Can Clinical Findings Predict Orbital Fractures and Treatment Decisions in Patients With Orbital Trauma? Derivation of a Simple Clinical Model

J Craniofac Surg. 2017 Oct;28(7):e661-e667. doi: 10.1097/SCS.0000000000003823.


The aim of this study was to determine the predictive value of 4 clinical signs (periorbital ecchymosis, periorbital emphysema, diplopia, and inferior orbital nerve hypoesthesia) for the diagnosis of orbital fractures (OFs) and the need for surgical treatment in patients with orbital trauma. The investigators designed and implemented a retrospective cohort study composed of patients with orbital trauma. The primary and secondary outcomes were respectively the diagnosis of OFs and the need for surgical treatment. Multivariable logistic regressions including the clinical signs, age, and causes of injury were used to determine the independent contribution of each clinical sign to the prediction of the outcomes and to obtain weights to compute OF and surgery scores. The outcomes were assessed by receiver-operating characteristic (ROC) curves, sensitivity, specificity, and positive and negative predictive values. A total of 912 patients were included. All clinical signs except periorbital ecchymosis were significantly associated with OFs and the need for surgical treatment (P < 0.001). The predictive power of each clinical sign taken separately was moderate for the 2 outcomes (area under ROC curve [AUC] <0.7). A better predictive value was found when all clinical signs were used together (AUC >0.7). Patients with an OF score >3 were likely to have an OF and patients with a surgery score ≤2 were unlikely to have surgery.The present study demonstrated that our OF and surgery scores resulted in an effective model that allowed the stratification of patients with orbital trauma based on their risk of having OFs and risk of needing a surgical treatment.

MeSH terms

  • Decision Support Systems, Clinical*
  • Diplopia / etiology*
  • Humans
  • Models, Statistical*
  • Orbital Diseases / etiology*
  • Orbital Fractures / complications
  • Orbital Fractures / diagnosis*
  • Orbital Fractures / epidemiology
  • ROC Curve
  • Sensitivity and Specificity