D70. Development and Validation of a Risk Calculator for Enophthalmos Following Orbital Fracture Repair

This is the authors' abstract. We don't add key points for this paper.

PURPOSE: Enophthalmos is an undesired complication of orbital fracture repair (OFR). Risk factors for postoperative enophthalmos remain unclear. We aim to develop and validate a risk calculator for better prediction of this outcome. METHODS: We retrospectively reviewed adult trauma patients who underwent OFR at two centers from 2015 to 2022. Patients with postoperative follow-up <2 weeks were excluded. Our primary outcome was incidence of postoperative enophthalmos at ≥2 weeks following OFR. The C-statistic (area under the ROC curve) and Hosmer-Lemeshow tests were used to assess the risk model accuracy. RESULTS: Of 501 patients with OFR, 349 met the inclusion criteria. Twenty-seven (7.7%) patients had postoperative enophthalmos. The risk model was developed and partially validated using patient data from 2015 to 2019 (n=253), and fully validated from 2020 to 2022 (n=96). Risk factors for postoperative enophthalmos were: older age at injury, preoperative enophthalmos, medial wall involvement, near-total wall defects, and delayed surgical repair after 1 week from injury. There was an incremental increase in the odds of postoperative enophthalmos with increased duration from injury to surgery. Our risk model showed very good discrimination (C-statistic=84.5%) and calibration (P=0.83). It was highly accurate in predicting enophthalmos using the validation dataset (C-statistic=0.89). For a risk cut-off of 9.5%, the sensitivity and negative predictive value were 85% and 99%, respectively. CONCLUSION: We designed the first validated risk calculator that can be used to rule out enophthalmos following OFR. Our risk calculator enables surgeons to preoperatively predict the absence of postoperative enophthalmos in 99% of cases.