- Email: [email protected]
Second stage after initial hybrid palliation for hypoplastic left heart syndrome: Arterial or venous shunt?
CONGENITAL HEART DISEASE: HYPOPLASTIC LEFT HEART SYNDROME
Second stage after initial hybrid palliation for hypoplastic left heart syndrome: Arterial or venous shunt? Mohamed S. Nassar, PhD, FRCS,a,b,c Srinivas A. Narayan, MD, MRCPCH,b Andrew Nyman, MRCPCH,b Caner Salih, MD, FRCS,b Conal B. Austin, MD, FRCS,b David Anderson, MD, FRCS,a,b and Tarique Hussain, MD, PhDa,b ABSTRACT Objective: Hybrid palliation for hypoplastic left heart syndrome has been developed as an alternative to neonatal Norwood surgery. At the second stage, a source of pulmonary blood flow has to be established. Options include an arterial modified Blalock–Taussig or a venous superior cavopulmonary shunt. Methods: We retrospectively reviewed patients who received second-stage palliation after the initial hybrid. Patients were stratified according to the source of pulmonary blood supply into the arterial shunt (n ¼ 17 patients) or venous shunt (n ¼ 26 patients).
CHD
Results: Age and weight at second stage were lower in the arterial group (85 [45268] days vs 152.5 [61-496] days, P ¼ .001 and 3.6 [2.7-9.4] kg vs 5.1 [2.97-9.4] kg, P ¼ .001, respectively). All recorded surgical times were shorter in the arterial group. Mechanical ventilation and intensive care stay were shorter in the venous group (5.82 [2.01-14.9] days vs 2.42 [0.56-13.67] days, P ¼ .005 and 8.5 [3.623.7] vs 5.75 [0.8-17.6] days, P ¼ .036, respectively) There was no difference in mortality (2/17 vs 5/26; P ¼ .685) or incidence of complications between the 2 groups. There was a tendency toward a higher need for intervention in the immediate postoperative period in the venous group, but this did not reach significance (6/17 vs 13/26, P ¼ .342). The arterial group has shown better development of the branch pulmonary arteries with a higher lower lobe index (158.38 39.43 mm2/m2 vs 113.33 43.96 mm2/m2, respectively, P ¼ .037). Conclusions: Both arterial and venous shunts are viable options with mortality and morbidity results comparable to those in the literature. The arterial shunt pathway (2-stage Norwood I) may offer better pulmonary arterial growth than the venous shunt (comprehensive/combined Norwood I and II). (J Thorac Cardiovasc Surg 2015;150:350-7)
From the aDepartment of Cardiovascular Imaging, Division of Imaging Sciences and Biomedical Engineering, King’s College, London, United Kingdom; bEvelina London Children Hospital, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom; and cDepartment of Cardiothoracic Surgery, Alexandria University, Alexandria, Egypt. Financial support was received from the Department of Health via the National Institute for Health Research comprehensive Biomedical Research Centre award to Guy’s & St Thomas’ NHS Foundation Trust in partnership with King’s College London and King’s College Hospital NHS Foundation Trust. This work was also funded by the British Heart Foundation Award RE/08/003. Received for publication Jan 27, 2015; revisions received March 28, 2015; accepted for publication April 18, 2015; available ahead of print June 6, 2015. Address for reprints: Mohamed S. Nassar, PhD, FRCS, Paediatric Cardiology and Cardiac Surgery, Evelina London Children’s Hospital, Guy’s & St Thomas’ NHS Foundation Trust, Westminster Bridge Road, London SE1 7EH, UK (E-mail: [email protected]). 0022-5223/$36.00 Copyright Ó 2015 by The American Association for Thoracic Surgery http://dx.doi.org/10.1016/j.jtcvs.2015.04.039
350
Schematic representation of overall outcomes of patients. Central Message This observational study show that a 2-stage Norwood I seems to be noninferior to a comprehensive Norwood I and II and may give better pulmonary arterial growth. Perspective Hybrid palliation for hypoplastic left heart syndrome is a relatively new technique. Our results have shown that a 2-stage Norwood I is a valid option compared with the comprehensive Norwood II. Over the years, we have developed criteria that can help to define which patient would benefit from a 2-stage Norwood I.
See Editorial page 286.
Originally described to reduce the risk of neonatal surgery, hybrid palliation for hypoplastic left heart syndrome (HLHS) has been introduced as an alternative to the Norwood procedure, aiming to defer a more complex surgery to a later date with an older child who can withstand major operation with relatively lower risk.1 Since 2005, it has been our institutional policy to use the hybrid operation in patients with HLHS presenting with low birth weight (<2.5 kg), postnatal collapse, restrictive atrial septum, or possible biventricular repair. The second stage after the initial hybrid involves construction of the aortic arch with establishment of pulmonary blood supply. A superior cavopulmonary anastomosis has been the pulmonary blood flow source of choice for this (ie, combined/comprehensive I and II as the second procedure), thus avoiding the transition from an arterial to a venous shunt seen in the traditional Norwood pathway.2
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
Abbreviations and Acronyms AVV ¼ atrioventricular valve HLHS ¼ hypoplastic left heart syndrome MRI ¼ magnetic resonance imaging PA ¼ pulmonary artery POD ¼ postoperative day RPA ¼ right pulmonary artery SCPC ¼ superior cavopulmonary connection Concerns have been raised regarding this approach considering the significant challenge in the reconstruction of the pulmonary arteries (PAs) at the second stage after initial hybrid palliation.3 In addition, reports have shown lower branch PA size in the hybrid pathway compared with the Norwood pathway.4 Some centers have reported their experience with initial PA banding and duct patency maintenance with prostaglandin infusion followed by the Norwood I intervention, the so-called rapid 2-stage Norwood operation, for high-risk patients.5-7 Others have shifted their practice from a venous type of shunt to an arterial shunt after initial banding and maintenance of duct patency.8,9 We currently have a mixed practice of arterial or venous shunt. We reserve the arterial shunt simply because of the surgeon’s preference or for those patients with suspected increased pulmonary vascular resistance, such as those developing a restrictive interatrial septum; those presenting with an indication for an early second stage, such as tight bands or retrograde coarctation; those with anomalous pulmonary venous drainage, in whom intervention on the superior vena cava or longer complex procedure is anticipated; those in whom better development of branch PA is required, as with hypoplastic PAs or interrupted IVC; and those who require a bail out of a high-pressure cavopulmonary anastomosis. On the other hand, in the absence of any of the previous indications, we tend to allocate all other patients, especially those with aberrant subclavian artery, to a venous type of shunt. In the current study, we compare these 2 cohorts of patients in terms of second-stage results, morbidity, mortality, and candidacy for Fontan completion. MATERIALS AND METHODS This is a retrospective study of a single institution. After institutional review board approval, the department database (Heartsuite XP 3.9.14, Systeria, Glasgow, UK) was used to identify patients who reached the second-stage palliation after an initial hybrid between the start of the program in November 2005 and November 2014. Those who underwent biventricular repair, before the second stage, were excluded. Patients were divided into 2 groups according to the source of pulmonary blood flow: Group A received a systemic modified Blalock–Taussig shunt, and group B received a venous hemi-Fontan shunt. Not all patients were preassigned to either group at the time of hybrid. Postoperatively, patients received brain magnetic resonance imaging (MRI) or screening of vocal cords only if clinically indicated.
SURGICAL TECHNIQUE We have previously reported our surgical technique for hybrid and arch reconstruction.10,11 At the time of the second-stage palliation, an arterial or venous shunt is constructed as the source of pulmonary blood flow. The arterial shunt is usually constructed between the first branch of the arch and the right pulmonary artery (RPA), and the venous shunt is a hemi-Fontan type of superior cavopulmonary anastomosis constructed in most cases under a period of total circulatory arrest. The ductal stent is usually peeled off the descending aorta after resection of the isthmus. In patients in whom the stent has been in place for a long time, the part of the descending aorta with the embedded stent is usually resected. The Sano-type right ventricle to PA shunt was not performed in any patients because of the institutional preference to avoid ventriculotomy of the systemic ventricle. We tend not to patch the site of banding because of the friable nature of tissue. Instead, we free all adventitial bands and stretch the vessel from within. Patch material (a pulmonary homograft patch) is used only if the arterial wall is injured while breaking adhesions or the band has been in place for a long time and stretching was insufficient to dilate the constriction point. The PA stump is usually closed directly, and in a few cases, a small patch is used to close the stump to avoid branch PA distortion. At the time of Fontan completion, a 4-mm fenestrated lateral tunnel Fontan is constructed. Any additional intervention on the atrioventricular valve (AVV), arch, or branch PA was recorded. ASSESSMENT OF VENTRICULAR AND ATRIOVENTRICULAR VALVE FUNCTION Patient records were reviewed. Echocardiographic qualitative assessments of ventricular function and degree of AVV regurgitation before operation and hospital discharge were recorded. For comparison, ventricular function was graded as 1 for normal/mildly depressed, 2 for mildmoderate/moderately depressed, and 3 for moderatesevere/severely depressed. AVV regurgitation was graded as 1 for trivial/mild, 2 for mild-moderate/moderate, and 3 for moderate-severe/severe regurgitation. PRE-FONTAN ASSESSMENT Currently, MRI is our investigation of choice for single ventricle palliation.12 Our MRI technique has been reported.10 The following data were recorded: 1. Indexed lower lobe area of RPA and left pulmonary artery (LPA), because the lower lobe index was used to compare branch PA growth between the 2 groups. 2. Differential flow between both branch PAs. 3. PA pressure (measured through a superior vena cava line at the time of MRI).
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
351
CHD
Nassar et al
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
4. Any arch distortion or narrowing was recorded. 5. End-diastolic volume indexed to body surface area. 6. Ejection fraction as an indicator of single ventricle function. 7. Oxygen saturation at the time of MRI study. Statistics Data were summarized as mean standard deviation or as median and range for continuous variables. For categoric variables, number and percentage were used. The distribution of continuous and scale variables was tested by using the Kolmogorov–Smirnov test for normality. Variables with normal distribution were compared with independent samples t test. Mann–Whitney U test was used for variables with non-normal distribution. Categoric variables were compared by the Fisher exact test. IBM SPSS version 22 (IBM, Armonk, NY) was used for all statistical calculations.
CHD
RESULTS During the study period, a total of 146 Norwood and 79 hybrid procedures were performed in our center. Of those 79 patients, 11 were excluded because they had biventricular repair. A total of 43 patients have reached the second stage of univentricular palliation. Seventeen patients received an arterial shunt (group A), and 26 patients received a venous shunt (group B). During the first half of our experience (2005-2009), we tried to push all patients to the comprehensive stage (venous shunt), with 9 patients receiving a venous shunt and only 2 patients receiving an arterial shunt. Because our experience grew overtime, we developed the previously mentioned indications for arterial shunt. Of the 17 patients who received an arterial shunt, 4 had anomalous systemic venous drainage, 2 had partial anomalous pulmonary venous drainage, 5 had developed restrictive inter-atrial septum, 1 had retrograde coarctation, 1 had narrowing of ductal stent, 1 had a small LPA, and 1 had a bail out. Preoperative Data Group A patients had significantly lower age and weight at the time of operation. There was no significant difference in preoperative echocardiography rank between the 2 groups. Data are summarized in Table 1. Surgical Data At the second stage, all recorded times were significantly shorter in the arterial shunt group. There was no difference in the need for branch PA plasty or for a second run. There was a higher need for delayed sternal closure in the arterial group, but none of the patients have shown any evidence of mediastinitis. Two patients in the venous shunt group received an additional arterial shunt because of persistent 352
Nassar et al
TABLE 1. Preoperative characteristics Arterial shunt Venous shunt P (17 patients) (26 patients) value Age (d) 85 (45-268) Weight (kg) 3.6 (2.7-9.4) Diagnosis Aortic atresia/mitral atresia 5 Aortic atresia/mitral stenosis 1 Aortic stenosis/mitral atresia 2 Aortic stenosis/mitral stenosis 6 0 Aortic stenosis, endomyocardial elastofibrosis Tricuspid atresia, transposition 1 of the great arteries Unbalanced atrioventricular 0 septal defect Aortic atresia, large inlet VSD, 1 hypoplastic arch Other diagnoses 1 PAPVD 2 ASVD Left SVC 3 Interrupted IVC 1 Hypoplastic LPA 1 Primary indication for hybrid: Low birth weight 9 Postnatal collapse 2 Restrictive atrial septum 2 Possible biventricular repair 2 Others 2 Echocardiography function: Normal/mildly depressed 13 Mild-moderate/moderately 3 depressed Moderate-poor/poor 1 Echocardiography AVV incompetence: Normal/mildly 13 Mild-moderate/moderate 4 Moderate-severe/severe 0
152.5 (61-496) 5.1 (2.97-9.4)
.001 .001
5 2 1 3 6
.481 1.000 .552 .122 .066
1
1.000
2
.511
2
1.000
4 0
.633 .151
2 0 1
.369 .395 1.000
8 5 2 6 5
.205 .685 1.000 .446 .685 .683
16 8 2 1.000 19 7 0
VSD, Ventricular septal defect; PAPVD, partial anomalous pulmonary venous drainage; ASVD, anomalous systemic venous drainage; SVC, superior vena cava; IVC, inferior vena cava; LPA, left pulmonary artery; AVV, atrioventricular valve.
low saturation, and both died. There was 1 intraoperative crossover in each group. An arterial shunt was changed to a venous shunt because of poor function and evidence of coronary steal, but the patient did not survive. One patient with a venous shunt required takedown to an arterial shunt because of unacceptably high PA (superior vena cava) pressure; this patient survived and subsequently had successful Fontan completion. Four patients (3 in group A and 1 in group B) required a second bypass run, within the same procedure, for arch redo. One patient in group A required arch redo on postoperative day (POD) 1 because of a kink in the arch that was
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
TABLE 2. Second-stage surgical intervention data Procedure
Arterial shunt (17 patients)
Venous shunt (26 patients)
P value
CPB (min) Crossclamp (min) Circulatory arrest (min) Branch PA plasty AVV repair Delayed chest closure Second run
183.53 85.21 98.06 53.50 20 (3-137) 4 0 13 4
201.38 51.39 122.46 34.43 47 (0-89) 5 1 10 6
.048 .003 .006 1.000 1.000 .027 1.000
Second run refers to reinstitution of cardiopulmonary bypass within the same procedure. Delayed chest closure was unplanned and considered in case of bleeding, hemodynamic instability, or desaturation. CPB, Cardiopulmonary bypass; PA, pulmonary artery; AVV, atrioventricular valve.
discovered on return to the pediatric intensive care unit. Data are summarized in Table 2. Pediatric intensive care unit stay and duration of ventilation were shorter in the venous shunt group. At hospital discharge, there was no significant difference in echocardiography findings. Group A had significantly higher oxygen saturation at discharge. Data are summarized in Table 3. TABLE 3. Postoperative data Arterial shunt (17patients)
Venous shunt (26 patients)
PICU stay: Length of stay (d) 8.5 (3.6-23.7) 5.75 (0.8-17.6) Mechanical ventilation (d) 5.82 (2.01-14.9) 2.42 (0.56-13.67) Need for inotropes (d) 7 (3-23) 5 (0.8-17) Mortality 2 (12%) 5 (19%) Morbidity Brain injury 5 (29%) 8 (31%) Necrotizing enterocolitis 1 2 Left vocal cord palsy 1 2 Chylothorax 1 2 Echocardiography findings: Function: Normal/mildly depressed 12 13 Mild-moderate/moderate 3 8 Moderate-poor/poor 2 5 AVV incompetence: Normal/ mild 10 16 Mild-moderate/moderate 7 5 Moderate-severe/severe 0 5 Reintervention Diagnostic catheter 3 8 Catheter intervention 3 3 Surgery 2 6 Cardiac MRI/CT 3 4 Total interventions 11 21 No. of patients 6 (35%) 13 (50%) Hospital stay (d) 15 (8-30) 13 (1-57) Oxygen saturation at 85 (75-92) 80 (50-89) hospital discharge (%)
P value .036 .005 .10 .685 1.000 1.000 1.000 1.000 .246
.088
.480 .666 .446 1.000 .295 .342 .471 .014
PICU, Pediatric intensive care unit; AVV, atrioventricular valve; MRI, magnetic resonance imaging; CT, computed tomography.
Survival There were 2 early mortalities in the arterial shunt group. The first patient had a kink of the reconstructed arch that was discovered on return to the pediatric intensive care unit, and this patient underwent reoperation. She was extubated later, with echocardiography confirming good results of arch reconstruction, but on POD 9 she had a cardiac arrest during transfer for MRI of the brain. The second patient had a thrombosed 4-mm shunt, underwent emergency stenting of the shunt, and the branch PA was carried out, but his condition continued to deteriorate and he died on POD 15. In the venous shunt group, there were 5 early mortalities. Two of those were very early in the study period. The first had an RPA kink with formed thrombus in the superior vena cava. This patient had revision of the venous shunt and an additional arterial shunt, but a large cerebral infarction developed and the patient died on POD 9. The second patient required bilateral PA stenting, and later, the venous shunt was taken down and replaced with an arterial shunt, but the function deteriorated and this patient died on POD 16. The third patient had persistent low saturations, and an additional arterial shunt was constructed that thrombosed with thrombus extending to the PAs, and this patient died on POD 1. The fourth patient was returned to the operating room for bleeding, but cardiac function deteriorated and a significant cerebral infarct developed; the patient died on POD 17. The fifth patient was admitted 2 months after the initial hybrid with desaturation and poor right ventricular function. A rapid Norwood one I was planned but converted to a venous shunt because of volume load, ST depression, and significant AVV regurgitation at the end of the operation; his cardiac function deteriorated, and he died on POD 2. There was 1 late mortality in a patient in the venous shunt group who had an RPA kink that was addressed at the time of the Fontan completion. However, cardiac function deteriorated after surgery, subsequent multiorgan failure subsequently developed, and the patient died. Reintervention In the immediate postoperative period, apart from 1 arch reintervention, the branch PA or shunt revision was the target of all reinterventions. There was a trend toward a higher need for imaging and intervention in the venous shunt group. However, this did not reach statistical significance; data are summarized in Table 3. In the arterial shunt group, 14 of the 15 survivors have received a superior cavopulmonary connection (SCPC), with 1 patient awaiting surgery. Concomitant procedures at the time of SCPC included AVV repair in 2 patients, LPA plasty in 3 patients, and arch redo in 1 patient. Recorded times included cardiopulmonary bypass (96.2 25.35 minutes), crossclamp time (48.5 23.24 minutes), and circulatory arrest (20.4 13.59 minutes). At the stage of SCPC, there were no mortalities,
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
353
CHD
Nassar et al
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
TABLE 4. Magnetic resonance imaging findings at pre-Fontan assessment Arterial shunt (9 patients)
Venous shunt (13 patients)
EF (%) 58 (43-67) 65.5 (44-70) iEDV (mL/m2) 92.78 19.37 75.30 25.26 Differential 73/27 (45-90/10-55) 67/33 (50-85/15-50) RPA/LPA flow 114.40 28.64 69.55 25.23 iRLL area (mm2/m2) iLLL area (mm2/m2) 43.98 18.74 43.78 24.07 Lower lobe index 158.38 39.43 113.33 43.96 (mm2/m2) PAP (mm Hg) 13 2.12 12.0 1.41 Oxygen Saturation at 80.67 4.71 81.89 4.29 time of MRI
P value .250 .110 .650 .003 .930 .037 .257 .520
EF, Ejection fraction; iEDV, indexed end-diastolic volume; RPA, right pulmonary artery; LPA, left pulmonary artery; iRLL, indexed right lower lobe area; iLLL, indexed left lower lobe; PAP, pulmonary artery pressure; MRI, magnetic resonance imaging.
and the only recorded morbidity was a prolonged chylothorax in 1 patient. Before the Fontan completion, 3 patients in the arterial group and 1 patient in the venous group required catheter intervention on the branch PA. Overall, there was no significant difference in the number of patients who required intervention on the branch PA (41% vs 34%, respectively).
Pre-Fontan Magnetic Resonance Imaging Nine patients in group A and 13 patients in group B had pre-Fontan MRI studies. Lower lobe index measurement was possible in all but 3 patients in group B (because of quality of images in 2 and 1 was excluded because it had a single lung Fontan because of congenital unilateral lung hypoplasia). There was no evidence of arch obstruction in any patient at the pre-Fontan assessment. Results are shown in Table 4. Fontan Completion In the arterial group, 4 of the 15 survivors had Fontan completion, and in the venous group, 9 of 21 patients had Fontan completion (Figure 1). Additional procedures at the time of Fontan completion were 1 PA plasty in the venous group and 1 AVV repair in the arterial shunt group. There was no statistically significant difference between the 2 groups at the time of Fontan completion considering surgical times and postoperative stay. DISCUSSION Data comparing the 2 possible pathways after the initial hybrid are limited. Having a mixed practice, arterial or venous shunt at the second stage, has given us the chance
CHD FIGURE 1. Schematic representation of overall outcomes of patients. SCPC, Superior cavopulmonary connection.
354
Nassar et al
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
to compare these 2 cohorts. Our results have shown points of strength and weakness for each option. There was no significant difference in mortality between the 2 groups (11.7% vs 19.2%). The higher incidence of mortality in the combined group is in part because 2 of the 5 deaths were in our very early practice, perhaps representing a learning curve, and it is possible that this relates to evolution of a strategy for assigning the appropriate shunt type for each patient. Retrospective review of our patient cohort shows that some of the earlier cohort who underwent venous shunt pathway may have been assigned an arterial shunt in the current era (1 early mortality in the venous group, who had a restrictive septum). Our mortality results are still comparable to most of the reported series,13,14 bearing in mind that in our center, the hybrid procedure is reserved for higher-risk, more complex, and unstable patients in whom a Norwood procedure would carry an unacceptable high risk. A previous report from our center has confirmed the higher risk of our hybrid cases using a modified comprehensive Aristotle score.15 Three of the 5 mortalities in the venous shunt group were related to the venous shunt, supporting the notion that reconstruction of the PAs is a significant challenge at this stage. This was further reinforced by the difference in all recorded surgical times. There was 1 intraoperative crossover in each group. We thought it was more appropriate to assign patients to each group according to the type of shunt they received at the end of the operation because the complications are often related to the hemodynamic state produced by the shunt. Assigning those 2 patients on the basis of intention to treat would not affect the final mortality results (2 vs 5 for presented data [P ¼ .68] and 3 vs 4 for intention to treat analysis [P ¼ 1.00]). The higher incidence of unplanned delayed sternal closure in the arterial shunt group could be related to the early hemodynamic instability caused by the high pulmonary flow state, with possible right ventricle volume overload or coronary underperfusion, in comparison with the more stable situation and volume off-loading advantage of the venous shunt. On the other hand, this hemodynamic stability was at the expense of significantly lower saturations in the venous shunt group, proven by the higher need for PA imaging and lower saturations at discharge. This postoperative hypoxia could be due to not only the technically demanding reconstruction of the PAs but also the pulmonary vascular resistance, which may be elevated after a long bypass run.16,17 The effect of this postoperative hypoxia on neurology was not evident when major neurologic complications were compared between the 2 groups. Although we do not perform routine postoperative brain MRI, we have a low threshold for MRI whenever the clinical scenario (eg, change of motor tone, convulsions) suggests possible brain injury. In both groups, 30% of patients had MRI-proven brain abnormalities after clinically detected abnormal neurology in the immediate
postoperative period. This relatively high incidence is comparable to other reports studying the neurologic development in patients with HLHS. Reports on brain MRI studies in patients with HLHS show up to 73% incidence of new or worsened lesions.18 Moreover, it seems that the difference in oxygen saturation between the 2 groups became less evident with time, because at the time of pre-Fontan MRI, there was no significant difference between the 2 groups. Long-term neurodevelopmental delay has been demonstrated in patients with HLHS.19 Saiki and colleagues20 studied cerebral perfusion after bilateral banding. Their results were in favor of a shorter duration between the initial hybrid and the second stage. They have shown that in patients posthybrid, cerebral blood flow worsens over time.20 We could not find data on the incidence of neurologic complications in the few reports comparing post-hybrid Norwood I and comprehensive Norwood I and II after bilateral PA banding.8,14 Whether the longer duration until second-stage palliation and the lower postoperative saturations in the venous group would induce subtle brain changes making a difference between both groups in the long run requires further investigation. When it comes to the need for reintervention after the second stage, one of the disadvantages of the arterial shunt is the need for an additional surgery to establish superior cavopulmonary anastomosis as an intermediate step before the Fontan circulation. Although there was no difference on the rate of reintervention after second-stage hospital discharge (excluding SCPC stage), there was a higher, although not statistically significant, need for intervention in the venous shunt group immediately after surgery. All patients with an arterial shunt who went through the additional stage survived with very low morbidity and a short hospital stay. Arch reconstruction is a challenge in patients with HLHS. The presence of a ductal stent adds to the challenge of the second stage after a hybrid. Shorter duration between initial hybrid palliation and second stage (as in arterial shunt group) would theoretically make stent extraction, and subsequently arch reconstruction, less challenging compared with the well-embedded stent seen with longer interval between the initial hybrid and the second-stage palliation (in the venous shunt group). Arch obstruction is a known complication after arch reconstruction in patients with HLHS; 5 of our patients required a second bypass run for redo arch reconstruction. There was no difference in arch reintervention between the 2 groups; only 1 patient (1/43, 2.3%) required arch reintervention after hospital discharge. We have a low incidence of arch obstruction compared with reports from other centers. Results from the single ventricle trial and the Boston group have shown an incidence of arch obstruction up to 29%.21,22 We have previously reported our arch reconstruction technique,10 and we believe that the use of pulmonary homograft patch and the extension
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
355
CHD
Nassar et al
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
of the reconstruction far down into the descending aorta can avoid the future development of arch obstruction.
CHD
Pulmonary Arteries Baba and colleagues3 commented on the significant challenge in reconstructing the PAs at the comprehensive stage II palliation and in achieving subsequent reasonable PA architecture for Fontan completion. They have mentioned that the mechanism of branch PA stenosis mainly stems not only from anatomic distortion of the branch PA but also from the difficulty in geometric arrangement between the newly reconstructed aorta and the branch PA.3 The rigid graft and the higher systemic pressure in the arterial shunt can overcome any mild degree of branch PA or PA stump distortion, which would be a challenge to the compressible lowpressure venous shunt. This is supported by our results, which have shown a higher need for addressing the branch PA in the immediate postoperative period in the venous shunt group, compared with later intervention in the arterial shunt group (usually at the time of SCPC). Review of the mortality data presented supports the notion that significant residual branch PA narrowing requires prompt reintervention and consideration of anticoagulation. This inherent problem of the comprehensive stage II has shifted some centers to a rapid 2-stage N I operation (primary branch PA banding and maintenance of ductal patency with prostaglandin followed by an early Norwood procedure). When performing an SCPC after an arterial shunt, the branch PAs, in particular the LPA, are fixed within the mediastinum. This makes further distortion less likely while constructing the venous shunt compared with the comprehensive stage II. A surgical strategy of routine mobilization from hilum to hilum is used at some centers with more liberal use of patching material. The merits of that approach have not been assessed in this article. Our approach has been to use the hemi-Fontan procedure to achieve better central reconstruction and avoid patching. Likewise, the merit of performing a hemi-Fontan for this reason over a bidirectional Glenn, which may reduce bypass time, is beyond the scope of this article. Davies and colleagues23 have shown that there is a higher rate of PA intervention if bands were left in place for more than 90 days because early removal of branch PA bands would avoid scarring and narrowing at the band sites. We have not seen any difference in surgical branch PA plasty at the time of the second stage. As we have mentioned, we tend to avoid patching the branch PAs whenever possible. Previous reports on the development of branch PAs in the hybrid pathway after the comprehensive stage II have confirmed smaller vessels and a greater need for intervention when compared with the Norwood pathway.4 Ota and colleagues9 compared branch PA development after a rapid 2-stage N I operation to a classic Norwood pathway, and their results showed no difference. Our results are in 356
Nassar et al
harmony with previous studies showing a significantly higher lower lobe index in the arterial shunt group. Sakamoto and colleagues8 showed a tendency toward better branch PA development in the arterial shunt group in their series. However, Davies and colleagues14 have failed to demonstrate this in their cohort. However, it is worth mentioning that this report contains only a brief comment on the absolute diameter of branch PA on conventional angiography.14 In our series, there was a significant difference in branch PA development at the time of pre-Fontan assessment. However, this difference in vessel size was not reflected on the measured PA pressure. Therefore, the effect on long-term outcome is still unknown. Finally, the aim of the palliative surgery for patients with HLHS is to reach the Fontan circulation. Fontan candidacy could be compromised by deterioration of ventricular function, AVV malfunction, or distortion in branch PA geometry. The only difference in our series was the size of the branch PAs. None of the patients in our cohort were refused Fontan completion. Study Limitations The retrospective single-center nature of the study and the relatively small number of cases are the main limitations of the study. This was a retrospective study, so patients were not randomized to either pathway. A larger number of patients and a longer follow-up may reveal more significant differences between the 2 pathways. Because of the consistent operative management of these patients, no inferences could be drawn about potentially important determinants of outcome, such as the type of arch reconstruction or SCPC, among others. These important limitations call for large multicenter studies to better define the outcome of different management strategies. CONCLUSIONS The option of an arterial shunt (2-stage Norwood I) does not seem to be inferior to the commonly used venous shunt (comprehensive Norwood I and II). There is growing evidence to suggest that early intervention after bilateral branch PA banding is favorable, which is in favor of the arterial shunt pathway. Conflict of Interest Statement Dr Hussain reports grants from NIHR comprehensive Biomedical Research Centre award and grants from British Heart Foundation award RE/08/003 during the conduct of the study. All other authors have nothing to disclose with regard to commercial support. References 1. Gibbs JL, Wren C, Watterson KG, Hunter S, Hamilton JR. Stenting of the arterial duct combined with banding of the pulmonary arteries and atrial septectomy or septostomy: a new approach to palliation for the hypoplastic left heart syndrome. Br Heart J. 1993;69:551-5.
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
2. Michel-Behnke I, Akintuerk H, Marquardt I, Mueller M, Thul J, Bauer J, et al. Stenting of the ductus arteriosus and banding of the pulmonary arteries: basis for various surgical strategies in newborns with multiple left heart obstructive lesions. Heart. 2003;89:645-50. 3. Baba K, Kotani Y, Chetan D, Chaturvedi RR, Lee KJ, Benson LN, et al. Hybrid versus Norwood strategies for single-ventricle palliation. Circulation. 2012; 126(11 Suppl 1):S123-31. 4. Dave H, Rosser B, Knirsch W, Hubler M, Pretre R, Kretschmar O. Hybrid approach for hypoplastic left heart syndrome and its variants: the fate of the pulmonary arteries. Eur J Cardiothorac Surg. 2014;46:14-9. 5. Gomide M, Furci B, Mimic B, Brown KL, Hsia TY, Yates R, et al. Rapid 2-stage Norwood I for high-risk hypoplastic left heart syndrome and variants. J Thorac Cardiovasc Surg. 2013;146:1146-52. 6. Guleserian KJ, Barker GM, Sharma MS, Macaluso J, Huang R, Nugent AW, et al. Bilateral pulmonary artery banding for resuscitation in high-risk, single-ventricle neonates and infants: a single-center experience. J Thorac Cardiovasc Surg. 2013;145:206-14. 7. Russell RA, Ghanayem NS, Mitchell ME, Woods RK, Tweddell JS. Bilateral pulmonary artery banding as rescue intervention in high-risk neonates. Ann Thorac Surg. 2013;96:885-90. 8. Sakamoto T, Harada Y, Kosaka Y, Umezu K, Yasukochi S, Takigiku K, et al. Second-stage palliation after bilateral pulmonary artery bands for HLHS and its variants–which is better, modified Norwood or Norwood plus bidirectional Glenn? World J Pediatr Congenit Heart Surg. 2011;2:558-65. 9. Ota N, Murata M, Tosaka Y, Ide Y, Tachi M, Ito H, et al. Is routine rapid-staged bilateral pulmonary artery banding before stage 1 Norwood a viable strategy? J Thorac Cardiovasc Surg. 2014;148:1519-25. 10. Nassar MS, Bertaud S, Goreczny S, Greil G, Austin CB, Salih C, et al. Technical and anatomical factors affecting the size of the branch pulmonary arteries following first-stage Norwood palliation for hypoplastic left heart syndrome. Interact Cardiovasc Thorac Surg. 2015;20:631-5. 11. Venugopal PS, Luna KP, Anderson DR, Austin CB, Rosenthal E, Krasemann T, et al. Hybrid procedure as an alternative to surgical palliation of high-risk infants with hypoplastic left heart syndrome and its variants. J Thorac Cardiovasc Surg. 2010;139:1211-5. 12. Heathfield E, Hussain T, Qureshi S, Valverde I, Witter T, Douiri A, et al. Cardiovascular magnetic resonance imaging in congenital heart disease as an alternative to diagnostic invasive cardiac catheterization: a single center experience. Congenit Heart Dis. 2013;8:322-7. 13. Brescia AA, Jureidini S, Danon S, Armbrecht E, Fiore AC, Huddleston CB. Hybrid versus Norwood procedure for hypoplastic left heart syndrome:
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
contemporary series from a single center. J Thorac Cardiovasc Surg. 2014; 147:1777-82. Davies RR, Radtke W, Bhat MA, Baffa JM, Woodford E, Pizarro C. Hybrid palliation for critical systemic outflow obstruction: neither rapid stage 1 Norwood nor comprehensive stage 2 mitigate consequences of early risk factors. J Thorac Cardiovasc Surg. 2015;149:182-93. Lloyd DF, Cutler L, Tibby SM, Vimalesvaran S, Qureshi SA, Rosenthal E, et al. Analysis of preoperative condition and interstage mortality in Norwood and hybrid procedures for hypoplastic left heart syndrome using the Aristotle scoring system. Heart. 2014;100:775-80. Kirshbom PM, Jacobs MT, Tsui SS, DiBernardo LR, Schwinn DA, Ungerleider RM, et al. Effects of cardiopulmonary bypass and circulatory arrest on endothelium-dependent vasodilation in the lung. J Thorac Cardiovasc Surg. 1996;111:1248-56. Schultz JM, Karamlou T, Swanson J, Shen I, Ungerleider RM. Hypothermic low-flow cardiopulmonary bypass impairs pulmonary and right ventricular function more than circulatory arrest. Ann Thorac Surg. 2006;81: 474-80. Dent CL, Spaeth JP, Jones BV, Schwartz SM, Glauser TA, Hallinan B, et al. Brain magnetic resonance imaging abnormalities after the Norwood procedure using regional cerebral perfusion. J Thorac Cardiovasc Surg. 2006; 131:190-7. Mahle WT, Visconti KJ, Freier MC, Kanne SM, Hamilton WG, Sharkey AM, et al. Relationship of surgical approach to neurodevelopmental outcomes in hypoplastic left heart syndrome. Pediatrics. 2006;117:e90-7. Saiki H, Kurishima C, Masutani S, Tamura M, Senzaki H. Impaired cerebral perfusion after bilateral pulmonary arterial banding in patients with hypoplastic left heart syndrome. Ann Thorac Surg. 2013;96:1382-8. Ohye RG, Sleeper LA, Mahony L, Newburger JW, Pearson GD, Lu M, et al. Comparison of shunt types in the Norwood procedure for single-ventricle lesions. N Engl J Med. 2010;362:1980-92. Porras D, Brown DW, Marshall AC, Del Nido P, Bacha EA, McElhinney DB. Factors associated with subsequent arch reintervention after initial balloon aortoplasty in patients with Norwood procedure and arch obstruction. J Am Coll Cardiol. 2011;58:868-76. Davies RR, Radtke WA, Klenk D, Pizarro C. Bilateral pulmonary arterial banding results in an increased need for subsequent pulmonary artery interventions. J Thorac Cardiovasc Surg. 2014;147:706-12.
Key Words: Hypoplastic left heart syndrome, hybrid, shunt
Readers who found these articles interesting may also like to read the following papers found in recent and future issues of our sister publications, Seminars in Thoracic and Cardiovascular Surgery and Operative Techniques in Thoracic and Cardiovascular Surgery! Congenital Heart Disease Current Readings: Jeffrey Jacobs. Long Term Management of Patients Undergoing Successful Pediatric Cardiac Surgery. Semin Thorac Cardiovasc Surg. Summer 2014;26(2):132-144. Current Readings: Brian Kogon. Pulmonary Valve Replacement for Pulmonary Valve Insufficiency in Formerly Repaired Tetralogy of Fallot Patients. Semin Thorac Cardiovasc Surg. Expected publication August 2015. Original Submission: Meena Nathan. Technical Performance Score as a predictor for post discharge reintervention in valve sparing Tetralogy of Fallot repair. Semin Thorac Cardiovasc Surg. 2014;26(4):297-303. Editorial Commentary: Jonathan Chen. Technical performance anxiety: utility of the Technical Performance Scale in predicting later intervention after repair of Tetralogy of Fallot. Semin Thorac Cardiovasc Surg. 2014;26(4):304-305. Shi-Joon Yoo. MRI as a decision making tool in congenital heart disease surgery. Oper Tech Thorac Cardiovasc Surg. Summer 2014;19(2):152-163. Pascal Vouhe. Valve Sparing Konno and HOCM in Children. Oper Tech Thorac Cardiovasc Surg. Summer 2014;19(2):164-178. Jose Pedro da Silva. Pulmonary root translocation for anatomical repair of congenitally corrected transposition of the Great Arteries. Oper Tech Thorac Cardiovasc Surg. 2014;19(3):304-323.
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
357
CHD
Nassar et al
Second stage after initial hybrid palliation for hypoplastic left heart syndrome: Arterial or venous shunt? Mohamed S. Nassar, PhD, FRCS,a,b,c Srinivas A. Narayan, MD, MRCPCH,b Andrew Nyman, MRCPCH,b Caner Salih, MD, FRCS,b Conal B. Austin, MD, FRCS,b David Anderson, MD, FRCS,a,b and Tarique Hussain, MD, PhDa,b ABSTRACT Objective: Hybrid palliation for hypoplastic left heart syndrome has been developed as an alternative to neonatal Norwood surgery. At the second stage, a source of pulmonary blood flow has to be established. Options include an arterial modified Blalock–Taussig or a venous superior cavopulmonary shunt. Methods: We retrospectively reviewed patients who received second-stage palliation after the initial hybrid. Patients were stratified according to the source of pulmonary blood supply into the arterial shunt (n ¼ 17 patients) or venous shunt (n ¼ 26 patients).
CHD
Results: Age and weight at second stage were lower in the arterial group (85 [45268] days vs 152.5 [61-496] days, P ¼ .001 and 3.6 [2.7-9.4] kg vs 5.1 [2.97-9.4] kg, P ¼ .001, respectively). All recorded surgical times were shorter in the arterial group. Mechanical ventilation and intensive care stay were shorter in the venous group (5.82 [2.01-14.9] days vs 2.42 [0.56-13.67] days, P ¼ .005 and 8.5 [3.623.7] vs 5.75 [0.8-17.6] days, P ¼ .036, respectively) There was no difference in mortality (2/17 vs 5/26; P ¼ .685) or incidence of complications between the 2 groups. There was a tendency toward a higher need for intervention in the immediate postoperative period in the venous group, but this did not reach significance (6/17 vs 13/26, P ¼ .342). The arterial group has shown better development of the branch pulmonary arteries with a higher lower lobe index (158.38 39.43 mm2/m2 vs 113.33 43.96 mm2/m2, respectively, P ¼ .037). Conclusions: Both arterial and venous shunts are viable options with mortality and morbidity results comparable to those in the literature. The arterial shunt pathway (2-stage Norwood I) may offer better pulmonary arterial growth than the venous shunt (comprehensive/combined Norwood I and II). (J Thorac Cardiovasc Surg 2015;150:350-7)
From the aDepartment of Cardiovascular Imaging, Division of Imaging Sciences and Biomedical Engineering, King’s College, London, United Kingdom; bEvelina London Children Hospital, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom; and cDepartment of Cardiothoracic Surgery, Alexandria University, Alexandria, Egypt. Financial support was received from the Department of Health via the National Institute for Health Research comprehensive Biomedical Research Centre award to Guy’s & St Thomas’ NHS Foundation Trust in partnership with King’s College London and King’s College Hospital NHS Foundation Trust. This work was also funded by the British Heart Foundation Award RE/08/003. Received for publication Jan 27, 2015; revisions received March 28, 2015; accepted for publication April 18, 2015; available ahead of print June 6, 2015. Address for reprints: Mohamed S. Nassar, PhD, FRCS, Paediatric Cardiology and Cardiac Surgery, Evelina London Children’s Hospital, Guy’s & St Thomas’ NHS Foundation Trust, Westminster Bridge Road, London SE1 7EH, UK (E-mail: [email protected]). 0022-5223/$36.00 Copyright Ó 2015 by The American Association for Thoracic Surgery http://dx.doi.org/10.1016/j.jtcvs.2015.04.039
350
Schematic representation of overall outcomes of patients. Central Message This observational study show that a 2-stage Norwood I seems to be noninferior to a comprehensive Norwood I and II and may give better pulmonary arterial growth. Perspective Hybrid palliation for hypoplastic left heart syndrome is a relatively new technique. Our results have shown that a 2-stage Norwood I is a valid option compared with the comprehensive Norwood II. Over the years, we have developed criteria that can help to define which patient would benefit from a 2-stage Norwood I.
See Editorial page 286.
Originally described to reduce the risk of neonatal surgery, hybrid palliation for hypoplastic left heart syndrome (HLHS) has been introduced as an alternative to the Norwood procedure, aiming to defer a more complex surgery to a later date with an older child who can withstand major operation with relatively lower risk.1 Since 2005, it has been our institutional policy to use the hybrid operation in patients with HLHS presenting with low birth weight (<2.5 kg), postnatal collapse, restrictive atrial septum, or possible biventricular repair. The second stage after the initial hybrid involves construction of the aortic arch with establishment of pulmonary blood supply. A superior cavopulmonary anastomosis has been the pulmonary blood flow source of choice for this (ie, combined/comprehensive I and II as the second procedure), thus avoiding the transition from an arterial to a venous shunt seen in the traditional Norwood pathway.2
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
Abbreviations and Acronyms AVV ¼ atrioventricular valve HLHS ¼ hypoplastic left heart syndrome MRI ¼ magnetic resonance imaging PA ¼ pulmonary artery POD ¼ postoperative day RPA ¼ right pulmonary artery SCPC ¼ superior cavopulmonary connection Concerns have been raised regarding this approach considering the significant challenge in the reconstruction of the pulmonary arteries (PAs) at the second stage after initial hybrid palliation.3 In addition, reports have shown lower branch PA size in the hybrid pathway compared with the Norwood pathway.4 Some centers have reported their experience with initial PA banding and duct patency maintenance with prostaglandin infusion followed by the Norwood I intervention, the so-called rapid 2-stage Norwood operation, for high-risk patients.5-7 Others have shifted their practice from a venous type of shunt to an arterial shunt after initial banding and maintenance of duct patency.8,9 We currently have a mixed practice of arterial or venous shunt. We reserve the arterial shunt simply because of the surgeon’s preference or for those patients with suspected increased pulmonary vascular resistance, such as those developing a restrictive interatrial septum; those presenting with an indication for an early second stage, such as tight bands or retrograde coarctation; those with anomalous pulmonary venous drainage, in whom intervention on the superior vena cava or longer complex procedure is anticipated; those in whom better development of branch PA is required, as with hypoplastic PAs or interrupted IVC; and those who require a bail out of a high-pressure cavopulmonary anastomosis. On the other hand, in the absence of any of the previous indications, we tend to allocate all other patients, especially those with aberrant subclavian artery, to a venous type of shunt. In the current study, we compare these 2 cohorts of patients in terms of second-stage results, morbidity, mortality, and candidacy for Fontan completion. MATERIALS AND METHODS This is a retrospective study of a single institution. After institutional review board approval, the department database (Heartsuite XP 3.9.14, Systeria, Glasgow, UK) was used to identify patients who reached the second-stage palliation after an initial hybrid between the start of the program in November 2005 and November 2014. Those who underwent biventricular repair, before the second stage, were excluded. Patients were divided into 2 groups according to the source of pulmonary blood flow: Group A received a systemic modified Blalock–Taussig shunt, and group B received a venous hemi-Fontan shunt. Not all patients were preassigned to either group at the time of hybrid. Postoperatively, patients received brain magnetic resonance imaging (MRI) or screening of vocal cords only if clinically indicated.
SURGICAL TECHNIQUE We have previously reported our surgical technique for hybrid and arch reconstruction.10,11 At the time of the second-stage palliation, an arterial or venous shunt is constructed as the source of pulmonary blood flow. The arterial shunt is usually constructed between the first branch of the arch and the right pulmonary artery (RPA), and the venous shunt is a hemi-Fontan type of superior cavopulmonary anastomosis constructed in most cases under a period of total circulatory arrest. The ductal stent is usually peeled off the descending aorta after resection of the isthmus. In patients in whom the stent has been in place for a long time, the part of the descending aorta with the embedded stent is usually resected. The Sano-type right ventricle to PA shunt was not performed in any patients because of the institutional preference to avoid ventriculotomy of the systemic ventricle. We tend not to patch the site of banding because of the friable nature of tissue. Instead, we free all adventitial bands and stretch the vessel from within. Patch material (a pulmonary homograft patch) is used only if the arterial wall is injured while breaking adhesions or the band has been in place for a long time and stretching was insufficient to dilate the constriction point. The PA stump is usually closed directly, and in a few cases, a small patch is used to close the stump to avoid branch PA distortion. At the time of Fontan completion, a 4-mm fenestrated lateral tunnel Fontan is constructed. Any additional intervention on the atrioventricular valve (AVV), arch, or branch PA was recorded. ASSESSMENT OF VENTRICULAR AND ATRIOVENTRICULAR VALVE FUNCTION Patient records were reviewed. Echocardiographic qualitative assessments of ventricular function and degree of AVV regurgitation before operation and hospital discharge were recorded. For comparison, ventricular function was graded as 1 for normal/mildly depressed, 2 for mildmoderate/moderately depressed, and 3 for moderatesevere/severely depressed. AVV regurgitation was graded as 1 for trivial/mild, 2 for mild-moderate/moderate, and 3 for moderate-severe/severe regurgitation. PRE-FONTAN ASSESSMENT Currently, MRI is our investigation of choice for single ventricle palliation.12 Our MRI technique has been reported.10 The following data were recorded: 1. Indexed lower lobe area of RPA and left pulmonary artery (LPA), because the lower lobe index was used to compare branch PA growth between the 2 groups. 2. Differential flow between both branch PAs. 3. PA pressure (measured through a superior vena cava line at the time of MRI).
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
351
CHD
Nassar et al
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
4. Any arch distortion or narrowing was recorded. 5. End-diastolic volume indexed to body surface area. 6. Ejection fraction as an indicator of single ventricle function. 7. Oxygen saturation at the time of MRI study. Statistics Data were summarized as mean standard deviation or as median and range for continuous variables. For categoric variables, number and percentage were used. The distribution of continuous and scale variables was tested by using the Kolmogorov–Smirnov test for normality. Variables with normal distribution were compared with independent samples t test. Mann–Whitney U test was used for variables with non-normal distribution. Categoric variables were compared by the Fisher exact test. IBM SPSS version 22 (IBM, Armonk, NY) was used for all statistical calculations.
CHD
RESULTS During the study period, a total of 146 Norwood and 79 hybrid procedures were performed in our center. Of those 79 patients, 11 were excluded because they had biventricular repair. A total of 43 patients have reached the second stage of univentricular palliation. Seventeen patients received an arterial shunt (group A), and 26 patients received a venous shunt (group B). During the first half of our experience (2005-2009), we tried to push all patients to the comprehensive stage (venous shunt), with 9 patients receiving a venous shunt and only 2 patients receiving an arterial shunt. Because our experience grew overtime, we developed the previously mentioned indications for arterial shunt. Of the 17 patients who received an arterial shunt, 4 had anomalous systemic venous drainage, 2 had partial anomalous pulmonary venous drainage, 5 had developed restrictive inter-atrial septum, 1 had retrograde coarctation, 1 had narrowing of ductal stent, 1 had a small LPA, and 1 had a bail out. Preoperative Data Group A patients had significantly lower age and weight at the time of operation. There was no significant difference in preoperative echocardiography rank between the 2 groups. Data are summarized in Table 1. Surgical Data At the second stage, all recorded times were significantly shorter in the arterial shunt group. There was no difference in the need for branch PA plasty or for a second run. There was a higher need for delayed sternal closure in the arterial group, but none of the patients have shown any evidence of mediastinitis. Two patients in the venous shunt group received an additional arterial shunt because of persistent 352
Nassar et al
TABLE 1. Preoperative characteristics Arterial shunt Venous shunt P (17 patients) (26 patients) value Age (d) 85 (45-268) Weight (kg) 3.6 (2.7-9.4) Diagnosis Aortic atresia/mitral atresia 5 Aortic atresia/mitral stenosis 1 Aortic stenosis/mitral atresia 2 Aortic stenosis/mitral stenosis 6 0 Aortic stenosis, endomyocardial elastofibrosis Tricuspid atresia, transposition 1 of the great arteries Unbalanced atrioventricular 0 septal defect Aortic atresia, large inlet VSD, 1 hypoplastic arch Other diagnoses 1 PAPVD 2 ASVD Left SVC 3 Interrupted IVC 1 Hypoplastic LPA 1 Primary indication for hybrid: Low birth weight 9 Postnatal collapse 2 Restrictive atrial septum 2 Possible biventricular repair 2 Others 2 Echocardiography function: Normal/mildly depressed 13 Mild-moderate/moderately 3 depressed Moderate-poor/poor 1 Echocardiography AVV incompetence: Normal/mildly 13 Mild-moderate/moderate 4 Moderate-severe/severe 0
152.5 (61-496) 5.1 (2.97-9.4)
.001 .001
5 2 1 3 6
.481 1.000 .552 .122 .066
1
1.000
2
.511
2
1.000
4 0
.633 .151
2 0 1
.369 .395 1.000
8 5 2 6 5
.205 .685 1.000 .446 .685 .683
16 8 2 1.000 19 7 0
VSD, Ventricular septal defect; PAPVD, partial anomalous pulmonary venous drainage; ASVD, anomalous systemic venous drainage; SVC, superior vena cava; IVC, inferior vena cava; LPA, left pulmonary artery; AVV, atrioventricular valve.
low saturation, and both died. There was 1 intraoperative crossover in each group. An arterial shunt was changed to a venous shunt because of poor function and evidence of coronary steal, but the patient did not survive. One patient with a venous shunt required takedown to an arterial shunt because of unacceptably high PA (superior vena cava) pressure; this patient survived and subsequently had successful Fontan completion. Four patients (3 in group A and 1 in group B) required a second bypass run, within the same procedure, for arch redo. One patient in group A required arch redo on postoperative day (POD) 1 because of a kink in the arch that was
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
TABLE 2. Second-stage surgical intervention data Procedure
Arterial shunt (17 patients)
Venous shunt (26 patients)
P value
CPB (min) Crossclamp (min) Circulatory arrest (min) Branch PA plasty AVV repair Delayed chest closure Second run
183.53 85.21 98.06 53.50 20 (3-137) 4 0 13 4
201.38 51.39 122.46 34.43 47 (0-89) 5 1 10 6
.048 .003 .006 1.000 1.000 .027 1.000
Second run refers to reinstitution of cardiopulmonary bypass within the same procedure. Delayed chest closure was unplanned and considered in case of bleeding, hemodynamic instability, or desaturation. CPB, Cardiopulmonary bypass; PA, pulmonary artery; AVV, atrioventricular valve.
discovered on return to the pediatric intensive care unit. Data are summarized in Table 2. Pediatric intensive care unit stay and duration of ventilation were shorter in the venous shunt group. At hospital discharge, there was no significant difference in echocardiography findings. Group A had significantly higher oxygen saturation at discharge. Data are summarized in Table 3. TABLE 3. Postoperative data Arterial shunt (17patients)
Venous shunt (26 patients)
PICU stay: Length of stay (d) 8.5 (3.6-23.7) 5.75 (0.8-17.6) Mechanical ventilation (d) 5.82 (2.01-14.9) 2.42 (0.56-13.67) Need for inotropes (d) 7 (3-23) 5 (0.8-17) Mortality 2 (12%) 5 (19%) Morbidity Brain injury 5 (29%) 8 (31%) Necrotizing enterocolitis 1 2 Left vocal cord palsy 1 2 Chylothorax 1 2 Echocardiography findings: Function: Normal/mildly depressed 12 13 Mild-moderate/moderate 3 8 Moderate-poor/poor 2 5 AVV incompetence: Normal/ mild 10 16 Mild-moderate/moderate 7 5 Moderate-severe/severe 0 5 Reintervention Diagnostic catheter 3 8 Catheter intervention 3 3 Surgery 2 6 Cardiac MRI/CT 3 4 Total interventions 11 21 No. of patients 6 (35%) 13 (50%) Hospital stay (d) 15 (8-30) 13 (1-57) Oxygen saturation at 85 (75-92) 80 (50-89) hospital discharge (%)
P value .036 .005 .10 .685 1.000 1.000 1.000 1.000 .246
.088
.480 .666 .446 1.000 .295 .342 .471 .014
PICU, Pediatric intensive care unit; AVV, atrioventricular valve; MRI, magnetic resonance imaging; CT, computed tomography.
Survival There were 2 early mortalities in the arterial shunt group. The first patient had a kink of the reconstructed arch that was discovered on return to the pediatric intensive care unit, and this patient underwent reoperation. She was extubated later, with echocardiography confirming good results of arch reconstruction, but on POD 9 she had a cardiac arrest during transfer for MRI of the brain. The second patient had a thrombosed 4-mm shunt, underwent emergency stenting of the shunt, and the branch PA was carried out, but his condition continued to deteriorate and he died on POD 15. In the venous shunt group, there were 5 early mortalities. Two of those were very early in the study period. The first had an RPA kink with formed thrombus in the superior vena cava. This patient had revision of the venous shunt and an additional arterial shunt, but a large cerebral infarction developed and the patient died on POD 9. The second patient required bilateral PA stenting, and later, the venous shunt was taken down and replaced with an arterial shunt, but the function deteriorated and this patient died on POD 16. The third patient had persistent low saturations, and an additional arterial shunt was constructed that thrombosed with thrombus extending to the PAs, and this patient died on POD 1. The fourth patient was returned to the operating room for bleeding, but cardiac function deteriorated and a significant cerebral infarct developed; the patient died on POD 17. The fifth patient was admitted 2 months after the initial hybrid with desaturation and poor right ventricular function. A rapid Norwood one I was planned but converted to a venous shunt because of volume load, ST depression, and significant AVV regurgitation at the end of the operation; his cardiac function deteriorated, and he died on POD 2. There was 1 late mortality in a patient in the venous shunt group who had an RPA kink that was addressed at the time of the Fontan completion. However, cardiac function deteriorated after surgery, subsequent multiorgan failure subsequently developed, and the patient died. Reintervention In the immediate postoperative period, apart from 1 arch reintervention, the branch PA or shunt revision was the target of all reinterventions. There was a trend toward a higher need for imaging and intervention in the venous shunt group. However, this did not reach statistical significance; data are summarized in Table 3. In the arterial shunt group, 14 of the 15 survivors have received a superior cavopulmonary connection (SCPC), with 1 patient awaiting surgery. Concomitant procedures at the time of SCPC included AVV repair in 2 patients, LPA plasty in 3 patients, and arch redo in 1 patient. Recorded times included cardiopulmonary bypass (96.2 25.35 minutes), crossclamp time (48.5 23.24 minutes), and circulatory arrest (20.4 13.59 minutes). At the stage of SCPC, there were no mortalities,
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
353
CHD
Nassar et al
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
TABLE 4. Magnetic resonance imaging findings at pre-Fontan assessment Arterial shunt (9 patients)
Venous shunt (13 patients)
EF (%) 58 (43-67) 65.5 (44-70) iEDV (mL/m2) 92.78 19.37 75.30 25.26 Differential 73/27 (45-90/10-55) 67/33 (50-85/15-50) RPA/LPA flow 114.40 28.64 69.55 25.23 iRLL area (mm2/m2) iLLL area (mm2/m2) 43.98 18.74 43.78 24.07 Lower lobe index 158.38 39.43 113.33 43.96 (mm2/m2) PAP (mm Hg) 13 2.12 12.0 1.41 Oxygen Saturation at 80.67 4.71 81.89 4.29 time of MRI
P value .250 .110 .650 .003 .930 .037 .257 .520
EF, Ejection fraction; iEDV, indexed end-diastolic volume; RPA, right pulmonary artery; LPA, left pulmonary artery; iRLL, indexed right lower lobe area; iLLL, indexed left lower lobe; PAP, pulmonary artery pressure; MRI, magnetic resonance imaging.
and the only recorded morbidity was a prolonged chylothorax in 1 patient. Before the Fontan completion, 3 patients in the arterial group and 1 patient in the venous group required catheter intervention on the branch PA. Overall, there was no significant difference in the number of patients who required intervention on the branch PA (41% vs 34%, respectively).
Pre-Fontan Magnetic Resonance Imaging Nine patients in group A and 13 patients in group B had pre-Fontan MRI studies. Lower lobe index measurement was possible in all but 3 patients in group B (because of quality of images in 2 and 1 was excluded because it had a single lung Fontan because of congenital unilateral lung hypoplasia). There was no evidence of arch obstruction in any patient at the pre-Fontan assessment. Results are shown in Table 4. Fontan Completion In the arterial group, 4 of the 15 survivors had Fontan completion, and in the venous group, 9 of 21 patients had Fontan completion (Figure 1). Additional procedures at the time of Fontan completion were 1 PA plasty in the venous group and 1 AVV repair in the arterial shunt group. There was no statistically significant difference between the 2 groups at the time of Fontan completion considering surgical times and postoperative stay. DISCUSSION Data comparing the 2 possible pathways after the initial hybrid are limited. Having a mixed practice, arterial or venous shunt at the second stage, has given us the chance
CHD FIGURE 1. Schematic representation of overall outcomes of patients. SCPC, Superior cavopulmonary connection.
354
Nassar et al
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
to compare these 2 cohorts. Our results have shown points of strength and weakness for each option. There was no significant difference in mortality between the 2 groups (11.7% vs 19.2%). The higher incidence of mortality in the combined group is in part because 2 of the 5 deaths were in our very early practice, perhaps representing a learning curve, and it is possible that this relates to evolution of a strategy for assigning the appropriate shunt type for each patient. Retrospective review of our patient cohort shows that some of the earlier cohort who underwent venous shunt pathway may have been assigned an arterial shunt in the current era (1 early mortality in the venous group, who had a restrictive septum). Our mortality results are still comparable to most of the reported series,13,14 bearing in mind that in our center, the hybrid procedure is reserved for higher-risk, more complex, and unstable patients in whom a Norwood procedure would carry an unacceptable high risk. A previous report from our center has confirmed the higher risk of our hybrid cases using a modified comprehensive Aristotle score.15 Three of the 5 mortalities in the venous shunt group were related to the venous shunt, supporting the notion that reconstruction of the PAs is a significant challenge at this stage. This was further reinforced by the difference in all recorded surgical times. There was 1 intraoperative crossover in each group. We thought it was more appropriate to assign patients to each group according to the type of shunt they received at the end of the operation because the complications are often related to the hemodynamic state produced by the shunt. Assigning those 2 patients on the basis of intention to treat would not affect the final mortality results (2 vs 5 for presented data [P ¼ .68] and 3 vs 4 for intention to treat analysis [P ¼ 1.00]). The higher incidence of unplanned delayed sternal closure in the arterial shunt group could be related to the early hemodynamic instability caused by the high pulmonary flow state, with possible right ventricle volume overload or coronary underperfusion, in comparison with the more stable situation and volume off-loading advantage of the venous shunt. On the other hand, this hemodynamic stability was at the expense of significantly lower saturations in the venous shunt group, proven by the higher need for PA imaging and lower saturations at discharge. This postoperative hypoxia could be due to not only the technically demanding reconstruction of the PAs but also the pulmonary vascular resistance, which may be elevated after a long bypass run.16,17 The effect of this postoperative hypoxia on neurology was not evident when major neurologic complications were compared between the 2 groups. Although we do not perform routine postoperative brain MRI, we have a low threshold for MRI whenever the clinical scenario (eg, change of motor tone, convulsions) suggests possible brain injury. In both groups, 30% of patients had MRI-proven brain abnormalities after clinically detected abnormal neurology in the immediate
postoperative period. This relatively high incidence is comparable to other reports studying the neurologic development in patients with HLHS. Reports on brain MRI studies in patients with HLHS show up to 73% incidence of new or worsened lesions.18 Moreover, it seems that the difference in oxygen saturation between the 2 groups became less evident with time, because at the time of pre-Fontan MRI, there was no significant difference between the 2 groups. Long-term neurodevelopmental delay has been demonstrated in patients with HLHS.19 Saiki and colleagues20 studied cerebral perfusion after bilateral banding. Their results were in favor of a shorter duration between the initial hybrid and the second stage. They have shown that in patients posthybrid, cerebral blood flow worsens over time.20 We could not find data on the incidence of neurologic complications in the few reports comparing post-hybrid Norwood I and comprehensive Norwood I and II after bilateral PA banding.8,14 Whether the longer duration until second-stage palliation and the lower postoperative saturations in the venous group would induce subtle brain changes making a difference between both groups in the long run requires further investigation. When it comes to the need for reintervention after the second stage, one of the disadvantages of the arterial shunt is the need for an additional surgery to establish superior cavopulmonary anastomosis as an intermediate step before the Fontan circulation. Although there was no difference on the rate of reintervention after second-stage hospital discharge (excluding SCPC stage), there was a higher, although not statistically significant, need for intervention in the venous shunt group immediately after surgery. All patients with an arterial shunt who went through the additional stage survived with very low morbidity and a short hospital stay. Arch reconstruction is a challenge in patients with HLHS. The presence of a ductal stent adds to the challenge of the second stage after a hybrid. Shorter duration between initial hybrid palliation and second stage (as in arterial shunt group) would theoretically make stent extraction, and subsequently arch reconstruction, less challenging compared with the well-embedded stent seen with longer interval between the initial hybrid and the second-stage palliation (in the venous shunt group). Arch obstruction is a known complication after arch reconstruction in patients with HLHS; 5 of our patients required a second bypass run for redo arch reconstruction. There was no difference in arch reintervention between the 2 groups; only 1 patient (1/43, 2.3%) required arch reintervention after hospital discharge. We have a low incidence of arch obstruction compared with reports from other centers. Results from the single ventricle trial and the Boston group have shown an incidence of arch obstruction up to 29%.21,22 We have previously reported our arch reconstruction technique,10 and we believe that the use of pulmonary homograft patch and the extension
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
355
CHD
Nassar et al
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
of the reconstruction far down into the descending aorta can avoid the future development of arch obstruction.
CHD
Pulmonary Arteries Baba and colleagues3 commented on the significant challenge in reconstructing the PAs at the comprehensive stage II palliation and in achieving subsequent reasonable PA architecture for Fontan completion. They have mentioned that the mechanism of branch PA stenosis mainly stems not only from anatomic distortion of the branch PA but also from the difficulty in geometric arrangement between the newly reconstructed aorta and the branch PA.3 The rigid graft and the higher systemic pressure in the arterial shunt can overcome any mild degree of branch PA or PA stump distortion, which would be a challenge to the compressible lowpressure venous shunt. This is supported by our results, which have shown a higher need for addressing the branch PA in the immediate postoperative period in the venous shunt group, compared with later intervention in the arterial shunt group (usually at the time of SCPC). Review of the mortality data presented supports the notion that significant residual branch PA narrowing requires prompt reintervention and consideration of anticoagulation. This inherent problem of the comprehensive stage II has shifted some centers to a rapid 2-stage N I operation (primary branch PA banding and maintenance of ductal patency with prostaglandin followed by an early Norwood procedure). When performing an SCPC after an arterial shunt, the branch PAs, in particular the LPA, are fixed within the mediastinum. This makes further distortion less likely while constructing the venous shunt compared with the comprehensive stage II. A surgical strategy of routine mobilization from hilum to hilum is used at some centers with more liberal use of patching material. The merits of that approach have not been assessed in this article. Our approach has been to use the hemi-Fontan procedure to achieve better central reconstruction and avoid patching. Likewise, the merit of performing a hemi-Fontan for this reason over a bidirectional Glenn, which may reduce bypass time, is beyond the scope of this article. Davies and colleagues23 have shown that there is a higher rate of PA intervention if bands were left in place for more than 90 days because early removal of branch PA bands would avoid scarring and narrowing at the band sites. We have not seen any difference in surgical branch PA plasty at the time of the second stage. As we have mentioned, we tend to avoid patching the branch PAs whenever possible. Previous reports on the development of branch PAs in the hybrid pathway after the comprehensive stage II have confirmed smaller vessels and a greater need for intervention when compared with the Norwood pathway.4 Ota and colleagues9 compared branch PA development after a rapid 2-stage N I operation to a classic Norwood pathway, and their results showed no difference. Our results are in 356
Nassar et al
harmony with previous studies showing a significantly higher lower lobe index in the arterial shunt group. Sakamoto and colleagues8 showed a tendency toward better branch PA development in the arterial shunt group in their series. However, Davies and colleagues14 have failed to demonstrate this in their cohort. However, it is worth mentioning that this report contains only a brief comment on the absolute diameter of branch PA on conventional angiography.14 In our series, there was a significant difference in branch PA development at the time of pre-Fontan assessment. However, this difference in vessel size was not reflected on the measured PA pressure. Therefore, the effect on long-term outcome is still unknown. Finally, the aim of the palliative surgery for patients with HLHS is to reach the Fontan circulation. Fontan candidacy could be compromised by deterioration of ventricular function, AVV malfunction, or distortion in branch PA geometry. The only difference in our series was the size of the branch PAs. None of the patients in our cohort were refused Fontan completion. Study Limitations The retrospective single-center nature of the study and the relatively small number of cases are the main limitations of the study. This was a retrospective study, so patients were not randomized to either pathway. A larger number of patients and a longer follow-up may reveal more significant differences between the 2 pathways. Because of the consistent operative management of these patients, no inferences could be drawn about potentially important determinants of outcome, such as the type of arch reconstruction or SCPC, among others. These important limitations call for large multicenter studies to better define the outcome of different management strategies. CONCLUSIONS The option of an arterial shunt (2-stage Norwood I) does not seem to be inferior to the commonly used venous shunt (comprehensive Norwood I and II). There is growing evidence to suggest that early intervention after bilateral branch PA banding is favorable, which is in favor of the arterial shunt pathway. Conflict of Interest Statement Dr Hussain reports grants from NIHR comprehensive Biomedical Research Centre award and grants from British Heart Foundation award RE/08/003 during the conduct of the study. All other authors have nothing to disclose with regard to commercial support. References 1. Gibbs JL, Wren C, Watterson KG, Hunter S, Hamilton JR. Stenting of the arterial duct combined with banding of the pulmonary arteries and atrial septectomy or septostomy: a new approach to palliation for the hypoplastic left heart syndrome. Br Heart J. 1993;69:551-5.
The Journal of Thoracic and Cardiovascular Surgery c August 2015
Congenital Heart Disease: Hypoplastic Left Heart Syndrome
2. Michel-Behnke I, Akintuerk H, Marquardt I, Mueller M, Thul J, Bauer J, et al. Stenting of the ductus arteriosus and banding of the pulmonary arteries: basis for various surgical strategies in newborns with multiple left heart obstructive lesions. Heart. 2003;89:645-50. 3. Baba K, Kotani Y, Chetan D, Chaturvedi RR, Lee KJ, Benson LN, et al. Hybrid versus Norwood strategies for single-ventricle palliation. Circulation. 2012; 126(11 Suppl 1):S123-31. 4. Dave H, Rosser B, Knirsch W, Hubler M, Pretre R, Kretschmar O. Hybrid approach for hypoplastic left heart syndrome and its variants: the fate of the pulmonary arteries. Eur J Cardiothorac Surg. 2014;46:14-9. 5. Gomide M, Furci B, Mimic B, Brown KL, Hsia TY, Yates R, et al. Rapid 2-stage Norwood I for high-risk hypoplastic left heart syndrome and variants. J Thorac Cardiovasc Surg. 2013;146:1146-52. 6. Guleserian KJ, Barker GM, Sharma MS, Macaluso J, Huang R, Nugent AW, et al. Bilateral pulmonary artery banding for resuscitation in high-risk, single-ventricle neonates and infants: a single-center experience. J Thorac Cardiovasc Surg. 2013;145:206-14. 7. Russell RA, Ghanayem NS, Mitchell ME, Woods RK, Tweddell JS. Bilateral pulmonary artery banding as rescue intervention in high-risk neonates. Ann Thorac Surg. 2013;96:885-90. 8. Sakamoto T, Harada Y, Kosaka Y, Umezu K, Yasukochi S, Takigiku K, et al. Second-stage palliation after bilateral pulmonary artery bands for HLHS and its variants–which is better, modified Norwood or Norwood plus bidirectional Glenn? World J Pediatr Congenit Heart Surg. 2011;2:558-65. 9. Ota N, Murata M, Tosaka Y, Ide Y, Tachi M, Ito H, et al. Is routine rapid-staged bilateral pulmonary artery banding before stage 1 Norwood a viable strategy? J Thorac Cardiovasc Surg. 2014;148:1519-25. 10. Nassar MS, Bertaud S, Goreczny S, Greil G, Austin CB, Salih C, et al. Technical and anatomical factors affecting the size of the branch pulmonary arteries following first-stage Norwood palliation for hypoplastic left heart syndrome. Interact Cardiovasc Thorac Surg. 2015;20:631-5. 11. Venugopal PS, Luna KP, Anderson DR, Austin CB, Rosenthal E, Krasemann T, et al. Hybrid procedure as an alternative to surgical palliation of high-risk infants with hypoplastic left heart syndrome and its variants. J Thorac Cardiovasc Surg. 2010;139:1211-5. 12. Heathfield E, Hussain T, Qureshi S, Valverde I, Witter T, Douiri A, et al. Cardiovascular magnetic resonance imaging in congenital heart disease as an alternative to diagnostic invasive cardiac catheterization: a single center experience. Congenit Heart Dis. 2013;8:322-7. 13. Brescia AA, Jureidini S, Danon S, Armbrecht E, Fiore AC, Huddleston CB. Hybrid versus Norwood procedure for hypoplastic left heart syndrome:
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
contemporary series from a single center. J Thorac Cardiovasc Surg. 2014; 147:1777-82. Davies RR, Radtke W, Bhat MA, Baffa JM, Woodford E, Pizarro C. Hybrid palliation for critical systemic outflow obstruction: neither rapid stage 1 Norwood nor comprehensive stage 2 mitigate consequences of early risk factors. J Thorac Cardiovasc Surg. 2015;149:182-93. Lloyd DF, Cutler L, Tibby SM, Vimalesvaran S, Qureshi SA, Rosenthal E, et al. Analysis of preoperative condition and interstage mortality in Norwood and hybrid procedures for hypoplastic left heart syndrome using the Aristotle scoring system. Heart. 2014;100:775-80. Kirshbom PM, Jacobs MT, Tsui SS, DiBernardo LR, Schwinn DA, Ungerleider RM, et al. Effects of cardiopulmonary bypass and circulatory arrest on endothelium-dependent vasodilation in the lung. J Thorac Cardiovasc Surg. 1996;111:1248-56. Schultz JM, Karamlou T, Swanson J, Shen I, Ungerleider RM. Hypothermic low-flow cardiopulmonary bypass impairs pulmonary and right ventricular function more than circulatory arrest. Ann Thorac Surg. 2006;81: 474-80. Dent CL, Spaeth JP, Jones BV, Schwartz SM, Glauser TA, Hallinan B, et al. Brain magnetic resonance imaging abnormalities after the Norwood procedure using regional cerebral perfusion. J Thorac Cardiovasc Surg. 2006; 131:190-7. Mahle WT, Visconti KJ, Freier MC, Kanne SM, Hamilton WG, Sharkey AM, et al. Relationship of surgical approach to neurodevelopmental outcomes in hypoplastic left heart syndrome. Pediatrics. 2006;117:e90-7. Saiki H, Kurishima C, Masutani S, Tamura M, Senzaki H. Impaired cerebral perfusion after bilateral pulmonary arterial banding in patients with hypoplastic left heart syndrome. Ann Thorac Surg. 2013;96:1382-8. Ohye RG, Sleeper LA, Mahony L, Newburger JW, Pearson GD, Lu M, et al. Comparison of shunt types in the Norwood procedure for single-ventricle lesions. N Engl J Med. 2010;362:1980-92. Porras D, Brown DW, Marshall AC, Del Nido P, Bacha EA, McElhinney DB. Factors associated with subsequent arch reintervention after initial balloon aortoplasty in patients with Norwood procedure and arch obstruction. J Am Coll Cardiol. 2011;58:868-76. Davies RR, Radtke WA, Klenk D, Pizarro C. Bilateral pulmonary arterial banding results in an increased need for subsequent pulmonary artery interventions. J Thorac Cardiovasc Surg. 2014;147:706-12.
Key Words: Hypoplastic left heart syndrome, hybrid, shunt
Readers who found these articles interesting may also like to read the following papers found in recent and future issues of our sister publications, Seminars in Thoracic and Cardiovascular Surgery and Operative Techniques in Thoracic and Cardiovascular Surgery! Congenital Heart Disease Current Readings: Jeffrey Jacobs. Long Term Management of Patients Undergoing Successful Pediatric Cardiac Surgery. Semin Thorac Cardiovasc Surg. Summer 2014;26(2):132-144. Current Readings: Brian Kogon. Pulmonary Valve Replacement for Pulmonary Valve Insufficiency in Formerly Repaired Tetralogy of Fallot Patients. Semin Thorac Cardiovasc Surg. Expected publication August 2015. Original Submission: Meena Nathan. Technical Performance Score as a predictor for post discharge reintervention in valve sparing Tetralogy of Fallot repair. Semin Thorac Cardiovasc Surg. 2014;26(4):297-303. Editorial Commentary: Jonathan Chen. Technical performance anxiety: utility of the Technical Performance Scale in predicting later intervention after repair of Tetralogy of Fallot. Semin Thorac Cardiovasc Surg. 2014;26(4):304-305. Shi-Joon Yoo. MRI as a decision making tool in congenital heart disease surgery. Oper Tech Thorac Cardiovasc Surg. Summer 2014;19(2):152-163. Pascal Vouhe. Valve Sparing Konno and HOCM in Children. Oper Tech Thorac Cardiovasc Surg. Summer 2014;19(2):164-178. Jose Pedro da Silva. Pulmonary root translocation for anatomical repair of congenitally corrected transposition of the Great Arteries. Oper Tech Thorac Cardiovasc Surg. 2014;19(3):304-323.
The Journal of Thoracic and Cardiovascular Surgery c Volume 150, Number 2
357
CHD
Nassar et al