Efficacy of the Stripped AFX Aortic Cuff as a Scaffolding Bare Stent to Facilitate the Expansion of the Thoracoabdominal and Visceral Aorta during Thoracic Endovascular Aortic Repair for Complicated Stanford Type B Aortic Dissection

Abstract

Purpose:

During thoracic endovascular aortic repair for complicated Stanford type B aortic dissection, large bare stent placement for the abdominal aorta is sometimes necessary. In smaller abdominal aortic diameter cases, we used the stripped AFX aortic cuff as a scaffolding bare stent rather than the Zenith Dissection Endovascular Stent, which is a commercially available, large bare stent. In this study, we evaluated the feasibility of the stripped AFX aortic cuff and experiments were conducted to compare the stripped AFX and the Zenith Dissection Endovascular Stent.

Material and Method:

The type B aortic dissection patients treated with thoracic endovascular aortic repair using stripped AFX at three institutions between January 2014 and December 2017 were retrospectively reviewed. Clinical data, including technical success, perioperative complication, and overall survival, were evaluated. The experiment assessed the chronic outward force that reflected the load acting on the artery wall from the stent.

Result:

Eight cases (seven males) were reviewed. The median (interquartile range, IQR) age of the patients was 60 years (46.3-70.3). The technical success rate was 100%, and no perioperative complications were observed. The median (IQR) follow-up period was 28.9 months (17.5-31.5). During the follow-up, one patient died of septic shock unrelated to aortic events. The median (IQR) diameter of the stripped AFX on the last follow-up CT was 23.5 mm (21.9-25.0). The chronic outward force of the Zenith Dissection Endovascular Stent was two to three times that of the stripped AFX.

Conclusions:

The stripped AFX aortic cuff is feasible and safe as a scaffolding stent during thoracic endovascular aortic repair for Stanford Type B aortic dissection.

Introduction

Thoracic endovascular aortic repair (TEVAR) has been widely used for Stanford type B aortic dissection (TBAD) [1]. The application of large bare stents in the thoracoabdominal and visceral aorta during TEVAR was first described in 2005 [2]. The technique is called provisional extension to induce complete attachment (PETTICOAT) [3]. The Zenith Dissection Endovascular Stent (TX-D stent, Cook Medical, Bloomington, IN, USA) (Fig. 1a) was invented for this purpose, and a prospective trial study reported the effectiveness and feasibility of the TX-D stent [4]. However, a TX-D stent has the smallest diameter of 36 mm, which is very large compared to the collapsed true lumen or the whole aorta. The potential risk of the excessive pressure on the aortic wall and following vascular damage such as stent graft-induced new entry is a concern when using the excessively oversized stent. In practice, TX-D stent misalignment possibly due to its excessive oversize has been reported [5].

Figure 1.

(a) Zenith TX-D stent with a diameter of 36 mm and length of 164 mm.

(b) Stripped AFX with a diameter of 28 mm and length of 95 mm.

We developed a novel technique to use the modified AFX (Endologix, Irvine, CA, USA) aortic cuff as a smaller scaffolding stent and conducted clinical and experimental studies to assess the effectiveness and safety of the stripped AFX.

Material and Methods

Patients and indication of device usage

The records of all patients with TBAD treated by TEVAR at three institutions between January 2014 and December 2017 were retrospectively reviewed. The records of the patients who were treated with TEVAR using stripped AFX were extracted. We obtained the informed consent to use the device designed for AAA from each patient. Ethical approval was obtained from the ethics committee of each institution. Because this study was a retrospective review of charts and radiographic images, patient consent for publication was not required. Instead, an opt-out method was alternatively employed.

Patient follow-up, including clinical examinations and CT images, was performed at 1 week, 6 months, and 12 months after the procedures and annually thereafter.

The indications of stripped AFX usage were the following: (a) The aortic true lumen was still narrow after the coverage of the primary entry by the stent graft, confirmed by intravascular ultrasound. (b) The whole aortic diameter (including the true and false lumens) that was measured on preoperative CT at the infrarenal level was less than 28 mm (if the whole aortic diameter was more than 28 mm, the TX-D stent was used).

Device creation

The AFX aortic cuff was deployed from a delivery sheath on the side table. The graft was cut to within 1 cm at both ends using a cautery knife. Most of the graft was stripped easily, leaving both sutured ends intact, thus maintaining the original diameter of both ends (Fig. 1b). The size of the AFX aortic cuff ranged from 22 to 28 mm, enabling the use of smaller sized stents rather than the TX-D stent (especially in the distal part of the stent). The stripped AFX was reloaded in the delivery sheath after peeling the graft.

Experimentation

An experiment was conducted to evaluate the radial force, closely associated with aortic complications in the long follow-up. The TX-D, with a diameter of 36 mm, and the stripped AFX, with a diameter of 28 mm, were compared using the Mylar^Ⓡ (Dupont, Wilmington, DE, USA) film-testing method [6].

The stent for measurement was placed in the apparatus and wrapped with a film to reduce its diameter (Fig. 2). An autograph with 0.5 mm/s (crosshead speed: 1.57 mm/s) was used to pull the film to achieve stent diameters of 15, 20, and 25 mm. Subsequently, the film was gradually released so that the stent could return to its original diameter. The force of each process was then measured. The “radial resistive force” required to contract the stent reflects the load acting on the stent from the artery wall. Similarly, the “chronic outward force” required to release the stent reflects the load acting on the artery wall from the stent. The forces were measured and calculated using the following formulas.

Figure 2.

This schema shows the mechanism of the Mylar film test.

P₁ = F₁ ÷ S₁,

S₁ = (D₀ × π × L_s) − (M × L),

where P₁: force (N/mm²), F₁: measured load (N), S₁: stent surface area (mm²), D₀: initial diameter (mm), L_s: stent length (mm), M: crosshead movement distance (mm), and L: film length (mm).

We performed the measurements in a water bath maintained at a constant temperature of 37°C ± 2°C and in triplicates for each stent.

Results

Patient cohort and perioperative records

Eight cases were retrospectively reviewed. Seven of them were male. The median (interquartile range, IQR) age of the patients was 60 years (46.3-70.3). The indications for TEVAR were renal malperfusion (n = 3), lower extremity malperfusion (n = 3), acute expansion of the aorta (n = 1), and uncontrolled pain (n = 1). The median (IQR) duration between TBAD onset and TEVAR was 12 days (8.8-47.5). Table 1 describes the breakdown of the patient characteristics and preoperative anatomical data. Fig. 3 shows an actual case. All patients did not have connective tissue disease.

Table 1.

Patient Demographics.

Case	Age/Sex	Duration from onset (days)	Indication	Comorbidities	Whole aortic diameter at preoperative CT (mm)
					LSA	Midthoracic	Supraceliac	Infrarenal
1	41 M	12	Renal malperfusion	None	30.4	35.4	27.2	26.1
2	59 M	2	Renal malperfusion	HT	33.5	39.6	27	21.6
3	47 M	28	Renal malperfusion	None	32.4	40	25	25
4	44 M	0	Leg malperfusion	DM	39.5	33.2	27.5	27
5	71 M	249	Enlargement of aorta	HT, DM, CAD	31.9	41.3	27.2	22.6
6	70 M	11	Leg malperfusion	HT, DL	27.7	31.7	27	22.8
7	73 F	12	Uncontrollable pain	HT, DL	36.2	36.4	28	25.5
8	61 M	16	Leg malperfusion	None	35.3	32.4	27.3	22.3

HT, hypertension; DM, diabetes mellitus; CAD, coronary artery disease; DL, dyslipidemia; LSA, left subclavian artery.

Figure 3.

The details of case 2.

(a) Preoperative CT image showing the shrinkage of the aortic true lumen, and the whole aortic diameter (including the aortic false lumen) was 25.5 mm.

(b) IVUS at the left renal vein (arrow) level showing the shrinkage of the aortic true lumen (arrowhead) even after the deployment of the TX-2.

(c) IVUS at the left renal vein (arrow) level showing the expanded aortic true lumen (arrowhead) by the stripped AFX.

(d) Postoperative CT image showing the obvious dilation of the aortic true lumen and shrinkage of the aortic false lumen at the thoracic level with the stripped AFX fully expanded.

(e) Volume rendering image of postoperative CT showing good dilatation of TX-2 and stripped AFX.

All procedures were performed under general anesthesia. The technical success rate, which was defined as successful device deployment without any complications, was 100%. The stent grafts used were TAG, CTAG (W. L. Gore & Associates, Flagstaff, AZ, USA, n = 8), and TX-2 (Cook Medical, n = 3). Three TX-2 (diameter: 32, 36, and 38 mm) and five CTAG (diameter: 31 mm × 1 and 34 mm × 4) were used for primary entry closure. Additional three CTAG (diameter: 28 mm each) stent grafts were deployed to the distal descending thoracic aorta to cover the remaining entry at the distal thoracic aorta. Six patients required extra-anatomical bypass prior to TEVAR (right subclavian artery-left subclavian artery, n = 5; right subclavian artery-left subclavian artery, left carotid artery, n = 1). No perioperative complications were noted. The original diameter of the stripped AFX was 28 mm (the diameter of all devices used in this study was 28 mm), and its length was 95 mm. All patients received two stripped AFX deployment.

Follow-up records

The median (IQR) follow-up period was 28.9 months (17.5-31.5). During the follow-up period, one patient died of septic shock unrelated to aortic disease 15 months after TEVAR. The complete thrombosis of the false lumen was achieved in eight patients in the descending thoracic aorta (100%), four patients in the thoracoabdominal aorta (50%), and three patients in the visceral aorta (38%). The aortic diameters at the left subclavian artery, midthoracic, supraceliac, infrarenal aorta level, and stripped AFX size at last follow-up CT are described in Table 2. No obvious aortic enlargement more than 5 mm compared to the preoperative CT was observed in all cases and at all levels.

Table 2.

Aortic Diameter at Last Follow-up CT.

Case	Whole aortic diameter at last follow-up CT (mm)				Stripped AFX size at last follow-up CT (mm)	Follow-up duration (month)
	LSA	Midthoracic	Supraceliac	Infrarenal
1	32.3	32.1	24.8	17.7	23.9	55.7
2	35.2	32.5	32	22.1	23.5	34.7
3	33	34	27	25.5	23.5	30.4
4	38.4	34.1	29.4	26.1	29.3	30.0
5	33.4	36	30.6	23.3	21.5	27.8
6	29.9	28.3	26.5	23.5	19.9	18.3
7	36.3	32.2	29.1	30	26	8.2
8	32.5	29.4	27.4	22.4	23.1	14.9

LSA, left subclavian artery.

Experimental data

The results for each stent radial force are shown in Fig. 4 and 5. The “radial resistive forces” of the TX-D stent and stripped AFX were identical. However, the “chronic outward force” of the TX-D stent was two to three times that of the stripped AFX. Therefore, the load acting on the aortic wall or flap from the TX-D stent was higher than that from the stripped AFX.

Figure 4.

Summary of the experiment that investigated the “radial resistive forces” of the stripped AFX and TX-D stents.

Figure 5.

Summary of the experiment that investigated the “chronic outward forces” of the stripped AFX and TX-D stents.

Discussion

Currently, guidelines strongly recommend the application of TEVAR for TBAD with complications [7]. The efficiency of distal bare metal stents in promoting aortic true lumen expansion after entry tear coverage with TEVAR has been reported [3, 4], and the Zenith Dissection Endovascular Stent (TX-D stent, Cook Medical, Bloomington, IN, USA) was invented to meet this demand. In some cases, the E-XL (Jotec-GmbH, Hechingen, Germany) was used for this purpose in some papers [8]. The efficacy of the TX-D stent was proven in a prospective single-arm study [4], and that of E-XL was proven in a retrospective study [8]. However, there is a potential risk of distal aortic complications, such as aneurysmal degeneration or new dissection caused by the TX-D stent, because of its larger diameter than that of a healthy suprarenal aorta or the E-XL due to its strong radial force. Hofferberth et al. reported that adding the TX-D stent decreased late distal aortic complications compared to proximal covered stent graft alone [9]. The follow-up period of that study was only approximately four years, and long-term follow-up data after using the TX-D stent are still unknown. A case report suggested that excessive oversizing of the TX-D stent might cause stent misalignment [5]. Therefore, using a device with a low radial force and a small diameter is crucial to avoid device-related complications in the long follow-up period. Our device was proven to have a lower radial force and diameter than the TX-D stent, as shown in the present study. Even though data comparing the radial force between the E-XL and stripped AFX stents are lacking, the E-XL stent was reported to have more radial force than the TX-D stent [8]. Therefore, our device is expected to have the lowest radial force and cause minimal distal aortic complications.

Using a device with less radial force may raise another concern: the stent may not promote true lumen expansion. However, the aorta flap of the acute or subacute dissection was very fragile and was easily pushed with little force. The present study showed that the application of the physician-modified device worked well, almost dilating up to its original size in every patient, indicating that the device had a sufficient radial force to expand the true lumen.

Moreover, in some cases, a distal bare stent is insufficient to restore blood flow to the visceral arteries. In such cases, small bare metal stenting of visceral arteries is required. However, a sizable bare stent deployed in the suprarenal aorta can hamper the addition of bare metal stents to visceral arteries. Therefore, a sparser stent pattern is preferable for distal bare stents. The density of the stent mesh in this physician-modified device was lower than that of the TX-D stent or E-XL. Therefore, our physician-modified device is preferable for distal bare stents in terms of the stent pattern.

The material of the TX-D stent is stainless steel, which hampers magnetic resonance image scanning. By contrast, AFX is made of cobalt chromium and enables MRI scanning.

The process used to create this physician-modified device is easy and requires less than 15 min for preparation. The alteration does not impede its use.

In the experiment, the stent radial force was measured by compressing the stent using a Mylar film. Stent radial force is commonly measured by pinching the stent, but the behavior of the stent is different from that of the stent deployed in the human aorta. The stent behavior of our method is quite similar to that in clinical use.

There were some limitations in this study. This was a retrospective medical chart review, which might include patient selection bias. In addition, the number of the patients enrolled in this study was small. Therefore, further study is warranted.

Conclusion

The stripped AFX aortic cuff is feasible and safe as a scaffolding stent during TEVAR for TBAD.

Conflict of Interest

None

Author Contribution

Shinichi Iwakoshi, Shoji Sakaguchi, Mai Murata, Tomoki Nagata, Akimitsu Tanaka, and Ryosuke Kametani: Contributions to the submitted work: acquisition of data for the work.

Shinichi Iwakoshi, Arisa Kameda, Shinsaku Maeda, Takeshi Sato, Hideyuki Nishiofuku, Shigeo Ichihashi, Toshihiro Tanaka, and Kimihiko Kichikawa: Contributions to the submitted work: conception or design of the work.

Disclaimer

Shigeo Ichihashi and Toshihiro Tanaka are the Editorial Board members of Interventional Radiology. They were not involved in the peer-review or decision-making process for this paper.