Patients with a functionally single ventricle, unrestricted

Transcription

1 Mid-Term Results for Double Inlet Left Ventricle and Similar Morphologies: Timing of Damus-Kaye- Stansel Andrew J. B. Clarke, MBBS, FRACS, Shingo Kasahara, MD, David R. Andrews, MBBS FRACS, Stephen G. Cooper, MBChB, FRACP, Ian A. Nicholson, MBBS, FRACS, Richard B. Chard, MBBS, FRACS, Graham R. Nunn, MBBS, FRACS, and David S. Winlaw, MBBS, MD The Children s Hospital at Westmead, Adolph Basser Cardiac Institute, Westmead NSW, Australia Background. Patients with double inlet left ventricle/ltransposition and similar morphologies have their systemic outflow traverse a bulboventricular foramen (BVF), which has a propensity to narrow over time. A Norwood procedure may be performed as the initial palliation. We prefer aortic arch repair and pulmonary artery banding, delaying Damus-Kaye-Stansel (DKS) or BVF resection until the second palliation. The aims of this study were to compare our results with those reported for Norwood strategy and examine the development of systemic outflow obstruction. Methods. Retrospective study of patients with double inlet left ventricle, L-TGA or similar morphology presenting between 1990 and Follow-up with clinical assessment, echocardiography and catheter studies. Results. Twenty-five patients had initial palliation with pulmonary artery banding with repair of any associated arch obstruction. Twelve patients had DKS performed as part of their second stage procedure, and 3 had DKS performed later for recurrent stenosis after prior enlargement of BVF. Six patients had BVF resection without later restenosis and 4 patients did not develop BVF stenosis. There was one early death (4%) and two late (8%). Fontan completion was achieved in 20 of the 22 survivors. There were no cases of DKS obstruction, no pulmonary valve had more than mild regurgitation. Conclusions. Our approach achieves low operative mortality and morbidity and compares favorably with reported results for Norwood palliation. The significant rate of systemic outflow obstruction in those who did not undergo DKS at the second stage confirms the utility of early DKS in children with this morphology. (Ann Thorac Surg 2004;78:650 7) 2004 by The Society of Thoracic Surgeons Patients with a functionally single ventricle, unrestricted pulmonary blood flow and arch obstruction require staged palliation culminating in the Fontan circulation. Many patients in this group have double inlet left ventricle (DILV) and l-transposition (l-tga) morphology. Frequently, systemic outflow must traverse a bulboventricular foramen (BVF) that has a propensity to narrow over time. The initial operation for children with this complex must address arch obstruction and limit pulmonary blood flow. This may be achieved by the creation of a shunt-dependent circulation and pulmonary arteryaortic anastomosis with arch augmentation; similar to the Norwood procedure for hypoplastic left heart syndrome. Although some centers achieve very good results for first stage Norwood palliation, there remains significant associated mortality and morbidity [1, 2]. An alternative approach, preferred by our group, is to perform arch repair and pulmonary artery banding as the Accepted for publication Jan 7, Address reprint requests to Dr Winlaw, Cardiac Surgery Research, The Children s Hospital at Westmead, Locked Bag 4001, Westmead NSW 2145, Australia; davidw@chw.edu.au. first palliation. A Damus-Kaye-Stansel (DKS) connection may be created at the second operation at the same time as a superior cavopulmonary connection, addressing any actual or potential systemic ventricular outflow tract obstruction. We performed a retrospective study of all patients with DILV, l-tga or similar morphology, treated by our institution over a 10-year period from 1990 to The aim of the study was to determine the medium-term outcome of patients and compare this to reported results for initial Norwood palliation. A secondary aim was to examine the prevalence and development of systemic outflow obstruction, to evaluate our practice with regard to timing of the DKS connection. Material and Methods We undertook a retrospective review of patient records, echo and catheter studies. Specific attention was directed at finding evidence of DKS obstruction, ventriculoarterial valve incompetence, and ventricular function. Patients were clinically assessed for functional status. Arch repair and pulmonary artery banding (PAB) was performed by thoracotomy, otherwise PAB alone or in 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg CLARKE ET AL 2004;78:650 7 TIMING OF DAMUS-KAYE-STANSEL CONNECTION 651 combination with septectomy was performed through a sternotomy. Coarctation repair was performed as an extended end-to-end procedure in all patients. A 5-mm Dacron band (DuPont Pharmaceuticals, Wilmington, DE) was placed by passing the band through the transverse sinus then between aorta and pulmonary artery. The optimal position for the band was determined by identifying the origin of the coronary arteries from the aorta. By placing the PA band immediately above this point, a proximal band is obtained, usually positioned at or just above the level of the sinotubular junction. The band is held in position by the undisturbed adventitia and other tissue connecting the great arteries, preventing distal migration and branch PA distortion. No additional sutures are used to secure the band to the pulmonary artery. The tightness of the band is determined by saturations and hemodynamics and was guided by the Toronto formula [3], taking into consideration that the left lung is partly collapsed during the banding procedure. The DKS procedure was performed using conventional continuous flow cardiopulmonary bypass and moderate hypothermia with cardioplegic arrest. Thirteen of 15 patients had the DKS created by transection of the main pulmonary artery, usually at the site of the previous pulmonary artery band, and implantation of the proximal MPA stump into the side of the aorta, utilizing a triangular flap of aortic wall to create the posterior part of the connection. The remaining defect was closed with glutaraldehyde preserved pericardium, or Gore-Tex patch (W.L. Gore & Associates, Flagstaff, AZ) augmentation (Fig 1). The other 2 patients had both great vessels transected at the level of the pulmonary artery band. The proximal aorta and pulmonary artery were sewn together side to side for approximately one-third to one-half of their circumference and then the augmented distal aorta sewn to this double outlet ventricle (Fig 2). With this technique, in most patients, the connection can be created without any patch material, but if there appears to be excessive tension, then a patch of autologous pericardium is used. The latter technique is simple and leaves the aortic sinuses intact, and is now our preferred approach. Three of 15 patients required DKS construction after completion of Fontan and ligation of the main pulmonary artery. The valves were examined at the time of operation and found to have normal appearance. When widening of the BVF was performed, access was gained through a short ventriculotomy in the outflow chamber immediately beneath the aortic valve, avoiding adjacent coronary arteries. Resection was mostly performed on the free wall aspect of the BVF toward the apex, to minimize risk of injury to the conduction system. Superior cavopulmonary connections were created using standard techniques, using continuous flow cardiopulmonary bypass and moderate hypothermia. Lateral tunnel Fontan was performed early in the series (6 patients) with extracardiac Fontan conduits used during the later part of the series (14 patients). Extracardiac Fontan conduit is now our preferred technique. Patient follow-up data are reported as median (range) Fig 1. Aortic flap technique of creating a Damus-Kaye-Stansel connection. (Top) The main pulmonary artery is transected at the site of the previous pulmonary artery band. The proximal main pulmonary artery stump is then implanted into the side of the aorta, utilizing a triangular flap of aortic wall to create the posterior part of the connection. (Bottom) The remaining defect was closed with glutaraldehyde preserved pericardium, or Gore-Tex patch augmentation. and calculated using Microsoft Excel (Microsoft, Inc., Redmond, WA). Freedom from DKS connection was analyzed using SPSS for Windows (Chicago, IL). Results Patient Cohort Between 1990 and 2000, 25 patients underwent initial palliation for the complex of functionally single ventricle, ventriculoarterial discordance, and systemic outflow through a BVF, with or without arch obstruction (Table 1). Twenty-one patients had situs solitus, double inlet left ventricle (DILV), and l-transposed great arteries. Four had situs solitus, tricuspid atresia, and d-transposed great arteries. All patients had the systemic outflow traversing a BVF; no patient had significant outflow obstruction at birth. Follow-up was complete for 20 of 22 survivors (91%). The median follow-up from first palliation was 101

3 652 CLARKE ET AL Ann Thorac Surg TIMING OF DAMUS-KAYE-STANSEL CONNECTION 2004;78:650 7 Fig 2. Double-barrel technique of creating a Damus-Kaye-Stansel connection. (Left) Both great vessels are transected at the site of the previous pulmonary artery band. (Middle) The proximal aorta and pulmonary artery were sewn together side to side for approximately one-third to onehalf their circumference and (right) then the augmented distal aorta sewn to this double outlet ventricle. months ( months). Two patients were lost to follow-up after moving abroad. Both had completed Fontan circulation. One had a DKS connection, the other, who had no evidence of systemic outflow obstruction, did not have a DKS. Surgical Strategies: Outcomes Initial palliation with PAB was performed in association with other procedures including coarctation repair, atrial septectomy and interrupted arch repair (Table 1). The median age at initial palliation was 6 days (range 0 to 48 days). Fifteen of 25 patients had a DKS procedure at some point in their palliation (Fig 3). Twelve of 15 had a DKS performed as part of their second stage procedure (early DKS group) at a median age of 14 months old (2.4 to 26.1 months old) with a median time of 13.3 months (2.3 to 25.9 months) after initial palliation. Three of 15 had DKS performed late, after enlargement of BVF or resection of LV outflow, to address recurrent obstruction (late DKS group; Fig 3). Ten of 25 patients did not undergo a DKS procedure at any point in their palliation (Fig 4). Four patients did not develop BVF narrowing and progressed to Fontan circulation without obstruction ever developing. Six patients had BVF resections without requirement for later DKS, however 1 patient underwent a second BVF/LV outflow stenosis resection (Fig 5). Four patients underwent systemic to pulmonary artery shunting instead of a superior cavopulmonary connection as second stage palliation. Two patients had elevated pulmonary arterial pressures related to lung disease of Table 1. Initial Palliation Functionally single ventricle, transposed great arteries, systemic outflow via BVF (n 25) BVF bulboventricular foramen; 12 PAB Aortic Coarctation ( 10) / Arch interruption ( 2) 9 PAB alone 4 PAB Atrial septectomy PAB pulmonary artery band. prematurity. Both underwent DKS and central shunt creation at the second stage and one has subsequently undergoing superior cavopulmonary connection and is awaiting completion of the Fontan connection. The other has persisting pulmonary hypertension making him unsuitable for venous shunting. He has had a further modified Blalock-Taussig shunt. One patient was lost to follow-up abroad for 13 months (subsequent mortality, patient 1, Table 2). One infant initially palliated with an arch repair and pulmonary artery band required augmentation of pulmonary blood flow but was not thought suitable for progression to cavopulmonary connection at that age (2 months old, 3.2 kg; subsequent mortality, patient 3 [Table 2]). Fontan completion was achieved in 20 of 22 survivors. The median time to Fontan completion was 41.8 months (29.6 to 68.1 months) for the whole group, 38.2 months (29.6 to 52.6 months) in the early DKS group, and 62.7 months (44.6 to 68.1 months) in the late DKS group. Of the survivors that have not progressed to Fontan, one is scheduled for completion surgery, the other is not suitable because of pulmonary hypertension (shunt dependent circulation and receives home oxygen therapy). Before Fontan completion, the following hemodynamic data were obtained: cardiac index mean 5.2 L min 1 m 2 (range 3.7 to 8.2), left ventricular end diastolic pressure mean 4.2 (range 1 to 10). Mortality Data There was one early (30 day) death (4%) and two late deaths (8%) giving a total mortality of 12% (see Table 2 for details). Of the two late deaths, one was in the early DKS group and the other had had a BVF/systemic outflow tract resection. The early death had not had DKS or BVF/systemic outflow tract resection. Morbidity Four of 12 patients (33%) who underwent neonatal arch repair needed further arch surgery. One patient underwent reoperation 23 days after initial repair. The remainder required arch augmentation at the same time as

4 Ann Thorac Surg CLARKE ET AL 2004;78:650 7 TIMING OF DAMUS-KAYE-STANSEL CONNECTION 653 Fig 3. Palliation for DILV, TGA, or similar morphology that involved creation of DKS connection. (BDG bidirectional Glenn shunt; BVF bulboventricular foramen; DILV double inlet left ventricle; DKS Damus-Kaye-Stansel connection; LVOTO left ventricular outflow obstruction; PAB pulmonary artery band; TGA transposition of the great arteries.) subsequent palliations. Augmentation was achieved by patching with autologous pericardium briefly treated with glutaraldehyde. This necessitated a period of deep hypothermic circulatory arrest (core temperature 20 C for 20 minutes [16 23 minutes]). Four of 25 patients (16%) required pulmonary artery repair at second stage palliation attributable to the PAB. Three of these were due to distal migration of the PAB and the repair was performed using autologous glutaraldehyde preserved pericardium patch (2) or Gore-Tex patch (1). The other injury was due to the PAB cutting through the main pulmonary artery and associated with a false aneurysm. This was repair with a pulmonary artery z-arterioplasty and did not require patching (subsequent early mortality, patient 1 [Table 2]). Four pacemakers were required for the development of heart block, at times remote from surgical intervention. No episodes of heart block occurred in patients having BVF resection. There was one late stroke 2 months after DKS and superior cavopulmonary connection from a presumed paradoxical embolus. This child has normal ventricular and valve function, but has moderate residual neurologic impairment. One patient suffered cerebral hemorrhage before the initial palliation and suffers cerebral palsy, but is suitable for and is awaiting Fontan completion. There were no other cerebral complications noted. Development of Systemic Outflow Obstruction Nineteen of 25 patients developed LV outflow obstruction or BVF restriction between the first and second palliation (76%). Two further patients had a small BVF noted on echo before second stage palliation and had DKS performed in anticipation of developing restriction. Twelve patients had DKS procedure at second stage (early group). Nine patients had LV outflow tract resection or BVF enlargement. Four patients did not require DKS, LV outflow tract resection or BVF enlargement. Four patients who had LV outflow tract resections or BVF enlargement, subsequently required DKS connection [3] or repeat LV outflow resection/bvf enlargement for recurrent stenosis [1]. Quality of DKS Connection Current echocardiograph data were available for 13 of 15 of those who underwent DKS connection. There were no cases of DKS obstruction. There were no cases of moderate or severe pulmonary regurgitation (PR). PR was absent in 8 patients, trivial in 3 and mild in 2. The three patients who underwent late resurrection of the pulmonary valve after previous MPA ligation had normal pulmonary valve function. Two of 15 patients did not have current echocardiograph data due to migration abroad, however follow-up echocardiographs, at a median 5 months post completion

5 654 CLARKE ET AL Ann Thorac Surg TIMING OF DAMUS-KAYE-STANSEL CONNECTION 2004;78:650 7 Fig 4. Palliations for DILV, TGA, or similar morphology that did not involve creation of DKS connection. (BDG bidirectional Glenn shunt; BVF bulboventricular foramen; DILV double inlet left ventricle; DKS Damus-Kaye-Stansel connection; LVOTO left ventricular outflow obstruction; PAB pulmonary artery band; TGA transposition of the great arteries.) Fontan (2 to 6 months) revealed no PR, no evidence of DKS obstruction and normal ventricular function. Functional Limitations Excluding the neurologic events (2) discussed above, the deaths (3) and the patient who is unsuitable for cavopulmonary connection, there was follow-up of 17 of the remaining 19 patients. Fifteen patients have no apparent restriction in physical activities although no formal testing was carried out. Two patients had mild to moderate restriction in physical activity (New York Heart Association class II equivalent) but are otherwise well. that occurs following pulmonary artery banding is said to accelerate BVF narrowing and cause systemic outflow obstruction [8]. However, it is acknowledged that the Comment A total cavopulmonary connection (Fontan circulation) is the long-term goal in palliation for those with a functionally single ventricle. Particular to the group we describe, the systemic outflow must traverse a bulboventricular ventricular foramen on the way to the aorta. Narrowing at this site frequently develops, causing systemic outflow obstruction. Staged palliation involving pulmonary artery banding for the complex of DILV/l-TGA and similar morphologies is not favored by groups with good Norwood results [4, 5], although some such groups still prefer initial palliation by banding [6, 7]. The ventricular hypertrophy Fig 5. Freedom from Damus-Kaye-Stansel connection (DKS) according to age. Data from all 25 patients are included.

6 Ann Thorac Surg CLARKE ET AL 2004;78:650 7 TIMING OF DAMUS-KAYE-STANSEL CONNECTION 655 Table 2. Mortality Patient Age Initial palliation Circumstances Cause of death 1 Tricuspid Atresia, d-tga 16 mo PAB/Atrial septectomy 2 DILV, 1-TGA 16 mo PAB / Coarctation repair 3 Triscuspid atresia, d-tga 11 mo PAB / Coarctation repair Lost to followup for 13 mo. On investigation, PAB cut through main pulmonary artery causing false aneurysm and unrestricted pulmonary blood flow High pulmonary vascular resistance evident on cardiac catheterisation. Operation: division of MPA and repair false aneurysm/creation of central shunt/bvf - LVOTO resection Age 6 months - superior cavopulmonary connection / BVF - LVOTO resection Satisfactory early ventricular function, but progressive postoperative ventricular dysfunction of unknown aetiology. Listed for cardiac transplant Inadequate pulmonary blood flow at 2 months of age. Due to age and small size (3.2kg) and recurrent respiratory tract infections he was not felt suitable for cavopulmonary connection. Underwent DKS / atrial septectomy / central shunt Slow recovery, recurrent respiratory tract infections. Cardiac arrest 6 h post surgery with excessive pulmonary flow and poor systemic perfusion Died from cardiac failure awaiting transplant, six months after second stage surgery. Died 8 months post surgery from severe bronchiolitis. BVF bulboventricular foramen; DILV double inlet left ventricle; DKS Damus-Kaye-Stansel connection; LVOTO left ventricular outflow obstruction; MPA main pulmonary artery; PAB pulmonary artery band; d-tga dextro-transposition of the great arteries; l-tga levo-transposition of the great arteries. natural history of the morphologic lesion is for systemic outflow obstruction to develop, partly due to turbulent flow, even without banding. Optimal placement of the band may be more difficult than expected and pulmonary artery distortion is an unwelcome complication, particularly with a view to eventual cavopulmonary connection. Adequacy of arch repair is also a consideration that may influence surgical strategy. Taking into consideration these factors, many groups have chosen to perform a Norwood procedure, addressing arch obstruction and systemic ventricular outflow obstruction at the same time. Nevertheless, this strategy involves a significant period of cardiopulmonary bypass and a variable period of circulatory arrest with potential for cerebral injury [9, 10]. For non-hlhs indications, the Norwood procedure is associated with significant early morbidity and mortality (8% to 19%), with further attrition awaiting second stage surgery in the context of a shunt dependent circulation [1, 4, 11]. Recoarctation may occur in up to 30% of patients [12, 13], and in the traditional Norwood circulation the ventricle is volume loaded. Conventionally held views that the DILV/l-TGA group may be better Fontan candidates in the longer term may be untrue [14]. Our group has not performed Norwood palliation for this group of patients, preferring to perform arch repair and pulmonary artery banding as initial palliation. The raw operative mortality for the first stage (1/25, 4%) and overall (3/25, 12%) in this cohort would suggest that it is at least as safe and possibly better than published reports for first stage Norwood surgery [1, 4, 11]. The two cannot be directly compared because the pathology mix in reported series of Norwood surgery on nonhypoplastic left heart patients is different to this series. The high rate of achieving a Fontan circulation (20/22 survivors) also suggests that our approach is safe and effective. In the uncommon situation where DILV/l-TGA coexist with extreme arch hypoplasia, Norwood palliation is clearly the only acceptable pathway. Somewhat surprisingly, we observed a 30% rate of recoarctation requiring reoperation, similar to that in reported series for Norwood palliation [12, 13]. There were also 4 patients (4/25 to 16%) of pulmonary arterial distortion requiring repair at the second stage, this did not influence the likelihood of achieving a Fontan circulation. Similarly, left pulmonary artery narrowing is also a common sequel of the Norwood procedure, because of the bulk of the DKS connection and asymmetric PA growth from right-sided shunts. In these respects the adequacy of initial arch repair and pulmonary artery distortion approach seems superior. Significant pulmonary valvar incompetence and obstruction of the DKS connection were not features in our operative results. Other than the techniques we describe, we reestablished flow through a previously ligated main pulmonary artery on three occasions as previously reported [15]. There have been no failures of this strategy to date.

7 656 CLARKE ET AL Ann Thorac Surg TIMING OF DAMUS-KAYE-STANSEL CONNECTION 2004;78:650 7 Patients with pulmonary problems, including disease related to prematurity, present a particular challenge. We have taken the view that this group may do better without cardiopulmonary bypass at the first palliation, although it is acknowledged that when necessary very low birth weight premature infants can undergo major surgery involving bypass with good outcomes [16]. The difficulty with a strategy that initially involves banding comes when they outgrow their pulmonary artery band and become unacceptably cyanotic. If at this point they are not suitable for a venous shunt because of elevated pulmonary pressures, then a systemic to pulmonary artery shunt is required usually in conjunction with a DKS connection. In our experience, this is a difficult group and indeed one where progression beyond an arterial shunt dependent circulation was limited. There seems little to choose between the two approaches in this regard. This study does not conclusively suggest that all patients with a systemic outflow by a BVF require a DKS connection, although clearly the majority do. We have not identified a restrictive BVF at birth in this series, or in others with this pathology outside the time-frame reported. There may be some patients in whom the BVF is so large that obstruction at that site is unlikely, but only 4 of 25 in our series achieved a Fontan completion without either a BVF resection or DKS connection. We did not see any of the commonly reported complications associated with BVF systemic outflow tract resection, including heart block and injury to the septal artery [17]. Nevertheless, our surgical management in all patients with functionally single ventricles ideally involves performing extracardiac procedures rather than intracardiac ones. Abnormalities of rhythm (early and late), early postoperative ventricular dysfunction and distortion of AV valvar support are minimized by this approach. Similarly, we now employ extracardiac conduits for Fontan completion instead of the lateral tunnel or atriopulmonary connection. The development of ventricular hypertrophy after pulmonary artery banding is a concern, but may be minimized by early conversion to a cavopulmonary connection. Increasingly, we are performing this at a younger age with acceptable results. Tweddell and colleagues [18] described in a Norwood population that the rate of weight gain significantly decreases after 3 months, before falling saturations might normally trigger reassessment and conversion to a cavopulmonary shunt. Whether the systemic ventricle is volume loaded, as in the standard Norwood palliation, or pressure loaded, as occurs when banded, we must time our interventions to minimize harm to ventricular function. In the banded population, diastolic dysfunction is a significant concern, particularly in terms of the quality of the final Fontan circulation and so early conversion to a cavopulmonary shunt is important. Conclusion In this difficult group of patients we advocate staged palliation involving early limitation of pulmonary blood flow and performing a DKS connection at the time of creating the superior cavopulmonary connection. This strategy is achieved with low morbidity and mortality and a high rate of conversion to the Fontan circulation. The authors acknowledge Dr Timothy Cartmill who operated on three of the patients in this series. They thank Drs K. C. Lau, Richard Hawker, and Gary Sholler who were responsible for the clinical and echocardiographic follow-up of these patients. References 1. Daebritz SH, Nollert GD, Zurakowski D, et al. Results of Norwood stage I operation. comparison of hypoplastic left heart syndrome with other malformations. J Thorac Cardiovasc Surg 2000;119: Ashburn DA, McCrindle BW, Tchervenkov CI, et al. Outcomes after the Norwood operation in neonates with critical aortic stenosis or aortic valve atresia. J Thorac Cardiovasc Surg 2003;125: Trusler GA, Mustard WT. A method of banding the pulmonary artery for large isolated ventricular septal defect with and without transposition of the great arteries. Ann Thorac Surg 1972;13: Mosca RS, Hennein HA, Kulik TJ, et al. Modified Norwood operation for single left ventricle and ventriculoarterial discordance: an improved surgical technique. Ann Thorac Surg 1997;64: Tchervenkov CI, Shum-Tim D, Beland MJ, Jutras L, Platt R. Single ventricle with systemic obstruction in early life: comparison of initial pulmonary artery banding versus the Norwood operation. Eur J Cardiothorac Surg 2001;19: Daenen W, Eyskens B, Meyns B, Gewillig M. Neonatal pulmonary artery banding does not compromise the shortterm function of a Damus-Kaye-Stansel connection. Eur J Cardiothorac Surg 2000;17: Odim JN, Laks H, Drinkwater DC Jr, et al. Staged surgical approach to neonates with aortic obstruction and singleventricle physiology. Ann Thorac Surg 1999;68: Freedom RM, Benson LN, Smallhorn JF, Williams WG, Trusler GA, Rowe RD. Subaortic stenosis, the univentricular heart, and banding of the pulmonary artery: an analysis of the courses of 43 patients with univentricular heart palliated by pulmonary artery banding. Circulation 1986;73: Limperopoulos C, Majnemer A, Shevell MI, et al. Predictors of developmental disabilities after open heart surgery in young children with congenital heart defects. J Pediatr 2002;141: Newburger JW, Jonas RA, Wernovsky G, et al. A comparison of the perioperative neurologic effects of hypothermic circulatory arrest versus low-flow cardiopulmonary bypass in infant heart surgery. N Engl J Med 1993;329: Jacobs ML, Rychik J, Murphy JD, Nicolson SC, Steven JM, Norwood WI. Results of Norwood s operation for lesions other than hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 1995;110: Ishino K, Stumper O, De Giovanni JJ, et al. The modified Norwood procedure for hypoplastic left heart syndrome: early to intermediate results of 120 patients with particular reference to aortic arch repair. J Thorac Cardiovasc Surg 1999;117: Chessa M, Dindar A, Vettukattil JJ, et al. Balloon angioplasty in infants with aortic obstruction after the modified stage I Norwood procedure. Am Heart J 2000;140: McGuirk SP, Winlaw DS, Langley SM, et al. The impact of ventricular morphology on midterm outcome following completion total cavopulmonary connection. Eur J Cardiothorac Surg 2003;24: Broekhuis E, Brizard CP, Mee RB, Cochrane AD, Karl TR. Damus-Kaye-Stansel connections in children with previously transected pulmonary arteries. Ann Thorac Surg 1999; 67:

8 Ann Thorac Surg CLARKE ET AL 2004;78:650 7 TIMING OF DAMUS-KAYE-STANSEL CONNECTION Reddy VM, Hanley FL. Cardiac surgery in infants with very low birth weight. Semin Pediatr Surg 2000;9: O Leary PW, Driscoll DJ, Connor AR, Puga FJ, Danielson GK. Subaortic obstruction in hearts with a univentricular connection to a dominant left ventricle and an anterior INVITED COMMENTARY This report by Clarke and coworkers is an important contribution to the surgical literature. The authors report excellent results in a complex group of patients presenting with double inlet left ventricle and aortic arch obstruction. In some centers, infants presenting with this anatomic substrate are routinely treated with a Norwood-type reconstruction. The authors demonstrate excellent first-stage outcomes utilizing a conservative policy of primary arch reconstruction and pulmonary artery banding. Because many, if not most, of these infants eventually develop obstruction at the level of the bulboventricular foramen (BVF), a Damus-Kaye-Stansel (DKS) procedure is a useful second- or third-stage procedure. The authors approach avoids the use of cardiopulmonary bypass in the neonate and provides a low overall mortality. The authors have also demonstrated that short-term pulmonary artery banding is a reasonable method to control pulmonary blood flow in the neonate. In addition, they demonstrate that the DKS is a safe and durable approach for the treatment of systemic ventricular outlet obstruction in the setting of transposed great vessels. Although it can be argued that the authors are comparing their outcomes to the older, published results for the Norwood operation (utilizing a systemic to pulmonary artery shunt) in nonhypoplastic left heart syndrome patients, I am unaware of any data to suggest that the currently popular modified Norwood procedure utilizing a right ventricle-to-pulmonary artery (RV-PA) conduit is well suited to this anatomic substrate. The pediatric cardiology and cardiac surgery communities eagerly await mid- and long-term data comparing outcomes for the traditional Norwood procedure utilizing a systemic to pulmonary artery shunt when compared with the more modern version of the Norwood procedure utilizing a RV-PA conduit. In essence, the special anatomic considerations present in the setting of double inlet left ventricle with L-transposition of the great vessels may not be ideal for the RV-PA conduit approach. The authors approach should be particularly attractive to smaller centers where the outcome for a primary Norwood type reconstruction may be less favorable. The authors are fortunate that none of their patients had a BVF so small that a primary Norwood procedure would be the only option. The authors describe some late subaortic outlet chamber. Results of a staged approach. J Thorac Cardiovasc Surg 1992;104: Tweddell EA. Strategies to reduce in hospital and interstage mortality for stage one HLHS. Presented at the 39th Annual Meeting of Society of Thoracic Surgeons, San Diego, morbidity in terms of reoperation on the aortic arch. This finding is not unique to their approach, because some patients will require late arch surgery or balloon dilatation after a Norwood-type reconstruction. I have one technical concern about performing a DKS procedure several months after repair of a coarctation. In some patients, it is possible to create mild to moderate residual midtransverse arch obstruction unless the DKS patch is extended into the prior coarctation repair site. The authors have demonstrated that a carefully constructed DKS connection does not result in progressive distortion and malfunction of the semilunar valve. They report satisfactory mid-term semilunar valve function even when the pulmonary valve was resurrected in those patients that had previously undergone division of the main pulmonary artery. It is my bias that patients with the potential for systemic ventricular outlet obstruction at a BVF should be considered for a prophylactic DKS procedure whenever the pulmonary artery is divided or ligated. The authors did not present data on late ventricular function. We should hope that they continue to follow this important group of patients and that they have objective data in terms of exercise capacity and ventricular function when they present their long-term outcomes. A properly performed and timely DKS connection should result in better late systemic ventricular function when compared with the ventricular function of patients undergoing late relief of BVF obstruction by muscle resection performed through the aortic valve or by means of an incision in the right ventricular outflow chamber. In summary, the authors approach is well conceived and well executed. The excellent mid-term outcomes suggest that this group of patients will have an excellent longterm functional result. John Lamberti, MD Department of Cardiac Surgery Division of Pediatric Cardiology/Cardiac Surgery Children s Hospital of Oakland nd St Oakland, CA jlamberti@mail.cho.org 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /j.athoracsur