| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2003;24:723-730
© 2003 Elsevier Science NL
a Department of Clinical Physiology, Sahlgrenska University Hospital, SE-413 45 Gothenburg, Sweden
b Department of Cardiothoracic Surgery, Sahlgrenska University Hospital, SE-413 45 Gothenburg, Sweden
Received 24 May 2003; received in revised form 28 July 2003; accepted 30 July 2003.
* Corresponding author. Tel.: +46-31-342-2735; fax: +46-31-828978
e-mail: odd.bech-hanssen{at}klinfys.gu.se
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
Key Words: Echocardiography • Mitral regurgitation • 5-year survival
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
The treatment of patients with MR has changed in recent decades, with an increasing proportion undergoing valve repair instead of valve replacement. The reported excellent short- and long-term results of mitral repair [1–5] can potentially influence the policy regarding the timing of mitral surgery. Some authors now recommend surgery in patients without symptoms or signs of left ventricular dysfunction, if the likelihood of successful repair is high and the operative risks are low [6,7]. Operating on asymptomatic or mildly symptomatic patients requires the diagnosis of the anatomic lesion and assessment of the likelihood of repair to be correct and the operative mortality as well as the late failure of repair to be low. Previous reports that recommend early surgical intervention in patients with severe MR are often based on lessons learned from patients with flail leaflets and they reflect a surgical standard that may be not representative for many institutions [4,5,8]. Our surgical policy regarding the timing and choice of procedure (repair or replacement) has been relatively conservative and is probably representative of the standard at many institutions.
In the present retrospective study, we investigate all the patients undergoing mitral valve surgery from 1995 to 1999, with the aim of describing the importance of both clinical and preoperative echocardiographic variables in relation to the outcome.
|
2. Material and methods |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
2.2. Surgical procedure
The operation techniques included median sternotomy and cold blood cardioplegia for myocardial protection. The total subvalvular apparatus was preserved in patients undergoing valve replacement, when possible. In patients undergoing repair due to hypermobility of the median portion of the posterior leaflet, a quadrangular resection was performed. In patients with hypermobility of the anterior leaflet either transposition of chorda or a small triangular resections was used. No chordal shortening or artificial chorda procedures were performed. All patients undergoing valve repair received an annuloplasty with a ring inserted.
2.3. Assessment of anatomic lesions by echocardiography
The mitral anatomic lesions were divided into three groups as previously described: (1) functional with normal leaflets, (2) organic degenerated hypermobile leaflets or (3) organic degenerated leaflets without hypermobility [9]. The group with hypermobile leaflets comprises patients with prolapse, flail leaflets due to rupture of chorda and papillary muscle rupture. The group with organic degenerated valves includes rheumatic valve disease and endocarditis.
The reports from the physicians performing the preoperative echocardiograpic examinations were analyzed and the anatomic lesions and the mechanisms involved were classified into one of the three groups (functional, hypermobility or organic degenerated). The echocardiographic tapes were not reinterpreted.
2.4. Echocardiography
M-mode measurements were made according to the recommendations of the American Society of Echocardiography [10]. The ejection fraction was calculated either from M-mode according to Teichholz, or from two-dimensional recordings according to Simpson's rule. When image quality did not permit calculation, the ejection fraction was assessed by eyeballing. The mitral annulus diameter was assessed either by eyeballing or by measurements at the beginning of diastole from the parasternal length axis, apical four-chamber and two-chamber views.
All the patients were investigated by color Doppler, as well as by pulsed and continuous wave Doppler. Mitral flow was recorded at the tips of the mitral leaflets in the four-chamber view. Pulmonary venous flow velocities were obtained by placing the sample volume at the orifice of the upper right pulmonary vein. Continuous wave Doppler signals were recorded by a 2 MHz non-imaging probe to obtain an optimal signal-to-noise ratio and multiple windows were used.
The assessment of MR severity was based on a number of variables: color Doppler jet characteristics including jet width and area, continuous wave Doppler intensity and shape of the spectral recording, mitral inflow, pulmonary vein flow pattern by pulsed wave Doppler, left atrial size, left ventricular dimensions and pulmonary artery pressure.
2.5. Statistical analysis
The results are expressed as the mean±SD. An unpaired Student's t-test was used to compare appropriate data, while the chi-square test was used for proportions. A P-value of <0.05 was considered significant. Operative mortality was defined as death within the first month or during the same period of hospitalization.
5-year cumulative survival (Kaplan–Meier) was compared with the expected survival in an age- and gender-matched normal population. The expected number of deaths was calculated for the patients under the assumption that their risk coincided with that of the normal population. Age, sex and calendar time were taken into account and the comparisons between observed and expected numbers were performed using Poisson distributions.
The death hazard function was estimated using a Poisson model depending on a set of variables [11]. The hazard function was of the form exp(ß0+ß1x1+ß2x2+...+ßkxk), where the beta values are coefficients and x1,x2,...,xk are the values of the variables. The analysis was performed stepwise. The following variables were included: age, gender, time since surgical treatment, concomitant coronary artery bypass grafting (CABG), repair or prosthesis, left ventricular ejection fraction (EF) and NYHA class. In the last step, only the variables of significant importance (P<0.05) were included.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
|
|
3.2. Anatomic lesions and surgical strategy
The echocardiographic assessment of anatomic lesions and mechanisms responsible for the MR were functional MR in 19%, hypermobile leaflets in 63%, organic degenerated in 10%, previous mitral surgery in 4% and not possible to classify from report in 4%.
In the patients with functional MR, the left ventricular systolic function was abnormal in the majority of the patients with regional dysfunction in 66% and uniform dysfunction in 21%. Hypermobility engaged the posterior leaflet in 59%, the anterior leaflet in 23% and both leaflets in 16%. Four patients (2%) had a papillary muscle rupture. Three patients in the group with echocardiographic organic degenerated valves had typical rheumatic lesions (10%) and a fairly large proportion (45%) had vegetations.
The proportion of the patients undergoing repair was 58% (of these, 34% with isolated annuloplasty) and, among those who underwent valve replacement, 79% received a mechanical prosthesis. The proportion of repair in patients with functional MR, hypermobile leaflets and organic degenerated leaflets was 84, 58 and 28%, respectively. Hypermobility in the posterior leaflet led to repair in 78% of the patients. Patients with hypermobility in the anterior or both valves underwent repair in 29% and 23% of cases, respectively. The proportion with repair was 28% in patients with organic degenerative valve disease.
3.3. Operative mortality and reoperations
Thirteen patients (overall 4.4%) died within 1 month of operation or during the same period of hospitalization. In patients without CABG (n=172), in-hospital mortality was 2.2% compared with 7.6% in patients with CABG (P=0.03). Patients who died early or in hospital were older (71±5 vs. 64±11 years, P=0.01), they had concomitant CABG performed more frequently (69% vs. 39%, P=0.03) and had higher preoperative serum creatinine (133±67 vs. 106±29 mmol/l, P=0.004) compared with those who survived for longer. The significance of female gender was borderline (54 vs. 29%, P=0.06). Operative mortality was not affected by repair or valve replacement or anatomic lesion.
Reoperations were performed in 12 patients with primary repairs and three patients with primary valve replacement. Six patients in the group with primary repairs were reoperated on within the first month and a further three subjects were operated on within the first year. Among patients treated with repair undergoing a reoperation, six (50%) had a hypermobile posterior leaflet, two (17%) had endocarditis and two had functional MR. Freedom from reoperation during the follow-up time (median 45 months) was 92% in the repair group and 98% in the prosthesis group.
3.4. 5-year results
The overall survival after 5 years was significantly lower in the patient group compared with the expected survival in an age- and gender-matched normal population (P<0.001) (Fig. 1 and Table 3). Survival in patients without CABG did not differ significantly from that in the normal population, while patients with CABG had a lower 5-year survival than expected (Fig. 2)
. Patients with preoperative NYHA class III, IV had a lower 5-year survival than the normal population (Fig. 2, P<0.001), while survival in patients with NYHA class I or II did not differ from the survival expected in a normal population, Fig. 2. The observed survival was reduced in patients undergoing both valve replacement (P<0.001) and repair (P=0.03), Fig. 3 and Table 3.
|
|
|
|
|
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
4.1. Excess mortality due to heart dysfunction
All patients undergoing mitral valve surgery due to MR between 1995 and 1999 were included in the study. In overall terms, many patients had severe symptoms (NYHA III/IV), frequent atrial fibrillation, signs of left ventricular dysfunction (EF<60%) and pulmonary hypertension. Severe symptoms at the time of operation did not determine operative mortality, but NYHA class clearly influenced 5-year mortality. Severe symptoms thus indicate an increased risk of irreversible myocardial damage leading to excessive mortality compared with the normal population.
In contrast to other reports, we were not able to establish a link between preoperative left ventricular ejection fraction and outcome [12]. It is possible to speculate that, in patients with severe MR which unloads the ventricle into the left atrium, an apparently normal ejection fraction can obfuscate ventricular dysfunction [12,13]. Indeed, this finding underscores the limitation in the assessment of ventricular function by the ejection fraction in patients with severe MR.
4.2. Timing of surgery
The timing of surgical intervention in patients with severe MR cannot be generalized. According to the current ACC/AHA guidelines, there is general agreement that symptomatic patients with severe MR will benefit from surgical treatment [14]. The benefit of the surgical treatment of asymptomatic patients with severe MR and normal left ventricular function, on the other hand, has not been established [14,15]. However, some experienced centers now recommend surgery in this patient group [6,7,16]. Operating on asymptomatic or mildly symptomatic patients requires the diagnosis of the anatomic lesion and the assessment of the likelihood of repair to be correct, the operative mortality to be very low and the mortality rate after operation to be lower than the risk of death if uncorrected. The data from the present study show that it was possible, using echocardiography, to identify an important subgroup of patients (41% of total) with hypermobile valves without CABG, with low operative mortality (2.3%) and 5-year survival that was no different from that of the normal background population. The long-term outcome is known to be dependent upon preoperative symptoms [17], left ventricular function [18] and the presence of atrial fibrillation [19]. It is therefore conceivable that early intervention will also improve the long-term outcome in patients with hypermobile valves in whom repair can be performed.
From our data, we do not know the extent to which the patients had been followed with clinical and echocardiographic investigations during an asymptomatic phase or whether they were symptomatic at the time of the primary diagnosis. However, patients with hypermobile leaflets and MR are often recognized as a result of a heart murmur when they still are mildly symptomatic or asymptomatic. Approximately 50% of the patients with hypermobile valves were in functional class III/IV and it can therefore be argued that a large proportion of the patients with severe MR were operated on too late. Patients with ischemic functional MR, on the other hand, often present with symptoms primarily caused by their coronary artery disease and the timing of surgery must be individualized.
4.3. Importance of surgical procedure
In the present study, we found that patients undergoing repair had a better outcome than patients undergoing valve replacement. We found important baseline differences (NYHA class, atrial fibrillation) that can partly explain the observed difference. However, using the Poisson regression model we could demonstrate better outcome in the repair group independent of these baseline differences, and the difference between repair and replacement increased with time. Previous studies comparing mitral repair with replacement are conflicting. There are studies that suggest that repair is preferable, but differences in baseline characteristics complicate the interpretation of these results [20,21], while other reports with multivariate analysis do not confirm the superiority of repair [22,23]. However, some more recent reports [1,4] are in agreement with our findings that repair improves the outcome compared with valve replacement. Principally the observed difference in outcome between repair and replacement can be explained by more pronounced left ventricular function dysfunction or valve-related complications in the replacement group. The surgical procedure used during the study period preserved the subvalvular apparatus in order to avoid the deterioration in left ventricular function in patients without chordal support. However, we do not have any data on the postoperative left ventricular function and therefore residual myocardial dysfunction might contribute to the less favorable outcome in the replacement group. Further studies are needed to explain the difference in outcome between the repair and replacement group.
In the study period, the surgical strategy at our institution tended to be conservative, the percentage of repair was 58%, which can be compared with the 75–85% reported from some other institutions [4,9]. Patients with hypermobility in the posterior leaflet and functional MR were treated with repair on a level that can be compared with others [4,9]. However, patients with hypermobility in the anterior leaflet or both leaflets most frequently underwent replacement, as did patients with an infected mitral valve. The fairly low percentage of repair at our institution reflects a concern about the durability of procedures on the anterior leaflet or infected mitral valve. A prerequisite for a change in treatment strategy must be that these patients can be successfully treated with repair. The documented superiority of repair in terms of outcome [4,9], together with more recent good results regarding the durability of valve repair on anterior prolapses [4,8,24] and infected mitral valves [2] should, however, encourage efforts to increase the percentage of valve repair.
4.4. Study limitations
The main limitation of the present study that might have influenced the results is the retrospective design. The echocardiographic reports, the description of surgical findings and the treatment strategy in relation to the echocardiographic and surgical findings were not clearly defined. Several echocardiographers and surgeons participated and it is therefore conceivable that varying skills in echocardiography and surgery might have introduced misinterpretations of anatomic lesions.
4.5. Conclusions
In the study period, most patients were clearly symptomatic and often displayed signs of systolic dysfunction and pulmonary hypertension. It is important to note that, in patients with NYHA class I/II and patients with hypermobile leaflets without coronary artery bypass, the 5-year survival did not differ from that of a normal age- and gender-matched population. Patients undergoing repair had a better outcome compared with the replacement group. It is possible, using transthoracic echocardiography and clinical data, to identify patients with an excellent outcome. The findings encourage early intervention before severe symptoms occur, especially if repair is possible.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. J. M. Westenberg, S. D. Roes, N. Ajmone Marsan, N. M. J. Binnendijk, J. Doornbos, J. J. Bax, J. H. C. Reiber, A. de Roos, and R. J. van der Geest Mitral Valve and Tricuspid Valve Blood Flow: Accurate Quantification with 3D Velocity-encoded MR Imaging with Retrospective Valve Tracking Radiology, December 1, 2008; 249(3): 792 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Determinants of operative mortality in valvular heart surgery. J. Thorac. Cardiovasc. Surg., March 1, 2006; 131(3): 547 - 557.
|
|
|
|
 |
|
 J. J.M. Westenberg, J. Doornbos, M. I.M. Versteegh, J. J. Bax, R. J. van der Geest, A. de Roos, R. A.E. Dion, and J. H.C. Reiber Accurate quantitation of regurgitant volume with MRI in patients selected for mitral valve repair Eur. J. Cardiothorac. Surg., March 1, 2005; 27(3): 462 - 467. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
