Significance of Low Peak Doppler Velocity in the Proximal Sano Conduit in Hypoplastic Left Heart Syndrome Shyam K. Sathanandam, MD, Ranjit Philip, MD, Andrew Van Bergen, MD, David A. Roberson, MD, Wei Cui, MD, Michel N. Ilbawi, MD, Alexander J. Javois, MD, and Tarek S. Husayni, MD CONGENITAL HEART
The Heart Institute for Children, Advocate Hope Children’s Hospital, Oak Lawn, Illinois; and Department of Pediatrics, University of Tennessee Health Science Center, Memphis, Tennessee
Background. The Sano modification of the Norwood operation is a well-established first step palliation for hypoplastic left heart syndrome (HLHS). Theoretically, the first point of resistance to pulmonary flow should be in the proximal Sano, generating high Doppler flow velocity. Paradoxically, however, some patients have low gradients in the proximal Sano conduit. The objective of this study was to determine the hemodynamic and anatomic significance of low proximal Sano Doppler flow velocity and its clinical implications. Methods. Doppler-derived peak gradients in the proximal Sano conduits were measured in HLHS patients after Norwood-Sano surgery over a 4-year period and confirmed by cardiac catheterization within 2 to 4 weeks. Clinical outcomes of patients with proximal Sano gradients of 30 mm Hg or less (group 1) were compared with patients whose gradient was greater than 30 mm Hg (group 2).
Results. Of the 53 patients, 21 (40%) belonged to group 1. Patients in group 1 had smaller ostial right and left pulmonary artery (PA) diameter (3.2 ± 1.2 mm versus 4.5 ± 1.8 mm, p [ 0.03; and 3.4 ± 1.2 mm versus 5.6 ± 2.1 mm, p [ 0.01) when compared with patients in group 2. Patients (7 of 10) who needed either balloon angioplasty of a distal Sano or proximal branch PA were from group 1 (p [ 0.01). Patients in group 1 had higher rates of unintended PA interventions (33% versus 9%) and complications (48% versus 16%) compared with group 2. Conclusions. Low peak Doppler flow velocity in the proximal Sano correlates with the presence of either distal Sano stenosis or proximal branch PA stenosis. These patients require close follow-up in the interstage period and may need earlier intervention.
B
The concerns of the Sano modification are related to the ventriculotomy, which could contribute to ventricular dysfunction or aneurysm formation, or be a nidus for arrhythmias. Free conduit insufficiency may result in ventricular volume overload and inadequate PA growth, which may be due to the lack of forward flow during diastole [7, 8]. Several case series report improved short-term survival with the Sano modification [1, 3, 9, 10]. Retrospective reviews have demonstrated no early or midterm survival differences between the mBTS or the Sano modification [4, 11–13]. A study using cardiac magnetic resonance showed smaller right pulmonary artery growth and worsening ventricular function in the Sano cohort compared with the mBTS cohort [7]. The Single Ventricle Reconstruction Trial, which is a multicenter, randomized clinical trial, compared the intermediate outcomes of neonates undergoing the mBTS with those of the Norwood-Sano operation [8]. It concluded that neonates who have the Sano modification had smaller PA growth and higher incidence of unintended interventions. Hemodynamically, the first point of high resistance to pulmonary blood flow is in the proximal Sano conduit
etter understanding of anatomy and physiology of patients with hypoplastic left heart syndrome (HLHS) and advances in surgical techniques and postoperative management have contributed to a remarkable improvement in outcomes after the Norwood operation. The classic Norwood operation involved a modified Blalock-Taussig shunt (mBTS) for pulmonary blood flow. An alternative is the Sano shunt, which involves placing a valveless polytetrafluoroethylene conduit from the right ventricle (RV) to the pulmonary artery (PA) to provide pulmonary blood flow [1]. The advantages of the Sano modification (RV-PA conduit) are associated with the absence of diastolic runoff from the systemic circulation to the pulmonary circulation. Hence, it potentially provides a stable immediate postoperative course, increased coronary arterial flow, improved weight gain related to improved splanchnic perfusion, and lower interstage mortality [2–6].
Accepted for publication May 6, 2014. Address correspondence to Dr Philip, Division of Pediatric Cardiology, Le Bonheur Children’s Hospital, Heart Institute, 3rd Flr, 50 N Dunlap St, Memphis, TN 38103; e-mail: [email protected].
Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2014;98:1378–85) Ó 2014 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.05.049
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
Abbreviations and Acronyms BDG HLHS LPA MA-AA mBTS MS-AA MS-AS PA RPA RV S1P
= = = = = = = = = = =
bidirectional Glenn hypoplastic left heart syndrome left pulmonary artery mitral atresia and aortic atresia modified Blalock-Taussig shunt mitral stenosis and aortic atresia mitral stenosis and aortic stenosis pulmonary artery right pulmonary artery right ventricle stage 1 palliation
and should result in a high Doppler flow velocity signal at this point. Paradoxically, however, we observed that a number of patients had low gradients in the proximal Sano conduit. We postulated that a low flow velocity in the proximal Sano conduit is caused by a significant distal obstruction. The purposes of this study were to determine whether the low proximal Sano flow velocity is caused by significant distal obstruction, and to ascertain the hemodynamic, anatomic, and clinical implications of such findings.
Patients and Methods Review of medical records and computerized hospital data was approved by the Institutional Review Board, and the procedures followed were in accordance with institutional guidelines for retrospective record review and protection of patients’ confidentiality. The Institutional Review Board waived the need for patient consent.
Study Population Neonates with HLHS who were managed at our institution during a 4-year period were included in the study. HLHS was defined as the presence of mitral atresia and aortic atresia (MA-AA), mitral stenosis and aortic atresia (MS-AA), or severe mitral stenosis and aortic stenosis (MS-AS). All had mitral valve diameter, aortic valve size Z –4 or less. All cases with additional or other cardiac anomalies were excluded from this study, including HLHS with ventricular septal defect, unbalanced atrioventricular septal defect, double outlet right ventricle with mitral and aortic stenosis or atresia, critical aortic stenosis, endocardial fibroelastosis with normal or large left ventricle, and heterotaxy syndromes requiring a Norwood operation.
Operative Technique for the Sano Conduit Our surgical approach for the Sano conduit involves anastomosing the PA conduit at the partially augmented PA wall early during the cooling phase. During the rewarming phase, a modification of the anastomosis to the RV is made, and the conduit directed leftward to the aorta. A full thickness ventriculotomy is performed to construct a transmural opening of 4.5 mm to 5 mm. After
1379
the length of the polytetrafluoroethylene conduit is fashioned, its anterior wall is reverse beveled. The posterior wall is then sutured to the epicardial surface of the ventriculotomy. A “baseball-plate” tailored bovine pericardial hood is constructed to augment the anterior anastomotic site. This hood is fashioned purposely redundant. It is believed to reduce the diastolic regurgitant flow due to dynamic anterior wall effect and to prevent muscular hypertrophy, shunt intimal hyperplasia, and early stenosis at the ventricular end of the conduit.
Measurement Variables We measured Doppler-derived peak gradient in the proximal Sano conduit in HLHS survivors after Norwood-Sano surgery. Pulse-wave Doppler was performed in a high parasternal short-axis view or, rarely, from a subcostal sagittal view of the Sano conduit. These views provide the best pulse Doppler sampling of the proximal Sano conduit. All patients underwent cardiac catheterization within 2 to 4 weeks of the Doppler study before referral for the bidirectional Glenn (BDG) surgery. We confirmed Doppler-measured peak Sano gradients by cardiac catheterization. Patients who had Doppler proximal Sano gradient of 30 mm Hg or less (group 1) were compared with patients who had Sano gradient greater than 30 mm Hg (group 2). Patients were followed till Fontan completion. The McGoon ratio is the combined diameter of the right pulmonary artery (RPA) and left pulmonary artery (LPA) at the level of the hilum of the lung, to the diameter of the descending thoracic aorta at the level of the diaphragm. We calculated the McGoon ratio using the narrowest PA branch diameter, which is at the level of the ostium of each branch PA and also at the level of the hilum of the lungs. They are called ostial McGoon ratio and hilar McGoon ratio, respectively. The Nakata index is the sum of the cross-sectional areas of the RPA and LPA at the level of the hilum of the lungs, indexed to body surface area. We calculated an ostial Nakata index and a hilar Nakata index. These indexes were calculated in all patients from PA angiograms. Pulmonary angiography was performed by contrast injection in the proximal Sano conduit with the lateral camera in the 90-degree left anterior oblique projection and the anterior-posterior camera in a 45-degree caudal angulation. This view, called the “sleeping-bat view,” provides the best layout of the entire Sano conduit and the branch PAs without any foreshortening. The following were considered unintended interventions before the BDG surgery: balloon angioplasty or stent implantation in the Sano conduit or the branch PA, mBTS before the BDG, or PA arterioplasty during the BDG. Mortality, extracorporeal membrane oxygenation (ECMO) support or cardiac transplantation after the BDG were considered unintended complications after BDG. Similarly, PA intervention before or during Fontan or need for a fenestrated Fontan were considered unintended interventions for the Fontan surgery. After Fontan completion, mortality, ECMO support, cardiac
CONGENITAL HEART
Ann Thorac Surg 2014;98:1378–85
1380
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
Ann Thorac Surg 2014;98:1378–85
Fig 1. Flowchart showing patient distribution and outcomes. (mBTS ¼ modified Blalock-Taussig shunt; HLHS ¼ hypoplastic left heart syndrome; PA ¼ pulmonary artery.)
CONGENITAL HEART
transplantation, prolonged chest tube drainage, and protein-losing enteropathy (PLE) were considered unintended complications. The primary outcome of the study was the association of the Doppler-derived peak proximal Sano gradient and the need for unintended interventions. The secondary outcome was the association of the Doppler-derived peak proximal Sano gradient and unintended complications.
Statistical Analysis Data are expressed as mean SD for continuous variables and as median and range for categorical variables. The Fisher exact test was used to detect significant differences between groups for categorical variables. Unpaired one-tailed or two-tailed t tests were performed for comparison of continuous variables between groups. Survival analysis was performed using the Kaplan-Meier product limit method for all subgroups. Cox regression on survival was then performed to see whether there was a real difference when other confounding factors were accounted for. Statistical relationship between two variables was determined and expressed using Pearson’s correlation coefficient. The software IBM SPSS 15.0.1 for Windows (IBM Corporation, Armonk, NY) was used for statistical analysis.
Results Neonates with HLHS over a 4-year-period were included in the study. The male to female ratio was 2:1. Forty-two
percent of neonates belonged to the MS-AS subtype, 27% belonged to the MS-AA subtype, and 31% belonged to the MA-AA subtype. Fifty-six percent of the MS-AA subtype had associated coronary sinusoids. Seventeen percent had associated noncardiac anomalies, which were considered nonlethal. The size of the ascending aorta was 4.8 1 mm for the MS-AS subtype, 2.2 0.6 mm for the MS-AA subtype, and 2.0 0.4 mm for the MA-AA subtype. The age at stage 1 palliation (S1P) was 7 3.2 days. All study patients underwent the Norwood surgery as their initial procedure for S1P. Pulmonary blood flow during Norwood S1P was provided either with an mBTS or a Sano conduit based on our institutional selective approach. For all patients with preoperative ascending aortic diameter of 0.6 mm/kg or less, the Sano modification was chosen. A relative indication was aortic atresia with a relatively larger ascending aorta size. Based on this approach, 35 infants underwent Norwood S1P with mBTS, whereas 65 infants underwent the Sano modification. The hybrid strategy was reserved for patients with extreme prematurity (less than 34 weeks), extreme low birth weight (less than 1.5 kg), or with extracardiac findings. These patients (n ¼ 6) were either listed for transplant (n ¼ 2) or had a comprehensive stage II palliation and hence were not included in the study. Of the 65 infants with the Sano modification, 53 underwent detailed echocardiographic analysis and cardiac catheterization (Fig 1). The range of Sano conduit size was 4 mm (n ¼ 6), 5 mm (n ¼ 23), and 6 mm (n ¼ 24). The size of the conduit did not affect Doppler-derived peak Sano
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
gradient (log rank p ¼ 0.68). Based on the proximal Sano Doppler gradient, these 53 patients were further subdivided into group 1, who had a proximal Sano Doppler gradient of 30 mm Hg or less (40%, n ¼ 21), and group 2, who had a proximal Sano Doppler gradient of greater than 30 mm Hg (60%, n ¼ 32). The groups had similar demographics (Table 1). All 53 infants underwent cardiac catheterization 3 to 5 months after the S1P. Doppler-derived Sano gradients were confirmed by cardiac catheterization in 27 of the 53 patients (Fig 2) and found to have good correlation (R ¼ 0.8). Angiographically, we found the following pattern of obstruction to pulmonary blood flow: none at the Sano outlet; 16 with distal Sano conduit stenosis; 7 at the RPA ostium; and 16 with LPA ostial stenosis. The most common pattern was both distal Sano and proximal LPA stenoses, seen in 12 of the 21 patients (57%), followed by isolated LPA, isolated distal Sano, and combined distal Sano and RPA stenosis in 4 of 21 patients each (19%). The least common was the isolated RPA stenosis, seen in 3 of the 21 patients (14%). Patients in group 1 had smaller ostial RPA diameter (3.2 1.2 mm versus 4.5 1.8 mm, p ¼ 0.03) and smaller
1381
ostial LPA diameter (3.4 1.2 mm versus 5.6 2.1 mm, p ¼ 0.01), when compared with patients in group 2. Patients in group 1 also had smaller ostial McGoon ratio (1.38 0.26 versus 1.66 0.33, p ¼ 0.03) and ostial Nakata index (152 18.2 mm2/m2 versus 232 26.4 mm2/m2, p ¼ 0.01) compared with group 2, whereas patients in both groups had almost similar hilar McGoon ratio (1.6 0.32 versus 1.82 0.24, p ¼ 0.2) and hilar Nakata index (317 47 mm2/ m2 versus 368 34 mm2/m2, p ¼ 0.3; Table 2, Fig 3). Figure 4 shows the Doppler and corresponding angiographic findings. It demonstrates that patients with Doppler-derived peak proximal Sano gradients of 30 or less had either distal Sano or proximal PA stenosis, whereas patients with Doppler-derived peak proximal Sano gradients of greater than 30 had no stenosis anywhere in the Sano conduit nor in the branch PA. Pulmonary vascular resistance (PVR) could not be accurately determined in all patients, and therefore it was not included in the study. Seven of the 10 patients who needed either balloon angioplasty of the distal Sano stenosis or proximal branch PA were from group 1 (p ¼ 0.01). Patients in group 1 had higher rates of unintended PA intervention before or during the BDG or required
Table 1. Demographic Comparison Variable Preoperative variables Birth weight, kg, median SD Prenatal diagnosis Sex, male/female HLHS with IAS Diagnosis: MS/AS, MS/AA, MA/AA Size of AAO, mm, median SD Ventriculocoronary connections Associated noncardiac anomaly Operative variables for NS1P Age at SIP, days, median SD Type of S1P: mBTS/Sano CPB time, minutes, median SD Cross-clamp time, minutes, median SD Circulatory arrest time, minutes Antegrade perfusion time, minutes Postoperative variables after NS1P-Fontan Overall mortality/transplant ECMO Arrhythmia Postoperative renal dysfunction Postoperative RV dysfunction Postoperative moderate or greater TR HLOS after NS1P, days, median SD ICULOS after NS1P, days, median SD a
Group 2 compared with group 1, p ¼ 0.05.
b
Total n ¼ 100
Group 1 n ¼ 21
Group 2 n ¼ 32
3.27 0.57 80 (80%) 67/33 9 (9%) 42, 27, 31 3.7 1.0 15 (15%) 17(17%)
3.29 0.44 16 (76%) 14/7 0 (0%) 2, 9, 10 2.3 0.5 7 (33%) 4 (19%)
3.19 0.47 27 (84%) 21/11 0 (0%) 3, 12, 17 2.5 0.4 7 (22%) 6 (19%)
7 3.2 35/65 109 33 65 15 50.3 8.3 34.3 6.1
6.7 2.6 0/21 103 31.2 64 16.7 52 6.2 35 4.2
6.8 2.8 0/32 103 32.4 67 18.6 51 6.4 35 3.2
35 (35%) 21 (21%) 25 (25%) 38 (38%) 21 (21%) 12 (12%) 35 7.3 29 6.8
6 (29%)a 5 (24%)b 7 (33%) 4 (19%) 7 (33%) 5 (24%) 33 4.2 27 3.2
4 (12.5%)a 3 (9%)b 9 (28%) 6 (19%) 9 (28%) 6 (19%) 32 3.5 26 2.6
Group 2 compared with group 1, p ¼ 0.01.
AAO ¼ ascending aorta; CPB ¼ cardiopulmonary bypass; ECMO ¼ extracorporeal membrane oxygenation; HLOS ¼ hospital length of stay; IAS ¼ intact atrial septum; ICULOS ¼ intensive care unit length of stay; mBTS ¼ modified Blalock Taussig Shunt; NS1P ¼ Norwood stage 1 palliation; RV ¼ right ventricle; TR ¼ tricuspid valve regurgitation.
CONGENITAL HEART
Ann Thorac Surg 2014;98:1378–85
1382
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
Ann Thorac Surg 2014;98:1378–85
CONGENITAL HEART Fig 2. Correlation between Doppler-derived peak proximal Sano gradient and other variables. (A) Correlation between Doppler and catheterderived peak proximal Sano gradient. (B) Correlation between peak proximal Sano gradient and O2 saturation. (C) Correlation between peak proximal Sano gradient and mid right pulmonary diameter. (D) Correlation between peak proximal Sano gradient and mid left pulmonary artery diameter. (E) Correlation between peak proximal Sano gradient and ostial McGoon ratio. (F) Correlation between proximal Sano gradient and hilar McGoon ratio. (G) Correlation between proximal Sano gradient and ostial Nakata index. (H) Correlation between proximal Sano gradient and hilar Nakata index.
Ann Thorac Surg 2014;98:1378–85
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
1383
Variable RPA diameter, mm LPA diameter, mm Narrowest Sano diameter, mm RVEDP, mm Hg Ostial McGoon ratio Hilar McGoon ratio Ostial Nakata index, mm2/m2 Hilar Nakata index, mm2/m2 O2 saturations before BDG, % Balloon angioplasty of Sano or branch PA before BDGa PA plasty with BDG or need for mBTS or mortality before BDG or mortality/ECMO/ transplant after BDGb PA plasty with Fontan or need for fenestrated Fontanc PA plasty, prolonged chest tube drainage, PLE, mortality post-Fontand
Sano Doppler Gradient 30 n ¼ 21
Sano Doppler Gradient > 30 n ¼ 32
p Value
3.2 1.2 3.4 1.2 3.17 1.7 72 1.38 0.26 1.6 0.32 152 18.2 317 47 71 4 7/21 (33%) 10/21 (48%)
4.5 1.8 5.6 2.1 3.8 1.2 8.6 2.7 1.66 0.33 1.82 0.24 232 26.4 368 34 78 3 3/32 (9%) 5/32 (16%)
0.03 0.01 0.03 0.3 0.03 0.2 0.01 0.3
The Heart Institute for Children, Advocate Hope Children’s Hospital, Oak Lawn, Illinois; and Department of Pediatrics, University of Tennessee Health Science Center, Memphis, Tennessee
Background. The Sano modification of the Norwood operation is a well-established first step palliation for hypoplastic left heart syndrome (HLHS). Theoretically, the first point of resistance to pulmonary flow should be in the proximal Sano, generating high Doppler flow velocity. Paradoxically, however, some patients have low gradients in the proximal Sano conduit. The objective of this study was to determine the hemodynamic and anatomic significance of low proximal Sano Doppler flow velocity and its clinical implications. Methods. Doppler-derived peak gradients in the proximal Sano conduits were measured in HLHS patients after Norwood-Sano surgery over a 4-year period and confirmed by cardiac catheterization within 2 to 4 weeks. Clinical outcomes of patients with proximal Sano gradients of 30 mm Hg or less (group 1) were compared with patients whose gradient was greater than 30 mm Hg (group 2).
Results. Of the 53 patients, 21 (40%) belonged to group 1. Patients in group 1 had smaller ostial right and left pulmonary artery (PA) diameter (3.2 ± 1.2 mm versus 4.5 ± 1.8 mm, p [ 0.03; and 3.4 ± 1.2 mm versus 5.6 ± 2.1 mm, p [ 0.01) when compared with patients in group 2. Patients (7 of 10) who needed either balloon angioplasty of a distal Sano or proximal branch PA were from group 1 (p [ 0.01). Patients in group 1 had higher rates of unintended PA interventions (33% versus 9%) and complications (48% versus 16%) compared with group 2. Conclusions. Low peak Doppler flow velocity in the proximal Sano correlates with the presence of either distal Sano stenosis or proximal branch PA stenosis. These patients require close follow-up in the interstage period and may need earlier intervention.
B
The concerns of the Sano modification are related to the ventriculotomy, which could contribute to ventricular dysfunction or aneurysm formation, or be a nidus for arrhythmias. Free conduit insufficiency may result in ventricular volume overload and inadequate PA growth, which may be due to the lack of forward flow during diastole [7, 8]. Several case series report improved short-term survival with the Sano modification [1, 3, 9, 10]. Retrospective reviews have demonstrated no early or midterm survival differences between the mBTS or the Sano modification [4, 11–13]. A study using cardiac magnetic resonance showed smaller right pulmonary artery growth and worsening ventricular function in the Sano cohort compared with the mBTS cohort [7]. The Single Ventricle Reconstruction Trial, which is a multicenter, randomized clinical trial, compared the intermediate outcomes of neonates undergoing the mBTS with those of the Norwood-Sano operation [8]. It concluded that neonates who have the Sano modification had smaller PA growth and higher incidence of unintended interventions. Hemodynamically, the first point of high resistance to pulmonary blood flow is in the proximal Sano conduit
etter understanding of anatomy and physiology of patients with hypoplastic left heart syndrome (HLHS) and advances in surgical techniques and postoperative management have contributed to a remarkable improvement in outcomes after the Norwood operation. The classic Norwood operation involved a modified Blalock-Taussig shunt (mBTS) for pulmonary blood flow. An alternative is the Sano shunt, which involves placing a valveless polytetrafluoroethylene conduit from the right ventricle (RV) to the pulmonary artery (PA) to provide pulmonary blood flow [1]. The advantages of the Sano modification (RV-PA conduit) are associated with the absence of diastolic runoff from the systemic circulation to the pulmonary circulation. Hence, it potentially provides a stable immediate postoperative course, increased coronary arterial flow, improved weight gain related to improved splanchnic perfusion, and lower interstage mortality [2–6].
Accepted for publication May 6, 2014. Address correspondence to Dr Philip, Division of Pediatric Cardiology, Le Bonheur Children’s Hospital, Heart Institute, 3rd Flr, 50 N Dunlap St, Memphis, TN 38103; e-mail: [email protected].
Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2014;98:1378–85) Ó 2014 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.05.049
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
Abbreviations and Acronyms BDG HLHS LPA MA-AA mBTS MS-AA MS-AS PA RPA RV S1P
= = = = = = = = = = =
bidirectional Glenn hypoplastic left heart syndrome left pulmonary artery mitral atresia and aortic atresia modified Blalock-Taussig shunt mitral stenosis and aortic atresia mitral stenosis and aortic stenosis pulmonary artery right pulmonary artery right ventricle stage 1 palliation
and should result in a high Doppler flow velocity signal at this point. Paradoxically, however, we observed that a number of patients had low gradients in the proximal Sano conduit. We postulated that a low flow velocity in the proximal Sano conduit is caused by a significant distal obstruction. The purposes of this study were to determine whether the low proximal Sano flow velocity is caused by significant distal obstruction, and to ascertain the hemodynamic, anatomic, and clinical implications of such findings.
Patients and Methods Review of medical records and computerized hospital data was approved by the Institutional Review Board, and the procedures followed were in accordance with institutional guidelines for retrospective record review and protection of patients’ confidentiality. The Institutional Review Board waived the need for patient consent.
Study Population Neonates with HLHS who were managed at our institution during a 4-year period were included in the study. HLHS was defined as the presence of mitral atresia and aortic atresia (MA-AA), mitral stenosis and aortic atresia (MS-AA), or severe mitral stenosis and aortic stenosis (MS-AS). All had mitral valve diameter, aortic valve size Z –4 or less. All cases with additional or other cardiac anomalies were excluded from this study, including HLHS with ventricular septal defect, unbalanced atrioventricular septal defect, double outlet right ventricle with mitral and aortic stenosis or atresia, critical aortic stenosis, endocardial fibroelastosis with normal or large left ventricle, and heterotaxy syndromes requiring a Norwood operation.
Operative Technique for the Sano Conduit Our surgical approach for the Sano conduit involves anastomosing the PA conduit at the partially augmented PA wall early during the cooling phase. During the rewarming phase, a modification of the anastomosis to the RV is made, and the conduit directed leftward to the aorta. A full thickness ventriculotomy is performed to construct a transmural opening of 4.5 mm to 5 mm. After
1379
the length of the polytetrafluoroethylene conduit is fashioned, its anterior wall is reverse beveled. The posterior wall is then sutured to the epicardial surface of the ventriculotomy. A “baseball-plate” tailored bovine pericardial hood is constructed to augment the anterior anastomotic site. This hood is fashioned purposely redundant. It is believed to reduce the diastolic regurgitant flow due to dynamic anterior wall effect and to prevent muscular hypertrophy, shunt intimal hyperplasia, and early stenosis at the ventricular end of the conduit.
Measurement Variables We measured Doppler-derived peak gradient in the proximal Sano conduit in HLHS survivors after Norwood-Sano surgery. Pulse-wave Doppler was performed in a high parasternal short-axis view or, rarely, from a subcostal sagittal view of the Sano conduit. These views provide the best pulse Doppler sampling of the proximal Sano conduit. All patients underwent cardiac catheterization within 2 to 4 weeks of the Doppler study before referral for the bidirectional Glenn (BDG) surgery. We confirmed Doppler-measured peak Sano gradients by cardiac catheterization. Patients who had Doppler proximal Sano gradient of 30 mm Hg or less (group 1) were compared with patients who had Sano gradient greater than 30 mm Hg (group 2). Patients were followed till Fontan completion. The McGoon ratio is the combined diameter of the right pulmonary artery (RPA) and left pulmonary artery (LPA) at the level of the hilum of the lung, to the diameter of the descending thoracic aorta at the level of the diaphragm. We calculated the McGoon ratio using the narrowest PA branch diameter, which is at the level of the ostium of each branch PA and also at the level of the hilum of the lungs. They are called ostial McGoon ratio and hilar McGoon ratio, respectively. The Nakata index is the sum of the cross-sectional areas of the RPA and LPA at the level of the hilum of the lungs, indexed to body surface area. We calculated an ostial Nakata index and a hilar Nakata index. These indexes were calculated in all patients from PA angiograms. Pulmonary angiography was performed by contrast injection in the proximal Sano conduit with the lateral camera in the 90-degree left anterior oblique projection and the anterior-posterior camera in a 45-degree caudal angulation. This view, called the “sleeping-bat view,” provides the best layout of the entire Sano conduit and the branch PAs without any foreshortening. The following were considered unintended interventions before the BDG surgery: balloon angioplasty or stent implantation in the Sano conduit or the branch PA, mBTS before the BDG, or PA arterioplasty during the BDG. Mortality, extracorporeal membrane oxygenation (ECMO) support or cardiac transplantation after the BDG were considered unintended complications after BDG. Similarly, PA intervention before or during Fontan or need for a fenestrated Fontan were considered unintended interventions for the Fontan surgery. After Fontan completion, mortality, ECMO support, cardiac
CONGENITAL HEART
Ann Thorac Surg 2014;98:1378–85
1380
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
Ann Thorac Surg 2014;98:1378–85
Fig 1. Flowchart showing patient distribution and outcomes. (mBTS ¼ modified Blalock-Taussig shunt; HLHS ¼ hypoplastic left heart syndrome; PA ¼ pulmonary artery.)
CONGENITAL HEART
transplantation, prolonged chest tube drainage, and protein-losing enteropathy (PLE) were considered unintended complications. The primary outcome of the study was the association of the Doppler-derived peak proximal Sano gradient and the need for unintended interventions. The secondary outcome was the association of the Doppler-derived peak proximal Sano gradient and unintended complications.
Statistical Analysis Data are expressed as mean SD for continuous variables and as median and range for categorical variables. The Fisher exact test was used to detect significant differences between groups for categorical variables. Unpaired one-tailed or two-tailed t tests were performed for comparison of continuous variables between groups. Survival analysis was performed using the Kaplan-Meier product limit method for all subgroups. Cox regression on survival was then performed to see whether there was a real difference when other confounding factors were accounted for. Statistical relationship between two variables was determined and expressed using Pearson’s correlation coefficient. The software IBM SPSS 15.0.1 for Windows (IBM Corporation, Armonk, NY) was used for statistical analysis.
Results Neonates with HLHS over a 4-year-period were included in the study. The male to female ratio was 2:1. Forty-two
percent of neonates belonged to the MS-AS subtype, 27% belonged to the MS-AA subtype, and 31% belonged to the MA-AA subtype. Fifty-six percent of the MS-AA subtype had associated coronary sinusoids. Seventeen percent had associated noncardiac anomalies, which were considered nonlethal. The size of the ascending aorta was 4.8 1 mm for the MS-AS subtype, 2.2 0.6 mm for the MS-AA subtype, and 2.0 0.4 mm for the MA-AA subtype. The age at stage 1 palliation (S1P) was 7 3.2 days. All study patients underwent the Norwood surgery as their initial procedure for S1P. Pulmonary blood flow during Norwood S1P was provided either with an mBTS or a Sano conduit based on our institutional selective approach. For all patients with preoperative ascending aortic diameter of 0.6 mm/kg or less, the Sano modification was chosen. A relative indication was aortic atresia with a relatively larger ascending aorta size. Based on this approach, 35 infants underwent Norwood S1P with mBTS, whereas 65 infants underwent the Sano modification. The hybrid strategy was reserved for patients with extreme prematurity (less than 34 weeks), extreme low birth weight (less than 1.5 kg), or with extracardiac findings. These patients (n ¼ 6) were either listed for transplant (n ¼ 2) or had a comprehensive stage II palliation and hence were not included in the study. Of the 65 infants with the Sano modification, 53 underwent detailed echocardiographic analysis and cardiac catheterization (Fig 1). The range of Sano conduit size was 4 mm (n ¼ 6), 5 mm (n ¼ 23), and 6 mm (n ¼ 24). The size of the conduit did not affect Doppler-derived peak Sano
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
gradient (log rank p ¼ 0.68). Based on the proximal Sano Doppler gradient, these 53 patients were further subdivided into group 1, who had a proximal Sano Doppler gradient of 30 mm Hg or less (40%, n ¼ 21), and group 2, who had a proximal Sano Doppler gradient of greater than 30 mm Hg (60%, n ¼ 32). The groups had similar demographics (Table 1). All 53 infants underwent cardiac catheterization 3 to 5 months after the S1P. Doppler-derived Sano gradients were confirmed by cardiac catheterization in 27 of the 53 patients (Fig 2) and found to have good correlation (R ¼ 0.8). Angiographically, we found the following pattern of obstruction to pulmonary blood flow: none at the Sano outlet; 16 with distal Sano conduit stenosis; 7 at the RPA ostium; and 16 with LPA ostial stenosis. The most common pattern was both distal Sano and proximal LPA stenoses, seen in 12 of the 21 patients (57%), followed by isolated LPA, isolated distal Sano, and combined distal Sano and RPA stenosis in 4 of 21 patients each (19%). The least common was the isolated RPA stenosis, seen in 3 of the 21 patients (14%). Patients in group 1 had smaller ostial RPA diameter (3.2 1.2 mm versus 4.5 1.8 mm, p ¼ 0.03) and smaller
1381
ostial LPA diameter (3.4 1.2 mm versus 5.6 2.1 mm, p ¼ 0.01), when compared with patients in group 2. Patients in group 1 also had smaller ostial McGoon ratio (1.38 0.26 versus 1.66 0.33, p ¼ 0.03) and ostial Nakata index (152 18.2 mm2/m2 versus 232 26.4 mm2/m2, p ¼ 0.01) compared with group 2, whereas patients in both groups had almost similar hilar McGoon ratio (1.6 0.32 versus 1.82 0.24, p ¼ 0.2) and hilar Nakata index (317 47 mm2/ m2 versus 368 34 mm2/m2, p ¼ 0.3; Table 2, Fig 3). Figure 4 shows the Doppler and corresponding angiographic findings. It demonstrates that patients with Doppler-derived peak proximal Sano gradients of 30 or less had either distal Sano or proximal PA stenosis, whereas patients with Doppler-derived peak proximal Sano gradients of greater than 30 had no stenosis anywhere in the Sano conduit nor in the branch PA. Pulmonary vascular resistance (PVR) could not be accurately determined in all patients, and therefore it was not included in the study. Seven of the 10 patients who needed either balloon angioplasty of the distal Sano stenosis or proximal branch PA were from group 1 (p ¼ 0.01). Patients in group 1 had higher rates of unintended PA intervention before or during the BDG or required
Table 1. Demographic Comparison Variable Preoperative variables Birth weight, kg, median SD Prenatal diagnosis Sex, male/female HLHS with IAS Diagnosis: MS/AS, MS/AA, MA/AA Size of AAO, mm, median SD Ventriculocoronary connections Associated noncardiac anomaly Operative variables for NS1P Age at SIP, days, median SD Type of S1P: mBTS/Sano CPB time, minutes, median SD Cross-clamp time, minutes, median SD Circulatory arrest time, minutes Antegrade perfusion time, minutes Postoperative variables after NS1P-Fontan Overall mortality/transplant ECMO Arrhythmia Postoperative renal dysfunction Postoperative RV dysfunction Postoperative moderate or greater TR HLOS after NS1P, days, median SD ICULOS after NS1P, days, median SD a
Group 2 compared with group 1, p ¼ 0.05.
b
Total n ¼ 100
Group 1 n ¼ 21
Group 2 n ¼ 32
3.27 0.57 80 (80%) 67/33 9 (9%) 42, 27, 31 3.7 1.0 15 (15%) 17(17%)
3.29 0.44 16 (76%) 14/7 0 (0%) 2, 9, 10 2.3 0.5 7 (33%) 4 (19%)
3.19 0.47 27 (84%) 21/11 0 (0%) 3, 12, 17 2.5 0.4 7 (22%) 6 (19%)
7 3.2 35/65 109 33 65 15 50.3 8.3 34.3 6.1
6.7 2.6 0/21 103 31.2 64 16.7 52 6.2 35 4.2
6.8 2.8 0/32 103 32.4 67 18.6 51 6.4 35 3.2
35 (35%) 21 (21%) 25 (25%) 38 (38%) 21 (21%) 12 (12%) 35 7.3 29 6.8
6 (29%)a 5 (24%)b 7 (33%) 4 (19%) 7 (33%) 5 (24%) 33 4.2 27 3.2
4 (12.5%)a 3 (9%)b 9 (28%) 6 (19%) 9 (28%) 6 (19%) 32 3.5 26 2.6
Group 2 compared with group 1, p ¼ 0.01.
AAO ¼ ascending aorta; CPB ¼ cardiopulmonary bypass; ECMO ¼ extracorporeal membrane oxygenation; HLOS ¼ hospital length of stay; IAS ¼ intact atrial septum; ICULOS ¼ intensive care unit length of stay; mBTS ¼ modified Blalock Taussig Shunt; NS1P ¼ Norwood stage 1 palliation; RV ¼ right ventricle; TR ¼ tricuspid valve regurgitation.
CONGENITAL HEART
Ann Thorac Surg 2014;98:1378–85
1382
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
Ann Thorac Surg 2014;98:1378–85
CONGENITAL HEART Fig 2. Correlation between Doppler-derived peak proximal Sano gradient and other variables. (A) Correlation between Doppler and catheterderived peak proximal Sano gradient. (B) Correlation between peak proximal Sano gradient and O2 saturation. (C) Correlation between peak proximal Sano gradient and mid right pulmonary diameter. (D) Correlation between peak proximal Sano gradient and mid left pulmonary artery diameter. (E) Correlation between peak proximal Sano gradient and ostial McGoon ratio. (F) Correlation between proximal Sano gradient and hilar McGoon ratio. (G) Correlation between proximal Sano gradient and ostial Nakata index. (H) Correlation between proximal Sano gradient and hilar Nakata index.
Ann Thorac Surg 2014;98:1378–85
SATHANANDAM ET AL PROXIMAL SANO CONDUIT DOPPLER SIGNIFICANCE
1383
Variable RPA diameter, mm LPA diameter, mm Narrowest Sano diameter, mm RVEDP, mm Hg Ostial McGoon ratio Hilar McGoon ratio Ostial Nakata index, mm2/m2 Hilar Nakata index, mm2/m2 O2 saturations before BDG, % Balloon angioplasty of Sano or branch PA before BDGa PA plasty with BDG or need for mBTS or mortality before BDG or mortality/ECMO/ transplant after BDGb PA plasty with Fontan or need for fenestrated Fontanc PA plasty, prolonged chest tube drainage, PLE, mortality post-Fontand
Sano Doppler Gradient 30 n ¼ 21
Sano Doppler Gradient > 30 n ¼ 32
p Value
3.2 1.2 3.4 1.2 3.17 1.7 72 1.38 0.26 1.6 0.32 152 18.2 317 47 71 4 7/21 (33%) 10/21 (48%)
4.5 1.8 5.6 2.1 3.8 1.2 8.6 2.7 1.66 0.33 1.82 0.24 232 26.4 368 34 78 3 3/32 (9%) 5/32 (16%)
0.03 0.01 0.03 0.3 0.03 0.2 0.01 0.3
