|
|
||||||||
J Thorac Cardiovasc Surg 2008;135:339-346
© 2008 The American Association for Thoracic Surgery
Surgery for Congenital Heart Disease |
a Department of Cardiac Surgery, Childrens Hospital Boston, Boston, Mass
b Department of Cardiology, Childrens Hospital Boston, Boston, Mass.
Received for publication May 9, 2007; revisions received August 30, 2007; accepted for publication September 14, 2007. * Address for reprints: Francis Fynn-Thompson, MD, Department of Cardiac Surgery, Childrens Hospital Boston, Harvard Medical School, 300 Longwood Ave, Bader 273, Boston, MA 02115. (Email: francis.fynnthompson{at}cardio.chboston.org).
| Abstract |
|---|
|
|
|---|
Methods: From January 2001 through December 2006, all patients with hypoplastic left heart syndrome who underwent stage I palliation at Childrens Hospital Boston were retrospectively reviewed. The subgroup with the mitral stenosis–aortic atresia variant was studied separately. We evaluated preoperative echocardiographic data, operative characteristics, and postoperative factors associated with death or the need for transplantation. The Kaplan–Meier method was used to assess survival.
Results: Thirty-eight (23%) of 165 patients had mitral stenosis–aortic atresia. Hospital mortality or need for transplantation for patients with mitral stenosis–aortic atresia was significantly higher than for other anatomic subgroups (29% vs 7.9%, P = .002). Left ventricle–subepicardial coronary artery communications were present in 20 (53%) patients with mitral stenosis–aortic atresia and were associated with a significantly higher hospital mortality (50% vs 6%, P = .004). No difference in outcome was demonstrated between different sources of pulmonary blood flow. A longer cardiopulmonary bypass time (P = .02) and the need for postoperative extracorporeal membrane oxygenation support (P < .001) were associated with a higher mortality rate.
Conclusions: With improved outcomes in the management of neonates with hypoplastic left heart syndrome, those with the mitral stenosis–aortic atresia variant and left ventricle–subepicardial coronary artery fistulae have emerged as a higher-risk subgroup for failure of stage I palliation. Further investigation is required, and a change in clinical management strategy for this particular subgroup might be warranted.
| Introduction |
|---|
|
|
|---|
|
Hospital outcome after stage I surgical palliation for hypoplastic left heart syndrome (HLHS) in our institution and others has improved considerably in the last 2 decades (Figure 1).
1,2
The decrease in overall hospital mortality for these patients prompted a review of our more recent experience to investigate whether there are specific anatomic subsets of patients who have not benefited from this trend and might still be at increased risk. To date, few studies have looked at the specific anatomic variants of HLHS and their effect on surgical outcomes.1-4
Most published studies have focused primarily on other factors, such as weight, prematurity, the size of the ascending aorta, or the relationship of left ventricular (LV) to right ventricular (RV) volumes.1-3
|
The purpose of this review is to summarize the recent surgical outcomes for patients with HLHS and the MS-AA variant at our institution and to identify factors associated with hospital mortality and failure of stage I Norwood palliation within this subgroup.
| Materials and Methods |
|---|
|
|
|---|
As part of a larger review of our contemporary outcomes of stage I palliation for HLHS, we retrospectively identified and included in the study all infants with the diagnosis of HLHS and MS-AA who underwent stage I palliation at the Childrens Hospital Boston between January 2001 and December 2006.
Echocardiographic Evaluation
The anatomic diagnosis of the MS-AA variant of HLHS was identified by means of postnatal echocardiographic analysis in all patients. A further retrospective review of the preoperative echocardiogram was done in all 38 patients with MS-AA. All the echocardiograms were independently reviewed by 2 cardiologists in our institution, focusing on the status of the interatrial septum, measurements of the mitral valve annulus, mitral valve function, LV end-diastolic volume, septal thickness, size of the ascending aorta, and presence of LV-CA fistulae. The echocardiographic presence of LV-CA fistulae was defined as the presence of systolic and diastolic flow across the LV myocardium away from the mitral valve orifice.10-13
In the last 3 patients in our series, we confirmed the echocardiographic diagnosis of LV-CA fistulae with angiographic analysis. The echocardiographic findings in these 3 catheterized patients with angiographically documented fistulae were used during the retrospective review of echocardiograms in the remaining 35 patients with MS-AA (Figure 2).
|
Surgical Procedure
All patients were taken to the operating room after resuscitation and inotropic and ventilatory support, as necessary. Surgical intervention consisted of standard stage I Norwood palliation with neoaortic arch reconstruction and atrial septectomy. Pulmonary blood flow was restored with either a modified Blalock–Taussig shunt (mBTS) or a right ventricle–pulmonary artery (RV-PA) conduit based on the individual surgeons preference and on randomization in the ongoing Single Ventricle Reconstruction (SVR) trial since January 2004. A period of deep hypothermic circulatory arrest, selective antegrade cerebral perfusion, or both was used in all cases at the surgeons discretion.
Statistical Analysis
The primary outcome variables were failure of stage I palliation (defined as in-hospital mortality or need for heart transplantation) and interstage mortality occurring between stage I palliation and bidirectional Glenn (BDG) shunt. Patient and procedural variables assessed for association with these outcomes included prenatal diagnosis, presence of intact or highly restrictive atrial septal communication, anatomic subgroup, presence of coronary artery fistulae, degree of tricuspid valve regurgitation, age at surgical intervention, weight at surgical intervention, type of pulmonary blood flow supply, need for revision of pulmonary blood flow supply, cardiopulmonary bypass (CPB) time, circulatory arrest time, and postoperative extracorporeal membrane oxygenation (ECMO) requirement (Table 1). Continuous variables were summarized as means ± standard deviation or median (range). Characteristics were compared for patients who died in the hospital after the stage I procedure versus those who survived until discharge by using the Fisher exact test for categorical variables and either the 2-sample t test or the Wilcoxon rank sum test for continuous variables. Patients with and without LV-CA fistulae were compared in a similar manner. Survival probabilities were estimated by using the Kaplan–Meier method.
|
| Results |
|---|
|
|
|---|
This group with MS-AA represents 23% (38/165) of all neonates with HLHS presenting to our institution during the same period. Of these 38 patients comprising the study group, 21 (55%) were given prenatal diagnoses of HLHS. Twenty (53%) of the 38 patients had preoperative echocardiographic evidence of LV-CA fistulae. Of these 38 patients, 9 (26%) had an IAS and underwent creation or enlargement of an atrial septal defect in the catheterization laboratory before surgical intervention.
Stage I Palliation
The 38 patients comprising the study group had stage I palliation at a median of 4 days (range, 0–36 days) of age, with a median weight of 3.1 kilograms (range, 1.7–4.4 kg). An mBTS supplied pulmonary blood flow in 23 (61%) patients, and its diameter was 3.5 mm in 20 patients, 4 mm in 1 patient (with IAS), and 3 mm in 2 patients who weighed less than 2 kg. When an RV-PA conduit was placed, the conduit was constructed of either a 4-mm (n = 1) or 5-mm (n = 14) Gore-Tex tube. The infant who received a 4-mm RV-PA conduit weighed less than 2 kg. In 15 (40%) patients the source of pulmonary blood flow had been randomized in the ongoing SVR trial (mBTS, 5; RV-PA conduit, 10).
Mean CPB time was 160 ± 59 minutes, mean crossclamp time was 43 ± 32 minutes, and mean circulatory arrest time was 40 ± 24 minutes. The chest was initially left open in 33 (87%) patients after surgical palliation and was closed after a median of 3 days (range, 2–5 days).
Four (11%) of 38 patients required shunt revision (3 of these patients had an mBTS). Excessive pulmonary blood flow prompted anatomic reduction of the shunt diameter in 2 patients, whereas acute shunt thrombosis occurred in 1 patient. In the remaining patient the RV-PA conduit was changed to an mBTS because of the finding of stenosis at the proximal conduit anastomosis associated with a severely thickened right ventricle.
Nine (24%) patients were supported postoperatively with ECMO; 6 of them died, and 2 underwent transplantation. Indications for ECMO included inability to wean from CPB (n = 3), severe RV dysfunction (n = 3), cyanosis (n = 2), and acute shunt thrombosis (n = 1).
Hospital Survival and Predictors of Mortality
Hospital mortality and need for transplantation for patients with MS-AA was significantly higher than those for the other anatomic subgroups of patients with HLHS undergoing stage I palliation during the same period (29% vs 7.9%, P = .002). Twenty-seven patients survived stage I palliation. Seven patients died from RV failure, 3 from sepsis, and 1 from alveolar capillary dysplasia. Two additional patients underwent orthotopic heart transplantation after stage I palliation failed; 1 survived, and 1 did not. For the 27 hospital survivors to discharge, median hospital stay was 20 days (range, 4-82 days). The presence of LV-CA fistulae was associated with failure of stage I palliation (P = .004). Other variables significantly associated with hospital mortality were longer CPB time (P = .02) and postoperative ECMO requirement (P < .001, Tables 1 and 2).
Four patients, all with LV-CA fistulae, required prolonged CPB assistance in the operating room because of multiple failed attempts of weaning; in 3 patients ECMO support was started in the operating room. The presence of IAS or highly restrictive atrial septum was not significantly associated with increased hospital mortality (P = .12) in patients with MS-AA.
|
There was no hospital mortality associated with BDG procedures. One patient died 262 days after a BDG procedure of RV failure. Two patients are still waiting for BDG procedures. At most recent follow-up, 14 patients had undergone a fenestrated lateral tunnel Fontan operation at a median age of 31.3 months (range, 20.4–44.1 months). Survival estimates according to the presence of LV-CA fistulae are shown in Figure 3.
|
| Discussion |
|---|
|
|
|---|
We undertook the current review in an attempt to determine specific survival rates and predictors of mortality in recent years. The overall hospital survival at the Childrens Hospital Boston after stage I palliation during the last 6 years is 89% and is comparable with the recent published data in the literature.4,5,17,18
However, the risk of death and failure of stage I palliation is not homogenously distributed across the entire cohort of patients with HLHS, and patients with the MS-AA subtype and LV-CA fistulae had the highest risk of hospital mortality and failure of stage I palliation (P = .002).
In 1990, Murdison and colleagus19
reported that the MS-AA subgroup faired worse in the short and intermediate term after surgical palliation than other anatomic subgroups of HLHS. Subsequently, Jonas and associates,3
in 1994, speculated about a possible correlation between the diagnosis of MS-AA and increased hospital mortality after stage I palliation. Sugiyama and coworkers9
found that patients with MS-AA had a thicker endocardium, and, histologically, there were more abnormal findings, such as myocardial necrosis, calcification, and interstitial fibrosis, compared with other subtypes. These authors speculated that these findings were the result of relative coronary hypoperfusion at the time of surgical intervention. Sauer and colleagues,8
in 1989, also reported a pathologic study demonstrating the association of coronary artery anomalies in 42% of the hearts of patients with HLHS and MS-AA. They observed that the hearts of these patients had a higher percentage of endocardial fibroelastosis, as well as thicker and more tortuous coronary arteries with an increased intimal thickness and stenosis, especially in the proximal left coronary artery. This finding was suggestive of the presence of anatomically significant fistulae. They theorized that these coronary abnormalities might impair ventricular function.
Our study demonstrated that patients with MS-AA had a significantly higher mortality when compared with that of other patients with HLHS undergoing stage I palliation. However, on closer analysis, this increased risk appears to be limited to patients with associated LV-CA fistulae. None of the other studied variables were associated with significantly worse outcomes. In particular, the presence of IAS or highly restrictive atrial septum, although prevalent in this subgroup with MS-AA, was not related to hospital outcome in our analysis. This might be due to the small number of patients included in the study and will have to await confirmation by other studies.
The exact mechanisms for the increased rate of failure after stage I palliation in patients with MS-AA and fistulae are still unclear. We hypothesize that the presence of fistulae might interfere with adequate myocardial protection during CPB or might result in coronary hypoperfusion once the heart is decompressed on CPB. The fact that many of the patients who died had problems beginning in the operating room related to CPB (resulting in significantly longer bypass times for the group that died) possibly suggests such a mechanism. However, those hypotheses remain unproved.
In conclusion, the presence of LV-CA fistulae in patients with HLHS and MS-AA is a risk factor for increased hospital mortality after surgical palliation. Patients with this anatomic variant of HLHS represent an identifiable subgroup that is at significantly increased risk. Based on these findings, we have changed the institutional policy for these patients, performing preoperative angiography on all patients with MS-AA to confirm and better delineate the LV-CA fistulae. Further prospective evaluations with larger numbers of patients are required to recommend modification of the current surgical management of this anatomic variant of HLHS. Possible surgical alternatives include neonatal hybrid stage I palliation and heart transplantation.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
B. Stiller Exciting view on coronary arteries in children with hypoplastic left heart syndrome: finding new answers helps asking better questions Eur J Cardiothorac Surg, January 16, 2012; (2012) ezr307v1. [Full Text] [PDF] |
||||
![]() |
J. H. Hansen, A. Uebing, J. Scheewe, H.-H. Kramer, and G. Fischer Angiographic evaluation of the coronary artery anatomy in patients with hypoplastic left heart syndrome Eur J Cardiothorac Surg, December 9, 2011; (2011) ezr123v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Jacobs, A. Asante-Korang, S. M. O'Brien, P. J. Chai, G. H. Dadlani, G. L. Rodriguez-Fazzi, D. Vu, J. McCormack, D. E. McKenna, R. J. Boucek Jr, et al. Lessons Learned From 119 Consecutive Cardiac Transplants for Pediatric and Congenital Heart Disease Ann. Thorac. Surg., April 1, 2011; 91(4): 1248 - 1255. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Zhang, H. Holtby, S. Cai, O. Al Radi, and J. Li Aortic atresia is associated with an inferior systemic, cerebral, and splanchnic oxygen-transport status in neonates after the Norwood procedure Eur J Cardiothorac Surg, March 1, 2011; 39(3): e13 - e21. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. K. Sathanandam, A. C. Polimenakos, D. A. Roberson, C. F. elZein, A. Van Bergen, T. S. Husayni, and M. N. Ilbawi Mitral Stenosis and Aortic Atresia in Hypoplastic Left Heart Syndrome: Survival Analysis After Stage I Palliation Ann. Thorac. Surg., November 1, 2010; 90(5): 1599 - 1608. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ono, H. Goerler, H. Bertram, and T. Breymann Regression of Marked Myocardial Sinusoids Associated With Hypoplastic Left Heart Syndrome During Staged Fontan Approach Ann. Thorac. Surg., September 1, 2010; 90(3): 1007 - 1009. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Januszewska, R. Kozlik-Feldmann, J. Abicht, R. Dalla-Pozza, and E. Malec Right Ventricle-to-Pulmonary Artery Shunt in Norwood Procedure: Early Results World Journal for Pediatric and Congenital Heart Surgery, April 1, 2010; 1(1): 44 - 50. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. K. Furck, A. Uebing, J. H. Hansen, J. Scheewe, O. Jung, G. Fischer, C. Rickers, T. Holland-Letz, and H.-H. Kramer Outcome of the Norwood operation in patients with hypoplastic left heart syndrome: A 12-year single-center survey J. Thorac. Cardiovasc. Surg., February 1, 2010; 139(2): 359 - 365. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |