- Email: [email protected]
A Modular Branched Stent-Graft System for Sutureless Anastomoses in Extensive Aortic Arch Replacement—A Porcine Study
Basic Science Research A Modular Branched Stent-Graft System for Sutureless Anastomoses in Extensive Aortic Arch ReplacementdA Porcine Study Chien-Chang Chen,1,2 Yu-Chieh Tseng,1 Chih-Chan Lin,3 Chien-Fang Li,4 and Ming-Long Yeh,1 Tainan and Chiayi, Taiwan
Background: We developed a modular branched stent-graft system to test whether it is feasible for sutureless anastomoses in extensive aortic arch replacement. Methods: Extensive aortic arch replacement was performed using the three-branched stentgraft system in eight pigs. Under deep hypothermic circulatory arrest, sutureless anastomoses were established at the distal aortic arch and the two supra-aortic arteries with the modular branched stent-graft system. External bandings on the distal arches were applied in six pigs (banding group) and not applied in two pigs (control group). No external banding was applied on the two supra-aortic arteries. Results: Successful procedures were achieved in all pigs in the banding group, whereas failures were seen in the control group owing to leakage from the distal arch anastomoses. The anastomosis at each distal aortic arch was completed in 10 minutes in the banding group and in 5 minutes in the control group; the anastomosis of each supra-aortic artery was achieved in 5 minutes. Median durations of the circulatory arrest, aortic cross-clamping, and cardiopulmonary bypass were 30, 67, and 174 minutes, respectively. The postoperative computed tomography revealed adequate alignment of the stents and appropriate size matching between stent-graft and native aorta. Histological examinations revealed no pressure necrosis at the sutureless anastomotic sites. Conclusions: This study confirmed the technical feasibility of sutureless anastomoses with the modular branched stent-graft system in porcine extensive aortic arch replacement. An external banding is essential for the secure hemostasis of the distal arch anastomosis, but it is not required for the supra-aortic arteries.
INTRODUCTION 1
Division of Biomechanics, Institute of Biomedical Engineering, National Cheng-Kung University, Tainan, Taiwan. 2 Division of Cardiovascular Surgery, Chiayi Christian Hospital, Chiayi, Taiwan. 3 Department of Medical Research, Chi-Mei Medical Center, Tainan, Taiwan. 4 Department of Pathology, Chi-Mei Medical Center, Tainan, Taiwan. Correspondence to: Ming-Long Yeh, PhD, Institute of Biomedical Engineering, National Cheng-Kung University, Number 1 University Road, Tainan 70101, Taiwan; E-mail: [email protected] Ann Vasc Surg 2012; 26: 527–536 DOI: 10.1016/j.avsg.2012.01.003 Ó Annals of Vascular Surgery Inc.
Aortic arch replacement is a technical challenge for most cardiothoracic surgeons. During the past decade, several novel open techniques have been developed to reduce cerebral and myocardial ischemia and to completely eliminate the residual arch aneurysmal tissues. Examples of the techniques were antegrade selective cerebral perfusion, trifurcated graft technique and its modification, and T-graft technique for treating most of the aortic arch diseases.1e4 However, when the distal portion of the descending aorta should be included in the so-called ‘‘extensive aortic arch replacement,’’ the sutured anastomosis of the deep descending aortic 527
528 Chen et al.
stump becomes a very difficult task for the surgeons. To overcome the difficulty, Kato et al. proximalized the aortic cut end to the arch level and inserted the stented elephant trunk graft into the distal descending aorta (DAo) to achieve sutureless anastomosis. Then, the conventional sutured anastomosis was made at the aortic arch stump to include the whole stented elephant trunk.5,6 The similar concept known as frozen elephant trunk technique was also used to treat extensive aortic arch aneurysms. In brief, the hybrid procedure was performed by combining the conventional trifurcated graft technique for aortic arch replacement with open stent-grafting for descending aorta.7 Although these techniques have greatly reduced surgical difficulties, suture-based anastomoses at the aortic arch and the supra-aortic arteries by the skilled surgeons are still required. Although the suture-based anastomosis is the gold standard in vascular surgery, it has potential disadvantages, including high technical requirement, long anastomotic time, suture-hole bleeding, interstitch leakage, and vessel wall destruction. These disadvantages are especially evident for the less experienced surgeons when dealing with multiple deep and fragile supra-aortic and arch anastomoses in the narrow surgical spaces. Therefore, development of an easy, expeditious, and less destructive anastomotic method is necessary to further simplify the procedure and improve the surgical results of extensive aortic arch replacement. Since the successful treatment of aortic aneurysms with stent-grafts by Parodi et al. and Volodos et al. in 1991, endovascular aneurysm repair has revolutionized the field of aortic surgery.8,9 This success supports the idea that the selfexpanding stent-graft is able to make fast and secure sutureless ‘‘intravascular’’ end-to-end anastomoses. Whether the similar notion can be safely applied to the anastomoses at the open stumps of aortic arch and supra-aortic arteries remains unknown. In this porcine experimental study, we developed a modular three-branched stent-graft system to test the technical feasibility for sutureless anastomoses in open extensive aortic arch replacement.
METHODS All animal testing procedures were approved by the Institutional Animal Care and Use Committee and conducted according to the Animal Use Protocol of the Chi-Mei Medical Center, Tainan, Taiwan.
Annals of Vascular Surgery
Animals and Preoperative Computed Tomography Eight adult Landrace-Yorkshire-Duroc pigs were used for the extensive aortic arch replacement, and the average age of the studied pigs was 4 months. Their body weight, length, and height were recorded, and their body surface area was estimated using an automatic calculator. On sedation of the pig, the contrast-enhanced computed tomographic aortography (CTA) was taken for each pig. A reconstructed three-dimensional image of the thoracic aorta was used to verify the entire aortic configuration. The following data for experimental pigs were also secured: 1) the diameters of brachiocephalic arteries (BCAs), left subclavian arteries (LSCAs), distal arches adjacent to LSCA orifices, and distal DAos 5 cm downstream of the distal arch; 2) the lengths of respective BCA and LSCA; and 3) the lengths and diameters of ascending aortas (AAos) (Fig. 1). The Three-Branched Stent-Graft A three-branched Dacron aortic arch graft (Fig. 2) was constructed according to the parameters obtained from the preoperative CTA. The first sidearm was used for arterial cannulation (the perfusion sidearm). The right and left ‘‘head sidearms’’ were used for the BCA and LSCA, respectively; each of them was 3 cm in length. The external diameter of the main graft was determined according to the measured distal arch diameter (just distal to LSCA). The diameter of the main graft was 10% to 20% in excess of that of the distal aortic arch. Four Gianturco stainless-steel Z-stents, all 14 mm in length, were inserted onto the inner surface of the distal half of the main graft. The stents were affixed to the graft with 5-0 polypropylene sutures. A plastic suction tube was placed in the main graft. The stented part of the main graft was compressed with chain ties,10 and then sterilized with ethylene oxide for 24 hours. General Anesthesia Anesthetic induction was started with an intramuscular injection of Zoletil 50 mL/kg (Virbac, Carros, France). Bilateral auricular veins were used for intravenous access. The superficial femoral artery served as the site of an arterial line. Inhaled 2% to 5% Sevoflurane (Abbott, Queenborough, England) through endotracheal tube was used for maintenance of general anesthesia. Intravenous pancuronium (0.1 mg/kg) (Organon, Oss, Holland) was used every 3 hours for muscle relaxation. The
Vol. 26, No. 4, May 2012
A modular branched stent-graft system 529
Fig. 1. Preoperative computed tomographic aortography. (A) Transaxial view at the level of the two supra-aortic branches. (B) Sagittal view of the aortic arch and its branches.
arterial blood pressure, electrocardiography, and rectal temperature were continuously monitored. Intravenous 1% Fresofol (5 mg/kg) (Fresenius Kabi, Graz, Austria) was used for anesthetic maintenance during cardiopulmonary bypass (CPB). Operative Procedures Under general anesthesia, the mediastinum was exposed through a median sternotomy. After systemic heparinization (300 U/kg), CPB was initiated from the proximal aortic arch and the right atrium. The initial CPB flow was 2.5 L/min/m, which was then lowered with cooling of the systemic perfusate. When the rectal temperature reached 18 C, the AAo was cross-clamped. Antegrade 4 C crystalloid cardioplegia was administered through the aortic root cannula to induce electrical arrest of the myocardium. The aortic clamp was released, and the aortic arch was opened longitudinally under deep hypothermic circulatory arrest. The AAo was transected at the insertion site of aortic root cannula. The aortic arch at the level of the LSCA was partially transected. The BCA and LSCA were detached from the aortic arch for better exposure. The stented portion of the main graft was implanted into the DAo, and then deployed by releasing the chain ties. The external aortic band (Dacron strap, 10 mm in width) was wrapped around the distal aortic arch stump in six pigs (the banding group), whereas this banding procedure was not performed in two pigs (the control group). A balloon was inserted into the opened stent-graft under direct vision without using a wire. It was inflated to ensure good apposition of the stentgraft to the inner aortic wall for this sutureless anastomosis. The two head sidearms were slightly
inserted into the BCA and LSCA, with the overlapping lengths of 0.5 to 1.0 cm. The self-expanding Viabahns of 5 cm in length (Gore, Flagstaff, AZ) were inserted into the two head sidearms, BCA, and LSCA using the guidewire technique. They were deployed as interposition stent-grafts and then balloon-dilated under wire support for close attachment. Two 5-0 polypropylene fixation stitches were made on the proximal portion of each sidearm and Viabahn to prevent migration. The diameters of the Viabahns were 10% in excess of that of the incorporated sidearms. Therefore, sutureless anastomoses were achieved in the two head sidearms, BCA, and LSCA without external banding in all pigs. After all sutureless anastomoses were completed, the arterial cannula was repositioned to the perfusion sidearm and then CPB was restarted. The proximal stentless part of the main graft was clamped after evacuation of air in the Trendelenburg position. Myocardium was protected with direct catheterized coronary perfusion with 4 C crystalloid cardioplegia. Systemic perfusate was rewarmed. The proximal end of the graft was end-to-end anastomosed to the AAo with 4-0 polypropylene sutures (Fig. 3). The aortic clamp was released, and cardiac rhythm was restarted. The flow in the BCA and LSCA was detected with Doppler ultrasonography, and the patency of the DAo was determined by the superficial femoral arterial pressure. Postoperative Care and Follow-up The pigs were allowed to recover spontaneously from anesthesia without reversal of muscle relaxants. One week postoperatively, contrast-enhanced CTA was performed to visualize the entire thoracic aortic vasculature. Then, the entire aorta including
530 Chen et al.
Annals of Vascular Surgery
Definition of Technical Success A successful procedure was defined as the one that meets all of the following items:
Fig. 2. The three-branched stent-graft. The right and left head sidearms marked as brachiocephalic artery (BCAs) and left subclavian artery (LSCAs) are used for connecting to the brachiocephalic and left subclavian arteries, respectively. ‘‘C’’ indicates the first sidearm for the arterial cannulation (the perfusion sidearm).
1. Complete replacement of the AAo, the total arch, and the proximal DAo with the branched stent-graft. 2. No bleeding from any of the sutureless anastomoses. 3. No dislodgement of any of the sutureless anastomoses. 4. Maintenance of the patency of all sutureless anastomoses. 5. Successful weaning from CPB. 6. Survival of the animals after awakening from anesthesia.
RESULTS Physical Data of Test Pigs The mean body weight, length, height, and surface area of the pigs used for aortic arch replacement were 35.1 ± 6.5 kg, 108.3 ± 4.1 cm, 43.3 ± 6.1 cm, and 0.97 ± 0.1 m2, respectively. Vascular and Graft Parameters
Fig. 3. Illustration of the extensive aortic arch replacement with the modular branched stent-graft system. The arrow indicates external aortic banding at the distal aortic arch. The head sidearms were slightly inserted into the BCA and LSCA, with the overlapping lengths of 0.5 to 1.0 cm. Then, the Viabahns were deployed inside the sidearms, BCA, and LSCA to achieve sutureless anastomoses.
aortic tissue, anastomotic sites, and grafts were excised after killing each pig. The following four specimens were obtained: the BCA, LSCA, AAo, and distal arch anastomoses. These specimens were fixed in 4% formalin under 0 mm Hg for 24 hours. The arterial tissues at the anastomotic sites were microscopically examined with hematoxylin and eosin stain.
1) The diameter and length of BCA were 9.5 ± 0.8 mm and 29.0 ± 8.1 mm, respectively. The corresponding values with the LSCA were 8.5 ± 1.0 mm and 22.9 ± 4.7 mm, respectively. The diameter of the distal arch just lateral to the LSCA orifice was 18.5 ± 1.4 mm. The diameter of descending aorta 5 cm downstream from the distal arch was 17.6 ± 2.3 mm. The BCA was consistently larger and longer than the LSCA. 2) The mean external diameters of the right and left head sidearms and the main graft were 9.7 ± 0.8 mm, 8.8 ± 1.3 mm, and 21.3 ± 2.0 mm, respectively. The length of each head sidearm was 3 cm, and the length of the distal stented portion of the main graft was 6 cm. Surgical Parameters and Results Each BCA or LSCA anastomosis was performed in 5 minutes, and each distal aortic arch anastomosis was performed in 10 minutes in the banding group and in 5 minutes in the control group. Median durations of the circulatory arrest, aortic cross-clamping, and CPB were 30 minutes (range, 25e60 minutes), 67 minutes (range, 60e92 minutes), and 174 minutes (range, 120e185 minutes), respectively (Table I). Most of the circulatory arrest and aortic
Vol. 26, No. 4, May 2012
A modular branched stent-graft system 531
Table I. Surgical parameters and results Data of Pigs
Banding group (n ¼ 6)
Control group (n ¼ 2)
Number EAB TCA (min) AXC (min) CPB (min) Aortic arch leak BCA/LSCA leak Neurological complication Weaning from CPB
1 + 60 90 174
2 + 25 92 185
3 + 35 80 175
4 + 30 60 120
5 + 30 60 180
6 + 26 67 133
7 25 62 166 + +
8 32 67 174 +
+
+
+
+
+
+
+
+
EAB, external aortic banding; BCA, brachiocephalic artery; LSCA, left subclavian artery; TCA, total circulatory arrest time; AXC, aortic cross-clamping time; CPB, cardiopulmonary bypass time.
cross-clamping time was spent in the mobilization of the aortic arch and its branches. Most of the CPB time was spent in cooling and rewarming of the systemic perfusate. Successful procedures were achieved in all six pigs in the banding group. All porcine hearts were successfully weaned from CPB. There was minimal blood loss in all pigs in the banding group. The two pigs in the control group died due to bleeding from distal aortic arch anastomoses after weaning from CPB. The anastomoses of the BCA and LSCA were secure, without blood leakage in all eight experimental pigs. There was no slippage of the stentgraft components throughout the procedures in all pigs. Completion of the operation in the banding group is demonstrated in Figure 4. After recovering from anesthesia, all pigs opened eyes spontaneously with free movement of their four limbs, except for one pig in the control group that suffered from right hemiplegia. Postoperative CTA The postoperative three-dimensional CTA displayed: 1) adequate alignment of the stents, 2) appropriate size matching between the stent-graft, and 3) good patency of the aorta, BCA, LSCA, and related tributaries (Fig. 5). Histological Examinations Microscopic examinations 1 week after the operation showed a comparable thickness of the tunica media between the AAo (Fig. 6A) and the distal arch anastomoses (Fig. 6E). However, the infiltration of inflammatory cells around the needle hole was apparent in the AAo (Fig. 6B) but not in the distal arch (Fig. 6F). Of note, the tunica adventitia of the AAo showed remarkable active inflammation
together with aggregation of foreign body giant cells (see the inset) and formation of the granulation tissue (Fig. 6C), both of which were absent in the tunica adventitia of the distal arch (Fig. 6G). The architectures at the BCA (Fig. 6D) and LSCA (Fig. 6H) anastomoses were not altered. There were no signs of pressure necrosis in any of the anastomoses.
DISCUSSION A simple, fast, and secure anastomotic method in aortic arch replacement plays a pivotal role in reducing cerebral ischemia and anastomotic bleeding. The application of a modular branched stent-graft system for sutureless anastomoses at the stumps of aortic arch and supra-aortic arteries is, to the best of our knowledge, an unprecedented trial. Since the initial application of the open stentgrafting technique in distal arch and descending aorta by Kato et al., the stented or frozen elephant trunk techniques are now gaining an increasing acceptance in extensive aortic arch replacement.5e7,11e13 In Europe, the commercialized hybrid stented elephant trunks, E-vita open plus (Jotec, Hechingen, Germany), have also been successfully used in the cases with acute type A aortic dissection and extensive aortic arch aneurysms.14,15 The further development was open placement and inclusion of the branched stentgrafts in aortic arches and descending aortas, that is, the ‘‘inclusion methods.’’ With these methods, sutureless anastomoses were performed at the distal portions of descending aortas and supra-aortic arteries, and then the aortic stumps could be proximalized to thoracic aortic zone 2 to zone 0 (Criado classification)16 depending on whether single-, double-, or triple-branched stent-grafts were used.
532 Chen et al.
Annals of Vascular Surgery
Fig. 4. Completion of the extensive aortic arch replacement. The sutureless anastomoses were made at the distal arch stump (arrow), BCA, and LSCA. The external
band was obscured by the surrounding tissue. The stump of the perfusion sidearm was labeled as ‘‘C.’’
The observed satisfactory short-term results support technical feasibility of these techniques.17e20 All inclusion methods shared a common concept that conventional sutured anastsomoses were made for hemostasis at the aortic stumps, and thereby the prostheses were included in the aortic walls. However, for secure placement of the doublebranched or triple-branched stent grafts, the diameters of sidearm grafts and the intersidearm distances must be closely corresponded to the supra-aortic arteries. Otherwise, an endoleak from cervical branches might complicate the postoperative courses.19,20 This endoleak could not be seen during the procedures because systemic perfusion was initiated only after the inclusion of prostheses was completed. In the aortic dissection where an intimal tear is located within the stent-grafting range, it is possible that an endoleak may cause persistent false lumen postoperatively. In contrast to the previous stent-graft inclusion methods, the head sidearms of the present branched stent-graft were exposed in the pericardial cavities, with the Viabahns being used for sutureless anastomoses in the sidearms and the supra-aortic arteries (Fig. 4). Therefore, no endoleak could occur. Our data revealed that the Viabahns exerted adequate radial force for holding sutureless anstomoses at
both ends of the supra-aortic arteries and the head sidearms. This modular approach is also distinct from the customized unibody branched stent-graft,17e20 as it allows instant combinations of the stent-graft components. Similar to the modular concept of endovascular repair for abdominal aortic aneurysm that used bilateral iliac extension stent-grafts, the Viabahns were mounted on the two-head sidearms proximally, with their distal ends serving for sutureless anastomoses in the BCA and LSCA. This characteristic may provide convenience and immediateness for future emergent clinical application in open extensive aortic arch replacement. Hemostasis at the aortic stumps is important, as demonstrated by the previous inclusion methods that used sutured anastomoses to prevent the retrograde blood flow from intercostal arteries and small distal endoleaks.5,13,17,21 In contrast, sutureless anastomoses were used for the hemostasis in this experiment. However, it should be noticed that the main trunk of the present branched stent-graft exhibited front and cephalic motion on each ventricular contraction. A similar phenomenon has also been found by Akin et al. in the movement of native thoracic aorta.22 Because this motion can cause instability of the hemostatic zone at the
Vol. 26, No. 4, May 2012
Fig. 5. The CT scan 1 week postoperatively. It revealed adequate expansion of the branched stent-graft and patent distal arterial flow. The anastomotic zones of BCA and LSCA were in parallel shapes, whereas those of distal arches were in curved funnel shapes. The arrow indicates the fourth stent, which was used to increase graft support and reduce the risk of graft kink.
‘‘curved’’ distal arch, the use of the external banding is essential for stabilizing the area to achieve the secure sutureless anastomosis. On the other hand, the experimental results suggested that the external bandings were unnecessary for anastomoses at the BCA and LSCA. This discrepancy may be explained by the following reasons. First, the anastomotic zones of BCA and LSCA were in parallel shapes, whereas those of distal arches were in curved funnel shapes (Fig. 5). Therefore, apposition of the stent-grafts was better in the former cases. Second, the conventional sutures in the AAo anastomosis provided a proximal fixation of the main trunk of the branched stent-graft. When an external banding was also applied on the distal arch anastomosis, both ends of the main trunk were fixed, and the motion of the two head sidearms was reduced accordingly. These two factors resulted in more stable hemostatic zones in the two
A modular branched stent-graft system 533
supra-aortic arteries, and hence, the external banding was unnecessary. Recently, Lachet et al. proposed an innovative technique, Viabahn Open Revascularization Technique (VORTEC), for expeditious renal revascularization in hybrid endovascular aortic repair.23 The technique was further applied to the revascularization of superior mesenteric, celiac, and supraaortic arteries (the last was called the telescoping technique).24,25 In brief, the Viabahn was inserted from the anterior side of the target artery using the Seldinger technique and then deployed partly within the artery and partly outside the artery to attain sutureless anastomosis. The advantage of this technique is the obviation of difficult complete vessel exposure and graft anastomoses, thereby reducing the duration of flow interruption. However, the porcine BCA and LSCA were too short (mean lengths of BCA: 29.0 mm and LSCA: 22.9 mm) for stable landing of the Viabahns in this experiment, as the ‘‘side approach’’ of the technique should sacrifice part of the effective intravascular sealing length. To increase the landing length, we modified the technique to the ‘‘end approach’’ by inserting the Viabahns through the orifices of BCA and LSCA after opening the aortic arch. The tradeoff was consumption of circulatory arrest time. Nevertheless, the procedure was simple and swift and did not require much additional cerebral ischemic time. For the distal arch anastomosis, the aortic wall was sandwiched between the inner stent-graft and the external band to secure the blood sealing zone. It was possible that the vascular supply to the aortic wall could be jeopardized, which could result in aortic wall necrosis. However, the present study revealed no signs of necrosis in distal arch anastomosis, based on both the gross observation and the microscopic examination (Fig. 6EeG). Moreover, the foreign body giant cell reaction frequently observed in the sutured anastomosis (Fig. 6C) was absent in the sutureless anastomosis (Fig. 6G). Generally, the time required for each sutureless anastomosis was less than that for the sutured one. However, the durations of circulatory arrest and aortic cross-clamping in this experiment were only slightly less than those for the existing branched stent-graft inclusion techniques applied in humans (circulatory arrest time: 30 minutes vs. 31 minutes17 or 36 minutes20; aortic-clamping time: 67 minutes vs. 84 minutes17). The nonprominent surgical durations were mainly attributed to the unique porcine anatomy and our limited experiences in porcine aortic arch replacement. The porcine thoracic anatomy is different from the human anatomy in
534 Chen et al.
Fig. 6. Microscopic examinations of the four anastomotic sites 1 week postoperatively. (A) The AAo wall (100). The scale bar in the right lower corner is 1 mm. (B) Tunica media of the AAo (200). The arrow indicates the suture hole with surrounding inflammatory cells. (C) Tunica adventitia of the AAo (400). The arrow depicts the foreign body giant cells, with the magnified images in the inset in the right upper corner. (D) BCA
several aspects. As illustrated in Figure 7, the porcine heart is dextral, with the apex pointing toward the right anterior side. The porcine ascending aorta is short, approximately 2 cm in length, which is located obliquely behind the large right ventricular outflow tract and the pulmonary artery. The porcine aortic arch is deeply seated in left posterior part of the pericardial cavity where the working space is narrow. The porcine axillary arteries are too small to cannulate for antegrade selective cerebral perfusion. These characteristics make porcine aortic arch replacement much more difficult than that for humans. Therefore, most of the circulatory arrest and aortic cross-clamping time was spent in the mobilization of the aortic arch and its branches. Approximately one-fourth of the CPB time was consumed in cooling and rewarming of the systemic perfusate. One concern of the arch graft is that the unsupported portion could buckle, kink, or narrow over time. To prevent the drawback, we placed a fourth stent proximal to the BCA sidearm in addition to the other three distal stents to increase graft support and reduce the risk of graft kink (Fig. 5). The fourth stent was placed as close to the adjacent one as possible to minimize the unsupported portion. We did not place a stent between the sidearms to further increase graft support because the lengthened branched portion of the arch graft could not fit to the small porcine aortic arch anatomy in the narrow
Annals of Vascular Surgery
anastomotic site (100). (E) Distal arch (100). The same scale bar length as panel A in the right lower corner. (F) Tunica media of the distal arch (200). (G) Tunica adventitia of the distal arch (400). Note the scarcity of inflammatory cell infiltrations in comparison with the AAo adventitia (panel C). (H) LSCA anastomotic site (100).
Fig. 7. The anatomy of porcine heart. The porcine heart is dextral, with the apex pointing toward the right anterior side. In this figure, the ascending aorta and aortic arch cannot be seen, as they are located behind the large pulmonary artery and right ventricular outflow tract. The arrow indicates left anterior descending artery, which courses toward the apex (LA, left auricle; LV, left ventricle; RA, right auricle; RV, right ventricle; PA, pulmonary artery; SVC, superior vena cava).
pericardial space. Another issue is the hemodynamic forces over segments of grafts and intervening stents are unknown and could be harmful if there are graft kinks or tears that occur over time. For the sidearm branches, a durable and flexible stent-graft such as Viabahn can reduce the risks
Vol. 26, No. 4, May 2012
of kinks or tears, as described in the VORTEC technique.24 For the main body, future work should be focused on modification of the stent material such as nitinol, adjustment of the stent width and interstent distance, and addition of small intersidearm stents to improve flexibility, conformability, and supporting forces of the stent-graft. In this experiment, an external banding was essential for a secure aortic arch anastomosis. This was achieved by use of a Dacron strap wrapped around the distal arch stump, with its both ends sutured together for appropriate constraint of the external aortic circumference. The procedure took additional 5 minutes in the circulatory arrest period. To simplify the banding technique and shorten the procedural time, one may improve the external band with an adjustable length and autolock mechanism for handy manipulation in future studies. Study Limitations The average life span of Landrace-Yorkshire-Duroc pigs is 10 years. The body weight of a 1-year-old pig can be >100 kg after a normal growth. In practical situations, it is difficult to handle an old pig for an experiment. Therefore, the study was based on the 4-month-old porcine model of normal aorta (30e40 kg in body weight). The aim of the experiment was to test the technical feasibility of the concept of sutureless anastomosis, and therefore, only short-term results were obtained. Experiments on the diseased aortic model with long-term observation are warranted before application in humans.
CONCLUSIONS This study proved the technical feasibility of sutureless anastomoses in porcine extensive aortic arch replacement with a modular branched stent-graft system. The external banding is essential for the secure hemostasis of the distal arch anastomosis, but it is not necessary for the supra-aortic branches.
The authors are grateful to Ms. Pi-Chen Chien for conducting cardiopulmonary bypass. The experimental work was conducted in Chi-Mei Medical Center, Tainan, Taiwan. REFERENCES 1. Chen CC, Hsieh SR. Modified trifurcated graft in acute type A aortic dissection with the least brain ischemic time. Ann Thorac Surg 2007;83:e6e8. 2. Hsieh SR, Chen CC, Wei HJ. A novel arch-first t-graft technique for extensive aortic arch reconstruction. Ann Thorac Surg 2008;85:1814e6.
A modular branched stent-graft system 535
3. Panos A, Murith N, Bednarkiewicz M, Khatchatourov G. Axillary cerebral perfusion for arch surgery in acute type A dissection under moderate hypothermia. Eur J Cardiothorac Surg 2006;29:1036e9. 4. Spielvogel D, Strauch JT, Minanov OP, et al. Aortic arch replacement using a trifurcated graft and selective cerebral antegrade perfusion. Ann Thorac Surg 2002;74:S1810e4. discussion S1825e32. 5. Kato M, Ohnishi K, Kaneko M, et al. New graft-implanting method for thoracic aortic aneurysm or dissection with a stented graft. Circulation 1996;94:II188e93. 6. Kato M, Kuratani T, Kaneko M, et al. The results of total arch graft implantation with open stent-graft placement for type a aortic dissection. J Thorac Cardiovasc Surg 2002;124:531e40. 7. Karck M, Chavan A, Hagl C, et al. The frozen elephant trunk technique: a new treatment for thoracic aortic aneurysms. J Thorac Cardiovasc Surg 2003;125:1550e3. 8. Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491e9. 9. Volodos NL, Karpovich IP, Troyan VI, et al. Clinical experience of the use of self-fixing synthetic prostheses for remote endoprosthetics of the thoracic and the abdominal aorta and iliac arteries through the femoral artery and as intraoperative endoprosthesis for aorta reconstruction. Vasa Suppl 1991;33:93e5. 10. Shibata T, Hirai H, Fukui T, et al. Assembly and deployment of a branched arch stent graft using the transaortic approach. Ann Thorac Surg 2005;79:1790e2. 11. Miyamoto S, Hadama T, Anai H, et al. Stented elephant trunk method for multiple thoracic aneurysms. Ann Thorac Surg 2001;71:705e7. 12. Sun L, Qi R, Chang Q, et al. Surgery for acute type A dissection with the tear in the descending aorta using a stented elephant trunk procedure. Ann Thorac Surg 2009;87:1177e80. 13. Sun LZ, Qi RD, Chang Q, et al. Surgery for acute type A dissection using total arch replacement combined with stented elephant trunk implantation: experience with 107 patients. J Thorac Cardiovasc Surg 2009;138:1358e62. 14. Mestres CA, Fernandez C, Josa M, Mulet J. Hybrid antegrade repair of the arch and descending thoracic aorta with a new integrated stent-dacron graft in acute type A aortic dissection: a look into the future with new devices. Interact Cardiovasc Thorac Surg 2007;6:257e9. 15. Schoenhoff FS, Schmidli J, Eckstein FS, et al. The frozen elephant trunk: an interesting hybrid endovascularsurgical technique to treat complex pathologies of the thoracic aorta. J Vasc Surg 2007;45:597e9. 16. Criado FJ, Barnatan MF, Rizk Y, et al. Technical strategies to expand stent-graft applicability in the aortic arch and proximal descending thoracic aorta. J Endovasc Ther 2002;9(Suppl. 2): II32e8. 17. Chen LW, Dai XF, Lu L, et al. Extensive primary repair of the thoracic aorta in acute type A aortic dissection by means of ascending aorta replacement combined with open placement of triple-branched stent graft: early results. Circulation 2010;122:1373e8. 18. Chen LW, Dai XF, Yang GF, et al. Open-branched stent graft placement makes total arch replacement easier for acute type A aortic dissection. Ann Thorac Surg 2010;89:1688e90. 19. Chen LW, Dai XF, Zhang GC, Lu L. Total aortic arch reconstruction with open placement of triple-branched stent graft for acute type A dissection. J Thorac Cardiovasc Surg 2010;139:1654.e1e5.e1.
536 Chen et al.
20. Shimamura K, Kuratani T, Matsumiya G, et al. Hybrid endovascular aortic arch repair using branched endoprosthesis: the second-generation ‘‘branched’’ open stent-grafting technique. J Thorac Cardiovasc Surg 2009;138:46e52. discussion 52e3. 21. Sun LZ, Qi RD, Chang Q, et al. Is total arch replacement combined with stented elephant trunk implantation justified for patients with chronic stanford type A aortic dissection? J Thorac Cardiovasc Surg 2009;138:892e6. 22. Akin I, Ince H, Kische S, Nienaber CA. Implication of thoracic aortic stent-graft conformability on clinical outcomes. Acta Chir Belg 2009;109:20e6.
Annals of Vascular Surgery
23. Lachat M, Mayer D, Criado FJ, et al. New technique to facilitate renal revascularization with use of telescoping selfexpanding stent grafts: VORTEC. Vascular 2008;16:69e72. 24. Donas KP, Lachat M, Rancic Z, et al. Early and midterm outcome of a novel technique to simplify the hybrid procedures in the treatment of thoracoabdominal and pararenal aortic aneurysms. J Vasc Surg 2009;50:1280e4. 25. Donas KP, Rancic Z, Lachat M, et al. Novel sutureless telescoping anastomosis revascularization technique of supraaortic vessels to simplify combined open endovascular procedures in the treatment of aortic arch pathologies. J Vasc Surg 2010;51:836e41.
Background: We developed a modular branched stent-graft system to test whether it is feasible for sutureless anastomoses in extensive aortic arch replacement. Methods: Extensive aortic arch replacement was performed using the three-branched stentgraft system in eight pigs. Under deep hypothermic circulatory arrest, sutureless anastomoses were established at the distal aortic arch and the two supra-aortic arteries with the modular branched stent-graft system. External bandings on the distal arches were applied in six pigs (banding group) and not applied in two pigs (control group). No external banding was applied on the two supra-aortic arteries. Results: Successful procedures were achieved in all pigs in the banding group, whereas failures were seen in the control group owing to leakage from the distal arch anastomoses. The anastomosis at each distal aortic arch was completed in 10 minutes in the banding group and in 5 minutes in the control group; the anastomosis of each supra-aortic artery was achieved in 5 minutes. Median durations of the circulatory arrest, aortic cross-clamping, and cardiopulmonary bypass were 30, 67, and 174 minutes, respectively. The postoperative computed tomography revealed adequate alignment of the stents and appropriate size matching between stent-graft and native aorta. Histological examinations revealed no pressure necrosis at the sutureless anastomotic sites. Conclusions: This study confirmed the technical feasibility of sutureless anastomoses with the modular branched stent-graft system in porcine extensive aortic arch replacement. An external banding is essential for the secure hemostasis of the distal arch anastomosis, but it is not required for the supra-aortic arteries.
INTRODUCTION 1
Division of Biomechanics, Institute of Biomedical Engineering, National Cheng-Kung University, Tainan, Taiwan. 2 Division of Cardiovascular Surgery, Chiayi Christian Hospital, Chiayi, Taiwan. 3 Department of Medical Research, Chi-Mei Medical Center, Tainan, Taiwan. 4 Department of Pathology, Chi-Mei Medical Center, Tainan, Taiwan. Correspondence to: Ming-Long Yeh, PhD, Institute of Biomedical Engineering, National Cheng-Kung University, Number 1 University Road, Tainan 70101, Taiwan; E-mail: [email protected] Ann Vasc Surg 2012; 26: 527–536 DOI: 10.1016/j.avsg.2012.01.003 Ó Annals of Vascular Surgery Inc.
Aortic arch replacement is a technical challenge for most cardiothoracic surgeons. During the past decade, several novel open techniques have been developed to reduce cerebral and myocardial ischemia and to completely eliminate the residual arch aneurysmal tissues. Examples of the techniques were antegrade selective cerebral perfusion, trifurcated graft technique and its modification, and T-graft technique for treating most of the aortic arch diseases.1e4 However, when the distal portion of the descending aorta should be included in the so-called ‘‘extensive aortic arch replacement,’’ the sutured anastomosis of the deep descending aortic 527
528 Chen et al.
stump becomes a very difficult task for the surgeons. To overcome the difficulty, Kato et al. proximalized the aortic cut end to the arch level and inserted the stented elephant trunk graft into the distal descending aorta (DAo) to achieve sutureless anastomosis. Then, the conventional sutured anastomosis was made at the aortic arch stump to include the whole stented elephant trunk.5,6 The similar concept known as frozen elephant trunk technique was also used to treat extensive aortic arch aneurysms. In brief, the hybrid procedure was performed by combining the conventional trifurcated graft technique for aortic arch replacement with open stent-grafting for descending aorta.7 Although these techniques have greatly reduced surgical difficulties, suture-based anastomoses at the aortic arch and the supra-aortic arteries by the skilled surgeons are still required. Although the suture-based anastomosis is the gold standard in vascular surgery, it has potential disadvantages, including high technical requirement, long anastomotic time, suture-hole bleeding, interstitch leakage, and vessel wall destruction. These disadvantages are especially evident for the less experienced surgeons when dealing with multiple deep and fragile supra-aortic and arch anastomoses in the narrow surgical spaces. Therefore, development of an easy, expeditious, and less destructive anastomotic method is necessary to further simplify the procedure and improve the surgical results of extensive aortic arch replacement. Since the successful treatment of aortic aneurysms with stent-grafts by Parodi et al. and Volodos et al. in 1991, endovascular aneurysm repair has revolutionized the field of aortic surgery.8,9 This success supports the idea that the selfexpanding stent-graft is able to make fast and secure sutureless ‘‘intravascular’’ end-to-end anastomoses. Whether the similar notion can be safely applied to the anastomoses at the open stumps of aortic arch and supra-aortic arteries remains unknown. In this porcine experimental study, we developed a modular three-branched stent-graft system to test the technical feasibility for sutureless anastomoses in open extensive aortic arch replacement.
METHODS All animal testing procedures were approved by the Institutional Animal Care and Use Committee and conducted according to the Animal Use Protocol of the Chi-Mei Medical Center, Tainan, Taiwan.
Annals of Vascular Surgery
Animals and Preoperative Computed Tomography Eight adult Landrace-Yorkshire-Duroc pigs were used for the extensive aortic arch replacement, and the average age of the studied pigs was 4 months. Their body weight, length, and height were recorded, and their body surface area was estimated using an automatic calculator. On sedation of the pig, the contrast-enhanced computed tomographic aortography (CTA) was taken for each pig. A reconstructed three-dimensional image of the thoracic aorta was used to verify the entire aortic configuration. The following data for experimental pigs were also secured: 1) the diameters of brachiocephalic arteries (BCAs), left subclavian arteries (LSCAs), distal arches adjacent to LSCA orifices, and distal DAos 5 cm downstream of the distal arch; 2) the lengths of respective BCA and LSCA; and 3) the lengths and diameters of ascending aortas (AAos) (Fig. 1). The Three-Branched Stent-Graft A three-branched Dacron aortic arch graft (Fig. 2) was constructed according to the parameters obtained from the preoperative CTA. The first sidearm was used for arterial cannulation (the perfusion sidearm). The right and left ‘‘head sidearms’’ were used for the BCA and LSCA, respectively; each of them was 3 cm in length. The external diameter of the main graft was determined according to the measured distal arch diameter (just distal to LSCA). The diameter of the main graft was 10% to 20% in excess of that of the distal aortic arch. Four Gianturco stainless-steel Z-stents, all 14 mm in length, were inserted onto the inner surface of the distal half of the main graft. The stents were affixed to the graft with 5-0 polypropylene sutures. A plastic suction tube was placed in the main graft. The stented part of the main graft was compressed with chain ties,10 and then sterilized with ethylene oxide for 24 hours. General Anesthesia Anesthetic induction was started with an intramuscular injection of Zoletil 50 mL/kg (Virbac, Carros, France). Bilateral auricular veins were used for intravenous access. The superficial femoral artery served as the site of an arterial line. Inhaled 2% to 5% Sevoflurane (Abbott, Queenborough, England) through endotracheal tube was used for maintenance of general anesthesia. Intravenous pancuronium (0.1 mg/kg) (Organon, Oss, Holland) was used every 3 hours for muscle relaxation. The
Vol. 26, No. 4, May 2012
A modular branched stent-graft system 529
Fig. 1. Preoperative computed tomographic aortography. (A) Transaxial view at the level of the two supra-aortic branches. (B) Sagittal view of the aortic arch and its branches.
arterial blood pressure, electrocardiography, and rectal temperature were continuously monitored. Intravenous 1% Fresofol (5 mg/kg) (Fresenius Kabi, Graz, Austria) was used for anesthetic maintenance during cardiopulmonary bypass (CPB). Operative Procedures Under general anesthesia, the mediastinum was exposed through a median sternotomy. After systemic heparinization (300 U/kg), CPB was initiated from the proximal aortic arch and the right atrium. The initial CPB flow was 2.5 L/min/m, which was then lowered with cooling of the systemic perfusate. When the rectal temperature reached 18 C, the AAo was cross-clamped. Antegrade 4 C crystalloid cardioplegia was administered through the aortic root cannula to induce electrical arrest of the myocardium. The aortic clamp was released, and the aortic arch was opened longitudinally under deep hypothermic circulatory arrest. The AAo was transected at the insertion site of aortic root cannula. The aortic arch at the level of the LSCA was partially transected. The BCA and LSCA were detached from the aortic arch for better exposure. The stented portion of the main graft was implanted into the DAo, and then deployed by releasing the chain ties. The external aortic band (Dacron strap, 10 mm in width) was wrapped around the distal aortic arch stump in six pigs (the banding group), whereas this banding procedure was not performed in two pigs (the control group). A balloon was inserted into the opened stent-graft under direct vision without using a wire. It was inflated to ensure good apposition of the stentgraft to the inner aortic wall for this sutureless anastomosis. The two head sidearms were slightly
inserted into the BCA and LSCA, with the overlapping lengths of 0.5 to 1.0 cm. The self-expanding Viabahns of 5 cm in length (Gore, Flagstaff, AZ) were inserted into the two head sidearms, BCA, and LSCA using the guidewire technique. They were deployed as interposition stent-grafts and then balloon-dilated under wire support for close attachment. Two 5-0 polypropylene fixation stitches were made on the proximal portion of each sidearm and Viabahn to prevent migration. The diameters of the Viabahns were 10% in excess of that of the incorporated sidearms. Therefore, sutureless anastomoses were achieved in the two head sidearms, BCA, and LSCA without external banding in all pigs. After all sutureless anastomoses were completed, the arterial cannula was repositioned to the perfusion sidearm and then CPB was restarted. The proximal stentless part of the main graft was clamped after evacuation of air in the Trendelenburg position. Myocardium was protected with direct catheterized coronary perfusion with 4 C crystalloid cardioplegia. Systemic perfusate was rewarmed. The proximal end of the graft was end-to-end anastomosed to the AAo with 4-0 polypropylene sutures (Fig. 3). The aortic clamp was released, and cardiac rhythm was restarted. The flow in the BCA and LSCA was detected with Doppler ultrasonography, and the patency of the DAo was determined by the superficial femoral arterial pressure. Postoperative Care and Follow-up The pigs were allowed to recover spontaneously from anesthesia without reversal of muscle relaxants. One week postoperatively, contrast-enhanced CTA was performed to visualize the entire thoracic aortic vasculature. Then, the entire aorta including
530 Chen et al.
Annals of Vascular Surgery
Definition of Technical Success A successful procedure was defined as the one that meets all of the following items:
Fig. 2. The three-branched stent-graft. The right and left head sidearms marked as brachiocephalic artery (BCAs) and left subclavian artery (LSCAs) are used for connecting to the brachiocephalic and left subclavian arteries, respectively. ‘‘C’’ indicates the first sidearm for the arterial cannulation (the perfusion sidearm).
1. Complete replacement of the AAo, the total arch, and the proximal DAo with the branched stent-graft. 2. No bleeding from any of the sutureless anastomoses. 3. No dislodgement of any of the sutureless anastomoses. 4. Maintenance of the patency of all sutureless anastomoses. 5. Successful weaning from CPB. 6. Survival of the animals after awakening from anesthesia.
RESULTS Physical Data of Test Pigs The mean body weight, length, height, and surface area of the pigs used for aortic arch replacement were 35.1 ± 6.5 kg, 108.3 ± 4.1 cm, 43.3 ± 6.1 cm, and 0.97 ± 0.1 m2, respectively. Vascular and Graft Parameters
Fig. 3. Illustration of the extensive aortic arch replacement with the modular branched stent-graft system. The arrow indicates external aortic banding at the distal aortic arch. The head sidearms were slightly inserted into the BCA and LSCA, with the overlapping lengths of 0.5 to 1.0 cm. Then, the Viabahns were deployed inside the sidearms, BCA, and LSCA to achieve sutureless anastomoses.
aortic tissue, anastomotic sites, and grafts were excised after killing each pig. The following four specimens were obtained: the BCA, LSCA, AAo, and distal arch anastomoses. These specimens were fixed in 4% formalin under 0 mm Hg for 24 hours. The arterial tissues at the anastomotic sites were microscopically examined with hematoxylin and eosin stain.
1) The diameter and length of BCA were 9.5 ± 0.8 mm and 29.0 ± 8.1 mm, respectively. The corresponding values with the LSCA were 8.5 ± 1.0 mm and 22.9 ± 4.7 mm, respectively. The diameter of the distal arch just lateral to the LSCA orifice was 18.5 ± 1.4 mm. The diameter of descending aorta 5 cm downstream from the distal arch was 17.6 ± 2.3 mm. The BCA was consistently larger and longer than the LSCA. 2) The mean external diameters of the right and left head sidearms and the main graft were 9.7 ± 0.8 mm, 8.8 ± 1.3 mm, and 21.3 ± 2.0 mm, respectively. The length of each head sidearm was 3 cm, and the length of the distal stented portion of the main graft was 6 cm. Surgical Parameters and Results Each BCA or LSCA anastomosis was performed in 5 minutes, and each distal aortic arch anastomosis was performed in 10 minutes in the banding group and in 5 minutes in the control group. Median durations of the circulatory arrest, aortic cross-clamping, and CPB were 30 minutes (range, 25e60 minutes), 67 minutes (range, 60e92 minutes), and 174 minutes (range, 120e185 minutes), respectively (Table I). Most of the circulatory arrest and aortic
Vol. 26, No. 4, May 2012
A modular branched stent-graft system 531
Table I. Surgical parameters and results Data of Pigs
Banding group (n ¼ 6)
Control group (n ¼ 2)
Number EAB TCA (min) AXC (min) CPB (min) Aortic arch leak BCA/LSCA leak Neurological complication Weaning from CPB
1 + 60 90 174
2 + 25 92 185
3 + 35 80 175
4 + 30 60 120
5 + 30 60 180
6 + 26 67 133
7 25 62 166 + +
8 32 67 174 +
+
+
+
+
+
+
+
+
EAB, external aortic banding; BCA, brachiocephalic artery; LSCA, left subclavian artery; TCA, total circulatory arrest time; AXC, aortic cross-clamping time; CPB, cardiopulmonary bypass time.
cross-clamping time was spent in the mobilization of the aortic arch and its branches. Most of the CPB time was spent in cooling and rewarming of the systemic perfusate. Successful procedures were achieved in all six pigs in the banding group. All porcine hearts were successfully weaned from CPB. There was minimal blood loss in all pigs in the banding group. The two pigs in the control group died due to bleeding from distal aortic arch anastomoses after weaning from CPB. The anastomoses of the BCA and LSCA were secure, without blood leakage in all eight experimental pigs. There was no slippage of the stentgraft components throughout the procedures in all pigs. Completion of the operation in the banding group is demonstrated in Figure 4. After recovering from anesthesia, all pigs opened eyes spontaneously with free movement of their four limbs, except for one pig in the control group that suffered from right hemiplegia. Postoperative CTA The postoperative three-dimensional CTA displayed: 1) adequate alignment of the stents, 2) appropriate size matching between the stent-graft, and 3) good patency of the aorta, BCA, LSCA, and related tributaries (Fig. 5). Histological Examinations Microscopic examinations 1 week after the operation showed a comparable thickness of the tunica media between the AAo (Fig. 6A) and the distal arch anastomoses (Fig. 6E). However, the infiltration of inflammatory cells around the needle hole was apparent in the AAo (Fig. 6B) but not in the distal arch (Fig. 6F). Of note, the tunica adventitia of the AAo showed remarkable active inflammation
together with aggregation of foreign body giant cells (see the inset) and formation of the granulation tissue (Fig. 6C), both of which were absent in the tunica adventitia of the distal arch (Fig. 6G). The architectures at the BCA (Fig. 6D) and LSCA (Fig. 6H) anastomoses were not altered. There were no signs of pressure necrosis in any of the anastomoses.
DISCUSSION A simple, fast, and secure anastomotic method in aortic arch replacement plays a pivotal role in reducing cerebral ischemia and anastomotic bleeding. The application of a modular branched stent-graft system for sutureless anastomoses at the stumps of aortic arch and supra-aortic arteries is, to the best of our knowledge, an unprecedented trial. Since the initial application of the open stentgrafting technique in distal arch and descending aorta by Kato et al., the stented or frozen elephant trunk techniques are now gaining an increasing acceptance in extensive aortic arch replacement.5e7,11e13 In Europe, the commercialized hybrid stented elephant trunks, E-vita open plus (Jotec, Hechingen, Germany), have also been successfully used in the cases with acute type A aortic dissection and extensive aortic arch aneurysms.14,15 The further development was open placement and inclusion of the branched stentgrafts in aortic arches and descending aortas, that is, the ‘‘inclusion methods.’’ With these methods, sutureless anastomoses were performed at the distal portions of descending aortas and supra-aortic arteries, and then the aortic stumps could be proximalized to thoracic aortic zone 2 to zone 0 (Criado classification)16 depending on whether single-, double-, or triple-branched stent-grafts were used.
532 Chen et al.
Annals of Vascular Surgery
Fig. 4. Completion of the extensive aortic arch replacement. The sutureless anastomoses were made at the distal arch stump (arrow), BCA, and LSCA. The external
band was obscured by the surrounding tissue. The stump of the perfusion sidearm was labeled as ‘‘C.’’
The observed satisfactory short-term results support technical feasibility of these techniques.17e20 All inclusion methods shared a common concept that conventional sutured anastsomoses were made for hemostasis at the aortic stumps, and thereby the prostheses were included in the aortic walls. However, for secure placement of the doublebranched or triple-branched stent grafts, the diameters of sidearm grafts and the intersidearm distances must be closely corresponded to the supra-aortic arteries. Otherwise, an endoleak from cervical branches might complicate the postoperative courses.19,20 This endoleak could not be seen during the procedures because systemic perfusion was initiated only after the inclusion of prostheses was completed. In the aortic dissection where an intimal tear is located within the stent-grafting range, it is possible that an endoleak may cause persistent false lumen postoperatively. In contrast to the previous stent-graft inclusion methods, the head sidearms of the present branched stent-graft were exposed in the pericardial cavities, with the Viabahns being used for sutureless anastomoses in the sidearms and the supra-aortic arteries (Fig. 4). Therefore, no endoleak could occur. Our data revealed that the Viabahns exerted adequate radial force for holding sutureless anstomoses at
both ends of the supra-aortic arteries and the head sidearms. This modular approach is also distinct from the customized unibody branched stent-graft,17e20 as it allows instant combinations of the stent-graft components. Similar to the modular concept of endovascular repair for abdominal aortic aneurysm that used bilateral iliac extension stent-grafts, the Viabahns were mounted on the two-head sidearms proximally, with their distal ends serving for sutureless anastomoses in the BCA and LSCA. This characteristic may provide convenience and immediateness for future emergent clinical application in open extensive aortic arch replacement. Hemostasis at the aortic stumps is important, as demonstrated by the previous inclusion methods that used sutured anastomoses to prevent the retrograde blood flow from intercostal arteries and small distal endoleaks.5,13,17,21 In contrast, sutureless anastomoses were used for the hemostasis in this experiment. However, it should be noticed that the main trunk of the present branched stent-graft exhibited front and cephalic motion on each ventricular contraction. A similar phenomenon has also been found by Akin et al. in the movement of native thoracic aorta.22 Because this motion can cause instability of the hemostatic zone at the
Vol. 26, No. 4, May 2012
Fig. 5. The CT scan 1 week postoperatively. It revealed adequate expansion of the branched stent-graft and patent distal arterial flow. The anastomotic zones of BCA and LSCA were in parallel shapes, whereas those of distal arches were in curved funnel shapes. The arrow indicates the fourth stent, which was used to increase graft support and reduce the risk of graft kink.
‘‘curved’’ distal arch, the use of the external banding is essential for stabilizing the area to achieve the secure sutureless anastomosis. On the other hand, the experimental results suggested that the external bandings were unnecessary for anastomoses at the BCA and LSCA. This discrepancy may be explained by the following reasons. First, the anastomotic zones of BCA and LSCA were in parallel shapes, whereas those of distal arches were in curved funnel shapes (Fig. 5). Therefore, apposition of the stent-grafts was better in the former cases. Second, the conventional sutures in the AAo anastomosis provided a proximal fixation of the main trunk of the branched stent-graft. When an external banding was also applied on the distal arch anastomosis, both ends of the main trunk were fixed, and the motion of the two head sidearms was reduced accordingly. These two factors resulted in more stable hemostatic zones in the two
A modular branched stent-graft system 533
supra-aortic arteries, and hence, the external banding was unnecessary. Recently, Lachet et al. proposed an innovative technique, Viabahn Open Revascularization Technique (VORTEC), for expeditious renal revascularization in hybrid endovascular aortic repair.23 The technique was further applied to the revascularization of superior mesenteric, celiac, and supraaortic arteries (the last was called the telescoping technique).24,25 In brief, the Viabahn was inserted from the anterior side of the target artery using the Seldinger technique and then deployed partly within the artery and partly outside the artery to attain sutureless anastomosis. The advantage of this technique is the obviation of difficult complete vessel exposure and graft anastomoses, thereby reducing the duration of flow interruption. However, the porcine BCA and LSCA were too short (mean lengths of BCA: 29.0 mm and LSCA: 22.9 mm) for stable landing of the Viabahns in this experiment, as the ‘‘side approach’’ of the technique should sacrifice part of the effective intravascular sealing length. To increase the landing length, we modified the technique to the ‘‘end approach’’ by inserting the Viabahns through the orifices of BCA and LSCA after opening the aortic arch. The tradeoff was consumption of circulatory arrest time. Nevertheless, the procedure was simple and swift and did not require much additional cerebral ischemic time. For the distal arch anastomosis, the aortic wall was sandwiched between the inner stent-graft and the external band to secure the blood sealing zone. It was possible that the vascular supply to the aortic wall could be jeopardized, which could result in aortic wall necrosis. However, the present study revealed no signs of necrosis in distal arch anastomosis, based on both the gross observation and the microscopic examination (Fig. 6EeG). Moreover, the foreign body giant cell reaction frequently observed in the sutured anastomosis (Fig. 6C) was absent in the sutureless anastomosis (Fig. 6G). Generally, the time required for each sutureless anastomosis was less than that for the sutured one. However, the durations of circulatory arrest and aortic cross-clamping in this experiment were only slightly less than those for the existing branched stent-graft inclusion techniques applied in humans (circulatory arrest time: 30 minutes vs. 31 minutes17 or 36 minutes20; aortic-clamping time: 67 minutes vs. 84 minutes17). The nonprominent surgical durations were mainly attributed to the unique porcine anatomy and our limited experiences in porcine aortic arch replacement. The porcine thoracic anatomy is different from the human anatomy in
534 Chen et al.
Fig. 6. Microscopic examinations of the four anastomotic sites 1 week postoperatively. (A) The AAo wall (100). The scale bar in the right lower corner is 1 mm. (B) Tunica media of the AAo (200). The arrow indicates the suture hole with surrounding inflammatory cells. (C) Tunica adventitia of the AAo (400). The arrow depicts the foreign body giant cells, with the magnified images in the inset in the right upper corner. (D) BCA
several aspects. As illustrated in Figure 7, the porcine heart is dextral, with the apex pointing toward the right anterior side. The porcine ascending aorta is short, approximately 2 cm in length, which is located obliquely behind the large right ventricular outflow tract and the pulmonary artery. The porcine aortic arch is deeply seated in left posterior part of the pericardial cavity where the working space is narrow. The porcine axillary arteries are too small to cannulate for antegrade selective cerebral perfusion. These characteristics make porcine aortic arch replacement much more difficult than that for humans. Therefore, most of the circulatory arrest and aortic cross-clamping time was spent in the mobilization of the aortic arch and its branches. Approximately one-fourth of the CPB time was consumed in cooling and rewarming of the systemic perfusate. One concern of the arch graft is that the unsupported portion could buckle, kink, or narrow over time. To prevent the drawback, we placed a fourth stent proximal to the BCA sidearm in addition to the other three distal stents to increase graft support and reduce the risk of graft kink (Fig. 5). The fourth stent was placed as close to the adjacent one as possible to minimize the unsupported portion. We did not place a stent between the sidearms to further increase graft support because the lengthened branched portion of the arch graft could not fit to the small porcine aortic arch anatomy in the narrow
Annals of Vascular Surgery
anastomotic site (100). (E) Distal arch (100). The same scale bar length as panel A in the right lower corner. (F) Tunica media of the distal arch (200). (G) Tunica adventitia of the distal arch (400). Note the scarcity of inflammatory cell infiltrations in comparison with the AAo adventitia (panel C). (H) LSCA anastomotic site (100).
Fig. 7. The anatomy of porcine heart. The porcine heart is dextral, with the apex pointing toward the right anterior side. In this figure, the ascending aorta and aortic arch cannot be seen, as they are located behind the large pulmonary artery and right ventricular outflow tract. The arrow indicates left anterior descending artery, which courses toward the apex (LA, left auricle; LV, left ventricle; RA, right auricle; RV, right ventricle; PA, pulmonary artery; SVC, superior vena cava).
pericardial space. Another issue is the hemodynamic forces over segments of grafts and intervening stents are unknown and could be harmful if there are graft kinks or tears that occur over time. For the sidearm branches, a durable and flexible stent-graft such as Viabahn can reduce the risks
Vol. 26, No. 4, May 2012
of kinks or tears, as described in the VORTEC technique.24 For the main body, future work should be focused on modification of the stent material such as nitinol, adjustment of the stent width and interstent distance, and addition of small intersidearm stents to improve flexibility, conformability, and supporting forces of the stent-graft. In this experiment, an external banding was essential for a secure aortic arch anastomosis. This was achieved by use of a Dacron strap wrapped around the distal arch stump, with its both ends sutured together for appropriate constraint of the external aortic circumference. The procedure took additional 5 minutes in the circulatory arrest period. To simplify the banding technique and shorten the procedural time, one may improve the external band with an adjustable length and autolock mechanism for handy manipulation in future studies. Study Limitations The average life span of Landrace-Yorkshire-Duroc pigs is 10 years. The body weight of a 1-year-old pig can be >100 kg after a normal growth. In practical situations, it is difficult to handle an old pig for an experiment. Therefore, the study was based on the 4-month-old porcine model of normal aorta (30e40 kg in body weight). The aim of the experiment was to test the technical feasibility of the concept of sutureless anastomosis, and therefore, only short-term results were obtained. Experiments on the diseased aortic model with long-term observation are warranted before application in humans.
CONCLUSIONS This study proved the technical feasibility of sutureless anastomoses in porcine extensive aortic arch replacement with a modular branched stent-graft system. The external banding is essential for the secure hemostasis of the distal arch anastomosis, but it is not necessary for the supra-aortic branches.
The authors are grateful to Ms. Pi-Chen Chien for conducting cardiopulmonary bypass. The experimental work was conducted in Chi-Mei Medical Center, Tainan, Taiwan. REFERENCES 1. Chen CC, Hsieh SR. Modified trifurcated graft in acute type A aortic dissection with the least brain ischemic time. Ann Thorac Surg 2007;83:e6e8. 2. Hsieh SR, Chen CC, Wei HJ. A novel arch-first t-graft technique for extensive aortic arch reconstruction. Ann Thorac Surg 2008;85:1814e6.
A modular branched stent-graft system 535
3. Panos A, Murith N, Bednarkiewicz M, Khatchatourov G. Axillary cerebral perfusion for arch surgery in acute type A dissection under moderate hypothermia. Eur J Cardiothorac Surg 2006;29:1036e9. 4. Spielvogel D, Strauch JT, Minanov OP, et al. Aortic arch replacement using a trifurcated graft and selective cerebral antegrade perfusion. Ann Thorac Surg 2002;74:S1810e4. discussion S1825e32. 5. Kato M, Ohnishi K, Kaneko M, et al. New graft-implanting method for thoracic aortic aneurysm or dissection with a stented graft. Circulation 1996;94:II188e93. 6. Kato M, Kuratani T, Kaneko M, et al. The results of total arch graft implantation with open stent-graft placement for type a aortic dissection. J Thorac Cardiovasc Surg 2002;124:531e40. 7. Karck M, Chavan A, Hagl C, et al. The frozen elephant trunk technique: a new treatment for thoracic aortic aneurysms. J Thorac Cardiovasc Surg 2003;125:1550e3. 8. Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491e9. 9. Volodos NL, Karpovich IP, Troyan VI, et al. Clinical experience of the use of self-fixing synthetic prostheses for remote endoprosthetics of the thoracic and the abdominal aorta and iliac arteries through the femoral artery and as intraoperative endoprosthesis for aorta reconstruction. Vasa Suppl 1991;33:93e5. 10. Shibata T, Hirai H, Fukui T, et al. Assembly and deployment of a branched arch stent graft using the transaortic approach. Ann Thorac Surg 2005;79:1790e2. 11. Miyamoto S, Hadama T, Anai H, et al. Stented elephant trunk method for multiple thoracic aneurysms. Ann Thorac Surg 2001;71:705e7. 12. Sun L, Qi R, Chang Q, et al. Surgery for acute type A dissection with the tear in the descending aorta using a stented elephant trunk procedure. Ann Thorac Surg 2009;87:1177e80. 13. Sun LZ, Qi RD, Chang Q, et al. Surgery for acute type A dissection using total arch replacement combined with stented elephant trunk implantation: experience with 107 patients. J Thorac Cardiovasc Surg 2009;138:1358e62. 14. Mestres CA, Fernandez C, Josa M, Mulet J. Hybrid antegrade repair of the arch and descending thoracic aorta with a new integrated stent-dacron graft in acute type A aortic dissection: a look into the future with new devices. Interact Cardiovasc Thorac Surg 2007;6:257e9. 15. Schoenhoff FS, Schmidli J, Eckstein FS, et al. The frozen elephant trunk: an interesting hybrid endovascularsurgical technique to treat complex pathologies of the thoracic aorta. J Vasc Surg 2007;45:597e9. 16. Criado FJ, Barnatan MF, Rizk Y, et al. Technical strategies to expand stent-graft applicability in the aortic arch and proximal descending thoracic aorta. J Endovasc Ther 2002;9(Suppl. 2): II32e8. 17. Chen LW, Dai XF, Lu L, et al. Extensive primary repair of the thoracic aorta in acute type A aortic dissection by means of ascending aorta replacement combined with open placement of triple-branched stent graft: early results. Circulation 2010;122:1373e8. 18. Chen LW, Dai XF, Yang GF, et al. Open-branched stent graft placement makes total arch replacement easier for acute type A aortic dissection. Ann Thorac Surg 2010;89:1688e90. 19. Chen LW, Dai XF, Zhang GC, Lu L. Total aortic arch reconstruction with open placement of triple-branched stent graft for acute type A dissection. J Thorac Cardiovasc Surg 2010;139:1654.e1e5.e1.
536 Chen et al.
20. Shimamura K, Kuratani T, Matsumiya G, et al. Hybrid endovascular aortic arch repair using branched endoprosthesis: the second-generation ‘‘branched’’ open stent-grafting technique. J Thorac Cardiovasc Surg 2009;138:46e52. discussion 52e3. 21. Sun LZ, Qi RD, Chang Q, et al. Is total arch replacement combined with stented elephant trunk implantation justified for patients with chronic stanford type A aortic dissection? J Thorac Cardiovasc Surg 2009;138:892e6. 22. Akin I, Ince H, Kische S, Nienaber CA. Implication of thoracic aortic stent-graft conformability on clinical outcomes. Acta Chir Belg 2009;109:20e6.
Annals of Vascular Surgery
23. Lachat M, Mayer D, Criado FJ, et al. New technique to facilitate renal revascularization with use of telescoping selfexpanding stent grafts: VORTEC. Vascular 2008;16:69e72. 24. Donas KP, Lachat M, Rancic Z, et al. Early and midterm outcome of a novel technique to simplify the hybrid procedures in the treatment of thoracoabdominal and pararenal aortic aneurysms. J Vasc Surg 2009;50:1280e4. 25. Donas KP, Rancic Z, Lachat M, et al. Novel sutureless telescoping anastomosis revascularization technique of supraaortic vessels to simplify combined open endovascular procedures in the treatment of aortic arch pathologies. J Vasc Surg 2010;51:836e41.