OBJECTIVES: To assess the diagnostic performance of per-exercise echocardiography using a recent tilt table to investigate for coronary artery disease with a significance threshold for stenosis > or =70%. METHODS: 104 consecutive patients (93 men) referred for elective coronary angiography underwent tilt table exercise echocardiography in which second harmonic imaging was used systematically. Images were recorded and interpreted by the same operator. RESULTS: The investigation was contributory in 81 patients (target HR at least 85% of the age-predicted maximal HR or criterion for discontinuation). The most common reason for a non-contributory investigation (negative investigation but submaximal or uninterpretable investigation) was submaximal exercise (91%). The prevalence of significant coronary artery disease in the cohort of 81 contributory investigations was 38%. Sensitivity and specificity to detect significant lesions was 90% and 94% respectively. The positive and negative predictive values were 90% and 94% respectively. The three false negatives involved distal lesions. The three false positives occurred in patients with regional wall motion abnormalities in the context of various heart diseases (severe mitral incompetence, diabetic and hypertensive cardiomyopathy, left bundle branch block). CONCLUSION: Using a high performance echocardiography machine with second harmonic imaging per-exercise tilt table echocardiography offers excellent diagnostic performance to detect significant coronary artery disease under everyday clinical practice conditions.