Comparison of guidelines for diagnosing suspected stable angina and the additional value of the calcium score

Background The performance of current diagnostic algorithms of the American College of Cardiology/American Heart Association (ACC/AHA), National Institute for Health and Care Excellence (NICE) and European Society of Cardiology (ESC) in patients with stable chest pain and coronary artery calcium (CAC) remains a matter of debate. We compared their merits in patients with CAC and investigated the additional value of the CAC score to improve diagnostic accuracy and risk stratification. Methods and results Patient data were obtained from a prospective registry of 642 consecutive patients. Mean age 63 (SD 11) years, 50% male. According to the guidelines, low and intermediate/high pre-test probability groups were constructed. Patients were reclassified based on their CAC score. Obstructive coronary artery disease (CAD) was observed in 14%. All models performed modestly in accurately predicting CAD (c-statistic <0.65). After addition of the CAC score, the c-statistic of the NICE model increased to 0.75 (95% confidence interval (CI) 0.73-0.78) which was just non-significant compared to the ESC model (0.71 95% CI 0.67-0.74) and performed significantly better than ACC/AHA (0.68 (95% CI 0.64-0.72)). After reclassification more than 50% of patients were classified low risk in NICE and ESC, while the prevalence of obstructive CAD (4.8% and 5.2% respectively) did not increase. Conclusions Addition of the CAC score to the studied models improved the ability to safely rule-out obstructive CAD and identified other patients at high risk for future coronary artery events. These results suggest that incorporating CAC score will lead to substantially less downstream testing and lower costs.