Stroke affects 16.9 million individuals each year and is the second leading cause of death worldwide. Despite advances in pharmacologic therapy, morbidity , mortality and rates of hospitalization for stroke remain high. These data emphasize the importance of identifying all treatable conditions that could aggravate stroke. One such condition is obstructive sleep apnea (OSA).

Sleep-related breathing disorders, including obstructive and central sleep apnea, often coexist with stroke. Compared to the general population, in whom OSA is the most common form of this breathing disorder with recent prevalence estimates of 22% of male and 17% of female , in the stroke population, the prevalence of OSA is much greater at 70% . Several randomized controlledtrials on OSA patients with stroke in acute or sub-acute stage showed that treating OSA with continuous positive airway pressure (CPAP) improved motor and functional outcomes, accelerated neurological recovery.Apart from the benefits in better neurological outcomes, secondary analyses of SAVE study suggested that CPAP treatment potentially help to reduce recurrence of stroke. Nevertheless, we don't have evidence yet from randomized control studies to prove CPAP treatment would reduce the recurrence of cardiovascular or cerebrovascular events.

Traditionally, recurrence of cardiovascular or cerebrovascular events uses documented mortality, morbidity or hospitalization for heart failure, acute coronary syndrome or stroke as clinical endpoints. Recently, several studies showed that enlarged left atrium (LA) can serve as a predictor for recurrent stroke or cardiovascular events. On the other hand, a growing body of studies demonstrated that CPAP treatment reduce size of LA in those with OSA. Notably, all of these studies above are observational or retrospective in nature. To date, there are no prospective longitudinal randomized controlled trials reporting the effect of CPAP treatment of OSA on the change of size of LA. We therefore will undertake a randomized , controlled trial involving patients with stroke to test the primary hypothesis that treatment of OSA with CPAP would reduce the size of LA.

Inclusion Criteria:

1. Time from onset of stroke symptoms to hospital arrival <2 weeks .
2. Stroke is documented with brain magnetic imaging or computed tomography
3. Competency to provide informed consent.
4. Moderate to severe obstructive sleep apnea being established with the use of a home sleep-study screening device (ApneaLink).
5. Epworth sleepiness scale≦10.

Exclusion Criteria:

1. Having received CPAP for obstructive sleep apnea prior to admission.
2. History of pneumothorax or brain surgery.
3. Coexisting heart failure or renal failure or persistent atrial fibrillation.
4. Unable to wear a nasal or nasal-oral mask.
5. Concomitant uncontrolled infection.
6. Swallowing difficulty or episodes of choking due to stroke
7. Coexisting central nervous diseases such as dementia