Sex-specific differences in diagnostic approaches of inpatient sleep testing for obstructive sleep apnea

Few studies have investigated sex-specific disparities in inpatient sleep testing. We postulate that women are more likely to have a milder degree of obstructive sleep apnea (OSA) and lower extent of hypoxia on Type III sleep studies versus polysomnography (PSG).

The Cleveland Clinic Sleep laboratory registry was leveraged to identify all adult inpatient sleep studies performed for OSA. Demographics, comorbidities, and sleep study measures were collected and compared by sex and sleep study type. Logistic regression was used to examine sleep study type predictive of OSA (apnea hypopnea index [AHI; ≥5, ≥15 and ≥ 30]) and hypoxia, (median percentage of sleep time spent at <90% SaO2 [TST<90%,≥ 11%,] adjusted for covariates.

The sample 778 patients had a mean age of 56.1 ± 16.1 years; 44.5% were female and 72.2% Caucasian. At an AHI≥5, women showed an increase odds of OSA (adjusted, OR = 2.04,95%; CI:1.24-3.35, p = 0.005) with Type III sleep study vs PSG compared to men. At an AHI≥15, men had less odds of OSA (adjusted OR = 0.60,95%CI:0.39-0.90,p = 0.015) with Type III sleep study vs PSG compared to women (OR = 1.15,95%CI:0.72-1.85,p = 0.56), with an interaction p-value of 0.040. These results were attenuated when the analysis was restricted using the 3% hypopnea scoring rule. Men and women had higher odds of TST <90 ≥ 11% (OR:2.60,95%CI:1.60-4.21,p=<0.001; OR:3.46,95%CI:1.97-6.05,p < 0.001) with Type III sleep study versus PSG, albeit no sex-interaction was observed.

These results suggest that sex-specific differences in diagnostic performance of sleep testing type in the inpatient setting should be considered according to level of OSA severity, which are influenced by hypopnea-related desaturation extent.