Anthropometric 3D evaluation of the face in patients with sleep related breathing disorders.

ABSTRACT

PURPOSE: To evaluate craniofacial measurements on 3D-stereophotogrammetry and see if particular measurements are more typical in obstructive sleep apnea (OSA) and have a correlation with its severity. METHODS: Subjects included were adults undergoing a diagnostic polysomnography. Age, BMI, neck, abdominal and hip circumference (cm) were recorded. 3D-stereophotogrammetry was performed and landmarks were placed on the 3D-image. Different linear, angular and volume measurements were performed to gauge facial and neck anatomy. The relationship between these measurements and the severity of OSA, based on the obstructive apnea/hypopnea index (OAHI, events/h), was assessed by multiple linear regression, and adjusted for BMI and sex. RESULTS: Of 91 subjects included (61 male), mean age was 46 ± 12 years, BMI 30.1 ± 6.5 kg/m2, OAHI 19.3 ± 18.8/h. BMI was higher (p = 0.0145) in females (32.9 ± 7.7) than in males (28.6 ± 5.3). This was also true for hip circumference (118 ± 15 vs 107 ± 10, p = 0.0006), while the neck circumference was higher (p < 0.0001) in males (41 ± 4 vs 37 ± 4). The following parameters could predict the logOAHI (r2-adjusted = 0.51) sex (p < 0.0001), BMI (p = 0.0116), neck-depth/mandibular-length (p = 0.0002), mandibular-width angle (p = 0.0118), neck-depth euclidean distance/surface distance (E/S) (p = 0.0001) and the interaction terms between sex and neck-depth/mandibular-length (p = 0.0034), sex and neck-depth E/S (p = 0.0276) and BMI and neck-depth E/S (p = 0.0118). The interaction between sex and neck-depth/mandibular-length showed a steeper linear course in females. This is also true for the interaction term BMI with neck-depth E/S in patients with a higher BMI. With a same neck-depth ratio, the OAHI is larger in men. CONCLUSION: Measurements involving the width of the face and addressing the soft tissue in the upper neck were found to have a significant relation with OSA severity. We found remarkable differences between non-obese/obese subjects and between males and females.