Transcatheter Closure of Postoperative Residual Ventricular Septal Defects Using Amplatzer-Type Perimembranous VSD Occluders

Abstract: Objective. The reoperations of postoperative residual ventricular septal defects (VSDs) are associated with higher risks. Our aim is to assess the efficacy and safety of transcatheter closure of postoperative residual VSDs using perimembranous VSD occluders. Methods. Twenty-one patients with residual VSDs underwent transcatheter closure in our center from January 2005 to January 2012. The study population consisted of 9 males and 12 females whose ages ranged from 1.9 to 54 years (median age, 8.7 years). Eighteen cases had perimembranous VSD repair previously, 3 cases had tetralogy of Fallot surgical treatment. All patients had signs of left ventricle volume overload (Qp/Qs ≥1.5). Two types of perimembranous VSD occluders, symmetric and asymmetric, were used in 14 and 7 cases, respectively. The diameter of residual VSDs ranged from 4-16 mm (mean, 7.8 mm). The waist size of occluders ranged from 6-18 mm. Results. There were 0 deaths and 1 serious adverse event. Intravascular hemolysis occurred in 1 patient (4.8%), lasted for 7 days, and recovered with therapy. A trivial intraprosthetic residual shunt was observed in 2 patients (9.5%) after the procedure and 1 patient (4.8%) at 6 months. Two patients (9.5%) had transient left anterior hemiblock and recovered within the first week after the procedure. At the latest follow-up, no atrioventricular block and new-onset aortic regurgitation occurred. Conclusions. Transcatheter closure is a feasible and safe management option for patients with postoperative residual VSDs and obviates the need for a second surgery and cardiopulmonary bypass.

J INVASIVE CARDIOL 2013;25(8):402-405

Key words: transcatheter closure, occluder, residual ventricular septal defect

___________________________________

Postoperative residual ventricular septal defects (VSDs) are relatively common complications that may occur after surgery for isolated VSDs or complex cardiac diseases due to patch dehiscence, suture disruption, incomplete closure of the defect, or bacterial endocarditis. The incidence of postoperative residual VSD varies according to the type of defect being repaired, ranging from 5%-25%.^1,2 Although some of these residual VSDs are restrictive and well tolerated, they may result in left-to-right shunting, persistent left ventricular volume overload and elevated pulmonary vascular resistance, which require reintervention.³ Surgical repair remains the mainstay of treatment for postoperative residual VSD. However, reoperation is associated with higher risk and mortality; for instance, hemodynamic instability, the recurrence of the lesion, or cardiopulmonary bypass. In recent years, transcatheter closure techniques have developed; meanwhile, several kinds of devices have been applied for VSD closure, including buttoned devices, controlled release coils, and Amplatzer muscular occluders (AGA Medical Corporation).^4-6

However, there were few reports in the literature concerning transcatheter closure of postoperative residual VSDs, especially closure using perimembranous VSD occluders. In this report, we describe 21 patients with postoperative residual VSDs who underwent percutaneous closure using Amplatzer-type perimembranous VSD occluders, including symmetric and asymmetric occluders.

Methods

Patients. During the period of January 2005 to January 2012, a total of 21 patients with postoperative residual VSDs underwent transcatheter closure in our center. The study population consisted of 9 males and 12 females, whose ages ranged from 1.9 to 54 years (median age, 8.7 years), weighing from 11 to 69 kg (median weight, 21.6 kg). Concerning the type of defect repaired previously, 18 cases had perimembranous VSD repair; 3 cases had tetralogy of Fallot surgical treatment. One patient combined with atrial septal defect (ASD), and 1 with patent ductus arteriosus (PDA). All patients were assessed by transthoracic echocardiography (TTE) and had signs of left ventricle volume overload, with Qp/Qs ranging from 1.5-3.5 (median Qp/Qs, 2.1). The demographic and diagnostic data were listed in Table 1.

Perimembranous VSD occluders. A modified perimembranous VSD occluder (Shanghai Shape Memory Alloy) based on the Amplatzer occluder (AGA Medical Corporation) was used in this study.⁷ Unlike the Amplatzer occluder, there were two types of perimembranous VSD occluders, symmetric and asymmetric, which were used in 14 and 7 cases, respectively (Table 2). The right disks are identical in both types and 4 mm larger in size than the waists. The difference lies in the shape of the left disks. As to the asymmetric occluders, the left disks are secund toward the cardiac apex with no superior rim extending toward the aortic valve ring and 6 mm larger in size than waists. As for the symmetric occluders, they can be divided into several subtypes. Among them, two subtypes commonly used are A2B2 and A4B2 occluders, in which the left disks are symmetric and 4 mm and 8 mm larger in size than the waists, respectively.⁷ In recent years, there also appear two complementary subtypes, A3B2 and A5B2 occluders, which are not commonly used clinically and were not used in our study. Of 14 cases with symmetric occluders, A2B2 and A4B2 occluders were used in 11 cases and 3 cases, respectively (Table 2).

Transcatheter procedure. Transcatheter closure of postoperative residual VSD was performed under general anesthesia for children (≤10 years old) and local anesthesia for adult patients and children (>10 years old). Patients were given 100 IU/kg heparin intravenously after femoral venous and arterial paracentesis. According to the standard techniques described previously,^8,9 implantation of the VSD occluder and standard angiography of left ventricle and ascending aorta were performed in all cases. Because most postoperative residual VSDs occurred in the margin of the patch and were irregular, it was usually very difficult to introduce the track wire through them. In some of these patients, a novel wire-maintaining technique was used, with which the track wire was maintained in the delivery sheath during implantation of occluders. If the initial occluder was inappropriate in size, the delivery sheath could be reintroduced through the maintained wire. The diameter of residual VSDs was calculated by TTE and angiographic measurement, and the devices were selected to be 1 to 2 mm larger in size than residual VSDs. Symmetric or asymmetric occluders were determined depending upon the distance from the aortic annulus and the form of residual defects. Successful closures were made when the devices were implanted in the correct position with no significant complications (such as significant valve regurgitation) or significant residual shunt. Procedural data and devices used are listed in Table 2.

All patients underwent clinical examination and electrocardiographic monitoring for 24 hours after closure, and were discharged from hospital 1 week later if no complications occurred. Aspirin was prescribed 3-5 mg/kg/day for 6 months post procedure. TTE was performed regularly before hospital discharge, at 1, 3, 6, and 12 months post procedure, and yearly thereafter. 

Data were expressed as frequency or percentage for nominal variables, as median (range) for categorical variables, and as mean ± standard deviation for continuous variables. SPSS 11.0 (SPSS, Inc) was used for the statistical computations.

Results

All closures were successfully performed, with no additional procedures required. In 3 patients, the residual VSD with aneurysm formation or multiple outlets was closed with only 1 symmetric occluder (A4B2). One patient combined with ASD, and 1 with PDA, both of which were closed at the same time. 

There were 0 deaths and 1 serious adverse event. Intravascular hemolysis occurred in 1 patient (4.8%) who had intraprosthetic shunt in the margin of the occluder and history of high blood pressure, which lasted for 7 days and recovered with therapy. A trivial intraprosthetic residual shunt was observed in 2 patients (9.5%) before hospital discharge and 1 patient (4.8%) at 6 months post procedure. Two patients (9.5%) had transient left anterior hemiblock and recovered within the first week post procedure. At the latest follow-up, no AVB occurred in any patient. Ultrasonographic results showed that all occluders were well in shape, and no new-onset aortic regurgitation was found.

Discussion

It is well known that postoperative residual VSDs may result in persistent left ventricular volume overload, pulmonary hypertension, and risk of bacterial endocarditis. The impairment is usually determined by the amount of left-to-right shunt, which is especially larger following repair of tetralogy of Fallot and poorly tolerated. According to previous literature, it is generally accepted that a Qp/Qs ≥1.5 is an indication for reoperation.¹⁰ In this study, all patients with postoperative residual VSDs fulfilled this criterion. Although surgical repair is regarded as the standard method of treatment for all kinds of VSDs, it faces the challenges from postoperative residual VSDs, which are often significantly difficult to repair and associated with higher risk and mortality; for instance, hemodynamic instability, the recurrence of the lesion, or cardiopulmonary bypass. Recently, transcatheter closure has become an alternative strategy for treatment of postoperative residual VSD. Previous studies have proven the feasibility of closing the residual defects with devices, such as buttoned devices, controlled release coils, and Amplatzer muscular occluders. In this study, we described our experience with transcatheter closure of postoperative residual VSDs using a modified perimembranous VSD occluder.

In practice, residual VSDs mainly result from suture disruption, avulsion, or incomplete defect repair. Most of them occurred in the posteroinferior and superior margins of the patch, because the suture was easier for the surgeon to avulse in these areas. Due to the irregular form and margin of postoperative residual VSDs, the individual occluder should be chosen cautiously. In our study, we chose the occluders depending upon the form of residual defects and the distance between the aortic annulus and defects. If the residual defect occurred in the posteroinferior of the patch, a symmetric occluder (A2B2) would be chosen to close the shunt. In particular, when the residual defect was in the form of an aneurysm with the inlet larger than the outlet, or had multiple outlets in the right side, an A4B2 occluder was the most suitable device because of the larger left area to be covered. One of the most serious problems in transcatheter closure of residual defect is aortic regurgitation. If the residual defect was located at the superior edge of the patch and was close to the aortic annulus with a subaortic rim of 2 mm or less, an asymmetric occluder would be used. On one hand, the asymmetric device has no superior end, so it prevents interference with aortic valve function after implantation; on the other hand, the apical side of the left disk is relatively large and leans on the muscular septum, which ensures the stability of the device. In our series, symmetric (A4B2) and asymmetric occluders were used in 3 and 7 patients, respectively.

In contrast to normal VSDs, a common variation of postoperative residual VSD is the morphologic irregularity. We found that this significant variant was due to the previous patch repair in most of our patients. This variation might be associated with longer procedural times compared to normal VSD repair for several reasons. First, it is usually very difficult to introduce the track wire through residual VSD and establish the arteriovenous wire loop. Second, it is difficult to determine the diameter of residual VSD and choose the proper device size; if the central waist of the device appeared to be “pinched,” we should consider recapturing the device and using a smaller one. Third, more asymmetric occluders were used in patients with residual VSDs. The asymmetric occluder had to be repeatedly pushed and pulled back to ensure that the platinum marker on the left disk was kept toward the cardiac apex. Therefore, the wire-maintaining technique was used to avoid reestablishing the arteriovenous wire loop in some of our patients.

Hemolysis after VSD closure is a known complication. Usually, it is due to the presence of residual shunt. The mechanism is high-velocity blood flow past the device, which leads to mechanical fragmentation of erythrocytes.¹¹ It has been described as a rare and transient complication, for which conservative treatment is usually sufficient. In the literature, there have been only a few reports of hemolysis after VSD closure and no report of persistent hemolysis that required device removal.¹² One patient (4.8%) had intravascular hemolysis in our study. Due to fibrous formation, irregular margin of postoperative residual VSD, and worse shaping of occluder, the occurrence rate of hemolysis after closure might be higher in patients with residual VSDs. As to a trivial low-velocity residual shunt post procedure, it is not important and can close spontaneously. 

The most serious concern of perimembranous VSD closure is the occurrence of complete AVB (cAVB). The incidence of cAVB varied from 0%-5.7% in the literature.^13-16 The occurrence of cAVB is related to the proximity of the conduction system to the margins of the VSD. Various mechanisms have been suggested, eg, the occluder may cause direct compression trauma, provoke an inflammatory reaction, or form scarring in the conduction tissue.¹⁷ Because fibration or scar formation occurred in the margins of the VSD after previous operation, we thought that subsequent negative influence of the occluder mentioned above might decrease, and then the occurrence rate of cAVB after closure might be lower in patients with postoperative residual VSDs. In our series, there was no occurrence of cAVB. Only 2 patients (9.5%) had transient left anterior hemiblock and recovered within the first week post procedure. The physical stimulus of the conduction tissue caused by guidewire and catheter might be the major cause for left anterior hemiblock. 

Conclusion

Reoperations of residual VSDs are associated with higher risks and mortality. This study showed that transcatheter closures of postoperative residual VSDs were performed with few complications and successful results. We think that transcatheter closure is an effective and alternative management option to be used in patients with postoperative residual VSDs. Further analysis is required to evaluate its long-term safety and efficacy.

References

1. Bol-Raap G, Weerheim J, Kappetein AP, et al. Follow-up after surgical closure of congenital ventricular septal defects. Eur J Cardiothorac Surg. 2003;24(4):511-515.
2. Kirklin JW, Barrat-Boyes BG. Ventricular septal defect. In: Kouchoukos NT, Blacksgowe EH, Doty DB, Hawley FL, Karp RB, eds. Cardiac Surgery. 3rd ed. New York: Churchill Livingstone. 1993:850-910.
3. Pedra CA, Pontes SC, Pedra SR. Percutaneous closure of postoperative and post-traumatic ventricular septal defects. J Invasive Cardiol. 2007;19(11):491-495.
4. Sideris EB, Walsh KP, Haddad JL, Chen CR, Ren SG, Kulkarni H. Occlusion of congenital ventricular septal defects by the buttoned device: “Buttoned device” Clinical Trials International Register. Heart. 1997;77(3):276-279.
5. Latiff HA, Alwi M, Kandhavel G, Samion H, Zambahari R. Transcatheter closure of multiple muscular ventricular septal defects using Gianturco coils. Ann Thorac Surg. 1999;68(4):1400-1401.
6. Chessa M, Carminati M, Cao QL, et al. Transcatheter closure of congenital and acquired muscular ventricular septal defects using the Amplatzer device. J Invasive Cardiol. 2002;14(6):322-327. 
7. Qin YW, Chen JM, Zhao XX, et al. Transcatheter closure of perimembranous ventricular septal defect using a modified double-disk occluder. Am J Cardiol. 2008;101(12):1781-1786.
8. Rao PS. Perimembranous ventricular septal defect closure with the amplatzer device. J Invasive Cardiol. 2008;20(5):217-218.
9. Fu YC, Bass J, Amin Z, et al. Transcatheter closure of perimembranous ventricular septal defect using the new amplatzer membranous VSD occluder. Results of the US phase I trial. J Am Coll Cardiol. 2006;47(2):319-325.
10. Ammash NM, Warmes CA. Ventricular septal defects in adults. Ann Intern Med. 2001;135(9):812-824.
11. Spence MS, Thomson JD, Weber N, Qureshi SA. Transient renal failure due to hemolysis following transcatheter closure of a muscular VSD using an Amplatzer muscular VSD occluder. Catheter Cardiovasc Interv. 2006;67(5):663-667.
12. Martinez MW, Mookadam M, Mookadam F. A case of hemolysis after percutaneous ventricular septal defect closure with a device. J Invasive Cardiol. 2007;19(7):E192-E194.
13. Arora R, Trehan V, Kumar A, Kalra GS, Nigam M. Transcatheter closure of congenital ventricular septal defects. Experience with various devices. J Interv Cardiol. 2003;16(1):83-91. 
14. Masura J, Gao W, Gavora P, et al. Percutaneous closure of perimembranous ventricular septal defects with the eccentric Amplatzer device: multicenter follow-up study. Pediatr Cardiol. 2005;26(3):216-219. 
15. Yip WC, Zimmermann F, Hijazi ZM. Heart block and empirical therapy after transcatheter closure of perimembranous ventricular septal defect. Catheter Cardiovasc Interv. 2005;66(3):436-441. 
16. Carminati M, Butera G, Chessa M, Drago M, Negra D, Piazza L. Transcatheter closure of congenital ventricular septal defect with Amplatzer septal occluders. Am J Cardiol. 2005;96(12A):52-58. 
17. Li X, Li L, Wang X, Zhao HB, Zhang SY. Clinical analysis of transcatheter closure of perimembranous ventricular septal defects with occluders made in China. Chin Med J (Engl). 2011;124(14):2117-2122.