Transcatheter closure of congenital ventricular septal defects: experience with various devices.

Transcatheter closure of congenital ventricular septal defect (VSD) using various devices is gaining acceptance in selected cases of perimembranous and muscular defects, avoiding the inherent risks of cardiopulmonary bypass. The procedure was attempted in 137 patients having congenital defects using Rashkind Umbrella Device (RUD) in 29 patients, Amplatzer ventricular septal occluder (AVSO) in 107 patients, and Detachable Coil in one. All patients were selected using stringent criteria by detailed transthoracic echocardiography and/or transesophageal echocardiography. The location of VSD was perimembranous in 91 patients and was muscular trabecular in 46 patients. Seven patients had left ventricle (LV) to right atrium (RA) communication. Thirty-five patients with perimembranous and two with muscular VSD had aneurysm formation. The patients were 3 to 33 years old, and the diameter of VSD ranged from 3 to 12 mm. The pulmonary to systemic flow ratio was > or = 2:1 in 47 (34.3%) patients. The procedure was successful in 130 (94.8%) patients, with a success rate of 86.2% with RUD and 97.1% with AVSO. Residual shunt at 24 hours was seen in eight (32%) patients with RUD and in one patient (0.9%) with AVSO. Three (2.8%) developed transient bundle branch block, and two (1.9%) patients had complete heart block. New tricuspid stenosis and tricuspid regurgitation was observed in one patient each with AVSO. After immediate balloon dilatation, the mean pressure gradient across tricuspid valve decreased from 11 to 3 mmHg in the patient with tricuspid stenosis. On a follow-up of 1 to 66 (mean 35.2 +/- 10.7) months, the device was in position in all. None developed late conduction defect, aortic regurgitation, infective endocarditis, or hemolysis. At 9-month follow-up, the mean pressure gradient across the tricuspid valve was 3 mmHg in the patient with tricuspid stenosis. Complete occlusion of the shunt was achieved in 129(99.2%) patients. One patient with RUD having persistent residual shunt underwent a second procedure with AVSO. Three out of 107 patients with AVSO had an unsuccessful procedure where the defect was perimembranous with a superior margin of defect less than 3 mm away from the aortic valve, and the specially designed perimembranous AVSO had to be retrieved because of hemodynamic compromise due to significant acute aortic regurgitation, whereas in all others, the defect was either > or = 3 mm away from the aortic valve or had aneurysm formation. All seven patients with LV to RA communication showed complete abolition of the shunt. Thus, in properly selected cases of perimembranous and muscular ventricular septal defects, the transcatheter closure is safe and efficacious using appropriate devices. The success rate is higher with AVSO compared with the previously used devices, as well as more successful for the muscular defects than those that are perimembranous in location.