Why the study?
Multiple risk models predict obstructive CAD in chest pain, but their performance compared to an experienced cardiologist's assessment using coronary angiography as a reference remained unclear.
Does an experienced cardiologist's clinical assessment improve the prediction of obstructive coronary artery disease compared to standard clinical risk models in stable patients with chest pain?
Does an experienced cardiologist's clinical assessment improve the prediction of obstructive coronary artery disease compared to standard clinical risk models in stable patients with chest pain?
An experienced clinician's assessment outperforms most standard clinical risk models in predicting obstructive coronary artery disease in stable patients with chest pain.
Answers are grounded in this paper and cite their sources.
Clinician assessment matches Duke score and outperforms other models in women; supports hybrid use but leaves model optimization open in prospective cohorts.
INTRODUCTION: Multiple risk models are used to predict the presence of obstructive coronary artery disease (CAD) in patients with chest pain. We aimed to compare the performance of these models to an experienced cardiologist's assessment utilizing coronary angiography (CA) as a reference. MATERIALS AND METHODS: We prospectively enrolled patients without known CAD referred for elective CA. We assessed pretest probability of CAD using the following risk models: Diamond-Forrester (original and updated), Duke Clinical score, ACC/AHA, CAD consortium (basic and clinical) and PROMISE minimal risk tool. All patients completed self-administrative Rose angina questionnaire. Independently, an experienced cardiologist assessed the patients to provide a binary prediction of obstructive CAD prior to CA. Obstructive CAD was defined as >80% stenosis in epicardial coronary arteries by visual assessment, or fractional flow reserve <0.80 in intermediate lesions (30-80%). RESULTS: A total of 150 patients were recruited (100 women, 50 men). Mean age was 58 (32-78) years. Obstructive CAD was found in 31 patients (21%). The area under the curve (AUC) for all the clinical risk prediction models (except the Duke Clinical Score, AUC 0.73, P = 0.07) was significantly lower compared with the clinician's assessment (AUC 0.51-0.65 vs. 0.81, respectively, P < 0.01). The clinician's assessment had sensitivity comparable to the Duke Clinical score, which was higher than all other clinical models. There was no difference in prediction performance on the basis of sex in this predominantly female population. DISCUSSION/CONCLUSION: In stable patients with chest pain and suspected CAD, current clinical risk models which are universally based upon the characteristics of the chest pain, show suboptimal performance in predicting obstructive CAD. These findings have important clinical implications, as current appropriateness criteria for recommending CA are on the basis of these risk models.
No takes yet. Share an insight, caveat, or question.
Havistin et al. (2021) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: