Is Posterior Leaflet Extension and Associated Commissurotomy

Abstract

In this prospective study, the long-term effect of posterior leaflet extension with glutaraldehyde-preserved autologous pericardium and associated mitral valve commissurotomy was investigated in patients with mixed mitral valve disease of rheumatic origin.

Mitral commissurotomy and posterior leaflet extension using a pericardial patch were performed in 25 patients from 1 January 1994 through 31 December 1995 for mixed mitral valve disease. Preoperatively, no patient had chordal rupture or papillary muscle dysfunction. Four patients had left atrial thrombosis. The mean age was 35.7 ± 15.4 years. Associated procedures were tricuspid annuloplasty in 4 patients, aortic annuloplasty in 3, aortic and tricuspid annuloplasty in 1, and aortic homograft replacement in 1.

There were no early deaths. One patient died 2 years after surgery due to noncardiac causes. Mitral valve area increased from 1.53 ± 0.63 cm² to 2 ± 0.33 cm² (P = 0.09), and left atrial diameter decreased from 5.8 ± 1 cm to 4.86 ± 1.27 cm (P = 0.07) after 6.1 ± 0.7 years (range, 5.5 to 7.1 years). Mitral insufficiency was reduced significantly, from grade 2.65 ± 0.9 to grade 1.2 ± 0.9 (P = 0.007). Functional capacity improved in all patients (New York Heart Association functional class, 3 ± 0.58 preoperatively vs 1.44 ± 0.82 postoperatively; P = 0.001). Three patients required reoperation and valve replacement.

This type of reconstruction may be a good alternative for patients who are not able to use anticoagulant therapy. Long-term results of this technique are acceptable; however, the risk of reoperation is an important disadvantage in these young patients.

Reconstructive procedures are the 1st choice in surgical treatment of mitral valve disease. Valve reconstruction has been compared with valve replacement in different settings of mitral valve disease, and the superiority of reconstructive techniques has been demonstrated.^1–3 The most important advantages of mitral valve reconstruction are the avoidance of prosthetic valve complications¹ and the preservation of left ventricular function by sparing the subvalvular apparatus.⁴ Furthermore, the long-term results of reconstructive procedures are quite good.^5,6 Techniques in which an autologous pericardial patch is used for mitral valve reconstruction often yield acceptable results.^7,8 Posterior leaflet extension with a pericardial patch is one of these techniques, and it is easily applied in selected patients.^9–13

The aim of this prospective study was to assess the long-term results of mitral commissurotomy and posterior leaflet extension with use of a pericardial patch in patients with mixed mitral valve disease of rheumatic origin.

Patients and Methods

Mitral commissurotomy and posterior leaflet extension using a pericardial patch was performed in 25 patients who underwent surgery for mixed mitral valve disease during a 2-year period (1 January 1994 through 31 December 1995). Fifteen (60%) patients were female, and 10 (40%) were male. The mean age was 35.7 ± 15.4 years (range, 14 to 64 years). Associated cardiac diseases were noteworthy tricuspid regurgitation in 4 (16%) patients, moderate (grade 2) aortic insufficiency in 3 patients (12%), triple-valve disease in 1 (4%), and severe (grade 4) aortic regurgitation in 1 (4%). Preoperative echocardiographic evaluation revealed asymptomatic left atrial thrombosis in 4 patients (16%). Sixteen patients (64%) had preoperative atrial fibrillation. None of the patients experienced thromboembolic events before surgery. All patients were clinically symptomatic, and most of them were in New York Heart Association (NYHA) functional class III or IV. Although 4 patients were in NYHA class II, receiving full medical therapy, their mitral valve area was severely restricted (<1 cm²); hence the indication for surgery.

Mitral Valve Pathology

The cause of mitral valve disease in our series was rheumatic fever. The dominant condition in 21 patients was notable (<2 cm²) mitral valve stenosis with notable (grade 2 or greater) mitral valve regurgitation. Two patients had severe (<1 cm²) mitral valve stenosis with mild (grade 1) mitral valve regurgitation. These 2 patients had moderate thickening of the subvalvular apparatus. Only 2 patients had severe mitral valve insufficiency caused by annular dilatation and posterior leaflet retraction with mild mitral valve stenosis. No patient had chordal rupture or papillary muscle dysfunction.

Patient Selection

Patients who had pliable anterior leaflets with no motion restriction, in combination with coaptation defects of the mitral valve due to posterior leaflet retraction, were selected for reconstruction by means of posterior leaflet extension with a pericardial patch. The risks and advantages of both valve reconstruction and replacement were explained to all patients. Consent stating that they chose the reconstruction procedure was obtained from all patients.

Surgical Technique

After the initiation of standard cardiopulmonary bypass, myocardial protection was maintained by antegrade intermittent infusion of hypothermic crystalloid cardioplegic solution. Retrograde continuous infusion of blood cardioplegic solution was used in 5 patients (20%) who had undergone simultaneous aortic intervention. Venting from the right upper pulmonary vein was used for left heart decompression. After cross-clamping the aorta, we performed a left atriotomy. Left atrial thrombosis was detected in 4 patients, in whom thrombectomy was performed. The cause of mitral valve stenosis was fusion of both commissures, while the structure of the anterior leaflet was normal. Although thickened, the subvalvular apparatus was still in good condition in all patients. Bilateral commissurotomy was performed before posterior leaflet extension. In addition, we performed chordal splitting in 2 patients whose mitral stenosis was severe. When the coaptation of the mitral valve was checked, we observed that the main reason for regurgitation was retraction of the posterior leaflet. We extended the posterior leaflet using a pericardial patch. In the 2 patients with severe mitral valve regurgitation and mild stenosis, posterior leaflet retraction and annular dilatation were the causes of insufficiency. In these patients, in addition to posterior leaflet extension, we performed ring annuloplasty to remodel the dilated mitral valve annulus.

The pericardial patch was trimmed to a length of 10 cm and a width of 2 to 3 cm. Excessive tissue was cleaned from the patch, and it was treated with a 0.62% glutaraldehyde solution for 10 to 15 minutes. The patch was washed with sterile physiologic serum to reduce the possibility of débris embolization from the patch. The patch was then remodeled in accordance with the posterior leaflet and the coaptation defect of the patient.

The mitral valve posterior leaflet was cut from the anterolateral commissure to the posterolateral commissure, at a point 1 to 2 mm from the annulus and parallel to it (Fig. 1). The marginal chordae tendineae were preserved, while the secondary chordae were cut off if needed. The pericardial patch was then sutured to the annulus and to the posterior leaflet using 5/0 or 6/0 monofilament polyester material with a continuous technique. The ovoid patches were 3–5 cm × 1–2 cm, in correlation with the incision performed. The critical factor in this procedure was to avoid the systolic anterior motion of the posterior leaflet by preserving the 3:1 ratio of anterior-to-posterior leaflet width.⁹ At the end of the procedure, the valve was checked (Fig. 2) by injecting saline forcibly into the left ventricle. The aortic root was de-aired and distended by antegrade infusion of cardioplegic solution during testing. If the coaptation was acceptable, the left atrium was closed and the operation was completed in the usual manner.

Fig. 1 A) The scissors points to the incision in the posterior mitral leaflet. B) Extension with pericardial patch: sutures are placed in the superior margin of the defect in the posterior leaflet. The arrows show the pericardial patch.

Figure. 2. Intraoperative assessment of mitral valve competence. A) The coaptation of both leaflets is sufficient, not allowing regurgitant flow. B) Disruption of coaptation enables backflow.

Associated Cardiac Procedures

Left atrial thrombectomy and ligation of the left atrial appendix was performed in 4 patients (16%). The thrombi filled the left atrial appendices but did not extend into the left atrium. Tricuspid De Vega annuloplasty was performed in 1 patient and bicuspidization of the tricuspid valve in 3 patients. Aortic annuloplasty was performed in 3 patients. In 1 patient, both tricuspid Kay annuloplasty and aortic annuloplasty were added.

Aortic homograft replacement was performed in 1 patient who had severe aortic insufficiency.

Statistical Analyses

The patient data are presented as mean ± standard deviation. The reoperation-free life expectancy and freedom from death were analyzed with the Kaplan-Meier test, and the results are presented as mean ± standard error. The preoperative and postoperative late-term echocardiographic measurements and the functional capacity measurements were analyzed using the nonparametric Wilcoxon test. A commercial statistical software package (SPSS for Windows, version 10.0, SPSS Inc.; Chicago, Ill) was used for data analysis. A P value ≤0.05 was considered statistically significant for all comparisons.

The patients were followed up every 6 months with echocardiographic studies. Patients' data were collected from the hospital registry and from the patients themselves. Heparin administration for 24 hours and anticoagulation were started only in the 16 patients with preoperative atrial fibrillation; warfarin sodium was regulated by means of the international normalized ratio (INR). These 16 patients included those who underwent left atrial thrombectomy. Warfarin was also prescribed for the atrial fibrillation patient in whom sinus rhythm was restored after operation. For all 16 patients, warfarin was ordered as a lifelong regimen.

Results

All surviving patients were monitored for a mean follow-up period of 5.4 ± 2.2 years.

Early and Late Mortality

Early death did not occur, but 1 patient (4%) died during the 2nd postoperative year of noncardiac causes. Overall survival was 95.65% ± 4.25% at 8 years.

Postoperative Rhythm

Atrial fibrillation persisted after surgery in 15 of the 16 patients who had experienced atrial fibrillation preoperatively. Sinus rhythm was restored in only 1 of these. Postoperative rhythm disturbances were not detected in patients who were in sinus rhythm preoperatively.

Early and Late Complications

No nonsurgical complication (for example, thromboembolism or infective endocarditis) was identified. Complications associated with the use of warfarin were not detected in the 16 patients who had exhibited preoperative atrial fibrillation.

Reoperation

Severe mitral insufficiency after mitral valve reconstruction was detected in 3 patients (12%). The 1st patient underwent reoperation due to severe mitral insufficiency during the 1st postoperative week, and the 2nd and 3rd patients required reoperation during the 3rd and 6th postoperative months. The valves of all 3 patients were replaced. Reoperation-free life expectancy was 95.65% ± 4.25%, 91.1% ± 6%, and 85% ± 8.1% through years 1, 2, and 7.5, respectively.

Preoperative and Postoperative Echocardiographic Results

Echocardiographic studies performed on 21 patients who did not require reoperation at 6.1 ± 0.7 years (range, 5.5–7.1 years) after surgery showed only slight increases in valve area but significant reductions of insufficiency (Fig. 3). The patient who died during the 2nd postoperative year was also excluded from the echocardiographic study. The decrease in valve gradient was not statistically significant. The preoperative and late postoperative echocardiographic results are presented in Table I.

Figure. 3 Postoperative echocardiographic view shows the coaptation of the leaflets and the 3:1 ratio of anterior-to-posterior leaflet width.

AML = anterior mitral leaflet; LA = left atrium; LV = left ventricle; PML = posterior mitral leaflet

TABLE I. The Results of Preoperative and Late Postoperative Echocardiographic Studies in 21 Patients

Preoperative and Postoperative Functional Capacity

The New York Heart Association (NYHA) functional capacity of the patients improved significantly at 5.4 ± 2.2 years after surgery due to a decrease in mitral insufficiency and an increase in effective mitral valve area (P = 0.001) (Fig. 4). Functional capacity increased in 22 patients, and it decreased in 3 patients who subsequently underwent reoperation.

Figure. 4. The postoperative New York Heart Association (NYHA) functional capacity of the 25 patients improved significantly at 5.4 ± 2.2 years after surgery.

P = 0.001

Discussion

In Turkey, the most common cause of mixed mitral valve disease is rheumatic fever. This disease usually affects young patients. Mitral valve involvement is progressive and can lead to severe malformation of the leaflets. Reconstruction of the damaged valves is usually not possible; therefore, replacement of the valves is the main surgical option. Yet both mechanical prosthetic valves and bioprosthetic valves may not be suitable for some patients. The main problems with mechanical valves are life-long anticoagulant therapy, increased hemolysis, paravalvular leakage,¹⁴ a high transvalvular gradient, and a relatively small effective valvular area. These drawbacks are especially problematic for patients who are unable to use anticoagulant therapy, such as pregnant women. Bioprosthetic valves, on the other hand, are not recommended for young patients because they tend to develop degenerative changes that lead to high rates of dysfunction and reoperation.¹⁵ Hence, it is desirable to protect these young patients from the adverse effects of prosthetic valves by using reconstructive techniques in selected cases. In sum, mitral valve reconstruction is most appropriate in the elderly, in young women who wish to become pregnant, and in any patients who cannot tolerate anticoagulant therapy.

Due to its biologic structure, pericardium is often used in cardiac surgery.^16,17 Its thickness is 0.5 to 1 mm, and it is composed of a serous and a fibrous layer.¹⁸ The usable area of pericardium between the 2 phrenic nerves is approximately 90 cm². This amount of pericardium is enough for all kinds of cardiac procedures, but first it must be treated with glutaraldehyde for at least 10 minutes to provide better resistance.¹⁹ Using fresh pericardium without fixation has yielded disappointing short-term results. Therefore, the surgeon must be sure that the pericardium has adequate resistance before using it.^9,17 Experimental studies have shown that pericardial patches can be successfully used in reconstruction of mitral valves,²⁰ but for obvious reasons it is important to avoid using pericardial tissue that has been affected by rheumatic fever. Its gross appearance at surgery may indicate whether it has been involved. The thickness, the resistance, and the structure of the pericardium are all important criteria that affect the success of the reconstruction. There is need for objective tests that can be used intraoperatively to determine the durability of the pericardium.²¹

The reoperation rate (3 patients) in our series was similar to that reported in the literature (8.3%–16.2%).^9,21–23 Our reoperation rate was 2% per year (12% in 5.4 years), which is similar to that reported by Chauvaud and associates.²⁴ This rate is higher than the reoperation rate among patients in whom mitral valve replacement using a mechanical prosthetic device was performed.²⁵ The reason for reoperation in our 3 patients was severe mitral insufficiency. The 1st patient, who also was the 1st patient in our series, had minimal mitral insufficiency at the end of the initial operation, which worsened rapidly during the 1st postoperative week. During reoperation, we found the pericardial patch to be well positioned; however, some marginal chordae tendineae of the posterior leaflet were ruptured, and the posterior leaflet had prolapsed into the left atrium. In subsequent operations, we discontinued cutting the chordae. The other 2 patients, who experienced late-term mitral insufficiency, underwent replacement of the valve. The cause of the insufficiency was progression of the disease, which affected both the valve and the annulus. The pericardial patch seemed not to be involved, and no calcification, thickening, or narrowing was detected on the patch.

The most interesting finding in our patients upon echocardiography was the absence of thickening and calcification on the posterior leaflet. Although the valve area increased 1:3 in ratio, this was not statistically significant. The significant decrease in mitral valve insufficiency corresponded with clinical improvement. Approximately 7.5 years after surgery, the gradients of the mitral valve returned to preoperative values. Nevertheless, the decreases in both mitral insufficiency and left heart dimension show us the success of this procedure. The low incidence of early and long-term complications is also an advantage over valve replacement.

Although we performed this procedure in a relatively small number of patients, we conclude that posteri- or leaflet extension with a pericardial patch, together with commissurotomy, reduces mitral insufficiency in selected patients with mixed mitral valve disease, although it does not significantly increase the effective mitral valve area. The relatively high risk of reoperation restricts the indications for this technique. However, the use of this technique in selected patients who manifest insufficiency rather than stenosis as the dominant mitral valve condition will improve quality of life by avoiding the complications associated with prosthetic valves.