Diagnostic yield of current referral strategies for elective coronary angiography in suspected coronary artery disease—An analysis of the ACROSS registry

INTRODUCTION AND OBJECTIVES: The purpose of this study was to assess the diagnostic yield of current referral strategies for elective invasive coronary angiography (ICA). METHODS: We performed a cross-sectional observational study of consecutive patients without known coronary artery disease (CAD) undergoing elective ICA due to chest pain symptoms. The proportion of patients with obstructive CAD (defined as the presence of at least one ≥50% stenosis on ICA) was determined according to the use of noninvasive testing. RESULTS: The study population consisted of 1892 individuals (60% male, mean age 64±11 years), of whom 1548 (82%) had a positive noninvasive test: exercise stress test (41%), stress myocardial perfusion imaging (36%), stress echocardiogram (3%) or coronary computed tomography angiography (3%). Referral without testing occurred in 18% of patients. The overall prevalence of obstructive CAD was 57%, higher among those with previous testing (58% vs. 51% without previous testing, p=0.026) and when anatomic rather than functional tests were used (81.3% vs. 57.1%, p=0.001). A positive test and conventional risk factors were all independent predictors of obstructive CAD, with adjusted odds ratios (95% confidence interval) of 1.34 (1.03-1.74) for noninvasive testing, 1.05 (1.04-1.06) for age, 3.48 (2.81-4.29) for male gender, 1.86 (1.32-2.62) for current smoking, 1.74 (1.38-2.20) for diabetes, 1.30 (1.04-1.62) for hypercholesterolemia, and 1.39 (1.08-1.80) for hypertension. CONCLUSIONS: More than 40% of patients without known CAD undergoing elective ICA did not have obstructive lesions, even though four out of five had a positive noninvasive test. These exams were relatively weak gatekeepers; functional tests were more often used but appeared to be outperformed by the anatomic test.