Snoring worsens with high alcohol consumption. It is unclear whether moderate alcohol intake worsens sleep and breathing in subjects with obstructive sleep apnoea syndrome (OSAS), and whether alcohol increases the pressure requirement for nasal continuous positive airway pressure (CPAP). Fourteen adult males with untreated OSAS but without heart or lung disease were studied (age 53+/-9 yrs, body mass index (BMI) 33+/-5 kg x m(-2) (mean+/-SD). The subjects underwent overnight polysomnography on four occasions: control, alcohol, CPAP, and alcohol + CPAP. On the alcohol nights, the subjects drank 1.5 mL x kg(-1) body weight (BW) vodka (40% alcohol by volume) (blood alcohol with and without CPAP 0.45+/-0.1 and 0.47+/-0.2 mg x mL(-1) (mean+/-SD)). On the CPAP nights, the pressure required to prevent apnoea, snoring, and silent inspiratory airflow limitation was determined using an autotitrating nasal CPAP system (ResCare AutoSet). Alcohol and control nights were performed in random order. Without CPAP, alcohol produced a small non-significant decrease in the percentage of rapid eye movement (REM) sleep (control 11+/-2 vs alcohol 8+/-1% (mean+/-SEM)), but with CPAP there was no such effect (control 15+/-2 vs 17+/-2%; CPA x alcohol interaction p=0.015). With CPAP, slow-wave sleep in the first 2 h increased slightly with alcohol (control 39+/-6 vs alcohol 51+/-4%; p=0.004). Arousal index without CPAP increased slightly with alcohol (control 43+/-5 vs alcohol 49+/-6 events x h(-1); p=0.02). There was little or no effect of alcohol on other sleep stages, arousal index, apnoea index, apnoea/hypopnoea index, mean or longest event duration, mean or worst arterial oxygen saturation, with or without CPAP, either for the full night or for the first 2 h. There was no change in the pressure requirement for CPAP (full night: control 11.9+/-0.9 vs alcohol 12.5+/-0.9 cm H2O; first 2 h: 10.9+/-0.6 vs 11.1+/-0.8 cm H2O). Moderate alcohol intake (in the form of vodka) has little effect on breathing or saturation during sleep in subjects with mild-to-severe obstructive sleep apnoea, and no effect on the pressure required for continuous positive airway pressure in order to prevent apnoea, snoring, and flow limitation. These results cannot be extrapolated to other doses or forms of alcohol, or to subjects with concurrent heart or lung disease.