Three‐Dimensional Condylar Subtyping and Its Association With Obstructive Sleep Apnea Severity in Adults: Implication for Craniofacial Skeletal Phenotype
Min‐Ji Jeon, Kun‐Woo Park, Yoon Jeong Choi, Yoon‐Goo Kang, Hyo‐Won Ahn, Su‐Jung KimABSTRACT
Objectives
To investigate the association between three‐dimensional (3D) condylar resorption pattern and obstructive sleep apnea (OSA) severity in adults, and to explore its structural implications within the craniofacial skeletal phenotype (CSP) framework.
Methods
Condylar morphology was classified using a two‐axis system integrating the degree of resorption and cortical remodelling pattern. Patients were grouped into no resorption (NR), adaptive remodelling (AR), and severe resorption (SR) categories. Dentoskeletal, temporomandibular joint (TMJ), and polysomnographic variables were compared among groups. Multivariable regression analyses were adjusted for age, sex, and body mass index, and hierarchical regression was used to examine the contribution of condylar grouping beyond CSP.
Results
Severe condylar resorption was identified in 23.9% of the sample and was more prevalent (54.0%) in patients with CSP features. The SR group demonstrated greater skeletal discrepancy and higher apnea–hypopnea index (AHI), respiratory disturbance index (RDI), and arousal index (ArI) values (all, p < 0.05). After adjustment for age, sex, and body mass index and false discovery rate correction, SR remained significantly associated with higher AHI, RDI, and ArI, whereas TMJ pain history, condylar angulation, and oxygenation‐related indices showed no significant associations. In hierarchical models, the overall incremental contribution of condylar grouping beyond CSP was not significant after multiple‐testing correction; however, SR remained significantly associated with higher AHI and RDI after additional adjustment for CSP (both, FDR‐adjusted p = 0.045).
Conclusions
Severe condylar resorption was associated with greater skeletal discrepancy and higher OSA burden. Three‐dimensional TMJ assessment may complement conventional skeletal phenotyping in adult OSA.