Study of association of severity of sleep disordered breathing and functional outcome in stroke patients

Sleep disordered breathing (SDB) is a prevalent yet underrecognized condition that may have major adverse consequences for those affected by it. We performed a prospective observational study to seek a correlation of severity of SDB with the severity of stroke and its functional outcome.

Patients with history of recent-onset stroke were recruited and underwent overnight polysomnography (PSG) after the acute phase of the stroke was over; for defining hypopneas, 3% and 4% desaturation limits were used, and the apnea-hypopnea index was respectively calculated as AHI_3% and AHI_4%. Stroke severity was graded using the Scandinavian Stroke Scale. Functional disability and neurological impairment was evaluated six weeks after the PSG using the Barthel Index (<80 = functional dependence; ≥80 = functional independence) and modified Rankins Scale (>2 = poor outcome; ≤2 = good outcome).

A total of 50 patients were enrolled, 30 (60%) with ischemic stroke and 20 (40%) with hemorrhagic strokes. Of the patients, 39 (78%) had an AHI_4% of >5/h, 23 (46%) had an AHI_4% of >15/h, and 9 (18%) had an AHI_4% of >30/h. Multivariate analysis showed that body mass index (odds ratio [OR] = 1.26; 95% confidence interval [CI] = 1.04-1.54, p = 0.019) and Scandinavian Stroke Scale score (stroke severity) (OR = 0.86; 95% CI = 0.76-0.96, p = 0.009) were significant risk factors for predicting SDB (AHI_4% > 15) in patients of stroke. When we looked for factors predicting outcomes, only AHI_4% (OR = 1.20; 95% CI 1.01-1.43, p value 0.041) was predictive of the functional dependence (based on Barthel Index) of the patient and AHI_4% (OR = 1.14; 95% CI 1.03-1.25, p = 0.008) and body mass index (OR = 0.75; 95% CI 0.59-0.96, p = 0.024) were found to be predictive of poor outcome (based on modified Rankins Scale). We obtained similar results, regardless of the hypopnea definition used.

In conclusion, given the high frequency of SDB in stroke patients and its correlation with poor outcome, screening for obstructive sleep apnea in all stroke and transient ischemic attack patients may be warranted.