Ventricular function in aortic stenosis: how low can you go?

In 1980, we reported that patients with aortic stenosis (AS) who had a low mean gradient ( 30 mm Hg) and low ejection fraction ( 0.30) fared poorly following aortic valve replacement (AVR) (1). We also demonstrated that while ejection fraction may be reduced in AS either because of afterload mismatch or contractile dysfunction, in patients with low transvalvular gradient, afterload mismatch is rarely the explanation for reduced ejection performance. Rather these patients suffer from severe muscle dysfunction in turn explaining their poor outcome. In our small sample, all patients with a low gradient did poorly but only four such patients were included. Subsequent reports confirm a worse prognosis for such patients (2–4), but these larger series have universally demonstrated that some patients with low gradient/low ejection fraction gain substantial benefit by improvement in symptoms following AVR. Benefit occurs despite only a modest improvement in the ejection performance (4). The failure of ejection performance to improve very much is consistent with contractile dysfunction rather than afterload mismatch as the major cause of reduced preoperative performance. Indeed when afterload mismatch is the major culprit, ejection performance improves dramatically. Persistent left ventricular dysfunction in turn no doubt explains why both early and midterm postoperative mortality is substantially greater than is seen in patients with aortic stenosis with preserved ventricular function.