Heritability of craniofacial structures in normal subjects and patients with sleep apnea.

OBJECTIVES Accumulating evidence has shown that there is a genetic contribution to obstructive sleep apnea (OSA).The objectives were to use magnetic resonance imaging (MRI) cephalometry to (1) confirm heritability of craniofacial risk factors for OSA previously shown by cephalometrics; and (2) examine the heritability of new craniofacial structures that are measurable with MRI. DESIGN A sib pair "quad" design examining apneics, apneic siblings, controls, and control siblings. The study design used exact matching on ethnicity and sex, frequency matching on age, and statistical control for differences in age, sex, ethnicity, height, and weight. SETTING Academic medical center. PATIENTS We examined 55 apneic probands (apnea-hypopnea index [AHI]: 46.8 ± 33.5 events/h), 55 proband siblings (AHI: 11.1 ± 15.9 events/h), 55 controls (AHI: 2.2 ± 1.7 events/h), and 55 control siblings (AHI: 4.1 ± 4.0 events/h). INTERVENTIONS N/A. MEASUREMENTS AND RESULTS Five independent domains reflecting different aspects of the craniofacial structure were examined. We confirmed heritability of sella-nasion-subspinale (38%, P = 0.002), saddle angle (55%, P < 0.0001), mandibular length (24%, P = 0.02) and lower facial height (33%, P = 0.006) previously measured by cephalometry. In addition, the current study added new insights by demonstrating significant heritability of mandibular width (30%, P = 0.005), maxillary width (47%, P < 0.0001), distance from the hyoid bone to the retropogonion (36%, P = 0.0018) and size of the oropharyngeal space (31%, P = 0.004). Finally, our data indicate that heritability of the craniofacial structures is similar in normal patients and those with apnea. CONCLUSIONS The data support our a priori hypothesis that the craniofacial structures that have been associated with obstructive sleep apnea (OSA) are heritable. We have demonstrated heritability for several intermediate craniofacial phenotypes for OSA. Thus, we believe that future studies should be able to identify genes associated with these intermediate craniofacial phenotypes.