nine datasets (n = 727) evaluated methicillin-resistant Staphylococcus aureus (MRSA) identification. Against the microbiological culture reference standard, the pooled summary estimates for detection of both MSSA [sensitivity: 0.89 (95% confidence interval [CI] 0.84–0.93), specificity: 0.99 (95% CI 0.97–0.99), PLR: 34.13 (95% CI 20.54–56.73), NLR: 0.19 (95% CI 0.12–0.3), and DOR: 283.37 (95% CI 129.49–620.1)] and MRSA [sensitivity: 0.81 (95% CI 0.67–0.91), specificity: 1.0 (95% CI 0.99–1.0), PLR: 62.1 (95% CI 24.5–157.6), NLR: 0.33 (95% CI 0.16–0.69), and DOR: 300.25 (95% CI 85.01–1060.5)] were comparable. Heterogeneity was moderate. GeneXpert was frequently used among NAA tests, and its diagnostic accuracy was in line with the overall pooled summary estimates. The heterogeneity in diagnostic efficacy (P >0.05) could not be explained by a meta-regression and subgroup analysis of the research design, sample condition, and patient selection technique. Conclusions: Our study suggested that NAAT can be applied as the preferred prescreening test for the timely diagnosis of staphylococcal strains associated with BJI in osteoarticular samples for successful antimicrobial therapy.