| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;71:1251-1254
© 2001 The Society of Thoracic Surgeons
a Division of Pediatric Cardiology, New York, New York, USA
b Division of Pediatric Cardiovascular Surgery, Babies and Children’s Hospital of New York, New York Presbyterian Hospital, and the Departments of Pediatrics and Surgery, Columbia University, New York, New York, USA
Accepted for publication November 19, 2000.
Address reprint requests to Dr Pass, Babies and Children’s Hospital of New York, 3959 Broadway, 2 North, New York, NY 10032
e-mail: Pediheart{at}aol.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentReferences |
|---|
Methods. From June 1990 to June 1999, 9 patients had direct BVF resection performed at our institution. The median age at surgery was 16.5 years (range 1 month to 27 years). Diagnoses in these patients were {S,L,L} single LV (n = 8) and {S,D,D} single LV tricuspid atresia (n = 1). Eight of 9 patients had pulmonary artery bands placed either before BVF resection or at the same time as this procedure. Three patients required reoperation for reobstruction at the BVF (12 total operations in 9 patients).
Results. Median preoperative peak systolic gradient across the BVF measured at cardiac catheterization was 47 mm Hg (range 10 to 63 mm Hg). The median peak gradient measured by Doppler echocardiography was 44 mm Hg (range 5 to 125 mm Hg). Eight of 9 patients survived the operation to discharge from the hospital and 7 of 9 are alive at follow-up. At a median follow-up of 22 months (range 5 to 76 months), 8 of 8 surviving patients had an unobstructed BVF as determined by qualitative two-dimensional echocardiography and Doppler color flow imaging. There was one perioperative and one late death 5 months postoperatively (secondary to fungal sepsis). No patient developed new or worsened aortic insufficiency after BVF resection. Eight of 9 patients had no change in AV nodal conduction after surgery. One patient developed Mobitz II heart block requiring postoperative implantation of a pacemaker.
Conclusions. Direct resection of an obstructive BVF can be performed with total relief of obstruction although reoperation may be required. Atrioventricular nodal function can be preserved in most patients with this operative approach, including those with {S,L,L} segmental anatomy.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
The DKS procedure was first proposed for the treatment of transposition of the great vessels in 1975 [5–7]. As extended for use in patients with obstruction of the BVF, this procedure involves anastomosis of the isolated proximal main pulmonary artery to the ascending aorta providing two outflow channels from the systemic single ventricle, thus obviating the need for intracardiac repair of complex subaortic obstruction. Potential benefits expressed by advocates of this approach include less reobstruction at the BVF and less heart block [1, 3, 4, 8–14]. However, there is a potential for increased semilunar insufficiency after this approach [1, 3, 10, 13]. In addition, in cases in which a septation is undertaken for complex single ventricles, a semilunar valve is needed for each ventricle making the DKS an untenable option [15]. Finally, rare reports have demonstrated higher than expected rates of heart block after the DKS [3].
In experienced hands, direct resection of the BVF may be advantageous because it is an easier operation than the DKS procedure and may have lower semilunar valvar insufficiency rates. Although there are many reviews of the results of the DKS in BVF obstruction in the pediatric surgical literature, relatively little data assessing direct BVF resection exist [1, 16]. Therefore, we retrospectively reviewed our experience with direct BVF resection over a 9-year period at Babies and Children’s Hospital of New York and Columbia University.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
|
The surgical anatomy of the BVF was identified in all cases by an aortotomy and resected through the aortic valve with care used to avoid resection of tissue that was related to atrioventricular conduction. The technique of BVF resection varied depending on whether the patient was L or D transposed. In L-transposed patients the inferior, apical, and leftward BVF borders were resected with care to avoid the anterosuperior rim of the defect where the conduction system is usually located [16]. In D-transposed patients, the anterosuperior and leftward borders of the defect were extensively resected. Three patients who had other intracardiac repairs at the time of BVF resection also had their defects resected through the tricuspid valve.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
BVF resection was performed as an isolated procedure in 5 of 12 operations performed in this patient group. BVF resection was performed concomitant with a Fontan or bidirectional Glenn procedure in 4 of 12 operations and with a pulmonary artery banding plus atrial septectomy procedure in 3 of 12 operations. All L-transposed patients had pulmonary artery bands placed before or at the time of BVF resection. The single D-transposed patient in this study did not have a pulmonary artery band. The median operative total cardiopulmonary bypass time was 87 minutes (mean 92; range 60 to 150 minutes) with a median aortic cross-clamp time of 39 minutes (mean 47; range 14 to 99 minutes).
There was one perioperative and one late postoperative death in this series. The perioperative death was patient 2, a 3-year-old boy with {S,L,L} segmental anatomy and left-sided atrioventricular valvar atresia. He died after a combination of BVF resection and Fontan procedure. The preoperative gradient across the BVF was 40 mm Hg and he was the only patient with a degree of ventricular hypertrophy characterized as moderate on his preoperative echocardiogram. He had a previous operation for BVF enlargement after Fontan and a second operation for BVF resection; the patient had poor ventricular function with low cardiac output. He underwent cardiac transplantation within 1 week of Fontan palliation, but ultimately died within days of this surgery. Pathologic examination of the explanted heart in this patient demonstrated a widely patent BVF measuring 2 x 1.3 cm in diameter.
The second death occurred in patient 9, 5 months after BVF resection performed concomitant with a partial septation procedure. This patient had {S,L,L} segmental anatomy with a double-inlet left ventricle. Death resulted from complications of fungal sepsis. A postmortem examination was not performed.
At the most recent follow-up (median 22 months; range 5 to 76 months), 8 of 8 operative survivors of BVF resection operations were considered to be unobstructed on the basis of qualitative echocardiographic evaluation using a combination of two-dimensional and color Dopper flow imaging. No patient had more than mild aortic insufficiency preoperatively and there was no change in the degree of postoperative aortic insufficiency in the 8 survivors. None had changes in qualitative ventricular systolic function; all remained within normal limits or were only mildly depressed.
Three patients required reoperation for BVF obstruction; all were L transposed, 2 patients with left atrioventricular valvar atresia and 1 patient with a double-inlet left ventricle. The median time to reoperation was 3.8 years with a range of 2.8 to 6.3 years. Immediate postoperative echocardiograms on these patients after their first BVF resection demonstrated unobstructed flow in 2 patients and moderately obstructed flow in 1 patient. The latter patient was still moderately obstructed; this was the same child who ultimately died perioperatively after his second BVF resection attempt. Before the reoperation, the measured peak systolic ejection gradients at the time of cardiac catheterization in the 3 patients were 10, 30, and 40 mm Hg. At follow-up, the 2 surgical survivors of reoperation after the BVF are unobstructed on echocardiography.
Seven of 9 patients in this series had either normal atrioventricular conduction with normal PR intervals or first degree heart block before BVF resection. Two of 9 patients had permanent pacemakers in place before operations for preexistent Mobitz II or third degree heart block. Atrioventricular conduction was preserved and unchanged in 6 of 7 patients who did not have preoperative pacemakers after BVF resection. One patient with {S,L,L} segmental anatomy and double inlet left ventricle (patient 5) developed high grade second degree block in the immediate postoperative period requiring implantation of a ventricular pacemaker. There was no new development of complete, third degree heart block in the 7 patients without antecedent second or third degree block.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
In recent years, the Damus-Kaye-Stansel procedure has emerged at many centers as the treatment of choice for the management of BVF obstruction based on two major formulations. First, direct BVF resection has been previously associated with relatively high mortality and incidences of heart block [1, 4]. Second, a great deal of literature exists suggesting relatively satisfactory hemodynamic results with the DKS approach [3, 4, 8–14]. However, there are disadvantages to the DKS approach. Postoperative semilunar valvar insufficiency is of particular concern, with quoted incidences between 10% and 52% [3, 4]. In addition, Lui and colleagues [3] reported a higher incidence of heart block in postoperative DKS patients (26%) than in patients who had undergone direct BVF resections (11.4%). This may seem unexpected because of the extracardiac nature of the DKS; however, since the majority of patients with obstructed BVFs are L transposed, no matter what form of repair is undertaken, postoperative heart block must be considered a risk in this patient population.
Finally, although uncommon, septation procedures, in which a single ventricle is divided into two functioning ventricles, require that there be two functioning semilunar valves. This form of septation operation would therefore disqualify the DKS as an option for BVF obstruction [15].
The present study reviews our experience with direct BVF resection in 9 patients. There was a single death related to BVF resection. This occurred in a patient who also underwent Fontan palliation at the time of BVF resection. It may be suggested that the combination of a Fontan palliation at the time of BVF resection, when there is a moderate to severe degree of obstruction, may be unwise because the ventricle would not normally have enough time to remodel after the relief of BVF obstruction. The patient who died after the combined Fontan-BVF resection was the only patient in this series with moderate ventricular hypertrophy. Three other patients in our study underwent BVF resection for mild to moderate obstruction at the time of cavopulmonary anastomosis or Fontan procedure with good hemodynamic results. Thus, the presence of a significant (20 to 40 mm Hg) gradient alone in the absence of ventricular hypertrophy did not represent a risk factor for BVF resection in this small group. In cases with moderate to severe ventricular hypertrophy, a staged approach with separate BVF resection and Fontan may be warranted.
For the 8 surgical survivors, the ultimate hemodynamic results achieved were acceptable. At latest follow-up, 8 of 8 patients had no significant obstruction across the BVF on echocardiographic evaluation. However, 2 of 8 patients in this surgical survivor group required a second operation after the BVF in order to achieve these results, and the single death in this group occurred during an attempted re-resection of the BVF. Thus, although ultimate hemodynamic relief of obstruction was adequate, a significant percentage of patients required a second surgery to achieve this end. Additionally, given that the median follow-up time in this group was only 22 months, follow-up may not be long enough to accurately describe the true incidence of reobstruction in patients treated in this manner.
No patient in this study had more than qualitatively mild aortic insufficiency before surgery and no patient had worsened aortic insufficiency after BVF resection. In addition, no patient had changes in qualitative echocardiographic ventricular systolic function, including the 2 surgical survivors of re-resection.
The electrophysiological results of BVF surgery were acceptable, particularly considering that 8 of 9 patients had {S,L,L} segmental anatomy. Although 2 patients had antecedent heart block requiring pacemaker implantation before BVF resection, none of the other 7 patients developed complete heart block. One patient required implantation of a pacemaker for second degree, Mobitz II block after surgery. These results are particularly important because heart block may appear in a large, increasing proportion of L-transposition of the great arteries in patients at long term follow-up, regardless of whether an operation has been undertaken [17]. Our surgical results further support data from Cheung and colleagues [16] demonstrating that the location of the conduction system is fairly consistent in this patient group, making the avoidance of the conduction system with surgical resection, and therefore of heart block, seem feasible in most cases. Clearly, however, knowledge of the usual anatomy does not completely remove all risk of heart block in this group as evidenced by the relatively high incidence of pacemaker requirement (1 in 7) after BVF resection in this small group.
Direct BVF resection has been performed in our center with good results. This may be the procedure of choice in older patients with previous pulmonary artery banding or in patients with single ventricles undergoing septation procedures. However, there may be patients with obstruction of the BVF in which the DKS may be more applicable; one such group may be previously unoperated infants with very small BVFs. Selection of this subset of infants can be made using careful echocardiographic measurements of the defect, indexed to body surface area [2]. Given the relatively high incidence of need to reintervene with direct BVF resection, in the absence of a direct comparison between the two techniques, it is difficult to draw accurate conclusions as to the best surgical strategy in treating this complex patient group.
In summary, this study demonstrates the feasibility of direct BVF resection for most patients with BVF obstruction. In this small series, direct BVF resection resulted in adequate relief of subaortic obstruction with a relatively low mortality rate, without worsening of aortic insufficiency, and without postoperative complete heart block. This technique represents a reasonable alternative to the DKS in this complex and challenging group of patients.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. D. Rodefeld, M. Ruzmetov, M. S. Schamberger, D. A. Girod, M. W. Turrentine, and J. W. Brown Staged surgical repair of functional single ventricle in infants with unobstructed pulmonary blood flow Eur. J. Cardiothorac. Surg., June 1, 2005; 27(6): 949 - 955. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hiramatsu, Y. Imai, H. Kurosawa, Y. Takanashi, M. Aoki, T. Shin'oka, and T. Sakamoto Midterm results of surgical treatment of systemic ventricular outflow obstruction in Fontan patients Ann. Thorac. Surg., March 1, 2002; 73(3): 855 - 860. [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 |
