Longitudinal reliability of upper airway assessment using cone-beam computed tomography: a retrospective study

To explore the consistency of upper airway measurements derived from cone-beam computed tomography (CBCT) images in adults without intervening therapy. CBCT images of 30 patients who underwent microscrew implantation or individual tooth restoration were retrospectively analysed. Scans were acquired at 2 time points using an indentical standardized protocol, with no intervening treatment affecting the dentofacial complex. The volumes and heights of the total upper airway, and the nasopharyngeal, palatopharyngeal, glossopharyngeal, and hypopharyngeal segments, along with cranio-cervical angle at the C2, C3 and C4 levels, were measured by 2 independent examiners. Intra- and inter-examiner reliability were assessed using intraclass correlation coefficients (ICCs) and interclass correlation coefficients, respectively, with 95% confidence interval (CI). Intra- and inter-examiner reliability was good to excellent for all measured parameters (ICCs > 0.80). Significant differences between time points were observed for nasopharyngeal volume (6577.79 ± 2061.56 mm3 vs. 6942.93 ± 1836.83 mm3, P = 0.013), total airway height (73.79 ± 9.12 mm vs. 71.42 ± 7.04 mm, P = 0.024) and hypopharyngeal height (13.64 ± 2.44 mm vs. 12.55 ± 2.93 mm, P = 0.023). No significant differences were observed for other volumetric or linear airway parameters, nor for cranio-cervical angles (P > 0.05). When a standardized scanning protocol is applied, CBCT provides good longitudinal consistency for most upper airway dimensions in untreated adults. However, nasopharyngeal volume and airway height measurements demonstrate greater temporal variability and should be interpreted with caution in longitudinal assessments.