Sleepiness in obstructive sleep apnea using hypopneas defined by a 3% oxygen desaturation or arousal but not by 4% or greater oxygen desaturation

Study objectives This analysis determined the association between obstructive sleep apnea (OSA) with hypopneas defined by a 3% O-2 desaturation or arousal (3%A), but not by a hypopnea criterion of >= 4% (4%), and subjective sleepiness. Methods Data were analyzed from Sleep Heart Health Study participants who had polysomnography (N = 6307) regarding OSA and subjective sleepiness scores (Epworth Sleepiness Scale, ESS). The apnea hypopnea index (AHI) was classified based on 3% only, 3%A only, and 4% definitions of hypopneas. Results Of the 3326 participants without 4% OSA, 67.6% (n = 2247) had 3%A only OSA. The ESS score was higher in 3%A only OSA than in those without OSA (7.44 +/- 4.2 vs 7.07 +/- 4.3, P = 0.02). Of those without 4% OSA, 40.2% (n = 1336) had 3% only OSA. The ESS score was higher in those with 3% only OSA than those with no OSA (7.72 +/- 4.3 vs 7.05 +/- 4.2, P < 0.001). A linear regression model demonstrated a trend towards significance for the 3%A only AHI as an independent predictor of ESS when controlled for age, BMI, and sex (P = 0.051). The association of 3% only AHI with the ESS was stronger (P = 0.003). However, 23.3% of the participants with hypersomnia would not qualify as having OSA if arousals were excluded from the definition of OSA. Conclusion The current study found that the presence and severity of OSA is associated with higher ESS scores in persons with 3%A only OSA. Hence, the use of the 4% hypopnea definition will result in the failure to identify and treat a significant number of individuals with OSA who have subjective sleepiness.