- Email: [email protected]
Prosthesis for Aortic Arch Substitution
Prosthesis for Aortic Arch Substitution Stefano Nazari, MD, Susanna Salvi, MD, Alessandro Aluffi, MD, Ettore Visconti, MD, Giuseppe Rescigno, MD, and Paolo Buniva, MD Department of Surgery, University of Pavia, Istituto di Ricovero e Cura a Carattere Scientifico “San Matteo”, and Foundation “Alexis Carrel,” Pavia, Italy
Background. The risk of neurologic complications in aortic arch prosthetic substitution is directly related to the duration of the circulatory arrest. The purpose of this article is to report the experiments on animals of a device for simplifying and quickening the vascular anastomosis in aortic arch substitution. Methods. The device consists of expandable loops of stainless steel wire, sewn to the proximal end of a Dacron prosthesis. An actuating removable guide allows the stainless steel wire loops to be expanded and tightened, in such a way that the prosthesis diameter is varied, while maintaining a regular cylindric shape. The prosthesis end is then transformed into a rigid cylindrical ring, approximately half the maximal diameter in length, with a variable and controllable diameter. A composite graft was prepared, fitted with the expandable device at the distal end of the main prosthesis as well as at each end of the branches for the supraaortic trunks. Cardiopulmonary bypass was established by cannulation of the
right atrium and left iliac artery. The prosthesis was positioned very easily and quickly during a brief hypothermic circulatory arrest; ascending aorta anastomosis was carried out by the standard technique after central nervous system reperfusion was resumed. Acute experiments were carried out in 5 swine. Results. Four of 5 animals survived the procedure without detectable neurologic sequelae. At sacrifice the prosthesis was found to be properly sited without lumen distortion or thrombosis. Conclusions. The main advantages of this device and modality of arch substitution in a clinical setting would include drastic reduction of the circulatory arrest time, easy and reliable hemostasis of the anastomosis line, and accurate and firm approximation of the dissection layers in case of dissecting aneurysms.
T
Material and Methods
horacic aortic substitution still carries the risk of serious complications; in particular, substitution of the aortic arch remains a major challenge, both because of the technical difficulties that may be encountered, especially in emergency operations for dissection, and because of the high risk of serious brain damage due to ischemia or emboli. In spite of the many sophisticated maneuvers attempted for their prevention (eg, deep hypothermic arrest, retrograde cerebral perfusion), the incidence of central nervous system complications is still significant even in very experienced centers [1– 4]. Although the central nervous system complications may have multifactorial causes, there is general agreement that the risk proportionally increases with the increased duration of the circulatory arrest time required for the prosthetic substitution. Because this phase is nearly entirely spent in performing the vascular anastomosis, it seems reasonable to focus the research on devices that might allow the simplification and quickening of this part of the procedure. The purpose of this article is to report the animal experimentation on a device for aortic arch prosthetic substitution.
Accepted for publication May 22, 1997. Presented at the Aortic Surgery Symposium V, New York, NY, April 24 –26, 1996. Address reprint requests to Dr Nazari, Cso Garibaldi 1, 27100 Pavia, Italy.
© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
(Ann Thorac Surg 1997;64:1339 – 44) © 1997 by The Society of Thoracic Surgeons
Device The device [5] consists of loops of stainless steel wire, sewn to the proximal end of a Dacron prosthesis. A removable actuating guide allows the stainless steel wire loops to be expanded and tightened in such a way that the prosthesis diameter varies while maintaining a regular cylindric shape (Fig 1). A standard vascular prosthesis of appropriate diameter and length is sutured to the expandable segment (Fig 2). The prosthesis end then becomes a rigid cylindrical ring, approximately half the maximal diameter in length, with a variable and controllable diameter. The prosthesis end, when reduced in diameter, is easily and quickly positioned into the vascular stump and then expanded to its final appropriate diameter. An external ligature stabilizes the prosthesis and guarantees hemostasis. A composite graft was prepared, fitted with the expandable device at the distal end of the main prosthesis as well as at each end of the branches for the supraortic trunks (Fig 3). For this experimental series a main prosthesis (16 mm) with only two lateral branches (8 mm) was used, due to the particular conformation of the supraaortic trunks (Fig 4A) in the employed animals.
Experimental Model Acute experiments were carried out in 5 swine (20 to 45 kg) in substerile conditions (only the extracorporeal 0003-4975/97/$17.00 PII S0003-4975(97)00991-0
1340
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
Fig 1. The working principle of the device relies on a number of metallic wires, sewn to a Dacron prosthesis, arranged in several circular loops. Traction on the wires by an actuating guide allows the loops to be expanded and tightened in such a way that the diameter of the prosthesis varies, while maintaining a regular cylindric shape. The prosthesis end then becomes a rigid cylindrical ring with a variable and controllable diameter. (Reproduced with permission from Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996;10:1003–9. © 1996 Springer-Verlag GmbH & Co.)
circuit was sterile). Cardiopulmonary bypass was established by cannulation of the right atrium (double-stage cannula) and left iliac artery (8-mm aortic cannula). The distal aorta and each of the supraortic trunks were isolated at the origin and encircled with umbilical tapes. During a brief period of hypothermic (28°C) circulatory arrest the ascending aorta and arch were opened at the anterior surface. The expandable ends of the prosthesis were then positioned into the corresponding aortic sites and opened as widely as possible (see Fig 3), the prepositioned ligatures were tied, and the femoral artery perfusion was resumed. After prosthesis deairing, the proximal end of the main prosthesis was clamped and the anastomosis with the ascending aorta was accomplished with a standard running suture while the animal was rewarmed. The expandable ends’ actuating guides were then simply cut out as closely as possible to the prosthesis. In 2 animals the expandable device was also prepared at the proximal end of the main prosthesis (see Fig 4A), and the ascending aortic anastomosis was also carried out with the device during the circulation arrest phase. At the end of the procedure, the ability of the animal to stand on its legs was considered as a proof of lack of significant neurologic deficits. The animals were then killed and the position and patency of the prosthesis were checked.
Results Four of 5 animals survived the procedure, one death being due to technical misoperation of the extracorporeal circuit. The procedure was very simple and quick. In particular, the insertion of the retracted expandable ends of the prosthesis into each vascular stump was very easy, pro-
Ann Thorac Surg 1997;64:1339 – 44
vided the surgical field was maintained bloodless. The prosthesis (Albograft, Sorin, Saluggia, Italy; and BARD, ACG, Boston, MA) was in the site and the brain perfusion was resumed in less than 10 minutes in all animals. The total cardiopulmonary bypass duration was 30 ⫾ 10 minutes. The procedure was also simplified by the fact that the swine strain provided to our laboratory (large white) has a constant arch conformation with only two supraortic trunks. The 4 surviving animals showed normal leg mobility after extubation, without detectable signs of neurologic deficits. The postmortem examination showed the prosthesis to be correctly sited with all the expandable ends patent (see Fig 4).
Comment The general histologically benign nature of aortic diseases might imply a durable result when prosthetic substitution is successfully achieved; this makes the devastating, if not fatal, neurologic complications of these operations particularly striking and dreadful. Such complications are still largely unpredictable and unavoidable in spite of the sophisticated maneuvers attempted for their prevention.
Fig 2. The device is realized in different sizes and is fitted with a flexible guide with its actuating handle (top left). The prosthesis segment required for substitution is cut to the appropriate length and sutured inside the expandable segment before cross-clamping (top right, bottom left). The metallic wires of the expandable prosthesis end constitute a rigid ring, approximately half its maximal diameter in length. When blood flow is reestablished, the guide is cut as close as possible to the expandable end and removed. The device allows the prosthesis end diameter to be reduced up to 20% to 40% of its maximal diameter (bottom left). The optimal opening of the device when in the site is within 70% and 90% of its maximal aperture. (Reproduced with permission from Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996;10:1003–9. © 1996 Springer-Verlag GmbH & Co.)
Ann Thorac Surg 1997;64:1339 – 44
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
1341
Fig 3. For arch substitution a composite graft can be prepared, fitted with the expandable device at the distal end of the main prosthesis as well as at each end of the branches for the supraortic trunks. The retracted expandable ends are positioned into the corresponding branches and then opened, during a brief hypothermic circulatory arrest; ascending aortic anastomosis can be carried out with the standard technique (or with the quick anastomotic ring shown in Figure 6) during the rewarming phase, while the central nervous system perfusion is resumed.
Ideally, if the cross-clamping time could be kept within the ischemic tolerance of the nervous tissue (5 to 7 minutes in normothermia, 20 to 30 minutes in deep hypothermia [6]), it seems reasonable to expect a virtually complete prevention of neurologic complications, at least if critical tributaries to the nervous tissues are not included in the substituted segment or are revascularized within this time lapse. Because nearly all the crossclamping time is spent for anastomosis accomplishment, it seems reasonable to focus the research on devices for quickening this procedure. In line with this hypothesis, a few years ago intraluminal ringed prostheses were quite extensively used [2, 7, 8] in thoracic aortic substitution, resuming Payr’s very old
concept for vascular anastomosis [9]. These devices, however, did not enter current surgical practice, mainly because although the prosthesis fixation by an external ligature is a very quick and easy task, the positioning of a prosthesis of appropriate diameter into the vascular stump is not. Some factors may explain this (Fig 5): (1) there is in fact a substantial reduction of the vascular lumen when the aorta is clamped, due to the absence of blood pressure distention [9 –11]; (2) the thickness of the aortic wall further reduces the viable prosthesis diameter; (3) due to the floppy consistency of the vascular wall, a significant gap between the internal aortic wall and the external ring diameter must be left to allow the ring to be easily slipped into the vascular stump, without friction;
Fig 4. (A) The device used in this experimental series includes only two supraaortic branches, due to the particular arch conformation of the swine strain used. (B) The anterior appearance of the arch with the device at sacrifice. (C) The prosthesis and arch viewed from the upper side.
1342
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
Fig 5. With the intraluminal ringed prosthesis a diameter rather smaller than that appearing appropriate must be used to keep the clamping time shorter than that achievable with a standard suture. The aorta in fact significantly reduces its diameter when clamped; moreover, due to the floppy consistency of the aortic wall, a significant gap must be left between the external diameter of the intraluminal ring and the internal aortic wall to allow introduction without friction. The expandable prosthesis overcomes all these adverse conditions. (Reproduced with permission from Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996; 10:1003–9. © 1996 Springer-Verlag GmbH & Co.)
and (4) the thickness of the prosthesis ring further reduces the final viable blood flow surface. Thus, if the cross-clamping time has to be kept shorter than that attainable with a handmade suture, a ringed prosthesis with a diameter significantly smaller than that appearing appropriate must be used. The expandable-end prosthesis [5, 11] overcomes all these factors. In fact, the insertion of the rigid end in a closed position is very easy, without friction against the vascular walls; the subsequent expansion allows the final prosthesis diameter to be set at its maximum, it even being possible to dilate the vascular wall to a certain extent. This guarantees the best match between the diameters of the aorta and the prosthesis. Finally, the expandable end wall is much thinner than that of the commercially available intraluminal prostheses, thus only marginally interfering with the final blood flow surface. Previous experiences have shown the reliability of the device in descending aortic anastomosis and highlighted some important details in its use [5]. In this regard, it was shown that the device in the present embodiment must be used in a bloodless field. In fact, when the expandable end is wetted with blood or other liquid, its diameter control becomes more difficult and sometimes impossible. This is due to the higher friction generated by the sliding of the many wire loops during actuation in a solid-liquid interface rather than in a solid-aerial one. On an intuitive basis the underlying mechanisms may be compared with that blocking the common watch functioning after exposure to water. Thus care must be taken
Ann Thorac Surg 1997;64:1339 – 44
to obtain a bloodless field until device opening is completed. This small drawback is compensated by the fact that no further wire fixation is required, after blood flow is reestablished and the guides are cut. This particular feature in fact guarantees that no late shrinking of the prosthesis end diameter is possible as the result of the prolonged action on the wire loops of the external ligature, thus preventing late diameter reduction and prosthesis dislocation. The previous experience also clearly showed that a reliable hemostasis cannot be obtained by simple overdistention of the wired prosthesis end against the inner aortic wall as might be hypothesized. Although this probably could be easily achieved by a more strongly wired device, we think that over-distention of the diseased aortic wall in these patients may be undesirable and dangerous to some extent; thus we think that an external ligature is always necessary both to assure hemostasis and to prevent prosthesis dislocation. Care must also be taken in device size selection. In fact, although the expandable segment lumen remains regularly cylindric within a quite wide range of diameters, the increasing thickness of its wall as the diameter is reduced, due to fabric folding, significantly narrows the effective inner flow surface. The wall thickness/diameter ratio seems to be acceptable for clinical use when the final aperture is set not inferior to 70% of its maximal aperture. In practice, it is assumed that the optimal aperture of each single expandable prosthesis end, defined as 90% of its maximal opening, will fit the inner diameter of the perfused aorta as estimated on angiographic contrast computed tomographic scan. This offers a good wall thickness/lumen diameter ratio, while keeping an extraaperture reserve in case of measure mismatch. Careful preoperative assessment of the required diameters, which with this device can be different at the two ends of the prosthetic segment, is mandatory to avoid both suboptimal hemodynamic results and intraoperative loss of time. In these experiences the external diameters of the devices used ranged from 8 to 16 mm; although the difficulties in manufacturing and operation increase at low diameters, these proved reliable, at least in this acute setting, in terms of thrombotic occlusion risk. Moreover, because no suturing is carried out and only vascular Dacron is in contact with the bloodstream, embolic and thrombotic long-term potential is predictably equal or even inferior to that of a standard vascular prosthesis of the same diameter. Further advantages of the device over the standard technique are the reliability of the anastomosis line hemostasis and approximation of dissected layers, both provided by the external ligature; these may be very difficult to achieve with standard (or even pledgeted) suturing in a dissected and fragile aortic wall. In particular, in dissection the external ligature against the rigid prosthesis end may enhance glue action in fixation of the layers; of course, in case of chronic or acute dissection, when the obliteration of the distal false lumen might not be advisable, the device can be put in place after the
Ann Thorac Surg 1997;64:1339 – 44
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
1343
Fig 6. Other possible uses of the expandable prosthesis may include aneurysms of descending thoracic aorta: (A) A modified version of the device (quick anastomotic ring) in which both ends of a short prosthetic segment are actuated simultaneously by only one guide [5] may be used in the repair of traumatic isthmic rupture (B), or for end-to-end prosthesis or vascular anastomosis (coarctation).
A
B
inner membrane has been appropriately fenestrated, with the same hemodynamic result as the standard procedure. We think that the main field of use of the device is aortic arch substitution, in which the duration of blood flow interruption so closely parallels the onset of serious central nervous system complications. This may shift the outcome of using the device from a purely technical improvement to a significant therapeutic achievement; as a matter of fact, the effects on central nervous system complications of such a brief hypothermic arrest time (⬍10 minutes), in no way achievable with a standard suture even in very experienced hands, can hardly be overlooked. However, this principle of vascular anastomosis allows us to hypothesize other useful applications. Figure 6 schematizes two other quite frequent clinical conditions in which the device could be advantageously used, both of them having being successfully tested in experimental models [5]. A modified device (quick anastomotic ring) was realized under the same working principle but with a different configuration [5] for use in short-segment descending aortic substitution (ie, isthmic traumatic rupture repair or coartation in the adult). In this version the actuating guide enters a short prosthesis segment at the middle and controls expansion of both ends simultaneously. The different configuration of the device also allows end-to-end anastomosis between vascular or prosthetic segments [5]. A quite relevant technical detail implied in the use of this prosthesis is the length of the vascular stump required for expandable end positioning, when applied in clamped vessels. Due to the linearization of the vessel diameter induced by the clamp, the length of the vascular stump distal to it must be significantly longer than that
usually required for standard suture; in practice it is advisable to isolate the vessel for a length exceeding its diameter (Fig 7). Of course, care must be taken when entering the aneurysm not to extend the incision to both the distal and proximal ends to allow the opening of the prosthesis into an intact vascular stump. It may be argued that when the aneurysm reaches the origin of an important collateral branch, eg, left subclavian or renal arteries, without a distinct healthy neck, the device cannot be appropriately positioned. This is not true; the device in fact can be expanded within the
Fig 7. When a clamp is applied, it is important to keep the length of the vascular stump significantly longer than usual to allow full expansion of the expandable end.
1344
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
Ann Thorac Surg 1997;64:1339 – 44
anastomosis line hemostasis as well as stable approximations of dissected layers even in the friable and fragile aortic wall often encountered in these patients. The device could be advantageously used in many other clinical conditions in which aortic prosthetic substitution is required.
References
Fig 8. The expandable end also can be positioned against aneurysmal wall, provided that its distal end would reach the healthy vascular wall. Thrombosis of the tract between the ligature and the prosthesis end will soon move the effective anastomosis line (*) where it would be with standard suture.
aneurysmal wall and fixed there by the external ligature, provided that its end reaches the healthy vascular wall limit, where ideally the standard suture would be placed (Fig 8). In fact the thrombosis soon developing in the tract between the ligature and the prosthesis end will exclude the brief tract of aneurysmal wall from the bloodstream, the final result being definitely identical to that achieved by standard suture. In conclusion, the present experimental series showed that the device allows arch substitution with a very significant decrease of circulatory arrest time, with a consequently proportional expected decrease of the rate of central nervous system ischemic complications. Further advantages of the device include safe and reliable
1. Crawford ES, Rubio PA. Reappraisal of adjuncts to avoid ischemia in the treatment of aneurysms of descending thoracic aorta. J Thorac Cardiovasc Surg 1973;66:693–704. 2. Crawford ES, Crawford JL. Diseases of the aorta. Baltimore: Williams and Wilkins, 1984. 3. Crawford ES, Crawford JL, Safi HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining intermediate and long-term results in 605 patients. J Vasc Surg 1986;3:389 – 404. 4. Gott VL. Heparinized shunts for thoracic vascular operations. Ann Thorac Surg 1972;14:219–20. 5. Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996;10:1003–9. 6. Kirklin JW, Barratt-Boyes BG. Hypothermia, circulatory arrest and cardiopulmonary bypass. In: Kirklin JW, BarrattBoyes BG, eds. Cardiac surgery, 2nd ed. New York: Churchill Livingstone, 1993:chapter 2, p 74. 7. Lemole GM, Strong MD, Spagna PM, Karilomicz MD. Improved results for dissecting aneurysms: intraluminal sutureless prosthesis. J Thorac Cardiovasc Surg 1982;83:249–55. 8. Berger RL, Romero L, Chaudhry AG, Dobnik DB. Graft replacement of the thoracic aorta with a sutureless technique. Ann Thorac Surg 1983;35:231–9. 9. Payr E. Beitrage zur Technik der Blutgefass und Nervennaht nebst Mittheilungen uber die Verwendung eines resorbirbaren Metalles in der Chirurgie. Arch Klin Chir 1900;62: 67–93. 10. Nazari S, Luzzana F, Carli F, et al. Aortic wall structural strengthening by intraluminal net prosthesis to arrest aneurysm progression and to prevent dissection and rupture. Experimental assessment of a new therapeutic approach. Eur J Cardiothorac Surg 1996;10:264–72. 11. Matsumae M, Oz MC, Lemole GM. A flexible sutureless intraluminal graft that becomes rigid after placement in the aorta. J Thorac Cardiovasc Surg 1990;100:787–92.
Background. The risk of neurologic complications in aortic arch prosthetic substitution is directly related to the duration of the circulatory arrest. The purpose of this article is to report the experiments on animals of a device for simplifying and quickening the vascular anastomosis in aortic arch substitution. Methods. The device consists of expandable loops of stainless steel wire, sewn to the proximal end of a Dacron prosthesis. An actuating removable guide allows the stainless steel wire loops to be expanded and tightened, in such a way that the prosthesis diameter is varied, while maintaining a regular cylindric shape. The prosthesis end is then transformed into a rigid cylindrical ring, approximately half the maximal diameter in length, with a variable and controllable diameter. A composite graft was prepared, fitted with the expandable device at the distal end of the main prosthesis as well as at each end of the branches for the supraaortic trunks. Cardiopulmonary bypass was established by cannulation of the
right atrium and left iliac artery. The prosthesis was positioned very easily and quickly during a brief hypothermic circulatory arrest; ascending aorta anastomosis was carried out by the standard technique after central nervous system reperfusion was resumed. Acute experiments were carried out in 5 swine. Results. Four of 5 animals survived the procedure without detectable neurologic sequelae. At sacrifice the prosthesis was found to be properly sited without lumen distortion or thrombosis. Conclusions. The main advantages of this device and modality of arch substitution in a clinical setting would include drastic reduction of the circulatory arrest time, easy and reliable hemostasis of the anastomosis line, and accurate and firm approximation of the dissection layers in case of dissecting aneurysms.
T
Material and Methods
horacic aortic substitution still carries the risk of serious complications; in particular, substitution of the aortic arch remains a major challenge, both because of the technical difficulties that may be encountered, especially in emergency operations for dissection, and because of the high risk of serious brain damage due to ischemia or emboli. In spite of the many sophisticated maneuvers attempted for their prevention (eg, deep hypothermic arrest, retrograde cerebral perfusion), the incidence of central nervous system complications is still significant even in very experienced centers [1– 4]. Although the central nervous system complications may have multifactorial causes, there is general agreement that the risk proportionally increases with the increased duration of the circulatory arrest time required for the prosthetic substitution. Because this phase is nearly entirely spent in performing the vascular anastomosis, it seems reasonable to focus the research on devices that might allow the simplification and quickening of this part of the procedure. The purpose of this article is to report the animal experimentation on a device for aortic arch prosthetic substitution.
Accepted for publication May 22, 1997. Presented at the Aortic Surgery Symposium V, New York, NY, April 24 –26, 1996. Address reprint requests to Dr Nazari, Cso Garibaldi 1, 27100 Pavia, Italy.
© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
(Ann Thorac Surg 1997;64:1339 – 44) © 1997 by The Society of Thoracic Surgeons
Device The device [5] consists of loops of stainless steel wire, sewn to the proximal end of a Dacron prosthesis. A removable actuating guide allows the stainless steel wire loops to be expanded and tightened in such a way that the prosthesis diameter varies while maintaining a regular cylindric shape (Fig 1). A standard vascular prosthesis of appropriate diameter and length is sutured to the expandable segment (Fig 2). The prosthesis end then becomes a rigid cylindrical ring, approximately half the maximal diameter in length, with a variable and controllable diameter. The prosthesis end, when reduced in diameter, is easily and quickly positioned into the vascular stump and then expanded to its final appropriate diameter. An external ligature stabilizes the prosthesis and guarantees hemostasis. A composite graft was prepared, fitted with the expandable device at the distal end of the main prosthesis as well as at each end of the branches for the supraortic trunks (Fig 3). For this experimental series a main prosthesis (16 mm) with only two lateral branches (8 mm) was used, due to the particular conformation of the supraaortic trunks (Fig 4A) in the employed animals.
Experimental Model Acute experiments were carried out in 5 swine (20 to 45 kg) in substerile conditions (only the extracorporeal 0003-4975/97/$17.00 PII S0003-4975(97)00991-0
1340
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
Fig 1. The working principle of the device relies on a number of metallic wires, sewn to a Dacron prosthesis, arranged in several circular loops. Traction on the wires by an actuating guide allows the loops to be expanded and tightened in such a way that the diameter of the prosthesis varies, while maintaining a regular cylindric shape. The prosthesis end then becomes a rigid cylindrical ring with a variable and controllable diameter. (Reproduced with permission from Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996;10:1003–9. © 1996 Springer-Verlag GmbH & Co.)
circuit was sterile). Cardiopulmonary bypass was established by cannulation of the right atrium (double-stage cannula) and left iliac artery (8-mm aortic cannula). The distal aorta and each of the supraortic trunks were isolated at the origin and encircled with umbilical tapes. During a brief period of hypothermic (28°C) circulatory arrest the ascending aorta and arch were opened at the anterior surface. The expandable ends of the prosthesis were then positioned into the corresponding aortic sites and opened as widely as possible (see Fig 3), the prepositioned ligatures were tied, and the femoral artery perfusion was resumed. After prosthesis deairing, the proximal end of the main prosthesis was clamped and the anastomosis with the ascending aorta was accomplished with a standard running suture while the animal was rewarmed. The expandable ends’ actuating guides were then simply cut out as closely as possible to the prosthesis. In 2 animals the expandable device was also prepared at the proximal end of the main prosthesis (see Fig 4A), and the ascending aortic anastomosis was also carried out with the device during the circulation arrest phase. At the end of the procedure, the ability of the animal to stand on its legs was considered as a proof of lack of significant neurologic deficits. The animals were then killed and the position and patency of the prosthesis were checked.
Results Four of 5 animals survived the procedure, one death being due to technical misoperation of the extracorporeal circuit. The procedure was very simple and quick. In particular, the insertion of the retracted expandable ends of the prosthesis into each vascular stump was very easy, pro-
Ann Thorac Surg 1997;64:1339 – 44
vided the surgical field was maintained bloodless. The prosthesis (Albograft, Sorin, Saluggia, Italy; and BARD, ACG, Boston, MA) was in the site and the brain perfusion was resumed in less than 10 minutes in all animals. The total cardiopulmonary bypass duration was 30 ⫾ 10 minutes. The procedure was also simplified by the fact that the swine strain provided to our laboratory (large white) has a constant arch conformation with only two supraortic trunks. The 4 surviving animals showed normal leg mobility after extubation, without detectable signs of neurologic deficits. The postmortem examination showed the prosthesis to be correctly sited with all the expandable ends patent (see Fig 4).
Comment The general histologically benign nature of aortic diseases might imply a durable result when prosthetic substitution is successfully achieved; this makes the devastating, if not fatal, neurologic complications of these operations particularly striking and dreadful. Such complications are still largely unpredictable and unavoidable in spite of the sophisticated maneuvers attempted for their prevention.
Fig 2. The device is realized in different sizes and is fitted with a flexible guide with its actuating handle (top left). The prosthesis segment required for substitution is cut to the appropriate length and sutured inside the expandable segment before cross-clamping (top right, bottom left). The metallic wires of the expandable prosthesis end constitute a rigid ring, approximately half its maximal diameter in length. When blood flow is reestablished, the guide is cut as close as possible to the expandable end and removed. The device allows the prosthesis end diameter to be reduced up to 20% to 40% of its maximal diameter (bottom left). The optimal opening of the device when in the site is within 70% and 90% of its maximal aperture. (Reproduced with permission from Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996;10:1003–9. © 1996 Springer-Verlag GmbH & Co.)
Ann Thorac Surg 1997;64:1339 – 44
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
1341
Fig 3. For arch substitution a composite graft can be prepared, fitted with the expandable device at the distal end of the main prosthesis as well as at each end of the branches for the supraortic trunks. The retracted expandable ends are positioned into the corresponding branches and then opened, during a brief hypothermic circulatory arrest; ascending aortic anastomosis can be carried out with the standard technique (or with the quick anastomotic ring shown in Figure 6) during the rewarming phase, while the central nervous system perfusion is resumed.
Ideally, if the cross-clamping time could be kept within the ischemic tolerance of the nervous tissue (5 to 7 minutes in normothermia, 20 to 30 minutes in deep hypothermia [6]), it seems reasonable to expect a virtually complete prevention of neurologic complications, at least if critical tributaries to the nervous tissues are not included in the substituted segment or are revascularized within this time lapse. Because nearly all the crossclamping time is spent for anastomosis accomplishment, it seems reasonable to focus the research on devices for quickening this procedure. In line with this hypothesis, a few years ago intraluminal ringed prostheses were quite extensively used [2, 7, 8] in thoracic aortic substitution, resuming Payr’s very old
concept for vascular anastomosis [9]. These devices, however, did not enter current surgical practice, mainly because although the prosthesis fixation by an external ligature is a very quick and easy task, the positioning of a prosthesis of appropriate diameter into the vascular stump is not. Some factors may explain this (Fig 5): (1) there is in fact a substantial reduction of the vascular lumen when the aorta is clamped, due to the absence of blood pressure distention [9 –11]; (2) the thickness of the aortic wall further reduces the viable prosthesis diameter; (3) due to the floppy consistency of the vascular wall, a significant gap between the internal aortic wall and the external ring diameter must be left to allow the ring to be easily slipped into the vascular stump, without friction;
Fig 4. (A) The device used in this experimental series includes only two supraaortic branches, due to the particular arch conformation of the swine strain used. (B) The anterior appearance of the arch with the device at sacrifice. (C) The prosthesis and arch viewed from the upper side.
1342
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
Fig 5. With the intraluminal ringed prosthesis a diameter rather smaller than that appearing appropriate must be used to keep the clamping time shorter than that achievable with a standard suture. The aorta in fact significantly reduces its diameter when clamped; moreover, due to the floppy consistency of the aortic wall, a significant gap must be left between the external diameter of the intraluminal ring and the internal aortic wall to allow introduction without friction. The expandable prosthesis overcomes all these adverse conditions. (Reproduced with permission from Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996; 10:1003–9. © 1996 Springer-Verlag GmbH & Co.)
and (4) the thickness of the prosthesis ring further reduces the final viable blood flow surface. Thus, if the cross-clamping time has to be kept shorter than that attainable with a handmade suture, a ringed prosthesis with a diameter significantly smaller than that appearing appropriate must be used. The expandable-end prosthesis [5, 11] overcomes all these factors. In fact, the insertion of the rigid end in a closed position is very easy, without friction against the vascular walls; the subsequent expansion allows the final prosthesis diameter to be set at its maximum, it even being possible to dilate the vascular wall to a certain extent. This guarantees the best match between the diameters of the aorta and the prosthesis. Finally, the expandable end wall is much thinner than that of the commercially available intraluminal prostheses, thus only marginally interfering with the final blood flow surface. Previous experiences have shown the reliability of the device in descending aortic anastomosis and highlighted some important details in its use [5]. In this regard, it was shown that the device in the present embodiment must be used in a bloodless field. In fact, when the expandable end is wetted with blood or other liquid, its diameter control becomes more difficult and sometimes impossible. This is due to the higher friction generated by the sliding of the many wire loops during actuation in a solid-liquid interface rather than in a solid-aerial one. On an intuitive basis the underlying mechanisms may be compared with that blocking the common watch functioning after exposure to water. Thus care must be taken
Ann Thorac Surg 1997;64:1339 – 44
to obtain a bloodless field until device opening is completed. This small drawback is compensated by the fact that no further wire fixation is required, after blood flow is reestablished and the guides are cut. This particular feature in fact guarantees that no late shrinking of the prosthesis end diameter is possible as the result of the prolonged action on the wire loops of the external ligature, thus preventing late diameter reduction and prosthesis dislocation. The previous experience also clearly showed that a reliable hemostasis cannot be obtained by simple overdistention of the wired prosthesis end against the inner aortic wall as might be hypothesized. Although this probably could be easily achieved by a more strongly wired device, we think that over-distention of the diseased aortic wall in these patients may be undesirable and dangerous to some extent; thus we think that an external ligature is always necessary both to assure hemostasis and to prevent prosthesis dislocation. Care must also be taken in device size selection. In fact, although the expandable segment lumen remains regularly cylindric within a quite wide range of diameters, the increasing thickness of its wall as the diameter is reduced, due to fabric folding, significantly narrows the effective inner flow surface. The wall thickness/diameter ratio seems to be acceptable for clinical use when the final aperture is set not inferior to 70% of its maximal aperture. In practice, it is assumed that the optimal aperture of each single expandable prosthesis end, defined as 90% of its maximal opening, will fit the inner diameter of the perfused aorta as estimated on angiographic contrast computed tomographic scan. This offers a good wall thickness/lumen diameter ratio, while keeping an extraaperture reserve in case of measure mismatch. Careful preoperative assessment of the required diameters, which with this device can be different at the two ends of the prosthetic segment, is mandatory to avoid both suboptimal hemodynamic results and intraoperative loss of time. In these experiences the external diameters of the devices used ranged from 8 to 16 mm; although the difficulties in manufacturing and operation increase at low diameters, these proved reliable, at least in this acute setting, in terms of thrombotic occlusion risk. Moreover, because no suturing is carried out and only vascular Dacron is in contact with the bloodstream, embolic and thrombotic long-term potential is predictably equal or even inferior to that of a standard vascular prosthesis of the same diameter. Further advantages of the device over the standard technique are the reliability of the anastomosis line hemostasis and approximation of dissected layers, both provided by the external ligature; these may be very difficult to achieve with standard (or even pledgeted) suturing in a dissected and fragile aortic wall. In particular, in dissection the external ligature against the rigid prosthesis end may enhance glue action in fixation of the layers; of course, in case of chronic or acute dissection, when the obliteration of the distal false lumen might not be advisable, the device can be put in place after the
Ann Thorac Surg 1997;64:1339 – 44
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
1343
Fig 6. Other possible uses of the expandable prosthesis may include aneurysms of descending thoracic aorta: (A) A modified version of the device (quick anastomotic ring) in which both ends of a short prosthetic segment are actuated simultaneously by only one guide [5] may be used in the repair of traumatic isthmic rupture (B), or for end-to-end prosthesis or vascular anastomosis (coarctation).
A
B
inner membrane has been appropriately fenestrated, with the same hemodynamic result as the standard procedure. We think that the main field of use of the device is aortic arch substitution, in which the duration of blood flow interruption so closely parallels the onset of serious central nervous system complications. This may shift the outcome of using the device from a purely technical improvement to a significant therapeutic achievement; as a matter of fact, the effects on central nervous system complications of such a brief hypothermic arrest time (⬍10 minutes), in no way achievable with a standard suture even in very experienced hands, can hardly be overlooked. However, this principle of vascular anastomosis allows us to hypothesize other useful applications. Figure 6 schematizes two other quite frequent clinical conditions in which the device could be advantageously used, both of them having being successfully tested in experimental models [5]. A modified device (quick anastomotic ring) was realized under the same working principle but with a different configuration [5] for use in short-segment descending aortic substitution (ie, isthmic traumatic rupture repair or coartation in the adult). In this version the actuating guide enters a short prosthesis segment at the middle and controls expansion of both ends simultaneously. The different configuration of the device also allows end-to-end anastomosis between vascular or prosthetic segments [5]. A quite relevant technical detail implied in the use of this prosthesis is the length of the vascular stump required for expandable end positioning, when applied in clamped vessels. Due to the linearization of the vessel diameter induced by the clamp, the length of the vascular stump distal to it must be significantly longer than that
usually required for standard suture; in practice it is advisable to isolate the vessel for a length exceeding its diameter (Fig 7). Of course, care must be taken when entering the aneurysm not to extend the incision to both the distal and proximal ends to allow the opening of the prosthesis into an intact vascular stump. It may be argued that when the aneurysm reaches the origin of an important collateral branch, eg, left subclavian or renal arteries, without a distinct healthy neck, the device cannot be appropriately positioned. This is not true; the device in fact can be expanded within the
Fig 7. When a clamp is applied, it is important to keep the length of the vascular stump significantly longer than usual to allow full expansion of the expandable end.
1344
NAZARI ET AL PROSTHESIS FOR AORTIC ARCH
Ann Thorac Surg 1997;64:1339 – 44
anastomosis line hemostasis as well as stable approximations of dissected layers even in the friable and fragile aortic wall often encountered in these patients. The device could be advantageously used in many other clinical conditions in which aortic prosthetic substitution is required.
References
Fig 8. The expandable end also can be positioned against aneurysmal wall, provided that its distal end would reach the healthy vascular wall. Thrombosis of the tract between the ligature and the prosthesis end will soon move the effective anastomosis line (*) where it would be with standard suture.
aneurysmal wall and fixed there by the external ligature, provided that its end reaches the healthy vascular wall limit, where ideally the standard suture would be placed (Fig 8). In fact the thrombosis soon developing in the tract between the ligature and the prosthesis end will exclude the brief tract of aneurysmal wall from the bloodstream, the final result being definitely identical to that achieved by standard suture. In conclusion, the present experimental series showed that the device allows arch substitution with a very significant decrease of circulatory arrest time, with a consequently proportional expected decrease of the rate of central nervous system ischemic complications. Further advantages of the device include safe and reliable
1. Crawford ES, Rubio PA. Reappraisal of adjuncts to avoid ischemia in the treatment of aneurysms of descending thoracic aorta. J Thorac Cardiovasc Surg 1973;66:693–704. 2. Crawford ES, Crawford JL. Diseases of the aorta. Baltimore: Williams and Wilkins, 1984. 3. Crawford ES, Crawford JL, Safi HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining intermediate and long-term results in 605 patients. J Vasc Surg 1986;3:389 – 404. 4. Gott VL. Heparinized shunts for thoracic vascular operations. Ann Thorac Surg 1972;14:219–20. 5. Nazari S, Luzzana F, Banfi C, et al. Expandable prosthesis for sutureless anastomosis in thoracic aorta prosthetic substitution. Eur J Cardiothorac Surg 1996;10:1003–9. 6. Kirklin JW, Barratt-Boyes BG. Hypothermia, circulatory arrest and cardiopulmonary bypass. In: Kirklin JW, BarrattBoyes BG, eds. Cardiac surgery, 2nd ed. New York: Churchill Livingstone, 1993:chapter 2, p 74. 7. Lemole GM, Strong MD, Spagna PM, Karilomicz MD. Improved results for dissecting aneurysms: intraluminal sutureless prosthesis. J Thorac Cardiovasc Surg 1982;83:249–55. 8. Berger RL, Romero L, Chaudhry AG, Dobnik DB. Graft replacement of the thoracic aorta with a sutureless technique. Ann Thorac Surg 1983;35:231–9. 9. Payr E. Beitrage zur Technik der Blutgefass und Nervennaht nebst Mittheilungen uber die Verwendung eines resorbirbaren Metalles in der Chirurgie. Arch Klin Chir 1900;62: 67–93. 10. Nazari S, Luzzana F, Carli F, et al. Aortic wall structural strengthening by intraluminal net prosthesis to arrest aneurysm progression and to prevent dissection and rupture. Experimental assessment of a new therapeutic approach. Eur J Cardiothorac Surg 1996;10:264–72. 11. Matsumae M, Oz MC, Lemole GM. A flexible sutureless intraluminal graft that becomes rigid after placement in the aorta. J Thorac Cardiovasc Surg 1990;100:787–92.