| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Pediatrics, University of Michigan School of Medicine, Ann Arbor, Michigan
b Division of Pediatric Cardiac Surgery, Department of Surgery, University of Michigan School of Medicine, Ann Arbor, Michigan
Accepted for publication August 7, 2009.
* Address correspondence to Dr Gaies, L1242 Women's, SPC 5204 1500 E Medical Center Dr, Ann Arbor, MI 48109-5204 (Email: mgaies{at}med.umich.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: A retrospective review was performed on all patients undergoing anatomic repair between January 1993 and January 2009. The primary outcome was in-hospital mortality. Variables potentially associated with outcome were identified a priori. Bivariate analyses were performed to determine the association between these variables and all outcome measures.
Results: In 65 patients who underwent anatomic repair, 35 had Senning/arterial switch and 30 had Senning/Rastelli. Early and intermediate survival rates for Senning/arterial switch operations were 94% and 91%, respectively. Repairs were successful in patients with tricuspid regurgitation, left ventricular outflow obstruction, and left ventricular dysfunction. Predictors of outcome were not identified in this subset. Early and intermediate survival rates for Senning/Rastelli operations were 77% and 60%, respectively. Longer aortic cross-clamp (p = 0.03) and cardiopulmonary bypass times (p = 0.01) were associated with mortality. Ventricular septal defect enlargement was associated with surgical heart block (p < 0.01). Age, prior procedures, atrial-apical discordance, and tricuspid regurgitation were not associated with outcome.
Conclusions: Senning/arterial switch operations can be performed with excellent intermediate-term outcomes in patients with lesions previously thought to confer higher risk. Candidates for Senning/Rastelli procedures may be at increased risk for postoperative morbidity and mortality. More data are necessary to determine factors influencing outcome after anatomic repair.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
This issue has stimulated increasing interest in anatomic repair techniques that restore the morphologic left ventricle and mitral valve to the systemic circulation. Several centers have now reported short-term and intermediate-term results of patients undergoing anatomic repair of CCTGA [1–6]. The results suggest that these anatomic repairs can be performed with acceptable short-term morbidity and mortality, but limited data are available on patient characteristics and other risk factors associated with certain outcomes [6].
Physicians who care for patients with CCTGA are usually faced with multiple potential management strategies, ranging from no intervention to full anatomic repair. As such, it is necessary to determine which patients are best served by a particular approach, and conversely, which factors place patients at incremental risk. The purpose of this study was to examine the anatomic, physiologic, and operative variables that affect outcome after anatomic repair procedures in a large, single-center series.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
|
Techniques described by Senning [8] for the treatment of D-transposition of the great arteries are used, with some modifications, for CCTGA. For those patients with decreased pulmonary blood flow, the left atrium is generally quite small, and surgical reconstruction of the pulmonary venous atrium may be difficult and routine enlargement is done with a patch. The potential to injure the atrioventricular node exists when the systemic venous atrium is constructed because the node is located in a more anterior position and may be crossed by the suture line. Atrial-apical discordance (atrial situs inversus with levocardia, or atrial situs solitus with dextrocardia) is thought to complicate the Senning procedure, but has not, in our experience, led to greater technical difficulties. After initiation of cardiopulmonary bypass (CPB), the heart is rotated into the left pleural cavity in the case of dextrocardia or to the right pleural cavity for patients with levocardia and situs inversus, to gain adequate exposure.
The techniques used for the arterial switch are generally the same as those used for D-transposition. Similar principles for coronary transfer apply in CCTGA. A Lecompte maneuver is used, but the leftward position of the proximal neopulmonary artery makes the connection to the pulmonary bifurcation more difficult. Closure of the bifurcation and extension of the incision towards the left facilitates this anastomosis.
Patients with subvalvar left ventricular outflow tract (LVOT) obstruction are considered suitable candidates for a Senning/arterial switch procedure if an adequate resection is technically feasible. Our experience suggests that the LVOT widens after anatomic repair because the septum moves into the morphologic right ventricle, allowing for a less aggressive resection in the subvalvar area. When a Rastelli procedure is necessary, a right ventriculotomy is placed at the infundibulum, through which a patch can be inserted to channel the ventricular septal defect to the aortic valve.
When needed, enlargement of the defect cannot be done in an anterosuperior direction in hearts with L-looping because complete heart block will result. Fashioning the patch from a tube of stretch polytetrafluoroethylene is a useful maneuver, because this material will conform easier along the curving pathway. A cryopreserved allograft is generally preferred for the right ventricle to pulmonary artery conduit and is typically placed to the left of the ascending aorta. Bovine jugular vein conduits are being used with greater frequency and have proven to be a useful choice.
Outcomes
The primary outcome was in-hospital mortality after anatomic repair. Secondary outcomes were late mortality and major in-hospital morbidity, including the need for mechanical circulatory support, dialysis, or a major neurologic event (seizure or cerebrovascular accident). Data were also collected on surgically acquired complete heart block, length of stay in the intensive care unit and hospital, and reoperation or interventional catheterization for significant residual lesions.
Statistical Analysis
The two surgical groups were analyzed separately because Senning/Rastelli patients were not candidates for a Senning/arterial switch. Univariate analyses were performed describing the frequency of all preoperative and intraoperative variables. Bivariate analyses were performed to determine the association between these variables and all outcome measures. Wilcoxon rank sum tests were used for nonparametric continuous variables and t tests for normally distributed variables. Results are expressed as means for normally distributed variables, and as medians for nonparametric continuous variables. Analysis by 
2 was used for categoric variables, and the Fisher exact test was used for variables with small numbers, such as death. Analyses were performed using SAS 9.1 software (SAS Institute Inc, Carey, NC).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
|
|
|
Mortality
Of the 35 patients who underwent a Senning/arterial switch procedure, 33 (94%) survived to hospital discharge and 32 (91%) survived to the last follow-up. One patient underwent cardiac transplantation for left ventricular diastolic dysfunction and is a long-term survivor. One who did not survive to discharge was a 31-month-old patient with mild tricuspid valve regurgitation and mild stenosis secondary to a supravalvar ring. This patient received a bolus of amiodarone for junctional ectopic tachycardia that directly led to ventricular fibrillation requiring mechanical circulatory support. A hypoxic-ischemic brain injury occurred during the resuscitation and support was eventually withdrawn. The other in-hospital death was a 5-year-old girl who had been previously banded and required mechanical circulatory support on the first postoperative night. She was successfully weaned from this support, but ventricular arrhythmias later developed and she died after a cardiac arrest. The one out-of-hospital death occurred in the previously described 2-month-old infant who was in severe congestive heart failure secondary to anatomic right ventricular failure and tricuspid valve regurgitation at the time of transfer to our institution requiring preoperative mechanical circulatory support. A pulmonary artery band was placed for 2 weeks before a Senning/arterial switch procedure. After discharge, congestive heart failure developed and the patient died 4 months after the repair.
Morbidity and reintervention
Major in-hospital morbidity was documented in 6 of the 33 patients in the Senning/arterial switch group who survived to discharge. Extracorporeal membrane oxygenation was required by 5 survivors, 2 required dialysis, and 5 experienced a neurologic event. Surgically acquired heart block developed in 3 patients. Intensive care unit and hospital length of stay, and reintervention rates are reported in Table 2. Of those who underwent reoperation, 3 required repair of pulmonary venous baffle obstruction, all of whom were operated on before 2002. Since that time, the use of bovine pericardium to augment the pulmonary venous baffle was abandoned in favor of other materials (autologous pericardium, cryopreserved homograft, or polytetrafluoroethylene), resulting in no further incidences of late pulmonary venous baffle obstruction requiring surgical or catheter-based intervention. Reintervention was required on the systemic venous pathway in 3 patients, comprising 2 operations and 1 catheter-based therapy.
Senning/Rastelli Group
Mortality
Of the 30 patients who underwent a Senning/Rastelli procedure, 23 (77%), survived to hospital discharge. There were 5 late deaths, for an overall late survival of 60%. One postoperative death occurred in an 11-month-old infant with CCTGA, dextrocardia, pulmonary atresia, and a hypoplastic tricuspid valve and right ventricle, who was not a suitable candidate for a Fontan palliation. A Senning/Rastelli procedure was judged to be the only surgical option other than transplantation. The patient experienced low cardiac output syndrome requiring mechanical circulatory support and died after a long hospital course. Another in-hospital death was the result of a hyperkalemic arrest. Two in-hospital deaths were characterized by early postoperative hypotension resulting in neurologic injury that ultimately led to the withdrawal of care, despite eventual hemodynamic recovery.
Of the 5 late deaths, 1 was a 17-month-old child with severe preoperative tricuspid valve regurgitation who died from complications of tricuspid valve endocarditis. A successful Senning/Rastelli procedure was performed in another patient, with double-outlet right ventricle, criss-cross atrioventricular relationship, and a previous right ventricular-to-pulmonary artery conduit, but a respiratory infection developed several months later and cardiac arrest resulted in severe ventricular dysfunction. The patient died awaiting transplant. After a successful anatomic repair, a third patient died at an another institution several months later of hyperkalemic arrest secondary to iatrogenic error. The circumstances of the remaining late deaths are not available.
Morbidity and reintervention
Of the 23 patients in the Senning/Rastelli group who survived to discharge, 6 experienced major in-hospital morbidity. Extracorporeal membrane oxygenation was required by 1, dialysis by 1, and 4 experienced a neurologic event. Surgically acquired heart block developed in 6 patients, 5 of whom required enlargement of their ventricular septal defect to construct an intraventricular baffle. Intensive care and hospital length of stay, and reintervention rates are reported in Table 2. Three patients required surgical reintervention on the pulmonary venous baffle, and 1 patient required an operation and percutaneous intervention to relieve superior vena caval and systemic venous baffle obstruction.
Risk Factor Analysis
Because of the infrequent events in the Senning/arterial switch group, no associations could be determined for morbidity or mortality for the various risk factors tested. Moderate or severe tricuspid valve regurgitation was present in 16 Senning/arterial switch patients, with only 1 death and 2 complications occurring among survivors. No deaths and only 1 complication occurred in the 8 Senning/arterial switch patients with mild to severe subvalvar pulmonary stenosis. Four patients in the Senning/arterial switch group had mild to moderate left ventricular dysfunction; all survived and only 1 had a significant postoperative complication.
The association of various risk factors with death in the Senning/Rastelli group is summarized in Table 3. The only risk factors associated with in-hospital death in this group were longer times for CPB (p = 0.01) and aortic cross-clamp (p = 0.03). Neither age at operation nor number of prior surgical procedures were related to early death or morbidity among survivors. The era of operation was also analyzed as a potential risk factor and was not associated with outcome.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Our analysis failed to identify any anatomic or physiologic risk factors that predicted in-hospital death. This was partly because of the low number of events in the Senning/arterial switch group and the small number of patients in the overall cohort. The degree of tricuspid regurgitation was not associated with short- or intermediate-term death, though this is usually an important factor in determining whether a patient should undergo anatomic vs conventional repair. As previously demonstrated by Shin'oka and colleagues [6], CPB time was associated with increased death in the Senning/Rastelli group, and the need to enlarge the ventricular septal defect in these patients significantly correlated with surgical heart block. The small number of patients at any one center limits the ability to draw significant conclusions from such a risk factor analysis, underscoring the need for collaborative research between centers performing anatomic repair.
The survival rate for the Senning/Rastelli group was substantially less than for those patients undergoing a Senning/arterial switch procedure. Senning/Rastelli patients generally had complex underlying anatomy and had undergone a wide variety of palliative operations before referral to our institution for anatomic repair. In some cases, these palliative procedures limited options for ultimate repair. Other surgical approaches, including biventricular conventional repair or single-ventricle palliation, were not judged to be feasible at a reasonable risk. Cardiologists and surgeons are usually faced with multiple surgical options when managing these patients, underscoring the need to determine a plan of palliation beginning in infancy to facilitate eventual repair, regardless of the pathway chosen.
Our review demonstrated that the overall rate of neurologic injury in patients undergoing Senning-Rastelli operations was 33%, almost twice the rate in the Senning/arterial switch group. This led us to further investigate potential hemodynamic factors in the early postoperative period for patients undergoing a Senning/Rastelli that place them at greater risk for this complication. The combination of elevated central venous pressure and low systemic arterial pressure early after repair may result in decreased cerebral perfusion pressure (mean arterial pressure – central venous pressure), jeopardizing blood flow to the brain. This scenario is more likely to occur in patients after Senning/Rastelli operations compared with those undergoing Senning/arterial switch due to the right ventriculotomy and intraventricular baffle, adverse effects of a Senning on right atrial transport (even absent superior vena caval obstruction), and compromised right ventricular diastolic dysfunction secondary to long CPB times.
Our in-depth review of hourly hemodynamic data in the first 48 hours after CPB demonstrated that the risk of neurologic injury is higher in patients undergoing Senning/Rastelli operations when low cerebral perfusion pressures (< 30 mm Hg) develop within 48 hours of their operation, although these patients had a technically successful repair and eventually recovered normal hemodynamic function. This pathophysiologic process appeared to contribute to the rate of neurologic injury in patients undergoing a Senning/Rastelli and most likely affected the overall mortality rate in this subset, with or without neurologic injury.
We have concluded that a lower threshold for mechanical circulatory support may be warranted in the setting of low cerebral perfusion pressure. Furthermore, we have considered both the intraoperative creation of a communication between the systemic and pulmonary venous baffles, or the establishment of a bidirectional Glenn connection when there are heightened preoperative concerns about the adequacy of the tricuspid valve or right ventricular size, or both.
Our results are consistent with those of Shin'oka and colleagues [6], who demonstrated that enlargement of the ventricular septal defect to construct the interventricular baffle during the Senning/Rastelli procedure is a significant risk factor for surgical complete heart block, although not a risk factor for hospital death. Complete heart block developed in 5 of our 8 patients who required enlargement of the defect and pacemaker placement was required. Of these, 3 had S,L,L anatomy, and other 2 had D-looped ventricles. Our practice has been to assume an anterior position of the conduction system in L-looped hearts and a posterior position for patients with D-looped ventricles, and to resect the myocardium accordingly when enlarging an existing ventricular septal defect. However, in the aforementioned study, patients with I,D,D anatomy had an anterior conduction system 80% of the time, and 10% of patients with S,L,L anatomy had a posterior conduction system determined by preoperative electrophysiologic study [6]. Patients undergoing Senning/Rastelli procedures may benefit from precise mapping of the dominant conduction system if defect enlargement is likely to be necessary.
Patients in our study who underwent pulmonary artery banding to achieve left ventricular training before a Senning/arterial switch procedure had similar outcomes to those who did not. These findings are consistent with those of Quinn and colleagues [10]. However, these authors also demonstrated that up to 50% of patients who underwent left ventricular training before anatomic repair had moderate to severe left ventricular dysfunction at intermediate follow-up compared with a prevalence of 20% in patients who did not require training. Given that most of our patients are not followed up at our institution, we had limited echocardiographic data to assess the development of left ventricular dysfunction. More data are necessary to determine optimal patient selection, timing, and duration for left ventricular training before Senning/arterial switch operations.
The enthusiasm for anatomic repair as an alternative to conventional repair of CCTGA has largely been based on literature demonstrating poor long-term survival usually associated with significant dysfunction of the systemic right ventricle [11–14], especially in patients with tricuspid regurgitation [6, 11, 15]. Although anatomic repair can be performed with short-term outcomes similar to those achieved after conventional repair, the long-term risks and benefits of anatomic repair remain unknown. Previous studies have demonstrated that right ventricular function and the degree of tricuspid regurgitation improve after anatomic repair [16, 17]. Other authors have shown excellent postoperative left ventricular function [2, 4, 6, 18] and New York Heart Association functional status after anatomic repair [2, 3, 5].
In contrast, exercise capacity has been shown to be no different [18] or reduced [19] at intermediate follow-up in patients who have undergone anatomic repair compared with patients repaired conventionally. Furthermore, newer data suggest that left ventricular dysfunction and neoaortic valve regurgitation develop frequently [10, 20, 21]. As these patients enter adulthood and the known complications related to atrial switch procedures become more prevalent, comparison of functional capacity and quality of life between anatomically and conventionally repaired patients will be needed.
This study has some limitations. The numbers of patients in each group limited the statistical power and thus the ability to make conclusions about the effect of different risk factors on outcomes. Data were retrospectively reviewed, and echocardiographic data were not reevaluated. Results from our center may not be applicable to centers that have less experience with these patients. Follow-up physiologic and imaging data were limited because most of our patients were referred from other institutions where they received outpatient care after their operation, thus limiting our ability to analyze these outcomes.
In conclusion, we have demonstrated in this series that Senning/arterial switch procedures can be performed with excellent rates of short-term and intermediate-term morbidity and mortality, even in patients with tricuspid regurgitation, left ventricular dysfunction, and significant LVOT obstruction. The outcomes for patients undergoing a Senning/Rastelli repair are less favorable, and more data are necessary to determine which patients are optimal candidates for this procedure. Now that anatomic repair has been shown to be an option for patients with CCTGA, it is important to determine the long-term quality of life and functional outcomes in patients who undergo these procedures. Comparisons of these findings between anatomically and conventionally repaired patients will provide critical information for cardiologists and surgeons caring for patients with this disease when choosing a management strategy.
|
Appendix |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Murtuza, D. J. Barron, O. Stumper, J. Stickley, D. Eaton, T. J. Jones, and W. J. Brawn Anatomic repair for congenitally corrected transposition of the great arteries: A single-institution 19-year experience J. Thorac. Cardiovasc. Surg., December 1, 2011; 142(6): 1348 - 1357.e1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Hraska, A. Mattes, C. Haun, H. C. Blaschczok, J. Photiadis, P. Murin, P. Zartner, and B. Asfour Functional outcome of anatomic correction of corrected transposition of the great arteries Eur J Cardiothorac Surg, November 1, 2011; 40(5): 1227 - 1234. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. G. Gaies, C. S. Watnick, J. G. Gurney, E. L. Bove, and C. S. Goldberg Health-related quality of life in patients with congenitally corrected transposition of the great arteries J. Thorac. Cardiovasc. Surg., July 1, 2011; 142(1): 136 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Hraska, P. Murin, C. Arenz, J. Photiadis, and B. Asfour The modified Senning procedure as an integral part of an anatomical correction of congenitally corrected transposition of the great arteries MMCTS, January 1, 2011; 2011(0224): mmcts.2009.004234 - mmcts.2009.004234. [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 |
