Histological analysis in the diagnosis of periprosthetic joint infection of the hip and knee : a systematic review and meta-analysis.

AIMS: The aim of this systematic review and meta-analysis was to assess the diagnostic value of the histological analysis of deep tissue samples in the diagnosis of periprosthetic joint infection (PJI) following total hip (THA) or knee arthroplasty (TKA). The efficacy of the most prevalent diagnostic thresholds (≥ 23 polymorphonuclear neutrophils (PMNs)/ten high-power fields (HPFs), ≥ five PMNs/HPF, and ≥ ten PMNs/HPF) was investigated to determine the optimal threshold to differentiate between septic and aseptic cases.

METHODS

PubMed (MEDLINE) and Embase were searched for studies evaluating the performance of histology to diagnose PJI in THAs and TKAs. A meta-analysis of the 43 included studies determined the pooled sensitivity, specificity, the diagnostic odds ratio (DOR) and the area under the summary receiver operating characteristic curve (AUSROC) of permanent (formalin-fixed) and frozen sections.

RESULTS

The performance of permanent sections was evaluated in 22 studies (n = 2,697; PJI 761/2,697; 28%). When considering only studies analyzing intraoperatively collected tissue samples (n = 17), the pooled sensitivity, specificity, DOR, and AUSROC were 82.0% (95% CI 80.4 to 83.5), 96.0% (95% CI 95.1 to 96.7), 153.7 (95% CI 69.3 to 340.9), and 0.965 (standard error (SE) 0.01). The threshold of ≥ five PMNs/HPF demonstrated the best diagnostic performance (sensitivity 82.0% (95% CI 80.0 to 84.0), specificity 94.7% (95% CI 93.5 to 95.8), DOR 133.5 (95% CI 41.6 to 428.6), and AUSROC 0.963 (SE 0.02)). The performance of intraoperatively collected frozen sections was evaluated in 25 studies (n = 3,137; PJI 538/3,137; 17%). The same diagnostic estimates were 67.8% (95% CI 66.1 to 69.4), 94.3% (95% CI 93.4 to 95.1), 47.1 (95% CI 27.7 to 80.2), and 0.960 (SE 0.01), respectively.

CONCLUSION

Due to their high accuracy, permanent sections of intraoperatively collected samples can be recommended as a confirmatory criterion for diagnosing PJI in THAs and TKAs. While frozen sections demonstrated lower sensitivities, specificities remained robust and comparable with those of permanent sections. Thus, they can also be used to confirm PJI, particularly when the findings of other preoperative diagnostic tests are inconclusive. In order to differentiate septic from aseptic cases, a threshold of ≥ five PMNs/HPF in each of at least five HPFs is advocated. High-quality prospective multicentre studies are needed to validate these findings.