eComment. Choice of bioprosthesis for aortic valve replacement

We read with great interest the best evidence topic by Yap et al. regarding the best bioprosthesis for aortic valve replacement [1]. We would like to present the results of our relative prospective multicentre randomized study. The aim of the study was to compare the haemodynamic performance and the sizing characteristics of the Medtronic Mosaic Ultra porcine and Carpentier-Edwards Perimount Magna bovine pericardial bioprosthetic stented aortic valve in relation to the patient's true aortic annulus size [2]. After the aortic valve excision, the annulus was measured by blinded sizers prior to the randomization of 141 patients (Ultra, n = 72; Magna, n = 69) [2]. The median patient age was 75 years, and 89% of the patients had aortic valve stenosis. A good left ventricular function was present in 75% of patients, and the EuroSCORE-predicted mortality was equal to 9%. Concomitant procedures (coronary artery bypass grafting, mitral/tricuspid valve repair, septal myectomy, modified maze procedure with radiofrequency) were performed in 61% of patients. The in-hospital mortality was 3%, and at six months postoperatively, 96% of the patients were in NYHA class I/II, with no inter-group differences. The mean 'true aortic annulus' size was 23.0 ± 1.4 mm for the Ultra valve, and 22.6 ± 1.8 mm for the Magna valve (P = not significant). The implanted labeled valve size was > or = 23 mm for 83% of Ultra valves, and for 52% of Magna valves (P <0.01), and smaller than the measured true aortic diameter (44% Magna vs 33% Ultra). The mean echo gradients were lower with Magna valves (11 ± 6 mmHg) than with Ultra (17 ± 6 mmHg; P <0.01), while the effective orifice area (EOA) was higher with Magna than with Ultra (1.6 ± 0.4 versus 1.4 ± 0.4; P <0.01). Both groups showed a similar left ventricular mass regression (Ultra -48 ± 83 g; Magna -42 ± 70 g). Trivial to moderate regurgitation was noted in 24% of Ultra valves compared to 48% of Magna valves (P <0.01) [2]. According to the results of our study, the selection of the Ultra bioprosthetic valve allowed the implantation of larger valve sizes. However, when compared to the 'true aortic annulus', the Magna was associated with lower transprosthetic gradients and larger EOAs. However, the long-term significance of these observations remains inconclusive in terms of bioprosthesis selection [2]. In conclusion, the choice of bioprosthetic heart valve should not be influenced solely by the early systolic gradients, but also by the expected long-term durability [2]. Rahimtoola has suggested that bovine valves have a lower incidence of structural valve degeneration based on a 2001 meta-analysis of both Hancock and Carpentier-Edwards valves [2, 3]. A recent report on the Hancock II bioprosthesis reported a freedom from structural valve degeneration at 20 years of 85% in patients aged 60-70 years, and of 99.8% in patients aged >70 years [2,4]. The results of in vitro studies have also suggested that the Perimount Magna ThermaFix anticalcification treatment may be inferior to the Mosaic Ultra anti-calcification, and would result in an earlier reversal of the initially favorable Magna gradients [5]. Conflict of interest: none declared