|
|
||||||||
Eur J Cardiothorac Surg 2007;32:582-587. doi:10.1016/j.ejcts.2007.06.037
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
Department of Cardiopulmonary Surgery, La Timone Children's Hospital, Marseille, France
Received 2 August 2006; received in revised form 20 June 2007; accepted 28 June 2007.
* Corresponding author. Address: Hôpital Timone Enfants, Service de Chirurgie Thoracique et Cardio-vasculaire, 264 rue Saint-Pierre, 13385 Marseille, France. Tel.: +33 491 386 676; fax: +33 491 384 576. (Email: dmetras{at}ap-hm.fr).
| Abstract |
|---|
|
|
|---|
Key Words: Subaortic stenosis Left ventricular outflow tract Mitral valve Coarctation of the aorta Bicuspid aortic valve Left superior vena cava
| 1. Introduction |
|---|
|
|
|---|
SSS has been reported after surgical repair of several congenital heart defects with or without an initial LVOT obstruction: ostium primum defect, univentricular heart requiring the Fontan procedure [6], ventricular septal defect [7,8], tetralogy of Fallot [7,8], transposition of the great arteries, double outlet right ventricle [7,9,10], discrete subaortic stenosis and Shone's syndrome [11], patent ductus arteriosus, abnormal ventriculoarterial connections [6], partial and complete atrioventricular septal defect [12], common atrium and aorticopulmonary window [8]. Nevertheless, none of these studies have investigated the conditions associated with SSS regardless of the initial congenital heart defects, although such a study could be helpful in understanding this serious postoperative complication.
The purpose of this study, based on a 20-year retrospective cohort, was to highlight the anatomical conditions and surgical techniques associated with the development of secondary subaortic stenosis after surgery on congenital heart defects with no initial LVOT obstruction.
| 2. Materials and methods |
|---|
|
|
|---|
2-test and Student's t-test were chosen to study the qualitative and quantitative data. Significance was accepted at p
< 0.05. | 3. Results |
|---|
|
|
|---|
3.1 Patient characteristics and initial heart defects
Sixty-one percent of the patients were female. None had a family history of subaortic stenosis. Two patients (with AVSD) were trisomic. The age at initial surgery varied between 4 days and 47 years, with a mean age of 32 ± 106 months (median: 2 months). The age of the patients without SSS operated on in our institution during the same period was 74.9 ± 148 months (p
< 10–4). The initial surgery was performed in 43% of the patients before the age of 1 month, and in 68% before the age of 6 months.
The initial heart defects are summarised in Table 1 . The initial heart malformation was complex (association of multiple defects) in 24 patients (86%). Coarctation of the aorta was present in 10% of the patients without SSS operated on in our institution during the same period, versus 40% of the patients with SSS (p < 10–4). Similarly, left superior vena cava and bicuspid aortic valve were, respectively, present in 14% and 21% of the SSS cases versus 4% and 6% in the patients without SSS operated on in our institution during the same period (p = 0.02 and p = 0.002, respectively). Echocardiographic reports and intra-operative findings indicated anatomical lesions of the LVOT and the mitral valve, respectively, in 93% and 32% of the cases before the occurrence of SSS (Tables 2 and 3 ).
|
|
|
|
|
The mean ICU and hospital stays for the SSS surgeries were 2 days and 4.5 days, respectively. Postoperative care was generally simple. One patient had an atrioventricular block that required a pacemaker. No patient died.
3.4 Further reoperations
Five patients (four coarctations and one AVSD) developed a second SSS after a mean of 7.4 years (from 1 year to 19 years). Four of them had an isolated subaortic membrane as the first SSS and were treated with an isolated membranectomy without myomectomy. One patient (with a coarctation) developed a third SSS after 19 years. Eighty-four percent of these patients with two or three successive SSS initially had a coarctation of the aorta. All of these patients underwent repeat surgery. The mean number of operations per patient (including initial surgery and SSS) was 2.8.
| 4. Discussion |
|---|
|
|
|---|
In our series, 71% of SSS appeared more than 2 years after the initial surgery, with a mean interval of 4.4 years. SSS could indeed occur much later: 19 years after the initial surgery in one of our patients. All the patients were asymptomatic, whereas symptoms have been present in 23–40% of the patients in previous reports [2,5]. This discrepancy between our series and others may be due to early detection in our patients through long-term follow-up.
Low age at the initial surgery was a significant risk factor for developing SSS, whatever the initial heart defect (32 vs 74.9 months, p < 10–4). The mean age at initial surgery of patients with one, two and three SSS was 38 months, 10 months and 1 month, respectively. This association of SSS with low age at the time of initial surgery has previously been reported only with discrete subaortic stenosis [13] and ostium primum atrial septal defect [14]. According to the literature, SSS does not seem to be related to a genetic syndrome.
The pathophysiological theories for the development of subaortic stenosis without a previous heart operation (turbulence theory and geometric theory) [2,4] can be extrapolated to SSS. Anatomical elements causing turbulent abnormal flow patterns in the subaortic region were present in 93% of our patients and thus may have contributed to the development of SSS through stimulation of the endothelium. These anatomical and geometric abnormalities of the LVOT, as well as their prevalence, are summarised in Table 2. They can create turbulence and shear stress in the subaortic region [2,15,16] or may result in disproportion between the inlet and outlet dimensions of the left ventricle [17].
Subaortic stenosis without previous heart operation has been described as an acquired disease due to a pre-existing anatomical alteration in the mitral valve apparatus [18]. In our series, an increased prevalence of mitral valvular abnormalities also exists in heart defects, which then develop secondary subaortic stenosis (32% of our patients) (Table 3). An abnormal rotation of the mitral valve apparatus was identified in only one case (4%) in our series, whereas the frequency in earlier reports has varied from 83% [18] to 100% [19].
Flow limitation in the left ventricle seems to enhance the development of SSS. Indeed, blood flow in cardiac cavities causes the development of these cavities. Thus, LV flow limitation explains the occurrence of LV and aortic hypoplasia, as well as obstructive lesions in the LV, including SSS. In our series, LV flow limitation conditions were present in 90% of the cases: aortic coarctation, left superior vena cava and bicuspid aortic valve were the most frequent abnormalities. Mitral valvular defects were found in 32% of our patients. Agnoleti et al. showed that persistence of the left superior vena cava can perturb the normal development of the left ventricle and is strongly associated with obstructions of left ventricular inflow and outflow, with a significantly increased frequency of subaortic stenosis [20]. The prevalence of LSCV is 2/1000 in the general population versus 14% in our series. LSVC and bicuspid aortic valve were significantly more frequent in the cases with SSS (14% and 21%) compared with the patients without SSS operated on in our institution during the study period (4% and 6%, p < 0.05). The appearance of SSS was also significantly influenced by previous coarctation repair (p < 10–4). Aortic coarctation was initially present in 40% of all patients, in four out of five patients with two postoperative SSS and in the only patient with three postoperative SSS (p < 10–4).
Similarly, the high frequency of anatomical and flow abnormalities during foetal life in our series and the rapid appearance and progression of SSS support this theory. Indeed, some authors have suggested that a narrowed LVOT alters the blood flow patterns in the heart during early embryogenesis and leads to an accumulation of cells near the crest of the ventricular septum, which differentiates into fibromuscular tissue and obstructs the subaortic region [21].
The initial surgery involved the LV-to-aorta pathway in 39% of our cases. SSS development could be related to scar formation in the LVOT. The initial surgical procedure may contribute to the development of SSS in many ways and could explain the detection of an early pressure gradient across the LVOT in the immediate postoperative course of 7 of our cases.
The left ventricle undergoes geometric change after the Rastelli operation or intraventricular repair [22]. The sinuous shape of this tunnelisation may lead to turbulent flow beneath the aortic valve, resulting in either septal or subarterial conal muscle hypertrophy, which can lead to a left ventricular outflow tract gradient [10].
AVSD repair at the ventricular level can lead to subaortic stenosis [18]. Cleft closure can decrease the normal systolic displacement of the valve leaflets away from the septum and can narrow the subaortic area, although some authors have found no correlation with an increased risk of subaortic obstruction [12,23,24].
Previous repair of coarctation of the aorta can reveal a silent subaortic stenosis by increasing blood flow through the aorta. This can explain the detection of a significant early pressure gradient across the LVOT in the immediate postoperative course in one of our patients.
After an operation for isolated VSD or tetralogy of Fallot, LV hypertrophy and extension of the VSD patch into the LVOT can also lead to SSS formation.
As described in the literature, SSS can occur after a Fontan procedure even in the absence of previous PA banding [9]. In our series, one patient with single left ventricle (double outlet left ventricle) associated with malposition of the great arteries first had PA banding and then an atriopulmonary anastomosis. (A Damus-Kaye-Stansel operation was considered unfeasible because of an abnormal pulmonary valve.) The progression of a restrictive bulboventricular foramen led to the development of SSS 11 years later, which was treated by enlarging the foramen.
Development of SSS, therefore, seems to be related to the combination of multiple factors. This would explain why SSS is more frequent in complex heart defects (86%) than in simple ones (14%).
SSS is a serious complication that causes severe left ventricular hypertrophy and significant aortic insufficiency requiring open-heart surgery. Nevertheless, strategies to prevent SSS remain elusive. Development of such a complication is unavoidable if the subaortic area is essentially narrowed. Moreover, despite the increasing sophistication of echocardiography, the cause of altered flow patterns is sometimes unclear. However, recent data highlight how the effects of mechanical stress on endothelial cells [15] may contribute to the development of SSS, which could provide a new basis for preventive management of SSS.
This work did not compare the technical aspects of the initial surgical procedure between patients with and without SSS, and a prospective study stratified on age and heart defect type remains necessary.
| 5. Conclusion |
|---|
|
|
|---|
| Appendix A |
|---|
|
|
|---|
Dr F. Lacour-Gayet (Denver, Colorado, USA): You have not found in your patients any patient with transposition or double outlet right ventricle. It seems that the rest of the operation has been shown as a good provider of left ventricular outflow obstruction on the long term as well as DORV, because rerouting the VSD to the aorta can be associated with a subaortic stenosis.
Why do you believe in your experience that you have not observed any subaortic obstruction in transposition or double outlet right ventricle?
Dr Kalfa: Actually, we did observe secondary subaortic stenosis after LV aorta rerouting in eight cases of double outlet right ventricle and TGA. That's the fourth group of surgical procedure I presented. And we think that we had subaortic stenosis in this case because of the anatomic change in the left ventricle after LV aorta rerouting. This surgery increases LV afterload, which leads to a LV and conal septum hypertrophy, and thus to SSS.
Moreover, the tunnel can have a sinuous shape which can enhance development of SSS, too.
And finally, we can think that the tunnel will not grow, whereas the aortic valve and aortic annulus will grow. So there will be a disproportion and this can lead to SSS, too. So there are a lot of anatomical risk factors to develop a SSS after LV–aorta rerouting in double outlet right ventricle.
Dr R. Jonas (Washington, D.C., USA): What are your thoughts about unroofing a left SVC, if that is one of the risk factors? I like the way you have grouped your risk factors together, but what about eliminating a left SVC by unroofing the coronary sinus and, if necessary, connecting the left SVC through a Warden-type atriocaval anastomosis. Is that something that you think might reduce the late risk subaortic stenosis?
You told us that a left SVC is a risk factor for the development of subaortic stenosis. Presumably that is caused by the enlarged coronary sinus which reduces the orifice of the mitral valve. If you unroof the coronary sinus – and I did see a report recently, I dont know if that author is here – where they suggested unroofing the coronary sinus to enlarge the effective mitral valve orifice. I wonder if we shouldnt be doing that more frequently to prevent late subaortic stenosis?
Dr Kalfa: I think that formation of SSS depends largely on what happens during foetal life, that is to say the left heart flow limitation during foetal life. As a matter of fact, the fast appearance and the fast evolution of SSS after birth can let us suppose that what happens during foetal life plays a major role in SSS development. So, if you try to treat LVSC, you will treat something at the time of operation; but the haemodynamic effects of what happened during foetal life will be still present and this has already favoured the occurrence of SSS. That's too late!
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Edwin, M. M. Tettey, K. Entsua-Mensah, and K. Frimpong-Boateng eComment: Discrete subaortic stenosis following repair of atrioventricular septal defects Interactive CardioVascular and Thoracic Surgery, July 1, 2009; 9(1): 87 - 88. [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 |