Optimal Synovial Fluid Leukocyte Count Cutoff for Diagnosing Native Joint Septic Arthritis After Antibiotics: A...

Urgent treatment of septic arthritis is key in preventing devastating morbidity or mortality. Accurate diagnosis is critical, and the standard diagnostic cutoff of 50,000 synovial leukocytes may be altered by previous administration of antibiotics. Our objective was to identify and compare a cutoff synovial leukocyte count with a high sensitivity and specificity for diagnosis of septic arthritis in patients who received antibiotics and those who had not. A receiver operating characteristic (ROC) curve was used to provide a discriminate cutoff value for diagnosing septic arthritis.

A retrospective chart review of 383 patients was done over a 13-year period including those who had arthrocentesis of any joint. Two groups were created, those who had not been given antibiotics within 2 weeks (control) and those who received intravenous or oral antibiotics within 2 weeks before arthrocentesis. Relevant data included synovial leukocyte count and differential cell count. Additional metrics included temperature, erythrocyte sedimentation rate, and C-reactive protein. A ROC curve determined the optimal synovial white blood cell cutoff for diagnosing septic arthritis in native joints for each group.

The ROC curve determined that patients who received antibiotics had an optimal cutoff of >16,000 cells (sensitivity = 82%, specificity = 76%), and a neutrophil percentage cutoff of >90% (sensitivity = 73%, specificity = 74%). The control group had an optimal synovial leukocyte cutoff of >33,000 cells (sensitivity = 96%, specificity = 95%). The optimal neutrophil percentage cutoff in the control group was >83% neutrophils (sensitivity = 89%, specificity = 79%).

When a patient is given antibiotics before arthrocentesis, a diagnostic value of >16,000 synovial leukocytes should be used to guide treatment of septic arthritis. A diagnostic value of >33,000 synovial leukocytes yields the highest accuracy for diagnosis of septic arthritis in patients who have not been given antibiotics before arthrocentesis.

III.