Periventricular Closure of Ventricular Septal Defects Without Cardiopulmonary Bypass Zahid Amin, MD, Xiaoping Gu, MD, James M. Berry, RDMS, Jack L. Titus, MD, Samuel S. Gidding, MD, and Albert P. Rocchini, MD Division of Pediatric Cardiology, and Department of Cardiovascular Radiology, University of Minnesota, Minneapolis, Minnesota
Background. Minimally invasive techniques are currently in use to close atrial and ventricular septal defects (VSD). Cardiopulmonary bypass (CPB) is instituted via the femoral vessels, which may cause injury to these vessels, especially in younger patients. The objectives of this study were to demonstrate the feasibility of periventricular closure of muscular VSD (MVSD) and paramembranous VSD (PVSD) without CPB, using the Amplatz VSD device. Methods. Five Yucatan pigs with naturally occurring PVSD (3- to 7-mm diameter) and 5 dogs with surgically created MVSD (6- to 14-mm diameter) were subjects of this study. The VSDs were closed intraoperatively with a 7-French delivery sheath inserted through the free wall
of the right (n ⴝ 5) or left ventricle (n ⴝ 5), under epicardial echocardiogram guidance. The animals were followed for 3 months. Results. There was no operative mortality. All MVSD closed after placement of the device. Closure rate of PVSD was 4 of 5 after placement and 3 of 5 after 3 months. One pig developed aortic incompetence at the last follow-up. Conclusions. Perventricular closure of MVSD and PVSD is feasible. Avoidance of CPB can decrease recovery time, its complications, and trauma to the femoral vessels. (Ann Thorac Surg 1999;68:149 –54) © 1999 by The Society of Thoracic Surgeons
I
Valley, MN) [4, 5]. Both devices were constructed from 0.004-inch wire mesh (Nitinol). The wires were laser welded and then braided with heat treatment. The device was woven to form two discs with a connecting waist. Important features of the device are that it is selfexpandable, self-centering, and retrievable, has a low profile, and can be delivered through a 6- or 7-French sheath. It has a microscrew on one (right or left) disc for attachment to the delivery cable. The steps involved in loading and deployment of the device have been discussed elsewhere [4, 5].
solated ventricular septal defects (VSD) are one of the most common congenital cardiac defects, accounting for approximately 29% of all congenital heart disease [1]. Surgical closure is recommended for hemodynamically significant defects. With the recent surge of minimally invasive techniques in adults, a few institutions have attempted to repair atrial and ventricular septal defects, under endoscopic guidance, by placing the patient on cardiopulmonary bypass (CPB) through femoral vessels [2, 3]. This technique may injure the femoral vessels and the risk of CPB persists. In addition, this technique cannot be applied in younger patients because of small vessel size. We utilized a new technique for closure of paramembranous VSD (PVSD) and muscular VSD (MVSD) without CPB. This technique is simple, and the procedure can be accomplished with a small incision.
MVSD Device
All animals received humane care in compliance with the “Guide for the Care and Use of Laboratory Animals” (NIH Publication No. 86-23, revised 1985).
The diameter of the device waist corresponded to the size of the VSD, and the length of the waist to the thickness of the ventricular septum. The right and left ventricular discs were of the same size. The flange size measured approximately 7 mm (Fig 1A). The discs and the connecting waist were filled with polyester fiber to enhance thrombogenicity. The device has already been used for percutaneous closure of muscular VSD [6].
The Devices
PVSD Device
Material and Methods
The PVSD and MVSD devices were modified from the Amplatzer Atrial septal occluder (AGA Medical, Golden Presented at the Thirty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 25–27, 1999. Address reprint requests to Dr Amin, Division of Pediatric Cardiology, Medical College of Georgia, 1120 15th St, BAA 800 W, Augusta, GA 30912; e-mail:
[email protected].
© 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
The PVSD device was modified from the MVSD device. The length of the waist and the flange size were reduced. Two types of PVSD devices were used. A concentric left ventricular retention disc was used in 4 pigs (Fig 1B), and an eccentric left ventricular disc in 1 pig (Fig 1C). The flange size was reduced to 3 mm in the concentric device. In the eccentric device, the flange on the left ventricular 0003-4975/99/$20.00 PII S0003-4975(99)00519-6
150
AMIN ET AL CLOSURE OF VSD WITHOUT CPB
Ann Thorac Surg 1999;68:149 –54
Fig 2. Delivery catheter inserted through the free wall of the right ventricle across the VSD.
Fig 1. Amplatz MVSD and PVSD devices. (A) MVSD device; (B) PVSD device with concentric left retention disc attached to the delivery cable. (C) PVSD device with eccentric left retention disc.
To close the MVSD, a 7-French delivery sheath with the dilator was pushed through the free wall of right ventricle and aimed toward the VSD (Fig 2). The sheath was advanced across the VSD and the dilator was removed. The device was screwed onto the delivery cable, and was advanced through the sheath to the left ventricle. After confirming its position by epicardial echocardiogram, the delivery cable was pushed and the left disc was deployed. The sheath and the delivery cable were pulled until the left disc approximated the ventricular septum. The delivery cable was kept stable and the sheath was withdrawn to deploy the right ventricular disc. The position of the device was rechecked by echocardiogram, and the device was disconnected from the cable by counterclockwise rotation (Fig 3). The pericar-
disc was reduced to 2 mm toward the aortic and increased to 5 mm toward the ventricular side.
Animal Model The animal model consisted of 5 adult mongrel dogs weighing 20 –29 kg and 5 pigs weighing 18 –25 kg.
Creation and Closure of Muscular Ventricular Septal Defects The method employed in the creation of muscular ventricular septal defects has been discussed elsewhere [6, 7]. Briefly, under general endotracheal anesthesia, the chest was entered via median sternotomy. A sharp punch was used to create the defect through a limited right ventriculotomy. The size of the defect was measured with the help of an epicardial echocardiogram. It ranged from 6 to 14 mm. Three defects were located in the midmuscular septum, one in the anterior muscular, and one in the apical septum.
Fig 3. Echocardiogram still frame showing the device (white arrow) in mid-MVSD.
Ann Thorac Surg 1999;68:149 –54
Fig 4. Left ventricle free wall exposed through left thoracotomy. A delivery catheter has been introduced through the free wall.
dium was partially approximated. A mediastinal and, if needed, pleural chest tubes were placed. The sternum was closed in the standard fashion. The animal was extubated in the operating room. Intravenous fluids were maintained until oral intake was adequate. Intravenous antibiotics and analgesia were provided for 3 days.
AMIN ET AL CLOSURE OF VSD WITHOUT CPB
151
Fig 5. Right ventricular septum with an MVSD and a PVSD. A wire has been inserted through the PVSD into the pulmonary artery (PA). An MVSD with a delivery catheter across the defect is also shown.
confirmed. A purse-string suture was placed around the catheter insertion site if there was bleeding. The pericardium was closed with interrupted sutures. The lung was
Closure of PVSD Closure of PVSD was attempted in 5 Yucatan miniature pigs. These pigs have naturally occurring PVSD similar to human beings. The VSD size ranged from 3 to 7 mm. Under general endotracheal anesthesia, the animal was placed on the operating table with its left side up. Using aseptic technique, the left chest was prepped and draped. A transverse incision was made and the chest was entered through the fourth intercostal space. The pleura was incised, and the lung was retracted with the help of malleable retractors. The pericardium was opened parallel to the phrenic nerve, and stay sutures were applied to retract the pericardium. An epicardial echocardium was performed to delineate the size of the defect and to measure the distance from the superior edge of the VSD to the aortic valve (Fig 4). The free wall of the left ventricle was punctured with an 18-gauge needle. With the needle pointing toward the VSD, a J-tipped guidewire was introduced through the needle into the left ventricle and across the VSD into the right ventricle. The wire was advanced to the main pulmonary artery (Fig 5). The needle was removed and the delivery sheath was introduced over the wire to the right ventricle. The wire was removed; the appropriate size device was screwed on the delivery cable, advanced through the delivery sheath, and deployed as outlined in Figure 6. The device was released once its stability was
Fig 6. Deployment of the PVSD device. (a) The delivery catheter advanced to the right ventricle. The wire is not shown. (b) The right ventricular disc is deployed. (c) The left ventricular disc is deployed.
152
AMIN ET AL CLOSURE OF VSD WITHOUT CPB
inflated and a chest tube was placed. The chest was closed in the standard fashion. The pig was extubated in the operating room. Analgesia and antibiotics were used for at least 3 days.
Follow-up All animals were followed for a period of 3 months. An echocardiogram was obtained after 1 week and 1 month postclosure. Cardiac catheterization and angiography was performed after 3 months. Under general endotracheal anesthesia, the animal was prepped and draped. A cut-down was performed on the right or the left groin. The femoral vessels were exposed. A 5-F Berman catheter (Arrow International Inc, Reading, PA) was advanced through the femoral vein to the right heart structures to obtain hemodynamics. A 5-F pigtail catheter (Cook, Bloomington, IN) was advanced through the femoral artery into the left ventricle; saturations and left ventriculogram were obtained. Catheters were removed and the animal killed.
Pathology The heart and the lungs were removed and sent for gross pathologic examination.
Results MVSD Perventricular closure of MVSD was attempted in 5 dogs. All had successful placement of the device. The closure rate was 100% after placement. There was no shunt by epicardial echocardiogram. Cardiac catheterization performed 3 months after placement of the device revealed no shunt by oximetry or by left ventricular angiogram.
PVSD Perventricular closure of PVSD was attempted in 5 pigs. Placement was successful in all animals. The closure rate was 80% immediately after deployment and 60% (3 of 5) after 3 months. Mild tricuspid regurgitation developed in 3 pigs. Aortic valve remained competent in 4 of 5 pigs at 3 months follow up (Fig 7). There were no occurrences of atrial arrhythmia, ventricular arrhythmia, or heart block. One pig developed residual VSD at the 3-month follow-up.
Ann Thorac Surg 1999;68:149 –54
clot was seen at the inferior margin of the left ventricular side of the device. In 2 other pigs, the microscrew on the left and right ventricular side was not covered with neoendocardium, although the disc was completely covered. The results were comparable with our previous report, where percutaneous technique was used to close PVSD [8]. In the pig that developed aortic insufficiency, the aortic rim of the concentric device was found to impinge upon the noncoronary cusp of the aortic valve. The device had tilted from its original position.
Comment Recent reports of minimally invasive techniques to close atrial and ventricular septal defect are encouraging [2, 3]. In the pediatric population, minimally invasive techniques have some drawbacks, at least for now. Femoral vessels are subjected to trauma to avoid a larger incision on the chest. The risks of cardiopulmonary bypass [9] and possible blood transfusion persist. The procedure cannot be applied in infants. On the other hand, perventricular closure is a collaborative approach between the surgeon and the cardiologist to circumvent the above problems, especially in smaller patients. In a previously published report of intraoperative closure of VSD on CPB bypass [10], the results were disappointing because of high mortality and residual defects. Transesophageal or epicardial echocardiography was of little help to assess residual shunt in a flaccid heart. The delivery system used by the authors was rigid and the device was not self-centering, whereas, the delivery system of the Amplatz VSD device is relatively flexible and small (6 or 7-F). The shape of the device is circular, it closes the defect by stenting the hole, and therefore oversizing is not required. There is no danger of disc separation and the flange size is small (5 to 7 mm). The mortality rate after repair of isolated VSD ap-
MVSD Pathology Gross pathological examination of the device revealed complete expansion of both discs. The device was covered with a thin, gray lining of neoendocardium on the left and the right ventricular side. The neoendocardial layer was continuous with the surrounding endocardium. The results were comparable with our previous report, where this device was used to close MVSD by percutaneous approach [6].
Paramembranous VSD Pathology Gross pathologic examination of the heart revealed the device covered with neoendocardium. In 1 pig, organized
Fig 7. Ascending aortogram at 3-month follow-up shows competent aortic valve. The paramembranous device (white arrow) is in place.
Ann Thorac Surg 1999;68:149 –54
proaches zero [11]. However, the mortality can be as high as 30% if the defects are multiple or associated with other cardiac anomalies [12]. In some recent studies, location of the defect and residual shunts were significant risk factors for intraoperative death and reoperation [13, 14]. Perventricular technique is a simplified way of approaching difficult to close residual defects. The primary difference between the previous and this study is that the defects were closed directly through the free wall of the ventricle without CPB. This technique decreases the chance of major complications should the device embolize during the procedure. There is no exposure to radiation and the procedure time is short. In infants, percutaneous catheter closure of septal defects is difficult because suitable catheters are not yet available. In one study [15], attempted closure of PVSD and MVSD resulted in a failure rate of 28%. The youngest patient was 4 years old, although the delivery system was 7–9-F. Other devices may require larger delivery system making catheter closure undesirable. In this series, closure of PVSD through the left ventricle was successfully accomplished via left thoracotomy. This step will help avoid CPB and possibly a future operation in patients who have coarctation of aorta with VSD. Results of paramembranous VSD closure were disappointing because of incomplete endothelialization and residual shunt in 2 pigs. These results are comparable with other series [16]. The following factors may have contributed to these results. (1) The thickness of the rim around the VSD is uneven, which may make the device unstable. (2) Chordal attachment on the right ventricular side can hinder expansion of the right disc and pull the waist of the device toward the right ventricle. This may result in residual shunt. (3) The length of the waist of the device was larger than the thickness of the septum, separating the discs from the septum. This probably led to incomplete endothelialization and tilting of the device. (4) None of these animals received aspirin or any other antiplatelet medication that may have contributed to clot formation on one device. We strongly believe that after placement of the device, aspirin therapy should be initiated for at least 6 months to prevent thromboembolization. (5) The design of the device may have influenced our results. Since two types of PVSD devices (eccentric and concentric) were used, an objective comparison could not be made because of small number. However, in another study [8], the eccentric device had fewer complications when compared with a concentric device. In this study, the pig with the eccentric device did not have residual shunt or aortic regurgitation, but did develop mild tricuspid insufficiency. An ideal paramembranous device will have all the qualities of Amplatzer device and
AMIN ET AL CLOSURE OF VSD WITHOUT CPB
153
possibly an uneven and shorter waist. This would prevent tilting of the device. The shorter waist will enhance endothelialization by approximating the discs to the septum. In summary, perventricular closure of VSD is feasible in MVSD and PVSD. This approach has been successfully applied in 1 baby [7]. The procedure can be performed without CPB with little exposure. More animal studies with longer follow-up are necessary, especially for the PVSD device.
References 1. Mitchell SC, Korones SB, Berendes HW. Congenital heart disease in 56,109 births. Circulation 1971;43:323–32. 2. Chang C, Lin PJ, Chu JN, et al. Video-assisted cardiac surgery in closure of atrial septal defect. Ann Thorac Surg 1996;62:697–70. 3. Lin PJ, Chang C, Chu J, et al. Minimally invasive cardiac surgical techniques in the closure of ventricular septal defect: an alternative approach. Ann Thorac Surg 1998;65:165–70. 4. Sharafuddin MJ, Gu X, Titus JL, Urness M, Cervera-Ceballos J, Amplatz K. Transvenous closure of atrial septal defects: preliminary results with a new self-expanding Nitinol prosthesis in a swine model. Circulation 1997;95:2162– 8. 5. Thanapoulos BD, Laskari CV, Tsaousis GS, Zarayelyan A, Vekiou A, Papasoulos GS. Closure of atrial septal defect with the Amplatzer device: preliminary results. J Am Coll Cardiol 1998;31:1110– 6. 6. Amin Z, Gu X, Berry JM, et al. A new device for closure of muscular ventricular septal defects in a canine model. Circulation 1999;(in press). 7. Amin Z, Berry JM, Foker JE, Rocchini AP, Bass JL. Intraoperative closure of muscular ventricular septal defect in a canine model and application of the technique in a baby. J Thorac Cardiovasc Surg 1998;115:1374– 6. 8. Gu X, Han Y, Amin Z, et al. Transcatheter closure of membranous ventricular septal defect with a new Nitinol prosthesis [abstract]. Circulation 1998;98(Suppl I):753. 9. Ferry PC. Neurologic sequelae of open-heart surgery in children. Am J Dis Child 1990;144:369 –73. 10. Fishberger SB, Bridges ND, Keane JF, et al. Intraoperative device closure of ventricular septal defects. Circulation. 1993;88(Part 2):205–9. 11. Kirklin JW, Barratt-Boyes BG, eds. Cardiac Surgery. New York: Churchill Livingstone, 1993:789–90. 12. Stark J, Sethia B. Closure of ventricular septal defect in infancy. J Cardiac Surg 1986;1:135–50. 13. Serraf A, Roux D, Lacour-Gayet F, et al. Reoperation after the arterial switch operation for transposition of great arteries. J Thorac Cardiovasc Surg 1995;110:892–9. 14. Kleinert S, Sano T, Weintraub RG, Mee RB, Karl T, Wilkinson JL. Anatomic features and survival strategies in doubleoutlet right ventricle. Circulation 1997;96:1233–9. 15. Sideris EB, Walsh KP, Haddad JL, Chen C, Ren SG, Kulkarni H. Occlusion of congenital ventricular septal defects by the buttoned device. Heart 1997;77:276–9. 16. Rigby ML, Redington AN. Primary transcatheter umbrella closure of perimembranous ventricular septal defect. Br Heart J 1994;72:368–71.
DISCUSSION DR BENSON R. WILCOX (Chapel Hill, NC): It was not clear to me how this device is fixed. The fixation ring, does it have teeth that are into the ventricular septum or is it just a question of a balance of pressures against it?
DR AMIN: It is a double disk device, which corresponds to the thickness of the ventricular septum. It is made of Nitinol; it is self-expanding. The width of the device, the area which fixes it, and the disks on both sides will prevent its embolization.