Accuracy of cone-beam computed tomography and periapical radiography in detecting small periapical lesions

The aim of this study was to determine the diagnostic accuracy of 2 cone-beam computed tomography (CBCT) machines and periapical (PA) radiography in detecting simulated apical lesions created with the smallest dental burs available.

By using mandibles from human cadavers, simulated apical lesions were created and then progressively enlarged in 16 roots by using sizes #1/4, #1/2, #1, #2, #4, and #6 round burs. Imaging was obtained after each enlargement with a Kodak 9000 3D (Kodak) CBCT, a Veraviewpocs 3De (Morita) CBCT, and intraoral digital PA radiography. Specificity and sensitivity at variable decision thresholds were calculated and plotted on receiver operator characteristic curves. The area under the curve (AUC) served as an estimate of diagnostic accuracy.

The overall AUCs for Kodak, Morita, and PA radiography were 0.767 (95% confidence interval [CI], 0.743-0.792), 0.753 (95% CI, 0.728-0.779), and 0.584 (95% CI, 0.554-0.615), respectively. The AUCs for Kodak and Morita were both statistically significantly larger than the AUC for all corresponding simulated lesion sizes imaged with PA radiography. Between Kodak and Morita, there were no statistically significant differences in AUCs for any of the corresponding simulated lesion sizes.

Both CBCT devices demonstrated poor accuracy in detecting simulated lesions smaller than 0.8 mm in diameter, fair to good accuracy when simulated lesion diameter was between 0.8-1.4 mm, and excellent accuracy when simulated lesions were larger than 1.4 mm in diameter. PA radiography, at best, demonstrated poor diagnostic accuracy for all simulated lesion sizes.