Construction of Preoperative Diagnostic Model of Periapical Cyst by Multiparameter Magnetic Resonance Imaging: A...

Due to the lack of a predictable and noninvasive method for distinguishing different types of apical periodontitis, it is difficult to estimate the actual prognosis of periapical cysts and the effectiveness of nonsurgical treatment for them. This pilot study aimed to establish a diagnostic model to differentiate periapical cysts and nonperiapical-cyst lesions based on parameters derived from magnetic resonance imaging (MRI).

Before endodontic microsurgery, MRI scans were performed for the patients with periapical lesions, including T2-weighted images (T2WI), noncontrast-enhanced T1-weighted images (T1WI), diffusion-weighted imaging (DWI) and contrast-enhanced T1-weighted images (T1WI + C). Sixteen parameters were collected, and thirteen derived parameters were calculated. The patients were divided into training and validation cohorts according to time order. The histopathological results were gold standards. Univariate, multivariate regression and multicollinearity analyses were used to filter variables and select predictors. Multivariate logistic regression was applied to develop the nomogram. Discrimination and calibration were evaluated by bootstrapping with 1000 resamples.

Eighty-four patients underwent MRI examinations followed by endodontic surgeries. Lesion length, enhanced area and enhanced-to-lesion area ratio were independent predictors for the diagnosis of periapical cysts. Area under the receiver operating characteristic (ROC) curve (AUC) of the training cohort was 0.861. Sensitivity and specificity based on ROC curves were 80.6% and 75.9%, respectively. The calibration curve of the nomogram was close to the ideal diagonal line. Decision curve analysis indicated positive net benefit in the model. The validation cohort proved the reliability of the prediction nomogram (AUC: 0.930; sensitivity: 81.8%; specificity: 76.9%).

The findings indicated that the proposed nomogram reliably identified periapical cysts through multiparameter MRI. The length of the lesion, enhanced area and enhanced-to-lesion area ratio were important variables for diagnosing periapical cysts. Multiparameter MRI was implicated as an efficient and noninvasive method to differentiate periapical lesions.

The study is registered on the Chinese Clinical Trial Registry website (ChiCTR2000035273).