Prevalence and risk factors of low respiratory arousal threshold in patients with severe obstructive sleep apnea

<bold>Objective:</bold> Low respiratory arousal threshold (ArTH) has been shown to be associated with lower positive airway pressure (PAP) adherence in selected patients with obstructive sleep apnoea (OSA). We aimed to determine the prevalence of low respiratory ArTH in patients with severe OSA (which is currently unknown), to describe their clinical characteristics and its consequences on PAP adherence. <bold>Methods:</bold> Within the multicentre Pays de la Loire Sleep Cohort, we identified patients with severe OSA treated with PAP and for whom follow-up data at 12 months were available. Patients with low ArTH were identified using surrogates of low respiratory ArTH derived from diagnostic sleep recordings: time with SpO2<90% [T90] < 3.7% and/or fraction of hypopnea over apnoea [Fhypopnea] ≥ 77. <bold>Results:</bold> Of 2,157 diagnosed with severe OSA and established on PAP with at least 12-month follow-up, 1,116 (52%) had a low respiratory ArTH. When compared to those with a high respiratory ArTH, patients with a low respiratory ArTH had a lower age (57 vs. 60 years; p<0.01), a lower body mass index (31 vs. 33kg/m2, p<0.01), were more frequently current or former smoker (64% vs. 69%, p:0.03), female (35 vs. 21%; p<0.01) and active worker (45% vs. 38%; p<0.01). They had a lower prevalence of diabetes and hypertension (31% vs. 36%, 62 vs. 69%; p<0.01%) and a lower alcohol use (43 vs. 50%, p<0.01). After adjustment for confounders, patients with low respiratory ArTH were at higher risk of non-adherent to PAP: odds-ratio [95%CI: 1.64 [1.20–2.27] (p<0.01). <bold>Conclusion:</bold> In patients with severe OSA, a low respiratory ArTH is frequent and is associated with lower adherence to PAP therapy.