Agreement between 2D and 3D radiographic outcome assessment one year after periapical surgery

To compare 2D with 3D radiography in assessing the treatment outcome 1 year after periapical surgery.

In this prospective study, periapical radiographs (PA) and cone beam computed tomography (CBCT) were performed 1 year after periapical surgery. Three calibrated observers independently evaluated the radiographs for the presence and type of periapical radiolucencies. Ratings in PA were compared to those in bucco-lingual and mesio-distal CBCT images (coronal and sagittal planes), and the ratings of the latter two were also compared between each other. Further, maximum size diameters of radiolucencies were measured on CBCT scans, and the calculated means were correlated with the types of radiolucency. Statistical analysis was completed using Friedman rank sum tests, the Wilcoxon signed rank test and the Pearson correlation coefficient.

A total of 61 roots in 54 patients were eligible for the final assessment. On average, the intra-observer ratings were identical in 59.6% when comparing PA and CBCT (kappa 0.112 to 0.192). A very high intra-observer agreement (93.4%) was noted when comparing bucco-lingual and mesio-distal CBCT ratings (kappa 0.797 to 1). Interobserver agreement was higher for PA (68.8%) than for CBCT (bucco-lingual 45.9%, mesio-distal 47.5%), but without reaching significant differences. The calculated mean size of persistent radiolucencies in CBCT scans correlated well with the assigned types of radiolucency.

CBCT images showed in nearly a third of the evaluated cases a worse situation than PA. There is a need to define criteria to assess the 'radiographic healing' in CBCT following periapical surgery.