The objective of this study was to examine the relationship between sleep apnoea, measures of obesity and blood pressure. The study was cross-sectional in design, involving 1464 consecutive men undergoing sleep studies at two non-hospital sleep laboratories. Detailed sleep studies, body mass index (BMI), neck, hip and waist circumferences and morning and evening blood pressures were measured in each patient. Twenty-eight per cent of patients were obese (BMI > 30 kg/m2) and 47% were overweight (BMI 26-30 kg/m2). Body fat distribution in these patients was typically central with a mean waist-hip ratio (WHR) for the entire group of 0.99 with over 80% of patients having a WHR more than 0.94. Increasing severity of sleep apnoea measured either by respiratory disturbance index (RDI) or minimum oxygen saturation in sleep (MOS) was associated with increasing central obesity and morning but not evening blood pressure. Normal weight patients (BMI < 25 kg/m2) with sleep apnoea were characterized by increased waist and hip circumferences and increased morning diastolic blood pressure compared to patients without sleep apnoea. The best explanatory variables for sleep apnoea were waist (r2 = 0.156, P < 0.001) and age (r2 = 0.013, P = 0.003) only. Morning but not evening blood pressure was related to sleep apnoea independently of obesity and age. We concluded that patients with sleep apnoea are centrally obese. Waist circumference is a better predictor for sleep apnoea than neck circumference or BMI suggesting that the link between obesity and sleep apnoea cannot be explained solely by neck fat deposition.(ABSTRACT TRUNCATED AT 250 WORDS)