- Email: [email protected]
Multilayered Parallel Endografting for Urgent Endovascular Repair of a Severely Angulated Thoracoabdominal Aortic Aneurysm
Selected Technique Multilayered Parallel Endografting for Urgent Endovascular Repair of a Severely Angulated Thoracoabdominal Aortic Aneurysm Sukgu M. Han, Sarah Wartman, Sung W. Ham, Eric C. Kuo, Vincent L. Rowe, and Fred A. Weaver, Los Angeles, California
Total endovascular repair of TAAA using branched, fenestrated stent grafts have been performed with promising midterm results. However, severe angulation of the aorta as well as close proximity of the visceral and renal artery ostia pose a significant technical challenge in designing and implanting branched, fenestrated stent grafts. Parallel grafting offers an alternative technique, allowing an urgent, or emergent total endovascular repair of symptomatic, or ruptured TAAA. We describe a technique of 4-vessel incorporation in a total endovascular repair of TAAA, using multilayered parallel endografting via bilateral femoral and unilateral brachial access. A 76-year-old male with severe chronic obstructive pulmonary disease and coronary artery disease presented with a symptomatic 9 cm extent IV thoracoabdominal aortic aneurysm. The thoracic, and paravisceral segments of his aorta, as well as the iliac arteries were severely angulated, whereas the superior mesenteric and the celiac arteries had a common origin. An urgent total endovascular aortic repair was performed. The aorta and the iliac arteries were straightened by placing stiff wires from bilateral femoral arteries in a ‘‘buddy’’ fashion. In addition, a brachiofemoral ‘‘body-floss’’ wire was established. Over this body-floss wire, thoracic stent grafts were deployed in multiple layers, alternating with parallel branch stents into visceral and renal arteries. Distally, a bifurcated modular stent graft was deployed down to the common iliacs, achieving complete aneurysm exclusion. Patient recovered well without complications and was discharged home in 5 days. Postoperative computed tomography scan showed patent visceral and renal stents and complete exclusion of the aneurysm without evidence of endoleak.
INTRODUCTION Open repair of thoracoabdominal aortic aneurysms (TAAAs) carries a combined morbidity and mortality
The authors have no conflicts of interest or financial disclosures relevant to the manuscript. Division of Vascular Surgery and Endovascular Therapy, Comprehensive Aortic Center, CardioVascular Thoracic Institute Keck Medical Center of University of Southern California, Los Angeles, CA. Correspondence to: Sukgu M. Han, MD, Division of Vascular Surgery and Endovascular Therapy, Department of Surgery, University of Southern California, Los Angeles, CA 90033, USA; E-mail: sukgu. [email protected] Ann Vasc Surg 2017; -: 1–6 http://dx.doi.org/10.1016/j.avsg.2017.02.007 Ó 2017 Elsevier Inc. All rights reserved. Manuscript received: November 15, 2016; manuscript accepted: February 11, 2017; published online: - - -
risks of up to 30%.1 Although custom manufactured fenestrated and branched stent grafts have been used to achieve total endovascular repair, access to these devices remains limited to a few centers in the United States. To eliminate the delay in treatment associated with custom manufacturing, some have performed ‘‘physician-modified endovascular grafting’’ by creating fenestrations or branches to the existing devices.2e4 The impact of such alterations on the durability of repair remains unknown. Currently, one size fits all; off-the-shelf manufactured devices are under investigation. However, under investigational protocol, these devices can still take up to 50 days from manufacturing to implantation.5 Alternatively, various parallel grafting techniques have been used to treat pararenal aneurysms and TAAA, incorporating 1 to 4 abdominal aortic branches 1
2 Han et al.
Annals of Vascular Surgery
Fig. 1. (A) Severely angulated paravisceral aorta and iliac arteries. (B) Paravisceral aorta makes 2 consecutive 90 turns. (C) Bilateral external iliac arteries making near complete coil. (D) The celiac and superior
mesenteric arteries were found to have a common trunk. (E) Right renal artery origin is at the same level as the common visceral trunk.
during endovascular repair.6e8 For true TAAA, the previously described ‘‘Sandwich’’ technique involves 4 simultaneous branch stents placed between 2 aortic stent grafts and includes multiple sheaths placed across the aortic arch.8 Although innovative, this configuration has been associated with a higher risk of stroke and reports of large gutter leak formation from crossing branch stents.6 In this article, we describe a modified version of the sandwich technique, ‘‘multilayered parallel endografting’’ (MPEG), designed to mitigate the aforementioned limitations and achieve total endovascular repair of an angulated TAAA using commercially available off-the-shelf devices. The patient gave an explicit consent to use and publish his deidentified clinical information.
(Fig. 1A). The paravisceral aortic segment was noted to make 2 90 turns (Fig. 1B), whereas the external iliac arteries made a near complete turn (Fig. 1C). The superior mesenteric and the celiac arteries had a common origin (Fig. 1D), and the right renal artery origin was at the same level as this common trunk (Fig. 1E).
CASE A 79-year-old male with severe chronic obstructive pulmonary disease, coronary artery disease, hypertension, and hyperlipidemia presented to an outside hospital with abdominal pain radiating to the back. Computed tomography (CT) scan revealed a 9 cm extent IV TAAA, and the patient was transferred to our aortic center. The patient’s comorbidities made him extremely high risk for open repair. Given the symptomatic nature of his aneurysm, the patient was offered an urgent total endovascular repair using off-the-shelf components. A detailed review of the CT angiography showed that the thoracic, and paravisceral aortic segments, as well as the iliac arteries were severely angulated
TECHNIQUE Under general anesthesia, open exposure of the left proximal brachial and bilateral femoral arteries was performed. After full systemic heparinization, the brachial artery was accessed with a 5-Fr sheath and the bilateral femoral arteries with 7-Fr sheaths. The aorta and the iliac arteries were straightened by placing stiff wires from bilateral femoral arteries in a ‘‘buddy’’ fashion. The right femoral artery was sequentially dilated up to 16-Fr sheath and the left to 22-Fr sheath over stiff wires. With bilateral stiff wires in place, a brachiofemoral through-andthrough ‘‘body-floss’’ access was established by capturing the brachial flexible 0.035-inch hydrophilic wire with an endovascular snare introduced through the left femoral sheath. Over this through-and-through wire, the left brachial access was upsized to a long 12-Fr Flex Ansel Guiding Sheath (Cook Medical, Bloomington, IN), and the distal tip of the sheath was positioned at the middescending thoracic aorta. The superior mesenteric artery (SMA) was precannulated from the brachial approach with a 0.018-inch wire as a marker.
Volume
-, -
2017
Endovascular TAAA repair using layered parallel endografting 3
Fig. 2. (A) The first layer aortic endograft (cTAG) deployed 1 cm above the common visceral trunk. (B) Celiac artery catheterization inside the first layer aortic endograft, followed by deployment of the second layer aortic endograft inside the celiac branch stent. This creates a parallel branch graft to the celiac artery. (C)
Repeated branch stenting alternating with aortic endograft deployment creates layered parallel endografts, incorporating 4 visceral and renal branches. (D) The TAAA repair is completed with deployment of bifurcated endograft in standard fashion.
The first conformable TAG (cTAG; W.L. Gore & Associates, Flagstaff, AZ) was deployed so that the distal end of the endograft was 1 cm above the common trunk origin of the celiac and SMA (Fig. 2A). Puncturing the valve of the 12-Fr brachial sheath, a long coaxial 7-Fr sheath was introduced into the cTAG. Using a 5-Fr angled tip catheter and a 0.035-inch flexible wire, we cannulated the celiac artery from the brachial approach. The 7-Fr brachial sheath position in the celiac artery was maintained,
as the second layer cTAG was deployed above the common visceral trunk. This configuration provided a stable platform to position the appropriately sized Viabahn (W.L. Gore & Associates) covered stents extending 2 cm above the second layer cTAG, creating a parallel branch stent graft (Fig. 2B). We repeated these steps for the SMA, and the bilateral renal arteries to create multiple layered parallel endografts, each time leaving a 0.018-inch wire behind within the branch stents before moving
4 Han et al.
Annals of Vascular Surgery
surgery, the patient developed respiratory failure from exacerbation of the underlying chronic obstructive pulmonary disease and expired.
DISCUSSION
Fig. 3. Multiple punctures of the 4 leaflets of the sheath valve allow wire access to all visceral and renal vessels, through a single 12-Fr brachial sheath.
onto the next target vessel with the coaxial 7-Fr sheath (Fig. 2C). Inside the last cTAG extending from suprarenal to infrarenal aorta, an Excluder bifurcated stent graft (W.L. Gore & Associates) was deployed. The bifurcated modular endovascular aortic aneurysm (EVAR) was completed in the standard fashion, sealing into the common iliac arteries bilaterally (Fig. 2D). Complete 4-vessel, visceral, and renal branch incorporation was performed with a single 12-Fr brachial access. This was enabled by sequential layering of the branch stents through repeated puncturing of the valve of the 12-Fr brachial sheath, creating a multiple coaxial system (Fig. 3). The list of implanted devices is provided in Table I. A kissing balloon technique was used to balloon mold the proximal seal and the overlap zones, protecting the SMA branch stent with a balloon. The right renal branch stent appeared to have a sharp angulation as it entered the native artery. This was corrected with additional reinforcement with an uncovered self-expanding stent. The seal between the lowest cTAG and the bifurcated endograft was reinforced with a cuff. The final angiography showed patent branch stents and complete exclusion of the TAAA without endoleak. At the completion of the procedure, the patient was awakened moving all 4 extremities in stable condition. Postoperatively, patient had uncomplicated recovery with full neurological function and was discharged home on postoperative day 5. Postoperative CT scan obtained at 6 months demonstrated patent branch stents and complete aneurysm exclusion without endoleak or gutter leaks (Fig. 4). The maximum aneurysm sac diameter regressed from 9 to 8.6 cm. Nine months after his
First described as a bail out maneuver for inadvertent high deployment of infrarenal aortic endograft,9 parallel grafting technique in ‘‘snorkel/ chimney’’ configuration is typically used to raise the proximal seal zone during endovascular repair of a juxtarenal or pararenal abdominal aortic aneurysm. Evolving from the unilateral or bilateral renal artery incorporation of ‘‘snorkel/chimney’’ technique, the ‘‘sandwich’’ technique was described to treat true TAAA by placing multiple branch stents simultaneously between 2 thoracic stent grafts.8 This technique required bilateral brachial artery access with 2 7- or 8-Fr sheaths across the aortic arch, cannulating 2 vessels from each side. Dorsey et al. sought to separate the 4-branch stents by groups of 2 between layers in ‘‘terrace’’ configuration. This also required 3 separate open arterial cutdowns in the left upper extremity and simultaneous 4-way kissing balloon molding. Both the ‘‘sandwich’’ and ‘‘terrace’’ techniques require multiple sheaths introduced in the upper extremities, often placing one or more sheaths across the aortic arch. Multiple sheath placement across the aortic arch has been implicated in higher perioperative stroke rates seen in multiple snorkel EVARs.6 In addition, multiple snorkel EVARs suffer from large volume, high flow gutter leaks as the branch stents may cross each other, creating a large gutter.8 Our ‘‘MPEG’’ technique mitigates both of these issues. By layering the branch stents one at a time, we are able to convert the simultaneous 4-vessel access into 4, sequential single-vessel parallel grafts. This modification avoids the increased risk of stroke associated with having to cross the aortic arch with multiple sheaths. In addition, layering of each branch between the 2 thoracic aortic endografts of the same diameter minimizes the potential for gutter leaks. This idea was born out of our previous experience with pararenal snorkel EVARs. We observed that a single snorkel EVAR seldom resulted in gutter leak, whereas obvious leaks were more frequently seen in 2 or more snorkels. In our opinion, parallel endografts should not be used as a replacement of fenestrated, branched devices manufactured through sound engineering principles and tested against stringent mechanical durability standards. However, parallel endografts can have technical advantages over fenestrated,
Volume
-, -
2017
Endovascular TAAA repair using layered parallel endografting 5
Table I. Devices used to complete MPEG Device
Specifications
Aortic components cTAG 34 mm 10 cm cTAG 34 mm 20 cm Excluder cuff 36 mm 4.5 cm Excluder Ipsi 35mm 14.5mm 16 cm Trunk Excluder Iliac 20 mm 13.5 cm limb 18 mm 9.5 cm Branch components Viabahn 8 mm 15 cm 7 mm 15 cm 6 mm 15 cm Visipro 6 mm 4 cm
Number
4 1 1 1 1 1 1 1 2 1
branched devices in certain anatomic configurations. Tortuosity of the branch bearing aortic segments, as well as access vessels, presence of occlusive disease of the target branch vessels, and proximity of the target vessel orifice can all present significant challenges to design and implantation of the fenestrated, branched endografts. As demonstrated in our case, MPEG can enable total endovascular repair of TAAAs containing such features using readily available, off-the-shelf endovascular components. Although the degree of difficulty in the initial implantation of MPEG may be less than the branched, fenestrated counterparts, several technical aspects deserve careful consideration. First, we have found that a brachiofemoral ‘‘body-floss’’ access establishes a more stable platform for branch cannulation in tortuous aortic anatomy. As the TAAA enlarges both in diameter and length, a type 3 aortic arch is commonly encountered. The through-and-through access and coaxial introduction of sheaths prevent prolapse of the sheath and enhance pushability and steerability during visceral and renal cannulations. Second, we have also observed that the same through-and-through wire provides an excellent platform for introducing and deploying the aortic endografts in multiple layers. By modulating the tension of the through-and-through wire during introduction of the aortic endografts, one can avoid catching the edge of the previous endografts. Third, preservation of wire access to each branch stent throughout the case is crucial. Once the wire access is lost, re-engaging the top of the branch stents is typically very difficult and often impossible. Our choice of devices used in MPEG is based on detailed knowledge of each component. Our preferred aortic endografts are cTAGs. Unlike other
Fig. 4. Six-month postoperative CT scan showing complete aneurysm exclusion without endoleak. All 4branch stents remained patent.
commercially available thoracic endografts constructed with discrete rows of stents sewn onto the fabric material, cTAG design uses a tightly overlapping self-expanding nitinol stents in helical configuration, allowing more opportunity to conform around the branch stents. We prefer to use Viabhans for branch stent grafts (W.L. Gore & Associates). Similar to cTAGs, Viabahns are self-expanding covered stents that contain many overlapping helical nitinol stent rows. The mechanical properties of the Viabhan stents show higher flexibility and greater radial force providing an improved kink and compression resistance. Despite the similar design constructs, the interaction between the aortic endografts and branch stents is likely a complicated function of many variables including the aortic contour, the course of the branch stents along the aortic endografts, the relative diameters of the aortic endografts, and branch stents with the respective radial force and compression resistance. These complex interactions are difficult to predict and may result in gutter leak or branch compression. The ideal cross-sectional configuration of a branch stent may be oval rather than a perfect circle, resulting from gentle, harmonious partial compression. When snorkeling 2 or more branches, 30% oversizing of the aortic endograft is typically recommended.7,8 However, we have chosen to use cTAGs of the same diameter to create multiple layers in our case. This reduces the additive effects of compressive radial force generated by the multiple layers of aortic endografts on the branch stents placed in the outer layers. Consequently, when necessary, the mechanical resistance to compression or kinking of the branch stents can be
6 Han et al.
augmented by placing additional reinforcing layer of uncovered self-expanding stents. The ideal length of the overlap between the aortic endografts and the branch stents remains unknown. The original sandwich technique suggested a minimum of 5 cm of overlap for each branch.8 Undoubtedly, the optimal length of the overlap zone for each branch is a competitive balance between the branch stent length and gutter length. Although lengthening of the gutter will decrease the likelihood of persistent gutter leak, the longer branch stent may compromise the patency by the same mechanism. This technique has several limitations. First, the long-term durability of parallel EVARs for TAAA is unknown, as the literature is dominated by juxtarenal and pararenal abdominal aortic aneurysms.6,7,10 Second, the cost of multiple components required for this technique is high. We have chosen to consider the device cost against the technique that enabled total endovascular repair of a high-risk patient with complex anatomy, who otherwise may be deemed unsalvageable. We have performed this procedure and demonstrated its feasibility. In our opinion, the procedural merit of this technique should not be negated by its high cost. Third, secondary intervention, particularly endovascular branch stent revision, is extremely difficult, as the visibility of the top of each layered branch stent is poor. In the event of branch occlusion, secondary intervention may require open abdominal approach to retrograde stent placement or surgical bypass to visceral and renal arteries. Despite these limitations, MPEG has the advantage of using off-the-shelf endovascular components. This technique provides a way to achieve urgent or emergent total endovascular repair even
Annals of Vascular Surgery
in thoracoabdominal aortic severely angulated anatomy.
aneurysms
with
REFERENCES 1. Rigberg DA, McGory ML, Zingmond DS, et al. Thirty-day mortality statistics underestimate the risk of repair of thoracoabdominal aortic aneurysms: a statewide experience. J Vasc Surg 2006;43:217e22. 2. Oderich GS, Fatima J, Gloviczki P. Stent graft modification with mini-cuff reinforced fenestrations for urgent repair of thoracoabdominal aortic aneurysms. J Vasc Surg 2011;54: 1522e6. 3. Scali ST, Neal D, Sollanek V, et al. Outcomes of surgeonmodified fenestrated-branched endograft repair for acute aortic pathology. J Vasc Surg 2015;62:1148e1159.e2. 4. Sweet MP, Starnes BW, Tatum B. Endovascular treatment of thoracoabdominal aortic aneurysm using physicianmodified endografts. J Vasc Surg 2015;62:1160e7. 5. Fernandez CC, Sobel JD, Gasper WJ, et al. Standard off-theshelf versus custom-made multibranched thoracoabdominal aortic stent grafts. J Vasc Surg 2016;63:1208e15. 6. Donas KP, Torsello G, Bisdas T, et al. Early outcomes for fenestrated and chimney endografts in the treatment of pararenal aortic pathologies are not significantly different: a systematic review with pooled data analysis. J Endovasc Ther 2012;19:723e8. 7. Dorsey C, Chandra V, Lee JT. The ‘‘terrace technique’’e totally endovascular repair of a type IV thoracoabdominal aortic aneurysm. Ann Vasc Surg 2014;28:1563.e11e6. 8. Donas KP, Lee JT, Lachat M, et al., PERICLES Investigators. Collected world experience about the performance of the snorkel/chimney endovascular technique in the treatment of complex aortic pathologies: the PERICLES registry. Ann Surg 2015;262:546e53. 9. Lobato AC, Camacho-Lobato L. A new technique to enhance endovascular thoracoabdominal aortic aneurysm therapydthe sandwich procedure. Semin Vasc Surg 2012;25:153e60. 10. Greenberg RK, Clair D, Srivastava S, et al. Should patients with challenging anatomy be offered endovascular aneurysm repair? J Vasc Surg 2003;38:990e6.
Total endovascular repair of TAAA using branched, fenestrated stent grafts have been performed with promising midterm results. However, severe angulation of the aorta as well as close proximity of the visceral and renal artery ostia pose a significant technical challenge in designing and implanting branched, fenestrated stent grafts. Parallel grafting offers an alternative technique, allowing an urgent, or emergent total endovascular repair of symptomatic, or ruptured TAAA. We describe a technique of 4-vessel incorporation in a total endovascular repair of TAAA, using multilayered parallel endografting via bilateral femoral and unilateral brachial access. A 76-year-old male with severe chronic obstructive pulmonary disease and coronary artery disease presented with a symptomatic 9 cm extent IV thoracoabdominal aortic aneurysm. The thoracic, and paravisceral segments of his aorta, as well as the iliac arteries were severely angulated, whereas the superior mesenteric and the celiac arteries had a common origin. An urgent total endovascular aortic repair was performed. The aorta and the iliac arteries were straightened by placing stiff wires from bilateral femoral arteries in a ‘‘buddy’’ fashion. In addition, a brachiofemoral ‘‘body-floss’’ wire was established. Over this body-floss wire, thoracic stent grafts were deployed in multiple layers, alternating with parallel branch stents into visceral and renal arteries. Distally, a bifurcated modular stent graft was deployed down to the common iliacs, achieving complete aneurysm exclusion. Patient recovered well without complications and was discharged home in 5 days. Postoperative computed tomography scan showed patent visceral and renal stents and complete exclusion of the aneurysm without evidence of endoleak.
INTRODUCTION Open repair of thoracoabdominal aortic aneurysms (TAAAs) carries a combined morbidity and mortality
The authors have no conflicts of interest or financial disclosures relevant to the manuscript. Division of Vascular Surgery and Endovascular Therapy, Comprehensive Aortic Center, CardioVascular Thoracic Institute Keck Medical Center of University of Southern California, Los Angeles, CA. Correspondence to: Sukgu M. Han, MD, Division of Vascular Surgery and Endovascular Therapy, Department of Surgery, University of Southern California, Los Angeles, CA 90033, USA; E-mail: sukgu. [email protected] Ann Vasc Surg 2017; -: 1–6 http://dx.doi.org/10.1016/j.avsg.2017.02.007 Ó 2017 Elsevier Inc. All rights reserved. Manuscript received: November 15, 2016; manuscript accepted: February 11, 2017; published online: - - -
risks of up to 30%.1 Although custom manufactured fenestrated and branched stent grafts have been used to achieve total endovascular repair, access to these devices remains limited to a few centers in the United States. To eliminate the delay in treatment associated with custom manufacturing, some have performed ‘‘physician-modified endovascular grafting’’ by creating fenestrations or branches to the existing devices.2e4 The impact of such alterations on the durability of repair remains unknown. Currently, one size fits all; off-the-shelf manufactured devices are under investigation. However, under investigational protocol, these devices can still take up to 50 days from manufacturing to implantation.5 Alternatively, various parallel grafting techniques have been used to treat pararenal aneurysms and TAAA, incorporating 1 to 4 abdominal aortic branches 1
2 Han et al.
Annals of Vascular Surgery
Fig. 1. (A) Severely angulated paravisceral aorta and iliac arteries. (B) Paravisceral aorta makes 2 consecutive 90 turns. (C) Bilateral external iliac arteries making near complete coil. (D) The celiac and superior
mesenteric arteries were found to have a common trunk. (E) Right renal artery origin is at the same level as the common visceral trunk.
during endovascular repair.6e8 For true TAAA, the previously described ‘‘Sandwich’’ technique involves 4 simultaneous branch stents placed between 2 aortic stent grafts and includes multiple sheaths placed across the aortic arch.8 Although innovative, this configuration has been associated with a higher risk of stroke and reports of large gutter leak formation from crossing branch stents.6 In this article, we describe a modified version of the sandwich technique, ‘‘multilayered parallel endografting’’ (MPEG), designed to mitigate the aforementioned limitations and achieve total endovascular repair of an angulated TAAA using commercially available off-the-shelf devices. The patient gave an explicit consent to use and publish his deidentified clinical information.
(Fig. 1A). The paravisceral aortic segment was noted to make 2 90 turns (Fig. 1B), whereas the external iliac arteries made a near complete turn (Fig. 1C). The superior mesenteric and the celiac arteries had a common origin (Fig. 1D), and the right renal artery origin was at the same level as this common trunk (Fig. 1E).
CASE A 79-year-old male with severe chronic obstructive pulmonary disease, coronary artery disease, hypertension, and hyperlipidemia presented to an outside hospital with abdominal pain radiating to the back. Computed tomography (CT) scan revealed a 9 cm extent IV TAAA, and the patient was transferred to our aortic center. The patient’s comorbidities made him extremely high risk for open repair. Given the symptomatic nature of his aneurysm, the patient was offered an urgent total endovascular repair using off-the-shelf components. A detailed review of the CT angiography showed that the thoracic, and paravisceral aortic segments, as well as the iliac arteries were severely angulated
TECHNIQUE Under general anesthesia, open exposure of the left proximal brachial and bilateral femoral arteries was performed. After full systemic heparinization, the brachial artery was accessed with a 5-Fr sheath and the bilateral femoral arteries with 7-Fr sheaths. The aorta and the iliac arteries were straightened by placing stiff wires from bilateral femoral arteries in a ‘‘buddy’’ fashion. The right femoral artery was sequentially dilated up to 16-Fr sheath and the left to 22-Fr sheath over stiff wires. With bilateral stiff wires in place, a brachiofemoral through-andthrough ‘‘body-floss’’ access was established by capturing the brachial flexible 0.035-inch hydrophilic wire with an endovascular snare introduced through the left femoral sheath. Over this through-and-through wire, the left brachial access was upsized to a long 12-Fr Flex Ansel Guiding Sheath (Cook Medical, Bloomington, IN), and the distal tip of the sheath was positioned at the middescending thoracic aorta. The superior mesenteric artery (SMA) was precannulated from the brachial approach with a 0.018-inch wire as a marker.
Volume
-, -
2017
Endovascular TAAA repair using layered parallel endografting 3
Fig. 2. (A) The first layer aortic endograft (cTAG) deployed 1 cm above the common visceral trunk. (B) Celiac artery catheterization inside the first layer aortic endograft, followed by deployment of the second layer aortic endograft inside the celiac branch stent. This creates a parallel branch graft to the celiac artery. (C)
Repeated branch stenting alternating with aortic endograft deployment creates layered parallel endografts, incorporating 4 visceral and renal branches. (D) The TAAA repair is completed with deployment of bifurcated endograft in standard fashion.
The first conformable TAG (cTAG; W.L. Gore & Associates, Flagstaff, AZ) was deployed so that the distal end of the endograft was 1 cm above the common trunk origin of the celiac and SMA (Fig. 2A). Puncturing the valve of the 12-Fr brachial sheath, a long coaxial 7-Fr sheath was introduced into the cTAG. Using a 5-Fr angled tip catheter and a 0.035-inch flexible wire, we cannulated the celiac artery from the brachial approach. The 7-Fr brachial sheath position in the celiac artery was maintained,
as the second layer cTAG was deployed above the common visceral trunk. This configuration provided a stable platform to position the appropriately sized Viabahn (W.L. Gore & Associates) covered stents extending 2 cm above the second layer cTAG, creating a parallel branch stent graft (Fig. 2B). We repeated these steps for the SMA, and the bilateral renal arteries to create multiple layered parallel endografts, each time leaving a 0.018-inch wire behind within the branch stents before moving
4 Han et al.
Annals of Vascular Surgery
surgery, the patient developed respiratory failure from exacerbation of the underlying chronic obstructive pulmonary disease and expired.
DISCUSSION
Fig. 3. Multiple punctures of the 4 leaflets of the sheath valve allow wire access to all visceral and renal vessels, through a single 12-Fr brachial sheath.
onto the next target vessel with the coaxial 7-Fr sheath (Fig. 2C). Inside the last cTAG extending from suprarenal to infrarenal aorta, an Excluder bifurcated stent graft (W.L. Gore & Associates) was deployed. The bifurcated modular endovascular aortic aneurysm (EVAR) was completed in the standard fashion, sealing into the common iliac arteries bilaterally (Fig. 2D). Complete 4-vessel, visceral, and renal branch incorporation was performed with a single 12-Fr brachial access. This was enabled by sequential layering of the branch stents through repeated puncturing of the valve of the 12-Fr brachial sheath, creating a multiple coaxial system (Fig. 3). The list of implanted devices is provided in Table I. A kissing balloon technique was used to balloon mold the proximal seal and the overlap zones, protecting the SMA branch stent with a balloon. The right renal branch stent appeared to have a sharp angulation as it entered the native artery. This was corrected with additional reinforcement with an uncovered self-expanding stent. The seal between the lowest cTAG and the bifurcated endograft was reinforced with a cuff. The final angiography showed patent branch stents and complete exclusion of the TAAA without endoleak. At the completion of the procedure, the patient was awakened moving all 4 extremities in stable condition. Postoperatively, patient had uncomplicated recovery with full neurological function and was discharged home on postoperative day 5. Postoperative CT scan obtained at 6 months demonstrated patent branch stents and complete aneurysm exclusion without endoleak or gutter leaks (Fig. 4). The maximum aneurysm sac diameter regressed from 9 to 8.6 cm. Nine months after his
First described as a bail out maneuver for inadvertent high deployment of infrarenal aortic endograft,9 parallel grafting technique in ‘‘snorkel/ chimney’’ configuration is typically used to raise the proximal seal zone during endovascular repair of a juxtarenal or pararenal abdominal aortic aneurysm. Evolving from the unilateral or bilateral renal artery incorporation of ‘‘snorkel/chimney’’ technique, the ‘‘sandwich’’ technique was described to treat true TAAA by placing multiple branch stents simultaneously between 2 thoracic stent grafts.8 This technique required bilateral brachial artery access with 2 7- or 8-Fr sheaths across the aortic arch, cannulating 2 vessels from each side. Dorsey et al. sought to separate the 4-branch stents by groups of 2 between layers in ‘‘terrace’’ configuration. This also required 3 separate open arterial cutdowns in the left upper extremity and simultaneous 4-way kissing balloon molding. Both the ‘‘sandwich’’ and ‘‘terrace’’ techniques require multiple sheaths introduced in the upper extremities, often placing one or more sheaths across the aortic arch. Multiple sheath placement across the aortic arch has been implicated in higher perioperative stroke rates seen in multiple snorkel EVARs.6 In addition, multiple snorkel EVARs suffer from large volume, high flow gutter leaks as the branch stents may cross each other, creating a large gutter.8 Our ‘‘MPEG’’ technique mitigates both of these issues. By layering the branch stents one at a time, we are able to convert the simultaneous 4-vessel access into 4, sequential single-vessel parallel grafts. This modification avoids the increased risk of stroke associated with having to cross the aortic arch with multiple sheaths. In addition, layering of each branch between the 2 thoracic aortic endografts of the same diameter minimizes the potential for gutter leaks. This idea was born out of our previous experience with pararenal snorkel EVARs. We observed that a single snorkel EVAR seldom resulted in gutter leak, whereas obvious leaks were more frequently seen in 2 or more snorkels. In our opinion, parallel endografts should not be used as a replacement of fenestrated, branched devices manufactured through sound engineering principles and tested against stringent mechanical durability standards. However, parallel endografts can have technical advantages over fenestrated,
Volume
-, -
2017
Endovascular TAAA repair using layered parallel endografting 5
Table I. Devices used to complete MPEG Device
Specifications
Aortic components cTAG 34 mm 10 cm cTAG 34 mm 20 cm Excluder cuff 36 mm 4.5 cm Excluder Ipsi 35mm 14.5mm 16 cm Trunk Excluder Iliac 20 mm 13.5 cm limb 18 mm 9.5 cm Branch components Viabahn 8 mm 15 cm 7 mm 15 cm 6 mm 15 cm Visipro 6 mm 4 cm
Number
4 1 1 1 1 1 1 1 2 1
branched devices in certain anatomic configurations. Tortuosity of the branch bearing aortic segments, as well as access vessels, presence of occlusive disease of the target branch vessels, and proximity of the target vessel orifice can all present significant challenges to design and implantation of the fenestrated, branched endografts. As demonstrated in our case, MPEG can enable total endovascular repair of TAAAs containing such features using readily available, off-the-shelf endovascular components. Although the degree of difficulty in the initial implantation of MPEG may be less than the branched, fenestrated counterparts, several technical aspects deserve careful consideration. First, we have found that a brachiofemoral ‘‘body-floss’’ access establishes a more stable platform for branch cannulation in tortuous aortic anatomy. As the TAAA enlarges both in diameter and length, a type 3 aortic arch is commonly encountered. The through-and-through access and coaxial introduction of sheaths prevent prolapse of the sheath and enhance pushability and steerability during visceral and renal cannulations. Second, we have also observed that the same through-and-through wire provides an excellent platform for introducing and deploying the aortic endografts in multiple layers. By modulating the tension of the through-and-through wire during introduction of the aortic endografts, one can avoid catching the edge of the previous endografts. Third, preservation of wire access to each branch stent throughout the case is crucial. Once the wire access is lost, re-engaging the top of the branch stents is typically very difficult and often impossible. Our choice of devices used in MPEG is based on detailed knowledge of each component. Our preferred aortic endografts are cTAGs. Unlike other
Fig. 4. Six-month postoperative CT scan showing complete aneurysm exclusion without endoleak. All 4branch stents remained patent.
commercially available thoracic endografts constructed with discrete rows of stents sewn onto the fabric material, cTAG design uses a tightly overlapping self-expanding nitinol stents in helical configuration, allowing more opportunity to conform around the branch stents. We prefer to use Viabhans for branch stent grafts (W.L. Gore & Associates). Similar to cTAGs, Viabahns are self-expanding covered stents that contain many overlapping helical nitinol stent rows. The mechanical properties of the Viabhan stents show higher flexibility and greater radial force providing an improved kink and compression resistance. Despite the similar design constructs, the interaction between the aortic endografts and branch stents is likely a complicated function of many variables including the aortic contour, the course of the branch stents along the aortic endografts, the relative diameters of the aortic endografts, and branch stents with the respective radial force and compression resistance. These complex interactions are difficult to predict and may result in gutter leak or branch compression. The ideal cross-sectional configuration of a branch stent may be oval rather than a perfect circle, resulting from gentle, harmonious partial compression. When snorkeling 2 or more branches, 30% oversizing of the aortic endograft is typically recommended.7,8 However, we have chosen to use cTAGs of the same diameter to create multiple layers in our case. This reduces the additive effects of compressive radial force generated by the multiple layers of aortic endografts on the branch stents placed in the outer layers. Consequently, when necessary, the mechanical resistance to compression or kinking of the branch stents can be
6 Han et al.
augmented by placing additional reinforcing layer of uncovered self-expanding stents. The ideal length of the overlap between the aortic endografts and the branch stents remains unknown. The original sandwich technique suggested a minimum of 5 cm of overlap for each branch.8 Undoubtedly, the optimal length of the overlap zone for each branch is a competitive balance between the branch stent length and gutter length. Although lengthening of the gutter will decrease the likelihood of persistent gutter leak, the longer branch stent may compromise the patency by the same mechanism. This technique has several limitations. First, the long-term durability of parallel EVARs for TAAA is unknown, as the literature is dominated by juxtarenal and pararenal abdominal aortic aneurysms.6,7,10 Second, the cost of multiple components required for this technique is high. We have chosen to consider the device cost against the technique that enabled total endovascular repair of a high-risk patient with complex anatomy, who otherwise may be deemed unsalvageable. We have performed this procedure and demonstrated its feasibility. In our opinion, the procedural merit of this technique should not be negated by its high cost. Third, secondary intervention, particularly endovascular branch stent revision, is extremely difficult, as the visibility of the top of each layered branch stent is poor. In the event of branch occlusion, secondary intervention may require open abdominal approach to retrograde stent placement or surgical bypass to visceral and renal arteries. Despite these limitations, MPEG has the advantage of using off-the-shelf endovascular components. This technique provides a way to achieve urgent or emergent total endovascular repair even
Annals of Vascular Surgery
in thoracoabdominal aortic severely angulated anatomy.
aneurysms
with
REFERENCES 1. Rigberg DA, McGory ML, Zingmond DS, et al. Thirty-day mortality statistics underestimate the risk of repair of thoracoabdominal aortic aneurysms: a statewide experience. J Vasc Surg 2006;43:217e22. 2. Oderich GS, Fatima J, Gloviczki P. Stent graft modification with mini-cuff reinforced fenestrations for urgent repair of thoracoabdominal aortic aneurysms. J Vasc Surg 2011;54: 1522e6. 3. Scali ST, Neal D, Sollanek V, et al. Outcomes of surgeonmodified fenestrated-branched endograft repair for acute aortic pathology. J Vasc Surg 2015;62:1148e1159.e2. 4. Sweet MP, Starnes BW, Tatum B. Endovascular treatment of thoracoabdominal aortic aneurysm using physicianmodified endografts. J Vasc Surg 2015;62:1160e7. 5. Fernandez CC, Sobel JD, Gasper WJ, et al. Standard off-theshelf versus custom-made multibranched thoracoabdominal aortic stent grafts. J Vasc Surg 2016;63:1208e15. 6. Donas KP, Torsello G, Bisdas T, et al. Early outcomes for fenestrated and chimney endografts in the treatment of pararenal aortic pathologies are not significantly different: a systematic review with pooled data analysis. J Endovasc Ther 2012;19:723e8. 7. Dorsey C, Chandra V, Lee JT. The ‘‘terrace technique’’e totally endovascular repair of a type IV thoracoabdominal aortic aneurysm. Ann Vasc Surg 2014;28:1563.e11e6. 8. Donas KP, Lee JT, Lachat M, et al., PERICLES Investigators. Collected world experience about the performance of the snorkel/chimney endovascular technique in the treatment of complex aortic pathologies: the PERICLES registry. Ann Surg 2015;262:546e53. 9. Lobato AC, Camacho-Lobato L. A new technique to enhance endovascular thoracoabdominal aortic aneurysm therapydthe sandwich procedure. Semin Vasc Surg 2012;25:153e60. 10. Greenberg RK, Clair D, Srivastava S, et al. Should patients with challenging anatomy be offered endovascular aneurysm repair? J Vasc Surg 2003;38:990e6.