Histological Findings in Infected and Noninfected Second Stage Revision Knee Arthroplasties

Tissues from a periprosthetic joint infection (PJI) of the knee contain a heavy neutrophil polymorph (NP) infiltrate (> 5 NPs per high-powered field [HPF] by Musculoskeletal Infection Society [MSIS] criteria). PJI of the knee can be treated by a two-stage procedure and our aim was to determine whether the MSIS histological criteria for PJI diagnosis are valid in a second-stage revision knee arthroplasty. Periprosthetic tissues from 45 second-stage revision knee cases were analyzed histologically by hematoxylin-eosin and chloroacetate esterase (CAE) staining for the identification of NPs. The number of NPs was determined semiquantitatively and results correlated with the microbiological and clinical findings. In 9 of the 45 cases, an organism was cultured in two or more samples, meeting MSIS microbiological criteria for a definite diagnosis of PJI; histologically, seven of these cases contained > 5 per NPs per HPF on average, with the remaining two cases containing 1 NP and 2 NPs per HPF. In noninfected second-stage revisions, NPs were not seen in 30 cases with 6 cases showing less than 1 NP per HPF on average. The sensitivity, specificity, accuracy, and positive and negative predictive values of MSIS histological criteria (> 5 NPs per HPF) to diagnose PJI were 78%, 100%, 96%, 100%, and 95%, respectively. MSIS histological criteria for the diagnosis of PJI are valid for most but not all infected second-stage revision knee arthroplasties. Correlation of histology with clinical, microbiology and other laboratory findings is required to establish a diagnosis of PJI in second-stage revision knee arthroplasties.