Novel strategy to predict high risk of inferior alveolar nerve injury during extraction of lower third molars based on assessment of computed tomographic images of multiple anatomical features.

Preoperative assessment is essential to prevent inferior alveolar nerve (IAN) injury during surgical extraction of the lower third molar (LM3). Here, we aimed to establish an assessment system to predict IAN injury during surgical extraction of the LM3. We conducted a retrospective cohort study on 115 patients diagnosed as 'high-risk' based on our previous risk assessment method involving three anatomical features of the inferior alveolar canal using computed tomographic (CT) images. We evaluated the occurrence of neurosensory impairment in these high-risk patients, and its association with novel anatomic features based on CT images. Neurosensory impairments were observed in 19 patients (16.5%). The inferior alveolar canal major diameter (p < 0.0001) and lingual bone thickness (p = 0.0039) were significantly associated with the occurrence of neurosensory impairment during LM3 extraction. Receiver operating characteristic curves were used to determine cut-off values of these quantitative factors to specifically predict IAN injury. Preoperative risk assessment with quantitative factors based on anatomical features observed on CT images may facilitate more appropriate surgical planning for patients at a high risk of IAN injury.