Postoperative lipid-lowering therapy and bioprosthesis structural valve deterioration: justification for a randomised trial?

Authors:
  • RMRoy G. Masters
    Roy G. Masters
  • PBPierre Bédard
    Pierre Bédard
  • PHPaul J. Hendry
    Paul J. Hendry
  • BLB.-Khanh Lam
    B.-Khanh Lam
  • FRFraser D. Rubens
    Fraser D. Rubens
  • TMThierry G. Mesana
    Thierry G. Mesana

Abstract
Bioprosthesis structural valve deterioration (SVD) is an incompletely understood process involving the accumulation of calcium and lipids. Whether this process could be delayed with lipid-lowering therapy (LLT) is currently unknown. The purpose of this observational study was to evaluate if an association exists between early LLT and a slowing of bioprosthesis SVD, with a view to designing a prospective trial. We followed 1193 patients who underwent aortic valve replacement with contemporary bioprostheses between 1990 and 2006 (mean follow-up 4.5+/-3.1 years, maximum 17.3 years). Of these patients, 150 received LLT (including statins) early after surgery. Prosthetic valve haemodynamics on echocardiography and freedom from re-operation for SVD were compared between patients who did and did not receive postoperative LLT. After bioprosthetic implantation, the progression of peak and mean trans-prosthetic gradients during echocardiographic follow-up (mean 3.3 years) was equivalent between patients treated with and without LLT (peak increase: 0.9+/-7.7 vs 1.1+/-10.9 mmHg, LLT vs no LLT, P=0.87; mean increase: 0.8+/-4.1 vs 0.2+/-5.9 mmHg, LLT vs no LLT, P=0.38). The annualised linear rate of gradient progression following valve replacement was also similar between groups (peak increase per year: 2.0+/-12.1 vs 1.0+/-12.9 mmHg per year, LLT vs no LLT, P=0.52; mean increase per year: 0.5+/-2.2 vs 0.6+/-6.0 mmHg per year, LLT vs no LLT, P=0.94). The incidence of mild or greater aortic insufficiency on the most recent echocardiogram was comparable (16.3% vs 13.8%, LLT vs no LLT, P=0.44), and there was no difference in the 10-year freedom from re-operation for SVD between the two groups [98.9% (95% confidence interval (CI): 91.9%, 99.8%) vs 95.4% (95% CI 90.5%, 97.9%), LLT vs no LLT, P=0.72]. In this observational study, there was no association demonstrated between early postoperative LLT and a slowing of bioprosthesis SVD. With the excellent durability of bioprostheses in the current era, a prospective randomised trial of statin therapy to prevent bioprosthetic SVD does not appear to be justified, let alone feasible.