Changes in pulmonary artery pressure within an obstructive apnoea and elevations of transmural pulmonary artery pressure (Ppa,tm) towards the end of apnoea are well known. The purpose of our study was to examine which factors contribute to the increase of Ppa,tm in an apnoea. In addition, the time course of Ppa,tm and associated factors during a sleep study was investigated. We analysed the association of changes in arterial oxygen saturation (Sa,O2), oesophageal pressure (Poes) to estimate intrathoracic pressure, systolic blood pressure (BPsys) to estimate left ventricular afterload, apnoea duration and the change in Ppa,tm (deltaPpa,tm) during the course of obstructive apnoeas. Consecutive apnoeas in nonrapid eye movement (NREM)-sleep at the beginning, the middle and the end of the sleep study were analysed in six patients with obstructive sleep apnoea. The mean systolic Ppa,tm was 28.0+/-12.1 mmHg at the beginning of apnoea and 38.6+/-15.5 mmHg at the end (deltaPpa,tm 10.5+/-7.4 mmHg; p<0.0001). DeltaSa,O2 (p<0.0001; odds ratio (OR) 1.45; confidence interval (CI) 1.20-1.76) and deltaPoes (p<0.0001; OR 1.22; CI 1.11-1.34) were independently associated with deltaPpa,tm in a multiple regression analysis. Apnoea duration as well as deltaPoes, deltaPpa,tm and deltaSa,O2 were all significantly higher (p<0.05) in apnoeas at the middle of the sleep study than at the beginning or the end. In conclusion, hypoxaemia and mechanical factors as an increase in negative thoracic pressure contribute to elevations of the transmural pulmonary artery pressure during an obstructive apnoea. The time course of pulmonary haemodynamics within a steep study reveals that the highest transmural pulmonary artery pressure occurs in the middle of the night with no progressive increase towards the end of the sleep study.