The aim of this study was to evaluate if the configuration of canal orifices can predict the presence of a second mesiobuccal canal (MB2) in maxillary second molars with fused roots.
Maxillary second molars with fused roots (n=150) were scanned in a micro-CT device (pixel size: 9 μm) and evaluated regarding the root fusion type and the incidence of MB2 canal. The centers of the canal orifices were connected and the angles formed by the intersection of these lines, as well as, their distances were measured and statistically compared (1-way ANOVA). Specimens were then assigned to 2 groups according to the presence (n=65) or absence (n=50) of the MB2 canal and compared regarding the angles and interorifice distances using the Welch’s t-test. A binary logistic regression model estimated the association between the interorifice distance, the fusion type, and the presence of MB2, while ROC curve analysis was performed to evaluate the diagnostic abilities of significant variables. Significance level was set at 5%.
Most common fusion types were 1 and 6, and MB2 canal incidence was 47.3%. No statistical difference was observed in the angles or interorifice distances among teeth with different fusion types (P > .05). Distances from mesiobuccal (MB) to distobuccal (DB) and palatal (P) canal orifices, as well as, MB and P angles, were significantly higher in the presence of the MB2 (P < .05). Binary logistic regression model revealed that MB-DB and MB-P distances were significant in the presence of the MB2 canal (P < .05). ROC curve analysis indicated that MB-P distance had acceptable diagnostic accuracy for the prediction of MB2 presence. Sensitivity (specificity) calculated from the MB-P distances of 4.0, 4.5 and 5.0 mm were 89.2% (36.5%), 76.9% (58.4%) and 44.4% (71.4%), respectively.
MB-P interorifice distance is a high predictor of the presence of MB2 canal in maxillary second molars with fused roots.

Author