Correlation Between Oxygen Saturation and Pulse Tracing Patterns on Overnight Oximetry With Normal Desaturation Index Is an Independent Predictor of Obstructive Sleep Apnea.

STUDY OBJECTIVES: Overnight pulse oximetry (OPO) is commonly used as a screening test for obstructive sleep apnea. Heart rate variability (HRV) correlates well with apnea-hypopnea index during polysomnography (PSG). We hypothesized that visual correlation of episodic increase in HRV with minimal oxygen desaturations on normal OPO (oxygen desaturation index less than 5 events/h) is predictive of OSA. METHODS: A retrospective analysis of patients undergoing OPO and PSG in 1 year was performed. We included only OPO performed on room air and interpreted as normal. Visual correlation between simultaneous increase in HRV and minimal oxygen desaturation was independently assessed by three raters, resulting in the consensus agreement. The primary outcome was presence of OSA on the subsequent PSG. RESULTS: Of 936 patients with OPO and PSG, 109 patients had normal overnight oximetry study on room air. Of these, 65 (60%) were females, median (interquartile range) age was 54 years (44, 67), body mass index was 29 kg/m2 (25, 32), and the median oxygen desaturation index was 1.8 events/h (1, 2.7). Consensus agreement identified 54 patients with visual correlation between pulse and minimal oxygen desaturations. Thirty-two patients (29%) were found to have OSA on PSG, of which 24 (75%) could have been accurately predicted by the consensus agreement (odds ratio 4.70, 95% confidence interval 1.87-11.8, P < .001). When adjusted for pertinent clinical and demographic variables, consensus agreement was independently associated with diagnosis of OSA on subsequent PSG (odds ratio 5.6, 95% confidence interval 1.76-20.9, P = .003). CONCLUSIONS: Visual correlation between episodic increase in HRV and minimal oxygen desaturations on OPO is an independent predictor of OSA, and promising marker for clinical use.

Sleep Apnea in Patients Hospitalized With Acute Ischemic Stroke: Underrecognition and Associated Clinical Outcomes.

STUDY OBJECTIVES: To evaluate clinical recognition of sleep apnea and related outcomes in patients hospitalized with acute ischemic stroke. METHODS: A retrospective study of all patients hospitalized with acute ischemic stroke from April 2008 to December 2014. The primary predictor and outcome variables were sleep apnea and hospital mortality, respectively. Secondary outcomes were mechanical ventilation, hospital length of stay, and the survivor's functional level by the modified Rankin scale. A sensitivity multivariate regression analysis included the propensity score for cardiovascular comorbidities and sleep apnea. RESULTS: Of 989 patients, 190 (19%) were considered to have sleep apnea. Only 42 patients (22%) received any treatment for sleep apnea during the hospital stay. Despite higher prevalence of cardiovascular comorbidities, the patients with sleep apnea had lower hospital mortality, 1% versus 5.6% in patients without sleep apnea (odds ratio [OR] 0.18; 95% confidence interval [CI], 0.03-0.58, P = .002). Only the National Institutes of Health Stroke Scale (NIHSS) and the Glasgow coma scale (GCS) were significant predictors of adjusted hospital mortality (OR 1.06, 95% CI 1.01-1.11, P = .01 and OR 0.61, 95% CI 0.51-0.69, P ≤ .001, respectively). A composite clinical propensity score for sleep apnea and cardiovascular comorbidities was significantly associated with decreased mortality, independent to either NIHSS (OR 0.11, 95% CI 0.017-0.71; P = .02) or GCS (OR 0.07, 95% CI 0.01-0.52; P = .01). CONCLUSIONS: Prevalence of sleep apnea in our study was low, likely because of clinical underrecognition. Despite having more cardiovascular disease, the patients with acute stroke and sleep apnea had less severe neurological injury and lower unadjusted mortality than those without a history of sleep apnea.