To evaluate the acceptance, effectiveness in preventing upper airways obstruction, and haemodynamic effects of continuous positive airway pressure (CPAP) in acute stroke.

Twelve patients (4 M, and 8 F; mean (SD), 75.2 (5.5) years) within 48h of acute stroke onset underwent: (1) sleep studies (1st night: auto-CPAP mode; 2nd night: diagnostic); (2) nocturnal non-invasive blood pressure studies (1st night during CPAP; 2nd night during spontaneous breathing (SB)); and (3) daytime cerebral blood flow velocity measurement in middle cerebral artery (FV) with transcranial Doppler during SB and with CPAP (5, 10, 15cmH₂O).

Ninety percent, 60% and 50% of stroke patients had a respiratory disturbance index (RDI) of ≥5, ≥10 and ≥15 events per hour, respectively (18.2 (11.3)). CPAP acceptance was 84%; 42% used CPAP more than 6h and 42% between 1–3h with a mean use of CPAP of 5.2h (4.0). Compared to SB, CPAP reduced, though not significantly, RDI, time with SaO₂<90%, mean blood pressure and mean blood pressure dips (10mmHg)/h. Compared with SB, any level of CPAP progressively and significantly reduced systolic and mean FV; drop in diastolic FV was significant at CPAP10 and CPAP15. The partial pressure of end-tidal CO₂ was significantly lowered by all levels of CPAP.

According to this pilot study, CPAP is reasonably well tolerated by patients with acute stroke for at least one night. Despite its possible beneficial effect on obstructive sleep-disordered breathing and blood pressure variability, CPAP use in acute stroke should be still considered with caution due to possible harmful haemodynamic effects at higher pressures.