Percutaneous Closure of Postoperative and Post-Traumatic Ventricular Septal Defects


Carlos A.C. Pedra, MD, Sérgio C. Pontes, Jr, MD, Simone R.F. Pedra, MD, *Lucia Salerno, MD,
*J. Breno Sousa, MD, Marly A. Miaira, MD, Ana Luisa Guerra, MD, M. Virginia T. Santana, MD,
M. Aparecida Silva, MD, Valmir F. Fontes, MD

Surgery has been regarded as the gold standard method of treatment for all kinds of ventricular septal defects (VSDs).1 The incidence of postoperative residual shunts varies according to the type of defect being repaired, with malalignment (tetralogy of Fallot), muscular and postinfarction types probably producing the higher rates.1 Although some of these leaks are restrictive and well tolerated, especially after perimembranous VSD repair,2 they may result in significant left-to-right shunting with persistent left ventricular volume overload, which requires reintervention.1,2 Post-traumatic defects after chest trauma are rarer, but can be even more problematic.3 Recent publications have reported the feasibility, safety and efficacy of percutaneous closure of native muscular and perimembranous defects using Amplatzer VSD devices (AGA Medical Corp., Plymouth, Minnesota).4–10 Controlled release coils made of a reinforced nitinol wire (PFM, Product fur Medizine, Germany) have also been employed for the same purposes with encouraging results.11,12 However, there are few reports in the literature of transcatheter closure of postoperative and post-traumatic VSDs.13–18 In this paper, we describe 3 patients with such defects who underwent percutaneous closure using Amplatzer devices.

Case 1. This patient was a non-dysmorphic 4-year-old boy (weighing 15 kg) with a large perimembranous ventricular septal defect who was referred for surgical repair. A transthoracic Doppler echocardiogram revealed a 14 mm perimembranous VSD with inlet extension (Figure 1A). There was significant left atrial and ventricular volume overload. Pulmonary arterial pressure was estimated at systemic levels. In the catheterization laboratory, the pulmonary vascular resistance, the Qp/Qs and the pulmonary vascular resistance/systemic vascular resistance were estimated at 3.6 U Wood X m-2, 2.9 and 0.23, respectively, with further improvements after the administration of nitric oxide. Left ventriculography in the long-axial view showed a large perimembranous defect measuring 13 mm at its maximal diameter (Figure 2A). Pulmonary arterial angiography demonstrated progressive tapering of the pulmonary arteries with satisfactory opacification of the peripheral vessels. Surgical repair was carried out using standard techniques. Postoperatively, signs of persistent left-to-right shunting were detected invasively and confirmed by echocardiography. A 5–6 mm residual defect was observed through the upper portion of the patch near the atrioventricular valves (Figure 1B). On the 8th postoperative day, the child underwent reoperation; however, the surgeon was unable to identify the site of the residual leak. After a month, a decision was made to attempt closing the residual leak using transcatheter techniques. Informed consent was obtained from the parents. Under general anesthesia, vascular access via the groin was obtained. Hemodynamics showed mild residual pulmonary arterial hypertension and the Qp/Qs was estimated at 1.8:1. Left ventriculography confirmed the echocardiographic findings (Figure 2B). After establishing an arterial-venous loop using a 260 cm GlideWire (Terumo Cardiovascular Systems Corp., Somerset, New Jersey), a 7 Fr Flexor braided sheath (Cook Cardiology,Bloomington, Indiana) was advanced to the ascending aorta across the defect from the vein. Under transesophageal echocardiographic (TEE) guidance, the latest version of a premounted PFM coil (with reinforced nitinol wire and Dacron fibers, 12 x 6 mm) was delivered into the defect. Final position of the coil was assessed by echocardiography (Figure 1C) and angiography (Figure 2C). Because the left ventricular loops protruded into the left ventricular outflow tract and there was significant residual leak, the device was recaptured inside the long sheath and removed from the body. A decision was then made to attempt closing the defect using an 8 mm Amplatzer membranous VSD occluder (AGA Medical Corp., Plymouth, Minnesota). Good device position was confirmed by both echocardiography and angiography. Aortic and atrioventricular valve functions were preserved. After device release, a tiny residual leak was seen through the superior portion of the device (Figures 1D and 2D). The patient was discharged home the following day on aspirin (5 mg/kg/day). After a year, he has been asymptomatic on no medications. Serial transthoracic echocardiograms (TTE) revealed complete closure of the defect with no aortic insufficiency or tricuspid regurgitation. The child remained in sinus rhythm with no signs of left or right bundlebranch block.

Case 2. A 52-year-old male who received a gun-shot to the thorax was admitted to the emergency room with hemorrhagicshock. Via lateral thoracotomy, the posterior left ventricular and the anterior right ventricular walls were sutured. On the following day, a harsh and loud systolic regurgitant murmur was heard on the left-lower sternal border. TTE revealed an 8–10 mm midmuscular VSD with significant left ventricular volume overload and mild-to-moderate left ventricular dysfunction. He developed a systemic inflammatory response syndrome with multiple organ failure requiring prolonged inotropic and mechanical ventilatory support and peritoneal dialysis. His clinical condition improved slowly but steadily, and he was discharged home 1 month after admission. Due to the hemodynamic stability at that time, a decision was made to attempt closing the defect using transcatheter techniques. Cardiac catheterization was performed under general anesthesia 2 months after hospital discharge. Hemodynamics showed the following pressures (in mmHg): right atrium: 10; right ventricle: 54/10; main pulmonary artery: 54/20; aorta: 100/60; left ventricle: 100/18. The Qp/Qs was estimated at 2.0:1. Left ventriculography showed a 12 mm defect (Figure 3). Because he had risk factors for coronary artery disease, selective angiograms were also performed, which demonstrated severe lesions in the mid-to-distal right coronary artery. Under TEE guidance and through 9 Fr jugular venous access, a 14 mm Amplatzer muscular VSD device was successfully implanted in the defect using previously published techniques10 (Figure 3). There was immediate closure of the defect confirmed by echocardiography and angiography. This was followed by successful coronary stent implantation. After the procedure, the patient developed a retroperitoneal hematoma which was managed clinically. After a year, he has been asymptomatic on aspirin, with no major adverse coronary event. Repeat echocardiograms demonstrated complete closure of the defect, normal left ventricular function and dimensions, and normal estimated right ventricular pressure.

Case 3. An 18-year-old adolescent suffered a severe blunt chest trauma after a car accident. He was initially managed at a local hospital, and after clinical stabilization, he was referred to our center for further cardiac assessment. Neurological and abdominal lesions were discarded. A harsh and loud systolic regurgitant heart murmur was heard at the left lower sternal border. A TTE demonstrated a 30 mm mid-muscular VSD with outlet extension and significant left ventricular enlargement. Pulmonary arterial pressure was estimated at two-thirds of systemic. The defect was patch closed using standard surgical techniques 4 weeks after the trauma. The patient had an uneventful recovery, however, the systolic regurgitant heart murmur was still heard upon discharge, albeit with less intensity and with a higher pitch. Repeat echocardiograms showed a well-positioned pericardial patch with 2–3 residual leaks directed toward the right ventricular outflow tract and mild ventricular volume overload. Due to the stable hemodynamic conditions, a decision was made to attempt closing the defect using transcatheter techniques on an elective basis. The interventional catheterization was performed 5 months after hospital discharge. Under general anesthesia, normal pulmonary artery pressures were recorded and the Qp/Qs was estimated at 1.5:1. Left ventriculography in the long-axial view showed 2 residual leaks in the upper portion of the pericardial patch with significant opacification of the right ventricular outflow tract (Figure 4). The superior and inferior holes measured 4–5 mm and 2–2.5 mm, respectively. There was a 4–5 mm distance between both defects. The superior defect was crossed retrogradely and a 260 cm coronary angioplasty wire was advanced toward the left pulmonary artery where it was snared to establish the arterial- venous loop. Under TEE guidance and through the new 7 Fr braided Amplatzer® TorqVue® sheath (AGA Medical), an 8 mm Amplatzer muscular VSD device was successfully implanted in the larger defect with the right ventricular disc providing satisfactory coverage of the more inferior defect (Figure 4). The patient was discharged home on low-dose aspirin for 6 months. A TTE performed on the following day demonstrated a 1–2 mm residual leak at the superior edge of the device and slight protrusion of the right disc toward the right ventricular outflow tract with no local gradient generation. Serial echocardiograms revealed complete closure of the residual leak at 1-year follow up.

Significant residual leaks after surgical VSD repair may result in persistent left ventricular volume overload, pulmonary arterial hypertension and a sustained risk of bacterial endocarditis. To avoid these complications, it is generally accepted that a Qp/Qs > 1.5 is an indication for reintervention.1,13 Our 2 patients with postoperative leaks fulfilled this criterion. However, reoperation requires a new sternotomy and cardiopulmonary bypass with its attendant morbidity, increased hospital stay and possible long-term neurological impairment. In addition, adequate exposure of a residual defect may not be achieved using the standard right atrial approach, as seen in one of our patients described here. The evolving experience with transcatheter closure of postinfarction19,20 and native perimembranous and muscular defects4–10 made this less invasive means an attractive alternative to manage this problem. Using double-umbrella devices, Preminger et al13 were likely the first to report an attempt to close percutaneously intramural residual interventricular defects after repair of conotruncal malformation. That peculiar type of defect and the inherent limitations of the device may have accounted for the disappointing results. More recently, Walsh and colleagues15 reported the results of transcatheter closure of various types of postoperative ventricular septal defects in 9 patients using Amplatzer devices. All procedures were successful, with abolition or significant decrease in the magnitude of the residual shunts and limited morbidity in all patients. These excellent outcomes can probably be explained on the basis of evolving experience and technique in a referral center, appropriate patient selection, better catheter equipment and use of better-designed devices. Giardini et al14 also reported transcatheter closure of a residual postoperative VSD after a Rastelli operation that resulted in a dramatic decrease in the right ventricular outflow tract pressure gradient, delaying reoperation for the obstruction.
In this brief report involving a very small series of patients, we showed that postoperative and post-traumatic VSDs can be closed using Amplatzer devices with good outcomes. In the first case described here, the residual defect was single and located at the superior edge of the patch toward the crux of the heart. Despite being close to the atrioventricular valves, we felt there was enough room surrounding the defect to accommodate a device without interfering with atrio-ventricular valve function. Also, we decided to attempt the transcatheter approach because the child had undergone unsuccessful reoperation. The PFM coil was employed initially in this case under a local study protocol to assess its safety and efficacy. Although clinical experience with it is still very limited11,12,16 (about 100 implantation procedures with the latest version; unpublished data; Dr. Trong-Phi Le, personal communication), it seems to work well for restrictive defects, especially those associated with aneurysm formation. In the case described here, the PFM coil remained in an inadequate position, protruding into the left ventricular outflow tract, probably because it approached the septum (patch) from above, coming from the aorta (inappropriate case selection for the type of device). After safe coil recapturing and removal, subsequent use of the Amplatzer membranous occluder was a natural choice. Initial clinical experience with this device has been encouraging.4–9 The rate of complete closure is high (> 90%), and aortic and tricuspid valve function are preserved, at least in the short- to mid-term.4–9 However, complete heart block seems to occur with an incidence of 1–3%.4–9 Whether this is related to the radial forces exerted by the central waist of the device on the defect edges, especially after oversizing, or due to an inflammatory reaction or the endothelialization process, is speculative. Other factors such as younger age and inletextension of the defect may play a role in the development of this complication. In our experience, complete heart block did occur between 3–6 months after the procedure in a single case out of 39 implantation procedures (unpublished personal data). In the case presented here, the Amplatzer device worked nicely because it approached the patch at a proper angle, coming from the left ventricular apex. Besides, the longer inferior portion of the left disc engaged well within the patch with the waist remaining inside the residual defect itself, which was likely responsible for the complete closure observed at follow up.
VSD formation after a penetrating chest trauma is a rare event, as few of these patients survive the wound, as our patient did (Case 2). Because of progressive deterioration of his clinical conditions, surgical or transcatheter repair of the defect was not carried out in the acute phase after the trauma. The increase in the defect size could be explained by the occurrence of ongoing tissue necrosis due to the associated involvement and compromise of the septal coronary arteries after the bullet perforated the interventricular septum. The resulting defect was very similar to a native mid-muscular defect. Performing the interventional procedure in the chronic phase after the trauma was an important factor in this case, as it is for patients with postinfarction defects.19,20 This bought sufficient time for the development of fibrotic scarring on the rims of the defect, facilitating adequate delineation of the defect on angiography and TEE, which was important for optimal device selection and fixation. Severe blunt trauma to the chest occasionally results in rupture of the interventricular septum,3 as seen in the third patient of this report.
Although percutaneous closure of such defects has been reported previously,17,18 the resultant defect in this patient was considered to be too large to be closed safely with currently available VSD Amplatzer devices. Because our patient had not sustained additional injuries and was clinically stable, surgical repair was carried out uneventfully. In this patient, some transcatheter technical issues merit discussion. The use of a coronary wire was necessary to negotiate the passage through the defect and the right ventricular outflow tract to reach the left pulmonary artery for snaring. In order to close both adjacent holes with a single device, a slightly oversized device was selected, resulting in pinching of the central waist and mild protrusion of the right disc towards the right ventricular outflow tract, albeit with no local obstruction. However, based on the findings of the serial echocardiograms, this proved to be the right decision since there was spontaneous closure of the residual leak due to ongoing endothelialization. Although transcatheter closure of VSDs has been successfully performed under intracardiac echocardiography monitoring,21 which obviates the need for endotracheal general anesthesia in larger patients, this imaging modality was not available at the institutions in which the procedures were carried out.

In conclusion, percutaneous closure of post-traumatic and postoperative VSDs using Amplatzer devices was feasible, safe and effective in the selected patients presented here. More experience is warranted before the widespread use of this technique can be recommended.


Echocardiographic images of the defect. ( A) Four-chamber view (preoperative
tranthoracic echocardiogram [TTE]): Large perimembranous defect with inlet extension
measuring 14 mm at its maximal diameter. (B) Modified four-chamber view (postoperative


1. Kouchoukos NT, Blackstone EH, Doty DB, et al. Ventricular septal defect. In: Kouchoukos NT, Blackstone EH, Doty DB, et al (eds). Kirklin/Barrat- Boyes: Cardiac Surgery. London: Churchill Livingstone. 2003, pp. 850–910.

2. Tynan M, Anderson RH. Ventricular septal defect. In: Anderson RH, Baker EJ, Macartney FJ, et al (eds). Paediatric Cardiology. London: Churchill Livingstone, 2002, pp. 983–1014.

3. Tiao GM, Griffith PM, Szmuszkovicz JR, Mahour GH. Cardiac and great vessel injuries in children after blunt trauma: An institutional review. J Pediatr Surg 2000; 35: 1656– 1660.

4. Hijazi ZM, Hakim F, Haweleh AA, et al. Catheter closure of perimembranous ventricular septal defects using the new Amplatzer membranous VSD occluder: Initial clinical experience. Catheter Cardiovasc Interv 2002; 56: 508– 515.

5. Thanopoulos BD, Tsaousis GS, Karanasios E, et al. Transcatheter closure of perimembranous ventricular septal defects with the Amplatzer asymmetric ventricular septal defect occluder: Preliminary experience in children. Heart 2003; 89: 918– 922.

6. Bass JL, Kalra GS, Arora R, et al. Initial human experience with the Amplatzer perimembranous ventricular septal occluder device. Catheter Cardiovasc Interv 2003; 58: 238– 245.

7. Pedra CA, Pedra SRFF, Esteves CA, et al. Percutaneous closure of perimembranous ventricular septal defects with the Amplatzer device: technical and morphological considerations. Catheter Cardiovasc Interv 2004; 61: 403– 410.

8. Fu YC, Bass J, Amin Z, et al. Transcatheter closure of perimembranous ventricular septal defects using the new Amplatzer membranous VSD occluder: Results of the U.S. phase I trial. J Am Coll Cardiol 2006; 47: 319– 325.

9. Hijazi ZM. Device closure of ventricular septal defects. Catheter Cardiovasc Interv 2003; 60: 107– 114.

10. Holzer R, Balzer D, Cao QL, et al. Device closure of muscular ventricular septal defects using the Amplatzer muscular ventricular septal defect occluder: Immediate and mid-term results of a U.S. registry . J Am Coll Cardiol 2004; 43: 1257– 1263.

11. Le TP, Vaessen P, Freudenthal F, et al. Transcatheter closure of subaortic ventricular septal defect (VSD) using a nickel-titanium spiral coil (NitOcclud): Animal study and initial clinical results. Progr Pediatr Cardiol 2001; 14: 83– 88.

12. Ewert P, Kretschmar O, Peters B, et al. Transcatheter closure of congenital ventricular septal defects. Z Kardiol 2004; 93: 147– 155.

13. Preminger TJ, Sanders SP, van der Velde ME, et al. Intramural residual interventricular defects after repair of conotruncal malformations. Circulation 1994; 89: 236– 242.

14. Giardini A, Donti A, Gargiulio G, et al. Transcatheter residual ventricular septal defect closure after Rastelli operat ion. Cathe te r Cardiovasc Interv 2005; 64: 209– 212. 15. Walsh MA, Coleman DM, Oslizlok P, Walsk KP. Percutaneous closure of postoperative ventricular septal defects with the Amplatzer device. Catheter Cardiovasc Interv 2006; 67: 445– 451.

16. Michel-Behke I, Le TP, Waldecker B, et al. Percutaneous closure of congenital and acquired ventricular septal defects – Considerations on selection of the occlusion device. J Interv Cardiol 2005; 18: 89– 99.

17. Bauriedel G, Redel DA, Schmitz C, et al. Transcatheter closure of a posttraumatic ventricular septal defect with an Amplatzer occluder device. Catheter Cardiovasc Interv 2001; 53: 508– 512.

18. Cowley CG, Shaddy RE. Transcatheter treatment of a large traumatic ventricular septal defect. Catheter Cardiovasc Interv 2004; 61: 144– 146.

19. Lock JE, Block PC, McKay RG, et al. Transcatheter closure of ventricular septal defects. Circulation 1988; 78: 361– 368.

20. Holzer R, Balzer D, Amin Z, et al. Transcatheter closure of postinfarction ventricular septal defects using the new Amplatzer muscular VSD occluder: Results of a U.S. registry. Catheter Cardiovasc Interv 2004; 61: 196– 201.

21. Cao QL, Zabal C, Koenig P, et al. Initial clinical experience with intracardiac echocardiography in guiding transcatheter closure of perimembranous ventricular septal defects: Feasibility and comparison with transesophageal echocardiography. Catheter Cardiovasc Interv 2005; 66: 258– 267.

Add new comment

Back to top