Background Preseptal cellulitis can be difficult to distinguish from orbital cellulitis in children. The majority of patients with periorbital infections are admitted for intravenous antibiotics. This study aimed to investigate the risk of missing orbital cellulitis and the outcomes of missed patients.

Methods A prospective cohort study of children aged 3 months to 18 years diagnosed with preseptal cellulitis over 5 years. Data were collected prospectively, including demographics, clinical features and outcomes.

Results There were 216 children diagnosed with preseptal cellulitis. 75 (35%) were treated with oral antibiotics and 141 (65%) with intravenous antibiotics. 5 (2%) children who were hospitalised were subsequently determined to have orbital cellulitis. All 5 children were either a young infant with difficult eye examination, or had headache or vomiting.

Conclusion The risk of missing orbital cellulitis is low. Young infants with difficult eye examination or the presence of headache or vomiting should increase suspicion of orbital cellulitis.