Abstract 11136: Obstructive Sleep Apnea in Aortic Dissection

Abstract

Background: Obstructive sleep apnea (OSA) imposes an afterload burden on the left ventricle and increases the pressure gradient across the intrathoracic aorta. This might increase the likelihood of aortic dissection (AD). We hypothesized that patients with acute AD would have a high prevalence of previously undiagnosed OSA and a largely nocturnal presentation. Methods: 40 consecutive, consenting subjects with acute AD who survived to hospital discharge were collected from 4 institutions. Mean age was 57 (range 28-75), 24 were male. 25 had type A dissection and 15 type B. 37 completed the modified Berlin Questionnaire and 30 had attended overnight polysomnography. Presenting CT scans had aortic diameter measured at a) the sinotubular junction b) ascending aorta c) origin of innominate d) mid-arch e) origin of left subclavian f) mid-descending aorta g) level of diaphragm; CT scans at 3 months were available in 24 subjects. Results: AHI ranged from 0 - 89; prevalence of OSA (defined by apnea-hypopnea index [AHI] > 5) was 63%. There were no differences in aortic diameter at any level comparing OSA vs non-OSA, regardless of AHI cutoff (>5, >15, >30; figure). Over time aortic diameter decreased at the level of the sinotubular junction and ascending aorta (expected - due to surgery). Whether dissection occurred during overnight hours (10 PM - 7 AM) did not differ by presence/absence of OSA (regardless of AHI cutoff). In those with high-risk Berlin questionnaires (>=2 categories positive) 10/21 (48%) dissections occurred during sleeping hours vs 2/13 (15%) in low risk Berlin group (p = 0.08, Fisher’s exact test). Conclusions: Among patients with acute AD the prevalence of OSA was high. Aortic diameters, however, were not different according to level of AHI. Presence of OSA defined by polysomnography was not associated with a nocturnal presentation in this sample.