The atrial switch operation was the treatment of choice

Transcription

1 Systemic Right Ventricular Failure After Atrial Switch Operation: Midterm Results of Conversion Into an Arterial Switch Sabine H. Daebritz, MD, Andreas R. Tiete, MD, Jörg S. Sachweh, MD, Wolfgang Engelhardt, MD, Götz von Bernuth, MD, and Bruno J. Messmer, MD Departments of Thoracic and Cardiovascular Surgery and Pediatric Cardiology, University Hospital RWTH, Aachen, Germany Background. Failure of the systemic right ventricle after atrial switch operation can be treated by conversion into an arterial switch operation. Methods. Four patients, age 38 to 59 months, presented with right ventricular failure after Senning operation and ventricular septal defect closure. One patient had elevated left ventricular pressure; in the other three patients the left ventricle was retrained to a left ventricular/right ventricular pressure ratio of 0.8 or greater by pulmonary artery banding in 12 to 24 months. Results. Postoperative course after arterial switch operation was prolonged, but clinical condition was good at discharge. Fractional shortening ranged from 20% to 28%. Trace-to-moderate aortic regurgitation was present; only 1 patient had preserved sinus rhythm. After a mean follow-up of 43.5 months 1 patient had died due to left ventricular dysfunction. The survivors are in New York Heart Association functional class I to II. Fractional shortening has improved (29% to 37%); aortic regurgitation has not increased. No patient has undisturbed sinus rhythm. Conclusions. Conversion of an atrial into an arterial switch is an alternative to cardiac transplantation in childhood. However, the procedure is demanding. Longterm morbidity is caused by rhythm disturbances. Aortic valve performance and left ventricular function require close observation. (Ann Thorac Surg 2001;71:1255 9) 2001 by The Society of Thoracic Surgeons The atrial switch operation was the treatment of choice for patients with transposition of the great arteries (TGA) before introduction of the arterial switch operation (). The major disadvantage of an atrial switch procedure is that the anatomically right ventricle is left as systemic ventricle. Long-term follow-up shows development of right ventricular (RV) failure in up to 7% to 10% of patients per 10 years with a higher incidence in patients with additional ventricular septal defect (VSD) [1 4]. Concomitant tricuspid regurgitation due to stretching of the originally noncircular tricuspid ring, organic damage as a result of VSD patching, or failure of systolic leaflet coaptation [5] is frequent. The treatment options are tricuspid valve replacement, orthotopic heart transplantation, and secondary with takedown of the atrial switch. Replacement of the systemic atrioventricular valve, however, has not shown satisfying results [6]. Heart transplantation carries the problems of rejection, coronary artery disease, and immunosuppressive therapy [7 9]. A conversion of the atrial into an arterial switch can be performed in selected patients [9 12]. However, in most of these patients this procedure cannot be done in one step. The left ventricle has to be retrained by pulmonary artery banding to be Accepted for publication Nov 6, Address reprint requests to Dr Daebritz, Department of Cardiac Surgery, LMU, University Hospital Grosshadern, Marchioninistr 15, D Munich, Germany; sabine.daebritz@hch.med.uni-muenchen.de. able to sustain the systemic circulation. We describe our experience with this procedure in 4 patients. Material and Methods Four male patients were referred to our institution with echocardiographic RV dysfunction 36 to 66 months after Senning operation. Underlying anatomy was a TGA with VSD. Correction had been performed in other institutions at the age of 36 to 89 days. findings, echocardiographic and catheterization data are listed in Table 1. All 4 patients underwent a conversion into an arterial switch. Follow-up was obtained by contacting the pediatric cardiologists and reviewing the clinical and echocardiographic findings. Surgical Technique for Pulmonary Artery Banding All banding procedures were performed through median sternotomy without cardiopulmonary bypass. In 2 patients, adjustable banding devices were used in the first banding step. Unfortunately, both devices broke, therefore they had to be replaced by a Teflon band [13]. The band was tightened to a gradient of 30 mm Hg to 40 mm Hg in each banding step. It was loosened in case of a remarkable increase of CVP, a decrease in oxygen saturation, or the development of rhythm disturbances. We did not insert intracardiac lines; transesophageal echocardiography was not available. The 2001 by The Society of Thoracic Surgeons /01/$20.00 Published by Elsevier Science Inc PII S (00)02662-X

2 1256 DAEBRITZ ET AL Ann Thorac Surg SYSTEMIC RV FAILURE AFTER 2001;71: Table 1. Data Before Pulmonary Artery Banding Age Condition LV/RV Pressure Ratio LVPW Thickness (mm) TR Rhythm 1 70 NYHA II 0.66 n.m. Moderate SR 2 68 NYHA I Severe SR 3 61 NYHA I Mild SR 4 59 NYHA II Moderate AV-nodal AV atrioventricular; LV left ventricle; LVPW left ventricular posterior wall; n.m. not measured; NYHA New York Heart Association functional class; RV right ventricle; SR sinus rhythm; TR tricuspid regurgitation. patients were observed on the intensive care unit for 1 day. Surgical Technique for the Conversion of the Atrial Into the Arterial Switch The conversion was performed under standard cardiopulmonary bypass techniques. The atrial baffle was taken down and the atrial septum was reconstructed with autologous tissue (3 patients) or with a Dacron patch (1 patient). The was performed in the standard fashion with a LeCompte maneuver in all 4 patients. Transesophageal echocardiography and left atrial pressure measurements were not performed. Results Two patients received an adjustable banding device, which had to be replaced due to breaking after 3 and 9 months. Intensive care stay was 1 day after each banding step. The banding data are listed in Table 2. Before the all 3 patients with a pulmonary arterial band had a left ventricular/right ventricular (LV/RV) pressure ratio at cardiac catheterization of 0.8 or greater, a posterior wall thickness of 6 to 7 mm, and 2 patients showed a decrease of tricuspid regurgitation. Systemic RV function had improved in echocardiography, but was not quantified. Three patients were in sinus and one in intermittent atrioventricular nodal rhythm. The was performed after a mean banding period of 19 months at a mean age of 78 months. Cardiopulmonary bypass time ranged from 141 to 258 minutes, crossclamp from 100 to 152 minutes; 2 patients had circulatory arrest of 33 and 59 minutes. All patients came off cardiopulmonary bypass with high inotropic support and stable hemodynamics. Intensive care stay was prolonged due to recurrent episodes of congestive heart failure and a Candida sepsis in 1 patient. Postoperative data are listed in Table 3. During a mean follow-up of 43.5 months 1 patient presented with a sudden onset of severely impaired LV function after 18 months with a fractional shortening of 8%. He died 19 months after before heart transplantation. The electrocardiogram did not reveal any signs of myocardial ischemia. He had had a period of digoxin intoxication with a peak blood level of 6.3. No further information was available. In particular, myocarditis was not ruled out. Permission for autopsy was not given. The other 3 patients are in good clinical condition and have satisfactory or good LV function (Fig 1). and echocardiographic data at follow-up are listed in Table 4. Comment The atrial switch procedure has been almost completely abandoned in the treatment of TGA due to severe rhythm disturbances and development of systemic RV failure with tricuspid valve regurgitation [1]. The morphologic right ventricle has not been designed to work as a systemic pumping chamber. It is a one coronary ventricle with only one conduction system radiation and without two well-balanced papillary muscles. The multiple papillary muscles of the tricuspid valve, which partly arise from the septal wall, are pulling the TV leaflets apart in the presence of RV dilatation. This applies particularly if the tricuspid orifice becomes circular as it does in the systemic circulation and if the septum is shifted to the left. Tricuspid valve replacement, although useful in congenitally corrected TGA with Ebstein-like anomaly of the TV, has not shown long-term efficacy after atrial switch [6, 12]. Cardiac transplantation is associated with the complications of immunosuppression and limited long- Table 2. Pulmonary Artery Banding Data Banding Period of Banding Procedures Banding Gradient Before (mm Hg) LV/RV Pressure Ratio Before TR Before Moderate 2 24 a 2 a Moderate Mild 4 21 a 3 a Mild a Use of an adjustable banding device. arterial switch operation; LV left ventricle; RV right ventricle; TR tricuspid regurgitation.

3 Ann Thorac Surg DAEBRITZ ET AL 2001;71: SYSTEMIC RV FAILURE AFTER 1257 Table 3. Data After Arterial Switch Operation Age at ICU Stay Inotropic Support Enoximone Ventilation Hospital FS Stay Rhythm b AR b TR b (%) Condition b a AV-nodal Trace No 20 Good SR Mild Mild 25 Excellent AF No No 22 Good AV-nodal Mod. No 29 Good a Prolonged due to candida infection. b At discharge. AF atrial flutter; AR aortic regurgitation; arterial switch operation; FS fractional shortening; SR sinus rhythm; TR tricuspid regurgitation. term outcome especially in children. Conversion of an atrial into an arterial switch was first described by Mee [10]. The procedure has been successfully performed in patients with RV failure after atrial switch [5, 10 12] and in congenitally corrected TGA [14 20] as double switch. In patients with elevated LV pressure the operation can be performed in one step without concerns about age. However, the majority of patients require previous pulmonary arterial banding to retrain the left ventricle. A neonate s left ventricle can be trained in 1 week [21], but the capacity and the rapidity of LV hypertrophy decreases with age. In neonates, increased workload leads to angiogenesis, myocyte hyperplasia, and hypertrophy. At a later age the response is hypertrophy alone induced by protooncogene expression, shift of myosine isoenzymes and other molecular mechanisms [22]. Thus far there is no information available concerning the time period that is necessary to train a left ventricle at different ages. Thus, the operative risk of a secondary or a double switch operation is age related; the age limit seems to be somewhere in the teens or low twenties [12, 14]. In addition, sudden imposition of a pressure overload to the myocardium by pulmonary banding may cause myocardial damage. Differences in LV function after primary compared to with previous pulmonary arterial banding support this theory [23, 24]. We observed severe deterioration of LV function in the follow-up in 1 patient who died waiting for transplantation. This has not been reported in the literature and may be an exceptional development not directly related to the operative procedure. However, it may have been induced by a too tight banding step causing myocardial fibrosis or an adult type of myocardial response to pulmonary artery banding. Technically, the banding procedure is challenging and carries the risk of overbanding with congestive heart failure. Urgent loosening and tightening of the band early after implantation have been described [12]. An adjustable banding device would be advantageous, but failed in our experience [13]. Intraoperative transesophageal echocardiography, observation on the intensive care unit and administration of inotropes are recommended to aid pulmonary artery banding adjustment, and to support the left ventricle. Although there is a uniform recommendation for the LV/RV pressure ratio before of 0.7 or greater, there are no guidelines for the length of the banding period. Cochrane and colleagues [12] reported 12 patients with a mean age of 7 years with retraining of the left ventricle in a mean time period of 26 months. In contrast to this, Chang and associates [11] described 4 patients of the same age with a mean banding period of 2 months. Fig 1. Fractional shortening (FS) after secondary atrial switch operation over time. 4 died due to unclear deterioration of left ventricular function. In the other 3 patients fractional shortening is normal or close to normal.

4 1258 DAEBRITZ ET AL Ann Thorac Surg SYSTEMIC RV FAILURE AFTER 2001;71: Table 4. Data at Last Follow-up Time After FS (%) AR TR PM Implantation After Rhythm After PM Implantation Condition Medication Trace No 13 PM rhythm (VVI) Good Captopril Mild No 5 SR Good Digoxine, sotalol No No... AF Good Digoxine, amiodarone 4 a 19 8 Mod. Mod. 6 AF Bad Digoxine, amiodarone a Last data before death. AF atrial flutter; AR aortic regurgitation; arterial switch operation; dig. digoxin; FS fractional shortening; mod. moderate; PM pacemaker; SR sinus rhythm; TR tricuspid regurgitation. Helvind and colleagues [14] reported 17 patients with retraining of left ventricle in one to five banding steps in 11.2 months. Their study, however, consisted mainly of patients with congenitally corrected TGA of younger age. In our small series the mean banding period was 19 months and was prolonged in 2 patients due to breaking of an adjustable banding device. Our assessment of LV function was restricted to measurement of posterior wall thickness and the banding gradient. More sophisticated methods, such as calculation of LV mass and wall stress, would be useful. The value of magnetic resonance imaging in this setting is not yet defined [14]. We observed improvement of tricuspid regurgitation and systemic RV function after pulmonary artery banding. It is of note, that this is due to a septal shift and does not reflect true recovery of RV function. Our results confirm the high incidence of rhythm disturbances observed in the literature after a conversion procedure or a double switch [12, 14]. One-third of the patients with an atrial switch have lost sinus rhythm after 10 years and atrial flutter is a risk factor for late death [4]. The conversion operation with its additional suture lines in the sensitive area of the right atrium may add to this problem [25]. Cardiac transplantation offers a major advantage. Aortic valve incompetence after secondary, although mostly trace or mild and so far not progressive, was noted in 3 of our patients. It was also a frequent finding in other series [11], necessitating aortic valve replacement in some patients [12]. Pulmonary artery banding may be a contributing factor by valve distortion, dilatation of the pulmonary annulus, or scarring. In conclusion, the conversion of an atrial into an arterial switch is a challenging procedure, which can be performed with good early results in selected patients. However, follow-up reveals substantial morbidity regarding LV function, aortic valve incompetence, and particularly rhythm disturbances. Therefore, patients undergoing a conversion operation should be selected carefully with special regard to age and preexisting rhythm disturbances. Conversion pathway has to be started early after systemic RV deterioration to have time to retrain gradually the left ventricle. If LV function deteriorates after secondary, early transplantation is recommended because there is no hope of ventricular recovery. References 1. Turina M, Siebermann R, Nussbaumer P, Senning A. Longterm outlook after atrial correction of transposition of the great arteries. J Thorac Cardiovasc Surg 1988;95: Turina M, Siebermann R, von Segesser L, Schoenbeck M, Senning A. Late functional deterioration after atrial correction for transposition of the great arteries. Circulation 1989; 80(3 Pt 1):I Merlo M, de Tommasi SM, Brunelli F, et al. Long-term results after atrial correction of complete transposition of the great arteries. Ann Thorac Surg 1991;51: Sarkar D, Bull C, Yates R, Wright D, et al. Comparisons of long-term outcomes of atrial repair of simple transposition with implications for a late arterial switch strategy. Circulation 1999 Nov 9;100(19 Suppl):II Van Son JAM, Reddy M, Silvermann NH, Hanley FL. Regression of tricuspid regurgitation after two-stage arterial switch operation for failing systemic ventricle after atrial inversion operation. J Thorac Cardiovasc Surg 1996;111: Van Son JAM, Danieson GK, Huhta JC, et al. Late results of systemic atrioventricular valve replacement in corrected transposition. J Thorac Cardiovasc Surg 1995;109: Kanter KR, Tam VK, Vinsent RN, Cuadrado AR, Raviele AA, Berg AM. Current results with pediatric heart transplantation. Ann Thorac Surg 1999;68: Shumway SJ. Heart transplantation on the pediatric agegroup and for end-stage congenital heart defects. Mayo Clin Proc 1998;73: Bouchart F, Gundry SR, Bailey LL. Infant heart transplantation Clin Transpl 1991; Mee RBB. Severe right ventricular failure after Mustard or Senning operation. Two-stage repair: pulmonary artery banding and switch. J Thorac Cardiovasc Surg 1986;92: Chang AC, Wernovsky G, Wessel DL, et al. Surgical management of late right ventricular failure after Mustard or Senning repair. Circulation 1992;86(Suppl 2):II Cochrane AD, Karl TR, Mee RBB. Staged conversion to arterial switch for late failure of the systemic right ventricle. Ann Thorac Surg 1993;56: Dabritz S, Sachweh J, Tiete A, Engelhardt W, von Bernuth G, Messmer BJ. Experience with an adjustable pulmonary artery banding device in two cases: initial success midterm failure. Thorac Cardiovasc Surg 1999;47: Helvind MH, McCarthy JF, Imamura M, et al. Ventriculoarterial discordance: switching the morphologically left ventricle into the systemic circulation after 3 months of age. Eur J Cardiothorac Surg 1998;14: Imai Y. Double-switch operation for congenitally corrected transposition. Adv Card Surg 1997;9: Yagihara T, Kishimoto H, Isobe F, et al. Double switch operation in cardiac anomalies with atrioventricular and

5 Ann Thorac Surg DAEBRITZ ET AL 2001;71: SYSTEMIC RV FAILURE AFTER 1259 ventriculoarterial discordance. J Thorac Cardiovasc Surg 1994;107: Karl TR, Weintraub RG, Brizard CP, Cochrane AD, Mee RBB. Senning plus arterial switch operation for discordant (congenitally corrected) transposition. Ann Thorac Surg 1997;64: Metras D, Kreitmann B, Fraisse A, et al. Anatomic repair of corrected transposition or atrioventricular discordance: report of 8 cases. Eur J Cardiothorac Surg 1998;13: Reddy VM, McElhinney DB, Silvermann NH, Hanley FL. The double switch procedure for anatomical repair of congenitally corrected transposition of the great arteries in infants and children. Eur Heart J 1997;18: Stumper O, Wright JG, 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: Jonas RA, Giglia TM, Sanders SP, et al. Rapid, two-stage arterial switch for transposition of the great arteries with intact ventricular septum beyond the neonatal period. Circulation 1989;80(3 Pt 1):I Di Donato RM, Fujii AM, Jonas RA, Castaneda AR. Agedependent ventricular response to pressure overload. Considerations for the arterial switch operation. J Thorac Cardiovasc Surg 1992;104: Boutin C, Wernovsky G, Sanders SP, Jonas RA, Castaneda AR, Colan SD. Rapid two-stage arterial switch operation. Evaluation of left ventricular systolic mechanics late after an acute pressure overload stimulus in infancy. Circulation 1994;90: Colan SD, Boutin C, Castaneda AR, Wernovsky G. Status of the left ventricle after arterial switch operation for transposition of the great arteries. Hemodynamic and echocardiographic evaluation. J Thorac Cardiovasc Surg 1995;109: Vetter VL, Tanner CS, Horowitz LN. Electrophysiologic consequences of the Mustard repair of D-transposition of the great arteries. J Am Coll Cardiol 1987;10: INVITED COMMENTARY The potential for physiologic growth of the myocardium (hyperplasia and angiogenesis) is lost very early in life. Myocyte response to pressure overload is considered the result of intermingling phenomena of cell replication and hypertrophy on one side, and myocyte necrosis and apoptosis on the other side, a process known as plasticity of the myocardium. Noteworthy, a few adult cardiac myocytes seem capable to reexpress fetal proteins and proliferate following sudden pressure overload. However, after the first few weeks of age, cell hypertrophy remains the dominant type of pressure-induced myocyte response. The deterioration of diastolic function during the retraining process causes myocyte death, which, in turn, propitiates left ventricular fibrosis and dilation, aggravating the magnitude of stress on the remaining viable cells. Fiber stretching in dilated myocardium may, in fact, promote myocyte apoptosis. In contrast to the adult, pressure-induced ventricular augmentation in the neonate involves increased angiogenesis, as in fetal life. Prolonged overload, however, causes decreased capillary density and coronary vascular reserve. These vascular changes probably reflect a switch from myocyte hyperplasia to hypertrophy as the primary mode of cardiac enlargement but may also indicate vascular abnormalities, eg, inhibition of normal vascular growth or an increase in the wall/lumen ratio because of vascular remodelling. The extrapolation of in vitro observations to the in vivo state requires considerable caution, especially because experimental models of aortic banding determine supravalvular (ie, supracoronary) aortic stenosis, whereas clinical application of pulmonary banding in transposition is equivalent to creating aortic valvar stenosis. Nevertheless, the currently used criteria for selection of patients as well as timing and modalities of left ventricular reconditioning are still arbitrary. As of today, the result of a two-stage conversion from atrial to arterial switch, even if undertaken at a relatively young age, eg, in the four patients of Daebritz and colleagues, cannot be defined a truly anatomic repair. It relies on a structurally and functionally subnormal neosystemic pump and may be complicated by progressive atrial arrhythmias and neoaortic valve incompetence. I believe that efforts to boost up this alternative to heart transplantation for a difficult group of patients should continue. Maybe, a protocol of serial myocardial biopsies could help in directing therapeutic approaches. Maybe, strategies to genetically engineer myocyte and vascular replication could be developed. At the moment, however, we cannot rule out that a majority of these patients will eventually need a heart transplantation. Roberto M. Di Donato, MD Department of Pediatric Cardiac Surgery Bambino Gesù Hospital Piazza S. Onofrio, Rome, Italy didonato@opbg.net by The Society of Thoracic Surgeons /01/$20.00 Published by Elsevier Science Inc PII S (01)