| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiovascular and Thoracic Surgery, Deutsches Herzzentrum Berlin, Berlin, Germany
b Department of Congenital Heart Disease/Pediatric Cardiology, Deutsches Herzzentrum Berlin, Berlin, Germany
Accepted for publication March 12, 2008.
* Address correspondence to Dr Delmo Walter, Deutsches Herzzentrum Berlin, Augustenburger Platz 1, Berlin, 13353, Germany (Email: delmo-walter{at}dhzb.de).
Presented at the Poster Session of the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods: Between June 1987 and December 2006, 111 children (mean age, 7.5 ± 5.9 years) with congenital and acquired MV diseases underwent MV reconstruction. Six children were aged younger than 3 months old, 28 were 3 months to 2 years, and 77 were 2 to 18 years old. Congenital MV lesions were found in 84.6%, isolated MV disease was found in 54.1%, and MV insufficiency was the predominant pathophysiology in 80%. Various standard repair techniques and our own modifications were used according to the lesions.
Results: Early mortality was 4.5%, and late mortality was 7.3%. Actuarial survival at 10 years was 77.4%. Actuarial reoperation-free survival at 10 and 15 years was 79.2%. At 19 years, freedom from MV replacement was 81.8% ± 7.5%, and freedom from repeat reconstruction 91% ± 1.5%. Mean follow-up was 5.4 years. Age younger than 3 months, urgency of operation, concomitant procedures, and coexisting anomalies were strong predictors of poor overall freedom from reoperation and decreased early and late survival. The highly satisfactory results were achieved by careful structural and functional assessment of the valve, avoidance of prosthetic material, and use of a spectrum of repair techniques tailored to the individual case that address all components of the valve lesion.
Conclusions: Mitral valve reconstruction in children using various surgical techniques provides satisfactory early and long-term survival and clinical outcome with low reoperation rates.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In our institution, MV reconstruction is the preferred technique for any kind of MV disease in infants, children, and adolescents. This avoids the need for valve replacement with all its drawbacks, particularly in infants and small children, in the face of the complete lack of a prosthesis that is suitable for this age group. Even when the primary repair result is not optimal, time is gained for repeated repair until a definitive adult-size prosthesis can be implanted.
We believe that reconstruction allows for valve growth without the need of anticoagulation. This is best achieved by using a spectrum of repair techniques applied individually and avoiding any prosthetic material. We reviewed our 19-year experience with surgical reconstruction of MV in pediatric patients to assess our techniques and determine early and long-term survival and freedom from reoperation and valve replacement.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Patients
Between June 1987 and December 2006, we surgically treated 437 pediatric patients (age < 18 years) with MV diseases, and 173 (39.6%) underwent operation for either congenital or acquired MV diseases. Excluded from this study were 62 patients with type II mitral insufficiency in whom the severely dysplastic annulus, leaflets, and deformed subvalvular apparatuses were initially repaired but intraoperative echocardiography showed persistent severe mitral insufficiency; hence, after several reconstructive attempts, the decision to replace the MV was made during the same surgical procedure. Also excluded were 264 patients with left-sided atrioventricular valve anomalies associated with septal defects and those with Marfan syndrome and other degenerative diseases. Data were thus available for 111 (25.5%) children who had MV reconstruction using different surgical techniques. Demographic data, surgical techniques used, postoperative course, and follow-up were reviewed.
Demographic Data
The series comprised 67 boys and 44 girls whose mean age was 7.5 ± 5.9 years (range, 1 day to 17.8 years). Children were divided into three age groups: 6 were aged younger than 3 months old, 28 were 3 months to 2 years old, and 77 were 2 to 18 years old. At the time of surgery, 32 patients (27.9%) were in New York Heart Association (NYHA) functional class II, 45 (40.5%) were in class III, and 34 (30.6%) were in class IV. Degree of MV lesions, left ventricular function, and associated anomalies were assessed by echocardiography or cardiac catheterization, or a combination.
Classification of Mitral Valve Lesions
Because several anomalies may coexist, the predominant one was used to classify the lesion according to Carpentier's functional classification [3]. The distribution of children according to this classification is reported in Table 1.
|
Mitral stenosis was present in 22 children (19.8%), including as a component of Shone anomaly in 8 children who all had parachute valves, isolated parachute valve in 4, presence of supramitral membrane in 2, and anomalies of the subvalvular apparatus in 8, such as papillary muscle fibrosis, shortened papillary muscles, chordal agenesis, chordal thickening, and chordal fusion. Severity was evaluated on the basis echocardiographic measurements of MV orifice area (cm2): none, –4 to 6 cm2; mild, 2 to 4 cm2; moderate, 1–2 cm2; and severe, less than 1 cm2.
Associated Lesions
Mitral valve lesions were associated with other congenital defects in 107 patients (96.3%), and 28 (25.2%) had previous operations. Among 8 patients with Shone anomaly, 6 had undergone repair of coarctation of the aorta. Two of these had additional aortic valve balloon dilatation. Another 2 patients had had aortic valve reconstruction and previous aortic valve balloon dilatation, 1 of whom had previous mitral commissurotomy and aortic valve balloon dilatation. Previous operations, either of the mitral or aortic valve or other cardiac anomalies, had also been done in 17 patients with MVI and 3 with mitral stenosis.
Concomitant repair of associated lesions was performed in 74 patients (66.7%), which included repair of complex cardiac anomalies in 47 (63.5%). The other 27 (36.5%) had concurrent repair of the aortic, tricuspid, or pulmonary valve, or both.
Follow-Up
Follow-up data were provided by the Department of Congenital Heart Disease/Pediatric Cardiology and Department of Clinical Studies, Deutsches Herzzentrum Berlin, and by the referring physicians.
Mitral Valve Reconstruction
All MV reconstruction was performed through a median sternotomy under cardiopulmonary bypass and moderate systemic hypothermia (rectal temperature, 28° to 32°C). Antegrade intermittent cold crystalloid cardioplegia with topical hypothermia was used for myocardial protection. Through a left atriotomy along the interatrial groove, annulus, leaflets, chordae tendinea, and papillary muscles were exposed and meticulously inspected to determine the precise nature of the lesion and plan the procedure. Particular attention was given to the leaflet motion and position of the papillary muscles. The concept followed in our institution is preservation of the native valves and avoidance of any prosthetic materials, except for suture materials, whenever possible.
Depending on the underlying valve pathology, various reconstruction techniques were used. Because malformation usually implies several anomalies, several repair steps may have to be used in the same patient. Suture used for repair in children was 5-0 to 7-0 polypropylene, according to age. Whenever necessary, pledgets and annular reinforcement strips from untreated autologous pericardium were used.
Intravalvular saline injection and intraoperative transesophageal echocardiography (TEE) were routinely performed to assess the adequacy of repair. Postoperative transthoracic echocardiography was done annually or if clinically indicated by symptoms.
Regardless of the underlying pathology and techniques used, no patient was discharged from the hospital with more than mild MVI.
Modified Kay-Wooler annuloplasty was most frequently used in small infants with annular dilatation. It was performed by shortening the segments of the posterior annulus next to both trigones by polypropylene sutures pledgeted with untreated autologous pericardium (author modification, R. H.; Fig 1A, 1B).
|
|
|
|
Resection of long blocks of septal myocardium between the two incisions is started just below the aortic annulus of the right coronary sinus and the commissure between the right and the left coronary sinuses (author modification, R. H., Fig 4E). The incision should be continued apically beyond the point of mitral-septal contact, usually marked by a fibrous band. This wide incision beneath the aortic valve improves exposure of the important area towards the apex. The mean (range) intraoperative preseptal myectomy pressure gradient was 60 ± 25 (40 to 105) mm Hg and the mean (range) postseptal myectomy gradient was 5 ± 6 (0 to 18) mm Hg.
After septal myectomy and ALRP, the aortic and mitral valves were inspected to ensure that they had not been injured. Pressures were remeasured in the left ventricle and aorta, and the TEE evaluation was repeated after weaning from cardiopulmonary bypass. If myectomy has been successful, there will be little or no residual gradient and little or no SAM of the MV. Overall, postmyectomy MVI was reduced to a regurgitant fraction of 0% to 10%. There were no early or late deaths or reoperation for repeat myectomy or repeat MV repair or replacement. No instance of mitral stenosis occurred.
Repair of Hammock Valve
In the repair of a hammock valve (Fig 5A) in the absence of any papillary muscle, a suitably thick part of the posterior left ventricular wall carrying the rudimentary chordae is carved off the wall (Fig 5B). Then it must be ensured that both the remaining left ventricular wall and the "new papillary muscles" maintain sufficient muscle thickness to perform their function (Fig 5C).
|
|
Mitral valve reconstruction was performed as follows: anterior commissuroplasty (Kay-Wooler technique, author modification, R. H.; Fig 1A, 1B) in 2 patients, posterior commissuroplasty with leaflet resection (Fig 2A) in another 2 patients, posterior commissuroplasty with pericardial strip reinforcement (author modification, R. H., Fig 2B, 2C) in 3 patients, and chordal rupture was repaired by chordal reimplantation in 1 patient.
Other Repair Strategies
Flexible ring annuloplasty was used in 4 patients (3.6%) in our early years. Mitral cleft in this series was restricted to the anterior leaflet. The posterior leaflet appeared normal in size. There was no accompanying annular dilatation, abnormal subvalvular apparatus, or subaortic obstruction due to the chordal attachments of the cleft. The isolated anterior mitral leaflet cleft was corrected by a direct suture technique, either completely or partially closed, as necessary, according to minimal acceptable age-dependent MV diameter to avoid mitral stenosis. Suture of the cleft was impossible in 2 patients in our series because of retraction of both parts of the anterior MV leaflet. One patient underwent augmentation of the anterior leaflet with a pericardial patch, and the other underwent an Alfieri procedure. Leaflet perforation was directly sutured or closed with a pericardial patch. Supramitral membrane, when found, was excised. All concomitant congenital heart anomalies were repaired accordingly.
Statistical Analysis
All data were analyzed with SPSS 12.0 software (SPSS Inc, Chicago, IL). Categoric data are expressed as absolute and percentage frequency values and continuous data as mean ± standard deviation. Early mortality was defined as death in the hospital or within 30 days after operation. Categoric variables and the association of risk factors with mortality and reoperations were analyzed with the Pearson 
2 test, two-tailed Fisher exact tests of the displayed proportions, and odds ratios. Cox regression analysis was used to determine the multivariable predictors of survival and freedom from reoperation. A value of p 
0.05 was considered significant. Survival and freedom from reoperation were analyzed according to Kaplan-Meier estimates with 95% confidence intervals (CI).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Late Death
Nine (8.1%) late deaths occurred among the 106 patients who were discharged from hospital. A 2-month-old infant with congenital mitral stenosis who underwent MV replacement with a heterograft 4 months after the initial reconstruction did not survive the second procedure. A child with severe anterior mitral prolapse who was 7 months old at the time of initial operation died of a noncardiac event 5 years postoperatively. A 1-year-old child who had myocardial infarction due to an anomalous coronary artery died 11 months later. A patient with isolated parachute valve who was 2 years old at the time of the initial operation underwent repeat MV reconstruction 5 years postoperatively. The patient had MV replacement 2 years later, but died 8 years postoperatively. An 8-year-old patient with Shone anomaly died from a noncardiac event 13 years postoperatively. Two 10-year-old children with cardiomyopathy died of heart failure, respectively, at 2 and 7 months postoperatively. Another 13-year-old patient, who had Shone anomaly, died of unknown causes 2 years postoperatively. A 17-year-old girl with multiple endocrinologic abnormalities and severe MVI from a dysplastic posterior leaflet died from a noncardiac cause 6 years after successful initial MV reconstruction.
Actuarial survival was 95.55% ± 2.6% at 30 days, 88.4% ± 3.2% at 1 year, 85.5% ± 3.7% at 5 years and 77.4% ± 5.1% at both 10 and 19 years (Fig 7A). Overall survival rates by age group during a 19-year period are shown in Figure 7B.
|
Eight patients eventually underwent MV replacement at a mean of 6.1 years (range, 4 months to 17 years) postoperatively. Overall freedom from MV replacement was 96.3% ± 1.8%, 93.2% ± 2.8%, 87.4% ± 4.9%, and 81.1% ± 7.5% at 1, 5, 10, and both 15 and 19 years, respectively (Fig 7E).
Multivariate analysis revealed that age at operation, concomitant operations, urgency of operation, operative times, and valve status on discharge were risk factors for death and reoperation (Table 2).
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
We encountered MVs significantly lacking in valve tissues or severely dysplastic, with severe deformation of the subvalvular apparatus that rendered reconstruction impossible, as in 62 patients who were excluded from this study. Attempts to reconstruct and preserve their valves proved futile because satisfactory functional results could not be achieved; hence, their valves were replaced during the same surgical procedure. From this experience, we learned that the main goal of surgical repair should be to achieve a satisfactory, if not ideal, MV function, rather than an ideal anatomic or morphologic reconstruction.
The key to a satisfactory outcome is thorough preoperative evaluation and understanding of the valve abnormality with meticulous attention to the anatomic and functional features of the MV apparatus and the precise mechanisms causing stenosis or insufficiency. Attempts to preserve the native MV should be encouraged, especially in infants and young children, because MV repair offers the advantages of avoiding thromboembolism, preserving chordal and subvalvular apparatus function, and making reoperation unnecessary [11].
Annular dilatation and prolapsed leaflet were frequently present in this population. We tried to avoid placement of a rigid ring prosthesis for stabilization or to correct annulus abnormalities because of concerns about anticoagulation, subsequent somatic growth problems, and the risk of a rigid prosthesis causing distortion of the heart cavities or contributing to LVOT obstruction, or both [7].
Several studies consider ring annuloplasty for MV incompetence obligatory in children aged older than 2 years [12–14]. This concept is supported by their experience of a 25% incidence rate of significant residual mitral regurgitation after repair without ring insertion [12]. Other groups demonstrated that other types of annuloplasty techniques can be used successfully in children and that prosthetic rings are not indispensable for achieving favorable results [15, 16]. In most of our patients, we used the commissure plication annuloplasty to correct the annular dilatation, which was reported to yield adequate long-term functional results [17]. We found that annuloplasty using the techniques of Kay-Wooler, Paneth, and Gerbode with reinforcement by an autologous pericardial strip [18] was highly satisfactory.
We found that hammock MV, dysplastic with shortened chordae directly inserted in a muscular mass of the posterior wall of the left ventricle, resulting in tethering of both leaflets, was one of the most challenging malformations to correct, as was also stated by other authors [5, 8, 15, 19, 20]. The key to successful reconstruction was mobilization of these shortened chordae by splitting or incising them off the posterior muscular ventricular wall.
Surgical repair of congenital mitral stenosis has been reported to be associated with greater postoperative mortality and morbidity [15, 19]. Of the 22 patients in this series with mitral stenosis, 12 had parachute MVs. Leaflets and commissures were normal but the chordae tendinea were short and thickened, reducing leaflet motion. Reconstruction was a combination of splitting of the parachute, commissurotomy, and fenestration. All patients did well.
In parachute and hammock valves, the degree and extent of incision or fenestration and commissurotomy is determined by measurement with a Hegar dilator, based on the minimal age-related acceptable MV diameter to avoid mitral stenosis. Adequate reconstruction becomes a balance between residual stenosis and induced insufficiency, assessed intraoperatively with TEE.
Shone anomaly required an equally challenging surgical strategy. Understanding its MV morphology is critical to determine the reconstructive approach. Parachute MV and supravalvular mitral ring are the most prevalent variants of mitral stenosis in this disease [21]. Supraannular fibrous ring, a characteristic feature, rarely occurs as an isolated lesion and is not usually severely obstructive. Resection of this ring should be straightforward [22]. Although recurrence has been described [22], our experience and that of others [22–24] showed that it is rare. Even when the parachute is associated with significant subaortic stenosis, we believe that such lesions are amenable to repair because the main obstructive mitral element is subvalvar.
Most of our patients presented initially in the neonatal period with coarctation of the aorta and some type of LVOT obstruction effectively treated by transaortic resection. Outcomes are related to the degree to which mitral stenosis can be relieved. Satisfactory hemodynamic results were achieved in this series by splitting the papillary muscle and release and separation of the chordae tendinea to open the obliterated chordal spaces and increase the effective mitral orifice.
Operations in children with HOCM were technically challenging because of the difficulty of exposure of the smaller structures. Our 9 patients with HOCM had associated moderate to severe MVI. Our reconstructive approach has been modified to avoid the potential development of SAM after septal myectomy and mitral leaflet repair. We addressed the prolapsing anterior leaflet by performing anterior leaflet retention plasty, which we found to be excellent in restricting MV motion, allowing more complete relief of subaortic obstruction and MVI and avoiding SAM. It is very important to evaluate the presence of abnormalities of the subvalvular apparatus in HOCM, because failure to recognize and treat them may be fatal or lead to incomplete relief of obstruction [25]. Symptomatic improvement of the 9 patients was gratifying, and all remain improved by at least one functional class. There has been no incidence of mitral stenosis after ALRP in this series.
In clefts with an otherwise normal MV, most repairs are accomplished by direct and complete suturing the edges of the cleft by using an age-related minimal normal valve diameter as a guide to prevent valve stenosis.
Our study demonstrates that MV reconstruction can be safely performed in patients with MV lesions as complications of acute and chronic endocarditis. Our results are in agreement with the findings of Muehrcke and colleagues [26], who advocated early intervention and repair to prevent leaflet destruction and vegetation embolization and to preserve left ventricular function. Likewise, Talwar and associates [27] found acceptable long-term results in repairing rheumatic valves in their large series of 278 patients.
No patients who underwent MV reconstruction received anticoagulation, except for 2 patients who had undergone previous aortic valve replacement. There was no incidence of thromboembolism in this series. In contrast, Aharon and colleagues [4] reported a patient in their series who had transient right-sided paralysis despite adequate anticoagulation after MV repair and aortic valve replacement.
Few reports on MV repair in children have identified predictors for poor outcome in this group, probably because of the small numbers of patients in each report. We identified age younger than 3 months, urgency of operation, concomitant procedures, operative times, and valve status on discharge as risk factors for death and reoperation.
Durability of repair is a major setback of MV reconstruction in children. With meticulous intraoperative assessment of valve morphology and careful selection of appropriate reconstruction strategy, repair can be long lasting. Our actuarial freedom from reoperation at 10 and 19 years are encouraging, especially in this population where 30% of patients were younger than 2 years old. We have continued to modify our surgical techniques to optimize our results.
In conclusion, MV reconstruction is the surgical technique of choice in our institution for any kind of mitral disease in childhood. We believe that reconstruction allows continuous somatic and valve growth, delays or eliminates the need for future valve replacement and lifelong anticoagulation, and obviates the known complications of valve replacement, which frequently requires subsequent reoperation to implant a larger prosthesis. It must be assumed that most, if not all, valves repaired during childhood, will eventually have to be replaced at some time in life. The concept of repair in childhood is primarily aimed at growth of the patient to an age when, if necessary, an adult-sized prosthesis can be implanted. Strong predictors for poor overall survival and freedom from reoperation are age younger than 3 months, urgency of surgery, concomitant procedures, and long operative times.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. M. B. Delmo Walter, H. Siniawski, and R. Hetzer Sustained improvement after combined anterior mitral valve leaflet retention plasty and septal myectomy in preventing systolic anterior motion in hypertrophic obstructive cardiomyopathy in children Eur. J. Cardiothorac. Surg., September 1, 2009; 36(3): 546 - 552. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Komoda, M. Huebler, F. Berger, and R. Hetzer Growth of mitral annulus in the pediatric patient after suture annuloplasty of the entire posterior mitral annulus Interactive CardioVascular and Thoracic Surgery, August 1, 2009; 9(2): 354 - 356. [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 |
