Refining Indications for Orbital Floor Fracture Reconstruction: A Risk-Stratification Tool Predicting Symptom Development and Need for Surgery

BACKGROUND: Challenges in orbital floor fracture management include delayed symptom onset and controversial surgical indications based on radiographic findings. This study assessed which imaging characteristics most reliably predict symptomatology to generate a tool quantifying individual need for surgery on initial presentation. METHODS: The clinical course for all patients with isolated orbital fractures at a single institution from 2015 to 2017 were reviewed. Trauma mechanism, computed tomographic scan findings, and symptoms necessitating surgery (diplopia, enophthalmos) were noted. Univariable and multivariable regression modeling was used to generate a predictive risk model for operative fractures. RESULTS: One hundred twenty-one patients with isolated orbital fractures were identified. Mechanism of injury included falls (41 percent), assault (37 percent), and vehicular trauma (17 percent). Patient follow-up averaged 4.4 ± 4.8 months. Average orbital floor fracture area was 2.4 cm2 (range, 0.36 to 6.18 cm2), and orbital volume herniation averaged 0.70 cm3 (range, 0.01 to 4.23 cm3). Twenty-one patients (17.3 percent) required surgical intervention for symptomatic fractures. The strongest predictors of symptoms were orbital volume increase greater than 1.3 cm3 (OR, 10.5; p = 0.001) and inferior rectus displacement within/below the fracture line (OR, 3.7; p = 0.049). Mechanical fall was risk-reducing (OR, 0.08; p = 0.005). Symptom risk was stratified from low (3.6 percent) to high risk (71 percent) (C-statistic = 0.90). The volume of herniated orbital contents was significantly more predictive of symptoms than fracture area (C-statistic = 0.81 versus C-statistic = 0.66; p = 0.02). CONCLUSIONS: The proposed risk tool allows highly accurate, early prediction of symptomatic orbital floor fractures. Findings suggest that orbital volume change, not fracture area, more reliably informs operative indications, along with inferior rectus muscle caudal malposition. A simplified stepwise decision algorithm demonstrates the potential utility of this risk-assessment tool. CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, III.