Several tests are used before reimplantation to detect persistent infection in patients with periprosthetic joint infection (PJI) undergoing two-stage revision. However, there is no consensus as to the optimal tests for excluding persistent infection or predicting successful reimplantation by stage. We aimed to determine the accuracy of different tests used to detect persistent infection after the first stage, and/or predicting failure following reimplantation in patients with PJI.

We conducted a systematic review and meta-analysis of PubMed, Embase and the Cochrane Library databases. Two reviewers independently conducted quality assessments and data extractions to estimate pooled sensitivity and specificity, diagnostic odds ratio and area under the receiver operating characteristic curves (AUSROC) for each test.

We included 24 studies published between May 1999 and September 2017. Synovial fluid polymorphonuclear neutrophils (PMN)% had the highest sensitivity of 0.70, followed by serum erythrocyte sedimentation rate (0.57) and spacer sonication fluid culture (0.53). Synovial fluid culture had the highest specificity of 0.97, followed by frozen section (0.93) and the Musculoskeletal Infection Society criteria (0.92). Spacer sonication fluid culture was the most accurate test with an AUSROC of 0.8089, followed by synovial fluid culture (0.7749) and frozen section (0.7819).

Spacer sonication fluid culture had a relatively high diagnostic accuracy. We emphasize that no test can be used alone to exclude persistent infection beyond the first stage and/or predict failed reimplantation beyond the second stage.

II, systematic review and meta-analysis of level 2 to level 4 studies with inconsistent results.