An unusual case of aortic rupture after deployment of a bare stent in the treatment of aortic dissection in a patient with giant-cell arteritis.

Giant-cell arteritis is associated with a higher risk of aortic aneurysm and aortic dissection formation. We present a women with aortic dissection type B treated with a stent graft and bare-metal stent implantation. After the stent deployment we noticed aortic rupture, which was successfully treated with implantation of an additional stent graft. This report highlights the difficulty of endovascular therapy in patients with giant-cell arteritis. We have to bear in mind that chronic inflammation of the aorta leads to a more fragile aortic wall than normal. We recommend the use of a stent graft over a bare-metal stent and gentle use of a balloon catheter.

Introduction

Giant-cell arteritis (GCA) is associated with a higher incidence of aortic dissection (AD) [1-3]. However, most cases of AD have an etiology other than GCA. One of the recommended methods of treatment of AD is the coverage of the entry tear with a stent graft [4]. Some vascular surgeons have expanded this method by the use of an additional bare stent deployed distally [5-7]. This report highlights the difficulty of such therapy in a patient with GCA.

Case report

In September 2016, a 70-year-old woman with GCA was admitted for treatment of a type B aortic dissection in the thoracic and abdominal aorta. Her past medical history was significant for polymyalgia rheumatica, a fatty liver, hypertension, and cirrhosis of the right kidney on a vascular background. She was diagnosed with giant-cell arteritis 12 years ago. At the beginning the disease had a self-limited course and she was only on non-steroid anti-inflammatory drugs. For the last 7 years she had had aortic complications, thickening of the aortic wall, and stenosis of the aortic arch branches, renal arteries and common iliac arteries. The wall in the ascending aorta was approximately 0.98 cm and 0.44 cm in the abdominal aorta. Regular courses of steroids and cyclophosphamide were instituted to reduce the inflammatory process of the aorta, but failed to improve the stenosis in the aortic branches. Furthermore, in subsequent years cirrhosis of the right kidney was found. Some 5 months ago control computed tomography angiography (CTA) revealed thoracoabdominal dissection type B. Although no symptoms in the chest and abdominal region were found, the patient nonetheless developed progressive intermittent claudication, categorized as stage 3 on the Rutherford scale. The false lumen in the right lateroposterior region was found to have an entry tear in the lower thoracic segment, a re-entry tear in the vicinity of the celiac trunk and another re-entry approximately 3 cm above the aortic bifurcation. The false lumen was supplied by a posterior intercostal artery in the thoracic area. No part of the major vessel was supplied by a false lumen. The true lumen was stenosed to 6 mm about 3 cm above the aortic bifurcation (Photo 1).

Photo 1

3D reconstruction showing patent false lumen

The procedure was planned to cover the entry tear with a stent graft in the thoracic region (Valiant Captiva, Medtronic, USA) and a bare-metal stent (Sinus-XL Stent, Optimed, Germany) in the abdominal region to expand the true lumen and compress the false lumen. Afterwards a stent was deployed and expanded with a balloon catheter (Reliant, Medtronic, USA) to restore the whole lumen. However, complications arose when an anesthesiologist reported a sudden drop in the patient’s blood pressure. Control angiography was then performed and an aortic rupture was revealed in the middle between the renal ostia and the abdominal aortic bifurcation (Photo 2). It was decided to implant another stent graft inside the previous mentioned stent. An iliac extension was used for this purpose (Endurant II, Medtronic, USA). Although a residual bruise was detected following this process (Photo 3), the blood pressure subsequently normalized, so a decision not to perform additional procedures was reached. After closure of the wound, 10 mg of protamine sulfate was given to efficiently reverse (based on ACT) the effect of 3000 units of heparin which was administered at the beginning of the procedure.

Photo 2

Angiography after bare stent deployment showing aortic rupture

Photo 3

Angiography showing residual bruise after additional stent graft implantation

The patient was admitted to the intensive care unit for a duration of 3 days. After 7 days of hospitalization she was discharged without any other problems.

Discussion

Giant-cell arteritis is chronic inflammation of medium and large-sized vessels containing elastic fibers. The most often affected vessel is the temporal artery. It is for this reason that GCA is sometimes called temporal arteritis. Inflammation of the intracranial arteries seldom occurs [8]. The incidence is the highest in Scandinavian countries and in people of Scandinavian descent and is approximately 20–30 per 100 000 persons aged 50 years and older [9, 10]. Women are affected 2–3 times more often than men [11, 12]. Typically patients are over 50 years of age and morbidity increases with age, reaching a peak in octogenarians [11, 12].

In contrast, non-GCA AD is more often found in men, with a men-to-women ratio of 4 : 1 [13]. Hypertension and age are the main risk factors [14]. The incidence is about 2.9–3.5 per 100 000 person-years, with the peak between 50 and 70 years of age [14].

A patient with GCA is 17.3 times more likely to develop a thoracic aortic aneurysm and 2.3 times more likely to develop an abdominal aneurysm [1]. Other studies suggest a 9.5–18% chance of developing aortic aneurysm or dissection [2, 3]. Twenty-five – forty percent of aortic dissections will develop into an aneurysm [15, 16]. Factors associated with aneurysmal progression include poorly controlled hypertension, patent false lumen and an aorta diameter of 4 cm or more [17].

The pathogenesis of GCA is complex and not fully understood. However, the inflammation starts from activated dendritic cells in the adventitia, which attract CD4 T lymphocytes. The release of cytokines including interferon γ brings monocytes to the place, which transform into multinucleated giant cells – the hallmark of GCA. Macrophages produce matrix metalloproteinases and reactive oxygen species which damage the media [18-21]. Thus the aortic wall is weakened and impaired due to hypertension and may lead to aortic dissection. In such a case the fragility of the wall could contribute to aortic rupture after stent expansion. Additionally, although the chronic administration of steroids may be responsible for the weakness of the aortic wall [22-24] and could play a role, steroids are nevertheless needed in complicated GCA because they inhibit the inflammation. The most common localization of the entry tear in type B AD is in the vicinity of the left subclavian artery [25, 26]. It is caused by considerable dP/dt pressure fluctuations in that region. Therefore, the mechanical force cleaves to the layers of the aortic wall. When, as in the reported case, the entry tear was situated at a greater distance from the left subclavian artery in the same region, lower dP/dt pressure fluctuations were noticed. This suggests that the fragile wall contributed more to AD than mechanical stress.

The combination of stent grafting with bare-metal stenting in AD was first proposed by Mossop et al. [5, 6]. The idea is to cover the entry tear with a stent graft and then, distally, expand the true lumen with a bare stent promoting false lumen thrombosis. The use of a bare stent also has the advantage of protecting the vessels’ ostia from occlusion, thus preserving the inflow. The concept was named PETTICOAT, an abbreviation of Provisional ExTension To Induce COmplete ATtachment.

Mossop et al. in their subsequent studies proved the safety of bare stents in AD. They did not report aortic rupture as a complication of bare-metal stenting in AD [5, 6]. Other authors encountered one aortic rupture [7]. However, the rupture was located distally to the distal end of the bare stent, and it concerned only the false lumen, with the intimal flap between both lumens still intact.

Usually aortic ruptures, if and when they happen, following a subsequent endovascular aortic repair (EVAR), are due to endoleaks which cause constant pressurization of the aortic wall. This results in aneurysmal growth and eventually rupture [27-31]. This process is extended in time. By contrast, in the described case the rupture occurred immediately.

The amount of stent oversizing may also play a role. Most instructions for use (IFU) advise about 10%. We used an 18 mm stent so oversizing was approximately 12% in the upper part of the aorta, but the aorta decreased in size in the lower part region where the focal stenosis (6 mm) was located, resulting in considerable oversizing. This oversizing in the lower part of the aorta does not appear to be responsible for the rupture, because the rupture was located 2 cm above the region that had the greatest stenosis. Anyway, we could not calculate the stent size according to stenosed area, because the stent would be free floating in the upper part of the aorta. It may also be the case that the excessive expansion caused by a balloon catheter also contributed to the aortic rupture. Nevertheless, a low pressure balloon catheter is considered part of a more reliable procedure since it is safer. Despite the fact that some authors do not advise post-dilatation in AD [32], we nonetheless perform it routinely in non-GCA AD, and we have never observed aortic rupture before. Postdilatation compresses the false lumen, which limits the inflow from re-entries, thus preventing its further expansion and compression of the true lumen. Fanelli et al. observed increased aortic remodeling after postdilatation in AD [33]. A similar effect was obtained by Kölbel et al., who constructed a stent graft designed to treat AD which has a unique shape, a hump in the midsection that obstructs the false lumen and prevents a continuous false lumen perfusion [34].

To cover the rupture we used an additional stent graft. In our opinion a covered stent should be the first line strategy in patients with GCA and aortic dissection. Even if endovascular treatment causes a rupture, it will automatically be covered by a stent graft and it will prevent extravasation of the blood.

Conclusions

In patients with GCA and aortic dissection the aortic wall is more fragile than in patients with non-GCA aortic dissection. Therefore, careful expansion of endovascular devices is recommended. Furthermore, we recommend the use of stent grafts over bare stents.

Conflict of interest

The authors declare no conflict of interest.