Morphologic determinants favoring surgical aortic valvuloplasty versus pulmonary autograft aortic valve replacement in children

AbstractThe pulmonary autograft is being used with increasing frequency to replace the diseased aortic valve in the pediatric population. Attempted surgical aortic valvuloplasty with an unacceptable result and return to cardiopulmonary bypass for aortic valve replacement with a pulmonary autograft results in prolonged bypass time and increased potential for morbidity. Therefore, the ability to predict an unsuccessful outcome for valvuloplasty would be of significant clinical benefit. This issue is addressed in the present study. Methods: Twenty-two patients (median age 5.7 years, range 3 weeks to 14 years) with bicuspid (n = 11), tricuspid (n = 9), or quadricuspid (n = 2) aortic valves underwent valvuloplasty for aortic stenosis (n = 9), aortic regurgitation (n = 7), or a combination (n = 6). Previous related procedures included balloon aortic valvuloplasty (n = 3) and open surgical valvotomy (n = 1). Median pressure gradient across the aortic valve was 80 mm Hg. Surgical valvuloplasty techniques included thinning of leaflets (n = 18), commissurotomy (n = 15), suspension of reconstructed leaflet to the aortic wall (n = 10), closure of leaflet fenestration (n = 5), shortening of free edge of prolapsed cusp (n = 4), repair of torn leaflets (n = 3), and augmentation of scarred leaflets with autologous pericardium (n = 3). Concomitant subvalvular and supravalvular stenosis were repaired in nine and four patients, respectively. In five patients, during the same hospital stay, a failed valvuloplasty was converted into a valve replacement with a pulmonary autograft because of residual or resultant stenosis (n = 3) or regurgitation (n = 2). Results: No early or late deaths occurred. At a median follow-up of 16.3 months the median pressure gradient across the aortic valve in the 15 patients with preoperative stenosis or combined stenosis and regurgitation was 16 mm Hg (p < 0.01 versus preoperative gradient). Of the 22 patients, the aortic valve functioned normally (defined as ≤ mild stenosis or regurgitation, or both) in 14 patients (including five patients with valve replacement); four patients had stenosis (gradients 40, 45, 60, and 60 mm Hg), two patients had regurgitation, and two patients had combined stenosis (gradients 40 and 50 mm Hg) and regurgitation. Three of the patients with recurrent stenosis underwent secondary surgical valvuloplasty without improvement. Outcome after valvuloplasty was examined according to valve structure: six of nine tricuspid valves functioned normally, whereas only three of 13 nontricuspid valves functioned normally (p = 0.07). Patients with a nontricuspid aortic valve and regurgitation had a high probability of requiring immediate valve replacement (p = 0.009). The actuarial freedom from significant native valve stenosis or regurgitation at 24 months was 82% for tricuspid valves and 36% for nontricuspid valves (p = 0.007). Conclusions: (1) Surgical aortic valvuloplasty should be the preferred approach when the aortic valve is tricuspid. (2) In contrast, aortic valve replacement with a pulmonary autograft should be the preferred strategy in the presence of a nontricuspid aortic valve (especially when the aortic valve is regurgitant) and after failed surgical valvuloplasty. (J THORAC CARDIOVASC SURG 1996;111:1149-57