[Angiographic correlates of the high-risk criteria for conventional exercise testing and the Duke treadmill score].
OBJECTIVE: To compare the sensitivity, the specificity, the positive and negative predictive value and the predictive accuracy of the Duke Treadmill Score, the Spanish Society of Cardiology (SEC) and American College of Cardiology/American Heart Association (ACC/AHA) high-risk criteria for exercise testing in the detection of left main disease, three vessel disease and two vessel disease involving the proximal left anterior descending artery. PATIENTS AND METHOD: A cohort of 199 patients (age 75 years) consecutively admitted to hospital for unstable angina was studied. All patients underwent an exercise stress test and coronariography. RESULTS: The SEC high-risk Criteria showed a sensitivity of 69.2% and a specificity of 49.0%. The ACC/AHA high-risk Criteria demonstrated a sensitivity of 98.1% and a specificity of 23.8% and the Duke Treadmill Score presented a sensitivity of 30.8% and a specificity of 90.5%. In patients with moderate risk in the Duke Treadmill Score we found a sensitivity of 62.9% and a specificity of 39.8% for the SEC high-risk criteria, while the ACC/AHA high-risk Criteria presented a sensitivity of 100.0% and a specificity of 5.8%. CONCLUSIONS: The ACC/AHA high-risk Criteria showed a higher sensitivity while the Duke Treadmill Score presented a higher specificity for the detection of left main disease, three vessel disease and two vessel disease involving the proximal left anterior descending artery. The ACC/AHA and SEC high-risk Criteria were found to be very useful in the group of patients with moderate risk in the Duke Treadmill Score.