Accuracy of cone-beam computed tomography and periapical radiography in apical periodontitis diagnosis

This study aimed to investigate the correlation and the agreement between periapical radiography (PR) and cone-beam computed tomography (CBCT) correlating to histologic findings in the diagnosis of apical periodontitis (AP).

One hundred thirty-four premolar root canals from 10 dogs were treated after AP induction. Four months later, the animals were killed, and standard digital PRs were obtained. The area of AP was measured by using ImageJ software. CBCT (i-CAT) images from each arch were obtained, and AP area and volume were measured by using Osiri-X software. The apical inflammatory infiltrate was evaluated under light microscopy. The correlation between imaging methods was evaluated by using the Pearson coefficient. The Bland-Altman method was used to assess the agreement between PR and CBCT data. The Spearman coefficient was used to correlate the imaging data and histologic findings.

Despite a strong correlation between PR and CBCT areas, the agreement limits were very broad (95% limits of agreement, 0.19-1.08). PR only measured, on average, 63% of CBCT values. Although there was a strong correlation between PR area and CBCT volume, the Bland-Altman method suggests that the larger the CBCT volume, the more underestimated the PR value. When APs had a volume smaller than 6 mm(3), the PR estimation of CBCT data was unpredictable. A positive correlation was found for PR area, CBCT area, CBCT volume, and histology data.

The diagnosis of AP based on PR data is clinically limited, and it should not be used for scientific investigations.