Accuracy of Cone Beam CT Measurements for Assessment of Posterior Mandible: A Cadaver Study

Objectives: This study is aimed at analysing the accuracy and reliability of the cone beam CT (CBCT) measurements and direct physical measurements of the posterior mandible. Materials and methods: Eighteen cadaveric hemi-mandibles were dissected from the soft tissues and the CBCT images of the mandibles were taken. Direct physical and cone beam CT measurements of six landmarks which includes height of ramus (R), distance of lingula to sigmoid notch (LS), distance of lingula to inferior border (LI), position of lingula in relation to occlusal plane (L-OP),  ramus thickness at crestal level (RT-C), and ramus thickness at midway between sigmoid notch and lingula (RT-M) were determined. Accuracy and reliability of the measurements were tested. Results: Four landmarks showed high accuracy when measuring the posterior mandible, while two landmarks, LI and RT-M, showed statistically significant weaker accuracy (p<0.05). Inter-reliability were good for all landmarks when measured directly physically on mandibles (ICC>0.7 and p>0.05), but were low on two landmarks, LI and RT-C, on CBCT measurement (ICC<0.5 and p<0.05). Conclusion: A generally strong accuracy between direct physical and CBCT measurements for most landmarks on posterior mandible were found. Reliability between two researchers were high on direct physical measurements. Meanwhile, two landmarks on CBCT which include LI and RT-C showed low inter-reliability. Hence, CBCT measurements proved to be a good tool for pre-operative assessment, since high inter-reliability and strong accuracy corresponding to direct physical were recorded.