Back to Annual Meeting Program
Twenty Years of Pediatric Orbital Roof Fractures: Incidence, Patterns and Indications for Operation
Devin Coon, M.D., Nance Yuan, BS, Danielle Jones, BS, Lori K. Howell, MD, Michael P. Grant, MD PhD, Richard J. Redett, MD.
Johns Hopkins University, Baltimore, MD, USA.
PURPOSE: Little data is available regarding patterns or sequelae of pediatric orbital roof fractures. In particular, long term outcomes and indications for operative intervention remain unclear. We sought to examine a large cohort to provide evidence based guidelines for the management of these injuries.
METHODS: IRB approval was obtained to review pediatric trauma patients presenting to Johns Hopkins Hospital over a twenty year period from 1991 to 2011. All patients with an ICD-9 diagnosis indicating orbital or anterior skull base fractures underwent review to determine orbital roof involvement. Patients with roof fractures were assessed for demographics, management and outcomes.
RESULTS: A total of 1128 pediatric patients had diagnoses of an orbital or anterior skull base fracture. After review, 134 patients (12%) were found to have orbital roof fractures. The average age was 5.7±4.1 years old, and 83 patients (62%) were male versus 38% female (p<0.01). Unilateral orbital roof fractures were seen in 117 patients (87%) while seventeen had bilateral fractures.
The most common cause was fall (40%), followed by motor vehicle collision (24%) and pedestrian struck (17%). Eleven patients (18%) had concomitant ocular injuries while 70 (54%) had intracranial injuries. Eight patients (6%) underwent orbital roof repair. The only complication was lid ptosis in a patient with a large frontal bone defect.
Sixty-one patients had PACS-available maxillofacial CTs with three-plane reconstruction. Most fractures entered through the supraorbital rim and exited the orbital apex. The fracture pattern was exclusively vertical in 57% and horizontal-only in 25%. Horizontal displacement was more common (87%, mean 2.3mm) than vertical (44%, mean 2.8mm). Thirty-six percent of fractures were comminuted (average surface area 4cm2).
Ninety-eight patients returned after discharge with an average followup time of 17 months. Two patients (95% CI 0.2-5.4%) developed vertical dystopia at three weeks and ten months respectively. Both had nonoperative management of comminuted fractures (Fig 1; 5.6% of comminuted fractures developed dystopia versus 0% of linear; p=0.06). All five patients who developed CSF leaks had comminuted fractures (p<0.01). Falls were more likely to result in linear fractures (p=0.03) and no fractures resulting from a fall required surgical repair (p=0.02). No patients developed encephalocele.
CONCLUSIONS: In the first large cohort study to examine pediatric orbital roof fractures, we found that the most common pattern is a non- or minimally displaced vertical fracture from the supraorbital rim to orbital apex. The majority of fractures can be managed conservatively with good long term outcomes. Vertical dystopia occurs in less than 5% of patients and was only seen in large comminuted fractures. Comminuted fractures with a surface area over 1.5 cm2 represent high risk fractures that must be closely followed for possible intervention based on the development of sequelae such as CSF leakage, vertical dystopia or encephalocele.
Figure 1. A one year old boy who sustained a comminuted orbital roof fracture (left) and developed vertical dystopia after ten months (right).
Back to Annual Meeting Program