2017 Sec 1 Green Book

Volume 137, Number 4 • Management of Orbital Fractures

at a high-volume ophthalmologic trauma referral center with the goal of analyzing outcomes and bet- ter understanding when intervention is warranted. PATIENTS AND METHODS Institutional reviewboardapprovalwasobtained to conduct a retrospective review of patients in the prospectively maintained pediatric trauma registry (0 to 18 years) at the Wilmer Eye Institute of Johns Hopkins Medical Institutions who were diagnosed with orbital fractures by International Classification of Diseases, Ninth Revision codes 802.4, 802.6, 802.7, and 802.8. The medical records were reviewed for each patient who was diagnosed with an orbital fracture between the years 2003 and 2013. Patients were excluded from this study if records did not include a full ophthalmologic examination imme- diately after presentation or if they did not follow- up after their injury. Data were extracted from each medical record for information about demograph- ics, mechanism of injury, physical examination at presentation, surgical intervention, and final out- comes. Patients were excluded if they had any facial fractures outside of the orbit, with the exception of nondisplaced zygoma and frontal bone fractures representing the continuation of orbital fractures. The t test and Mann-Whitney U test were used for statistical comparison of dichotomous groups for normally and nonnormally distributed vari- ables, respectively. Univariate logistic regression was used to examine the impact of individual fac- tors on the development of particular complica- tions, with multivariate regression for possible confounders. All statistical tests were two-sided, and significance was set to the level of p < 0.05. Demographics One hundred fifty patients met the selection criteria. The average patient age at the time of trauma was 12.6 ± 4.3 years. One hundred six- teen patients (77 percent) completed all recom- mended follow-up and were discharged from care, with an average follow-up time of 309 days. The majority of patients were male [110 (73.3 per- cent)], and the most common causes were sports (34 percent) and assault (32 percent) (Table 1). Fracture Characteristics One hundred twenty-seven patients (85 per- cent) had orbital fractures including the orbital floor. Fifty-eight (39 percent) had medial wall involvement, 12 (8 percent) included the roof, RESULTS

Table 1. Mechanisms of Injury ( n = 150)

No. (%) 50 (33.8) 47 (31.8) 18 (12.2) 13 (8.8) 10 (6.8) 5 (3.4) 5 (3.4)

Sports Assault

Play Fall

Motor vehicle accident

ATV/motorbike

Other

ATV, all-terrain vehicle.

and five (3 percent) had lateral wall involvement. The subset of patients without floor involvement ( n  = 23) was much less likely to undergo surgical management (30.4 percent versus 78.7 percent; p  < 0.001). Similarly, patients with evidence of extraocular muscle restriction on examination were much likelier to have a floor component of their orbital fracture (OR, 5.2; p  = 0.001). Twelve patients had extension of their fractures outside the orbit. Five had extension of a roof frac- ture into the frontal bone and four had zygoma involvement, all of which were nondisplaced. Three patients had nasoorbitoethmoid fractures: two with Markowitz-Manson type 1 fractures and one with a Markowitz-Manson type 2 fracture. On presentation, 43 patients (29 percent) showedevidenceof radiographicmuscleentrapment (i.e., herniation of a portion of the inferior rectus muscle belly into the defect) on computed tomog- raphy. Lack of diplopia and extraocular movement restriction on initial presentation were relatively sensitive for ruling out muscle entrapment on imag- ing (sensitivity, 95 percent; specificity, 45 percent; 95 percent CI, 84.2 to 99.4 percent). Twenty-two patients (15 percent) showed evidence of enoph- thalmos preoperatively, 19 of whom developed it in the acute fracture period. Of the remaining three patients, two were initially managed conservatively but developed enophthalmos, and one had surgery delayed because of ophthalmologic sequelae (i.e., commotio retinae and traumatic iritis). Surgical Cohort Overall, 107 patients (71 percent) underwent surgery for their fracture. Of these, 96 (90 per- cent) underwent repair within the acute period ( ≤ 3 weeks), whereas 11 (10 percent) underwent delayed fracture repair (22 to 1399 days; median, 49 days). Thirty-nine patients (37 percent) had surgery within the first 48 hours after their injury. Patients received surgical or conservative treat- ment based on their relative indications (Fig. 1), with the exception of the previously mentioned patients presenting with major eye injuries.

25