Graft failure or progressive native vessel disease can be a serious problem after coronary artery bypass grafting. However, because of poor image quality it may be difficult to evaluate these patients by transthoracic stress echocardiography. The purpose of this study, therefore, was to evaluate the effectiveness of dobutamine stress echocardiography in the detection of myocardial territories with compromised vascular supply (due to either an obstructed native vessel without graft, and obstructed graft, or a native vessel obstructed distal to bypass graft insertion with < or = 50% luminal diameter reduction on angiography) after coronary artery bypass grafting and to determine additional information obtained by biplane transoesophageal stress echocardiography. Sixty patients (54 men, mean age 59 +/- 8.5 years) who had undergone coronary bypass grafting (total number of graft vessels 198) were evaluated from 6 months to 14 years (mean 6.2 years) after surgery. Transthoracic dobutamine stress echocardiography, biplane transoesophageal dobutamine stress echo, and coronary angiography were performed and evaluated by independent examiners. An infusion of dobutamine up to a maximum of 40 micrograms.kg-1.min-1 was administered, and additional atropine (0.25-1.0 mg) was given if 85% of age-predicted maximal heart rate was not reached. Biplane transoesophageal echocardiography was performed in the transgastric short-axis view as well as transoesophageal 4- and 2-chamber views, allowing division of the left ventricle into a 14-segment scheme. Wall motion abnormalities induced with dobutamine stress were used to predict regional vascular insufficiency. A 4-point scale, ranging from 'excellent' (1) to 'impossible' (4) was used to assess each system's ability to evaluate all left ventricular segments. Forty-five patients, of whom 35 were identified by transthoracic echocardiography (sensitivity 78%), had at least one territory with a compromised vascular supply. In 15 patients, the vascular supply was uncompromised, with 13 showing no wall motion abnormalities inducible by transthoracic echocardiography (specificity 86%). However, biplane transoesophageal echocardiography had a higher sensitivity and specificity than transthoracic echocardiography in detecting compromised vascular supply, 93% and 93%, respectively. The former system correctly classified the vascular supplies in 113 of 120 vascular territories (94%), according to whether they were compromised or uncompromised. This was significantly more (P < 0.05) than by classification with transthoracic dobutamine echocardiography, by which system only 102 of the 120 vascular territories were correctly assessed (85%). Compared with the conventional transgastric monoplane short-axis view, examination using three different views via a biplane probe results in a higher sensitivity (93% vs 84%). Assessed on a 4-point scale, the ability to evaluate all left ventricular segments was 2.3 +/- 0.7 (mean +/- SD) for transthoracic echocardiography and 1.7 +/- 0.7 (P < 0.01) for biplane transoesophageal echocardiography. After coronary artery bypass grafting transthoracic dobutamine stress echocardiography has acceptable accuracy in the detection of regional vascular insufficiency. However, this accuracy can be improved using the higher image quality of transoesophageal echocardiography, combined with the advantages of several different views obtained by biplane transoesophageal echocardiography.