Radiographic predictors of visual outcome in orbital compartment syndrome

The goal of this study is to evaluate parameters on orbital CT as predictors of visual outcome in orbital compartment syndrome from retrobulbar hematoma or orbital cellulitis. The study will assess the assertion that certain patients are anatomically predisposed to vision loss in these conditions.

A retrospective chart review of consecutive patients with the diagnosis of either orbital cellulitis or orbital hematoma from the clinic of a single provider in an academic practice from 2006 to 2009 was performed. Exclusion criteria included preexisting vision loss, lack of CT scan for analysis, or lack of 1-month follow up for final visual acuity. Measurements of final visual acuity, medial wall length, lateral wall length, distance from the globe to the apex, and a novel measurement of posterior globe tenting (stretch angle) were obtained. Patients were divided into 2 groups: normal visual acuity and vision loss. Statistical analysis was performed to identify significant differences between the 2 groups.

The normal vision group consisted of 11 patients, all with vision of 20/30 or better. The average length of the medial and lateral wall was 43.9 and 41.6 mm, respectively. The average distance from the globe to the apex was 26.3 mm in the uninvolved eye and 30.3 mm in the involved eye, resulting in an average difference of 4.18 mm. The average stretch angle measurement was 28.9° in the uninvolved eye and 28.5° in the involved eye, resulting in an average difference of 0.41°. The vision loss group consisted of 4 patients, all with vision of count fingers or worse. The average length of the medial and lateral wall was 46.9 and 45.7 mm, respectively. The average distance from the globe to the apex was 32.2 mm in the uninvolved eye and 36.7 mm in the involved eye, resulting in an average difference of 4.50 mm. The average stretch angle measurement was 32.3° in the uninvolved eye and 21.1° in the involved eye, resulting in an average difference of 11.2°. The difference in stretch angle between eyes in the vision loss and normal vision groups were found to be highly significant with a p value of less than 0.001. The difference between the 2 orbits for globe to apex is a rough measurement of proptosis and was not statistically different in the 2 groups (p = 0.71), whereas the length from the globe to the apex in the uninvolved eye was statistically different between the vision loss group and normal vision group (p = < 0.001).

Orbital compartment syndrome is a potentially vision-threatening condition. Minimal objective data are currently available in the literature to guide physicians in making clinical judgments regarding these patients. The results of this study indicate that comparing the novel stretch angle between the patient's 2 eyes and measuring the length from the globe to the orbital apex can help identify patients at risk for poor visual outcome. This study provides objective measurements that can aid ophthalmologists and radiologists in determining the relative threat to vision in patients presenting with orbital compartment syndrome from orbital cellulitis or retrobulbar hematoma. Hopefully, the data can help select patients that may benefit from more aggressive intervention and will ultimately result in superior visual outcomes.