Mitral regurgitation may occur as a severe lesion (sometimes combined with aortic regurgitation) during the acute rheumatic process associated with extensive myocarditis and sometimes pericarditis and pancarditis. Anular dilation is the primary cause of regurgitation in this circumstance, with the valve leaflets frequently showing edema only and virtually normal chordae. After remission of the acute process, regurgitation may spontaneously regress, the heart becomes smaller, and anular dilation regresses. In most cases, however, there is progressive leaflet thickening, particularly of the posterior leaflet, which becomes retracted and rolled with shortening of chordae. The anterior leaflet is less thickened, and major chordae are frequently elongated, allowing leaflet prolapse. The posterior commissural leaflets are obliterated and fused, but the commissures remain more or less open. Calcification is uncommon. Anular dilation is almost invariably progressive and produces increasing regurgitation.

Mitral valve prolapse is a billowing of one or both leaflets into the left atrium during ventricular systole, with or without mitral regurgitation. Prolapse of a mitral valve leaflet occurring as an isolated abnormality * is a relatively common and complex entity, ^, occurring in 1% to 2.5% of the population. Familial mitral valve prolapse is inherited as an autosomal trait. Primary mitral valve prolapse occurs with increased frequency in patients with Marfan syndrome and certain other connective tissue disorders. The classification of Carpentier and others has enhanced our understanding of the anatomic details of prolapsing anterior and posterior leaflet components. In its severe form, mitral prolapse results in important mitral regurgitation (10% of patients ). Nonetheless, in the United States, mitral valve prolapse has been reported as the most common cause of surgically treated isolated mitral regurgitation. The primary pathologic conditions are leaflet redundancy and myxomatous thickening . The redundant and elongated leaflets no longer meet properly to support each other during systole and begin to overshoot into the left atrium. Not only is the valve thereby rendered regurgitant, but abnormal strain is also placed on the chordae. The chordae elongate and, ultimately, some rupture, increasing regurgitation. These histologic changes and severe valve redundancy are especially pronounced in younger patients with Barlow syndrome. Older patients with degenerative mitral regurgitation are more likely to have fibroelastic deficiency and less redundant valve tissue. These differences have important surgical implications because marked leaflet redundancy with more severe myxomatous changes requires more extensive reconstructive techniques. Calcifications may occur in the mitral anulus but do not appear to contribute to mitral valve dysfunction.

“Idiopathic” and more or less localized chordal rupture is usually a variant of mitral valve prolapse syndrome, in which a considerable portion of leaflet tissue is uninvolved by the myxomatous process. In most cases, the posteromedial portion of the posterior leaflet (P ₂ and P ₃ ) is involved; after chordal rupture, this becomes redundant and flail. More extensive posterior chordal rupture sometimes occurs. Localized chordal rupture may also occur in patients with Marfan syndrome.

Mitral anular calcification may occur in older patients without evident disease of the leaflets or chordae, but it may be more common in patients with myxomatous degeneration and prolapse of the mitral leaflets. Anular calcification is probably a degenerative disease, more common in elderly patients and, apparently, in females. It is also seen in patients with LV hypertrophy, particularly those with obstructive hypertrophic cardiomyopathy (see Chapter 19 ). The process involves the posterior portion of the anulus more often than other portions. Degenerative anular calcification often extends into the adjacent ventricular myocardium, and it may secondarily produce mitral regurgitation or stenosis by displacing or immobilizing the posterior leaflet. Anular calcification considerably complicates mitral repair or replacement. ^,

Papillary muscle dysfunction or rupture resulting from myocardial infarction or ischemic fibrosis can produce severe mitral regurgitation (see Chapter 10 ).

Endocarditis is a relatively uncommon cause of pure mitral regurgitation compared with its etiologic frequency in aortic regurgitation. When the aortic valve is infected and regurgitant, the regurgitant jet may infect the central portion of the anterior mitral leaflet, often producing perforation and mitral regurgitation. In the absence of aortic valve disease, a normal or abnormal mitral valve may become infected, with the destruction of cusps, chordae, or both (see Chapter 14 ).

Submitral LV aneurysms frequently result in mitral regurgitation. This unusual type of aneurysm is not ischemic in origin and occurs most often among the southern and western African black population. It may be multiloculated and have a well-defined neck immediately beneath the posterior mitral leaflet. Mitral regurgitation often coexists because of aneurysmal distortion of the posterior leaflet and leaflet prolapse. In rare instances, the aneurysm bulges into the left atrium from behind, partly obstructing the mitral orifice.

Patients with mitral regurgitation are often asymptomatic for many years, during which time LV size may steadily increase and LV contractility decrease. Eventually, effort intolerance develops, and symptoms of pulmonary venous hypertension evolve. Fluid retention and chronic heart failure, occasionally with cardiac cachexia, are characteristic of the late stage of the disease; by then, atrial fibrillation and secondary tricuspid regurgitation are usually evident.

As with mitral stenosis, important mitral regurgitation can usually be diagnosed based on history, physical examination, chest radiograph, ECG, and echocardiogram. The classic apical systolic murmur of mitral regurgitation is pansystolic, loudest at the apex, and radiates to the left axilla and left lung base. Classical auscultatory findings in mitral valve regurgitation from prolapse include one or more midsystolic clicks and a late or holosystolic murmur of mitral regurgitation.

In severe chronic mitral regurgitation, the chest radiograph usually is highly characteristic. The left atrium generally is more enlarged than in patients with mitral stenosis, and the left atrial appendage is usually prominent. Rarely, there may be marked enlargement of the left atrium (giant left atrium), although this seems to occur only in patients with rheumatic mitral regurgitation rather than other causes. The left ventricle may be enlarged, and there may be varying degrees of right atrial enlargement, depending on the amount of associated tricuspid regurgitation. The ECG may remain normal even in the presence of severe mitral regurgitation. However, a pattern of LV hypertrophy is common.

As with mitral stenosis, 2D echocardiography demonstrates the details of leaflet pathology. Both transthoracic (TTE) and transesophageal echocardiography (TEE) are useful. Echocardiographic diagnosis of mitral valve prolapse requires prolapse of 2 mm or more above the anulus in the long-axis parasternal view. Leaflet thickness of 5 mm or more increases the likelihood of mitral valve prolapse. Prolapse of a specific leaflet can be visualized (see Chapter 6 ), and Doppler color flow imaging can identify the location, direction, and magnitude of the mitral regurgitant flow. Echocardiography may also be used to estimate both the degree of LV enlargement and, by quantification of shortening fraction, ventricular contractility. Newer echocardiographic methods using 3D techniques offer precise and accurate evaluation of leaflet physiology and specific areas of regurgitation within the valve apparatus. The effective regurgitant orifice area (EROA) and regurgitant volume ( RV ) can be measured echocardiographically and have been identified as predictors of outcome following mitral valve repair (see Indications for Operation later in this section) . The American Society of Echocardiography has recommended grading mitral regurgitation as mild (EROA <20 mm ² , RV <30 mL), moderate (EROA 20-39 mm ² , RV 30-59 mL), and severe (EROA ≥40 mm ² , RV ≥60 mL). The terms “mild to moderate” and “moderate to severe” can be used to refine intermediate levels.

Left ventriculography also demonstrates the regurgitant process at the mitral valve and can show leaflet prolapse. The degree of regurgitation can usually be estimated with reasonable accuracy, although if left atrial or LV enlargement is severe, the estimate is less valid. Magnetic resonance imaging (MRI) has become a standard method for noninvasive quantification of mitral regurgitation (MR) and may be especially helpful in assessing valve regurgitation when there are multiple jets or poor imaging windows that complicate assessment by Doppler echocardiography. Mitral RV is calculated from MRI as the difference between left ventricular stroke volume (LVSV) and forward flow. With cardiac MRI, it is possible to check for errors in quantification with aortic (Ao) and pulmonary artery (PA) flows and LVSV. In the absence of an intracardiac shunt, Ao and PA flows are similar, and if there is no valvular regurgitation, then the LVSV = right ventricular stroke volume. Some studies suggest that MRI is more accurate than echocardiography in assessing the severity of MR.

Mitral regurgitation may develop acutely because of chordal rupture or infective endocarditis or may complicate the course of acute myocardial infarction (see Chapter 10 ). Symptoms and signs of severe pulmonary venous hypertension suddenly appear. The left atrium and left ventricle are normal in size or only slightly enlarged. The chest radiograph is dominated by signs of pulmonary venous hypertension, and left atrial pressure is high, as is the v wave. A mitral regurgitation murmur is often midsystolic and higher pitched compared with the pansystolic murmur of chronic mitral regurgitation.

Patients with rheumatic mitral regurgitation are more likely to have had a previous severe attack of rheumatic fever than those with mitral stenosis. The interval before the appearance of regurgitation also is shorter than for stenosis. Patients with surgically untreated but hemodynamically important rheumatic mitral regurgitation survive similarly to those with mitral stenosis. The curve is different in different environmental and genetic situations, as it is in mitral stenosis. In San Francisco, for example, the survival of such patients 5 years after initial evaluation was 80%, with a 10-year survival of 60%, whereas in Venezuela, 5-year survival was only 46%. Accelerated forms of rheumatic mitral regurgitation also occur in the same geographic areas where severe mitral stenosis appears in the pediatric population (i.e., Sub-Saharian Africa, Polynesian in New Zealand, African and Indian Subcontinent) with important symptoms by age 10 years.

The natural history of isolated mitral valve prolapse without regurgitation is highly variable, and the majority of patients have age-adjusted survival similar to that of the general population. ^, However, moderate to severe mitral regurgitation is the major predictor of cardiac mortality. Mitral regurgitation associated with mitral valve prolapse has a complex natural history that entails more than leakage of the mitral valve. Mitral valve regurgitation is not the only event in patients with mitral prolapse. Serious but rarely fatal arrhythmias may occur in patients with only mild regurgitation, and in some patients, symptoms of anxiety may mimic thyrotoxicosis, hyperadrenergic states, or hypoglycemia. Patients often have higher than normal catecholamine levels and other evidence of high adrenergic tone. Patients with hyperthyroidism have an increased prevalence of mitral prolapse.

Patients with the classic form of mitral prolapse, which includes thickening of the leaflets as well as prolapse, have an increased prevalence of infective endocarditis ^, ; those with normal leaflets (in the absence of regurgitation) do not. Both groups have a higher prevalence of stroke.

Severe mitral regurgitation requiring valve surgery rarely develops before age 50. Thereafter, the prevalence increases steeply, particularly in men. However, even men with mitral valve prolapse who have reached age 70 years have only about a 5% chance of getting severe regurgitation requiring mitral valve repair or replacement. Once important mitral regurgitation appears, however, it tends to progress. As prolapse worsens, support in systole provided by the closing of the two leaflets against each other (“stacked rifle effect”) is lost. This puts an abnormally large load on the chordae, which elongate, become thick, and eventually may rupture. This process worsens the regurgitation and accelerates the natural history of the disease.

Because of the increasing ability of cardiac surgeons to repair mitral regurgitation successfully, the natural history of severe mitral regurgitation secondary to mitral valve prolapse is of considerable importance. Two Mayo Clinic studies in the mid-1990s called attention to the increased mortality in patients with chronic flail leaflets treated medically. ^, A 2008 multicenter European study examined the natural history of severe regurgitation caused by one or both flail leaflets. Involvement was confined to the posterior leaflet in 314 patients (79%), anterior leaflet in 31 (8%), both in 46 (12%), and unspecified in 3 (1%). The long-term outcome with medical treatment included an important likelihood of major adverse events by 8 years, including atrial fibrillation, heart failure, and cardiovascular death ( Fig. 11.4 ). The need for mitral valve surgery (or death from cardiovascular causes) was nearly unavoidable 8 years after diagnosis. A subgroup analysis of asymptomatic patients with normal ventricular systolic function revealed a 5-year survival of 97% with medical treatment, but the combined occurrence of atrial fibrillation, heart failure, or cardiovascular death was 42% at 8 years. A Mayo Clinic analysis of asymptomatic patients with severe organic mitral regurgitation identified an effective regurgitant orifice (by echocardiography) >40 mm ² as a major predictor of late mortality.

Cumulative incidence of atrial fibrillation (AFib), heart failure (HF), or mitral valve (MV) surgery/cardiovascular death (CVD) during nonsurgical management of patients ( n = 394) with mitral regurgitation due to flail leaflet.

(From Grigioni F, Tribouilloy C, Avierinos JF, et al. Outcomes in mitral regurgitation due to flail leaflets a multicenter European study. JACC Cardiovasc Imaging . 2008;1:133-141.)

Patients with mitral regurgitation and ruptured chordae tendineae may have a slow, insidious development of symptoms. Ruptured chordae of the anterior or posterior leaflet or both are often found at operation, and the mitral valve leaflets have the appearance of myxomatous degeneration. Ruptured chordae may be present in patients with prolapse without important symptoms. Grenadier and colleagues found 11 (8%) of 134 patients with mitral valve prolapse to have ruptured chordae and few or no symptoms.

By contrast, important mitral regurgitation produced acutely by chordal rupture may occur in patients with no previous valve leakage. This happens predominantly in middle-aged men. Often, it is a complication in the life history of patients with midsystolic clicks and only trivial murmurs, but without previous evidence of mitral regurgitation. The anterior mitral leaflet and its chordae are frequently entirely normal, with the disease process limited to the medial aspect of the posterior mitral leaflet (P ₂ and/or P ₃ ). In this group of patients with acute and severe symptoms, presumably initiated by sudden chordal rupture, the left atrium and left ventricle are small, left atrial pressure is high, the v wave is greatly accentuated, and there is substantial clinical and radiologic evidence of pulmonary venous hypertension. TTE and TEE are diagnostic. If the patient survives the acute event, the left ventricle and left atrium enlarge moderately over time, and symptoms may lessen with appropriate medical management (as shown experimentally as well ). The patient gradually regains a feeling of well-being. One year later, left atrial and ventricular enlargement may not have progressed, the left ventricle and left atrium seemingly adapting to the volume overload. Years may pass before the self-aggravating tendency of mitral regurgitation results in increased regurgitation volume. After this, the classic natural history of important mitral regurgitation evolves.

In other patients with severe acute manifestations, symptoms improve only mildly with intense medical treatment. Although most patients survive, LV and left atrial enlargement progress steadily in the months after onset. Such patients have a large mitral regurgitant flow, and without surgical intervention, they die within 2 to 5 years.

Infective endocarditis on a previously mildly abnormal mitral valve may produce acute mitral regurgitation. The natural history of that condition is similar to that described for acute chordal rupture, except that early mortality is higher. Infrequently, death is related to uncontrolled infection. Infective endocarditis is discussed in detail in Chapter 14 .

Tricuspid valve regurgitation may be associated with any type of mitral valve disease. Consequently, tricuspid anuloplasty or, rarely, replacement may have to be performed concomitantly with mitral valve surgery. Techniques for these operations are described in Chapter 13 .

Mitral valve replacement begins with exposure of the mitral valve as described for mitral commissurotomy or repair, using one or two valve hooks to display the leaflets. To excise the valve, an incision is begun with the knife in the center of the anterior mitral leaflet at approximately the 12-o’clock position about 2 mm from the anulus because, at that point, the leaflet tissue is typically pliable and free of disease ( Fig. 11.16 ). The incision is carried leftward and rightward with a knife or scissors and onto the commissural leaflet tissue at the anterolateral and posteromedial commissures. The incisions through the commissural leaflet tissue are kept next to the anulus so that the anterior and posterior leaflets stay together. The underlying papillary muscle and fused chordae are cut just in front of the incision for better exposure.

Mitral valve replacement: Access to the left atrium and exposure of the valve as for repair (see Figure 11.5 ). As described in the text, incision in mitral leaflet is begun with knife anteriorly and about 2 mm from anulus, where leaflet usually is pliable and relatively free of disease. As incision is carried leftward with knife or scissors toward anterolateral commissure, underlying papillary muscle and fused chordae come into view and are incised. As incision is carried across anterolateral (illustrated here) and posteromedial commissural areas, care is taken to stay close to anulus so valve is kept in one piece. This greatly facilitates completing the valve excision. Ao, Ascending aorta; SVC, superior vena cava.

Classically, the excision is continued to the posterior leaflet from both sides. Ordinarily, the posterior leaflet and its chordae are left in place when they are thin and pliable. Thus, only the anterior leaflet is fully excised (but see “ Chordal Sparing Procedure ” later in this section). If this is not possible because of extensive disease, the secondary chordae that tether the posterior leaflet to the underlying ventricular myocardium are cut. Ideally, at least the tertiary chordae attached to the anulus should be preserved. Preservation of the subvalvular apparatus of the mitral valve has been linked to better LV performance and patient survival (see later). When subanular calcification is present and can be excised without disturbing the anulus or the myocardium, it is removed. Otherwise, calcification should be left in situ because overzealous efforts may damage the circumflex coronary artery or precipitate postrepair ventricular rupture.

The mechanical prosthesis or bioprosthesis can be sewn into place with well-distributed interrupted simple sutures, with horizontal mattress sutures using pledgets on the atrial side, or with a continuous size 0 polypropylene suture ( Figs. 11.17 and 11.18 ). All methods are associated with an extremely low prevalence of periprosthetic leakage (see “ Periprosthetic Leakage ” under Modes of Death later in this section). A technique employing interrupted pledgeted mattress sutures with the pledgets on the ventricular side may be chosen when heavy calcification remains in some areas of the anulus or, rarely, when exposure is particularly difficult. With these techniques, all sutures are placed in the heart first and then passed through the valve sewing ring. The prosthesis can then be lowered into position, and the sutures tied and cut. Care is to be taken to avoid protruding suture ends as they can damage the leaflets of bioprostheses or impinge the leaflets of mechanical valves.

Mitral valve replacement: continuous suture technique. (A) One end of a double-armed size 0 polypropylene (or polyester) suture buttressed with a felt pledget is passed through mitral anulus just posterior to anterolateral commissure. Suture is then passed through prosthetic sewing ring, and valve without holder is lowered into place. Using about four throws, suture line is carried to the left, anteriorly passing stitches from anulus to sewing ring and taking deep bites, but avoiding noncoronary cusp of aortic valve. Suture is held midway across distance to posteromedial commissure. (B) Other end of the suture, having been placed through mitral anulus as a mattress, is placed into prosthetic sewing ring and continued from anulus to sewing ring with four or five throws. It is helpful to move prosthesis in and out of anulus to accommodate needle passage. Suture is held. A second double-armed pledgeted suture is begun as a horizontal mattress just posterior to posteromedial commissure. Its ends are carried with four or five throws in each direction, first anteriorly and then posteriorly, to meet and tie with previously held ends. Knots then lie behind leaflet guards of the St. Jude Medical prosthesis depicted here.

Mitral valve replacement: interrupted suture technique. (A) Double-armed size 0 polyester sutures are generally placed with pledgets on atrial aspect of anulus. It is convenient to place one arm of a mattress suture just behind posteromedial commissure, reserving other end for placement later and holding suture for exposure. (B) Posterior sutures are sequentially placed counterclockwise beginning at anterolateral commissure and may either be held or placed in prosthetic sewing ring. (C) Anterior suture line usually proceeds clockwise from anterolateral commissure to posteromedial commissure. Along whole circumference, suture bites encompass base of residual leaflet. Clockwise from anterolateral commissure are the aortic cusps, conduction system, atrioventricular septum, coronary sinus, and circumflex coronary artery, which are at risk when bites are taken too deeply. With all sutures in place and on tension, valve is lowered into place and sutures tied, with about 12 mattress sutures used. Before tying, device struts are inspected to ensure that no suture is looped around a strut or caught within the device. If the tissues a have good consistency, single-bite interrupted sutures, from anulus to prosthesis, can also be used, with extra care for uniform spacing.

After completing the valve insertion, a catheter is passed through the valve into the left ventricle to act as a frustrator. The steps for exiting from the left side of the heart and de-airing are as described for mitral stenosis.

Information based on experimental observation and clinical experience indicates that retention (rather than resection) of the mitral tensor apparatus at mitral valve replacement results in better LV function postoperatively (see “ Cardiac Performance ” later in this section). To insert a prosthesis while retaining both the anterior and posterior chordal attachments, the anterior leaflet may be split centrally, or using a triangular resection of A ₂ , and folded laterally, and the posterior leaflet left as is ( Fig. 11.19 A-F). Alternatively, both the intact anterior and intact posterior leaflets can be folded toward their bases, and then a prosthesis inserted. These maneuvers (including folding the anterior leaflet) do not result in LV outflow tract obstruction or prosthetic obstruction when done for mitral regurgitation. The operative decision should be individualized based on the patient’s anatomy, pathology, and ventricular function; therefore, surgeons should be familiar with more than one surgical preservation technique. Although appropriate for functional and degenerative mitral regurgitation, the role of chordal sparing in rheumatic valve disease is less well understood, with limited evidence supporting total chordal sparing.

Mitral valve replacement: chordal sparing procedure. (A–B) When it is appropriate to preserve anterior (and posterior) tensor apparatus, midportion of anterior leaflet is removed as a trapezoid, triangle, or rectangle. Lateral and medial aspects of anterior leaflet remain and retain their chordal attachments. (C–F) Residual portions of leaflet are folded back to be sutured to anulus. Leaving anterior leaflet totally intact may risk left ventricular outflow tract obstruction or, in the case of replacement for mitral stenosis, residual left ventricular inflow obstruction. Prosthesis is sutured into place using previously described interrupted or continuous suture technique. In either case, sutures surround retained leaflets, adding strength and purchase to the repair.

The design of mechanical valve prostheses has evolved over the last 5 decades. A discussion of modern mechanical valves is complicated by two factors: (1) changing availability of specific valves because of continuing introduction and withdrawal of new models and (2) variability in prevalence of use in different countries, in part because of differing regulatory requirements. Here, we discuss mechanical valves currently available and in use in at least several countries.

The major types of mechanical heart valves were ball and cage, tilting disc, and bileaflet valves. The first two designs are no longer used. The Starr-Edwards (S-E) mitral ball-valve prosthesis (model 6120; Fig. 11.21 ), introduced in 1965; the Bjork-Shiley Monostrut valve ( Fig. 11.22 ), in the market since 1982; the Medtronic-Hall valve (1978); and the Omniscience/Omnicarbon valves (1978) have since been discontinued. Bileaflet valves have variable design features, but all contain two leaflets that swing apart during opening, providing three separate flow areas. The leaflets are guided by a hinge or pivot mechanism that acts to retain the leaflets and defines their opening angle. Nearly all valves are rotatable following implantation to address potential impingement on leaflet closure by surrounding tissues.

Starr-Edwards mitral ball-valve prosthesis.

Bjork-Shiley Monostrut mitral valve prosthesis.

Current mechanical valve design has focused on excellent hemodynamics, lifetime durability, and maximal resistance to thromboembolism. Orifice diameter, occluder characteristics, the opening angle, and leaflet or occluder orientation to the plane of the mitral orifice influence transvalvar gradients. Dynamic regurgitation is a feature of all mechanical prosthetic valves and is increased with larger effective orifice size and time needed for disc closure. A small right ventricle is beneficial because it minimizes stasis and decreases platelet aggregation.

The St. Jude Medical (SJM) mitral valve ( Fig. 11.23 ), available worldwide, was first implanted in October 1977. It is a bileaflet device, and both the leaflets and orifice ring are fabricated from pyrolytic carbon. The leaflets are orifice oriented, and a pivot system supports closing forces. The pivot guards are raised above the housing, and leaflet motion is by rotation. Two relatively high-velocity regurgitant jets (seen on echocardiography) wash the pivot recess, producing approximately 10% regurgitation. The original SJM was not rotatable, but the newer Masters Series allows rotation to an “anatomic” or “antianatomic” position of the leaflets. The sewing cuff on the mitral prosthesis has a supra-anular configuration and is provided in a standard (M-101) or expanded (MEC-102) polyester fiber cuff. It is also available with a PTFE fiber cuff (MT-103). Hemodynamic performance is excellent, as reflected in a relatively large effective orifice area. The favorable hemodynamics in the smaller sizes is an advantage in small children. Regurgitant flow may be greater than optimal, particularly at low heart rates. Between 88% and 96% of patients with device placement in the mitral position are free from a thromboembolic event at 5 years after operation ( Fig. 11.24 and Table 11.2 ). Linearized rate of the first thromboembolic event is about 1.5% to 1.75% per patient-year. Up to 75% of thromboembolic episodes occur when anticoagulation is inadequate (international normalized ratio [INR] < 2.5). Thrombosis is uncommon. Incidence is approximately 0.1% per patient-year and is usually associated with inadequate anticoagulation. Mechanical failure is uncommon. The SJM mitral valve is provided in sizes 19 to 33 mm.

St. Jude Medical mitral valve prosthesis.

Freedom from major thromboembolism (including thrombosis and fatal) following mitral valve replacement with CarboMedics and St. Jude Medical prostheses. CM, CarboMedics; PNS, not statistically significant; SE, standard error; SJM, St. Jude Medical.

(From Jamieson WR, Miyagishima RT, Tyers GF, Lichenstein SV, Munro AI, Burr LH. Bileaflet mechanical prostheses in mitral and multiple valve replacement surgery: influence of anticoagulant management on performance. Circulation . 1997;96:II134.)

The CarboMedics (Livanova) mitral valve ( Fig. 11.25 ) is a low-profile bileaflet prosthesis constructed of pyrolytic carbon. It has been in use since 1985 and was approved by the Food and Drug Administration (FDA) in 1992. The leaflet retention mechanism is within the pyrolyte housing and has no pivot guards, struts, or orifice projections. Leaflet motion is by rotation, with relatively complete seating but allowing four small regurgitant jets (typically noted on echocardiography). The valve housing is rotatable within a carbon-coated polyester sewing ring. A modification of the sewing ring in another model (CarboMedics Optiform) allows flexibility in supra- or subanular implantation. The valve has a favorable record for freedom from thromboembolism , with about 93% of patients event-free at 3 years (see Table 11.2 and Fig. 11.24 ), and the linearized rate of thromboembolism is 1.0% to 2.5% per patient-year. Thrombosis is uncommon. Prevalence is approximately 0.5 per 100 patient-years. Hemodynamics are generally good with this valve, but the 25-mm CarboMedics valve has been found to have a higher diastolic gradient than other bileaflet prostheses, especially at high flows. This valve is not recommended for patients with a small mitral valve orifice. The mitral device is provided in sizes 23 to 33 mm (small sizes also available).

CarboMedics mitral valve prosthesis.

The Medtronic Open Pivot, a successor of the ATS valve, in clinical use since 2000, is a bileaflet ( Fig. 11.26 ) solid pyrolytic carbon valve with a unique pivot design in which the pivot areas are entirely within the orifice ring, and the valve leaflets hinge on convex pivot guides on the ring. This differs from other mechanical valves that have cavities in the hinge area. The open pivot design feature is intended to decrease blood stasis and thrombus formation near the hinge points. The valve design minimizes overall valve height and generates a larger orifice area. Valve noise is reportedly reduced by this design. The mitral prosthesis is available in sizes 25 to 33 mm.

Medtronic ATS mitral valve prosthesis.

The On-X mechanical valve (CryoLife) ( Fig. 11.27 ), approved by the FDA in 2002, is a pure pyrolytic carbon valve prosthesis with a bileaflet design similar to other bileaflet prosthetic valves. The pyrolytic carbon structure is stronger than the silicon-alloyed pyrolytic carbon used in other mechanical prostheses. The On-X valve contains a flared inlet that produces a higher volume of flow with increased washing to minimize flow stagnation. The leaflets open 90 degrees, with “soft landing” leaflets designed to reduce blood element stress. The aortic On-X valve is currently considered to be the least thrombogenic of all bileaflet valves. Indeed, the On-X aortic prosthesis is the only mechanical valve with FDA and CE approval for use with lower INR (1.5 to 2.0). The AHA and ACC guidelines state that a lower INR may be reasonable for patients with the mechanical On-X Aortic Valve, ^, but it is unclear whether this indication applies to the mitral prostheses. The On-X mitral prosthesis comes in sizes 23 to 33 mm and is available with an intra- and supraanular sewing ring.

On-X mechanical mitral valve prosthesis.

A number of stent-mounted bioprosthetic devices are in clinical use for mitral valve replacement, including those with leaflets made of xenograft aortic valves, bovine or equine pericardium, and allograft aortic valves, fascia lata, and dura mater. Commercially available stented bioprosthetic mitral valves contain either porcine aortic valve leaflets or leaflets constructed from pericardium. These stented valves are designed to mimic flow characteristics of the in situ aortic valve. Pericardial prostheses have been greatly modified since the original Ionescu-Shiley valve, which demonstrated poor durability, manifested frequently by leaflet tearing. Bioprosthetic valves are preserved with glutaraldehyde, which cross-links collagen fibers and reduces normal turnover of extracellular matrix tissues. Glutaraldehyde fixation of porcine valves is achieved at high (60–80 mmHg), low (<3 mmHg), or zero pressure conditions. Lower fixation pressures in newer-generation porcine valves may reduce the tendency for calcification. Current pericardial valves utilize glutaraldehyde fixation at low pressure and mounting of the pericardium completely within the stent, producing less leaflet abrasion and potentially greater durability.

The major advantage of bioprosthetic mitral valves is their resistance to thromboembolism, which is sufficiently uncommon that long-term anticoagulation with warfarin is not recommended for most patients unless other risk factors are present (see “Management of Anticoagulation with Prosthetic Valves” under Special Situations and Controversies). The advantage of not requiring anticoagulation is accompanied by major susceptibility to structural valve deterioration in both pericardial and porcine valves. The end result of valve deterioration is progressive mitral stenosis and/or regurgitation, which may (rarely) occur as early as 4 to 5 years postoperatively. By 10 years, prevalence of primary tissue valve failure is about 30%, and by 15 years, 35% to 65%. The degeneration process is greatly accelerated in younger patients.

The Medtronic-Hancock glutaraldehyde-preserved porcine xenograft (standard model) , introduced in 1970, consists of glutaraldehyde-preserved porcine aortic valves mounted on a polypropylene stent with polyester fabric and a molded silicone rubber sewing ring. It is FDA-approved and available for use in the mitral position worldwide. The Hancock II porcine bioprosthesis, the current updated version of the Hancock I, was first implanted in 1982. The leaflets are fixed in glutaraldehyde and treated with sodium dodecyl sulfate to retard calcification. Hemodynamic, thromboembolic, thrombotic, and tissue failure characteristics are similar to those of the Carpentier-Edwards prosthesis. Khuri and colleagues summarized hemodynamic characteristics of the Hancock valve. The reported actuarial freedom from structural valve deterioration in the mitral position was 76% at 15 years for patients aged 65 years or older. The valve is available in sizes 25 to 35 mm.

The Medtronic Mosaic bioprosthetic mitral valve ( Fig. 11.28 ) is a porcine stented tissue valve introduced into the United States in 2000. It incorporates the stent of the Medtronic Hancock II valve, with the addition of α-amino-oleic acid treatment to reduce the potential for calcification.

Medtronic Mosaic mitral valve prosthesis.

The Carpentier-Edwards PERIMOUNT mitral pericardial valve ( Fig. 11.29 ) is a glutaraldehyde-fixed stent-mounted valve introduced into clinical use in 1984 and approved by the FDA in 2000. It has a profile and configuration similar to the Carpentier-Edwards aortic pericardial device. Hemodynamic performance is comparable with other mitral bioprostheses. Ten-year freedom from thromboembolism was 93% ± 3.0% in the series reported by Poirer and colleagues. Freedom from structural valve deterioration at 10 years was 81% ± 7.0%, although no failures were noted in patients aged 70 years or older. The newer Magna and Magna Ease mitral valve adds a nonreversible fixation process (ThermaFix) to reduce residual glutaraldehyde and phospholipids that contribute to pericardial leaflet calcification. In addition, the Magna Mitral bioprosthesis is designed with an asymmetric sewing cuff to maximize anular conformity. It is available in sizes 25 through 33 mm.

Carpentier-Edwards PERIMOUNT Plus mitral valve prosthesis.

(Courtesy Edwards Lifesciences, Irvine, Calif.)

The St. Jude Medical/Abbott Biocor stented porcine heterograft ( Fig. 11.30 ) incorporates three separate porcine leaflets with low-pressure fixation. Actuarial freedom from reoperation due to structural valve degeneration has been reported at 96% at 15 years for patients older than age 60. This prosthesis is characterized by low stent posts in the mitral position. This prosthesis was recently withdrawn from the market and substituted by the St. Jude Epic stented porcine valve that carries the same design as the Biocor valve, with the addition of a proprietary anticalcification treatment designed to increase valve durability. To date, no long-term data are available on the effect of this anticalcification treatment on valve durability in humans. In 2021, the FDA approved the St. Jude Epic Plus, an enhanced design of both the valve and the holder.

St. Jude Medical Biocor stented mitral valve prosthesis.

In some circumstances, valves designed for aortic valve replacement may be used for mitral valve replacement with special modification of the techniques.

Over the last two decades, a gradually increasing number of cardiac surgical centers have gained experience with limited access approaches to the mitral valve, either with direct vision or video-assisted techniques (most recently with robotic technology). Successful repair of mitral regurgitation resulting from degenerative disease of the posterior leaflet, anterior leaflet, or both has been reported with a variety of these techniques. In experienced centers, medium-term survival and preservation of valve competence has been equivalent to standard techniques via median sternotomy. ^{, ,} A right paramedian incision with division of the third and fourth costal cartilages, a “J” incision beginning at the sternal notch and extending to the fourth intercostals, ^, and a partial right lower sternotomy have all gained popularity, with good reported outcomes.

More recently, numerous reports have recommended a small right anterior thoracotomy, which appears to be currently the most widely applied limited access approach. ^, Right thoracotomy was employed by Lillehei and colleagues in their first open operations for mitral valve disease and continues to be used by some surgeons. Because the anular plane of the mitral valve lies in the sagittal plane of the body, surgical intervention on the mitral valve is particularly well suited to a right thoracotomy approach as the valve will be viewed en face. The right thoracotomy approach tends to heal well as a result of coverage by the overlying pectoralis muscle and avoids operating in the xiphisternal region, the area most susceptible to wound breakdown after sternotomy. A particular advantage of this approach may be in reoperations after median sternotomy when coronary artery bypass grafting (CABG) was performed and when the patent venous or arterial grafts are at risk for injury during resternotomy. Particularly in females, this approach may be aesthetically more appealing by using a submammary skin incision.

Although certain details are variable among experienced surgeons, the following principles are generally applicable. The small right thoracotomy is usually performed through a 4- to 6-cm right inframammary incision ( Fig. 11.31 ), and the chest is usually entered through the fourth intercostal space (occasionally the third). Double-lumen endotracheal tube intubation is employed to enhance visualization, and use of a TEE probe is routine. Right femoral artery cannulation for CPB is most commonly employed, although direct aortic cannulation can be nearly routinely applied through a third intercostal space approach. A long, vacuum-assisted, femoral vein cannula designed for vacuum-assisted venous return, with or without a second 15 Fr to 17 Fr cannula, is inserted via the right internal jugular vein using the Seldinger technique. In primary cases, aortic occlusion is managed with direct external clamping (often with a flexible clamp or Chitwood clamp) or endoaortic balloon occlusion (less reliable). Cardioplegia can be administered antegradely directly into the ascending aorta (and the catheter subsequently used for de-airing procedures) or retrogradely via a catheter placed through the right atrium into the coronary sinus under TEE guidance. Alternatively, a coronary sinus catheter for retrograde cardioplegia can be inserted through the internal jugular vein with positioning by TEE, but this requires anesthesiologist expertise and is less reliable for routine use. A left atrial incision in the right interatrial groove anterior to the entrance of the right pulmonary veins provides exposure of the mitral valve. De-airing is facilitated by flooding the surgical field with CO ₂ , placing a urinary catheter or vent across the mitral valve during initial de-airing, and venting the ascending aorta through the antegrade cardioplegia catheter.

Minimally invasive exposure and cannulation technique via right anterolateral thoracotomy. Additional subcentimeter ports are added for cardiotomy suction, Chitwood clamp, camera, and retraction.

Under certain circumstances, mitral valve surgery on the beating heart at near normothermia has been advocated. As long as the necessary precautions for preventing ejection of air into the aorta are understood and followed, this technique can be an important option in the presence of prior CABG, severe LV dysfunction, or extensive arteriosclerotic disease of the ascending aorta.

Left thoracotomy has been used in the past, but because venous cannulation is difficult, this approach cannot currently be recommended.

In the evolution of limited-access approaches to the mitral valve, current surgical philosophies are divided between a greater emphasis on techniques for direct visualization using more standard surgical techniques and instrumentation, and video-assisted endoscopic or robotic approaches. Whether the enhanced visualization using thoracoscopic and robotic techniques provides sufficient benefit to justify the more expensive technology will await further long-term studies.

The first robotic endoscopic mitral valve repair was performed by Carpentier in 1998 using an early prototype of the da Vinci instrument, and in 2002, the FDA approved the da Vinci™ platform for mitral valve surgery in the United States. Although robotic surgery is still only performed in selected centers, the number of surgeons interested in the technique is increasing. In a recent paper, Rao and colleagues from Philadelphia reported a series of 786 patients who underwent minimally invasive mitral surgery in their institution. They performed a matched analysis and concluded that mitral valve surgery via a classic endoscopic approach yielded similar clinical outcomes when compared to robotic endoscopic surgery. They demonstrated that both classic endoscopic and robotic endoscopic approaches allow repair of degenerative mitral valves with excellent short- and medium-term outcomes in a tertiary referral center. Patients undergoing robotic endoscopic mitral repair had a significantly longer CPB time when compared to the classic endoscopic cohort, with 148 minutes of CPB in the robotic endoscopic cohort compared to 133 minutes in the classic endoscopic group, P =.03 ( Fig. 11.32 ).

Patients undergoing robotic mitral valve repair had significantly longer CPB time compared to traditional endoscopic approaches.

(From Rao A, Tauber K, Szeto WY, et al. Robotic and endoscopic mitral valve repair for degenerative disease. Ann Cardiothorac Surg . 2022;11(6):614-621.)

Cross clamp time was not statistically significant between robotic endoscopic and classic endoscopic groups, 148 ± 37 and 133 ± 42 minutes, respectively. Longer CPB and ischemic times remain a problem, but with experience, these times are expected to decrease. Both techniques introduce a degree of complexity to the procedure, hence requiring experienced surgeons and mandating careful patient selection.

Occasionally, in large patients with small left atria, exposure is not optimal with the approach described earlier under Technique of Operation. The “superior approach” through the most superior aspect of the left atrium, where it appears in the transverse sinus, is an attractive alternative ( Fig. 11.33 ). ^, The superior vena cava is mobilized and retracted laterally, and the aorta is retracted to the left (after clamping it and injecting cardioplegic solution because retraction may make the aortic valve regurgitant) ( Fig. 11.33 A). A transverse incision is made in the roof of the left atrium as far from the aortic root origin as possible ( Fig. 11.33 B).

“Superior approach” involves retracting superior vena cava laterally and aorta to the left. Roof of left atrium is opened with incision directed leftward and posteriorly, leaving an adequate rim of atrial tissue anteriorly so as not to infringe on the mitral anulus.

The mitral valve is found to be very accessible; often, only stay sutures are required for retraction. Because of this, studying the competence of the valve after commissurotomy or repair is easier. Closure of this incision must be accurate, catching all layers (including the endocardium) with each stitch.

The superior approach is also useful when using a short, limited upper sternotomy. A single-stage venous cannula is inserted through the right atrial appendage, and distal ascending aortic cannulation is combined with antegrade cardioplegia.

Approach through the right atrium and across the atrial septum can be useful, especially when additional procedures are necessary for the tricuspid valve or the atrial septum. Exposure may be limited, and retraction is impeded by concern about injury to the AV node. Exposure can be improved by extending the incision onto the roof of the right atrium. A variation of this approach is that of Guiraudon and colleagues ( Fig. 11.34 ). Incision is begun high on the right atrial free wall and is extended caudally on the interatrial septum. It is completed by extending the septal incision to the superior aspect (roof) of the left atrium. Exposure is excellent, although closure is somewhat tedious. It is useful for patients with deep chests and those with small left atria.

Biatrial approach of Guiraudon and colleagues, or “transplant” incision. (A–B) Right atrium is opened transversely near atrioventricular groove. Atrial septum is incised vertically through fossa ovalis, and incision is extended superiorly to meet the right atrial incision. Where these two incisions meet, incision is directed leftward through roof of left atrium toward left atrial appendage. Walls of atrial septum and left atrium are deflected by stay sutures. Mitral valve exposure is excellent. (C) Atrial septum and roof of left atrium are closed with separate sutures, meeting superiorly. Right atrial portion is then closed. Ao, Ascending aorta; LA, left atrium; PT, pulmonary trunk; RA, right atrium.

Good results have also been reported with a “minitransseptal” incision in the atrial septum extending from the inferomedial edge of the fossa ovalis up toward the medial base of the superior vena cava without incising the superior surface of the left or right atrium.

In the current era, hospital mortality after either closed or open commissurotomy approaches zero, and late survival is similar in risk-adjusted comparisons. ^, A difference exists only in prevalence of postcommissurotomy mitral regurgitation (see “ Mitral Regurgitation ” in text that follows). Intermediate-term survival is good in those with favorable immediate hemodynamic results ( Fig. 11.35 ). In many institutions, percutaneous catheter valvotomy has almost completely replaced surgical commissurotomy for isolated mitral stenosis. Procedure mortality approaches zero, and risk of complications (bleeding, severe regurgitation) requiring urgent operation is low. However, open surgical commissurotomy appears to result in better valve orifice areas that should lead to longer durability.

Survival considering cardiovascular-related deaths after percutaneous mitral commissurotomy. Survival was poorer in patients who had poor immediate results, and notably, only 19% ± 4.0% were alive and free from surgery at 5 years (curve not shown).

(From Iung B, Garbarz E, Michaud P, et al. Late results of percutaneous mitral commissurotomy in a series of 1024 patients. Analysis of late clinical deterioration: frequency, anatomic findings, and predictive factors. Circulation . 1999;99:3272.)

Mitral commissurotomy is not curative with either open or closed (or balloon) commissurotomy, with survival progressively diverging from that of the general population ( Fig. 11.36 ). However, few late deaths result directly from the effects of recurrent or residual mitral stenosis or regurgitation. Rather, they result from thromboembolism or early or later sequelae of reoperation and mitral valve replacement.

Survival after mitral commissurotomy by either closed or open technique. (A) Survival. Each circle represents a death, positioned according to Kaplan-Meier estimator, vertical bars represent 70% confidence limits (CL), and numbers in parentheses are patients at risk. Blue line enclosed within dashed CLs is the parametric survival estimate. Red curve is survival of an age/gender/ethnicity–matched general population. (B) Hazard function for death. Note steadily rising single hazard phase. Dashed lines are CLs. Red line is hazard for an age-gender-ethnicity–matched general population.

(From Hickey MS, Blackstone EH, Kirklin JW, Dean LS. Outcome probabilities and life history after surgical mitral commissurotomy: implications for balloon commissurotomy. J Am Coll Cardiol . 1991;17:29.)

Mitral regurgitation is a risk of mitral commissurotomy by any technique but occurs in only 2% to 5% of patients who undergo open commissurotomy and in about 10% of patients who undergo closed commissurotomy. Rarely does the newly developed regurgitation require immediate operation, but it may lead to reoperation within a few months. Mild postcommissurotomy mitral regurgitation has little effect on survival or need for mitral valve replacement, but important postcommissurotomy regurgitation adversely affects both ( Fig. 11.37 ). Prevalence of new important mitral regurgitation is about 10% after percutaneous balloon mitral commissurotomy and may require early surgery.

Nomograms representing solutions of multivariable equations, illustrating effect of postcommissurotomy mitral regurgitation on (A) risk-adjusted survival, and (B) freedom from mitral valve replacement. Dashed lines are 70% confidence limits. OR, Operating room.

(For details of equations, see Hickey MS, Blackstone EH, Kirklin JW, Dean LS. Outcome probabilities and life history after surgical mitral commissurotomy: implications for balloon commissurotomy. J Am Coll Cardiol . 1991;17:29.)

Increase in calculated mitral valve area (or orifice size) produced by commissurotomy varies greatly and depends not only on the surgical opening but also on leaflet pliability and extent of subvalvar obstruction from fused chordae.

Somewhat better orifices can be obtained by open than closed commissurotomy. Antunes and colleagues achieved valve areas of 2.89 ± 0.49 cm ² in patients with pliable valves and minimal subvalvar disease, as well as good but smaller orifices in patients with extensive subvalvar diseases. After a mean follow-up of 8.5 years, the mean valve area was 2.37 ± 0.42 cm ² ( Fig. 11.38). Ben Farhat and colleagues reported an equivalent increase of mitral valve area (0.9–2.2 cm ² ) for open commissurotomy and percutaneous balloon valvotomy that was greater than for closed commissurotomy (0.9–1.6 cm ² ). The superior hemodynamic results continued over a 7-year follow-up period.

Post open mitral commissurotomy areas, before, after and at late follow-up. Note that after commissurotomy, areas are significantly greater than those commonly reported after balloon commissurotomy. Mean valve area was still 2.37 cm ² after a 10-year follow-up.

(From Antunes MJ, Magalhães MP, Colsen PR, Kinsley RH. Long-term follow-up of open mitral commissurotomy: the continuing role for surgical treatment in developing countries. J Heart Valve Dis. 2001;9(4):472–477)

A secondary effect of increased orifice size is lowering of left atrial pressure, although it frequently remains above normal. On average, left atrial pressure at rest is about 12 mmHg after valvotomy, increasing to about 17 mmHg on exercise. LV end-diastolic pressure often is modestly higher after commissurotomy.

Cardiac output is usually increased by operation, and the increase at rest and exercise correlates well with the increase in calculated valve area. Rp usually falls immediately, especially in young patients, as verified by Block and Palacios during percutaneous balloon commissurotomy. Pulmonary artery pressure usually falls, which correlates well with the decrease in left atrial pressure and Rp.

Successful mitral commissurotomy may reduce the likelihood of thromboembolism, although accurate comparisons are not available. About 90% of patients are free of a thromboembolic event 8 to 10 years after open commissurotomy. ^, The linearized rate of thromboembolism has been reported at 1% to 2% per patient-year. ^{, , , ,} Atrial fibrillation, older age, and a history of thromboembolism preoperatively are reported risk factors for postcommissurotomy thromboembolism. ^, Also, a postcommissurotomy thromboembolic event predisposes the patient to further events. The type of surgical commissurotomy, open or closed, is not a risk factor.

In properly selected patients, successful open or closed mitral commissurotomy results in dramatic relief of symptoms. More than 90% of patients are in NYHA functional class I or II during the first 1 or 2 postoperative years ( Fig. 11.39 ). ^{, , ,} Redevelopment of symptoms results from gradual loss of leaflet pliability, progression of subvalvar pathology, and increase of valvar calcification resulting from continuing rheumatic activity and scarring process. Although recurrence of rheumatic fever may accelerate this pathology, progression seems to occur even without further rheumatic episodes. Thus, NYHA functional class correlates well with estimated area of the mitral orifice late postoperatively; the mean value is 2.0 cm ² in class I patients, 1.7 in class II, and 1.6 in class III. Eventually, although not until 20 years postoperatively in some patients, most patients lose their good functional status and return with restenosis or new-onset regurgitation. Absence of leaflet pliability in the presence of valvar calcification is a risk factor for the rate of decline in functional status after all types of commissurotomy. Type of surgical commissurotomy (open or closed) has not been shown to be a risk factor.

Comparison of preoperative and postoperative New York Heart Association functional status in 123 patients surviving an average of 48 months after open mitral commissurotomy. Class IIa, Breathlessness with unusually strenuous activity; class IIb, breathlessness with ordinary activity.

(From Smith WM, Neutze JM, Barratt-Boyes BG, Low JB. Open mitral valvotomy: effect of preoperative factors on result. J Thorac Cardiovasc Surg . 1981;82:738.)

As a consequence of this, most patients undergoing mitral commissurotomy by any technique will require another procedure at some time, generally mitral valve replacement, because of gradual loss of leaflet pliability, progression of subvalvar pathology, and increase of valvar calcification. ^{, , , ,} About 20% of patients undergoing surgical commissurotomy require valve replacement within 10 years, and about half require it by 20 years ( Fig. 11.40 ).

Mitral valve replacement after open or closed commissurotomy. Format is as in Fig. 11.36A. (A) Time-related freedom. (B) Hazard function for mitral valve replacement after mitral commissurotomy.

(From Hickey MS, Blackstone EH, Kirklin JW, Dean LS. Outcome probabilities and life history after surgical mitral commissurotomy: implications for balloon commissurotomy. J Am Coll Cardiol . 1991;17:29.)

The same variables that cause survival to vary probably cause prevalence of postcommissurotomy mitral valve replacement to vary ( Table 11.3 ). Again, type of surgical commissurotomy (open or closed) has not been demonstrated to be a risk factor for subsequent mitral valve replacement. Morphology of the mitral valve powerfully affects the time-related prevalence of mitral valve replacement after surgical commissurotomy as well as after percutaneous balloon valvotomy. Management strategy for patients with mitral stenosis is illustrated in Fig. 11.41 .

Management strategy for patients with mitral stenosis. *Repair, commissurotomy, or valve replacement. AF , atrial fibrillation; CVC, Comprehensive Valve Center; MR, mitral regurgitation; MS, mitral stenosis; MV, mitral valve; MVA, mitral valve area; NYHA, New York Heart Association; PASP, pulmonary artery systolic pressure; PMBC, percutaneous mitral balloon commissurotomy.

(From Otto CM, Nishimura RA, Bonow RO, et al. 2020 ACC/AHA Guideline for the Management of Patients with Valvular Heart Disease: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation . 2021;143(5):e72-e227.)

Incremental risk factors for premature death after repair of mitral regurgitation are the same as those after replacement. In the UAB experience, neither procedure was a risk factor vis-à-vis the other. Akins and Dujardin and their colleagues, however, indicate that mitral valve replacement is a risk factor for late mortality compared with mitral repair in univariable analysis but not by multivariable analysis. ^, In most studies, this reflects choice of replacement for those patients who are older or in a more advanced NYHA functional class. However, Gillinov and colleagues found by both multivariable analysis and propensity adjustment that valve replacement was a risk factor for late death.

Most patients have little or no residual mitral regurgitation after repair. ^, Alvarez and colleagues reported that only 4.5% of 155 repair patients experienced repair failure within 6 months. In a group of more than 1000 patients analyzed for late failure, only 30 repair patients needed late reoperation (freedom from reoperation at 10 years 93%; CL 91%–94%). Recurrent regurgitation in these and other series is caused by either disease progression or inadequate operation, including suture dehiscence. In most patients, residual regurgitation is present immediately after repair rather than developing later. Although rate of reoperation may underestimate the prevalence of late regurgitation, various groups report freedom from reoperation of 80% to 96% at 10- to 15-year follow-up, suggesting good durability of repair. David and colleagues recently reported an incidence of reoperation on the MV of 4.6% after 20 years of follow-up.

A clear relationship has not been established between technique of repair and prevalence of residual regurgitation. However, as discussed earlier, repair is clearly more effective for patients with degenerative (myxomatous) disease than for those with restrictive (rheumatic) disease ( Fig. 11.45 ). ^, The best results may be in patients with ruptured chordae to the posterior leaflet (P ₂ ), ^, although Orszulak and colleagues have found almost equally good results when the ruptured chordae belong to the anterior leaflet. ^, Lawrie and colleagues reported a contemporary experience in which outcomes were equally good for artificial chord replacement in anterior leaflet prolapse as in posterior leaflet prolapse. For complex repairs and those involving the anterior leaflet, chordal replacement is superior to chordal shortening. ^, Historically, surgical repair of posterior leaflet prolapse has been more durable than repair of anterior leaflet prolapse, but results of the latter have improved in the current era. In patients with isolated posterior leaflet prolapse, chordal replacement plus anuloplasty ring appears to provide midterm valve competence equivalent to that of traditional quadrangular resection plus ring. Others have extended leaflet resection to the severely prolapsing anterior leaflet with good midterm valve function. ^, Many surgeons avoid leaflet resection and utilize artificial chordae to repair both anterior and posterior leaflet prolapse (i.e., “respect rather than resect.”)

Freedom from reoperation after mitral valve repair in patients with rheumatic and degenerative etiologies for mitral regurgitation. A number of patients at risk in each group is shown below curves. Vertical bars indicate one standard deviation.

(From Lessana A, Carbone C, Romano A, et al. Mitral valve repair: results and the decision-making process in reconstruction. Report of 275 cases. J Thorac Cardiovasc Surg . 1990;99:622.)

LV performance responds to mitral valve repair in the same general manner as it does to mitral valve replacement. Late postoperatively, some regression of LV hypertrophy occurs, with decreased heart size, LV and LA volumes, and muscle mass. Preoperative LV systolic function declines early postoperatively but improves over the first year postoperatively as the LV volume decreases. Suri and colleagues reported that patients with an ejection fraction (EF) of less than 50% at hospital dismissal following mitral valve repair were 3.5-fold less likely to recover normal ejection fraction during long-term follow-up ( P <.001). This same study found that a preoperative EF of greater than 65% (hazard ratio [HR], 1.8; P <.001) and a left ventricular end-systolic diameter (LVESD) of less than 36 mm (HR, 2.0; P <.001) resulted in the greatest likelihood of postoperative recovery of normal ejection fraction (EF ≥ 60%).

Some studies document LV performance to be better after mitral valve repair than after replacement. ^, Corin and colleagues found that systolic and diastolic function returned to normal in mitral repair patients, but global and regional systolic and diastolic functions were depressed in replacement patients. This difference is probably attributable to preservation of the tensor apparatus in the repaired valve ^, but may also reflect earlier intervention in patients undergoing repair versus replacement.

An important issue is the potential reversal of preexisting LV dysfunction following mitral valve repair. De novo postoperative LV dysfunction is not uncommon in patients with “normal” preoperative EF undergoing mitral valve repair. LV dysfunction can persist, impairing recovery of LV size, function, and survival. Among patients with preserved EF preoperatively, large right ventricle (>80 mL by quantitative Doppler and proximal isovelocity surface area method) is predictive of LV dysfunction (EF < 50%) following mitral valve repair. In terms of long-term changes, lowest mortality was observed in patients with pre-LV ejection fraction (LVEF) of 60% to 70%.

Patients undergoing mitral valve repair are generally free of late thromboembolic complications, even though they rarely receive anticoagulants late postoperatively. In patients without AF, the risk of thromboembolic events varies from 0.4% to 1.6% per year, but reaches 2.5% during the first postoperative month, even with routine anticoagulation therapy, and appears to be similar to that observed after mitral valve repalacement. The early use of anticoagulation (warfarin or NOACs) versus antiplatelet agents remains controversial. Van der Wall and colleagues reported that vitamin K antagonists and aspirin therapy showed a similar event rate during 3 months after mitral valve repair in patients without a prior history of AF. In both treatment groups, thromboembolic event rate was low, and major bleeding rates were comparable.

In about 5% to 10% of patients with mitral regurgitation associated with mitral valve prolapse, abnormal systolic anterior motion (SAM) of the mitral valve develops immediately after mitral anuloplasty with a Carpentier ring. ^, Related to this, gradients of 60 mmHg across the LV outflow tract have been measured. This complication appears to be limited to patients with myxomatous degeneration ^{, ,} and excessive redundancy of the anterior and posterior leaflets, although presence of a small hyperdynamic left ventricle and excessive ventricular hypertrophy may also contribute. Other independent predictors of SAM are a thick basal interventricular septum (>15 mm), a short distance between the leaflet coaptation point and the interventricular septum (<25 mm), a narrow aortomitral angle (<120 degrees), an anterior displacement of the papillary muscles, the presence of excessive leaflet tissue (like in Barlow’s disease, where the posterior leaflet is typically very high), and a ratio between the heights of the anterior and posterior leaflets ≤1.3.

SAM is now believed to result from anterior displacement of leaflet coaptation. If the possibility of SAM is anticipated, methods to reduce the height of the posterior leaflet are leaflet resection with triangular excision of a portion of the middle (P ₂ ) scallop, the use of shortened neochordae, or folding plasty, and should be considered at the time of primary repair. Simple removal of the rigid anuloplasty ring may abolish obstruction while retaining competence of the mitral valve. Substitution by a larger ring, a flexible ring, or a half-ring, or addition of a posterior leaflet sliding plasty may also eliminate SAM. An edge-to-edge suture may be used and appears very efficacious. When a bulging subaortic septum is thought to contribute to the SAM, transaortic myectomy can be curative.

But before any surgical steps to correct SAM are undertaken, it is advisable to stop inotropic infusion, initiate volume infusion, and increase afterload. Grossi and colleagues noted a prevalence of 6.4% (CL 5.2%–7.8%) in their series of valve repairs. All patients were treated medically, with resolution of gradients in all patients and resolution of SAM in half of patients within a year. Similarly good outcomes of conservative management were reported by Ashikhmina and colleagues, who found that among 98 patients with SAM identified intraoperatively, SAM resolved completely in 70 patients (71%), and at a median of 6.2 years, only 7% of patients with early SAM had residual SAM and none had LV outflow tract obstruction (LVOT). In an adjusted analysis, there were no significant differences in mitral regurgitation grade (N = 448; P =.822) or LVOT gradient (N = 337; P =.234) according to the presence of in-hospital SAM.

Functional status of most patients is excellent after repair or replacement of a regurgitant mitral valve, with greater than 90% of surviving patients in NYHA class I or II. ^{, ,}

Freedom from reoperation late after mitral valve repair is generally 85% to 95% at 10 years, ^{, , , , ,} similar to that after mitral valve replacement. Prevalence of reoperation is greater for anterior leaflet repair than for posterior leaflet repair.

Factors that may increase risk of reoperation late after repair include rheumatic disease (as opposed to degenerative disease; see Fig. 11.45 ), advanced degenerative changes involving the anterior leaflet, and residual regurgitation at completion of the initial procedure. When patients with 1+ or 2+ (of possible 4+) early postoperative regurgitation were compared with those with “echo-perfect” results (no regurgitation), Fix and colleagues found 83% freedom from reoperation in the former group versus 96% in the latter group at late follow-up ( P =.07). Type of repair appears to have no effect on prevalence of reoperation. Reed, Chauvaud, and Ohno and their colleagues showed mitral repair to be as successful in children as in adults ( Fig. 11.46 ). ^{, ,}

Freedom from reoperation after mitral valve repair in children younger than age 12 years. Vertical bars represent one standard error.

(From Chauvaud S, Fuzellier JF, Houel R, Berrebi A, Mihaileanu S, Carpentier A. Reconstructive surgery in congenital mitral valve insufficiency (Carpentier’s techniques): long-term results. J Thorac Cardiovasc Surg . 1998;115:84.)

The hazard function of reoperation after repair of mitral regurgitation is low, constant, and different from that of replacement ( Fig. 11.47 ). Mitral valve repair does not have the peaking early hazard phase for reoperation, which is related to periprosthetic leakage and prosthetic valve endocarditis following mitral replacement. Furthermore, no rising late hazard phase for reoperation after repair has been demonstrated, indicating the durability of this approach. In fact, in a 20-year experience reported by the Mayo Clinic, the absolute prevalence of reoperation in the current era was 5% ± 2% for posterior leaflet repair and 10% ± 2% for anterior leaflet repair at 10 years. A more recent analysis from the Mayo Clinic examined reoperations after mitral valve repair over 35 years. New pathology was the cause for reoperation in about 55% of patients, and failure of mitral repair in most of the remainder. Mitral valve re-repair was possible in 44% of patients.

Separately determined hazards for mitral valve reoperation in patients with mitral regurgitation who have undergone repair or replacement (UAB group, 1975 to July 1983). (See original article for equations, P -values, and coefficients.)

(From Sand ME, Naftel DC, Blackstone EH, Kirklin JW, Karp RB. A comparison of repair and replacement for mitral valve incompetence. J Thorac Cardiovasc Surg . 1987;94:208.)

Infective endocarditis is rare after repair of mitral regurgitation. No cases were found in the UAB experience, with a follow-up of 21 to 120 months. This complication is more common when the affected valve is replaced rather than repaired. Most patients that have endocarditis after mitral valve repair require reoperation, especially if in the presence of drug-resistant organisms. However, if infection is limited to a leaflet, early reoperation may be unnecessary because antibiotics alone can eradicate the infection.