Synovial fluid cell count for diagnosis of chronic periprosthetic hip infection

Background: There is a paucity of data regarding the threshold of synovial fluid white blood-cell (WBC) count and polymorphonuclear cell (neutrophil) percentage of the WBC count (PMN%) for the diagnosis of chronic periprosthetic joint infection (PJI) after total hip arthroplasty. Despite this, many organizations have provided guidelines for the diagnosis of PJI that include synovial fluid WBC count and PMN%. We attempted to define a threshold for synovial fluid WBC count and PMN% for the diagnosis of chronic PJI of the hip using a uniform definition of PJI and to investigate any variations in the calculated thresholds among institutions. Methods: From 4 academic institutions, we formed a cohort of 453 patients with hip synovial fluid cell count analysis as part of the work-up for revision total hip arthroplasty. Using the definition of PJI from the Musculoskeletal Infection Society (MSIS), 374 joints were diagnosed as aseptic and 79, as septic. Intraoperative aspirations were performed as routine practice, regardless of the suspicion for infection, in 327 (72%) of the patients. Using receiver operating characteristic curves, the optimal threshold values for synovial WBC count and PMN% were identified. Results: For the diagnosis of chronic PJI of the hip, the threshold for the overall cohort was 3,966 cells/μL for WBC count and 80% for PMN%. Despite the high predictive accuracy for the cohort, there was notable institutional variation in fluid WBC count and PMN%. Furthermore, the rate of PJI was 14% (4 of 28) for patients with a WBC count of 3,000 to 5,000 cells/μL compared with 91% (20 of 22) for patients with a WBC count of >50,000 cells/μL. Similarly, the rate of PJI was 29% (14 of 49) for patients with a PMN% of 75% to 85% compared with 69% (33 of 48) for patients with a PMN% of >95%. Conclusions: Using the MSIS criteria, the optimal synovial fluid WBC count and PMN% to diagnose chronic PJI in the hip is closer to thresholds for the knee than those previously reported for the hip. This study validates the diagnostic utility of synovial fluid analysis for the diagnosis of periprosthetic hip infection; however, we also identified a clinically important "gray area" around the threshold for which the presence of PJI may be unclear. Level of Evidence: Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.