Predicting the Presence of Adjacent Infections in Septic Arthritis in Children.

BACKGROUND

The gold standard for treatment of septic arthritis is urgent surgical debridement. Preoperative magnetic resonance imaging (MRI) may identify osteomyelitis, subperiosteal abscesses, and intramuscular abscesses, which frequently occur with septic arthritis. If these adjacent infections are not recognized, initial treatment may be inadequate. The purpose of this study is to develop a prediction algorithm to distinguish septic arthritis with adjacent infections from isolated septic arthritis to determine which patients should undergo preoperative MRI.

METHODS

An IRB-approved retrospective review of 87 children treated for septic arthritis was performed. All patients underwent MRI. Sixteen variables (age, sex, temperature, WBC, CRP, ESR, ANC, hematocrit, platelet count, heart rate, systolic blood pressure, diastolic blood pressure, symptom duration, weight-bearing status, prior antibiotic therapy, and prior hospitalization) from admission were reviewed. Graphical and logistical regression analysis was used to determine variables independently predictive of adjacent infection. Optimal cutoff values were determined for each variable and a prediction algorithm was created. Finally, the model was applied to our patient database and each patient with isolated septic arthritis or adjacent infection was stratified based upon the number of positive predictive factors.

RESULTS

A total of 36 (41%) patients had isolated septic arthritis and 51 (59%) had septic arthritis with adjacent foci. Five variables (age above 3.6 y, CRP>13.8 mg/L, duration of symptoms >3 d, platelets <314×10 cells/μL, and ANC>8.6×10 cells/μL) were found to be predictive of adjacent infection and were included in the algorithm. Patients with ≥3 risk factors were classified as high risk for septic arthritis with adjacent infection (sensitivity: 90%, specificity: 67%, positive predictive value: 80%, negative predictive value: 83%).

CONCLUSIONS

Age, CRP, duration of symptoms, platelet count, and ANC were predictive of adjacent infections. Patients who met ≥3 criteria are at high risk for adjacent infection and may benefit from preoperative MRI.

LEVEL OF EVIDENCE: Level III—retrospective comparative study.