Acute iatrogenic type A aortic dissection following thoracic aortic endografting

Article Outline

• Abstract
• Case report
• Discussion
  • Incidence and onset
  • Etiologies
  • Thoracic aortic disease
  • Technical issues
  • Operative treatment and outcome
• Conclusion
• Author contributions
• References
• Copyright

Endovascular intervention has emerged as a less traumatic alternative treatment for several diseases of the thoracic aorta.¹, ² However, depending on the different aortic pathologies, procedure related complications have become increasingly evident: severe complications include type I endoleaks, migration, and endograft (EG) collapse, as well as those observed during conventional surgery (eg, stroke and paraplegia).³, ⁴

One of the emerging and most alarming complication of thoracic endografting is iatrogenic retrograde type A acute dissection (RTAAD).⁵ Retrograde type A acute dissection is defined as acute aortic dissection that originates distally to the ascending aorta with a retrograde flap progression into the ascending aorta.⁶ This complication has been previously described during conventional cardiac surgery with high mortality rate; previous case reports suggested that the fragility of the aortic wall and Marfan disease were predisposing factors to such a life-threatening complication.⁷, ⁸, ⁹

This report presents a case of iatrogenic RTAAD after EG repair of a descending thoracic penetrating aortic ulcer, requiring emergent surgical replacement of the ascending aorta and the aortic arch. The available English literature on RTAAD was also reviewed, in order to recognize potential predisposing factors and specific strategies to prevent it.

Back to Article Outline

Case report 

A 76-year-old male was referred to our department, presenting with a penetrating thoracic aortic ulcer; six-months before, this patient was admitted to our Emergency Department because of an acute aortic syndrome caused by a type B intramural hematoma that was managed medically (Fig 1, A). Medical history included hypertension, mild mitral valve regurgitation, and chronic inflammatory bowel disease. Clinically, the aortic ulcer was asymptomatic. The preliminary thoraco-abdominal computed tomographic (CT) scan with 3D-VR reconstruction showed a type 3 aortic arch (Fig 1, B), an enlarged ascending aorta [maximum diameter: 47 mm; body surface area (BSA): 1.75], and the presence of the penetrating ulcer located on the outer curve of the aorta, extending for 60 mm with a maximum penetration of 16 mm (Fig 1, C). The proximal neck was 17 mm in length from the ostium of the left common carotid artery; proximal and distal landing zones measured 36 mm, with the distal one located in the proximal third of the descending aorta. Abdominal scan revealed the presence of an enlarged left common iliac artery (maximum diameter: 30 mm), and a non-calcified suitable calibers of the femoral arteries. Informed consent was obtained for an endovascular exclusion. Logistic EuroSCORE predicted a mortality rate of 14.5%.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Computed tomography-angiography during urgent hospital admission for acute aortic syndrome showing type B intramural hematoma (A). Six-month 3D-VR reconstruction follow-up control detected a penetrating ulcer (sketched line) of the distal portion of a type 3 aortic arch (B), protruding 16 mm (C).

Under general anesthesia, the right common femoral artery was exposed in a standard fashion and a preliminary aortic angiography with an angulation of 60° confirmed the presence of a suitable proximal neck (Fig 2, A). A single endograft (TAG 36 mm × 15 cm; W.L.Gore & Associates, Flagstaff, Arizona, USA) was deployed during controlled hypotension (<90 mm Hg), with planned partial overstenting of the origin of the left subclavian artery. The angiographic assessment revealed a low flow reperfusion of the ulcer, because of the incomplete adherence of the endograft (EG) to the inner curve of the aorta. Hence, in order to avoid a “wind sock” effect and a potential collapse of the EG, we ballooned only once inside the proximal extremity of the EG to obtain the desired configuration. Final control confirmed the complete exclusion of the ulcer, and the patency of the left common carotid and subclavian arteries (Fig 2, B). Neither hemodynamic nor cardiorespiratory or neurologic adverse events were noted during the procedure.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Preliminary intraoperative angiography with 60° angulation (A) confirmed the origin of the ulcer (sketched line) just distally to the left subclavian artery. Final control after balloon angioplasty confirmed the EG correction and the absence of endoleak (B). Postoperative computed tomography-angiography: hemopericardium (arrow, C), true lumen compression (D), and entry tear at the proximal end of the EG (E).

The patient was transferred to the recovery room. Four hours later, while being extubated, profound and persistent hypotension and disparity of the pupils were noted. Suspecting a cerebrovascular complication, a computed tomography (CT) was done showing a type A aortic dissection: the intimal flap originated from the proximal end of the EG with a retrograde extension to the arch and the ascending aorta; additionally hemopericardium was also noted (Fig 2, C). Both the origin of the brachiocephalic trunk and the left common carotid artery were both dissected and the true lumen was compressed by false lumen (Fig 2, D). The EG was patent and well-conformed to the arch curvature, the ulcer was completely excluded without endoleak, and the left common carotid artery was patent. The left subclavian artery was slightly overstented (Fig 2, E). No active bleeding was noted.

Because of the potentially lethal extension of the dissection, immediate surgical repair was undertaken. Delay between the diagnosis and the intervention was 18 minutes. An arterial cannula was introduced into the right subclavian artery and a double stage venous cannula was placed into the right atrium, establishing the cardiopulmonary bypass. A hypothermic circulatory arrest was achieved, cooling the patient to 24°C. The ascending aorta and the proximal arch were opened, confirming the diagnosis of aortic dissection. The entry tear was located between the left common artery and the left subclavian artery ostium, in correspondence to the proximal end of the EG (Fig 3, A, B, and C). The intimal flap extended retrograde into the ascending aorta and stopped just proximally to the left main coronary artery. The ascending aorta and the proximal ventral part of the aortic arch were replaced with a 30 mm woven graft (Uni-graft W aortic arch; Braun, Melsungen, Germany) with epiaortic vessel reimplantation. The distal part of the graft was sutured to the proximal end of the EG with a polipropilene suture (Fig 3, D). After rewarming, weaning from cardiopulmonary bypass was obtained; the patient remained severely hypotensive despite massive doses of catecholamines and finally died.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Intraoperative finding: ascending and right pulmonary artery hematoma (A, arrow), opened arch with the intimal flap (B, arrow), proximal EG attachment site (C, sketched lines), and anastomosis between the vascular graft and the EG (D, arrow).

Back to Article Outline

Discussion 

Incidence and onset 

Retrograde type A acute aortic dissection (RTAAD) after thoracic EG implantation has been reported worldwide, ranging between 1% and 6.8%.³, ⁵, ¹⁰ Several descriptions of RTAAD after thoracic EG have been limited to isolated case reports or short series, or also from studies evaluating major complications of endovascular treatment of thoracic aortic diseases, during either elective or urgent cases (Table I). Moreover, RTAAD after thoracic EG implantation can occur not only during the procedure, but even days, weeks, or months following the same procedure, a fact that seems to be underreported.¹¹, ¹² The most extensive experience has been recently published by Dong et al¹² who reported 11 cases out of 443 EG implantation for type B aortic dissections through a seven-year experience with different devices, including both acute (≤ 14 days) cases and type A dissection discovered during the follow-up. In their study, the overall incidence rate was 2.5%, but is likely to be 0.7% (three cases) of true RTAAD. Similarly, Kpodonu et al¹¹ found that retrograde type A dissection accounted for 2.4% of their 287 thoracic EG intervention, but in their experience, the mean delay of detection was 202 days. Dealing with complications of thoracic EG, Neuhauser et al⁵ reported the highest incidence of retrograde type A dissection: in their initial experience of 73 patients, they reported an overall 6.8%, with three cases (4.1%) of acute dissection.

Table I. Reports of iatrogenic retrograde type A acute aortic dissection during thoracic endografting

n.s, Not specified.

Etiologies 

Possible causative mechanisms of RTAAD during EG intervention might be quite different: it could be caused from wires and sheaths manipulation in the aortic arch causing localized intimal tears, as suggested by Zhang et al,¹³ whereas Neuhauser et al⁷ mainly impeached reiterative balloon remodeling of the EG may cause intimal injuries extending in a retrograde manner, also because the too-rigid structure of the EG does not adapt perfectly to the aortic curve in an angulated aortic arch. In their cases, Dong et al¹² and Rubin et al⁹ described an aortic wall injury generated by pushing forward a partially-released EG and strongly concluded to avoid further attempts of device mobilization after the first contact of the EG with the aortic wall.

Thoracic aortic disease 

The limited experience published on RTAAD still leaves debate as to whether it is a complication of the EG procedure or an independent event, such as the natural progression of the disease.⁴, ⁵, ¹¹, ¹² Available data support the hypothesis of a combination of different causes rather than a single factor. Because most of the cases have been registered after EG treatment of type B dissections, several authors suggested the fragility of the aortic wall as the pathological background, and the endografting related factors as the provoking aspect.⁵, ⁹, ¹⁴, ¹⁵ Available data support the hypothesis of a combination of different causes rather than a single factor. Out of the 27 cases of RTAADs piled up in our review, nearly 70% involved a “fragile” aorta, meaning the presence of dissection or its variants, or rupture: Girdauskas et al⁴ suggested that intramural hematomas in the EG landing areas, especially in the distal aortic arch, could predispose to formation of retrograde dissection. In our case report, the penetrating ulcer would have been considered a “fragile” lesion since developed as a consequence of a previous intramural hematoma and not, as more frequently occurred, like an evolution of an atherosclerotic aorta. Noticeably, in the studies of Dong et al¹² and Won et al,¹⁶ RTAAD represented the most common complication among Marfan patients. We might expect that the delicate nature of the vessel wall, especially in the acute pathologies, predisposed to this sort of problems.

Anyway, RTAAD has been described also after EG repair of degenerative atherosclerotic disease: Verhoye et al¹⁷ and Pasic et al⁸ suggested that the stiffness of the EG device and the limited angulation capacity might have initiated the intimal tear. Undoubtedly, the impact force during EG deployment as well as the hemodynamic shear stress on the aortic wall has promoted the extension of the dissection.¹²

Technical issues 

It has been suggested that the configuration of the device itself may also contribute to the onset of RTAAD. However, the influence of the different EGs on intimal injury is questionable. Injury from the proximal bare spring of EG with proximal free-flow was taken into account first, because it has been designed to provide strong radial force to strengthen the proximal fixation.⁵, ⁸, ⁹, ¹⁰, ¹⁸, ¹⁹, ²⁰ By reviewing the published articles, Dong et al¹² found 37 patients who experienced retrograde dissection after EG repair for type B dissection, and noted that an EG with a proximal bare spring was used in 73% of cases. The authors included both acute and chronic retrograde dissection, whereas the present review dealt with acute retrograde dissection only. Out of 27 cases identified, the device configuration was definite in 17 cases (63%): the proximal bare spring was used in eight of them. Of note, the majority of the authors recommended further refinements directed towards lower profile, more flexible, and less traumatic EGs to minimize aortic arch injury, such as a fully covered proximal stent but solid enough to provide a tight circumferential seal.⁵, ¹², ¹⁴, ¹⁵, ¹⁶ Both types of EGs require balloon remodeling to match to the curved geometry of the aortic arch in order to obtain a tight seal. Since the description of Totaro et al²⁰ in 2001, several cases of RTAAD were reported to be triggered by repeated endoaortic balloon dilations performed to secure the fixation of the EG; Rubin et al⁹ experienced two deaths for ruptured dissections after endovascular repair associated with multiple balloon dilation. Similarly, Girdauskas et al⁴ suggested that extensive endoaortic maneuvers and balloon remodeling as the main causes of their retrograde intimal lesions. Our case involved a penetrating ulcer as the primal aortic disease, originating from a variant of the dissection such as intramural hematoma.², ⁴, ⁵ Hence, also in our experience, it is plausible that balloon remodeling caused the aortic intimal injury; in our case, balloon dilation was performed gently and in accordance with the indication for use (IFU), and was necessary to adapt the EG to the inner curve of the aortic arch in order to avoid the “wind sock” effect from the ascending aorta.

Routine oversizing >20% of the EG could contribute to the development of RTAAD. The cause may be related to the limited flexibility of the devices that produce forced wall stress at the outer curvature. Kpodonu et al¹¹ pointed out a 21.4% oversizing as the favoring factor in 28% of their retrograde dissections, but all three cases developed a mean of 23.3 months after the intervention; in the RTAAD cases, the mean oversizing was 9.6%.

An underreported factor that could contribute to the development of RTAAD was the placement of the EG at severely angulated aortic arch; this could be related to the inability to accommodate the too rigid EGs to the curved geometry of the distal aortic arch, especially when the angulation exceeds 60°.⁴, ¹², ¹⁴, ¹⁹ As a matter of fact, Girdauskas et al⁴ cited excessive angulation of the arch as much as 70° in one of their acute cases: the most likely explanation for RTAAD in their case was mechanical since the aortic neck angle was so extreme.

Operative treatment and outcome 

Dissections that extend in a retrograde manner into the aortic arch and ascending aorta may lead to cerebrovascular ischemia, aortic valve regurgitation, cardiac tamponade, and obstruction of the coronary artery; of the 27 RTAADs described in literature, sudden death occurred in two cases (7.4%).⁹, ¹² Hence, rapid diagnosis and subsequent open surgery should be the treatment of choice in an effort to avert these life-threatening complications, with ascending and arch graft replacement the preferred intervention. The main dilemma in these conversions was that the patient needed a total arch replacement with the EG in the proximal descending aorta. From a technical point of view, many authors sutured the vascular graft directly to the EG after the substitution of the entire aortic arch as described by Pasic et al⁸ and Savini et al²¹ We performed an end-to-end distal anastomosis by directly connecting the new arch vascular prosthesis to the EG because of the severe involvement of the supra-aortic vessels, and took deep bites also in the surrounding aortic wall for the suture reinforcement. We did not remove the EG because we were afraid to worsen the injured aortic wall. In contrast, Estrera et al⁶ removed the EG and placed a Dacron graft in its place using the modified elephant trunk technique in order to expedite future distal repair if required. Only in one case (3.7%), reported in the study of Verhoye et al,¹⁷ a conservative treatment was established for a focal and asymptomatic lesion. Overall, survival rate was 44.4%, but in those reports (29.4%) with at least two cases of RTAAD, the mortality rate was 8/14 (57.1%). Nonetheless, pondering the good outcome results of open conversion after thoracic EG failure, someone could argue that conventional therapy should not be completely abandoned even in this higher-risk population.³, ⁴, ⁶, ²² Langer et al³ converted eight patients out of a group of 106 EGs: all survived the operation and the majority had an uneventful course. In the experience of Grabenwoger et al,²² surgical reoperation became necessary in four patients (5%), and in-hospital mortality was null.

Back to Article Outline

Conclusion 

Retrograde type A dissection is a potentially lethal complication following EG repair of thoracic aortic pathologies; it may have an acute or delayed presentation. Final conclusions regarding precursors for developing a RTAAD could not be definitively identified; however, some strategies could be helpful to prevent it.

Close attention to aortic disease and device selection in term of specific anatomic landmarks remain major key points; a marked difference exists between EG repair for dissections and EG repair for atherosclerotic aneurysms. The poor quality of the aortic wall in patients with type B dissection or its variants seems to be strongly related to this severe complication.

Careful balloon dilation is mandatory to obtain a satisfactory EG adaptation. However, the self-expanding action of the EG, especially for those with a nitinol skeleton, results in self-adjustment over; therefore, excessive oversizing (>20%) of the EG could be avoided in those “fragile” aortas.

Further refinements should be directed towards more flexible and less traumatic devices.

Back to Article Outline

Author contributions 

Back to Article Outline

References 

1. Glade GJ, Vahl AC, Wisselink W, Linsen MA, Balm R. Mid-term survival and costs of treatment of patients with descending thoracic aortic aneurysms; endovascular vs. open repair: a case-control study. Eur J Vasc Endovasc Surg. 2005;29:28–34
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (129 KB)
   • CrossRef
2. Dick F, Hinder D, Immer FF, Hirzel C, Do DD, Carrel TP, et al. Outcome and quality of life after surgical and endovascular treatment of descending aortic lesions. Ann Thorac Surg. 2008;85:1605–1612
   • View In Article
   • CrossRef
3. Langer S, Mommertz G, Koeppel TA, Schurink GW, Autschbach R, Jacobs MJ. Surgical correction of failed thoracic endovascular aortic repair. J Vasc Surg. 2008;47:1195–1202
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (823 KB)
   • CrossRef
4. Girdauskas E, Falk V, Kuntze T, Borger MA, Schmidt A, Scheinert D, et al. Secondary surgical procedures after endovascular stent grafting of the thoracic aorta: successful approaches to a challenging clinical problem. J Thorac Cardiovasc Surg. 2008;136:1289–1294
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (335 KB)
   • CrossRef
5. Neuhauser B, Czermak BV, Fish J, Perkmann R, Jaschke W, Chemelli A, et al. Type A dissection following endovascular thoracic aortic stent-graft repair. J Endovasc Ther. 2005;12:74–81
   • View In Article
   • MEDLINE
   • CrossRef
6. Estrera AL, Shah P, Lee TY, Irani AD, Safi HJ. Repair of retrograde type A aortic dissection after thoracic endovascular aortic aneurysm repair using the modified elephant trunk technique. Vascular. 2009;17:116–120
   • View In Article
   • CrossRef
7. Yamashiro S, Kuniyoshi Y, Arakaki K, Inafuku H. Intraoperative retrograde type I aortic dissection in a patient with chronic type IIIB dissecting aneurysm. Interact Cardiovasc Thorac Surg. 2009;8:283–286
   • View In Article
   • CrossRef
8. Pasic M, Bergs P, Knollmann F, Zipfel B, Müller P, Hofmann M, et al. Delayed retrograde aortic dissection after endovascular stenting of the descending thoracic aorta. J Vasc Surg. 2002;36:184–186
   • View In Article
   • Abstract
   • Full-Text PDF (375 KB)
   • CrossRef
9. Rubin S, Bayle A, Poncet A, Baehrel B. Retrograde aortic dissection after a stent graft repair of a type B dissection: how to improve the endovascular technique. Interact Cardiovasc Thorac Surg. 2006;5:746–748
   • View In Article
   • CrossRef
10. Hansen CJ, Bui H, Donayre CE, Aziz I, Kim B, Kopchok G, et al. Complications of endovascular repair of high-risk and emergent descending thoracic aortic aneurysms and dissections. J Vasc Surg. 2004;40:228–234
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (169 KB)
    • CrossRef
11. Kpodonu J, Preventza O, Ramaiah VG, Shennib H, Wheatley GH, Rodriquez-Lopez JA, et al. Retrograde type A dissection after endovascular stenting of the descending thoracic aorta (Is the risk real?). Eur J Cardiothorac Surg. 2008;33:1014–1018
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (96 KB)
    • CrossRef
12. Dong ZH, Fu WG, Wang YQ, Guo da Q, Xu X, Ji Y, et al. Retrograde type A aortic dissection after endovascular stent graft placement for treatment of type B dissection. Circulation. 2009;119:735–741
    • View In Article
    • CrossRef
13. Zhang R, Kofidis T, Baus S, Klima U. Iatrogenic type A dissection after attempted stenting of a descending aortic aneurysm. Ann Thorac Surg. 2006;82:1523–1525
    • View In Article
    • CrossRef
14. Bethuyne N, Bove T, Van den Brande P, Goldstein JP. Acute retrograde aortic dissection during endovascular repair of a thoracic aortic aneurysm. Ann Thorac Surg. 2003;75:1967–1969
    • View In Article
    • MEDLINE
    • CrossRef
15. Shetty V, Vohra HA, Viola N, Brown I, Langley SM. Surgical intervention for retrograde type A aortic dissection caused by endovascular stent insertion for type B aortic dissection. J Thorac Cardiovasc Surg. 2008;135:1387–1388
    • View In Article
    • Full Text
    • Full-Text PDF (228 KB)
    • CrossRef
16. Won JY, Suh SH, Ko HK, Lee KH, Shim WH, Chang BC, et al. Problems encountered during and after stent-graft treatment of aortic dissection. J Vasc Interv Radiol. 2006;17(2 Pt 1):271–281
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (741 KB)
    • CrossRef
17. Verhoye JP, de Latour B, Heautot JF, Vola M, Langanay T, Corbineau H, et al. Mid-term results of endovascular treatment for descending thoracic aorta diseases in high-surgical risk patients. Ann Vasc Surg. 2006;20:714–722
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (760 KB)
    • CrossRef
18. Kaya A, Heijmen RH, Rousseau H, Nienaber CA, Ehrlich M, Amabile P, et al. Emergency treatment of the thoracic aorta: results in 113 consecutive acute patients (the Talent Thoracic Retrospective Registry). Eur J Cardiothorac Surg. 2009;35:276–281
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (279 KB)
    • CrossRef
19. Fanelli F, Salvatori FM, Marcelli G, Bezzi M, Totaro M, Vagnarelli S, et al. Type A aortic dissection developing during endovascular repair of an acute type B dissection. J Endovasc Ther. 2003;10:254–259
    • View In Article
    • MEDLINE
    • CrossRef
20. Totaro M, Miraldi F, Fanelli F, Mazzesi G. Emergency surgery for retrograde extension of type B dissection after endovascular stent graft repair. Eur J Cardiothorac Surg. 2001;20:1057–1058
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (47 KB)
    • CrossRef
21. Savini C, Casselman F, Ergenoglu MU, Degrieck I, Van Praet F, De Geest R, et al. Surgical management of progression to type A dissection from an intramural hematoma previously treated with endovascular stent graft placement. J Thorac Cardiovasc Surg. 2004;128:773–775
    • View In Article
    • Full Text
    • Full-Text PDF (197 KB)
    • CrossRef
22. Grabenwoger M, Fleck T, Ehrlich M, Czerny M, Hutschala D, Schoder M, et al. Secondary surgical interventions after endovascular stent-grafting of the thoracic aorta. Eur J Cardiothorac Surg. 2004;26:608–613
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (417 KB)
    • CrossRef