The perennial quest for an ideal prosthetic valve
- 1Department of Cardiology, Oxford University Hospitals NHS Trust, Oxford, UK
- 2Department of Cardiac Surgery, Oxford University Hospitals NHS Trust, Oxford, UK
- Correspondence to Dr Bernard D Prendergast, Department of Cardiology, Oxford University Hospitals NHS Trust, Oxford OX3 9DU, UK;
- Received 10 September 2012
- Revised 10 September 2012
- Accepted 12 September 2012
- Published Online First 19 October 2012
The evolution of prosthetic valve design
The first surgical heart valve replacement was undertaken over half a century ago since when there has been a continuing quest to develop a prosthesis with the perfect combination of durability, haematological compatibility and haemodynamic performance. Early caged ball mechanical valves were robust but associated with red cell trauma and rapidly followed by introduction of mono-leaflet and bi-leaflet designs which remain in use today. Introduced later in the 1960s, bioprosthetic valves consisting of pericardial or valve tissue mounted on a frame and sewing ring were initially conceived for use in elderly patients unlikely to outlive the valve, or in other patients at risk of bleeding complications associated with anticoagulation. Until recently, mechanical valves have remained the logical choice in patients <65 years of age given the minimal rates of primary valve failure.1 However, advances in the longevity and haemodynamic performance of modern bioprostheses and the emerging potential for transcatheter treatment when they degenerate seem set to challenge this paradigm.2
Stentless bioprosthetic valves offer the potential for (A) further improvements in the flow dynamics of cardiac output and consequent haemodynamic performance, (B) remodelling and dynamic adaptation to the anatomy of the aortic root with resulting progressive increase in the effective orifice area, and (C) enhanced regression of left ventricular hypertrophy with improved clinical outcomes. Although initial comparative studies demonstrated superior haemodynamics, prospective randomised controlled trials against modern stented bioprosthetic valves with lower profile frames, thinner sewing rings and excellent haemodynamic characteristics3 have shown no major differences in clinical and echocardiographic endpoints at 1-year follow up.4
Valve selection and assessment of function
Robust European guidelines for the management of patients with valvular heart disease include class I indications for valve replacement and advice on the choice of prosthetic valve.5 Selection of the optimal prosthesis for an individual patient depends on a number of factors including cardiac anatomy and pathophysiology, age, requirement for and/or suitability for anticoagulation, and lifestyle. However, while emphasising the importance of implanting an adequately sized valve to avoid the phenomenon of patient-prosthesis mismatch,6 the guidelines provide no specific discrimination between different bioprosthetic designs reflecting the limited available evidence and recognition that this decision is usually based on the patient's body size, anatomy and surgical preference.
The haemodynamic performance of a prosthetic valve should resemble that of a normal native valve. Excessive transvalve gradients, which may result from inferior design or undersizing relative to the patient's body surface area (so-called patient-prosthesis mismatch) may result in persistent postoperative symptoms. Although such symptoms are usually exertional, the majority of assessments of prosthetic valve function are undertaken at rest using conventional transthoracic echocardiography—identification of suboptimal haemodynamic performance requires dynamic assessment utilising exercise or pharmacological stress. Exercise echocardiography is the recommended approach but can be challenging in elderly patients and Dobutamine stress echocardiography (which replicates the high flow rates necessary to assess load-induced haemodynamics7) is an acceptable alternative, particularly when assessing the performance of small prostheses.8 To date, the majority of studies concerning the haemodynamic performance of prosthetic valves report limited echocardiographic data (transvalve gradient and effective orifice area) on use in the aortic position at rest. These measurements are prone to error due to flow rate dependency and difficulties measuring the left ventricular outflow tract diameter (valve size is not a valid surrogate). The dimensionless severity index (also known as the Doppler velocity index) incorporates the effect of flow on velocity and does not require anatomical measurement. Use of this more robust parameter at rest and during stress is now the recommended method for assessment of prosthetic valve haemodynamic performance.9
Khoo et al10 present rest and stress echocardiographic evaluation of the haemodynamic performance of five different prosthetic aortic valves and highlight controversies concerning designs associated with the best haemodynamic performance and challenges in assessment with controls for prosthesis size and flow rates. They demonstrate that stentless bioprostheses have superior haemodynamics (both at rest and during pharmacological stress) with parameters equivalent to mild native aortic stenosis (peak gradient 16±1 mm Hg and mean gradient 9±1 mm Hg) while stented bioprosthetic and mechanical valves (in this study a single tilting disc valve) are inferior with haemodynamic performance equivalent to moderate native aortic stenosis (peak gradient 24±3 mm Hg and mean gradient 13±2 mm Hg). Although the suggestion that these data will help cardiologists and cardiac surgeons choose the correct prosthesis for individual patients is perhaps overstated, their findings are important for three reasons: (A) five prosthetic valve designs were compared (rather than the usual two); (B) assessments were performed both at rest and during Dobutamine stress; (C) use of the dimensionless severity index controlled for variations in prosthetic valve size and stroke volume.
These data are consistent with earlier studies and therefore unsurprising but do provide contemporary assessment of the dynamic function of a range of prosthetic valves of different designs and from numerous manufacturers. Crucially, are these differences in haemodynamic performance of clinical relevance? Echocardiographic assessment of left ventricular mass demonstrated no apparent influence of valve haemodynamics at 1 year and clinical outcomes such as New York Heart Association (NYHA) class, 6 min walk testing or survival were not reported. Similarly, while numerous previous studies have demonstrated improved resting haemodynamics and favourable reduction in left ventricular mass up to 1 year following implantation of stentless bioprostheses,11 clinical advantages have not been proven.
Echocardiographic assessment of prosthetic valve function is challenging and stress imaging provides additional data on dynamic function. The demonstrated haemodynamic superiority of stentless valves is not yet convincingly associated with meaningful clinical advantages and decisions concerning selection of the optimal prosthesis for an individual patient will continue to be centred around the questions of prosthesis longevity and safety of anticoagulation. Secondary factors such as body surface area, small annular diameter and impaired left ventricular function may favour use of a stentless bioprosthesis or aortic root enlargement in individual cases to optimise cardiac output and avoid patient-prosthesis mismatch. Current data concerning the haemodynamic performance of prosthetic valves are largely based on resting Doppler data—increased use of stress imaging, widespread adoption of the dimensionless severity index and correlation with clinical outcomes are required to adequately evaluate prosthetic valve function and prove the potential advantages of stentless bioprostheses. Only then will cardiologists and cardiac surgeons be able to refine their choice of prosthetic valve even further.
Contributors All authors contributed equally to this editorial. BDP is responsible for the overall content as guarantor.
Competing interests None.