Comparison of visual assessment of coronary stenosis with independent quantitative coronary angiography: Findings from the PROMISE trial

Abstract Background The outcomes in patients by visual assessment and quantitative coronary angiography (QCA) for obstructive coronary artery disease (CAD) are not known. Our objective was to compare visual and QCA estimates of obstructive CAD, and assess their relationship to outcomes in stable patients with symptoms of CAD. Methods The Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) trial randomized 10,003 patients with CAD symptoms to anatomical or functional testing. Site reports of invasive angiography detailing visual stenosis and independent, blinded QCA were performed for obstructive CAD (≥50% stenosis). Disagreement between methods was determined and compared to outcomes (death, myocardial infarction, unstable angina hospitalization, or major procedural complications). Results Of 929 patients (9.3% of PROMISE cohort) with angiograms assessed by sites and QCA, 593 (64%) had obstructive CAD per site reports, while 428 (46%) had stenosis ≥50% per QCA. Results differed in 177 patients (disagreement rate 19.1%, κ = 0.63), of whom 171 had CAD per sites but not per QCA. One-year unadjusted Kaplan–Meier event rates were highest (5.1%) when QCA and visual assessment agreed for CAD, lowest (0.9%) when the 2 agreed for no obstructive CAD, and intermediate (3.1%) for patients who had CAD per visual assessment but not per QCA. Conclusions Visual estimation of angiograms results in more frequent diagnosis of obstructive CAD as compared with QCA. Concordance of results for presence or absence of obstructive CAD was associated with high and low event rates, respectively. Disagreement was associated with intermediate event rates, suggesting that cardiologists integrated clinical information into routine visual assessment of angiograms.