Left ventricular remodeling impacts the function of the Quattro stentless mitral valve bioprosthesis (a 4-year experience)
Article Outline
Objective
The St Jude Quattro stentless mitral valve prosthesis (QMV) is sutured to the mitral annulus and the papillary muscle heads, thereby preserving the subvalvular apparatus. After mitral valve replacement, remodeling of the left ventricle is often observed, causing a dilated ventricle to shrink in diameter. It was our objective to assess these changes in left ventricular (LV) geometry and evaluate its effects on the function of the QMV.
Methods
From September 1997 to October 2000, 24 patients received QMV at our institution. The patients were followed up at yearly intervals (mean 4.1 ± 2.2 years). All pre- and postoperative echocardiograms were evaluated, with attention focused on the subvalvular apparatus, leaflet morphology, and occurrence of late mitral regurgitation. In addition, all clinical outcomes and valve-related complications were recorded.
Results
Forty-one percent of patients (10/24) developed late mitral regurgitation (mild, n = 5; moderate, n = 5). The site of regurgitation was located at the 2 commissures in all cases. In 8 patients, changes in LV diameter had occurred. The point of leaflet coaptation had shifted away from the annulus in 4 patients. The overall mortality was 12.3%, and the postoperative stroke rate was 12.3%.
Conclusions
Midterm changes in LV geometry seem to affect the competence of the QMV. Predicting these changes and subsequently adapting the sizing procedure remain a challenging task. The high rate of late valve incompetence and poor clinical outcomes has prompted us to discontinue recruitment of patients for this trial.
Dilatation of the left ventricle after mitral valve replacement is a known risk factor for reduced postoperative ejection fraction and increased mortality. Preservation of the subvalvular apparatus has been shown to improve postoperative left ventricular (LV) function and to reduce LV diameters after mitral valve replacement.1 Depending on the etiology of mitral valve disease, native valve dysfunction, and the surgical procedure used to treat the pathology, different extents of LV remodeling occur postoperatively.2, 3
Similarly to changes in LV mass after aortic valve replacement, not only LV internal dimensions but also LV wall thicknesses are subject to remodeling after mitral valve replacement.2, 4
In patients with mitral valve stenosis, especially in the presence of shrinkage of the subvalvular apparatus, an increase in LV diameters occurs after mitral valve replacement. In patients with mitral valve incompetence, a reduction in LV size occurs postoperatively.
The St Jude Quattro stentless mitral valve bioprosthesis (St Jude Medical Inc, Minneapolis, MN) has the unique feature of papillary flaps that are attached to the papillary muscle heads (Figure 1). Thus, it provides a means of stabilizing and maintaining LV geometry and size, after the native valve and subvalvular apparatus have been excised. The papillary flaps form a continuity with the leaflets of the valve. The function of the stentless Quattro mitral valve is therefore prone to be influenced by changes in LV size and geometry.
Figure 1.
The St. Jude Quattro stentless mitral valve prosthesis.
The objective of our study was the assessment of changes in LV geometry after mitral valve replacement with the stentless Quattro bioprosthesis. Furthermore, we evaluated the effects of these changes on the function of the valve, with respect to the incidence of late mitral valve incompetence.
Methods
From September 1997 to October 2000, 24 patients received the investigational Quattro valve, at our institution. The mean age was 62 ± 16.3 years; there were 3 males and 21 females. The preoperative demographic data are summarized in Table I. All patients provided informed consent, and the study was approved by our institutional ethical committee.
Table I. Demographic data of 24 patients who received a quadrileaflet mitral valve prosthesis at our institution
| Age (y) | 62.2 ± 16.3 |
| Sex (M/F) | 3:21 |
| BSA (m2) | 1.73 ± 0.17 |
| Atrial fibrillation | 29% |
| Rheumatic | 41% |
| Degenerative | 59% |
| Coronary artery disease | 29% |
| NYHA | 3.06 ± 0.2 |
| EF | 64.1% ± 14.7% |
| Redo surgery | 24% |
The mean patient follow-up was 4.1 ± 2.2 years. Clinical examinations and transthoracic echocardiography were performed preoperatively, at discharge, after 3 to 6 months, and at yearly intervals thereafter. All pre- and postoperative echocardiograms were recorded on SVHS videotapes and evaluated. Special attention was focused on the dimensions of the heart chambers, the subvalvular apparatus, the plain of leaflet coaptation, the leaflet morphology, and the incidence and extent of late mitral valve regurgitation.
When assessing the heart chambers, LV end-systolic and end-diastolic diameters, as well as left atrial size, were measured in the transthoracic M-mode. In addition, changes in LV shape were looked for in 2-dimensional echocardiography. The preoperative and follow-up values were recorded and compared to each other, with respect to the native mitral valve dysfunction. Assessment of the subvalvular apparatus was done in 2-dimensional echocardiography. We looked for changes in the position of the papillary muscles, as well as the distance between the papillary muscle heads to the plane of the annulus, in the parasternal long axis and the apical 4-chamber view. We used the same views to determine the plane of leaflet coaptation, preoperatively and at follow-up. Special attention was paid to any dysfunction, such as prolapsing or retracted leaflets.
Mitral valve regurgitation was quantified using color Doppler. Special attention was focused on the incidence of late postoperative regurgitation. The maximum regurgitant jet area as well as the vena contracta was derived from the apical 4-chamber view and the parasternal long axis. The degree of mitral regurgitation was graded as trivial, mild, moderate, or severe, when the maximal regurgitant jet area, corrected for atrial size, was <10%, 10% to 20%, 21% to 40%, or 40%, respectively.
All outpatient department data regarding New York Heart Association (NYHA) functional class, cardiac rhythm, occurrence of clinical events, anticoagulation status, and hemolysis were also reviewed for all patients.
Surgical technique
Our surgical technique has been described previously,5 in short: access to the mitral valve was gained via a median sternotomy (n = 19) or a right-sided, anterolateral mini-thoracotomy in the fourth intercostal space (n = 5). Cardiopulmonary bypass was established. The mitral valve was approached via a left atriotomy through the intra-atrial groove. The valve was then excised leaving only one set of chordae to each papillary muscle, to allow for appropriate measurement of the distance from the native mitral annulus to the papillary muscle heads. The annulus was then sized, and the appropriate valve selected. Our sizing technique has been described previously.5 The papillary flaps were then attached to the 2 largest heads of the papillary muscles, by 2 pledgeted Cardionyl sutures (Cardionyl Peters Laboratories, Bobigny-Cedex, France) that are passed from the posterior to the anterior of each papillary muscle. Thereafter, the annulus of the Quattro valve was sutured to the native annulus with a continuous 3-0 polypropylene suture. Atriotomy and thorax were then closed in the usual way. The perioperative data are summarized in Table II.
Table II. Perioperative data of 24 patients who received a quadrileaflet mitral valve prosthesis
| Implants | n = 24 |
| Bypass time (min) | 138.3 ± 37 |
| X-clamp time (min) | 91.3 ± 26.3 |
| Minimally invasive surgery | 24% |
| Concomitant procedures | 53% |
| Mean ventilation time (h) | 12.2 ± 6.3 |
| Hospital stay (d) | 9.2 ± 4 |
Results
The dimensions of the heart chambers, the subvalvular apparatus, and the leaflets of the valve were easily visualized using transthoracic echocardiography. The pre- and postoperative hemodynamic measurements are summarized in Table III. Overall, there were 3 deaths. One death occurred intraoperatively as the papillary flap tore off the papillary muscle at the end of the operation causing intractable ventricular fibrillation. Another 2 patients died within the first 30 days because of cardiac reasons (ventricular fibrillation and congestive heart failure). Another 3 patients had a stroke, all within the first 6 months after the operation. They continued oral anticoagulation, and all neurological symptoms resolved fully. The clinical outcomes of patients and the valve-related complications are summarized in Table IV.
Table III. Echocardiographic data preoperatively, at discharge and at 4-year follow-up assessed by transthoracic echocardiography
| Preoperative | At discharge | 4-y follow-up | |
|---|---|---|---|
| Mitral valve area (cm2) | 1.3 ± 0.7 | 2.5 ± 0.4 | 2.2 ± 0.4 |
| Pmax (mm Hg) | 21.3 ± 14.3 | 10.4 ± 5.3 | 10.8 ± 3.3 |
| Pmean (mm Hg) | 13.6 ± 10.8 | 3.6 ± 2.0 | 3.8 ± 1.4 |
| Mitral regurgitation | n = 18 | n = 1 | n = 10 |
| Left atrium (mm) | 55.2 ± 7.3 | 46 ± 8.4 | 44 ± 11.3 |
| LVEDD (mm) | 49.1 ± 8.4 | 50.2 ± 9.2 | 48.7 ± 5.9 |
| LVESD (mm) | 36. 2 ± 8.1 | 38.1 ± 9.4 | 36.7 ± 8.4 |
| EF (%) | 64.1% ± 14.7% | 63% ± 10.3% | 62.4% ± 6.8% |
Table IV. Complications and mortality within a 4-year follow-up
| Cardiac-related mortality | N = 2 (early) | 8.2% |
| Valve-related mortality | N = 1 (early) | 4.1% |
| Total mortality | N = 3 | 12.3% |
| Stroke | N = 3 | 12.3% |
| Endocarditis | N = 1 | 4.1% |
| Major bleeding | N = 2 | 8.2% |
| Reoperation | N = 1 | 4.1% |
| Paravalvular leak | N = 0 | – |
| Structural deterioration | N = 0 | – |
| Thrombosis | N = 0 | – |
| Hemolysis | N = 0 | – |
Left ventricular size
Overall, LV size appeared not to change from preoperative to follow-up measurements. However, when separately assessing the LV size of patients grouped according to their native valve dysfunction, we found postoperative changes in LV dimensions in all groups. For patients with native mitral valve stenosis, mean LV end-diastolic diameters increased from 48.7 to 54.7 mm (P = .175). For patients with native mitral valve incompetence and mixed mitral valve disease, mean LV end-diastolic diameters decreased from 51.6 to 47 mm (P = .99) and 46.6 to 40 mm (P = .99), respectively.
Subvalvular apparatus and leaflets
Although there was remodeling of the left ventricle, as described above, the position of the papillary muscles within the ventricle did not appear to have changed.
The morphology of the mitral valve leaflets was intact in all patients. There was no dysfunction of the leaflets at discharge. We did, however, see dysfunction of leaflets in 4 patients at follow-up. In 3 patients, the plane of coaptation was prolapsing into the left atrium, and in 1 patient the plane of coaptation was retracted into the left ventricle.
Mitral valve regurgitation
Postoperatively, 1 patient had mild mitral valve regurgitation. After 1 year, 2 patients had mild regurgitation. At the 4-year follow-up, 10 patients had mild to moderate regurgitation (mild, n = 5; moderate, n = 5). The site of regurgitation was located at the commissures of the valve in all cases. We did not observe central incompetence in any of the patients in this study. Eight of the patients with mild to moderate regurgitation were symptomatic at follow-up. Five patients were in NYHA functional class II and 3 of the patients in NYHA class III.
Discussion
Remodeling of the left ventricle occurs after prosthetic mitral valve replacement.6 This is also true for the left ventricle after mitral valve replacement with the stentless Quattro bioprosthesis. Middlemost et al, one of the teams investigating this valve, report a reduction of LV diameters 1 year postoperatively.7 Walther et al also describe the same findings for their group of patients at 1 year after mitral valve replacement with the Quattro bioprosthesis.8 In our experience, LV remodeling took place in all of the patients. However, depending on the etiology of native valve dysfunction, different extents of remodeling occurred. In patients with mitral stenoses, the left ventricle increased in size postoperatively, and in patients with isolated mitral regurgitation or mixed disease, we noted a reduction of LV diameters. Considering that there is a continuity between the papillary muscle flaps that are attached to the papillary muscle heads and the leaflets of the valve, it only seems logical that changes in LV size will directly affect the coaptation area of the leaflets. So, if the valve is accurately sized at the time of implantation and postoperative remodeling causes an increase in LV diameters, the papillary flaps will pull on the leaflets, and the plane of coaptation will be retracted below the annular plane. On the other hand, billowing of the leaflets and displacement of the plane of coaptation above the annular plane will occur if the LV size decreases postoperatively. Depending on the extent of these changes, the function of the valve is impaired, and mitral valve regurgitation might result. In our group of patients, we identified 4 patients who showed displacement of leaflets. However, in none of these cases, central mitral valve regurgitation occurred as a result of changes in the plane of coaptation. Instead, we found regurgitation jets originating from the commissures of the valve.
Mild to moderate mitral valve regurgitation has been described by other teams investigating this valve. Walther et al8 report 9 of 23 patients with mild to moderate regurgitation at 12 months' follow-up. Similarly, Middlemost et al7 describe an incidence of mild regurgitation of 22% at 12 months' follow-up, in their group of patients. In our experience, there was a progressive incidence of late mitral valve regurgitation, from 1 patient at discharge to 10 patients at 4-year follow-up. Other than changes in LV size, we could not identify any other parameters associated with the valve that might have caused this incompetence.
After the evaluation of our results, it became obvious to us that we need to predict the postoperative changes in LV size and adapt the sizing procedure accordingly, if we want to prevent the occurrence of late mitral valve regurgitation. Thus, in a patient with mitral valve stenosis, in whom we predict the left ventricle to increase in size postoperatively, the length of the papillary flaps should be deliberately chosen a little longer than a perfect fit.
This will cause slight billowing of the leaflets in the immediate postoperative period. However, as the ventricle remodels and increases in size, the plane of coaptation will shift into the annular plane. On the other hand, if we predict a ventricle to decrease in size postoperatively, as it is commonly seen in patients with mitral regurgitation, the length of the papillary flaps should be chosen a little shorter at implantation.
However, before we can accurately predict the changes of a ventricle after mitral valve replacement, more extensive research will be necessary than our limited study. Also, sizing the valve and assessing the required length of the papillary flaps will remain a challenging task that will be difficult to express in exact numbers and centimeters, but will require a high level of experience and judgment by the implanting surgeon.
In conclusion, we would like to state that the clinical performance of the St Jude Quattro stentless mitral valve prosthesis in our series has been disappointing. Especially, the high incidence of late valve incompetence and the elevated stroke and mortality rates were worrying aspects. Subsequently, we have discontinued the recruitment of patients for this study.
References
- Postoperative assessment of chordal preservation and changes in cardiac geometry following mitral valve replacement. Eur J Cardiothorac Surg. 1996;10:734–744
- Differences in left ventricular response between rheumatic and myxomatous mitral valve disease following mitral valve replacement. Jpn J Thorac Cardiovasc Surg. 2000;48:751–756
- Mitral valve replacement versus miral valve repair. A Doppler and quantitative stress echocardiographic study. Circulation. 1994;89:132–137
- Performance of stentless vs. stented aortic valve prostheses in the elderly patient. A prospective randomized trial. Eur J Cardiothorac Surg. 2003;23:299–304
- Two years clinical experience with a quadrileaflet stentless bioprosthesis in the mitral position. J Heart Valve Dis. 2000;9:667–673
- Randomized trial comparing partial versus complete chordal sparing mitral valve replacement: effects on left ventricular volume and function. J Thorac Cardiovasc Surg. 2002;123:707–714
- The stentless quadrileaflet bovine pericardial mitral valve: early clinical results. J Heart Valve Dis. 1999;8:174–179
- Early clinical results after stentless mitral valve implantation and comparison with conventional valve repair or replacement. Circulation. 1999;100(Suppl II):II-78–II-83
PII: S0002-8703(05)00878-1
doi:10.1016/j.ahj.2005.08.022
© 2006 Published by Elsevier Inc.
