Introduction Haemodynamically significant coronary artery disease (CAD) is an important indication for revascularisation. Wall motion analysis during stress echocardiography (SE) is a noninvasive alternative to invasive fractional flow reserve (FFR) for evaluating hemodynamically significant CAD. We sought to determine the diagnostic accuracy of SE compared with invasive coronary angiography with FFR for the diagnosis of hemodynamically significant CAD.

Methods and Results Between January 2008 and April 2015, all patients that underwent clinically indicated FFR measurements during invasive angiography and SE in close succession were analysed. Patients were excluded if tests were not done within 6 months of each other, or an intervening percutaneous coronary procedure or acute coronary syndrome occurred. 184 patients (mean age 66.5yrs, 59 (32%) female) were identified. The majority of patients underwent coronary angiography following SE. The prevalence of known CAD, diabetes and chronic kidney disease were 46%, 43% and 13% respectively, and 14 (8%) patients had previous coronary artery bypass surgery. Exercise SE was performed in 84 (46%) patients and Dobutamine SE in 100 (54%) patients. Contrast was used in 158 patients (86%). In 108 patients (59%), the SE was positive for inducible ischaemia. From 217 vessels analysed, the Left Anterior Descending Artery, Right Coronary Artery, Left Circumflex Artery and Left Main Coronary artery were involved in 120 (55%), 47 (22%), 30 (14%), 18 (8%) respectively, with 2 vessels being grafts. 46 FFR measurements were positive (21%) and 171 were negative (79%), using a cut off of≤ 0.80. At the vessel level, the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of SE for identifying significant disease as assessed by FFR was 70%, 77%, 45% and 90% respectively. In 73 patients, there was single vessel disease on angiography. At the vessel level, the sensitivity, specificity, PPV and NPV were 85%, 68%, 37% and 95%.

Conclusion To date this is the largest study comparing SE and FFR for the assessment of the physiological significance of a coronary lesion, and reflects real world experience. SE demonstrates good diagnostic accuracy and excellent NPV for excluding flow-limiting disease. The low PPV is likely to represent the commencement of medical therapy following a positive SE, as well as referral bias (since only patients with positive SE underwent angiography) as well as the low prevalence of positive FFR measurements in this population. The presence of a haemodynamically significant stenosis can be accurately ruled out with SE.