How accurately do the European Society of Cardiology scores predict the likelihood of cardiovascular disease compared with the older Diamond-Forrester score?
The European approach to determining the pretest likelihood of coronary artery disease (CAD) in patients with chest pain is superior to that of the Diamond-Forrester approach recommended by U.S. guidelines, and will result in less need for immediate invasive treatment. It will identify more persons who are low risk and do not require further evaluation. The risk calculator is available at http://bit.ly/2je8FBc. (Level of Evidence = 1b)
Current guidelines from the American College of Cardiology and the American Heart Association recommend the use of the familiar Diamond-Forrester classification system to determine the pretest likelihood of CAD. It uses age, sex, and the type of angina (nonanginal, atypical, and typical) to place patients into risk groups for angina. However, the Diamond-Forrester system was developed with data from the 1970s, and much has changed in how we evaluate patients and manage chest pain. This includes changes in the management of risk factors, advances in treatment, and greater awareness of patients that should seek prompt evaluation for chest pain. The European Society of Cardiology has created a modified version of the Diamond-Forrester system that uses the same clinical variables (age, sex, angina), but was developed using contemporary data. They also created an enhanced version that adds additional risk factor variables.
In this study, the authors identified 2,274 patients without known cardiovascular disease who were referred for coronary computed tomographic angiography. Based on the U.S. guidelines, patients with a probability of less than 5% are classified as low risk and do not need further testing, those with a probability of 5% to 70% should undergo noninvasive testing, and those with a probability greater than 70% should undergo invasive angiography. The cutoffs in the European guidelines are 15% and 85% (instead of 5% and 70%), consistent with a less-aggressive approach to evaluation. The key finding was how poorly the Diamond-Forrester model fit the contemporary data. A calibration plot graphs the observed probability of CAD (defined as a greater than 50% lesion in one or more vessels) against the predicted probability. Although calibration was good (and similar) for the European scores, the Diamond-Forrester score greatly overestimated the likelihood of CAD. For example, at a predicted probability of 50%, the actual probability was only approximately 30%. The Diamond-Forrester score classified only 8% of patients as low risk, compared with 25% with the basic European score and 30% with the enhanced European score. Conversely, the Diamond-Forrester system was far more likely to classify a patient as high risk and requiring invasive angiography: 18% vs. 1% for both of the European scores.
Study design: Cohort (prospective)
Funding source: Foundation
Setting: Outpatient (specialty)
Reference: Bittencourt MS, Hulten E, Polonsky T, et al. European Society of Cardiology–recommended coronary artery disease consortium pretest probability scores more accurately predict obstructive coronary disease and cardiovascular events than the Diamond and Forrester score: The Partners Registry. Circulation. 2016;134(3):201-211.