The effect of adding orbital computed tomography findings to the Chandler criteria for classifying pediatric orbital cellulitis in predicting which patients will require surgical intervention

PURPOSE To assess the effect of adding orbital computerized tomography (CT) findings to the Chandler criteria for classifying pediatric orbital cellulitis in predicting which patients. will require surgical intervention. METHODS The medical records of patients with orbital CT at a tertiary pediatric hospital from January 2000 to March 2011 were reviewed retrospectively. CT images of cases with radiology report of postseptal orbital involvement were further reviewed by a neuroradiologist. RESULTS Of 101 cases of orbital cellulitis, 71 (mean age, 7.1 +/- 4.0) were successfully managed with systemic antibiotics alone; 30 patients (mean age, 7.2 +/- 4.3) required surgical intervention. Bony destruction on CT was significantly associated with surgical intervention (P = 0.02), and the size of the subperiosteal abscess (SPA) was significantly correlated with management outcome. Patients who were managed with systemic antibiotics alone had a mean SPA volume of 2.1 +/- 2.4 mL; those who had undergone surgical intervention had a mean SPA volume of 14.3 mL +/- 16.8 mL (P < 0.0001). If SPA volume is <3.8 mL, then the probability of surgery is 12%; if SPA is >3.8 mL, the probability of surgery is 71% (P < 0.0001). CONCLUSIONS Adding radiological characteristics such as presence of bony destruction on CT and size of SPA to the Chandler orbital cellulitis classification scheme increases the ability to more accurately predict which patients will require surgical intervention.