Radiographic Predictors of Postoperative Inferior Alveolar Nerve Injury in Mandibular Third Molar Surgery.

Background: Cone-beam computed tomography (CBCT) provides additional 3-dimensional information on the relationship between the mandibular third molar (M3M) and the inferior alveolar nerve (IAN). As such, CBCT is being increasingly utilized in preoperative M3M assessment.

Purpose: The purpose of the study was to compare the radiographic findings on panoramic and CBCT and their association with postoperative IAN paresthesia.

Study design, setting, sample: We conducted a retrospective cohort study in a sample of patients referred to the Department of Oral and Maxillofacial Surgery at the Royal Dental Hospital of Melbourne, for management of impacted M3Ms. Patients were included in this study if they had 1 or more high-risk findings on orthopantomogram (OPG), had both OPG and CBCT imaging taken and if at least 1 M3M had been extracted. Subjects were excluded from this study if their M3Ms were not extracted or if a CBCT was not indicated.

Independent variable: The independent variable was radiographic features identified on OPG (Rood and Shehab's radiographic signs, root morphology, impaction type, Pell and Gregory classification) and CBCT (number and location of roots, severity of IAN compression and the presence of ankylosis).

Main outcome variable: The outcome variable was postoperative neurosensory function defined as any reported paresthesia at 2-week postprocedural review.

Covariates: The covariates were patient demographic information (age, sex).

Analyses: Variables were initially assessed with univariate logistic regression analysis to determine factors related to developing postoperative paresthesia. Multivariate logistic regression analysis was then used to assess the association between positive univariate study variables and postoperative paresthesia, while adjusting for potential confounders. Covariates were assessed using an omnibus likelihood ratio test and included if they were statistically significant at the P < .05 level.

Results: The study sample consisted of 257 subjects who underwent surgical removal of n = 386 M3Ms. The mean age was 25.9 (standard deviation = 7.05). The panoramic features of narrowing of canal, diversion of canal and dark and bifid roots were identified as statistically significant associations of postoperative paresthesia. No CBCT features were significantly associated of postoperative paresthesia.

Conclusion and relevance: Certain panoramic features, along with patient age, are significant predictors of IAN paresthesia. CBCT findings were not significant predictors upon multivariate analysis, thus additional CBCT imaging did not significantly improve ability to predict paresthesia.