335 Development and Validation of a Predictive Score for Persistent Instability Following Conservative Treatment in Type II Odontoid Fractures

Abstract

INTRODUCTION: Odontoid fractures are common cervical spine fractures, however significant controversy exists regarding their treatment. While risk factors for failure of conservative therapy have been identified, no predictive risk score has been developed to aid in decision making regarding surgical intervention or conservative therapy. METHODS: A retrospective review was conducted of all patients evaluated at a level one trauma center from 2005 to 2021. Patients identified with Type II Odontoid Fractures as classified by the Anderson D’Alonzo Classification system who were treated with external orthosis were included in analysis. Patients were considered to have failed conservative therapy if they were offered surgical intervention due to nonunion and continued motion on flexion extension x-rays. A machine learning method (RiskSLIM) was then utilized to created a risk stratification score to identify patients at high risk for requiring surgical intervention due to persistent instability. RESULTS: In total, 138 patients were identified as presenting with type II odontoid fractures that were treated conservatively. Of these, 38 patients demonstrated persistent instability following conservative therapy with cervical orthosis. The Odontoid Fracture Predictive Model (OFPM) was created using a machine learning algorithm with a five-fold cross validation area under the curve of 0.7389 (95% CI: 0.671 to 0.808) and calibration error of 18.6%. Predictive factors were found to include anterior or posterior displacement, displacement greater than 5 mm, comminution at the fracture base, and history of smoking. The probability of persistent instability was <5% with a score of 0 and 88% with a score of 5. CONCLUSIONS: The OFPM model is a quick and accurate tool to assist in clinical decision making in patients with type II odontoid fractures.