Side-to-side sutureless vascular anastomosis with magnets

BASIC RESEARCH STUDIES

Side-to-side sutureless vascular anastomosis with magnets Detlev Erdmann, MD, PhD,a,c Ranya Sweis, MD,a Christoph Heitmann, MD,a Koji Yasui, MD,a Kevin C. Olbrich, PhD,a L. Scott Levin, MD,a A. Adam Sharkawy, PhD,d and Bruce Klitzman, PhD,a,b Durham, NC; and Fremont, Calif Objective: Abbe and Payr introduced vascular techniques and devices to facilitate vessel anastomosis over a century ago. Obora published the idea of a sutureless vascular anastomosis with use of magnetic rings in 1978. The purpose of this study was to assess the performance of a new magnetic device to perform a side-to-side arteriovenous anastomosis in a dog model. Material and methods: Male fox hounds (25 kg) were treated preoperatively and daily postoperatively with clopidogrel bisulfate (Plavix) and aspirin. The femoral artery and vein were exposed unilaterally in 3 dogs and bilaterally in 4 dogs (n ⴝ 11 anastomoses). A 4-mm arteriotomy was performed, and 1 oval magnet 0.5 mm thick was inserted into the lumen of the artery and a second magnet was applied external to the artery, compressing and stabilizing the arterial wall to create a magnetic port. An identical venous magnetic port was created with another pair of oval magnets. When the 2 ports were allowed to approach each other, they self-aligned and magnetically coupled to complete the arteriovenous anastomosis. Patency was assessed for the first hour with direct observation, again after 9 weeks with duplex ultrasound scanning, and at 10 weeks under direct open observation. The anastomoses were explanted after 10 weeks. Hydrodynamic resistance was measured ex vivo on the final 8 anastomoses by measuring the pressure drop across an anastomosis with a known flow rate. Results: After implantation, very high flow created visible turbulence and palpable vibration. All 11 anastomoses were patent under direct observation and palpation. Ten of 11 anastomoses were clearly patent on duplex scans, and patency of 1 anastomosis was questionable. Hydrodynamic resistance averaged 0.73 ⴞ 0.33 mm Hg min/mL (mean ⴞ SEM). Conclusions: Vascular anastomoses performed with magnets demonstrated feasibility; exhibited 100% patency after 10 weeks in a dog arteriovenous shunt model; lacked apparent aneurysm or other potentially catastrophic failure; demonstrated remodeling of the vessel wall after several weeks to incorporate the magnets, making the magnetic force unnecessary; and warrants further study in vessels with different sizes, flow rates, and locations. (J Vasc Surg 2004;40: 505-11.) Clinical Relevance: We present a magnet-based device used to perform side-to-side peripheral vascular anastomoses. Its advantages include the ability to anastomose vessels without requiring circumferential surgical exposure. Vascular anastomosis performed with these magnets demonstrated 100% patency in the dog, lacked apparent aneurysm or other potentially catastrophic failure, and demonstrated remodeling of the vessel wall after several weeks, to incorporate the magnets, making indefinite retention of field strength unnecessary. This technique could enable minimally invasive procedures, such as complex reconstructive and revascularizing surgery, and warrants further study in vessels with different sizes, flow rates, and locations.

Vascular surgery is closely associated with Alexis Carrel’s innovative visions of vascular anastomoses and organ From Kenan Plastic Surgery Research Laboratories, Division of Plastic, Reconstructive, Maxillofacial and Oral Surgery,a and Department of Biomedical Engineering, Duke University Medical Center,b Durham, Durham Veterans Affairs Medical Center,c Durham, and Ventrica, Inc,d Fremont. This study was funded in part by Ventrica, Inc, Fremont, Calif. Competition of interest: Ventrica, Inc, holds patents on the magnetic anastomotic device technology. Reprint requests: Bruce Klitzman, PhD, Kenan Plastic Surgery Research Laboratories, Duke University Medical Center, Box 3906, Durham, NC 27710 (e-mail: [email protected]). 0741-5214/$30.00 Copyright © 2004 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2004.05.026

transplantation.1 At the same time, the late 19th and early 20th century, various non-suture techniques were experimentally introduced to reestablish arterial flow after complete vessel transection.2-5 However, decades later the hand-sewn vascular anastomosis became routine, and because of its great success, non-suture techniques were subsequently abandoned. During the 1960s, with development and refinement of surgical instruments, suture material, and optical magnification, the vascular microanastomosis became a standard surgical reality.6,7 The sutured vascular anastomosis, in particular the microanastomosis, remains a time-consuming procedure and demands extensive training if high patency rates are to be achieved.8 The search for alternative methods that result in faster and easier vascular anastomoses included refinement both in tech505

506 Erdmann et al

Fig 1. Magnetic vascular positioner magnets on a plastic pedestal.

niques, such as tubes,9-11 cuffs,12 and rings,13-16 and in mechanical devices, such as ring couplers,17-21 staples,22 and clips.23-26 The purpose of this study was to assess a sutureless vascular anastomosis performed using a new magnetic vascular positioning device. Specific aims of this study of the magnetic anastomotic device include demonstration of feasibility, patency rate over a clinically relevant period, durability and necessity of maintaining the magnetic field strength of the device, and extent of tissue incorporation and encapsulation of the device. A side-to-side arteriovenous (AV) shunt was chosen as a model, to assess the feasibility of the device to perform an anastomosis and to evaluate patency over the first 10 weeks. Specific concerns included that the AV pressure gradient would produce forces exceeding the magnetic force, leading to failure of the anastomosis; compression of vascular tissue between the magnets would cause pressure necrosis, with subsequent failure of the vessel wall; direct contact between the blood and the magnet may cause thrombosis; the arterialvenous pressure gradient would cause turbulence, leading to an increased rate of thrombosis; and a foreign body response or endothelial hyperplasia may lead to unacceptable narrowing of the anastomosis over time. MATERIAL AND METHODS The magnets used had an outer diameter major axis of 7.03 mm and a minor axis of 3.05 mm (Fig 1). The inner diameter major axis was 4.85 mm, and the minor axis was 1.94 mm. The thickness was 0.48 mm. Scanning electron microscopy of magnet. Preliminary studies showed the feasibility of performing scanning electron microscopy on magnets (Fig 2). Gross distortion of the image was observed at the poles of the magnet (heel

JOURNAL OF VASCULAR SURGERY September 2004

and toe), particularly obvious by the distortion of the square grid on which the magnet rested. However, little distortion was apparent along the lateral sides of the magnet (Fig 2). Surgical procedure. The Institutional Animal Care and Use Committee of Duke University Medical Center approved the study before its initiation. Seven male fox hounds (approximately 25 kg) received 225 mg of clopidogrel bisulfate (Plavix) and 325 mg of aspirin at least 2 days preoperatively, then 75 mg of clopidogrel bisulfate and 325 mg of aspirin until the day of surgery. On the day of surgery the dogs received no antiplatelet medications until they had recovered postoperatively. For the surgery, in each animal an antecubital vein was cannulated and sodium pentothal injected. The dogs were subsequently intubated and anesthetized with 1.5% to 2.5% isoflurane in oxygen. With the animal in a supine position the medial thigh and groin were shaved and prepared with povidone iodine (right side only, n ⫽ 3; both sides, n ⫽ 4), and draped with sterile towels. Lactated Ringer solution (400500 mL) was administered via intravenous drip. The common femoral artery (2-3 mm inner diameter) and vein (3-4 mm inner diameter) were exposed unilaterally or bilaterally through a 10-cm longitudinal incision, and isolated under loupe magnification. The femoral artery and vein were both occluded temporarily at the proximal and distal ends with atraumatic vascular clamps or vessel loops. One arteriotomy and one venotomy (each 4 mm long) were performed at the middle part of each vessel. One MVP magnet (Ventrica, Inc) was inserted inside the artery, and a second magnet was applied external to the artery but precisely aligned with the intraluminal magnet (Fig 3, A-C). The 2 magnets attracted, compressed the vessel wall to hold it in place, and created a magnetic arterial port. An identical pair of magnets was used to create a similar magnetic port in the vein at a position adjacent to the arterial port. Subsequently selfalignment of the 2 magnetic ports was allowed, creating a magnet-based side-to-side AV shunt consisting of a stack of 4 magnets (Fig 3, D-E). Finally the arterial and venous clamps were released. The wounds were closed with 3-0 absorbable suture for the muscle and subcutaneous layers and 3-0 nylon suture for the cutaneous layer. Postoperatively the animals were allowed unrestricted activity in individual runs. The dogs received subcutaneous buprenorphine as needed for pain relief, and oral clopidogrel bisulfate (75 mg) and aspirin (325 mg) daily. The patency of each anastomosis was assessed clinically with palpation and duplex ultrasound scanning (Philips 5500; 7-MHz linear probe) after 9 weeks. Waveform recordings were made of each anastomosis. In most animals sedation with 1 mL of medetomidine hydrochloride administered subcutaneously was required during sonography, although 2 animals were calm enough to undergo imaging with no sedation. Explantation. After 10 weeks, animals were again anesthetized with isoflurane. All anastomoses were surgically exposed and evaluated for patency by means of direct

JOURNAL OF VASCULAR SURGERY Volume 40, Number 3

Erdmann et al 507

Fig 2. Scanning electron micrographs of magnet. A, Wide field micrograph (20⫻). B, Medium magnification (80⫻) of area indicated by small box in A. C, High magnification (640⫻) of area indicated by small box in B. Note gross distortion of image at the poles (heel and toe) of the magnets, but little apparent distortion along the sides.

visualization and a milking test after occluding all nonanastomosis inflow. Subsequently all anastomoses were gently flushed with Ringer solution and harvested. Short-term implantation. After explantation of the 3 unilateral AV shunts a second identical shunt was placed in the contralateral leg. After less than 5 minutes of perfusion, the anastomosis was explanted with the identical protocol as the 10-week anastomoses. Resistance measurement. Resistance measurements of the final 8 anastomoses were obtained with a Flex roller pump (Masterflex; Cole-Parmer Instrument Co) to pump saline solution at 90 mL/min through polyethylene tubing (PE240; inner diameter, 1.67 mm). A pressure transducer (Fiber Optic Sensor Technologies, Inc) was connected via stopcock to the inflow tubing, and measured the hydrodynamic pressure before vessel cannulation. The explanted femoral artery was then cannulated proximally with the PE240 tubing, and a large venotomy was created to enable the saline solution to exit. The exiting stream did not rise (“spurt”) more than about 0.5 cm above the vein, indicating insignificant kinetic energy in the saline solution as it exited. The hydrodynamic resistance of the anastomosis was calculated by dividing the measured inflow pressure (corrected for the pressure in the tube alone) by the flow rate. Macroscopic photographs were taken of each anastomosis after resistance measurements were made. Low magnification macrographs were taken with a macro lens. Higher magnification photos were taken through an operating microscope to visualize the lumen of the anastomosis. Histologic analysis. After the photography, all specimens were immersed in 10% buffered formalin for a mini-

mum of 6 hours. After fixation, 8 anastomoses were microdissected, and the magnets were removed from the vessel wall. Vessels were then embedded in paraffin, and 5-␮m thick cross-sections were cut through the anastomosis, and both proximal and distal to the anastomosis. Sections were stained with hematoxylin-eosin and examined under a bright-field microscope. The total wall thickness was measured at the 0-degree, 90-degree, 180-degree, and 270degree positions around the circumference of each artery and vein. Inasmuch as no differences were observed with position around the vessel circumference, wall thicknesses with position were pooled. Scanning electron microscopy of explanted anastomoses. Three specimens were fixed overnight in 1% glutaraldehyde rather than formalin, dehydrated in graded ethanol, and critical point dried in liquid carbon dioxide. They were then sputter-coated with gold-palladium, and photographed on a scanning electron microscope (model 200; JEOL). Data analysis. All data are expressed as mean ⫾ SE. Differences between groups were compared with the Student t test, with P ⱕ .05 required for statistical significance. RESULTS All 7 dogs survived placement of the magnets. Owing to aggressive pretreatment with anti-platelet medication, hemostasis was difficult. Deployment of the magnets was uneventful. When the arterial clamps were released the vein distended considerably, approximately 2 to 3 times normal diameter, and turbulence could clearly be observed through the venous wall. Turbulence could also be easily palpated over the anastomosis after closure. Many animals


508 Erdmann et al

Fig 3. Schema of the side-to-side anastomosing of the artery and vein. A, Insertion of the intravascular magnet. The extravascular magnet is held at some distance from the intravascular magnet to prevent premature attraction. B, Alignment of the intravascular and extravascular magnets. C, Release of the extravascular magnet onto the intravascular magnet. D, Alignment of the venous pair of magnets above the arterial pair of magnets. E, Self-alignment and attraction of the 4 magnets, completing the anastomosis.

were ambulatory after recovery on the day of surgery. All animals appeared to have normal gait after the first 1 or 2 days of recovery and throughout the remaining 10 weeks of the study. Ten of 11 anastomoses were clearly patent by palpation and duplex ultrasound scanning 9 weeks postoperatively, and the patency of 1 anastomosis was questionable. After 10 weeks all anastomoses were patent under direct observation and milking test on the day of sacrifice. At explantation, tissue fibrosis made surgery more difficult than the initial exposure. The magnets were not visible from the abluminal side, because of fibrosis and incorporation into the remodeled vessel wall, but were clearly visible when observed from the lumen. Scanning electron microscopy revealed the characteristic cobblestone appearance of endothelial nuclei on the luminal surface (Fig 4). The entire intravascular surface of the magnets appeared to be covered with a monolayer of endothelial cells. The brightfield macro photographs also indicated good endothelialization, as indicated by a glistening appearance (Fig 5, B-D). There was no obvious inflammatory reaction in the tissue adjacent to the anastomosis,

and there was no evidence of proximal or distal aneurysm formation, either at or adjacent to the anastomoses. The lumen of the anastomoses appeared to narrow along the major axes of the oval magnets, but very little along the minor axes (Fig 4, C and D). All but 1 anastomosis had limited ingrowth of tissue. The 1 anastomosis in which patency could not be definitively determined at palpation and duplex ultrasound scanning had a significantly narrowed lumen at macroscopic examination (Fig 4, D). Hydrodynamic resistance was not significantly different between the short-term and 10-week anastomoses. Individual anastomosis resistance averaged 0.73 ⫾ 0.33 mm Hg min/mL (mean ⫾ SEM). However, 1 anastomosis had 6-fold higher resistance (2.73 ⫾ 0.37 mm Hg min/mL) than the average of the other anastomoses (0.44 ⫾ 0.11 mm Hg min/mL). Histologic examination of the anastomoses revealed a continuity of the endothelium from the arterial vessel wall through the anastomotic port to the venous vessel wall. The magnet width represented only about one fifth of the vessel wall thickness, and the magnets were fully integrated into the vessel wall. There was no sign of compression necrosis at the site of the original vessel walls, although the tissue being compressed between the magnets was acellular (no apparent nuclei) at histologic examination. There was no evidence of thickening of the vascular tissue on the luminal side of the magnets, which would indicate myointimal hyperplasia, although the vessel wall on the abluminal side of the magnets thickened considerably. There was no evidence of a profound foreign body reaction, such as presence of giant cells, or of ongoing inflammatory response, such as with neutrophils or monocytes. The 10-week arterial wall thicknesses were 0.28 mm (proximal) and 0.22 mm (distal). The short-term proximal and distal arterial wall thicknesses were 0.29 and 0.29 mm, respectively (Fig 6). The venous wall adjacent to 10-week anastomoses (0.12 mm, both proximally and distally) was significantly thicker than the wall adjacent to acute anastomoses (0.02 mm, both proximally and distally). The proximal and distal vein had the largest thickening over the 10 weeks of implantation; the difference in thickness between the artery and vein seen over the short term became less pronounced over time. DISCUSSION This study demonstrates that vascular anastomoses performed with magnets are feasible, have good patency after 10 weeks, show no signs of aneurysm formation, and remodel the vessel wall to incorporate the magnets. With this device there is no need for broad surgical exposure of the vessels, freeing of the surrounding tissues, or eversion of the vessel walls. Once the arterial and venous ports are firmly in place, artery and vein are brought into close proximity, and the side-to-side vascular anastomosis is completed through attraction and self-alignment of the magnets. Follow-up of approximately 3 months did not reveal anastomotic stenoses, except in 1 anastomosis, or aneurysms. Although there was some reduction in the long


Erdmann et al 509

Fig 4. Scanning electron micrographs of luminal surface of explanted artery. Note cobblestone appearance of the surface, indicating endothelial cell coverage.

axis of the oval lumen of all anastomoses, the short axis seemed unaffected. As expected, this slight long axis narrowing did not increase the measured hydrodynamic resistance significantly between the short-term and 10-week explants. Microscopically, all intraluminal magnets were endothelialized, and only a small inflammatory or foreign body reaction was observed at histologic analysis. In 1893 Abbe2 questioned whether it is “impossible to restore an arterial blood supply once cut off.” Approaching the problem, he subsequently performed several animal studies using a glass tube to reestablish arterial flow. At the same time, similar non-suture techniques were introduced, primarily in the German medical literature, to restore arterial flow after complete vessel transection.3-5 A detailed description of the sutured vascular anastomosis is associated with Alexis Carrel in the early 20th century.27,28 However, it took several decades before his visions became reality. Because of the clinical success of the sutured vascular anastomosis, publications addressing non-suture methods declined for decades, until the end of World War II.10,29-31 With the development and refinement of surgical instruments, suture material, and the combination with optical magnification, the microvascular anastomosis became a standard clinical reality by the end of the 1960s.6,7 Even today the standard among all surgical specialties for creation of a vascular anastomosis is manual suturing. However, a sutured vascular anastomosis, particularly microanastomosis, remains a time-consuming procedure and demands extensive training if high patency rates are to be achieved.8 Therefore there still is demand for a faster, easier, minimally traumatic, and reliable procedure to create a vascular anastomosis. This demand is heightened by

the trends in various surgical subspecialties toward minimally invasive procedures.32-35 Among the various mechanical devices introduced to date,9-25 only a single publication describes an anastomotic device based on magnetic rings as a “preliminary report.” Obora et al36 successfully created end-to-end anastomoses in dogs and rats by passing the ends of a sectioned blood vessel through openings in ring-shaped magnets that later self-aligned. Although the idea was striking, the concept of sliding vessels through openings in magnets was not embraced. This technique needed broad surgical exposure, which may have decreased the enthusiasm for this method, although a currently used mechanical ring device requires similar passing of vessel ends through a mechanical anastomotic device. In addition, the techniques of tissue bonding with glue37-39 or with laser welding have been described.40-43 Biocompatible glues are often used in medicine to achieve hemostasis, close fistulas, immobilize foreign bodies such as hip prostheses, deliver drugs, and close cutaneous ulcers. Glue is able to seal small holes in vessels, but has not been effective enough to support a vascular anastomosis.38 Laser welding is commonly used in a hybrid technique in which only a few stitches are used and the laser completes the anastomosis. The technique requires less time than conventional suture techniques, and patency rates are reported to be equal.44 However, the reported incidence of pseudoaneurysm formation and rupture at the site of the anastomosis has been high in experimental studies, and has limited its clinical application.45 The surgical community has not embraced any of these methods as being significantly superior to manual suturing. Ring couplers have been accepted in the fields of plastic

510 Erdmann et al


Fig 5. A, Wide field photo of the short-term completed arteriovenous anastomosis, with magnets (arrow) visible between the vein (above) and the artery (below). B to D, Macro photographs of the 10-week explanted anastomosis, with magnets clearly visible from the luminal side. B, Low magnification view from the arterial lumen. C, View of a typical anastomosis from the luminal side. D, View of the anastomosis with the highest measured hydrodynamic resistance from the luminal side. Note the smaller lumen compared with that in C.

Fig 6. Comparison of vessel wall thickness of long-term (solid bars) and short-term (open bars) grafts at both the proximal and distal positions. *P ⬍ .05, long-term vs short-term grafts.

surgery and orthopedic surgery,46 but their clinical application has been limited because of the requirement of appropriate vessel size and pliability. In addition, a relatively long free segment of the vessel is necessary for eversion, and the ring coupler is most suitable for an end-to-end anastomosis. It is unclear as to the cause of the gross observation of narrowing of the lumen along the long axis of the oval magnets. Histologic examination did not give a definitive answer as to the cause of the narrowing. There did not appear to be aggressive myointimal hyperplasia in any anastomosis. Certainly the stiff magnets did not match the compliance of the adjacent native vessel, which may lead to vessel narrowing. The simplicity of the magnet technique and the high patency rate encourages further studies in vessels of different sizes, locations, and blood flow. A major advantage of


this device may eventually be its use in minimally invasive surgery, such as beating-heart coronary bypass surgery, peripheral vascular surgery, trauma surgery, and reconstructive plastic surgery, as well as approximation of nonvascular tissues. We thank Vasey McClory, Christiane Poellmann, Natalie Wisniewski, and Andrea E. Buckwalter for technical assistance; and Susan Rimmer and Shelly Walker, sonographers, Department of Medicine, Duke University Medical Center, for assistance with duplex ultrasound scanning.

Erdmann et al 511

23.

24.

25. 26.

27.

REFERENCES 1. Langer RM, Kahan BD. Alexis Carell’s legacy: visionary of vascular surgery and organ transplantation. Transplant Proc 2002;34:1061-6. 2. Abbe R. The surgery of the hand. NY Med J 1894;59:33-40. 3. Oueirolo GB. On the function of the liver as protection against intestinal intoxication [German]. Untersuch Naturlehre 1895;15:228. 4. Nitze M. Vascular anastomoses [German]. Centralbl Chir 1897;24: 1042. 5. Payr E. Contribution on the technique of blood vessel and nerve anastomoses with notes on the application of a resorbable metal in surgery [German]. Arch Klin Chir 1900;62:67-93. 6. Smith JW. Microsurgery: review of the literature and discussion of microtechniques. Plast Reconstr Surg 1966;37:227-45. 7. Cobbett JR. Microvascular surgery. Surg Clin North Am 1967;47:52142. 8. Werker PM, Kon M. Review of facilitated approaches to vascular anastomosis surgery. Ann Throrac Surg 1997;63:122-7. 9. Swenson O, Gross RE. Absorbable fibrin tubes for vein anstomoses. Surgery 1947;22:137-43. 10. Weiss EW, Lam CR. Tantalum tubes in the non-suture method of blood vessel anastomosis. Am J Surg 1950;80:452-4. 11. Egdahl RH, Hume DM. Nonsuture blood-vessel anastomosis: an experimental study using polyethylene as the prosthetic material. Arch Surg 1956;72:232-40. 12. Tschopp HM. Small artery anastomosis, using a cuff of dura mater and a tissue adhesive. Plast Reconstr Surg 1975;55:606-11. 13. Urschel HC, Roth EJ. Small arterial anastomosis. I: Nonsuture. Ann Surg 1961;153:599-610. 14. Goetz RH, Rohman M, Haller JD, Dee R, Rosenak SS. Internal mammary-coronary artery anastomosis: a nonsuture method employing tantalum rings. J Thor Cardiovasc Surg 1961;41:378-86. 15. Haller JD, Kripke DC, Rosenak SS, Dee R, Roberts T, Rohman M, et al. Long-term results of small vessel anastomoses with a ring technic. Ann Surg 1965;161:67-72. 16. Zhou G, Yi-chun L, Qun Q. Experience using microvascular anastomotic rings for the transplantation of 29 free flaps. Plast Reconstr Surg 1989;84:822-6. 17. Nakayama K, Tamiya T, Yamamoto K, Akimoto S. A simple new apparatus for small vessel anastomosis (free autograft of the sigmoid included). Surgery 1962;52:918-31. ¨ strup LT, Berggren A. The Unilink instrument system for fast and safe 18. O microvascular anastomosis. Ann Plast Surg 1986;17:521-5. ¨ strup LT, Lidman D. Mechanical anastomosis of small 19. Berggren A, O arteries and veins with the Unilink apparatus: a histologic and scanning electron microscopic study. Plast Reconstr Surg 1987;80:247-83. 20. Daniel RK, Olding M. An absorbable anastomotic device for microvascular surgery: experimental studies. Plast Reconstr Surg 1984;74:32936. 21. Daniel RK, Olding M. An absorbable anastomotic device for microvascular surgery: clinical applications. Plast Reconstr Surg 1984;74:33742. 22. Heijmen RH, Hinchliffe P, Borst C, Verlaan CW, Moues CM, van der Helm YJ, et al. A novel one-shot anastomotic stapler prototype for

28. 29. 30. 31. 32.

33.

34.

35.

36. 37. 38. 39.

40.

41.

42.

43.

44.

45.

46.

coronary bypass grafting on the beating heart: feasibility in the pig. J Thorac Cardiovasc Surg 1999;117:117-25. Kirsch WM, Zhu YH, Hardesty RA, Chapolini R. A new method for microvascular anastomosis: report of experimental and clinical research. Am Surg 1992;58:722-7. Nataf P, Hincliffe P, Manzo S, Simpson J, Kirsch WM, Zhu YH, et al. Facilitated vascular anastomoses: the one-shot device. Ann Thor Surg 1998;66:1041-4. Euler E, Wilker D, Eitel F, Mager S. Nonsuture microsurgical vessel anastomosis. J Reconstr Microsurg 1987;4:49-52. Zeebregts CJ, van den Dungen JJ, Kalicharan D, Cromheecke M, van der Want J, van Schilfgaarde R. Nonpenetrating vascular clips for small-caliber anastomosis. Microsurgery 2000;20:131-8. Carrel A. Surgical technique for vascular anastomoses and organ transplants [French]. Lyon Med 1902;98:859-63. Carrel A. The surgery of blood vessels. Johns Hopkins Hosp Bull 1907;18:18. Blakemore AH, Lord JW, Stefko PL. The severed primary artery in the war wounded. Surgery 1942;12:488-508. Blakemore AH, Lord JW. A non-suture method of blood vessel anastomosis. JAMA 1945;127:685-91. Johns TNP. A comparison of suture and nonsuture methods for the anastomosis of veins. Surg Gynecol Obstet 1947;84:939-42. Solem JO, Boumzebra D, Al-Buraiki J, Nakeeb S, Rafeh W, Al-Halees Z. Evaluation of a new device for quick sutureless coronary artery anastomosis in surviving sheep. Eur J Cardio Thorac 2000;17:312-8. Zegdi R, Lajos P, Ponzio O, Bruneval P, Fabiani JN. End-to-side anastomosis using a new sutureless vascular prothesis. ASAIO J 2001; 47:329-32. De Lorenzi F, Van der Hulst RWJ, Boeckx WD. VCS auto suture stapled microvascular anastomoses in lower leg free flaps. Plast Reconstr Surg 2002;109:2023-30. Li Y, Wood MB. End-to-side anastomosis in the dog using the 3M precise microvascular anastomotic system: a comparative study. J Reconstr Microsurg 1991;7:345-50. Obora Y, Tamaki N, Matsumoto S. Nonsuture microvascular anastomosis using magnet rings. Surg Neurol 1978;9:117-20. Hosbein DJ, Blumenstock DA. Anastomosis of small arteries using a tissue adhesive. Surg Gynecol Obstet 1964;118:112-4. Lerner R, Binur NS. Current status of surgical adhesives. J Surg Res 1990;48:165-8. Dumanian GA, Dascombe W, Hong C, Labadie K, Garrett K, Sawhney AS, et al. A new photopolymerizable blood vessel glue that seals human vessel anastomoses without augmenting thrombogenicity. Plast Reconstr Surg 1995;95:901-7. Travers V, Kanaujia RR, Ochi M, Ikuta Y. Microvascular anastomosis using carbon dioxide (CO2) laser: evaluation by SEM after corrosion cast. J Reconstr Microsurg 1987;4:61-5. Okada M, Shimizu K, Ikuta H, Horii H, Nakamura K. An alternative method of vascular anastomosis by laser: experimental and clinical study. Lasers Surg Med 1987;7:240-8. Reali UM, Gelli R, Giannotti, Gori F, Pratesi R, Pini R. Experimental diode laser-assisted microvascular anastomosis. J Reconstr Microsurg 1993;9:203-10. Bu¨ rger RA, Gerharz C-D, Draws J, Engelmann UH, Hohenfelder R. Sutureless laser-welded anastomosis of the femoral atery and vein in rats using CO2 and Nd:Yag lasers. J Reconstr Microsurg 1993;9:213-8. McCarthy WJ, LoCicero J, Hartz RS, Yao JS. Patency of laser-assisted anastomoses in small vessels: one year follow up. Surgery 1987;102: 319-26. Quigley MR, Bailes JE, Kwaan HC, Cerullo LJ, Brown JT. Aneurysm formation after low power carbon dioxide laser-assisted vascular anastomosis. Neurosurgery 1986;18:292-9. Ahn CR, Shaw WW, Berns S, Markowitz BL. Clinical experience with the 3M microvascular coupling anastomotic device in 100 free-tissue transfers. Plast Reconstr Surg 1994;93:1481-4.

Submitted Dec 10, 2003; accepted May 24, 2004. Available online Jul 22, 2004.