C-Reactive Protein As a Marker for Initiating Steroid Treatment in Children With Orbital Cellulitis

Purpose: To determine both the benefit of systemic steroids in pediatric patients with orbital cellulitis and to assess the usefulness of C-reactive protein (CRP) levels as a marker for starting steroids. Methods: Prospective, comparative interventional study. Pediatric patients aged 1 to 18 years admitted to a tertiary care children's hospital with a diagnosis of orbital cellulitis from October 2012 to March 2014 were included in the study. All patients were treated with intravenous antibiotics, and patients with subperiosteal abscess who met previously published criteria for surgical decompression underwent combined transorbital drainage and/or endoscopic sinus surgery. CRP was measured daily as a biomarker of inflammation, and when below 4mg/dl, patients were started on oral prednisone 1mg/kg per day for 7 days. Patients whose families did not consent to steroid treatment served as the control group. Patients were followed after discharge until symptoms resolved and all medications were discontinued. Results: Thirty-one children were diagnosed with orbital cellulitis during the study period. Of these 31 children, 24 received oral steroids (77%) and 7 did not (23%). There were 19 males and 5 females in the steroid group with an average age of 8.1 years, and 6 males and 1 female in the nonsteroid group with an average age of 7.1 years (p = 0.618). Thirteen patients (54%) in the steroid group and 2 patients (29%) in the nonsteroid group underwent sinus surgery with or without orbitotomy (p = 0.394). The average CRP at the onset of steroid treatment was 2.8mg/dl (range: 0.5-4). Patients who received oral steroids were admitted for an average of 3.96 days. In comparison, patients who did not receive steroids were admitted for an average of 7.17 days (p < 0.05). Once CRP was 4mg/dl, patients treated with steroids remained in the hospital for another 1.1 days, while patients who did not receive steroids remained hospitalized for another 4.9 days (p < 0.01). In the steroid group, 2 families reported increased hyperactivity in their children while on steroids. There was 1 case in each group of recurrence of symptoms after discharge from the hospital. Average follow-up time was 2.4 months in the steroid group and 2 months in the nonsteroid group (p = 0.996). At last visit, all patients returned to their baseline ophthalmic examination. There were no cases of vision loss or permanent ocular disability in either group. Conclusions: Our results give further evidence of the safety and benefit of systemic steroids in children with orbital cellulitis. Futhermore, this is the first study to suggest a standardized starting point (CRP 4mg/dl) and dosing schedule (oral prednisone 1mg/kg for 7 days) for children with orbital cellulitis. Patients who received systemic steroids after CRP dropped below 4mg/dl were discharged from the hospital earlier than patients who did not receive systemic steroids.