Should a Regurgitant Mitral Valve Be Replaced Simultaneously with a Stenotic Aortic Valve?

Abstract

Mitral valve regurgitation frequently accompanies aortic valve stenosis. It has been suggested that mitral regurgitation improves after aortic valve replacement alone and that the mitral valve need not be replaced simultaneously. Furthermore, mitral regurgitation associated with coronary artery disease, particularly in patients with poor left ventricular function, shows immediate improvement after coronary artery bypass grafting.

We studied 60 consecutive patients with aortic stenosis and mitral regurgitation to determine the degree of improvement in mitral regurgitation after aortic valve replacement alone versus aortic valve replacement combined with coronary artery bypass grafting. Thirty-six of the patients had normal coronary arteries (Group 1); the other 24 had symptomatic coronary artery disease requiring bypass surgery (Group 2). Echocardiography was performed preoperatively, 1 week postoperatively, and at follow-up.

In Group 1, left ventricular ejection fraction did not improve early or at 2.5 months postoperatively, but mitral regurgitation improved gradually during follow-up. In Group 2, mitral regurgitation showed improvement 1 week postoperatively (p <0.001), and left ventricular ejection fraction was improved at 2.5 months.

We conclude that patients with aortic valve stenosis and mild-to-severe mitral regurgitation, without echocardiographic signs of chordal or papillary muscle rupture and without coronary artery disease, should undergo aortic valve replacement alone. The mitral regurgitation will remain the same or improve. For patients with coexisting coronary artery disease, simultaneous aortic valve replacement and coronary artery bypass grafting are imperative; however, the mitral valve again requires no intervention, since mitral regurgitation improves significantly after the other 2 procedures.

Aortic valve stenosis progresses predictably over time, but systolic dysfunction appears to be an inconsistent marker of the hemodynamic consequences of severe aortic stenosis. ¹ Progressive hypertrophy, as an adaptive response to pressure overload, seems to prevent ventricular dilatation and mitral regurgitation development. On the other hand, progressive mitral regurgitation (MR) may act as a marker of impaired left ventricular performance in aortic stenosis and may also be a maladaptation to increasing aortic stenosis. ² Mitral regurgitation is a common finding in patients with aortic valve stenosis; an incidence as high as 67% has been reported in the literature. ³ Surgery is often required for aortic stenosis. However, double valve intervention (replacement, repair, or the 2 in combination) is associated with an elevation in hospital mortality rates (5% to 12.5%) and postoperative morbidity rates. ^4,5 Moreover, several studies have suggested that the severity of concomitant MR improves after aortic valve replacement. ^6–8

Aortic valve stenosis is also frequently accompanied by coronary artery disease. ⁹ It has been shown that in patients who have poor left ventricular function (left ventricular ejection fraction [LVEF] £0.25), symptomatic coronary artery disease, and ischemic MR, coronary artery bypass grafting (CABG) alone not only improves left ventricular function but also normalizes mitral valve function. ¹⁰ This immediate effect continues over the long term. ¹¹

We evaluated retrospectively the mitral valve function in 2 groups of patients: 1) those presenting with aortic valve stenosis combined with MR, who underwent aortic valve replacement alone; and 2) those with aortic stenosis, Mr, and symptomatic coronary artery disease, who underwent both CABG and aortic valve replacement. Neither group of patients underwent mitral valve replacement.

Patients and Methods

From January 1990 through January 1999, we treated 60 consecutive patients who had aortic valve stenosis and concomitant MR, with or without coronary artery disease. Preoperative coronary angiography and echocardiography were performed in all patients just before surgery. Thirty-six patients had aortic stenosis and moderate-to-severe MR (grades I through III) and underwent aortic valve replacement alone. Of those, 32 patients (89%) had coronary arteries without atherosclerotic changes, and 4 (11%) had 25% to 30% stenosis in the coronary arteries, which did not require myocardial revascularization. These 36 patients (Group 1) ranged in age from 36 to 81 years (64.0 ± 13.3 mean ± SD years), and 21 (58%) were men. The mean preoperative New York Heart Association (NYHA) functional class in this group was 3.0 ± 0.8.

The other 24 patients presented with aortic valve stenosis, MR (grades I through III), and symptomatic coronary artery disease. These patients underwent aortic valve replacement combined with CABG. These 24 patients (Group 2) ranged in age from 55 to 91 years (70.9 ± 9.2 years), and 14 (58%) were men. The mean NYHA functional class in this group was 3.3 ± 0.7.

In no patient was the surgery a reoperation. No patient showed echocardiographic signs of chordal or papillary muscle rupture (valve prolapse or flail valve) or evidence of morphologic changes in the mitral valve. We defined morphologic changes in the mitral valve as echocardiographic evidence of calcified nodules or vegetations anywhere on the valve; leaflet cusp lesions; and the following signs of leaflet damage: perforations, shortening, thickening, rigidity, retraction, and commissural fusion. No patient underwent surgical correction for MR. Preoperative patient characteristics are presented in Table I, and preoperative functional data are shown in Table II.

TABLE I. Preoperative Patient Characteristics in Group 1 (n = 36) and Group 2 (n = 24)

TABLE II. Preoperative Functional Data in the 2 Study Groups

In Group 2, 14 patients (58%) had triple-vessel coronary artery disease, 5 (21%) had double-vessel disease, and 5 (21%) had single-vessel disease.

Echocardiographic Analysis

A complete echocardiographic evaluation was performed just before the surgical procedure, 1 week after surgery, and during the follow-up period (at least 2 months after surgery). Echocardiographic equipment used during the study was either a Toshiba SSH-65A or a Hewlett-Packard 2500. The standard echocardiographic examination included M-mode, 2-dimensional (2-D), pulse, color-flow Doppler, and continuous Doppler echocardiography. Standard views were used, including parasternal long- and short-axis; and 2-, 3- and 4-chamber apical views. Pre- and postoperative studies were distributed randomly to echocardiographers for evaluation. The severity of MR flow was evaluated by visual assessment of a color Doppler echocardiogram—a simple, practical, and routine method of quantification—using a standard classification system ranging from mild (grade I) to severe (grade III) MR. The mean transvalvular aortic pressure gradient was measured pre- and postoperatively in at least 2 apical views. The left atrial diameter was measured in the long-axis parasternal view using M-mode echocardiography. The short-axis parasternal view allowed measurement of the thickness of the interventricular septum and the posterior wall. The left ventricular end-systolic and end-diastolic diameters were measured using the standard parasternal short-axis view. All echocardiographic films were reevaluated by a single cardiologist.

Statistics

The χ²-test (Fisher exact test) for nominal measurements, the median and Mann-Whitney tests for ordinal measurements, and the Student's t-test (independent) for metric measurements were used to evaluate statistical differences between groups where appropriate. A probability of p <0.05 was considered to be statistically significant.

Results

Group 1. Twenty-one patients (58%) in Group 1 received mechanical aortic valves, and 15 (42%) received biological aortic valves. The mean duration of cardiopulmonary bypass was 79.8 ± 13.5 minutes, and the mean ischemic time was 55.6 ± 11.6 minutes. One patient in this group (2.8%) died in the hospital. The mean stay in the intensive care unit was 2.4 ± 1.5 days (range, 2 to 8 days). The mean length of hospital stay was 12.7 ± 2.4 days (range, 10 to 19 days).

Group 2. Mechanical valve prostheses were implanted in 17 patients (71%) in Group 2, and biological prostheses were implanted in 7 patients (29%). A mean of 2.8 ± 1.6 distal coronary anastomoses (CABG; range, 1 to 5 anastomoses) were performed in this group. The mean duration of cardiopulmonary bypass was 122.1 ± 21.2 minutes, and the mean aortic cross-clamping time was 81.3 ± 10.0 minutes. One patient in this group (4.2%) died in the hospital. The mean stay in the intensive care unit was 4.4 ± 2.9 days (range, 2 to 12 days). The mean length of hospital stay was 13.3 ± 4.1 days (range, 7 to 20 days).

Group Comparisons

The preoperative LVEF values and cardiac indices were significantly lower (p <0.0001 and p = 0.0131, respectively) in Group 2 than in Group 1. Cardiopulmonary bypass and ischemic times (aortic cross-clamping) were significantly longer in Group 2 (p <0.0001). However, neither hospital mortality rates nor duration of hospitalization differed between the groups.

Evolution of the Mitral Valve Regurgitation

Group 1. Complete data on 35 of 36 patients in Group 1 were available (from the preoperative period through follow-up). Fifteen of these patients (43%) had grade II or III MR preoperatively. One week postoperatively, 15 patients remained in grade II or III. However, at the end of follow-up (mean, 7.2 ± 2.7 months; range, 2 to 29 months), this number had decreased to 12 patients (34%), all of whom had grade II MR. In no patient did the MR worsen during the study period, nor has any patient as yet required intervention for MR (Table III). In Group 1, the mean LVEF remained the same, from before surgery through the end of follow-up: 0.54 ± 0.10 (range, 0.42 to 0.74) and 0.55 ± 0.08 (range, 0.42 to 0.70), respectively.

TABLE III. Degree of Mitral Regurgitation in Survivors from the 2 Study Groups

Group 2. Complete data on 23 of 24 patients in Group 2 were available from the preoperative period through follow-up. In contrast to the patients in Group 1, Group 2 patients had significantly improved LVEFs from before surgery through the end of follow-up: 0.36 ± 0.12 (range, 0.25 to 0.62) and 0.51 ± 0.08 (range, 0.42 to 0.64), respectively (p = 0.0051). Preoperatively, 20 (90%) of the patients in Group 2 had MR grades of II or III. Just 1 week postoperatively, there was a significant improvement in MR, with only 20% (4) patients remaining in grade II and none in grade III (p <0.0001). During follow-up (2.7 ± 0.6 months; range 2 to 6 months), there was no further significant improvement or deterioration in MR, and no patient required surgical correction for remaining MR (Table III).

Other echocardiographic data are presented in Table IV. In both groups, the mean aortic transvalvular gradient decreased significantly after surgery. In Group 1, the left atrial diameter and the left ventricular end-systolic and end-diastolic diameters remained unchanged at the 1-week postoperative examination, in comparison with the preoperative values. By the end of follow-up, a slight reduction in the thickness of the left ventricular wall could be seen, particularly in the interventricular septum. In Group 2, the left atrial diameter and the left ventricular end-systolic and end-diastolic diameters improved significantly within 1 week after surgery. No further improvements were observed during the follow-up period in Group 2 patients (Table IV).

TABLE IV. Echocardiographic Values* Before and After Aortic Valve Replacement in Survivors from the 2 Study Groups

Discussion

A combination of valvular lesions makes it difficult to evaluate the true hemodynamic and symptomatic contributions of the individual lesions. ¹² Patients with aortic valve stenosis frequently have associated coronary artery disease, mitral regurgitation, and left ventricular dysfunction. Whereas isolated aortic valve replacement in elderly patients carries an acceptable mortality rate (approximately 1% to 2%), the operative risk is significantly increased when double valve surgery is performed, with or without myocardial revascularization. ^5,9 Mitral regurgitation may either be caused by morphologic changes in the mitral valve ¹³ or be a secondary effect of aortic stenosis or left ventricular dysfunction. Since these mechanisms are quite different, it is important to distinguish between them. In our study, by design, none of the patients in Group 1 had evidence of intrinsic mitral valve disease, and all were considered to have functional MR. Previous studies have shown that functional MR, when associated with aortic stenosis, improves after aortic valve replacement alone. ^7,8 Similar improvement of MR occurs when aortic stenosis is associated with coronary artery disease, and aortic valve replacement and CABG are performed together. ¹⁰ Many complex techniques have been introduced to measure the severity of mitral regurgitation, such as 3-D color Doppler echocardiography, ¹⁴ the combined use of a CO₂-rebreathing method and echocardiography, ¹⁵ and echocardiographic evaluation of numerous objective and morphologic elements. ⁸ One aim of our study was to determine whether the severity of the MR could be reliably evaluated on the basis of visual evaluation alone, there by simplifying preoperative evaluation in patients with combined aortic stenosis and MR, with or without coronary artery disease. When evaluating the severity of the MR, simple visual evaluation of the grade seems sufficient for the decision-making process. The severity of aortic stenosis, as reported by Come and colleagues, ¹ can be predicted accurately by noninvasive calculation of the aortic valve area.

We found that aortic valve replacement alone for aortic stenosis, in patients without coronary artery disease, did not improve left ventricular function 1 week postoperatively or during the follow-up period. Others have reported similar findings or even an early postoperative decline in LVEF. ^3,16,17 Our long-term findings are somewhat in contrast to those of Robiolio's group, ¹⁶ which showed a significant improvement in such patients 6 months after surgery. However, it should be noted that none of the patients in our series (Group 1) had severely depressed left ventricular function. On the other hand, gradual improvement of the MR was observed in these same patients, particularly among those who had a preoperative grade II or III MR. These results correspond well with those described by Harris and coworkers ⁸ but are in contrast to those described by Adams and Otto. ¹⁸

Our echocardiographic data correspond well with those described earlier. ⁸ Our findings have led us to conclude that aortic valve replacement alone is the correct treatment in patients with aortic stenosis, normal coronary arteries, and no evidence of morphologic changes in the mitral valve, regardless of the severity of the MR. When aortic valve stenosis and MR are associated with symptomatic coronary artery disease and poor left ventricular function (Group 2), the situation is somewhat different. In these patients, CABG in combination with aortic valve replacement is imperative. The mitral valve can be left without surgical correction, provided that there are no echocardiographic signs of morphologic mitral valve changes or of chordal or papillary muscle rupture. Our Group 2 patients experienced a significant and immediate improvement of the MR after surgery, together with significantly improved LVEFs at a mean of 2.5 months after surgery. These observations confirm previous findings. ^10,19 Aortic valve replacement for aortic stenosis combined with CABG did not significantly increase the surgical risk. Hospital mortality, length of stay in the intensive care unit, and the length of hospital stay did not differ significantly between the groups, despite longer cardiopulmonary bypass and aortic cross-clamping (ischemic) times in the group undergoing combined surgical procedures.

During the same period, another 15 patients with aortic stenosis and MR underwent aortic valve replacement combined with replacement of the mitral valve at our institution. Twelve of these patients (80%) had morphologic changes: 10 had combined mitral valve stenosis and regurgitation, and 2 had echocardiographic evidence of chordal rupture. Only 3 patients (20%) underwent mitral valve surgery solely on the basis of severity of MR (grade III). Clearly, the surgeons at our institution tend to operate on the mitral valve only when there is evidence of morphologic change, regardless of the severity of the MR.

In conclusion, patients with aortic valve stenosis and MR (without echocardiographic evidence of morphologic changes in the mitral valve or of chordal or papillary muscle rupture) and without symptomatic coronary artery disease (Group 1) require only aortic valve replacement. With this treatment, MR remains the same or improves slightly over time and does not require surgical correction; this seems to be a safer alternative to double valve replacement. If the aortic stenosis and MR are associated with symptomatic coronary artery disease (as in Group 2), myocardial revascularization is imperative, along with aortic valve replacement. In this group, as well, the mitral valve can be left without surgical intervention, provided that the MR is of an ischemic or functional type without papillary muscle rupture. Aortic valve replacement combined with CABG significantly improves the MR and the left ventricular function post-operatively. To date, no patient in our series has required mitral valve surgery.

Therefore, we recommend that MR be corrected only when there is preoperative echocardiographic evidence of morphologic change in the mitral valve or of chordal or papillary muscle rupture. Otherwise, patients can undergo aortic valve replacement, with or without combined myocardial revascularization, and experience the same or improved MR without mitral valve replacement.