Non-invasive assessment prior to invasive coronary angiography in routine clinical practice in Switzerland - Is it according to the guidelines?

Background Non-invasive testing is recommended as a basis to decide about the indication for invasive coronary angiography (ICA) in patients with suspected stenotic coronary artery disease (CAD). However, a recent study based on insurance claims data reported that one third of patients undergoing ICA in Switzerland did not have non-invasive testing beforehand. We aimed to re-evaluate the practice of testing prior to ICA in Switzerland by manual review of patient histories. Methods Retrospective analysis of all 816 consecutive patients (age 67 +/- 9 years, 70% males) undergoing elective ICA solely for the evaluation of stenotic CAD during the year 2015 in a single center in Eastern Switzerland. The proportion of patients undergoing a non-invasive test was assessed, and predictors of the lack of such a test were determined. Results 764/816 (94%) patients had a non-invasive test prior to ICA. The majority of patients (728/816; 89%) had an exercise stress test, one fifth (160/816; 20%) underwent a test other than an exercise stress test (6% scintigraphy, 4% stress echocardiography, 6% stress magnetic resonance imaging, 4% computed tomography coronary angiography), and 122/816 (15%) patients had two tests. The use of antianginal drugs other than beta-blockers [odds ratio 1.92 (95% confidence interval 1.01-3.66); p = 0.047] and a lower left ventricular ejection fraction [odds ratio 0.97 (95% confidence interval 0.94-0.99) per one % point increase; p = 0.005] were independent predictors of the lack of a non-invasive test. ICA revealed stenotic CAD in 72% of patients, and 54% of patients underwent revascularization. Patients with and without non-invasive tests did not differ with respect to ICA findings and management. Conclusions The present analysis suggests that patients are appropriately selected for ICA based on clinical judgement and non-invasive testing in Switzerland. There is no evidence for an overuse of ICA.