Reliability of Interproximal Bone Height Measurements in Bone- and Tissue-Level Implants: A Methodological Study for Improved Calibration Purposes.

Abstract

To evaluate the interobserver and intraobserver agreement between prosthodontists when measuring interproximal peri-implant bone levels from digital periapical radiographs and to introduce a radiographic quality index for periapical imaging assessment. Periapical radiographs of 122 single implants in the anterior and posterior regions with two categories of imaging quality (ie, optimal and suboptimal) were assessed. Six prosthodontists were asked to linearly measure the distance from the first bone-to-implant contact to the implant platform/shoulder (DIB) using an image processing program (ImageJ 1.48u4, NIH). The procedure was repeated after 3 to 4 weeks. Interobserver and intraobserver agreements were evaluated by intraclass correlation coefficient and kappa. A radiographic quality index developed for periapical imaging assessment has been introduced in this study. Each implant was classified into two categories according to the implant type and the quality of the radiographic image. There were significant interobserver differences (P < .001). Most of the discrepancies between repeated measures were below 0.5 mm (range: 0.37 [SD ± 0.76] to 0.55 [SD ± 0.68]). The interobserver and intraobserver agreements on the bone-level values were "fair to moderate" regardless of the implant type and radiographic quality. With optimal image quality in tissue-level images, "substantial agreement" could be achieved. There was no significant effect of the implant level type (P = .973). Image quality, as well as the interpreter, influenced the measurements' reproducibility by prosthodontists. Bone height assessments at bone-level implants seem to be slightly more variable compared with tissue-level assessments; however, there were no significant differences. Specific guidelines on how to estimate DIB for calibration purposes in the case of suboptimal radiographic image quality and how to obtain optimal images need to be developed. The radiographic quality index for periapical imaging assessment is expected to be adopted in future studies.