The approach to managing asymptomatic or questionably symptomatic patients for aortic stenosis is difficult. We aimed to determine whether cardiopulmonary exercise testing (CPET) is prognostically useful in such patients. Patients judged asymptomatic or questionably symptomatic for aortic stenosis with aortic valve area index <0.6 cm2/m2 and left ventricular ejection fraction ≥0.50 were managed conservatively provided they had either (group 1) normal peak oxygen consumption and peak oxygen pulse (>83% and >95% of the predicted values, respectively) or (group 2) subnormal peak oxygen consumption or peak oxygen pulse but with CPET data pointing to pathologies other than hemodynamic compromise from aortic stenosis. Increase in systolic blood pressure <20 mm Hg, ST depression ≥2 mm, or symptoms during the exercise test were allowed. Unexpected events included cardiac death or hospitalization with heart failure in patients who had not been recommended valve replacement. The median age of the study population (n = 101) was 75 years (interquartile range 65 to 79 years), and 67% were judged questionably symptomatic. During a follow-up at 24 ± 6 months, the rate of unexpected cardiac death and unexpected hospitalization with heart failure was 0% and 6.0%, respectively. All-cause mortality was 4.0% compared with 8.0% in the age- and gender-matched population. For group 1, 26 of 70 (37.1%) succumbed to cardiac death, or were hospitalized because of heart failure, or underwent valve replacement, and for group 2 this was 12 of 31 (38.7%). In conclusion, if CPET does not indicate a significant hemodynamic compromise because of aortic stenosis, an initially conservative strategy results in a good prognosis and an acceptable event rate.