|
|
||||||||
Eur J Cardiothorac Surg 2002;21:1042-1048
© 2002 Elsevier Science NL
a Department of Pediatric Cardiac Surgery, Eppendorf University Hospital (UKE), Martinistrasse 52, D-20246 Hamburg, Germany
b Marie Lannelongue Hospital, Paris, France
Received 21 September 2001; received in revised form 8 January 2002; accepted 11 February 2002.
* Corresponding author. Fax: +49-40-42803-4931
e-mail: flg{at}uke.uni-hamburg.de
| Abstract |
|---|
|
|
|---|
Key Words: Coronary heart disease Pediatric cardiac surgery Anatomopathology Double outlet right ventricle Arterial switch
| 1. Introduction |
|---|
|
|
|---|
The aim of this article is to present another surgical option with VSD rerouting to the pulmonary infundibulum and arterial switch, that seems almost always feasible in DORVncVSD in the absence of valvar pulmonary stenosis.
| 2. DORV anatomy |
|---|
|
|
|---|
The definition of DORV remains controversial. If there is unanimity around the 50% rule given by Anderson and colleagues [5,6], other definitions [79] remain valuable. The presence of a subaortic conus, as stated by Van Praagh et al. [8], remains the major landmark of this anomaly. Lecompte has proposed to change the term DORV for malposition of the great arteries [9].
2.2. DORVncVSD definition
The term non-committed VSD was introduced by Lev and Barati [7]. Van Praagh et al. [8] used the term non-committed VSD and mentioned the considerable distance between the VSD and the aorta. These terminologies have been quite vague and a precise definition was needed. Barbero-Marcial et al. [13] define it as a VSD either muscular inlet without perimembranous connection or a VSD of atrioventricular (AV) canal type. Belli et al. [14] define it as a VSD distant from the arterial valves by a distance superior to an aortic diameter; this definition includes perimembranous VSD.
The definition (Figs. 13) that we propose includes:
|
|
|
|
The coronary distribution (Fig. 4) shows an abnormal course with double looping [17], with a posterior location of the main trunk and a right coronary crossing anteriorly the aorta. Frequently, there is a single coronary right ostium giving the left main trunk and the RCA.
2.4. Associated lesions
Among the associated lesions, the subaortic obstruction is more a permanent lesion than an association. Almost constantly, one or several abnormal subaortic muscular bands is found that connect the right part of the aortic infundibular wall to the conal septum (Fig. 5B)
. A real DORVncVSD with AVSD seems quite rare; usually the AVSD is a type C of Rastelli with a large subaortic component of the VSD [18] that is therefore committed to the aorta. Nevertheless, AVSD with a VSD not extended below the aorta has been described in two patients with DORV by Pacifico et al. [19] and in one patient by Mee and colleagues [20].
|
Moderate LV hypoplasia allows biventricular repair. True single functioning ventricles are frequent and excluded from the study.
| 3. Patients |
|---|
|
|
|---|
All patients had a bilateral infundibulum, with a large persistent subaortic conus, D malposition of the aorta, side-by-side vessels and double loop coronary patterns. There were four patients with single coronary coming from the right ostium (Fig. 4). The VSD was perimembranous with inlet or trabecular extension. It was restrictive with an interventricular gradient greater than 20 mm in four patients. Subaortic obstruction was constant. The VSD was located far away from both the arterial valves by a distance greater than the aortic diameter (Fig. 5).
The associated lesions included: four ostium secundum, two non-connected LSVC, two restrictive mitral valve with normal subvalvular apparatus, two moderate LV hypoplasia with LV end-diastolic volume at 16 and 18 ml/m2, one muscular infundibular stenosis, two tricuspid valve straddling of type A and one left juxtaposition of appendages.
| 4. Methods |
|---|
|
|
|---|
The operation was performed on full-flow cardiopulmonary bypass at moderate hypothermia with repeated crystalloid cardioplegia. The mean cross clamping time was 122±21 min, ranging from 95 to 185 min. It was conducted through a triple approach; through the tricuspid valve, the pulmonary infundibulum and the aortic annulus. The RV incision was performed on the low part of the pulmonary infundibulum. A double exploration through the tricuspid valve and the infundibulum allows location of the VSD; it was always found in contact with the tricuspid valve and very close to the pulmonary infundibulum. The distance between the tricuspid valve and the conal septum was reduced in six patients. In five patients, tricuspid chordae attached to the conal septum or standing next to the infundibulum ostium were blocking the area of LV to aorta tunellization. The right ventricle was hypertrophied in all patients with previous PA banding. The VSD was enlarged superiorly (Fig. 3), towards the trabecula septo-marginalis, principally to meet the diameter of the aortic annulus. The almost permanent subaortic obstruction was released, by resection of multiple muscular bands located in the aortic conus reached through the tricuspid valve and/or the aortic annulus. The VSD was tunnelized to the pulmonary ostium using a Dacron or Gore-Tex patch; the VSD patch was sewn in contact with the tricuspid valve and the tunnel was not involved with the conal tricuspid chordae that were located more superiorly. This VSD prosthetic patch was short, measuring less than 2030 mm long. It was large and had a circular shape. The arterial switch technique had to manage a double coronary artery looping with side-by-side vessels. The Lecompte maneuver was always performed and included a right transfer of the PA bifurcation in order to prevent a compression of the left coronary button. In two patients with small LV size, a 4-mm interatrial fenestration was left in place.
| 5. Results |
|---|
|
|
|---|
| 6. Discussion |
|---|
|
|
|---|
The embryology is becoming a more precise science with the enormous interest brought by the use of embryonic cells. The introduction of immunohistochemistry in cardiac embryology research has allowed precise statements [1]. Using monoclonal antibodies targeted against different antigens expressed by the embryological tissues (GIN2), it is possible to follow the migration of cardiomyocytes along the embryological development [15]. These recent publications confirm that the primitive right ventricle is a DORV containing the conotruncus. The migration of the aorta toward the mitral valve is a complex procedure, named wedging, that allows the aortic side of the conotruncus to nestle between the tricuspid and the mitral valve [1]. These elements explain partly why the right ventricular structures are usually normal in DORV, this primitive RV bearing abnormally an additional structure: the entire conotruncus, including the aortic infundibulum, the infundibulum septum and the pulmonary infundibulum.
In DORVncVSD, the major landmark is the entire location of the great vessels on the right ventricle. To follow the Anderson rule, one can say that the great vessels are 200% on the RV. The entire conotruncus is on the RV including an aortic and a pulmonary conus separated by a conal septum that has nothing to do with the ventricular septation. The other forms of DORV contain only partially the great vessels; the Taussig Bing contains only 50% of the PA and the DORV-Fallot 50% of the aorta. These forms should better be named TGA or Fallot, respectively. It would be probably simpler to reserve the name DORV for patients with 200% vessels on the right ventricle, implying a double conus.
The definition of DORVncVSD is also controversial. The considerable distance [8] between the VSD and the two arterial valves needs to be quantified. Belli et al. [14] have proposed that in DORVncVSD, this distance should be greater than the aortic annulus diameter. If this definition is appropriate, it seems more precise to add two other conditions: 200% vessels on the right ventricle and a double conus.
The VSD, in DORVncVSD, is located below the trabecula septomarginalis (Fig. 3) as pointed out by Freedom and Yoo [16]. This allows to differentiate the DORVncVSD from the Taussig Bing hearts were the VSD is located above the TSM and therefore, close to the pulmonary annulus.
The definition given by Barbero-Marcial et al. [13] relies only on the location of the VSD that should be muscular, located in the inlet or trabecular septum without perimembranous extension; excluding all connections with the tricuspid valve. In fact, the VSD is usually perimembranous as mentioned by Becker and Anderson [5] and Freedom and Yoo [16], and as it was found in all of the patients in this series.
The considerable distance between the VSD and the great vessels is not due to an abnormal location of the VSD but is related to a very abnormal location of the aorta that is standing at the roof of the right ventricle. Therefore, it is more a remote aorta than a remote VSD that is a cause of DORVncVSD.
It is noticeable that in DORVncVSD, the right ventricular structures are in place; namely, the tricuspid valve, the infundibular septum and the pulmonary artery (Figs. 13). The disorder is created by the additional presence of the entire aortic conus that has remained in place as in primitive hearts. The stability of the right ventricular structures in DORVncVSD explains why the VSD is usually located next to the ostium infundibuli, making possible the new technique described.
Finally, the real anatomical anomaly of the DORV is a malposition of the aorta, the right structures being in place. Lecompte and Sidi [9] have proposed the term malposition of the great vessels that is partly correct. The term malposition of the aorta seems more appropriate and describes the real anomaly that is an abnormal remainder of the aortic conus.
6.2. Surgery
The intraventricular tunnel rerouting of the VSD to the aorta is the operation of reference for DORVncVSD.
In the series of 21 patients reported by Belli et al. [14], the overall mortality is 10% without late death. In two patients, the repair was abandoned during the operation because of conal tricuspid chordae. There have been six reoperations for late subaortic stenosis (29%) and no late deaths.
In the series of 18 patients reported by Barbero-Marcial et al. [13], the early mortality is 11% with no late subaortic stenosis but there were three late deaths, including two sudden deaths. An elegant multiple patch technique to prevent subaortic obstruction is presented. According to the authors, the presence of abnormal tricuspid chordae was not a contraindication and could be resolved by either reimplantation of the tricuspid chordae on the patch or sacrifice of one leaflet and bicuspidization of the tricuspid valve. A reduced pulmonary-tricuspid distance [23] was also not a contraindication, and the pulmonary valve was sacrificed and a valved conduit inserted.
The real risk of this technique, in addition to the potential impairment of the tricuspid and pulmonary valves, is the mandatory presence of a very long patch into the RV. The risk of subaortic obstruction is evaluated at 30% by Belli et al. [14,21,22] and seems to be correlated with the length of the patch. Furthermore, this long patch creates an important akinetic area in the LVOT obstruction that can impair the LV function.
The other option is the Fontan operation, preferred by many centers [3,4]. In comparison with the potential risks of the VSD to aorta rerouting, a Fontan, undertaken here on two ventricles, remains a satisfactory option. In cases of type AV valve straddling, overriding or severe LV hypoplasia, it is the only option.
The rerouting to the PA followed by arterial switch is our preferred solution.
The short length of the tunnel is a key issue and the longer the tunnel, the greater the risk of late subaortic obstruction. Our early experience allows confirmation that when the VSD is very far way from the aorta, it is almost always close to the ostium infundibuli (Figs. 1 and 2). The first advantage of this technique is to construct a shorter tunnel (Figs. 3 and 6) , reducing the risk of subaortic obstruction. The second advantage is that the rerouting to the PA is independent of the tricuspid valve to PA distance and that it is not involved with the tricuspid chordae (Fig. 6). Interestingly, in our series the VSD was always perimembranous. Barbero-Marcial et al. [13] exclude these forms in the definition of DORVncVSD. Further studies are required to clarify this point; our opinion is that an intact perimembranous septum is extremely rare in DORV for embryological reasons and that the VSD is not committed in DORVncVSD, mainly because of the remote position of the aorta. The arterial switch requires translation of coronary arteries with double loop coronary courses, side-by-side vessels and frequent single coronary. These types of transfer are now well controlled [17].They require, nevertheless, an adapted technique, the pulmonary bifurcation needing to be transferred towards the right PA [17] to prevent the compression of the left-sided coronary trunk, in cases of associated Lecompte maneuver. The arterial switch is therefore complex and may require a second learning curve in comparison to a normal coronary course switch.
|
| 7. Conclusion |
|---|
|
|
|---|
The intraventricular tunnel connecting the VSD to the aorta is the repair of reference. This technique seems nevertheless quite aggressive and is associated with an important risk of late subaortic obstruction when the VSD is very distant.
Our early experience, in a series of ten patients, with rerouting of the VSD to the ostium infundibuli followed by arterial switch, is satisfactory with one non-cardiac death and no late subaortic obstruction. It is our operation of choice when the aorta is very distant. This technique creates a smaller tunnel, is not involved by the presence of conal tricuspid chordae, and is independent of the distance between the pulmonary and the tricuspid valve.
| Footnotes |
|---|
| Appendix A. Conference discussion |
|---|
|
|
|---|
Dr Lacour-Gayet: Well, when you don't perform the LeCompte maneuver, there is a better relationship of the vessels. But you can never resolve the coronary relocation. When relocating the right coronary button posteriorly, you take a risk that this button is squeezed between the two vessels, particularly in this condition with frequent single coronary artery, where the left trunk runs behind the pulmonary artery. My experience has been to remain quite faithful to the LeCompte maneuver. Now, I know that others have reached a satisfactory result without the LeCompte maneuver.
Dr C. Tchervenkov (Montreal, QC, Canada): I don't intend to get into the discussion we had yesterday about complete AV canal and DORV. I would strictly like to focus on the fact that this technique is extremely innovative and offers an excellent solution for that subgroup of patients with non-committed VSD and DORV. However, Francois, you very well know that there are patients that have significant either valvular or subvalvular pulmonary stenosis, and some patients have VSD that is completely confined to the inlet such as in complete AV canal and DORV. Have you encountered any such patients in your series, and how would you deal with these patients?
Dr Lacour-Gayet: When the patient has an RVOT stenosis, if the stenosis is located at the site of the pulmonary valve, this is a contraindication for this technique. If it is located at the site of the infundibulum, this can be resected or patched and it is not a contraindication. If the operation of biventricular repair implies a sacrifice of the tricuspid and the pulmonary valve and construction of a very large tunnel inside the right ventricle, I think that the option of univentricular repair is more satisfactory.
As to your last question, regarding the AV canal and DORV, our experience is not in favor of a frequent association between DORV and the AV canal; these patients in our experience have been for the great majority of Fallot's tetralogy with AV canal, with a subaortic component of the VSD.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. K. Dharmapuram, S. Sundararaghavan, S. K. Swain, N. Ramdoss, P. Reddy, V. Okwulehie, R. Agarwal, and K. S. Murthy Morphosurgical correlation of outcomes in complete double outlet right ventricle Interactive CardioVascular and Thoracic Surgery, October 1, 2006; 5(5): 566 - 569. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. H. Artrip, H. Sauer, D. N. Campbell, M. B. Mitchell, C. Haun, M. C. Almodovar, V. Hraska, and F. Lacour-Gayet Biventricular repair in double outlet right ventricle: surgical results based on the STS-EACTS International Nomenclature classification. Eur. J. Cardiothorac. Surg., April 1, 2006; 29(4): 545 - 550. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Caffarena and J. M. Gomez-Ullate DORV with non-committed VSD and Taussig-Bing hearts. Controversial anatomic entities Eur. J. Cardiothorac. Surg., January 1, 2003; 23(1): 136 - 136. [Full Text] [PDF] |
||||
![]() |
R. H. Anderson Double outlet right ventricle Eur. J. Cardiothorac. Surg., November 1, 2002; 22(5): 853 - 853. [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 |