What Markers Best Guide the Timing of Reimplantation in Two-stage Exchange Arthroplasty for PJI? A Systematic Review and Meta-analysis

Background There is no consensus on the appropriate marker to use when deciding to perform reimplantation after two-stage exchange arthroplasty for periprosthetic joint infection (PJI). Questions/purposes What tests provide acceptable diagnostic value to guide appropriate timing of reimplantation in two-stage exchange arthroplasty for PJI? Methods A search of online databases (MEDLINE, EMBASE, OVID, and Cochrane database) was performed containing articles that provided sensitivity and specificity values for accuracy for predicting reimplantation of the hip and/or knee. Twelve articles were included for final analysis, which included data from 1047 patients. Data that described the diagnostic accuracy of markers for reimplantation were evaluated and categorized into four main entities according to diagnostic method (serologic, synovial, tissue, and diagnostic imaging). Twelve parameters were examined, including serum erythrocyte sedimentation (ESR) rate, serum C-reactive protein (CRP), serum white blood cell (WBC) count, synovial fluid Gram stain, synovial fluid culture, synovial fluid sonication culture, synovial fluid WBC, synovial fluid polymorphonucleocyte percentage (PMN%), tissue Gram stain, tissue culture, positron emission tomography scan, and leukocyte scan. Each of the included articles was independently analyzed for risk of bias and applicability by using QUADAS-2. Statistical heterogeneity was calculated by using the Cochran Q test, and an alpha of 0.10 was considered significant for heterogeneity. Results Tissue culture (sensitivity 0.82 [0.72-0.90], specificity 0.91 [0.89-0.95], diagnostic odds ratio (DOR) 46.87 [95% confidence interval {CI}, 22.03-99.69], synovial fluid PMN% (sensitivity 0.77 [0.46-0.95], specificity 0.74 [0.67-0.81], DOR 11.27 [95% CI, 2.89-43.61]), and synovial fluid culture (sensitivity 0.64 [0.52-0.74], specificity 0.96 [0.93-0.98], DOR 27.07 [95% CI, 2.55-288.00]) showed relatively high diagnostic performance. Other parameters had poorer diagnostic accuracy: ESR (sensitivity 0.56 [0.40-0.72], specificity 0.60 [0.53-0.66], DOR 2.41 [95% CI, 0.60-9.72), CRP (sensitivity 0.53 [0.39-0.67], specificity 0.72 [0.66-0.78], DOR 2.25 [95% CI, 0.09-4.63), and synovial fluid WBC count (sensitivity 0.37 [0.19-0.58], specificity 0.49 [0.41-0.57], DOR 0.94 [95% CI, 0.06-14.74). However, interpretation is limited, because only two to three studies were available for each pooled analysis. Both risks of bias and applicability concerns were low in the four domains assessed in QUADAS-2. Conclusions This meta-analysis suggests that no single marker was superior to all the others, and none (when used alone) is likely sufficient to confirm control of infection after the first stage of a two-stage protocol for PJI. Therefore, the current approach using multiple tools rather than a single marker is essential. Additionally, further studies must be conducted so that pooled analysis can be performed using multiple studies to determine ideal markers for reimplantation.