Intraoperative frozen section histopathology for the diagnosis of periprosthetic joint infection in hip revision surgery: the influence of recent dislocation and/or periprosthetic fracture.

To evaluate the accuracy of intraoperative frozen section histopathology for diagnosing periprosthetic joint infection (PJI) during hip revision surgery, both for patients with and without recent trauma to the hip. The study included all revision total hip replacement procedures where intraoperative frozen section histopathology had been used for the evaluation of infection in a single institution between 2008 and 2015. Musculoskeletal Infection Society criteria were used to define infection. 210 hips were included for evaluation. Prior to revision surgery, 36 hips had a dislocation or a periprosthetic fracture (group A), and 174 did not (group B). The prevalence of infection was 14.3% (5.6% in group A and 16.1% in group B). Using Feldman criteria, the sensitivity of histopathology was 50.0%, specificity 47.1%, positive predictive value 5.3% and negative predictive value 94.1% in group A. The sensitivity of frozen section histopathology was 75.0%, specificity 96.5%, positive predictive value 85% and negative predictive value 95.3% in group B. Intraoperative frozen section histopathology is reliable for the diagnosis of PJI if no dislocation or periprosthetic fracture has occurred prior to hip revision surgery.