| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;82:1585-1592
© 2006 The Society of Thoracic Surgeons
a Pediatric Cardiology, Necker Enfants Malades, Paris, France
b Cardiac Surgery, Necker Enfants Malades, Paris, France
Accepted for publication May 11, 2006.
* Address correspondence to Dr Agnoletti, Service de Cardiologie Pédiatrique, Groupe Hospitalier Necker Enfants Malades, AP-HP, 149, rue de Sèvres, 75743 Paris, France. (Email: gabriella.agnoletti{at}nck.ap-hop-paris.fr).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
METHODS: From January 1, 1990, to January 1, 2005, 65 neonates were referred for critical aortic stenosis. Among these 36 consecutive patients underwent surgical valvotomy by the same surgeon and 16 patients had balloon valvotomy. We examined clinical records to establish determinants of outcome and illustrate long-term results.
RESULTS: In the surgical group 6 patients had nonstenotic parachute mitral valve, 6 had aortic coarctation, 13 had endocardial fibroelastosis; and 15 had depressed fractional shortening. In the balloon valvotomy group most infants had associated mitral stenosis (n = 7), small aortic annulus (n = 5), multiple anomalies (n = 2), endocardial fibroelastosis (n = 9), and depressed fractional shortening (n = 13). In the surgical group global mortality was 19%; 47% in infants with depressed fractional shortening and 0% in infants with normal fractional shortening. Mortality in the balloon valvotomy group was 56%. Low fractional shortening, low mean aortic gradient, endocardial fibroelastosis, and small aortic annulus were associated with fatal outcome. In the surgical group 6 patients needed Ross operation at a median interval of 4 years from commissurotomy. In the balloon valvotomy group 3 patients needed early mitral replacement and 5 aortic valve surgery. At a median follow-up of 6 years, all children are in New York Heart Association class I, all have a normal ventricular function, 2 have aortic restenosis, and 5 have at least moderate aortic incompetence.
CONCLUSIONS: In a selected population of neonates with isolated aortic stenosis, surgical valvotomy offers an excellent long-term outcome to patients with a normokinetic left ventricle, while patients with a hypokinetic left ventricle have a poor outcome. Balloon valvotomy offers poor results when performed in patients with complex anomalies.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Satisfactory short-term and long-term results have been reported in neonates as well as in older children with percutaneous valvuloplasty [1] and surgical valvotomy [2]. In our institution we prefer elective surgical treatment, either in neonates or older children. In fact, if percutaneous balloon valvotomy usually causes rupture along lines of least resistance, either along underdeveloped commissures or into leaflet tissue [3], surgical valvotomy allows direct inspection of the valve, more fashioning of commissurotomies, and debridement of any excess tissue on the leaflets.
The aims of this study were to analyze our results in 36 consecutive infants undergoing commissurotomy and to possibly identify predictors of outcome.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
The present article concerns only 36 consecutive neonates undergoing surgical aortic commissurotomy, by a single surgeon (P.V.). During the same period 16 neonates underwent percutaneous valvuloplasty. The results in this group are briefly illustrated, even if, owing to the complexity of their disease, they could hardly constitute a control group. The policy of our institution was to propose valvotomy only to patients satisfying the Rhodes criteria for biventricular correction [4]. Exclusion criteria were any previous aortic procedure (percutaneous or surgical) and the presence of other intracardiac defects except aortic coarctation or patent ductus arteriosus. Only infants with nonstenotic parachute mitral valve underwent aortic commissurotomy, whereas patients with mitral annulus hypoplasia or mitral stenosis underwent other therapeutic options. When reintervention was necessary, surgical valvotomy was not repeated; Ross operation or balloon valvotomy, depending on the presence of aortic incompetence and valve morphology, was performed.
Patient recruitment did not change during the period of study. Hospital records, surgical files, and any document concerning last follow-up were reviewed to establish possible predictors of outcome. All children underwent two-dimensional echocardiography, Doppler investigation, and color Doppler mapping. Postoperative aortic regurgitation was evaluated by color Doppler and graded as mild, moderate, or severe, according to the criteria previously described [5]. This study was approved by our institutional review board (April 11, 2006).
Surgical Technique
The operations were performed by one single surgeon. The surgical technique did not change during the period of study. Median age and weight at operation were 16 days (range, 2 to 30 days) and 3.5 kg (range, 2.5 to 4.7 kg), respectively. The details of the initial surgical procedure were based on intraoperative inspection of the valve, as previously described [6]. After median sternotomy, hypothermic cardiopulmonary bypass was instituted. Myocardial protection was provided with blood warm cardioplegia and warm reperfusion. Short periods of circulatory arrest were used to improve visualization. After transverse aortotomy, the aortic valve was inspected, and a commissurotomy to the aortic annulus was performed. Only identifiable commissures were incised right back to the annulus to enlarge the valve orifice as much as possible. Rudimentary commissures were opened to the annulus when the valve was unicommissural. Any excessive nodular fibrous tissue below the commissure or on the free edge of the leaflets was excised. Eleven patients required concomitant ligation of patent ductus arteriosus, and 5 had terminoterminal repair of coarctation of the aorta.
Follow-Up
Duration of follow-up was defined as the period between the initial operation and the time of last patient contact. Follow-up echocardiography was performed at Necker Enfants Malades in most cases. For patients being monitored elsewhere, the local physician was contacted and asked to provide follow-up information including details on any further intervention and results of the most recent echocardiographic study.
Statistical Analysis
Descriptive statistics for the total population were obtained. Age and weight were quoted as median and a range; continuous variables were presented as the mean ± standard deviation. Comparison for individual variables before and after surgery was performed using the two-tailed paired Student's t test. Linear regression analysis was done using the correlation coefficient r. Analysis of postoperative time-related events was by the Kaplan–Meier method, using the log-rank test for group comparisons. Differences between proportions were evaluated using the 
2 test.
Independent risk factors for death and reintervention were sought using univariate regression analysis. Aortic annulus diameter, end-diastolic left ventricular diameter, end-systolic left ventricular wall thickness, fractional shortening (FS), and mean aortic gradient were analyzed as continuous variables. Presence of endocardial fibroelastosis, and number of aortic cusps in the preoperative and postoperative periods were analyzed as discrete variables. The relationship between time-related events and variables was investigated by using Cox proportional-hazards model, after proportionality of hazards had been checked. A two-sided probability value of 0.05 or less was considered to indicate statistical significance. Statistical analyses were conducted with R software, version 2.00 (Auckland, New Zealand).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
35%) in 21. No patient had more than trivial aortic incompetence. Thirteen infants were receiving prostaglandin E1 infusion, and 4 had retrograde aortic perfusion. End-diastolic left ventricular diameter was inversely correlated with FS (r = –0.7), but not with end-diastolic posterior left ventricular wall thickness.
|
|
|
illustrate survival curves. Overall survival was 80% at 13 years. However, infants with initially depressed FS had a significantly diminished survival (54% at 13 years; p < 0.05). At echocardiography, performed within 24 hours of surgery, FS increased in the global population, from 31.3% ± 9.7% to 36.1% ± 4.6% (p = 0.038). Infants with initially depressed ventricular function who survived recovered an almost normal ventricular function in the early postoperative period (FS from 21.3% ± 2.8% to 33.8% ± 1.9%; p < 0.001). Durations of extracorporeal circulation and cross-clamping were not correlated with postoperative FS.
|
|
In the immediate postoperative period, aortic incompetence appeared in 9 patients (25%) and was moderate in 4 and mild in 5. It occurred in 50% of patients having a unicuspid valve, in 8% of those having a bicuspid valve, and in 33% of those having a tricuspid valve. This difference was not statistically significant. Surgical technique insignificantly influenced the occurrence of aortic regurgitation, which appeared in 30% of patients having a bileaflet correction and in 22% of those having a trileaflet correction. Among patients who died, 2 had mild aortic regurgitation.
Patients having less than moderate aortic regurgitation and a mean aortic gradient less than 30 mm Hg (n = 23) were considered to have a good early surgical result. The percentage of infants having a good early surgical result did not change during the period of the study.
Outcome of Patients Undergoing Balloon Valvotomy
These data are illustrated in Table 2. There were 9 (56%) deaths caused by persistent left ventricular dysfunction and metabolic acidosis, despite intensive medical support (patients 2, 6, 7, 8, and 10), persistent pulmonary hypertension after Ross operation and mitral replacement (patient 5), plurimalformation and neonatal hepatitis (patient 9), and early Konno-Ross operation (patient 15). There was 1 sudden death (patient 4). Moderate to severe aortic incompetence appeared in 7 infants (44%). Severe pulmonary hypertension was a major finding in this group (patients 5, 6, 7, and 9).
Among 7 surviving infants, 4 did not have a surgical treatment and 3 had surgical valvotomy, aortic and mitral valve replacement, or Ross operation followed by mitral replacement, 6 days to 6 months after balloon valvotomy. Among the nonoperated-on patients, one had a periprocedural stroke without sequelae, one has a well-tolerated aortic incompetence, and none have pulmonary hypertension. Thus, at a median follow-up of 3 years (range, 1 to 14 years), 13% of patients have a good long-term result.
Follow-up of Patients Undergoing Surgical Valvotomy
Percutaneous aortic valve dilation was performed in 7% of patients, at the age of 10 and 4 years, respectively. In patient 15 it failed, and the patient had a Ross operation.
Ross operation was performed in 20% of patients. In patients 2 and 9 it was performed after 5 years and 3 months, respectively, owing to increase of aortic incompetence. In patients 15, 19, and 26 it was performed 4, 6, and 2 years, respectively, after commissurotomy, owing to increase of aortic gradient. Patient 2 was reoperated on 3 years after the Ross operation for substitution of the right ventricle to pulmonary artery conduit. There were no interventions on the mitral valve or on the aortic arch during the follow-up period.
At a median follow-up of 6 years (range, 1 to 14 years), all 29 survivors are in New York heart Association class I and have a normal ventricular function. Nine have mild, 3 have moderate, and 1 has severe aortic incompetence. Mean aortic gradient is 19.2 ± 9.1 mm Hg (range, 5 to 42 mm Hg). Four children have a mean aortic gradient higher than 30 mm Hg. Ross intervention will be considered in patient 1, who has a mean aortic gradient of 42 mm Hg and in patient 19, who has more than moderate aortic regurgitation. There are no patients with pulmonary hypertension.
Predictors of Outcome in Group of Patients Undergoing Surgical Valvotomy
Factors associated with increased mortality were size of the aortic annulus (p = =0.007), endocardial fibroelastosis (p = 0.05), FS less than 35% (p = 0.01), and low aortic gradient (p = 0.005). No factors were associated with reoperation only. The only factor associated with mortality or reintervention was the size of aortic annulus (p = 0.002). Association of predictor factors did not allow us to identify infants with a fatal outcome. Factors not associated with mortality or reintervention were left ventricular end-diastolic diameter, left ventricular end-diastolic wall thickness, and number of aortic cusps before or after surgery.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Mortality
In the surgical group all deaths were intrahospital, invariably attributable to low cardiac output, and occurred only in infants with initially depressed ventricular function. The unavailability of pediatric left ventricular assistance in our hospital probably increased the mortality in this group. Mortality in the balloon angioplasty group was extremely high compared with previous reports [15]; however, the minority of infants had normal ventricular function, nonstenotic mitral valve, and normal aortic annulus. In this group death was often multifactorial, pulmonary hypertension being a common clinical feature [16]. It is well known that congenital aortic stenosis can be assimilated to a chronic ischemic disease. Unfortunately, we are not able at present to distinguish reversible from irreversible ischemic damage in these patients; thus, we could not identify preoperatively the subset of patients with initially depressed ventricular function whose ventricular function improved after surgery or balloon angioplasty. A complete recovery of ventricular performance can be explained by the fact that in aortic stenosis, excess afterload may cause the ejection fraction to be reduced in the face of relatively normal contractility. The notion that afterload mismatch could exist in newborns with congenital aortic stenosis has never been studied before. It is possible that test infusion of inotropic drugs could temporarily improve ventricular function, identifying this subset of patients.
The association of diminished aortic annulus and depressed ventricular function was always fatal in our series. The high-risk association of these two conditions should be verified in larger series. A small aortic annulus offers limited possibility of repair. It is also possible that a small aortic annulus indicates an earlier onset of the disease, and this could explain the association with a poor ventricular function. When small aortic annulus and depressed ventricular function are associated, surgical or therapeutic options other than surgical commissurotomy could be considered, including balloon dilation as a bridge to surgery, neonatal Ross operation, Norwood operation, or possibly neonatal double switch operation.
None of our patients had preoperative aortic incompetence. Aortic incompetence was relatively frequent after commissurotomy; however, it was never severe and never required an early reoperation. Aortic incompetence was a frequent event in the balloon angioplasty group, leading to death or aortic replacement in all but 1 patient. Our bad results [15] can be explained by the fact that only complex patients were treated percutaneously.
Unlike other authors [10], we found that initial morphology of aortic valve and type of correction were not associated with an increased incidence of aortic incompetence. This could be related to either our relatively small series of patients or the fact that neonatal aortic valve allows different types of correction.
Mitral Valve and Coarctation of the Aorta
Mitral stenosis and small mitral annulus are often associated with aortic stenosis and increase mortality and morbidity of surgical treatment. On the contrary, it has been shown that the presence of nonstenotic parachute mitral valve is not associated with increased morbidity and is well tolerated at long-term follow-up [17]. We confirmed the benign nature of nonstenotic parachute mitral valve, as none of the infants of our surgical series underwent interventions on the mitral valve. On the contrary, all surviving infants with mitral stenosis had early mitral replacement.
Coarctation of the aorta is also associated with aortic stenosis, likely because of reduced blood flow through the aortic isthmus during fetal life. Although correction of coarctation of the aorta during the neonatal period has a 10% to 30% risk of recoarctation [18], we, as other authors, did not report any recoarctation in patients having had neonatal surgical aortic valvotomy and coarctation repair. This is likely because of the fact that sternotomy allows a better visualization of aortic arch, compared with thoracotomy, thus allowing optimization of surgical result.
Predictors of Outcome
In accordance with previous reports, we showed that small aortic annulus, depressed FS, low aortic gradient, and endocardial fibroelastosis were factors associated with increased mortality. On the contrary, left ventricular end-diastolic diameter and left ventricular end-diastolic wall thickness were not associated with mortality or reintervention. This can be explained by the fact that a dilated left ventricle can maintain a normal FS, and very few infants, likely having had an earlier onset of the disease, had a thin left ventricular posterior wall.
Follow-Up
We defined as good early surgical result a mean transvalvular aortic gradient less than 30 mm Hg and absent or more than mild aortic regurgitation. We did not include in our definition information on systolic left ventricular function, as all patients had a normal or almost normal ventricular function in the immediate postoperative period. Previous long-term studies of hospital survivors suggest that the beneficial outcome of surgical valvotomy are maintained at 15 to 20 years [2, 13]. We confirmed these data, as freedom from events in patients with good early result was 75% at 14 years' follow-up; however, we could have introduced a bias when defining good early surgical result.
Concerning patients undergoing a balloon valvotomy, because of the inclusion criteria in this series, only 2 of 16 patients are doing well at long-term follow-up.
Study Limitations
Our study has several potential limitations. Patients were not randomized, and balloon angioplasty was performed only in patients with complex diseases or very poor clinical condition. In addition, endocardial fibroelastosis resection and techniques prompting left ventricular growth were not performed in our department. The availability of pediatric left ventricular assistance could have provided a better understanding of reversibility of left ventricular dysfunction.
Echocardiography was not performed by the same operator, even if measurement of end-diastolic and end-systolic left ventricular diameter and calculation of FS are supposed to have a low interobserver variability. More sophisticated echographic investigations, including evaluation of diastolic left ventricular function, were not performed. Further studies aimed at identifying infants with irreversible ischemic damage of the left ventricle should be advocated.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  A. Kadner, O. Raisky, A. Degandt, D. Tamisier, D. Bonnet, D. Sidi, and P. R. Vouhe The Ross Procedure in Infants and Young Children Ann. Thorac. Surg., March 1, 2008; 85(3): 803 - 808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Hickey, C. A. Caldarone, E. H. Blackstone, G. K. Lofland, T. Yeh Jr, C. Pizarro, C. I. Tchervenkov, F. Pigula, D. M. Overman, M. L. Jacobs, et al. Critical left ventricular outflow tract obstruction: The disproportionate impact of biventricular repair in borderline cases. J. Thorac. Cardiovasc. Surg., December 1, 2007; 134(6): 1429 - 1437.e7. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Hraska, J. Photiadis, and C. Arenz Open valvotomy for aortic valve stenosis in newborns and infants MMCTS, June 19, 2007; 2007(0619): 2311. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Monro Invited commentary Ann. Thorac. Surg., November 1, 2006; 82(5): 1592 - 1593. [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 |
