Incongruence of the apnea\hypopnea index in detecting intermittent desaturations at very high altitude

<b>Rational:</b> Together with hypobaric hypoxia, obstructive (OSA) and central sleep apnea (CSA) at high altitude can occur in susceptible individuals. The night-time presence of sleep disordered breathing (SDB) can further worsen tonic hypoxia adding intermittent episodes of oxygen reduction. The aim of this study was to better understand if the apnea/hypopnea index (AHI) and the oxygen desaturation index (ODI) are able to accurately characterize SDB in condition of hypobaric hypoxia at very high altitude. <b>Methods:</b> We performed nocturnal polygraphy on healthy miners in the Chilean andes at 4100m altitude. To better evaluate the difference between ODI and AHI we also extracted with an automated software a sample of 32 subjects in a frequency matched method according to their age, BMI and AHI. These subjects performed home respiratory polygraphy (HRP) at sea-level for OSA suspicion. <b>Results:</b> 32 miners (age median [IQR] 54 [38.8 to 56.2], body max index 27.5 [25.7 to 30.1]) performed HRP at 4100m altitude. The AHI was 25.2 [11.8 to 58.2] with a prevalence of CSA, but the ODI (43.4 [22.8 to 61.9]) was significantly higher compared to the AHI. The matched group at sea level had an AHI of 23.6 [11.9 to 41.2] and an ODI of 25.5 [14.1 to 42.2]. 41% of subjects at high altitude presented &gt;9 events\hour between ODI and AHI compared to 3% at sea level (p&lt;0.01) and 50% of miners had a ratio (((AHI-ODI)/ AHI)*100) &gt; 49% compared to 6% at sea level (p&lt;0.01). <b>Conclusions:</b> At very high altitude the AHI is not able to accurately justify the intermittent desaturations on the top of hypobaric hypoxia. Novel metrics are needed to better capture the severity of OSA and CSA at high altitude.