Predictors of a true-positive stress echocardiography result for optimization of diagnostic algorithm in lowto intermediate-risk patients with acute chest pain

Background. Clinical suspicion of unstable angina in patients with previously unverified coronary artery disease (CAD) has limited efficacy in decisions whether invasive coronary angiography is necessary. A likelihood-based approach to selecting the optimal diagnostic test in evaluating chest pain offers distinct clinical benefits, but decision points for noninvasive angiography or functional testing for lowand intermediate-risk acute chest pain patients with previously unverified CAD remain undefined. Aim: To find decision-making point to select stress-echocardiography (SE) as the initial test in lowto intermediate-risk acute chest pain patients with previously unverified CAD. Methods. The study included 129 patients, aged 56 ± 11 years, of whom 83 (64%) were male and 97 (75%) had ≥ 3 risk factors for CAD. They underwent exercise SE. The diagnostic performance of SE was analyzed to identify true positive (TP) SE results; reference methods were invasive or noninvasive coronary angiography. TP SE was the target outcome in favor of SE as the initial test. All patients were clustered into phenogroups based on differences in serum triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and atherogenic index (AI). TC, non-HDL-C, and AI were used to determine thresholds for belonging to a phenogroup in which the odds of TP SE were higher. Results. The rate of TP SE was 8%. Patients with TP SE had higher levels of non-HDL-C (p = 0.001) and AI (p = 0.066) compared to the remaining patients. After clustering, 2 phenogroups were identified in the total study population, uniting the 36% of patients with the highest non-HDL-C and AI. The odds ratio for TP SE in this joint group was 7.2 (1.4–36.6). Non-LDL >4.42 mmol/l estimate joint group membership with sensitivity 0.91, specificity 0.88, and area under curve 0.97. Conclusion. Non-HDL-C >4.42 mmol/l may be considered in low to intermediate risk acute chest pain patients with previously unverified CAD as a criterion for using SE as a starting test.