Percutaneous closure of a large aortic paravalvular leak using two duct occluder devices

Abstract

A 21-year-old male presented with severe aortic paravalvular leak. He had undergone three cardiac surgeries and also had chronic kidney disease. It was decided for a trans-catheter closure owing to the risks of a fourth surgery and co-morbidity. The device was sized based on angiogram, balloon sizing and two dimensional transesophageal echo. There was significant residual leak after deployment of first device. Hence the defect was re-crossed and two duct occluder devices were positioned across the leak from two arterial access. After confirming position and satisfactory reduction in paravalvular leak, the devices were released in tandem. There was near abolition of leak. The patient is asymptomatic at three months follow up. Larger paravalvular leaks are better addressed with two devices of smaller size rather than a single large device. Technical considerations while deploying multiple devices are discussed.

1. Clinical summary

A 21-year-old male presented with class III dyspnea on exertion. He had undergone surgical closure of peri-membranous VSD (ventricular septal defect) complicated by prolapse of RCC (right coronary cusp) and consequent moderate AR (aortic regurgitation) at the age of 2 years. He underwent surgical closure of residual VSD and aortic valve repair at the age of 16 years. AR progressed over years and he eventually required valve replacement at the age of 20 years. A 21 mm St Jude bi-leaflet mechanical prosthesis was implanted in the aortic position. Mild paravalvular leak was noted at three months follow up. Gradually his symptoms worsened to class III. He had cardiomegaly, peripheral signs of severe AR, a long early diastolic murmur and a pansystolic murmur. Transthoracic echocardiography showed a small residual VSD, severe paravalvular leak and a dilated LV. The prosthetic valve was stable, leaflets were moving normally and there was no significant valvular leak, gradient, vegetation or thrombus. The paravalvular leak was posterior, near the non coronary sinus. There was no clinical or laboratory evidence of infective endocarditis. He also had chronic kidney disease. The decision was to proceed with device closure of the aortic paravalvular leak, considering the high risk of a fourth surgery.

The procedure was done under general anesthesia using 2D (two dimensional) TEE (transesophageal echocardiography) guidance. A track could be identified on TEE posteriorly in the 110° left ventricular outflow tract view measuring 6 mm in breadth. Angiogram in left anterior oblique – cranial view showed a paravalvular tract with a large circumferential extent (Supplemental video 1). Balloon sizing was done using a 9 * 20 mm Tyshak 2 balloon (NuMED Inc., Hopkinton, NY) and orthogonal views were recorded. The waist measured approximately 6.5 mm. It was decided to go ahead with a 10 * 8 mm Amplatzer duct occluder1 (AGA Medical Corporation, Plymouth, MN) device. The tract was crossed retrogradely using a 5F Right Judkins 3.0 coronary diagnostic catheter over a 035 inch 150 cm Terumo guide wire (GlideWire™, Terumo Inc., Japan). The terumo wire was exchanged for a 260 cm J tipped Amplatzer superstiff Teflon guidewire (Amplatz Extra Stiff, Cook Inc., Bloomington, IN) and the tip positioned at the left ventricular apex. A 10 * 8 mm Amplatzer duct occluder was loaded and deployed using a 7F delivery sheath. The device was held in position without releasing. There was significant residual shunt. The residual defect was re-crossed from the contralateral femoral artery (Fig. 1). A 10 * 8 mm Lifetech duct occluder device (Lifetech Scientific, Shenzhen, China) was deployed adjacent to the first device (Supplemental video 2, Fig. 2). TEE images showed near complete occlusion. The devices were released in tandem (Fig. 2). The patient is asymptomatic three months post procedure.

Fig. 1.

Intra procedural image – the first duct occluder device has been deployed across the aortic paravalvular leak and is attached to its cable. The defect has been re-crossed, the wire and catheter used for re-crossing is seen in situ.

Fig. 2.

Intra procedural image – two duct occluder devices after deployment across the aortic paravalvular leak seen in right anterior oblique view. Both the devices are attached to their cables and are yet to be released. The leaflets of the prosthetic aortic valve are in open position.

Supplementary video related to this article can be found at http://dx.doi.org/10.1016/j.ihj.2013.04.020

The following are the supplementary video related to this article:

Supplemental video 1

Aortic root angiogram in left anterior oblique-cranial view using a pigtail catheter. The prosthetic valve is seen with severe paravalvular leak posteriorly, with a gradation of intensity of contrast, indicating the large circumferential extent. The image also shows an aneurysmal ventricular septal defect patch with mild residual shunt and aortic valve regurgitation (catheter induced).

Supplemental video 2

The second duct occluder device is being taken across the aortic paravalvular leak through a second sheath while the first device is in situ attached to its cable.

2. Discussion

Paravalvular regurgitation affects 5–17% of all surgically implanted prosthetic heart valves. Owing to the frequent morbidity at presentation and risks of re-do surgery, percutaneous closure is being widely preferred; although plagued by residual leaks, hemolysis, need of repeat procedures and device embolizations.¹

There is no dedicated device available as yet for paravalvular leak closure. Device selection is based on the operator's assessment of the anatomy based on available imaging modalities and availability of devices.^2,3 Closure of para-aortic leak using two amplatzer septal occluders⁴ and one amplatzer muscular VSD occluder and amplatzer vascular plug⁵ have been previously reported.

We used 2D TEE and fluoroscopy for guidance during procedure. Although the angiogram and TEE suggested that more than one device might be needed, it was decided to deploy the first device and reassess the need of a second. As the angiogram and TEE showed significant residual leak, the defect was re-crossed and a second device was deployed from the contralateral femoral artery. Both devices were released once angiogram and TEE confirmed significant reduction in leak. Our immediate result was excellent; there was no significant residual leak or impingement on valve leaflets (Fig. 2).

Percutaneous closure with single device is advocated for smaller holes. When the leak is from a bigger orifice, two smaller devices are a better choice over a single large device as the chances of the discs of the device interfering with valve movement is lesser in the former. It also decreases the risk of erosion of adjacent structures from oversized devices.² It is also worth achieving complete closure in the first setting to negate congestive heart failure and accelerated hemolysis. When planned, multiple device delivery can be done simultaneously or sequentially retaining the wire across the leak, without necessitating re-crossing.¹ While using two devices in the same setting, both the devices are released after the second device is deployed and elimination of leak is confirmed. This would minimize chances of dislodgement of the first device and facilitate retrieval if necessary. If available, real time three dimensional echocardiography overcomes many of the fallacies regarding procedural planning based on 2D imaging.¹

Conflicts of interest

All authors have none to declare.