Is MRI a viable alternative to CT/CBCT to identify the course of the inferior alveolar nerve in relation to the roots of the third molars?

Abstract

Objectives To assess the reliability of judging the spatial relation between the inferior alveolar nerve (IAN) and mandibular third molar (MTM) based on MRI or CT/CBCT images. Methods Altogether, CT/CBCT and MRI images of 87 MTMs were examined twice by 3 examiners with different degrees of experience. The course of the IAN in relation to the MTM, the presence/absence of a direct contact between IAN and MTM, and the presence of accessory IAN were determined. Results The IAN was in > 40% of the cases judged as inferior, while an interradicular position was diagnosed in < 5% of the cases. The overall agreement was good (κ = 0.72) and any disagreement between the imaging modalities was primarily among the adjacent regions, i.e., buccal/lingual/interradicular vs. inferior. CT/CBCT judgements presented a very good agreement for the inter- and intrarater comparison (κ > 0.80), while MRI judgements showed a slightly lower, but good agreement (κ = 0.74). A direct contact between IAN and MTM was diagnosed in about 65%, but in almost 20% a disagreement between the judgements based on MRI and CT/CBCT was present resulting in a moderate overall agreement (κ = 0.60). The agreement between the judgements based on MRI and CT/CBCT appeared independent of the examiner’s experience and accessory IAN were described in 10 cases in MRI compared to 3 cases in CT/CBCT images. Conclusions A good inter- and intrarater agreement has been observed for the assessment of the spatial relation between the IAN and MTM based on MRI images. Further, MRI images might provide advantages in the detection of accessory IAN compared to CT/CBCT. Clinical relevance MRI appears as viable alternative to CT/CBCT for preoperative assessment of the IAN in relation to the MTM.

Comparison of Preoperative Cone-Beam Computed Tomography and 3D-Double Echo Steady-State MRI in Third Molar Surgery

“…It was necessary to correct the fusion manually through trial and error to allow the best possible evaluation. Therefore, we can only confirm the need for a perfect automatic fusion algorithm from other studies [ 23 ]. However, compared to the MRI protocols used in other studies, we achieved a much better image quality and more accurate evaluation using 3D-DESS MRI protocol.…”

“…This study confirms the trends observed by Deepho et al that there is a tendency for the “small pattern” (21%) to be more prevalent in the retromolar region. Nevertheless, it should be noted that the “same pattern” was most frequent in the retromolar (72%) and molar (76%) regions [ 23 ]. The large pattern was observed in only 7% of the cases, which is most likely explained by insufficient fusion or insufficient MRI image quality due to motion artifacts.…”

“…Additionally, the continuous detectability of the delimitation of the MC’s osseous cortical boundaries in the CBCT and the high signal intensity structure of the IAN in MRI were assessed for the retromolar and molar regions. According to the evaluation method of Deepho et al [ 23 ], the MC was considered as visible if its osseous cortical boundaries were seen as a circle on the cross-sectional coronal reconstruction and as invisible if a part or even the complete osseous cortical boundaries was missing. The same evaluation scale was also applied for the DESS images with the IAN to be considered visible if the IAN’s high signal intensity course could be followed continuously and as invisible if an interruption in the course of the IAN could be detected.…”

“…To investigate whether there are MRI related distortions and how the occupation of the MC by the IAN’s hyperintense signals varies over the defined regions of interest, the following three patterns were determined by both readers according to Deepho et al [ 23 ]: same pattern, if the area within the MC was occupied by the presence of IAN MRI signal hyperintensities; small pattern, if the area within the MC was only partially occupied by the presence of IAN MRI signal hyperintensities; and large pattern, if the IAN was sticking out of the MC area ( Figure 5 ).…”

“… To investigate the occupation of the mandibular canal (MC) by the inferior alveolar nerve’s (IAN) hyperintense signals, the following three patterns were determined according to Deepho et al [ 23 ]: ( A ) small pattern, if the area within the MC was only partially occupied by the presence of IAN MRI signal hyperintensities; ( B ) same pattern, if the area within the MC was occupied by the presence of IAN MRI signal hyperintensities; and ( C ) large pattern, if the IAN was sticking out of the MC area. …”