Diagnostic accuracy and reclassification potential of the acoustic CADScor algorithm in intermediate risk patients with suspected coronary artery disease

Abstract

Abstract Background Validation studies of the 2019 European Society of Cardiology pre-test probability model (ESC-PTP) for coronary artery disease (CAD) report that 35–40% of patients have intermediate pre-test risk (ESC-PTP 5-<15%). A clear strategy for deferral or referral in this group has not been established. Stratification tools with a high negative predictive value (NPV) are especially wanted to improve pre-test risk estimates. Acoustic detections of coronary stenosis are a new technology which could potentially be useful to supplement PTP stratification. One of the devices, the CADScor®System, has been shown to down-classify >40% of patients to low risk without increasing CAD prevalence. However, the clinical utility of using the CADScor algorithm (version (V)3.1) has not be validated. Purpose 1) To validate the diagnostic performance of the CADScor®System (V3.1), and 2) to study the reclassification potential of a clinical likelihood strategy by ESC-PTP estimation supplemented by a CAD-score. Methods In total, 1732 patients without known CAD but with symptoms suggestive hereof underwent coronary CTA as a first-line diagnostic test. Based on an interview prior to coronary CTA, the ESC-PTP model was applied and sound recordings were performed using the acoustic CADScor® System. Patients with a suspected >50% diameter stenosis in any coronary segment at coronary CTA were referred to investigation with Invasive angiography (ICA) with measurement of Fractional flow reserve (FFR). The ESC-PTP risk estimation was divided according to the recommended cut-offs of <5%, 5-<15% and >15% PTP of obstructive CAD. Haemodynamically obstructive CAD was defined as: (1) FFR value <0.80, (2) luminal diameter stenosis reduction >90%, or (3) luminal diameter stenosis reduction ≥50% if FFR was indicated but not performed. A predefined cut-off value of 20 was used for CAD-score values to rule-out CAD. Results A suspected stenosis was found in 439 patients (26%) after coronary CTA. The follow up with ICA with FFR showed significant stenoses in 198 patients (12%). In the entire cohort using the ≤20 CAD-score cutoff for CAD rule-out, sensitivity was 85.3% (95% CI 79.5–89.9%), specificity was 40.3% (95% CI 37.8–42.9%), the PPV was 5.9% (95% CI 13.8–18.3%)), and the NPV was 95.4% (95% CI 93.4–96.9%). Hence, the disease prevalence of obstructive CAD was 4.6% in the ruled-out patients. Applying the ≤20 CAD-score cutoff for CAD rule-out in intermediate risk patients (ESC-PTP 5-<15%) a total of 316 patients (48%) were down-classified to low risk with an obstructive CAD prevalence of 3.5%. Conclusion Having high NPV, the CADscor holds excellent rule-out power. Interestingly, the CADscor has reclassification properties in intermediate CAD risk patients where almost 50% can be deferred form further testing without increasing obstructive CAD risk. Thus, the CADscor can supplement clinical assessment to guide decisions on the need for further testing. Funding Acknowledgement Type of funding sources: Private company. Main funding source(s): The study was supported by the Health Research Fund of Central Denmark Region, Aarhus University Research foundation and by an institutional research grant from Acarix A/S, Denmark. Patient flowReclassification potential