Share this article on:

Minimally Invasive Segmental Artery Coil Embolization for Preconditioning of the Spinal Cord Collateral Network Before One-Stage Descending and Thoracoabdominal Aneurysm Repair

Luehr, Maximilian MD; Salameh, Aida MD; Haunschild, Josephina MS; Hoyer, Alexandro MD; Girrbach, Felix F. MD; von Aspern, Konstantin MD; Dhein, Stefan MD, PhD; Mohr, Friedrich-Wilhelm MD, PhD; Etz, Christian D. MD, PhD

Innovations:Technology and Techniques in Cardiothoracic and Vascular Surgery: January/February 2014 - Volume 9 - Issue 1 - p 60–65
doi: 10.1097/IMI.0000000000000038
Original Articles

Objective Paraplegia remains the most devastating complication after thoracic and thoracoabdominal aortic aneurysm (TAA/A) repair. The collateral network (CN) concept of spinal cord perfusion suggests segmental artery (SA) occlusion to mobilize redundant intraspinal and paraspinal arterial sources and ultimately trigger arteriogenesis, leading to spinal cord blood flow restoration within 96 to 120 hours. This principle is used by the two-staged approach to TAA/A-repair—which has lead to an elimination of paraplegia in an experimental model. However, the clinical implementation of a two-staged surgical procedure is challenging, particularly in the absence of an appropriate vascular segment for a “staged” open anastomosis or an appropriate endovascular landing zone. Selective, transfemoral minimally invasive SA coil embolization (MISACE) could provide the solution for one-stage repair of extensive aortic pathologies by triggering arteriogenic CN preconditioning and thereby allowing for recruitment of otherwise redundant arterial collaterals to the spinal cord.

Methods The feasibility of MISACE was explored in a single animal using an established piglet model. A 6F sheet was introduced via the femoral artery, and a 4F standard Judkins catheter was used for selective angiography and coil insertion. All thoracic and lumbar aortic SAs (15 pairs; Th4-L5) were successfully identified by dye injection. Pediatric platinum endovascular coils (Trufill Pushable Coils, 3 × 20 mm; Cordis, Waterloo, Belgium) were deployed to serially occlude the SA mimicking a CN preconditioning procedure.

Results All intercostal (thoracic) and lumbar aortic SAs (Th4-L5) were successfully identified and occluded by coil embolization. Successful SA coil embolization was verified intraoperatively by selective dye injection on angiography. No intraoperative coil dislodgement occurred. Autopsy revealed complete occlusion of all embolized SAs enhanced by early local thrombus formation. Thrombotic material was found only distally to the coils. No SA dissection was observed at the aortic SA origins.

Conclusions The MISACE technique allows for rapid serial endovascular occlusion of all thoracic and lumbar SAs. This new innovative approach bares the potential to CN preconditioning at the respective level of aortic pathology—to allow for adequate perioperative spinal cord blood supply—before conventional open or endovascular surgery. Selective, transarterial MISACE might lead to a dramatic reduction of ischemic spinal cord injury after open and endovascular TAA/A repair in the future.

Supplemental digital content is available in the text.

From the Department of Cardiac Surgery, Leipzig Heart Center–University of Leipzig, Leipzig, Germany.

Accepted for publication November 12, 2013.

A video clip is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site ( Please use Firefox when accessing this file.

Presented at the Annual Scientific Meeting of the International Society for Minimally Invasive Cardiothoracic Surgery, June 12–15, 2013, Prague, Czech Republic.

Disclosure: The authors declare no conflicts of interest.

Address correspondence and reprint requests to Christian D. Etz, MD, PhD, Department of Cardiac Surgery, Leipzig Heart Center–University of Leipzig, Struempellstr. 39, 04289 Leipzig, Germany. E-mail:

Paraplegia caused by ischemic spinal cord injury (SCI) after extensive descending and thoracoabdominal aortic aneurysm (TAA/A) repair remains—despite the consequent use of various neuroprotective strategies, such as deep-to-moderate hypothermia, cerebrospinal fluid drainage, perioperative monitoring of somatosensory- and motor-evoked potentials, and maintenance of a supranormal mean arterial pressure—a devastating complication with a real-world incidence of up to 32%.1–4

Reimplantation of aortic segmental arteries (SAs) to secure spinal cord perfusion—historically thought to be mainly dependent on one single artery, known as the Artery of Adamkiewicz, arising between the segments of Th7 and L1—is controversial because it does not reliably lower the incidence of ischemic SCI.3,5 On the contrary, complete SA occlusion has been demonstrated to result in an equal or even lower paraplegia rate—experimentally and in large retrospective clinical studies.5–7 On the basis of these observations, the modern collateral network (CN) concept of spinal cord perfusion has begun to replace the classic understanding of one single artery responsible for spinal cord perfusion8,9 in aortic surgery.

Consequentially, the recognition of an extremely redundant and efficient paraspinal and intraspinal CN that may compensate for acute—and chronic—loss of direct spinal blood support caused by SA occlusion during aortic surgery10,11 has lead to a recent paradigm shift: the two-staged repair has been suggested to offer superior neuroprotection for TAA/A repair—especially with regard to the maintenance of spinal cord viability.6 The most intractable issue with this new strategy, however, was the fact that undergoing aortic surgery twice for one aneurysm seemed not to justify the increased risk for morbidity and mortality for reoperative surgery. Therefore, we thought of an endovascular priming procedure to precondition the arterial CN and optimize spinal cord perfusion before complete SA occlusion—inevitable in thoracic endovascular aortic repair (TEVAR) and most open single-staged repairs.

Selective, transfemoral minimally invasive SA coil embolization (MISACE) may provide the solution for triggering arteriogenic CN preconditioning, thereby allowing for recruitment of otherwise redundant arterial collaterals to the spinal cord. Minimally invasive SA coil embolization—for successful CN preconditioning before conventional open or TEVAR—may enable the safe repair of extensive TAA/A as a one-stage operation, possibly solving what was the most devastating issue after aortic surgery for half a century: permanent paraplegia caused by acute ischemic SCI.

Back to Top | Article Outline


An acute experimental study was executed in one pig to explore the feasibility of MISACE for potential CN preconditioning using an established piglet model.

The animal used in our laboratory received humane care in compliance with the guidelines of “Principles of Laboratory Animal Care,” formulated by the National Society of Medical Research, and the Guide for the Care and Use of Laboratory Animals, published by the National Institutes of Health (publication no. 88-23, revised 1996). This acute experiment was approved by the local ethics committee of the University of Leipzig, Germany.

Back to Top | Article Outline

Pig Anatomy

The pig’s vasculature differs from that in humans in having 13 instead of 12 thoracic and 5 lumbar SAs, which ensure thorough spinal cord perfusion. The porcine aortic arch differs from that in humans in having only two supra-aortic branches. The first and the largest branch of the aortic arch is the brachiocephalic artery, which divides into the common bicarotid trunk and the right subclavian artery. The second branch is the left subclavian artery. Both subclavian arteries feed an extended vascular plexus located in the pig’s neck and the spinal cord until the level of the fourth thoracic vertebra (Th4) before these become the axillary arteries. In addition, the SAs arise as one common artery from the dorsal aorta before these subsequently divide, which is different from that in humans, in which the SAs arise separately at each segmental level.

Back to Top | Article Outline

Minimally Invasive SA Coil Embolization

After premedication with intramuscular midazolam (0.5 mg/kg), ketamine (15 mg/kg), and atropine (0.02 mg/kg), a juvenile pig (age, 3.5 months; weight, 35 kg) was intubated with an endotracheal tube, and ventilation was started (fraction of inspired oxygen, 0.5; PCO2, 35–40 mm Hg). The animal was placed on its right side. Sedation and analgesia were maintained by continuous intravenous administration of propofol (2%) and fentanyl at a rate of 25 to 35 mg/kg per hour and 1 to 4 μg/kg per hour, respectively. Blood pressure was continuously monitored during the entire experiment via a pressure catheter in the right brachial artery.

A 6F sheet was placed in the right femoral artery using the Seldinger technique. After heparinization (5000 IU), a standard 4F Judkins catheter was introduced to identify the respective aortic SA by selective angiography (Fig. 1A; Movie 1, available online at Subsequently, serial occlusion of 15 SA pairs (Th4-L5) was performed by coil embolization using pediatric platinum endovascular coils (Trufill Pushable Coils, 3 × 20 mm; Cordis, Waterloo, Belgium) to mimic a preoperative CN preconditioning procedure (Figs. 1B, C). Complete SA embolization was verified by dye injection (3–5 ml) after each coil deployment (Fig. 1D). Because aortic SAs in the pig usually arise as a single trunk that subsequently divides, one to two coils were required to completely embolize the SA at each segment.



After complete serial SA occlusion by MISACE, euthanasia was achieved by high-dose intravenous potassium during deep anesthesia.

Back to Top | Article Outline


To allow for complete exposure of the thoracic and thoracoabdominal aorta, a left-sided thoracotomy and extended laparotomy were performed after the animal was killed. After dissection of all aortic SAs, macroscopic verification of SA coil embolization was performed.

Back to Top | Article Outline


All intercostal (thoracic) and lumbar aortic SAs (15 pairs; Th4-L5) were successfully identified and occluded by coil embolization during the experiment. Successful SA coil embolization was verified intraoperatively by selective dye injection on angiography (Figs. 1A–D; Movie 1, available online at The coil-to-occlusion and catheter-positioning times were 30 to 45 seconds and 15 to 30 seconds per SA, respectively, with a total procedure time of 44 minutes for a complete (Th4–13, L1–5) serial MISACE procedure (excluding the required time for preparation of the experimental setup and femoral cannulation). Minimally invasive SA coil embolization with a single coil was feasible in seven SAs, whereas sufficient embolization required two coils in the remaining eight SAs. The total required amount of contrast dye was 280 mL, with a total fluoroscopy time of 22.41 minutes.

No intraoperative coil dislodgement occurred, either reversely into the aorta or antegradely.

Autopsy revealed complete occlusion of all embolized SAs, enhanced by early local thrombus formation. Thrombotic material was found only distally to the coils. No SA dissection was observed at the aortic SA origins (Fig. 2).



Back to Top | Article Outline


Paraplegia/paraparesis occurs equally often after open or endovascular aortic repair and often depends on the extent of repair, with associated arterial compromise in spinal cord blood supply.12,13

Thorough knowledge of the spinal cord’s vasculature and perfusion physiology is key to successfully avoid perioperative SCI: between 1957 and 1971, Lazorthes et al14–16 published a series of landmark studies on arterial supply of the spinal cord that challenged the classic theory of the Adamkiewicz artery. Despite these potential new insights in spinal cord blood supply, most surgical strategies for TAA/A repair remained unimproved with regard to spinal cord protection—often proclaiming the necessity of SA reimplantation to prevent perioperative paraplegia.17,18 However, a contemporary surgical strategy for TAA/A repair gaining more and more acceptance suggests aortic SA occlusion before opening of the aneurysm sack to prevent back bleeding with resulting steal syndrome and potential SCI—achieving an equal and even lower incidence of paraplegia in many reported series.5,19,20 These clinical observations have been supported by several experimental studies, eventually resulting in the development of the arterial CN concept of the spinal cord.10,11

Thoracic endovascular aortic repair is thought to be less invasive than conventional open repair and therefore has been increasingly used by many surgeons/interventionalists since its clinical implementation in 1996.21 However, endovascular stent grafting for TAA/A often leads to extensive SA coverage because of a required proximal and distal landing zone in the nondiseased aorta. Despite a reported low early postoperative mortality, TEVAR tends to result in a higher incidence of long-term complications such as endoleaks, device failure, or migration.22 Because TEVAR does not prohibit back bleeding from covered aortic SA, steal from the CN may occur (endoleak type II)—resulting in aneurysm progression and related death.22 Moreover, extensive aortic stent grafting has been associated with a significantly increased risk for paraplegia.23

The paravertebral and intraspinal CN is fed by various arterial sources that may be subdivided into three main source regions: (1) the subclavian and vertebral arteries, (2) the intercostal (thoracic) and lumbar SAs, and (3) the hypogastric (internal iliac) arteries9,11 (Fig. 3).



Experimentally, the CN has been demonstrated to comprise a vast paravertebral and intraspinal network of redundant arteries and arterioles communicating with the three main sources of spinal cord blood supply.10 Acute or chronic spinal cord ischemia can thus be alleviated across several thoracic or lumbar segments by redirecting blood flow to inadequately supplied cord areas.10 It has been demonstrated experimentally that SA occlusion may not only trigger arteriogenesis after chronic occlusion but also acutely increase small arterial vessel diameters, leading to complete restoration of spinal cord blood flow from alternative arterial sources within 96 to 120 hours.11,24

On the basis of these findings, the two-staged approach for TAA/A repair was developed. In a retrospectively reviewed series of 90 consecutive patients of which two thirds had undergone a regular open repair in one single surgery whereas one third had (unintentionally) received two separate surgeries for thoracic/TAA/A disease after a mean interval of 5 years (range, 3 months to 17 years)—mimicking a two-staged operation for TAA/A repair with an equal, nonsignificantly different number of occluded SA during repair—the paraplegia rate was 14.5% in the conventional repair group versus 0% with a staged repair.6 Subsequently, the superiority of the two-staged procedure was also successfully verified in a chronic piglet model experimentally.25

However, a two-staged operation is technically challenging, particularly in the absence of an appropriate vascular segment for a “staged” anastomosis or an endovascular landing zone. Moreover, any suitable two-staged TAA/A patient will have to be followed after the first operation/intervention and is expected to endure a second invasive procedure, potentially adding all the associated operative and perioperative risks of redo aortic surgery.

The CN of spinal cord blood supply may be amenable to arteriogenetic preconditioning by selective SA coil embolization (MISACE) before conventional open or endovascular repair by inducing ischemia in the affected aortic segments.7,10,24 Thus, an “activated” CN would not only be able to perioperatively maintain sufficient spinal cord perfusion and keep intraspinal perfusion pressures higher than levels critical causing SCI25,26 but also prevent intraoperative back bleeding from the aneurysm sack. Minimally invasive SA coil embolization—if applied before total endovascular repair—could therefore prevent related complications, such as aneurysm formation/further growth and bleeding by prevention of type II endoleaks after TEVAR.

Although it is technically feasible to embolize all thoracic and lumbar aortic SAs in one session, complete serial SA occlusion seems unsuitable to trigger CN remodeling because of the potential risk for ischemic SCI similar to extensive TEVAR. We therefore believe that—with regard to our experience with the two-staged approach for TAA/A repair—it might already be sufficient to embolize only a small number of lumbar or thoracic aortic SAs to efficiently induce CN preconditioning in the respective area of aortic pathology.

The authors are convinced that the implementation of this innovative and minimally invasive procedure into clinical practice might allow for complete TAA/A repair by TEVAR or conventional open surgery—after MISACE—with a significantly reduced risk for perioperative paraplegia in the future. However, we aim for a larger experimental series in a chronic model to investigate the neuroprotective potential of this new strategy for the spinal cord.

Back to Top | Article Outline

Limitations of the Study

Only one pig was used to show the feasibility of this new minimally invasive technique in an acute model. Therefore, no conclusions about a significant effect on CN preconditioning by MISACE can be drawn at this moment. No evaluation of motor function was performed postoperatively because of the fact that paraplegia after experimental SA occlusion has been shown to often occur delayed (up to 48 hours) postoperatively.

The vascularization of the pig differs as described in the text. Therefore, more than one coil would be necessary in humans to completely embolize the SA at each segment. This certainly would impact the total flouroscopy time, the amount of intra-arterial contrast agent, and the number of required coils for a successful MISACE preconditioning procedure.

Back to Top | Article Outline


Minimally invasive SA coil embolization allows for complete or selective serial endovascular occlusion of thoracic and lumbar aortic SAs, a strategy proven to generate “remodeling” of the arterial CN of spinal cord blood supply. Therefore, pretreatment by MISACE could enable awake preconditioning of the paraspinal and intraspinal arterial CN in any catheterization laboratory or hybrid room without the daunting side effects of a “first-stage” open procedure and subsequent reoperation—offering all beneficial effects of a staged repair. SA coil embolization represents an innovative and minimally invasive approach to CN preconditioning and might be the solution to dramatically reduce the incidence of ischemic SCI after open and endovascular TAA/A repair.

Back to Top | Article Outline


1. Jacobs MJ, Meylaerts SA, de Haan P, de Mol BA, Kalkman CJ. Assessment of spinal cord ischemia by means of evoked potential monitoring during thoracoabdominal aortic surgery. Semin Vasc Surg. 2000; 13: 299–307.
2. Svensson LG, Hess KR, Coselli JS, Safi HJ. Influence of segmental arteries, extent, and atriofemoral bypass on postoperative paraplegia after thoracoabdominal aortic operations. J Vasc Surg. 1994; 20: 255–262.
3. Griepp RB, Ergin MA, Galla JD, et al. Looking for the artery of Adamkiewicz: a quest to minimize paraplegia after operations for aneurysms of the descending thoracic and thoracoabdominal aorta. J Thorac Cardiovasc Surg. 1996; 112: 1202–1213.
4. Etz CD, Luehr M, Kari FA, et al. Paraplegia after extensive thoracic and thoracoabdominal aortic aneurysm repair: does critical spinal cord ischemia occur postoperatively? J Thorac Cardiovasc Surg. 2008; 135: 324–330.
5. Etz CD, Halstead JC, Spielvogel D, et al. Thoracic and thoracoabdominal aneurysm repair: is reimplantation of spinal cord arteries a waste of time? Ann Thorac Surg. 2006; 82: 1670–1677.
6. Etz CD, Zoli S, Mueller CS, et al. Staged repair significantly reduces paraplegia rate after extensive thoracoabdominal aortic aneurysm repair. J Thorac Cardiovasc Surg. 2010; 139: 1464–1472.
7. Etz CD, Homann TM, Plestis KA, et al. Spinal cord perfusion after extensive segmental artery sacrifice: can paraplegia be prevented? Eur J Cardiothorac Surg. 2007; 31: 643–648.
8. Adamkiewicz A. Die Blutgefaesse des menschlichen Rueckenmarks. S B Heidelberg Akad Wiss. 1882; 85: 101–130. Theil I + II.
9. Griepp RB, Griepp EB. Spinal cord perfusion and protection during descending thoracic and thoracoabdominal aortic surgery: the collateral network concept. Ann Thorac Surg. 2007; 83: S865–S869.
10. Etz CD, Kari FA, Mueller CS, Brenner RM, Lin HM, Griepp RB. The collateral network concept: remodeling of the arterial collateral network after experimental segmental artery sacrifice. J Thorac Cardiovasc Surg. 2011; 141: 1029–1036.
11. Etz CD, Kari FA, Mueller CS, et al. The collateral network concept: a reassessment of the anatomy of spinal cord perfusion. J Thorac Cardiovasc Surg. 2011; 141: 1020–1028.
12. Bavaria JE, Appoo JJ, Makaroun MS, Verter J, Yu ZF, Mitchell RS for the Gore TAG Investigators. Endovascular stent grafting versus open surgical repair of descending thoracic aortic aneurysms in low-risk patients: a multicenter comparative trial. J Thorac Cardiovasc Surg. 2007; 133: 369–377.
13. Greenberg RK, Lu Q, Roselli EE, et al. Contemporary analysis of descending thoracic and thoracoabdominal aneurysm repair: a comparison of endovascular and open techniques. Circulation. 2008; 118: 808–817.
14. Lazorthes G, Gouaze A, Zadeh JO, Santini JJ, Lazorthes Y, Burdin P. Arterial vascularization of the spinal cord. Recent studies of the anastomotic substitution pathways. J Neurosurg. 1971; 35: 253–262.
15. Lazorthes G, Poulhes J, Bastide G, Roulleau J, Chancholle AR. Research on the arterial vascularization of the medulla; applications to medullary pathology [in French]. Bull Acad Natl Med. 1957; 141: 464–477.
16. Lazorthes G, Poulhes J, Bastide G, Roulleau J, Chancholle AR. Arterial vascularization of the spine; anatomic research and applications in pathology of the spinal cord and aorta [in French]. Neurochirurgie. 1958; 4: 3–19.
17. Kuniyoshi Y, Koja K, Miyagi K, et al. Prevention of postoperative paraplegia during thoracoabdominal aortic surgery. Ann Thorac Surg. 2003; 76: 1477–1484.
18. Safi HJ, Miller CC III, Carr C, Iliopoulos DC, Dorsay DA, Baldwin JC. Importance of intercostal artery reattachment during thoracoabdominal aortic aneurysm repair. J Vasc Surg. 1998; 27: 58–68.
19. Acher CW, Wynn MM, Mell MW, Tefera G, Hoch JR. A quantitative assessment of the impact of intercostal artery reimplantation on paralysis risk in thoracoabdominal aortic aneurysm repair. Ann Surg. 2008; 248: 529–540.
20. Acher C. It is not just assisted circulation, hypothermic arrest, or clamp and sew. J Thorac Cardiovasc Surg. 2010; 140 (suppl): S136–S141; discussion S142–S146.
21. Mitchell RS. Stent grafts for the thoracic aorta: a new paradigm? Ann Thorac Surg. 2002; 74: S1818–S1832.
22. Svensson LG, Kouchoukos NT, Miller DC, et al, for The Society of Thoracic Surgeons Endovascular Surgery Task Force. Expert consensus document on the treatment of descending thoracic aortic disease using endovascular stent-grafts. Ann Thorac Surg. 2008; 85 (suppl): S1–S41.
23. Czerny M, Eggebrecht H, Sodeck G, et al. Mechanisms of symptomatic spinal cord ischemia after TEVAR: insights from the European Registry of Endovascular Aortic Repair Complications (EuREC). J Endovasc Ther. 2012; 19: 37–43.
24. Etz CD, Homann TM, Luehr M, et al. Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries. Eur J Cardiothorac Surg. 2008; 33: 1030–1038.
25. Zoli S, Etz CD, Roder F, et al. Experimental two-stage simulated repair of extensive thoracoabdominal aneurysms reduces paraplegia risk. Ann Thorac Surg. 2010; 90: 722–729.
26. Bischoff MS, Scheumann J, Brenner RM, et al. Staged approach prevents spinal cord injury in hybrid surgical-endovascular thoracoabdominal aortic aneurysm repair: an experimental model. Ann Thorac Surg. 2011; 92: 138–146.
Back to Top | Article Outline


This is an experimental proof-of-concept study on one animal. The authors demonstrated that they were able to selectively identify and occlude by coil embolization all intercostal and lumbar aortic spinal arteries in a healthy pig. They proposed that this may be a useful technique clinically to reduce ischemic spinal cord injury after open endovascular thoracoabdominal aortic aneurysm repair. Although this study shows feasibility in an animal model, the clinical utility of this approach has not been demonstrated and will require further studies. Moreover, although this was possible in a healthy pig, it may be very difficult to occlude these vessels in a diseased thoracoabdominal aorta. On top of this, a monitoring technique would have to be developed to allow an interventionalist to occlude enough of the spinal blood supply to allow for preconditioning without causing paraplegia.


Aortic surgery; Collateral network; Spinal cord; Descending aorta; Thoracoabdominal aneurysm; Paraplegia; Ischemic preconditioning; Coil embolization

©2014 by the International Society for Minimally Invasive Cardiothoracic Surgery