Risk factors observed in 2-dimensional radiographs for permanent injury of the inferior alveolar nerve after removal of mandibular third molars: a case-control study.

OBJECTIVE: To assess whether differences exist in signs observed in 2D radiographs of mandibular third molars between a case group of patients with and a control group without permanent sensory disturbance of the inferior alveolar nerve (IAN) after removal. STUDY DESIGN: Three observers blinded to patient status assessed radiographs from the case group (n=162) and the control group (n=172). Two new signs, craniocaudal relation of the roots and the mandibular canal and position of the canal over the roots; and 4 "classic" signs, interruption of the white borders of the canal, darkening of the roots, narrowing of the canal lumen, and diversion of the canal over the roots were registered. Chi-square tests assessed differences in distribution of radiographic signs between the groups. Odds ratios expressed the association between radiographic signs and permanent sensory disturbance. Inter- and intraobserver reliability values were calculated. RESULTS: We found significantly more teeth with roots positioned inferiorly to the canal borders (P<0.001; OR 4.1-5.3) and with the canal superimposed over the upper or middle third of the roots (P<0.001; OR 2.6-3.9) in the case group than in the control group. Inter- and intraobserver reproducibility was excellent for roots inferior to the canal borders and fair to good for canal superimposition. CONCLUSIONS: Two radiographic signs are valid predictors of permanent sensory disturbance of the IAN in 2D radiographs.