Tissue Valves: Stented

The valve most commonly used to replace the aortic valve is a stented bioprosthetic valve, either bovine pericardium or porcine ( Figures 14-2 and 14-3). The advantages to such a valve are: (1) ease of implantation and the rare occurrence of clinically significant patient prosthesis mismatch due to improving valve hemodynamics, (2) no lifelong need for anticoagulation with warfarin (unless the patient requires it for a different reason), (3) a relatively straightforward future reoperation for SVD, if necessary, and (4) the potential for a valve-in-valve procedure using a transcatheter heart valve for SVD. The most important disadvantage to tissue valves is the occurrence of SVD, which is primarily age dependent.

The technical aspects of AVR with a stented bioprosthetic valve are straightforward. The aortic valve can be exposed through a variety of aortotomies, including a hockey-stick, transverse, and oblique incisions. The aortic valve is excised and the annulus is extensively débrided of calcific plaques with care taken in the area of the conduction system (below the commissure between the noncoronary and right coronary cusps). Calcific extensions on the anterior leaflet of the mitral valve are removed. With adequate débridement of annular calcification, perivalvular leak is rare, and with current-generation bioprostheses, clinically significant patient-prosthesis mismatch is uncommon.

Bioprosthetic Valves: Stentless

In the 1990s, “stentless” bioprosthetic valves made from porcine aortic valves became available. The advantages of this type of valve versus stented bioprosthetic valves were thought to be: (1) avoiding anticoagulation with a low risk for stroke and (2) improved hemodynamics compared to stented and mechanical valves.5–7 The disadvantages to stentless valves were: (1) more complex operation requiring either a “mini-root” with reimplantation of the coronary ostia ( Figure 14-4) or subcoronary implantation ( Figure 14-5) and (2) data indicating concerns about freedom from SVD. ⁸

Some surgeons use stentless porcine valves primarily in patients with BAVs with aneurysms. In this case, the aortic valve, root, and a portion of the tubular ascending aorta are replaced. Current American College of Cardiology/American Heart Association (ACC/AHA) indicate that if a patient with a BAV requires AVR and has an ascending aortic diameter greater than 4.5 cm, then aortic replacement should be undertaken. ⁹ When this operation is performed with a bioprosthetic valve, current-generation stented valve must be sewn into a separate vascular graft, adding a few minutes to the procedure. With a stentless porcine valve, that step is not required because the porcine valve is packaged as a complete root.

Mechanical Valves

Mechanical valves have the advantages of long-term durability and a long track record with designs that have been durable for decades.10,11 The major disadvantages are: (1) the need for lifelong anticoagulation, currently with warfarin, (2) a higher risk of thromboembolism than with bioprosthetic valves, and (3) audible clicking in some patients with several of the mechanical valve types that may be troublesome.

A newer model of mechanical valve, the On-X prosthetic heart valve (On-X Life Technologies, Inc., Austin, Texas), first implanted in 1996, has been shown to have low adverse clinical event rates, including 0.6% thromboembolism per patient-year, 0.4% bleeding rate per patient-year, and 0% thrombosis rate when used in the aortic position. ¹² An ongoing clinical trial (Prospective Randomized On-X Anticoagulation Clinical Trial [PROACT]) is studying the safety of lower doses of warfarin in patients with high-risk for thromboembolism and antiplatelet drugs only (aspirin/clopidogrel) in patients with low-risk for thromboembolism. ¹³

Aortic Homografts

The first successful orthotopic placement of an aortic homograft was performed in 1962 by Donald Ross. ¹⁴ Much like the procedure for stentless bioprosthetic valves, the operation is more complex than straightforward implantation of a stented tissue valve, because a mini-root may be performed ( Figure 14-6) or the valve may be sewn in the subcoronary position, as with aortic homografts.

The perceived advantages to the homograft were: (1) freedom from anticoagulation and a low risk for thromboembolic events typical for bioprosthetic valves, (2) perception that the durability may be higher than that for stented or stentless tissue valves, and (3) belief that homografts are more resistant to reinfection. As to the last advantage, in the setting of endocarditis, most surgeons consider the homograft the valve of choice, although the data for this belief is not very robust. The disadvantages to a homograft are: (1) the increased complexity of implantation, (2) difficulty of reoperation in many patients because of calcification that develops in the wall, ¹⁵ and (3) a higher rate of SVD than originally hoped.^15,16

Ross Procedure

Donald Ross also developed the Ross procedure using the pulmonic valve and root autograft with homograft replacement of the patient’s own pulmonic valve ( Figure 14-7). The perceived advantages of this technique were freedom from anticoagulation and decreased risk of stroke. Also, in children, unlike with homografts and bioprosthetic valves, the tissue continues to grow with the patient. The disadvantages are: (1) much more complex operation than other procedures that simply replace the pathologic aortic valve, (2) the potential for dysfunction of two valves, the pulmonic homograft and the autograft, (3) the development of late aneurysms requiring reoperation, and (4) the potential for injury to the first septal perforator when mobilizing the pulmonary autograft. ¹⁷

Guidelines for Valve Choice

Choosing a valve according to the patient’s age is controversial. Findings of two major randomized clinical trials have not been consistent regarding difference in long-term survival between bioprosthetic and mechanical valves.18,19 Although both trials compared first-generation porcine valves and single-tilting disk Bjork-Shiley valves (neither valve is currently in use), the Veterans Affairs Cooperative Study ¹⁹ showed improved 15-year survival with mechanical valves than with bioprosthetic valves, but the Edinburgh Heart Valve Trial ¹⁸ showed no difference in 20-year survival. Regarding age threshold for valve choice, both the U.S. ⁹ and European ²⁰ guidelines recommend bioprosthetic valves for patients 65 years and older when only age is considered ( Table 14-1). The European guidelines, however, have a class I recommendation for mechanical valve in patients younger than 40 years and a class IIa recommendation for patients younger than 60 years, recognizing that in patients between 60 and 65 years, other factors impact valve choice. ²⁰ Guidelines for mechanical valves show that requirement for anticoagulation due to a mechanical valve in another position and a condition associated with high risk of thromboembolism are factors favoring the use of mechanical valves ( Table 14-2). On the other hand, contraindication to anticoagulation and planned pregnancy favors the use of bioprosthetic valves ( Table 14-3). Both guidelines acknowledge patient preference after informed consent, balancing the risk of long-term anticoagulation required for mechanical valves and the risk of SVD requiring re-intervention that is associated with bioprosthetic valves.

Long-term follow-up of commonly used bioprosthetic valves for the aortic position show good durability to more than 15 years in several large series ( Table 14-4). Freedom from SVD for stented bovine pericardial valves has been reported to be 82.3% at 15 years for the Carpentier-Edwards bovine pericardial valve (Edwards Lifesciences Corporation, Irvine, California) ²¹ and 62.3% at 20 years for the Mitroflow aortic pericardial heart valve (Sorin Group, Milan). ²² For stented porcine valves, the freedom from SVD has been reported to be 63.4% at 20 years for the Hancock II valve (Medtronic, Inc., Minneapolis, Minnesota) ²³ and the freedom from reoperation for SVD to be 61.1% for the Biocor valve (St. Jude Medical, Inc., St. Paul, Minnesota). ²⁴ All of these series stratified their results by patient age, which is the major determinant of durability, and found that 20-year freedom from reoperation for SVD in patients 70 years and older to be between 84.8% (Mitroflow) ²² and 100% (Hancock II). ²⁵

Controversy exists for valve choice in younger patients who would like to avoid the risk of complications associated with the use of long-term warfarin therapy required for mechanical valves. Even in patients as young as 45 years, freedom from SVD was about 85% at 10 years and 55% at 15 years with the Carpentier-Edwards pericardial valve in a study from the Cleveland Clinic. ²⁶ Also in a review of very young patients (mean age 22.7 ± 6.8 yrs), freedom from all bioprosthetic valve-related complications was 85.8% at 8 years. ²⁷ El Oakley et al ²⁸ calculated that for a 50-year-old patient, the risk of valve-related morbidity over the projected life expectancy of the patient was 108% with a mechanical valve, compared with 48% with a bioprosthesis. Different types of bioprosthetic valves also have been compared, and results may not be uniform. Rahimtoola ²⁹ compared reports of SVD with the use of bovine pericardial valves and porcine valves and found a much lower rate of SVD with pericardial valves.

For young patients who eventually develop SVD requiring intervention in the future, repeat AVR may not be mandatory. The possibility of TAVI for failed bioprosthetic valves (valve-in-valve procedure) has the potential of making AVR with a bioprosthetic valve more attractive to the patient (see Chapter 15). The feasibility of the valve-in-valve procedure has been demonstrated in single-center series.^30–33 Procedural success was 100% in one series of 23 patients, and another report of 47 patients noted one intraoperative death.^30,34 An international registry of 202 patients demonstrated 30-day mortality after valve-in-valve procedure of 8.4% and 1-year survival of 85.8%. ³⁵ Procedural concerns, however, included device malposition in 15.3% of patients, coronary ostial obstruction in 3.5%, and relatively high mean gradients, 15.9 ± 8.6 mm Hg.

Reintervention for failed bioprosthetic valves can, therefore, be performed as either an open surgical procedure or a transcatheter procedure. The safety of reoperative AVR is discussed later. Although currently considered an “off-label” use of transcatheter heart valves, the valve-in-valve procedure for failed bioprosthetic valves offers a potential application for future transcatheter technology. Safer future reinterventions may justify use of bioprosthetic valves in younger patients. Ultimately, the patient makes the decision after discussion with the surgeon and cardiologist about the risks and benefits as well as the perceived future of TAVI versus standard reoperation valve surgery.

Aortic Valve Repair

Aortic valve repair can be performed in selected patients with aortic regurgitation (AR). Unfortunately, valve repair in patients with aortic stenosis (AS) involving leaflet decalcification is not a feasible treatment option and has been associated with early postoperative AR due to leaflet scarring and late restenosis due to recalcification. As with mitral valve repair, the benefit of aortic valve repair over AVR is avoidance of prosthetic valve-related complications such as thromboembolism and infective endocarditis. Data on aortic valve repair durability are limited to experienced centers, but 10-year freedom from reoperation can be as high as 93% in tricuspid aortic valves.36–38 The durability of BAV repair, however, is less than that of tricuspid aortic valves, and repair for these patients with BAV remains controversial. No guidelines exist for aortic valve repair.

Repair of a normal, but regurgitant tricuspid aortic valve uses a combination of cusp repair and annuloplasty. Typically, one or more cusps are redundant which causes cusp prolapse. Central free margin plication at the nodulus of Arantius effectively shortens the cusp, resulting in a higher zone of coaptation with the other cusps ( Figure 14-8). ³⁹ An alternative method of cusp shortening is free margin resuspension with a continuous over-and-over suture from commissure to commissure ( Figure 14-9). This technique is also useful for closing cusp fenestrations, which are typically located near the commissures, where cusp stress is highest. The least common technique of cusp repair is cusp extension with pericardium in cases of inadequate cusp tissue ( Figure 14-10). Occasionally, leaflet perforations such as those that occur after healed endocarditis can be simply repaired with a pericardial patch ( Figure 14-11). A reduction annuloplasty may also be required in cases of annuloaortic ectasia. The simplest technique is the commissural plication ( Figure 14-12). This technique achieves narrowing of the interleaflet triangle below the commissure and reduces the diameter of the aortic root, thereby increasing coaptation of the cusp surfaces. Novel techniques utilizing an external aortic annulus ring for the purpose of downsizing the aortic annulus have been described, but these rings are not currently commercially available in the United States. ⁴⁰

Repair of the BAV can be performed with similar techniques of cusp repair and reduction annuloplasty. The goal of BAV repair is to restore a competent BAV rather than to create a tricuspid aortic valve. In cases with equal-size cusps and commissure oriented at 180 degrees to each other, repair can be performed readily as with a tricuspid aortic valve. However, in the more common type of BAV involving a conjoint (fused) cusp ( Figure 14-13), the raphe may be sclerosed and immobile, requiring additional techniques. In these cases, a triangular resection of the raphe can be performed with reapproximation of the edges to create a shortened and pliable cusp ( Figure 14-14). When tissue the conjoint cusp is inadequate, the raphe can be released from the commissure and shaved to improve cusp mobility. Judgment must be used in cases with severely sclerosed valves, because the durability of a repair of a diseased BAV may be less than even that of a bioprosthetic valve.