Annular Shape and Valve Interactions

Although of great value in evaluating valvular anatomy and function, echocardiography for many decades suffered from the limitations imposed by two-dimensional imaging modalities that made appreciation of true three-dimensional structures difficult. Although three-dimensional echocardiography has improved the overall imaging of valvular structures, the images are still displayed in two dimensions. This limitation can have an impact on the full appreciation of complicated valvular structure and function. For example, calculating the actual valve area of a saddle-shaped annulus is very complex.

Similarly, estimating the aortic-mitral annular angle describes the relationship between the aortic annular and mitral annular planes, but, in reality, neither of these annulae is planar. This angle is probably better approximated by the angle between the long axis of the left ventricle and the axis of the ascending aorta, which itself is only an estimation, because the aorta is not a straight tube. This angle can decrease steadily with increasing ventricular hypertrophy, and resolution to a more normal angle after aortic valve replacement may occur if ventricular hypertrophy resolves.

The aortic valve annulus is not planar, and representation of it as a planar circle is only an approximation of the true valvular effective orifice area. In fact, the aortic valve annulus (and some cardiac anatomists argue that there is not a true aortic valve annulus) rises to heights at the top of the commissures of the three leaflets and plunges down to the lowest portion of each leaflet. During systole, the portions of the annulus near the heights of the commissures move outward in a movement that allows the leaflet to get out of the path of the blood being ejected out of the left ventricle.

The mitral annulus is neither circular nor planar. Although the mitral valve starts as a tube during embryogenesis, the mid portion of the anterior leaflet is compressed downward and out of the plane of the rest of the annulus, presumably due to the presence of the high-pressure aorta just above it. The resulting true mitral annulus resembles a saddle whose shape changes during the course of the cardiac cycle.

The importance of these distinctions in relationship to the study by Tsang et al. is twofold. First, attempting to accurately measure effective orifices in complex three-dimensional structures that change in shape during the cardiac cycle, based upon measurements in several long-axis cut planes, is at best an approximation. Second, the investigators stated that they found that the mitral annular dimension and mitral valve area were reduced after aortic valve replacement and that they were somewhat surprised to find that the aortic annulus was smaller after aortic valve replacement even though they assumed that the cardiac surgeon had inserted the largest prosthesis possible. They suggested that valves be better designed to allow a bigger aortic annulus and also suggested that more flexible aortic valve prostheses might alter mitral annular function less. These suggestions reveal an incomplete understanding of the evolution of aortic prosthetic design and implantation techniques.

Prosthetic Valve Design

The design goals for an aortic prosthesis invariably result in trade-offs. For example, if prosthesis durability is the prime motivating factor, then mechanical prostheses are the best, but they require long-term anticoagulation because of their thrombogenic materials and designs. If the prime motivating factor is a desire to avoid anticoagulation, then bioprostheses that are constructed of either porcine aortic valves or bovine pericardium are advantageous but are less durable. These trade-offs extend to trying to achieve a maximal effective orifice area while still providing a durable design that can be safely implanted.

All initial aortic valve prosthetic designs were based on creating a basic rigid planar circle, which was either part of the housing of a mechanical valve or part of the stent of a bioprosthesis. For mechanical valves, this rigid circular portion was part of the housing that incorporated design features that allowed the occluder to be secured enough that it could not escape, while at the same time providing a surface against which the occluder seated to seal the orifice in diastole. Even though these housings are made of thin, very strong materials, usually titanium or pyrolytic carbon, they still take up space, i.e., they consume some of the area of the native annulus. They also force the cardiac surgeon to insert a planar prosthesis into a nonplanar annulus. Bioprosthetic designs have largely been based on providing a stent to which the bioprosthetic material is sewn and to which a sewing ring can be attached. Obviously, these stents and sewing rings also take up some of the native valve area.

Stentless valves have been developed in an attempt to minimize the aortic annular area consumed by the stent and the sewing ring. However, they still require some area of cloth to help secure the suture material. Implantation of these designs is much more technically demanding and fraught with potential implant errors. They also require longer cross-clamp ischemic times.

More recent bioprosthetic designs have featured stents that allow either more flexibility or deformation, either by changing the material used to make the stent or by using ingenious wire forms that do not form a real circle but rather trace the pattern of the native aortic annulus. The most recent aortic valve prostheses designed for transcatheter implantation have done away with a sewing ring and use a limited stent in an attempt to achieve better orifice area.

The only way to achieve almost true preservation of the native aortic annular size and shape with implantation of an aortic prosthesis is to place the entire prosthesis completely on top of the native annulus, such as with stentless root replacements. Unfortunately, this usually requires replacement of the proximal ascending aorta and reimplantation of the coronary arteries.

The only bioprosthesis that avoids both a stent and a sewing ring is a homograft. Correct implantation of a homograft requires advanced skills to avoid distorting the very flexible tissue. When implanted either in a free-hand fashion or as a root replacement, homografts yield very high effective orifice areas.