Homemade Candy Plug Using a Zenith Alpha Thoracic Stent-Graft for False Lumen Distal Occlusion in Acute-on-Chronic Type B Aortic Dissection

Introduction

In chronic type B aortic dissection, coverage of the proximal entry site alone during thoracic endovascular aortic repair (TEVAR) achieves complete false lumen thrombosis in only about 40% of cases because distal entry sites or branch vessels remain open to retrograde blood flow. ¹ Moreover, remodeling of the aorta is hampered by fibrosis in chronic aortic dissection, causing the distal false lumen site to remain patent. ²

The candy-plug technique developed by Kölbel’s group^2,3 is an effective means of achieving false lumen thrombosis by occluding distal retrograde flow. ³ Different types of stent-grafts have been customized into a candy plug, including Kölbel’s original design based on the Zenith TX2 Proform (Cook Medical, Bloomington, IN, USA), an Excluder aortic extender (W.L. Gore & Associates, Flagstaff, AZ, USA) by Ogawa et al, ⁴ and a Valiant Captivia FreeFlo (Medtronic, Santa Rosa, CA, USA) by Branzan and Schmidt. ⁵ This report details the use of a Zenith Alpha stent-graft (Cook Medical) as a candy plug to occlude the distal false lumen in a patient with a rapidly expanding complex chronic type B aortic dissection.

Case Report

A 50-year-old hypertensive male smoker had a complex chronic type B aortic dissection diagnosed in 2006 and underwent a left to right femoral-femoral bypass for acute right lower limb ischemia due to the dissection. He subsequently was lost to follow-up. He presented in September 2016 with chest pain, and computed tomography angiography (CTA) showed evolution of the lesion into a complicated chronic type B aortic dissection extending from distal to the origin of the left subclavian artery (LSA) to the left femoral artery. There was fusiform aneurysmal dilatation of the proximal descending aorta measuring up to 5.0 cm in diameter. Serial CTA 10 days later (Figure 1A-C) showed rapid enlargement of the false lumen; the aneurysmal dilatation of the proximal descending aorta had increased to 5.8 cm.

OPEN IN VIEWER

Figure 1.

Preoperative computed tomography angiography (CTA) scans showing (A) type B aortic dissection with the false lumen extending down to the left common femoral artery, (B) the 5.8-cm aneurysmal dilatation in the proximal descending aorta, and the proximal entry site distal to the left subclavian artery. (C) CTA 2 weeks after thoracic endovascular aortic repair showed a patent stent-graft with opacification in the false lumen, likely from retrograde flow (arrow).

The patient underwent left carotid–subclavian artery bypass with a 7-mm IMPRA graft (BD/Bard, Tempe, AZ, USA), TEVAR with a 34/30×209-mm tapered Zenith Alpha stent-graft (Cook Medical) to cover the proximal entry site. The left common carotid artery was revascularized with a 7×57-mm BeGraft covered stent (Bentley, Hechingen, Germany) as a chimney stent-graft, and the LSA was embolized with a 10-mm Amplatzer Vascular Plug 2 (AVP 2; Abbott Vascular, Redwood City, CA, USA). Completion angiography showed reduced flow into the false lumen.

Two weeks after the operation, the patient developed acute back pain; CTA showed persistent false lumen flow (Figure 1D) presumably from retrograde flow via a distal entry tear. The decision was made to occlude the distal large false lumen using the candy-plug technique.

On the back table, a 44×125-mm Zenith Alpha stent-graft was used for the candy plug. Three Ethibond 2/0 sutures (Ethicon, a Johnson & Johnson company, Somerville, NJ, USA) were tied around the middle stent to reduce the diameter to 10 mm (Figure 2A). The stent could not be resheathed due to barbs on the upper border of the first covered stent. The trigger wire was removed, allowing removal of the proximal bare stent and the first covered stent (Figure 2B), leaving the fabric intact (Figure 2C).

OPEN IN VIEWER

Figure 2.

To prepare the Zenith Alpha stent-graft, (A) 3 Ethibond 2/0 sutures were used to reduce the middle stent down to 10 mm in diameter. (B) The trigger wire, bare stent, and the first covered stent are removed, (C) leaving the fabric intact after stent removal.

The true lumen was accessed via the right common femoral artery (CFA) and a 30×155-mm Zenith Alpha stent-graft was introduced to extend the existing stent-graft to just above the celiac axis level. The false lumen was accessed via the left CFA. After placing a Lunderquist stiff wire (Cook Medical) in the false lumen, the candy plug was advanced into position with its lower edge corresponding to that of the stent-graft in the true lumen. The true lumen stent-graft was then deployed followed by deployment of the candy plug (Figure 3A). There was some difficulty in removing the delivery catheter as the nosecone was held at the constrained portion of the candy plug. Combining a push, pull, and twist maneuver to the delivery catheter was successful in retrieving the nosecone. The constrained portion was plugged with a 22-mm AVP2. Completion angiography showed mild proximal type Ia endoleak, most likely due to gutter leak. This was left alone as it was felt that it would stop spontaneously since the false lumen outflow was already occluded.

OPEN IN VIEWER

Figure 3.

(A) The candy plug (thick arrow indicates constrained portion) was advanced into position with the lower edge corresponding to that of the stent-graft in the true lumen (thin arrow). (B) Computed tomography angiography (CTA) after candy plug deployment showed persistent opacification in the false lumen of the proximal descending aorta, likely from type Ia endoleak (arrow). (C) CTA 1 month after embolization showed complete false lumen thrombosis of the descending aorta (thick arrow) above the candy plug (thin arrow). (D) CTA 2.5 years after surgery demonstrated the reduced size of the thrombosed false lumen (thick arrow). The candy plug (narrow arrow) remained in position.

Four days after the candy-plug procedure, the patient still complained of chest pain. CTA showed persistent opacification of the proximal descending aorta false lumen, likely due to a gutter leak giving rise to a type Ia endoleak (Figure 3B). The leak was embolized via a left brachial approach. A microcatheter was manipulated past the Amplatzer plug in the proximal LSA into the space between the proximal endograft, carotid chimney, and LSA. After coil embolization of the space between the chimney graft and the proximal endograft, there was still endoleak from inadequate sealing of the Amplatzer device in the proximal LSA. This leak was embolized successfully; a total of 18 coils [Interlock Coils (Boston Scientific, Marlborough, MA, USA) and Nester coils (Cook Medical)] were used for both leaks. Angiography after embolization showed residual minor type Ia endoleak, with no retrograde flow from the candy plug sealing site. A decision was made to leave the persistent endoleak because the procedure was prolonged, there was no ruptured aorta, and the flow was greatly lessened by the gutter embolization. The patient’s pain was alleviated, and he was monitored very closely after the operation. CTA at 1 month after embolization (Figure 3C) showed complete thrombosis of the false lumen in the descending aorta. A follow-up CTA 2.5 years after surgery (Figure 3D) showed that the candy plug was in position; there was no endoleak, and the thrombosed false lumen in the thoracic aorta was reduced in size (Figure 3C). The maximal size of the thoracic aorta was 5.1 cm at 1 month and 4.7 cm at 2.5 years after the operation.

Discussion

The candy-plug technique is a useful method of occluding a large distal false lumen in chronic aortic dissection.^2–7 Rohlffs et al ³ successfully deployed candy plugs in 18 patients with chronic thoracic aortic dissection; all but one had no persisting false lumen backflow. Others^5,7 have reported the successful use of the candy-plug technique in emergency treatment of false lumen rupture of chronic type B aortic dissection.

In this case we chose to use the Zenith Alpha stent-graft rather than the TX2 because the Alpha’s lower profile delivery catheter was more appropriate to the patient’s small CFA access. When modifying the Zenith Alpha into a candy plug, there were some difficulties in resheathing the device after applying the constraining ties, but the problem was solved by removing the metal stents. The other problem we did not foresee was infolding of the fabric at the constrained stent segment, which led to difficulty in retrieving the nosecone. One solution would be to tighten the constraining sutures with the nosecone sitting in the segment as a gauge.

Conclusion

The candy-plug technique is useful in facilitating complete occlusion of the false lumen in chronic aortic dissection, avoiding an open procedure. One of the learning points from our case is to use the nosecone as a gauge when tightening the constraining sutures on the homemade candy plug to prevent difficulty when retrieving the nosecone.