Aims: The aims of this study were to assess the accuracy of exercise testing in predicting symptom onset within 12 months in patients with asymptomatic aortic stenosis and to establish the criteria that define a positive test.
Methods and results: A total of 125 patients with aortic stenosis [effective orifice area (EOA) 0.9+/-0.2 cm2] were assessed by Specific Activity Scale (SAS) classification, transthoracic echocardiography, and treadmill exercise testing using the modified Bruce protocol. During follow-up, 36 patients (29%) developed spontaneous symptoms within 12 months. Of these, 26 (72%) had had symptoms revealed by exercise testing and 24 (67%) had severe stenosis (EOA < or = 0.8 cm2). Exercise-limiting symptoms were the only independent predictors of outcome at 12 months, and an abnormal blood pressure response or ST segment depression did not improve the accuracy of the exercise test. The positive predictive accuracy for exercise-induced symptoms was 57% in the whole population and 79% for patients aged <70 in SAS Class I. The negative predictive accuracy was 87% in the whole population and 86% in the subgroup.
Conclusion: A significant proportion of patients with apparently asymptomatic aortic stenosis experience limiting symptoms on treadmill exercise testing. The subsequent development of spontaneous symptoms is strongly related to the severity of stenosis and to limiting symptoms on exercise testing, but less so to an abnormal blood pressure response or ST segment depression.