HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Metras, D. Articles by Voisin, M. Search for Related Content PubMed PubMed Citation Articles by Metras, D. Articles by Voisin, M.

Eur J Cardiothorac Surg 1998;13:117-123
© 1998 Elsevier Science NL

Anatomic repair of corrected transposition or atrio-ventricular discordance¹

Report of 8 cases

^a Service of Cardiothoracic Surgery, La Timone Children’s Hospital, Boulevard Jean Moulin, 13385 Marseilles Cedex 05, France
^b Pediatric Cardiology Unit, Marseilles, France
^c Pediatric Cardiology Unit, Montpellier, France

Received 30 September 1997; received in revised form 3 December 1997; accepted 9 December 1997.

Corresponding author. Tel.: +33 491386676; fax: +33 491478170.

	Abstract

Top Abstract Introduction Patients and methods Results Discussion Appendix A. Conference... References

 
Background: It is well recognized that a right ventricle (RV) may eventually fail if working as systemic. This has led to arterial switch operation for transposition of the great arteries (TGA), but more recently to anatomic correction of ‘corrected transposition’ or atrio-ventricular discordance (AVD)combining an atrial switch and a ventricular outflow repair (arterial switch or Rastelli type). Methods: We report here 8 cases of such anatomic correction of AVD in patients from 3 months to 10 years. Of these patients, 6 had situs solitus, and 2 situs inversus; 7 patients had a ventricular septal defect (VSD); 5 had an increased pulmonary flow, and 3 had a decreased pulmonary flow with pulmonary stenosis (PS) or pulmonary atresia (PA). In all, 6 had discordant and 1 had concordant ventriculo-arterial (VA) relations, and 1 had double-outlet RV, 2 had an associated coarctation, and 1 had a cor triatriatum. Of the 8 patients, 5 had pulmonary artery hypertension, 3 a moderate to severe tricuspid insufficiency and 6 had undergone one or more previous operations. Correction was done between 3 months and 10 years of age, with 7 having a Senning procedure (twice left-sided) and 1 a Mustard type operation. VA outflows were repaired with: 4 arterial switches, 1 1eft ventricle to aorta rerouting, 1 right ventricular outflow tract (RVOT) repair and 2 Rastelli. A tricuspid annuloplasty was done once. Results: All patients survived the operation, but 2 patients needed a reoperation (both successful): late obstruction of pulmonary venous channel after a left-sided Senning, and residual VSD closure. Mean follow-up is 45 months (3–122). Of the 8 patients, 6 patients are in sinus rhythm, 2 have junctional rhythm. A patient with junctional rhythm and occasional tachycardia died suddenly 18 months after surgery 1 week after satisfactory evaluation. All have a normal left ventricle function as evidenced by echocardiography. Conclusions: Despite a more demanding procedure, the ‘anatomic repair’ of hearts with AVD is possible, even in infancy, with good early and mid-term result. Despite the potential late rhythmic problems of the Senning operation (somehow more frequent with AVD in our experience), it may become, with increasing experience, the procedure of choice in corrected transposition.

Key Words: Anatomic correction • Atrio-ventricular discordance • Senning procedure

	Introduction

Top Abstract Introduction Patients and methods Results Discussion Appendix A. Conference... References

 
It is now well recognized that a right ventricle (RV) functioning as systemic may eventually fail [1]. It has been one of the arguments to promote arterial switch in the classic transposition of the great arteries (TGA) [2]. Similarly, in corrected transposition or atrio-ventricular discordance (AVD), despite some satisfactory long-term results after conventional surgery, the mid-term or late failure of the anatomic RV can lead to death or transplantation and has promoted the ‘anatomic correction’, with atrial switch and arterial repair (switch or Rastelli depending upon the anatomic condition of the pulmonary outflow tract) [3] [4] to incorporate the anatomic left ventricle in the systemic circulation. In recent years, more experience has been reported on the early and mid-term results of this approach [5] [6] [7] [8].

The purpose of this paper is to report our experience in 8 cases operated upon with this approach, including 3 patients under 6 months, and to discuss results and indications.

	Patients and methods

Top Abstract Introduction Patients and methods Results Discussion Appendix A. Conference... References

 
From 1985 to 1997, 8 patients with AVD underwent an anatomic correction in our center. The anatomic features and associated malformations of 5 patients with an increased pulmonary flow and 3 patients with a decreased pulmonary flow (associated with pulmonary stenosis (PS) or pulmonary atresia (PA)) are listed in Table 1Table 2.

All the others patients had the various classic types of AVD with VA discordance (VAD), 2 patients having the situs inversus (I,D,D) type of segmental approach [9], the others the classic (S,L,L) type [9] of corrected transposition.

A total of 5 patients had undergone one or more previous palliative or ‘curative’ procedure (patient 2 underwent VSD closure, banding release and tricuspid plasty at age 1 year, 2 years before double switch).

All patients underwent the ‘anatomic’ correction between the ages of 3 months and 10 years (mean 40 months) under cardiopulmonary by-pass (CPB) with deep hypothermia and occasional periods of circulatory arrest (Table 3Table 4).

The atrial switch was done by a Senning procedure in 7 cases. It was done under a period of hypothermic circulatory arrest in 4 patients and under hypothermic CPB in 3 patients. The hypothermic circulatory arrest was chosen in cases of low weight and expected difficulties of canulation of both cavae. In 2 of those cases, with situs inversus and levocardia (patients 1 and 6) we have carried out a left-sided Senning, the surgeon standing on the left of the patient. It was felt that due to this particular situation, there was a lack of right atrial tissue. Thus, the neo-pulmonary venous channel was done with an augmentation patch.

In 1 case (patient 4) with associated cor triatriatum, a Mustard procedure was used, since a Senning was deemed unfeasible.

The VSD were approached through a right ventriculotomy in 5 patients and the VSD closed (patients 1 and 2) or a LV to aorta rerouting carried out through this approach with a large dacron patch inserted with interrupted pledgeted sutures and/or continuous suture (patients 6, 7 and 8). In patients 3 and 5 the VSD was approached through the right atrium with the classical recommendations [11]. In patient 4, the ventricular septum was intact.

The mitral cleft was closed with simple interrupted mattress sutures and the attempt of tricuspid valvuloplasty done with a simple pledgeted commissural stitch.

The right ventricular outflow tract (RVOT) repair was done in 2 cases with a Rastelli-type RV-PA connection with a 19 mm diameter cryopreserved homograft (one aortic and one pulmonary). In both cases, in spite of trying to avoid crossing the midline with the conduit to avoid anterior compression, it was felt necessary to leave the chest opened and to carry out a delayed sternal closure. The other case allowed a simpler RVOT repair with a monocusp homograft patch.

	Results

Top Abstract Introduction Patients and methods Results Discussion Appendix A. Conference... References

 
There was no operative or hospital death. The procedures were long, mean CPB time 234 min, and mean aortic cross-clamping time 136 min, with 3 patients being left with the chest opened, with a delayed sternal closure after 5 days. The patients had a somehow difficult postoperative course, with an ICU stay of 8–38 days, mean 14 days and a total hospital stay of 10–66 days, mean 29 days. Only 1 patient had an early (3 weeks) successful reoperation for residual VSD, but this prolonged the ICU stay to a total of 38 days.

Patient 1, with situs inversus, having undergone a left-sided Senning, presented with a severe late pulmonary venous pathway stenosis. He underwent a subsequent repair with a patch augmentation through a left thoracotomy under hypothermic circulatory arrest 1 year after initial operation. He is now 10 years after surgery in excellent condition.

All patients were carefully followed by cardiologists with EKG, Holter, echo-doppler and in 2 patients, by catheterization. The follow-up time is from 3 to 122 months, mean 45 months. The EKG showed a persistent and stable sinus rhythm in 6 patients, and junctional rhythm or occasional junctional tachycardia in 2 cases. These 2 patients were put under antiarrhythmic medication.

Of those 2 patients, 1 (patient 2) died suddenly 18 months after repair, 1 week after an extremely satisfactory evaluation of the cardiac function. A sudden arrhythmia was suspected. All other patients are in class 1, without medication (except the antiarrhythmic treatment in 1).

The last echo-doppler evaluation showed in all: (1) normal systemic LV function, (2) valvular anomalies (either AV or outlet), (3) normal RV function with systolic RV pressures estimated between 22 and 50 mmHg (mean 38 mmHg), and (4) the RV-PA homograft patch or conduit functioning properly. LV angiograms, when carried out, showed a normal function ( Fig. 1 ). Angiograms carried out in the right atrium and RV showed normally-functioning Senning construction, RV function, and Rastelli repair ( Fig. 2 ).

View larger version (101K): [in this window] [in a new window]   Fig. 1. Postoperative angiogram in the anatomic LV. It was a case of (I,D,D) with left-sided left ventricle. LV systolic function is normal (top, systole; bottom, diastole).

View larger version (115K): [in this window] [in a new window]   Fig. 2. Postoperative angiogram in the right-sided heart. It was a classical case of (S,L,L) with VSD and PA (top, the caval pathway after the Senning; bottom, RV injection and RV-PA homograft).

	Discussion

Top Abstract Introduction Patients and methods Results Discussion Appendix A. Conference... References

In simple TGAs, the late results after atrial switch (either Mustard or Senning) also show a progressive incidence of tricuspid insufficiency and RV failure [5] [8].

These late problems of a systemic RV and tricuspid valve have fully justified the development of arterial switch for simple TGV and support the concept of ‘anatomic correction’ for corrected transposition, with a potential best hope for longevity [5]. Since the first report of 2 cases undergoing a combined Mustard and Rastelli [3], more cases have been reported [4] [5] [8] and more recently the atrial switch and arterial switch has been done in the absence of PS or PA [6] [7]. Excellent reports have documented the satisfactory hemodynamics and normal LV function after mid-term follow-up.

The atrial switch
The Mustard has been reported as being more frequently used than the Senning procedure. The latter has been our preferred approach (7/8), the only case of Mustard being in a case with associated cor triatriatum where the Senning was not thought to be possible. However, in cases with situs inversus and levocardia or situs solitus with dextrocardia, the right atrial wall is smaller than usual and has led to augmentation patch of the pulmonary venous pathway. Patient 1 developed a severe retraction of xenopericardium used for this purpose and needed a reoperation 9 months after this left-sided Senning with autologous atrial patch plasty, and showed an excellent long-term (10 years) result. If an augmentation patch is necessary PTFE may be the best material.

We have experienced 2 patients with a late postoperative junctional rhythm or episodes of junctional tachycardia after the Senning; a much higher proportion than in our series of Senning in classic transposition. Due to the anterior location of the AV node, we hypothesized that the conductive tissue pathway may be different and carry a higher risk of junctional dysrhythmias but there is no clear answer to this (R. Anderson, personal communication, 1996). It must be mentioned, however, that spontaneous junctional rhythm has been reported to occur in AVD in 4/17 patients [20].

The VSD closure
The VSD closure or reconstruction of the LV to aorta pathway through a right ventriculotomy avoided any case of heart block in our experience, and, as reported in other papers, this approach has decreased considerably the occurrence of this complication and even avoided it completely [8].

Arterial switch
No special technical comment has been done in the 7 cases reported in the literature [6] [7], but it has been done in general with the classic Lecompte maneuver. In 3 of our cases, due to the respective position of the vessels, it was chosen not to do this maneuver and the pulmonary bifurcation was left posterior, with no evidence of obstruction.

The coronaries buttons, as reported by others [6] [7], were implanted without problem as it was supposed by the knowledge of coronary distribution in AVD [21].

Mid-term and late ventricular function
As already reported [6], the newly systemic anatomic left ventricle exhibits a normal contractility as shown by echocardiogram and angiogram in 2 patients in whom a postoperative cardiac catheterization was done. Together with the absence of complete AV blocks and the satisfactory RV function, these factors will probably result in a better long-term functional class and hopefully, survival.

The tricuspid insufficiency
The tricuspid insufficiency is a serious problem in AVD, and its repair has been reportedly done without great success [19]. In 1 of our patients it was attempted together with a classical type of repair (patient 2), done 1 year before the double switch. It was unsuccessful, and in the postoperative period elevated PA and LV pressures persisted, allowing the double-switch procedure, since the LV was thus already ‘prepared’ to sustain systemic conditions. Tricuspid insufficiency regressed postoperatively in all patients as already reported by others [7] [8] and is certainly an excellent indication for the anatomic repair.

Potential of the procedure
This relatively new approach appears therefore as potentially optimal in corrected transposition with RV systemic dysfunction and tricuspid insufficiency. Long-term results are still needed before it can be proposed as the best procedure for the vast majority of patients. In addition, in the cases associated with PS undergoing a conduit placement between the anatomic RV and the PAs, it must be stressed that a conduit replacement may be a difficult procedure according to the location of this conduit.

Are there currently recognized contraindications to this approach?
Some anatomic conditions may be a contraindication.

1. Very abnormal mitral valve (it has to be evaluated carefully) is certainly a contraindication, but a simple cleft is not.
   
2. Abnormal location of a VSD rendering difficult a rerouting from LV to AO in case of PS or PA. This has already been a matter of discussion [22], and it is possible that an inlet septum or trabeculated VSD, with in addition aberrant tricuspid chordae or some straddling may contraindicate this option. In cases where the VSD needs to be enlarged, one has to be clearly aware of the course of the conduction bundle [23].
   

Is age of the patient a problem?
We think that in cases of VSD and PS or PA, a shunt should be carried out when needed and the anatomic correction deferred until a reasonable size Rastelli repair can be done, although some have carried out a REV-type correction without interposition of a conduit, and this could change the age of correction to a younger one [5].

In cases with VSD and normal pulmonary outflow we think that a PA banding can be carried out if necessary shortly after birth, and complete correction deferred until 3 months of age. We have done this double-switch as a primary operation in 2 babies of 4 and 6 months, and we think it is feasible even earlier. As far as we know this procedure has not been reported at this age in the literature.

Even if long-term results are necessary to delineate more precisely the indications, it must be stressed than in all the 26 cases reported so far in the literature, and in our 8 cases, there was no complete heart block reported and the hospital mortality was much lower than in all the experiences of classical approach reported, so that the short and mid-term results are already better.

We therefore conclude that despite a more demanding procedure and limited recent experience, the double-switch procedure in AVD already achieves better early results than the ‘classical’ approach. It is likely that it will provide better long-term results. There are few contraindications, and it can be carried out in infancy if necessary. We consider it as an improvement on the previously employed procedures.

	Footnotes

 
Presented at the 11th Annual Meeting of the European Association for Cardio-thoracic Surgery, Copenhagen, Denmark, 28 September–1 October, 1997.

	Appendix A. Conference discussion

Top Abstract Introduction Patients and methods Results Discussion Appendix A. Conference... References

 
Dr Z. Al-Halees (Riyadh, Saudi Arabia): With these results, what would your approach be for, say, a patient who presents with corrected transposition and congenital complete heart block with no intracardiac lesion?

Dr D. Metras: At birth, you mean, with congenital heart block?

Dr Z. Al-Halees: With congenital heart block. The patient comes to you with impaired RV function (systemic ventricular dysfunction) and complete heart block with no intracardiac anomaly. What would you recommend for that patient? In other words, would you advise doing an anatomic correction on that child which of course will require prior preparation of the left ventricle by a pulmonary artery band?

Dr D. Metras: Well, I hadn’t thought of that kind of case before preparing this paper, but I think it’s a difficult problem. If you have no intracardiac disease, the LV pressure should be low, also the anatomic LV. So I think it’s a bit of a difficult problem. If you want to transform it, you have to prepare the LV plus put a pacemaker in the patient. So I think it needs more reflection than a quick answer in this meeting, but I’m not sure it should be incorporated into that kind of program.

Dr T. Ebels (Groningen, The Netherlands): Do you think there is a case for prophylactic pacemaker implantation in these children because of the natural tendency to develop AV block?

Dr D. Metras: Well, a quick answer is no because I guess that we can wait until a spontaneous block occurs and then there is the possibility of implanting an intravenous pacemaker. If you implant myocardial leads, one knows that you will have to one day reoperate on these patients anyway because of fibrosis around the electrode, and implanting a temporary pacing lead on the myocardium, on the epicardium is not a definitive solution, I would say.

Dr J. Stark (London, UK): I have just a very brief comment on the previous question. We recently had Professor Imai from Tokyo present a very large series, I think it was over 35 patients, with this procedure at our course in London. It seems that although in the natural history of corrected transposition late development of heart block has been described, Professor Imai didn’t report any late complete heart block. With this procedure, one approaches the VSD through the RV, which is an advantage, as the conduction is on the morphological LV side of the septum. So I would be reluctant to implant a prophylactic pacemaker because of all the problems that are related to pacemakers in children.

Dr C. Brizard (Melbourne, Australia): In the reconstruction of the RVOT, have you ever done a reconstruction without a conduit, I mean direct PA to RV anastomosis?

Dr D. Metras: One of our cases had a simple RVOT reconstruction with a monocusp outflow patch. It was a kind of double-outlet RV. But we haven’t gone to this direct anastomosis between the RV and the PA. That has been reported, I think, by Imai in his series. I think it’s a very good approach.

	References

Top Abstract Introduction Patients and methods Results Discussion Appendix A. Conference... References

 

1. Siebenmann R, VonSegesser L, Schneider J, Senning A, Turina M Late failure of systemic ventricle after atrial correction for transposition of great arteries. Eur J Cardio-thorac Surg 1989;3:119-124.[Abstract]
2. Jatene AD, Fontes VF, Paulista PP, Souza LCB, Neger F, Galantier M, Sousa JE Anatomic correction of transposition of the great vessels. J Thorac Cardiovasc Surg 1976;72:364.[Abstract]
3. Ilbawi MN, DeLeon SY, Backer CI, Duffy E, Muster AJ, Zales VR, Paul MH, Idriss FS An alternative approach to the surgical management of physiologically corrected transposition with ventricular septal defect and pulmonary stenosis or atresia. J Thorac Cardiovasc Surg 1990;100:410-415.[Abstract]
4. DiDonato RM, Troconis CJ, Marino B, Carotti A, Iorio FS, Rossi E, Marceletti C Combined Mustard and Rastelli operations; an alternative approach for repair of associated anomalies in congenitally corrected transposition in situs inversus. J Thorac Cardiovasc Surg 1992;104:1246-1248.[Abstract]
5. Imai Y, Sawatari K, Hoshino S, Ishihara K, Nakazawa M, Momma K Ventricular function after anatomic repair in patients with avioventricular discordance. Cardiovasc Surg 1994;107:1272-1283.
6. Yamagishi M, Imai Y, Hoshino S, Ishihara K, Koh Y, Nagatsu M, Shinoka T, Koide M Anatomic correction of atrioventricular discordance. J Thorac Cardiovasc Surg 1993;105:1067-1076.[Abstract]
7. Stümper O, Wright JGC, De Giovanni JV, Silove ED, Sethia B, Brawn WJ Combined atrial and arterial switch procedure for congenital corrected transposition with ventricular septal defect. Br Heart J 1995;73:479-482.[Abstract/Free Full Text]
8. Delius RE, Stark J Combined Rastelli and atrial switch procedure: anatomic and physiologic correction of discordant atrioventricular connection associated with ventricular septal defect and left ventricular outflow tract obstruction. Eur J Cardio-thorac Surg 1996;10:551-555.[Abstract]
9. VanPraagh R, VanPraagh S Isolated ventricular inversion: a consideration on the morphogenesis, definition and diagnosis of non transposed and transposed great arteries. Am J Cardiol 1966;17:395.[Medline]
10. Lecompte Y, Zannini L, Hanzan E, et al. Anatomic correction of transposition of the great arteries; a new technique without use of a prosthetic conduit. J Thorac Cardiovasc Surg 1981;82:629-631.[Abstract]
11. DeLeval M, Bastos P, Stark J, Taylor JF, Macartney FG, Anderson R Surgical technique to reduce risks of heart block following closure of ventricular septal defect in atrioventricular discordance. J Thorac Cardiovasc Surg 1979;78:515-526.[Abstract]
12. Szufladowicz M, Horvath P, DeLeval M, Elliott M, Wyse R, Stark J Intracardiac repair of lesions associated with atrioventricular discordance. Eur J Cardio-thorac Surg 1995;10:443-448.[Abstract]
13. Termignon JL, Leca F, Vouhé PR, Vernant F, Bical OM, Lecompte Y, Neveux JY ‘Classic’ repair of congenitally corrected transposition and ventricular septal defect. Ann Thorac Surg 1996;62:199-206.[Abstract/Free Full Text]
14. Colli AM, Freed MD, Lock JE, Jonas RA, Mayer JE, Castaneda A Outcome of patients with corrected transposition (Abstract). Cardiol Young 1993;3(Suppl 1):152.
15. Fox LS, Kirklin JW, Pacifico AD, Waldo AL, Bargeron LM Intracardiac repair of cardiac malformations with atrioventricular discordance. Circulation 1976;54:123-128.[Abstract/Free Full Text]
16. Marcelleti C, Maloney J, Ritter D, Danielson G, McGoon D, Wallace R Corrected transposition and ventricular septal defect. Surgical experience. Ann Surg 1980;191:751-759.[Medline]
17. Williams WG, Suri R, Shindo G, Freedom RM, Morch JE, Trusler GA Repair of major intracardiac anomalies associated with atrioventricular discordance. Ann Thorac Surg 1980;31:527-531.[Abstract]
18. Dimas AP, Moodie DS, Sterba R, Gill CC Long-term function of the morphologic right ventricle in adult patients with corrected transposition of the great arteries. Am Heart J 1989;118:526-530.[Medline]
19. Horvath P, Szufladowicz M, DeLeval M, Elliott M, Stark J Tricuspid valve abnormalities in patients with atrioventricular discordance: surgical implications. Ann Thorac Surg 1994;57:941-945.[Abstract]
20. Daliento L, Corrado D, Buja G, John N, Nava A, Thiene G Rhythm and conduction disturbances in isolated, congenitally corrected transposition of the great arteries. Am J Cardiol 1986;58:314-318.[Medline]
21. McKay R, Anderson RH, Smith A The coronary arteries in hearts with discordant atrioventricular connections. J Thorac Cardiovasc Surg 1996;111:988-997.[Abstract/Free Full Text]
22. DeLeval M Discussion of: ventricular function after anatomic repair in patients with atrioventricular discordance. J Thorac Cardiovasc Surg 1996;107:1283.
23. Anderson RH, Becker AE, Arnold R, Wilkinson JL The conducting tissues in congenitally corrected transposition. Circulation 1974;50:911-923.[Abstract/Free Full Text]

This article has been cited by other articles:

				M. Ly, E. Belli, B. Leobon, C. Kortas, O. E. Grollmuss, D. Piot, C. Planche, and A. Serraf Results of the double switch operation for congenitally corrected transposition of the great arteries Eur. J. Cardiothorac. Surg., May 1, 2009; 35(5): 879 - 884. [Abstract] [Full Text] [PDF]

				A. A. Alghamdi, B. W. McCrindle, and G. S. Van Arsdell Physiologic Versus Anatomic Repair of Congenitally Corrected Transposition of the Great Arteries: Meta-Analysis of Individual Patient Data Ann. Thorac. Surg., April 1, 2006; 81(4): 1529 - 1535. [Abstract] [Full Text] [PDF]

				M. N. Ilbawi, C. B. Ocampo, B. S. Allen, M. J. Barth, D. A. Roberson, P. Chiemmongkoltip, and R. A. Arcilla Intermediate results of the anatomic repair for congenitally corrected transposition Ann. Thorac. Surg., February 1, 2002; 73(2): 594 - 600. [Abstract] [Full Text] [PDF]

				S. H. Daebritz, A. R. Tiete, J. S. Sachweh, W. Engelhardt, G. von Bernuth, and B. J. Messmer Systemic right ventricular failure after atrial switch operation: midterm results of conversion into an arterial switch Ann. Thorac. Surg., April 1, 2001; 71(4): 1255 - 1259. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Metras, D. Articles by Voisin, M. Search for Related Content PubMed PubMed Citation Articles by Metras, D. Articles by Voisin, M.