Can CT predict the development of oroantral fistula in patients undergoing maxillary third molar removal?

In maxillary wisdom tooth extraction, the necessity of CT is unknown. The purpose of this study was to investigate whether CT adding to orthopantomography is useful for predicting oroantral perforation during maxillary third molar extraction.

Various risk factors for oroantral perforation during maxillary third molar extraction were investigated by univariate and multivariate analyses. We analyzed those of all patients and the patients who underwent CT, respectively. The proximity of the roots to the maxillary sinus floor (root-sinus [RS] classification) and Archer classification were assessed using panoramic radiography. The number of roots and vertical relationship were assessed using CT.

A total of 604 out of 3299 patients underwent CT adding to orthopantomography. In all cases, multivariate analyses except for CT findings showed that the RS classification type III/IV and the Archer classification Type B/C/D in panoramic findings were significantly correlated with oroantral perforation as radiological findings. In cases for which CT was performed, multivariate analyses showed that one root (OR 12.87) and the vertical relationship Type D (OR 5.63) in CT findings, besides the RS classification type III/IV (OR 4.47) in panoramic findings, were significantly related to oroantral perforation.

The RS classification and the Archer classification in panoramic findings can predict the risk of oroantral perforation. The usefulness of CT adding to orthopantomography is limited. However, when the relationship between the upper wisdom tooth and maxillary sinus floor (RS classification) is unclear, to check whether the number of roots is one and the apex of one root is projecting into the maxillary sinus in CT findings, is useful for the prediction.