A consecutive sample of 500 adults undergoing cardiac surgery was randomly allocated to extracorporeal circulation with either a Bard bubble oxygenator H1700 or a Bard membrane oxygenator HF5700 (Bard Ltd, Crawley, UK). Alveolar-arterial oxygen tension gradient (AaDO2) was calculated prebypass, then 20, 90, 180, and 420 minutes postbypass. Preoperative, initial postoperative, and first-day postoperative chest x-rays were assigned an extravascular lung water (EVLW) score and an atelectasis score. There was a comparable increase in AaDO2 after bypass in each group. The increase in EVLW score was significantly greater in the bubble group (mean 2.91, 95% CI 2.28-3.54) than the membrane group (mean 2.06, 95% CI 1.43-2.69) for the initial postoperative x-rays (P < 0.01) and also for the x-rays on the first postoperative day (P < 0.01). The increase in atelectasis score was significantly greater in the bubble group (mean 1.06, 95% CI 0.94-1.18) than the membrane group (mean 0.86, 95% CI 0.74-0.98) for the initial postoperative x-rays (P < 0.01) but not for the x-rays on the first postoperative day. There was no difference in duration of ventilation, intensive care, hospital stay, or hospital mortality between bubble and membrane groups. Although there was a statistically significant difference in x-ray scores between oxygenator groups, neither intrapulmonary shunting nor clinical outcome was influenced by the type of oxygenator used during bypass.