Objective: To date, no study has focused on the incidence and effects of prosthesis-patient mismatch in patients requiring aortic valve replacement for aortic insufficiency. We hypothesized that the incidence and implications of prosthesis-patient mismatch in patients with aortic insufficiency might be different than for aortic stenosis or mixed disease because the annulus is generally larger in aortic insufficiency and left ventricular remodeling might differ.
Methods: Ninety-eight patients with lone aortic insufficiency (>or=3+ with a preoperative mean gradient <30 mm Hg) were followed over 7.7 +/- 4.3 years (maximum, 17.5 years) with clinical and echocardiographic assessments. They were compared with 707 patients who had aortic valve replacement for aortic stenosis or mixed disease. Prosthesis-patient mismatch was defined as an in vivo indexed effective orifice area of 0.85 cm(2)/m(2) or less.
Results: Compared with patients with aortic stenosis/mixed disease, patients with aortic insufficiency had approximately half the incidence of prosthesis-patient mismatch (P = .003). Patients with prosthesis-patient mismatch had significantly higher transprosthesis gradients postoperatively. An independent detrimental effect of prosthesis-patient mismatch on survival was observed in patients with aortic stenosis/mixed disease who had preoperative left ventricular dysfunction (hazard ratio, 2.3; P = .03) but not in patients with aortic insufficiency, irrespective of left ventricular function (hazard ratio, 0.7; P = .7). In patients with aortic stenosis/mixed disease with left ventricular dysfunction, prosthesis-patient mismatch predicted heart failure symptoms by 3 years after aortic valve replacement (odds ratio, 6.0; P = .002), but there was no such effect in patients with aortic insufficiency (P = .8). Indexed left ventricular mass regression occurred to a greater extent in patients with aortic insufficiency than in patients with aortic stenosis/mixed disease (by an additional 29 +/- 5 g/m(2), P < .001), and there was a trend for prosthesis-patient mismatch to impair regression in patients with aortic insufficiency (by 30 +/- 17 g/m(2), P = .1).
Conclusions: The incidence and significance of prosthesis-patient mismatch differs in patients with aortic insufficiency compared with those with aortic stenosis or mixed disease. In patients with aortic insufficiency, prosthesis-patient mismatch is seen less frequently and has no significant effect on survival and freedom from heart failure but might have a negative effect on left ventricular mass regression.