| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
King Faisal Heart Institute, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia
Accepted for publication March 31, 2009.
* Address correspondence to Dr Alsoufi, King Faisal Heart Institute (MBC 16), King Faisal Specialist Hospital and Research Center, PO Box 3354, Riyadh, 11211, Saudi Arabia (Email: balsoufi{at}hotmail.com).
Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: Late echocardiographic and time-related clinical results of patients undergoing the Rastelli procedure were examined. Demographics and operative variables affecting outcomes were analyzed.
Results: The Rastelli cohort comprised 36 patients with d-TGA, VSD, and pulmonary stenosis. Median age at operation was 2.4 years (range, 0.3 to 8.3 years). Pulmonary stenosis was present in 31 and atresia in 5. Twenty-two patients had undergone a previous aortopulmonary shunt, and 6 had an atrial septectomy. No operative or late deaths occurred. Time-related freedom from permanent pacemaker implantation, recurrent LVOTO on echocardiogram, and conduit replacement at 10 years was 82%, 100%, and 49%, respectively. Systolic function was normal in all but 3 patients and 92% were in New York Heart Association functional class I and II. None of the patients had late arrhythmias or required heart transplantation.
Conclusions: Early and midterm survival after the Rastelli procedure is satisfactory. Aggressive resection of the infundibular septum to enlarge the VSD has mitigated the risk of LVOTO recurrence. Late conduit obstruction remains an important source of morbidity and frequently requires reintervention.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The policy at our institution is to perform routine aggressive enlargement of the VSD at the time of the Rastelli operation, with anterosuperior enlargement of the defect and routine resection of the infundibular septum. In the current series, we describe clinical and echocardiographic outcome in patients after the Rastelli operation and examine if routine VSD enlargement policy has mitigated the risk of development of LVOTO.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Inclusion Criteria
From 1992 to 2005, 36 consecutive children with d-TGA, VSD, and PS received the Rastelli biventricular repair at the King Faisal Specialist Hospital and Research Center in Riyadh, Saudi Arabia. Clinical, operative, and outcome data were abstracted from their medical records.
Operative Details
All operations are performed through a midline sternotomy. After adhesions are taken down and aortopulmonary shunts are exposed, conventional cardiopulmonary bypass is initiated with standard aortic and bicaval cannulation in all patients. The left heart is vented through the right superior pulmonary vein. Moderate hypothermia of 28° to 32°C is obtained while myocardial protection is provided, with antegrade cardioplegia infusion through the aortic root given every 20 minutes.
Once cardiopulmonary bypass is started, the pulmonary artery branches are adequately mobilized and the exposed modified Blalock-Taussig shunt is externally clipped and divided to avoid traction on the pulmonary artery as the child grows. After aortic cross-clamp application, the main pulmonary artery is transected distal to the pulmonary valve and is oversewn at the level of the pulmonary annulus.
A longitudinal right ventriculotomy is made in line with the proposed conduit path with care to avoid any injury to any major epicardial coronary vessel. Muscle bundles are resected on both sides of the ventriculotomy incision, and pledgeted retraction sutures are placed on the ventriculotomy edges to facilitate exposure of the VSD and the tricuspid valve apparatus (Fig 1A).
|
A redundant cylindrical patch, usually created from a vascular graft that is opened longitudinally, is used to construct the baffle. The baffle construction is usually made with running sutures to the right ventricular tissue, reinforced by pledgeted mattress sutures as needed (Figs 1C and D). Before closure of the superior margin of the patch, the baffle is inspected and sized with a Hagar dilator. If there is any concern regarding the adequacy of the baffle size, the ascending aorta is opened and the repair is examined and sized through the aortic valve.
A small oblique incision in the right atrium is made and the atrial septal defect is closed. The distal main pulmonary artery is opened longitudinally and the incision is carried into the proximal left pulmonary artery. The mobilized pulmonary artery (and subsequently the conduit) is preferably positioned to the left of the ascending aorta. The distal conduit anastomosis is usually made with the heart perfused and beating after air is removed and the cross-clamp is removed, except in smaller infants in whom it can be performed while the aortic cross-clamp is still in place.
Rewarming is completed as the proximal anastomosis is performed and cardiopulmonary bypass is shortly terminated after the completion of the proximal conduit anastomosis (Fig 1E). Intraoperative transesophageal echocardiography is routinely used to assess the repair.
Follow-Up
Late outcomes were determined from recent office visits at our center and from direct correspondence with patients' families. Mean follow-up was 8.3 ± 6.1 years (range, 1.3 to 17 years) and was complete in all but 2 patients who have left the country and in whom partial follow-up was available.
Statistical Analysis
All the data were analyzed with SAS 9 software (SAS Institute, Inc, Cary, NC). Data are presented as frequency (%), median with range, or mean ± SD, as appropriate, with the number of nonmissing values indicated. Estimates for survival, long-term freedom from reoperation, permanent pacemaker insertion, and LVOTO were made by the Kaplan-Meier method. Differences between survival curves were evaluated with the log-rank statistic. Cox regression was used to determine the independent predictors of late outcomes. The appropriateness of variable transformations was determined by means of univariate analysis. Variables with a univariate value of p < 0.05 or those with known biologic significance but failing to meet this critical c2 level were submitted to multivariable models.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Average cardiopulmonary bypass duration was 154 ± 56 minutes, and average aortic cross-clamp was 91 ± 26 minutes. Right ventricular outflow tract (RVOT) reconstruction was performed with pulmonary homografts in 25, aortic homografts in 10, and Contegra bovine jugular conduit in 1. Average conduit size was 19 mm (range, 17 to 22 mm). Additional procedures included pulmonary artery patch angioplasty im 5, tricuspid valve repair in 2, and coarctation repair, total anomalous pulmonary venous connection repair and common atrium septation in 1 patient each.
No operative or late deaths occurred. Six patients required permanent pacemaker implantation for heart block. Most were required within the first year after the Rastelli operation. Overall freedom from permanent pacemaker implantation at 5 years was 82% (Fig 2). The patient's age was not a risk factor for permanent pacemaker requirement. The incidence of complete heart block decreased in the later half in our series, with only 1 patient requiring a permanent pacemaker implantation. This is likely due to more careful VSD enlargement in the anterosuperior direction and more aggressive resection of the infundibular septum, which may carry a lower incidence of injury to the conduction system.
|
|
Additional operations at time of conduit change were required in few patients: tricuspid valve repair in 3, pulmonary artery patch angioplasty in 2, closure of residual VSD in 1, and muscle resection for right midventricular obstruction in 1. No patients required separate reoperations for cardiac causes other than conduit change.
Several demographic, anatomic, and operative variables were examined, and none was identified as a significant factor for conduit replacement on multivariable analysis, likely due to small sample size and the multitude of the factors that can influence graft longevity.
Follow-up echocardiograms were obtained and reviewed. No patients had any evidence of recurrent LVOTO. The median gradient across the LVOT baffle was 5 mm Hg (range, 2 to 14 mm Hg), and in only 1 patient did it exceed 10 mm Hg. The gradient was not higher in younger infants compared with older children. In addition, no aortic regurgitation was evident in any patient. Additional residual lesions on follow-up echocardiogram included small residual VSD in 2, mild tricuspid valve regurgitation in 3, and pulmonary artery branch stenosis in 3. In addition, 8 of the current homografts had evidence of mild stenosis with or without regurgitation, signaling potential need for conduit replacement in the future.
The LV function was normal in 33 patients and moderately depressed in 3. These 3 patients were all aged older than 4 years when they underwent the Rastelli repair.
At the last follow-up, 92 patients were in New York Heart Association functional class I or II. None of the remaining patients required orthotopic heart transplantation.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Early Death
Early operative survival has changed dramatically during the past 40 years. In the series from Mayo Clinic [5], early mortality decreased from 24.4% from years 1968 to 1977 to 4.8% from 1988 to 1997. The Boston group [4] similarly reported 30-day mortality of 11% from 1980 to 1989 and 5% from 1990 to 1998, with no deaths in the last 7 years of their report. Our operative mortality has been zero, which is in line with the current reported operative mortality of less than 5% for various procedures to correct this anomaly [8–11]. This superior operative survival is mainly due to improvement in technical experience with this complex biventricular repair, in perfusion and myocardial preservation strategies, and in postoperative care of those children in the intensive care unit.
Left Ventricular Outflow Tract Obstruction
Late LVOTO is an important late complication after the Rastelli procedure [4, 7]. The obstruction is usually at the level of the VSD or beneath the intracardiac baffle. The mechanisms likely responsible for this complication are related to failure of the VSD or of the baffle pathway to enlarge as the patient grows. Other possible explanations include reduced LV volume load after the repair resulting in a restrictive VSD or development of RV hypertension leading to septal hypertrophy and a restrictive VSD [4, 5, 7].
The reported reoperation risk for recurrent LVOTO after the Rastelli operation varies. Kreutzer and colleagues [4] from Boston reported an 84% freedom from LVOT reintervention at 15 years. In addition, the cause of early death was related to residual LVOTO in 3 patients. In contrast, Dearani and colleagues [5] from Mayo Clinic noted a 99% freedom from LVOT reintervention in hospital survivors, although they did not comment on the presence of LVOTO in the patients who died. The VSD was routinely enlarged anterosuperiorly in less than half of the patients in the Boston series compared with a lower threshold to enlarge the VSD anterosuperiorly in addition to excision of a large portion of the infundibular septum beneath the aortic valve in the Mayo Clinic series [4, 5]. This approach was adopted at our institution and we have not seen echocardiographic evidence of any significant gradient across the LVOT and thus no LVOT reintervention has been required to date.
Kreutzer and colleagues [4] reported that most LVOT reinterventions were in the first 5 years after the Rastelli operation (freedom of 88% at 5 years, 84% at 15 years). Time to LVOTO reintervention was somewhat shorter for infants aged younger than 1 year at the Rastelli repair [4]. Although our follow-up in the current series is shorter than that in their article, the freedom from LVOTO in our patients is promising considering that the majority of obstructions were evident early in their series.
An important variation on the theme of Rastelli was published by LeCompte and colleagues [12] in 1982. This operation was called "reparation a l'etage ventriculaire" or the REV procedure. In a departure from the classic Rastelli operation, the infundibular septum is totally resected, providing a short and direct tunnel from the LV to the aorta, thus avoiding the potential formation of a subaortic LVOTO. In addition, the pulmonary artery was directly connected to the RV without the use of a conduit. Reintervention for LVOTO was required in 2 of 111 patients; this similarly small number confirms the efficacy of aggressive VSD enlargement plus resection of the infundibular septum in reducing the risk of recurrent LVOTO [9, 12, 13].
Aortic translocation and biventricular outflow tract construction, the Nikaidoh operation, was proposed as an alternative to the Rastelli operation because it provides a more normal, better aligned RVOT and LVOT [14]. This procedure is technically challenging, such that it has not been widely adopted despite a recent increased interest by many groups. Yeh and colleagues [11] reported midterm outcomes in 19 patients after the Nikaidoh operation. No LVOTO was evident, and no patient had aortic insufficiency (AI) that was more than mild (mild in 9, trace in 3, and absent in 6) [11]. Conversely, Morell and colleagues [10, 15] reported midterm outcomes in 21 patients after a modified Nikaidoh procedure. LVOTO was similarly eliminated, although half of the patients had mild AI; at least 3 had moderate AI, with 1 requiring reoperation for severe AI.
The higher incidence in the later study may be related to technical differences as discussed by Yeh and Nikaidoh. They selectively reimplant only right coronaries, have not had to implant left coronaries, do not divide the ascending aorta, do not perform a LeCompte maneuver, and operate on slightly older patients [11] compared with Morell and colleagues. In comparison, the risk of AI is very small in patients undergoing the Rastelli operation, and none of the patients in our series had evidence of AI on follow-up echocardiograms.
In brief, our current study emphasizes findings from other reports that routine aggressive VSD enlargement with anterosuperior incision and resection of the infundibular septum mitigate the risk of LVOTO analogous to the more complex aortic translocation operation. Nonetheless, the Nikaidoh procedure remains indicated in patients with anatomic contraindications to the Rastelli procedure, such as those with a small RV, remote VSD, or a straddling tricuspid valve [10, 11, 14, 15].
The incidence of complete heart block requiring pacemaker requirement in the Mayo Clinic series was 11% compared with 3% in the Boston group, excluding patients who died with complete heart block before pacemaker implantation [4, 5]. This may be related to the lower threshold of VSD enlargement in the Mayo Clinic patients. We have had a similarly concerning high incidence of complete heart block and pacemaker implantation in our patients that may limit the value of this approach. Nonetheless, we believe that a more careful anterosuperior enlargement of the VSD and aggressive resection of the infundibular septum may decrease the risk of this complication, and that was noted in our more recent experience with decreased incidence of permanent complete heart block. We speculate that with this approach, the risk of complete heart block can be further reduced similar to that in patients undergoing the REV operation in which the incidence of permanent pacemaker implantation was very low [8, 9, 13].
Right Ventricular Outflow Tract Obstruction
Similar to previously published reports, a high incidence of RVOT reoperation for conduit change was documented in our series [3–5]. Our study is not powered to assess the risk factors for conduit replacement, but recent advances in our knowledge of the management of RVOT are noteworthy.
Valveless RVOT reconstruction such as in the REV operation or the original Nikaidoh procedure has been associated with diminished risk for RVOT reoperation. Nonetheless, the long-term results of free pulmonary regurgitation are of major concern. Earlier reoperation may be required, especially in patients with branch pulmonary artery stenosis, which is frequent in this cardiac pathology and may increase the severity of pulmonary regurgitation and its effect on RV dilatation and dysfunction [8, 9, 11, 13, 15].
The choice of conduit for RVOT reconstruction remains controversial. Dearani and colleagues [5, 16] reported improved conduit survival in those who received bioprosthetic valve conduits compared with homografts. Nonetheless, because we tend to operate earlier on those complex patients, the role of those valved conduits is limited. Moreover, proper conduit selection—and consequently graft survival—may improve in the current era as we learn from several recent reports that have shed some light on factors that may increase graft survival such as the proper sizing between Z score +1 and +3, valve type, and blood type, among others [16–18].
Finally, an important advance in the management of recurrent RVOTO is the evolution of techniques for percutaneous pulmonary valve implantation. Recent experience with this technique is encouraging and may decrease the risk for redo sternotomy for conduit change and may change the philosophy of our approach to recurrent RVOTO in the near future [19].
Ventricular Function and Late Survival
The largest available studies in the literature showed diminished long-term survival to between 50% and 60% at 20 years after the Rastelli operation [4, 5]. Despite shorter follow-up in the current series, the lack of late mortality to date is encouraging. Nonetheless, 3 patients who underwent the procedure at a relatively older age have persistent ventricular dysfunction [6]. Improved postoperative care, surgical, and perfusion techniques may have contributed to diminished late attrition, but early repair for those complex disorders, before development of ventricular changes that may persist despite successful biventricular repair, may be an important factor influencing late ventricular function and patient survival. A study by Graham and colleagues [6] demonstrated abnormal elevated mean end-systolic stress and LV mass and diminished contractile function after the Rastelli procedure. The risk of development of late LVOTO, historically higher in patients undergoing operations at younger age, has been mitigated with routine aggressive VSD enlargement, earlier biventricular repair is advocated in an effort to alleviate the effects of increased exposure to hypoxia, obstruction, and volume overloading over a prolonged period on the function of the LV.
The risk of late arrhythmias and sudden death after the Rastelli operation is well documented [4, 5]. These patients are usually followed up at our arrhythmia clinic where Holter monitoring and exercise testing are done every other year. No late tachyarrhythmias have developed in the patients in our series; however, longer follow-up is necessary because those arrhythmias may continue to arise late after the operation [4, 5].
Early and midterm results after the Rastelli operation have improved. The risk of recurrent LVOTO has been mitigated by routine enlargement of the VSD and excision of the infundibular septum. Neutralization of young age as a risk factor for development of recurrent LVOTO and the presence of persistent ventricular dysfunction despite successful operation especially in older children at time of biventricular repair support earlier definite repair to avoid the sequelae of hypoxia and chronic pressure and volume load on the myocardium. Therefore, our current policy is to perform the Rastelli operation when the child's weight is about 8 kg. The concerning incidence of complete heart block can be diminished by careful anterosuperior incisions and dependence on infundibular resection for VSD enlargement. Surgical alternatives using valveless direct RVOT reconstruction strategies such as the REV or Nikaidoh procedures may offer an advantage with regards to RVOT reintervention risk.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
In comparing this procedure to alternative procedures, it seems to me that the risk of acquiring heart block in this operation is related either to the placement of sutures for the baffle, which would be a risk shared by this procedure and alternative procedures, or to the technique of septal resection to enlarge the VSD [ventricular septal defect]. Would you please comment, and tell us whether over the course of this series you have changed your approach in terms of the details of septal resection to enlarge the VSD.
DR ALSOUFI: As stated, permanent pacemaker requirement was a major concern. As shown in the presentation, the incidence of complete heart block decreased in the later part of our experience. We believe that a more careful anterosuperior enlargement of the VSD and aggressive resection of the infundibular septum may have helped decreasing the risk of this complication as a result of avoiding septal incision in proximity to the conduction system. Existing literature may support our findings; in centers where they routinely enlarge the VSD anterosuperiorly such as the Mayo Clinic, permanent pacemaker implantation was required in about 11% of survivors, while it was required in only 3% of survivors in the Boston series, where they selectively enlarge the VSD, and finally it was required in less than 2% of patients in Lecompte's series where VSD enlargement is usually done by resecting the infundibular septum.
DR PIGULA (Boston, MA): Can you tell me again how long the follow-up was of these patients? And in those patients, pulmonary atresia, I think, is clearly understood. But those with stenosis that you decided needed a Rastelli rather than a straightforward arterial switch operation, some of these patients that will present with some degree of left ventricular outflow tract narrowing under that physiology are still able to be switched using a standard operation. So how did you decide?
DR ALSOUFI: All those patients were assessed at time of their initial presentation by the cardiologists and the cardiac surgeons and were thought not to be candidates to undergo the arterial switch operation. The decision to receive initial palliation is clear when the presenting lesion is pulmonary atresia; however, when there is pulmonary stenosis, the decision about their candidacy to receive the arterial switch operation depends on the size of the pulmonary valve annulus, the status of the cups, and the levels of obstruction. I cannot comment on how many patients during the same time period presented to our institution with d-transposition of the great arteries, VSD, and pulmonary stenosis and underwent a switch operation, but those usually have a reasonable size pulmonary valve annulus, normal cusps and no significant subvalvular obstruction. I'm sorry, what was your other question?
DR PIGULA: The other question was the duration of the follow-up. But also, was there a Z score that you have in mind? I guess I am just trying to get at how the decision was made, Rastelli vs in those borderline cases admittedly.
DR ALSOUFI: I don't have the information about the Z score of those patients with TGA [transposition of the great arteries], VSD, and pulmonary stenosis who had undergone the arterial switch operation successfully at our institution. Again, those usually would have normal cusps and no significant tunnel-like subvalvar obstruction. Of note, we have published our experience with the arterial switch operation in patients with bicuspid pulmonary valve and that is not a contraindication for the switch operation if the cusps are normal. If the diagnosis is made in the neonatal period, the child would undergo the switch operation vs a palliative shunt plus minus septectomy based on the nature of the obstruction of the left ventricular outflow tract.
The average follow-up in our series was about 8 years. This follow-up is relatively shorter compared to larger previous studies from Boston group or the Mayo Clinic. Nonetheless, in terms of recurrent LV outflow tract obstruction risk, if you look at the Boston series, you see that their 5-year freedom from LVOT obstruction was 88% and their 15-year freedom was 84%, so the majority of obstructions were evident within the first 3 to 4 years following operation. So even though our follow-up is not as long, we believe that our finding with regards to elimination of LVOT obstruction risk is very encouraging considering that historically those complications were evident in the early few years following surgery.
DR PIGULA: Well, the reason I am trying to pin you down a little bit on it is we have gone back and looked at these patients with left ventricular outflow tract obstructions in the transposition subgroup and found that a fair number of these, even with—admittedly it is a complex issue—but with Z scores of the pulmonary valve of –2.5 in some cases that that was suitable and actually, compared to the Rastelli, that was a preferential strategy in terms of left ventricular outflow tract lesions in the future. So I think that there is a little bit of a gray area there in terms of which strategy over the long term is going to be the most sustainable.
DR AL-HALEES (Riyadh, Saudi Arabia): I am the surgeon in the group. These patients were clearly nonswitchable. They had bad pulmonary valves, plus minus subvalvular obstruction. All these patients were not candidates for arterial switch.
DR JOSEPH FORBESS (Dallas, TX): One thing that I have gone through with my thinking on those cases, Frank, is that this is analogous to interrupted aortic arch with VSD, because the VSD is essentially the same as an interrupted aortic arch a posterior malalignment VSD. With a show of hands, would people do a switch on a 3-kg child with a 5.5-mm bicuspid pulmonary valve if the leaflets were fairly mobile?
(Show of hands.)
So not that many, but some. I would have to agree that readdressing the left ventricular outflow tract on a patient after arterial switch is clearly not an easy operation.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. W. Brown, M. Ruzmetov, D. Huynh, M. D. Rodefeld, M. W. Turrentine, and A. C. Fiore Rastelli Operation for Transposition of the Great Arteries With Ventricular Septal Defect and Pulmonary Stenosis Ann. Thorac. Surg., January 1, 2011; 91(1): 188 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Alsoufi, Z. Al-Halees, C. Manlhiot, A. Awan, M. Al-Ahmadi, B. W. McCrindle, M. Al-Joufan, and C. C. Canver Intermediate results following complex biventricular repair of left ventricular outflow tract obstruction in neonates and infants Eur J Cardiothorac Surg, October 1, 2010; 38(4): 431 - 438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Weyand, C. Haun, H. Blaschczok, N. Goetz-Toussaint, J. Photiadis, N. Sinzobahamvya, B. Asfour, and V. Hraska Surgical Treatment of Transposition of Great Arteries With Ventricular Septal Defect and Left Ventricular Outflow Tract Obstruction: Midterm Results World Journal for Pediatric and Congenital Heart Surgery, July 1, 2010; 1(2): 163 - 169. [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 |
