Sustained oximetric desaturation as a cardiac risk indicator remains correlated with nonspecific sleep apnea indices.

<bold>BACKGROUND:</bold> Sustained oximetric desaturation is a risk factor for cardiac morbidity and mortality in sleep apnea, Traditional sleep apnea indices are not as predictive. <bold>AIM:</bold> To model the effects of two indicators of sleep apnea severity on sustained oximetric desaturation. <bold>METHODS:</bold> Exploratory data analysis and an analysis of covariance (ANCOVA) were performed on an anonymized convenience sample of 5000 ambulatory sleep polygraphs. The ANCOVA modelled the effects of apnea-hypopnea index (AHI, events/h) and the percentage of recording time with inspiratory flow limitation (IFL, %) on the percentage of time with SpO2 < 90% (Under90, %). <bold>RESULTS:</bold> The medians (and interquartile ranges) for Under90, AHI, and IFL were 7.3 (0.01, 29.3)%, 12.7 (6.6, 23.6)/h, and 17.7 (6.4, 24.9)%. The sums of squares for AHI, IFL, AHI:IFL, and residuals were 52.5, 0.8, 0.3, and 332.1. The F scores and Pr(>F) for AHI, IFL, and AHI:IFL were 880.7 (p << 0.001), 13.4 (p = 0.002), and 5.6 (p=0.018). <bold>CONCLUSIONS:</bold> The time with SpO2 < 90%, a risk factor for cardiac morbidity and mortality, strongly correlated with AHI and less so with IFL, two insensitive markers for cardiac morbidity and mortality. Further modelling including epidemiological risk scores such as the Framingham score and other polysomnogram risk factors such as pulse wave amplitude variation may improve cardiac risk prediction with sleep apnea.