History of Mitral Valve Replacement

The advent of cardiopulmonary bypass in 1953 opened the door for open cardiac surgery, but the first mitral valve (MV) replacement was not performed until 1959 at the National Heart Institute by Nina Starr Braunwald (Fig. 35-1A), the first woman to perform cardiac surgery.¹ Modeling anatomy, she designed the valve using flexible polyurethane with Teflon chordae. The patient was discharged from the hospital and did well clinically for several months. However, Albert Starr (Fig. 35-1B), a cardiac surgeon in Portland, OR, and Miles “Lowell” Edwards (Fig. 35-1D), an engineer in California, abandoned the approach of mimicking the native valve structure and instead focused on valve function to develop a more hemodynamically consistent and durable valve made from a silastic ball contained within a wire cage arising from the valve housing.² This design debuted amidst a plethora of unique and distinct valve designs and revolutionized valve replacement, remaining the gold standard for MV replacement for much of that decade.

In pursuit of better hemodynamics and reduced thrombogenicity, the search for the perfect valve led to the development of other types of mechanical valves.³ Tilting disk valves were introduced by Jura Wada in 1966 (Wada hingeless) and by C. Walt Lillehei and Robert Kaster in 1967 (Lillihei–Kaster). Viking Björk worked with the Shiley laboratories to develop the Björk–Shiley tilting disk valve that gained widespread application in the 1970s. Unfortunately, this valve had a propensity for thrombosis. After the disk design was changed to decrease thrombogenicity, the struts became susceptible to fracture. The valve was subsequently taken off of the market and many were removed from patients. The Medtronic-Hall tilting disk valve introduced in 1977 was the most commonly used tilting disk valve until it was recently taken off of the market in the United States. The most commonly used mechanical valve in the world today is the St. Jude mechanical bileaflet valve first introduced in 1977.

Driven by the problem of thrombogenicity of mechanical valves, work on bioprosthetic valves yielded the Carpentier and Hancock porcine xenografts in 1969,^4,5 and Marian Ionescu introduced a glutaraldehyde-fixed bovine pericardial valve in 1971.⁶ These valves had promising early success, but structural valve dysfunction (SVD) was soon recognized to be a common complication, especially in younger patients. Further modification in the design and development of bioprosthetic valves has led to somewhat improved hemodynamics and durability, but SVD remains an obstacle to the use of these valves, especially in younger patients.

Mitral Regurgitation

Because of improvements in MV repair techniques, repair of acquired mitral regurgitation (MR) is now usually feasible and durable, with improved left ventricular (LV) function, operative mortality, and long-term survival compared to MV replacement.⁷ When repair is possible, survival is clearly better than with replacement if the MR is due to rheumatic, mixed, or degenerative disease, although the benefit of mitral surgery in ischemic MR is not as clear.⁸ According to the American College of Cardiology/American Heart Association (ACC/AHA) Guidelines, mitral surgery is recommended for symptomatic acute severe MR, chronic severe MR with New York Heart Association (NYHA) class II, III, or IV symptoms and no severe LV dysfunction, and asymptomatic patients with mild-to-moderate LV dysfunction and severe MR.⁹ Level II evidence suggests that MV surgery is reasonable for asymptomatic severe MR in some instances. The primary indication for mitral replacement in the setting of MR is when repair is not possible or the functional result would be inadequate. This may be due to valve pathology that prohibits repair, a failed attempt at repair, or surgeon experience. Pathologic and anatomic factors that may make replacement preferable to repair include rheumatic MR, the presence of calcific deposits, and shortened chordae or papillary muscles. In patients with endocarditis, destruction of the valve annulus, leaflets, chordae, or papillary muscles may preclude repair. The management of mitral valvular endocarditis is discussed in detail in Chapter 38, but repair is generally preferable to replacement when possible, although destruction of the valve apparatus may make repair impossible.¹⁰

Mitral Stenosis

As opposed to MR, acquired mitral stenosis (MS), usually due to rheumatic disease, more often requires replacement because the pathologic effects on the valve apparatus are not amenable to repair. According to the ACC/AHA Guidelines, the indications for MV replacement in MS are the following: (1) moderate (mean valve area ≤1.5 cm²) or severe MS (mean valve area ≤1.0 cm²) and NYHA functional class III or IV symptoms if they are not candidates for percutaneous balloon valvotomy or MV repair; and (2) severe MS with moderate–severe MR in symptomatic patients. Some evidence supports MV replacement in patients with severe MS with pulmonary hypertension and NYHA class I or II symptoms.⁹ Retrospective data suggest that the outcome with replacement for MS is better than with MV commissurotomy and valvuloplasty.¹¹

Hemodynamic Evaluation of Prosthetic Valves

The perfect valve remains elusive, but several characteristics define the ideal valve:

• 
  

  Minimal pressure drop

  
  
• 
  

  Trivial regurgitation

  
  
• 
  

  Minimal turbulence and stasis

  
  
• 
  

  Absence of shear stress

  
  
• 
  

  Lifelong durability

  
  
• 
  

  Biologically inert

  
  
• 
  

  Negligible thrombogenicity

All of these characteristics are related to valve hemodynamics to varying degrees. For example, greater turbulence and stagnation can contribute to thrombogenicity and shear stress can lead to hemolysis. These forces can damage tissue valves and contribute to calcification of all valve types and SVD of bioprosthetic valves. Therefore, a basic understanding of the hemodynamic assessment of prosthetic valves is fundamental to understanding the evolution and comparison of different types of prosthetic valves.^3,12,13 The normal valve area of the MV is 4 to 6 cm². The prosthetic valve housing, stent, and any other components that are within the housing decrease the true cross-sectional valve area through which blood must flow, defined as the effective orifice area (EOA). Therefore, the functional orifice of the valve, the EOA, is less than the geometric cross-sectional area of both the native annulus and the housing, and this will confer an inherent resistance and pressure gradient. The EOA may be determined by the Gorlin formula (derived from catheterization data and a hydraulic formula) or by the continuity equation, derived from acquired Doppler data.

These formulas are useful in comparing different valves in the laboratory. In the clinical setting, replacing a valve with an undersized prosthesis and a relatively small EOA relative to the patient size, also known as patient–prosthesis mismatch (PPM), is an area of controversy. One recent retrospective analysis from a single institution found that PPM was associated with recurrent congestive heart failure, postoperative pulmonary hypertension, and decreased survival after MV replacement.¹⁴ Another large retrospective study found that PPM predicted higher mortality at 6 and 12 years after MV replacement.¹⁵ However, no prospective data correlate EOA–patient size mismatch with long-term morbidity and mortality.

Another important parameter in evaluating hemodynamic function of prosthetic valves is the performance index, the ratio of the EOA to the sewing ring of the valve. The performance index is a surrogate measure for how efficiently the valve employs the total geometric area.

Prosthetic Valve Types

Prosthetic valves are classified as either mechanical or bioprosthetic. Historically, the types of mechanical valves are caged ball, tilting disk, and bileaflet valves, although only bileaflet valves are currently available in the United States. These include the St. Jude mechanical (Fig. 35-2A), Sorin mechanical bileaflet, Medtronic ATS Open Pivot (Fig. 35-2B), and On-X Mitral Standard (Fig. 35-2C). Although both stented and stentless bioprosthetic valves have been used for MV replacement,¹⁶ long-term follow-up is only available for stented bioprosthetic valves in the mitral position. Currently available bioprosthetic valves in the United States include the Carpentier–Edwards Perimount Magna pericardial bioprosthetic valve, Medtronic Hancock II (Fig. 35-2D) and Mosaic (Fig. 35-2E), as well as the St. Jude Epic mitral bioprosthetic valve (Fig. 35-2F). Homograft valves are not widely used for MV replacement. More than 50 different prosthetic valves have been introduced in the last 60 years. Unfortunately, insufficient data comparing different prosthetic valves in a prospective, randomized fashion exist to support the use of one type of valve over another. The individual characteristics of many of the different valves have been reviewed elsewhere.^3,17

Mechanical Versus Bioprosthetic Valves

Both patient and valve-related factors enter into the decision of whether to use a mechanical or a bioprosthetic valve.^17–19 Typically, mechanical valves have greater durability; however, they require lifelong anticoagulation. Bioprosthetic valves, on the other hand, are prone to SVD, but they are less thrombogenic and do not require anticoagulation. Hence, younger patients (<65–70 years old) are usually best served by a mechanical valve in order to avoid a reoperation for replacement of the bioprosthetic valve. However, patients who have contraindications to anticoagulation cannot have a mechanical valve. For example, women of child-bearing age who wish to conceive should receive a bioprosthetic valve, and then after they are no longer planning on becoming pregnant, can undergo reoperation and replacement with a mechanical valve. Patients who are likely to be noncompliant with anticoagulation should also receive bioprosthetic valves. A patient at higher risk for bleeding complications with anticoagulation should receive a bioprosthetic valve. Alternatively, if a patient already has another indication for anticoagulation, such as atrial fibrillation, a mechanical valve does not present any additional risk. However, an older patient may be at increased risk of hemorrhagic complications, so a bioprosthetic valve may be indicated. Bioprosthetic valves may be sufficient for young patients with reduced longevity such as recalcitrant intravenous drug abusers and patients on permanent hemodialysis. Technical factors must also be considered, such as a small LV cavity that may not accommodate the stents of a bioprosthetic valve. An algorithm for selection of a prosthetic valve is presented in Figure 35-3. Clearly, the decision of whether a mechanical or bioprosthetic valve should be used is complex and should involve the patient, the cardiologist, and the cardiac surgeon.