Atraumatic Vascular Anastomoses Using a Tourniquet Paul E. Collier, MD, Sewickley, Pennsylvania
Because of the occurrence of stenoses just beyond the distal anastomosis, lower leg exsanguination and the use of a mid-thigh tourniquet was adapted. From July 1, 1988 until June 30, 1990, 75 primary operations and 18 secondary operations on failing bypasses were performed using this technique. The primary operations were to the below-knee popliteai (4), anterior tibial (19), posterior tibial (16), peroneal (26), and inframaileolar arteries (10). The proximal anastomoses were also done while the tourniquet was inflated in 11 cases (three tibio-tibial and eight from the popliteal below the knee). Of the secondary procedures, 15 were patch angioplasties and three were distal extensions. The cuff was inflated to 400 mmHg from 22 to 73 minutes. This technique offers numerous advantages: only enough arterial surface is dissected for the anastomoses to be performed. Veins and arterial branches are left intact, thus decreasing the risk of bleeding. The artery is left in its bed to preserve its vasa vasorum. The lumen is not obliterated by clamps assuring the best anastomosis. Since there are no clamps in the field, suturing is less difficult. The tourniquet facilitates a faster, drier operation. No stenoses have been noted in the arteries just distal to the anastomoses on follow-up showing the value of atraumatic occlusion, especially on small vessels. No inflow stenoses occurred in the area under the tourniquet if the artery was patent at the time of operation. (Ann Vasc Surg 1992; 6:34-37). KEY WORDS:
Anastomoses; peripheral vascular disease; tourniquet.
shown that to be far from true [1,2]. Ultrasonographic and election microscopic examinations have demonstrated that all clamps injure arteries [3,4]. These injuries have ranged from endothelial imprinting to endothelial and medial injury. These injured areas attract platelets and fibrin and may cause immediate graft thrombosis or late stenoses in the inflow or outflow tract. The loss of the endothelium at the site of clamp injury was shown to persist at least 13 months in one study [3]. Silastic vessel loops have been shown to be the least injurious form of vessel occlusion. Studies have shown either no damage or only minimal intimal injury [2]. Loops have not been studied on either normal or diseased tibial vessels. Whittemore described localized arterial stenoses either just proximal or just distal to femoropopliteal
One of the keys to successful vascular operations is meticulous, atraumatic technique. This is especially true at the infrageniculate level where arteries are much smaller and frequently not normal. Many surgeons have advocated the use of microsurgical techniques. Vascular surgeons have always been trained in gentle, precise dissection with minimal tissue trauma. However, clamp technology has lagged behind. Although many manufacturers claim their vascular clamps are atraumatic, careful studies have From the Department of Surgery, Sewicktey Valley Hospital, Sewickley, Pennsylvania. Presented at the 16th Annual Meeting of the Peripheral Vascular Surgep:v Society, June 2, 1991, Boston, Massachusetts. Reprint requests: Paul E. Collier, MD, Suite 200, 301 Ohio River Boulevard, Sewickley, Pennsylvania 15143.
34
VOLUME 6 N o 1 - 1992
A TRA UMA TIC V A S C U L A R A N A S T O M O S E S USING A T O U R N I Q U E T
anastomoses and suggested "traumatic injury induced by a vascular clamp during initial surgery as the probable cause" [5]. Because of the occurrence of these arterial clamp injuries, Bernhard, Boren and Towne described the use of the pneumatic tourniquet as a substitute for vascular clamps [6]. They acknowledged that this had been previously described for repairing popliteal aneurysms [7]. Bernhard and associates however, did not provide any follow-up beyond the immediate postoperative period, nor have they subsequently reported it [6]. This technique seems to have been lost, and is never mentioned in reviews on techniques of arterial occlusion. In 1988 I switched from the use of Silastic vessel loops to the use of the pneumatic tourniquet when performing infrageniculate anastomoses. Intensive postoperative surveillance on all grafts was carried out with color Doppler imaging to ascertain whether there were adverse effects from the use of the pneumatic tourniquet, especially stenoses just beyond the anastomoses. This technique of atraumatic vascular occlusion and follow-up are the basis of this report.
MATERIAL AND METHODS From July 1, 1988 through June 30, 1990, 93 operations were performed using the pneumatic tourniquet. The saphenous vein was harvested in the usual fashion. All veins were either used in the reversed or the transposed, nonreversed fashion. The veins were placed in a chilled saline solution that contained 30 mg of papaverine and 5,000 units of heparin. The arteries were exposed utilizing standard incisions. Only enough of the anterior surface of the artery is exposed to perform the proposed anastomosis, leaving the artery in its bed with its branches and accompanying veins undisturbed. At this point, if the vein is to be used in the non-reversed fashion, the valves are cut using a Mills retrograde valvulotome. The patient is then given 5,000 units of heparin intravenously. At this point a laparotomy sponge moistened with antibiotic solution is placed in the vein harvest incision and several layers of cotton cast padding are wrapped around the mid- to distal thigh. A sterile, disposable tourniquet is wrapped snugly around this area (34 inch "Banana Cuff"*). The leg is then elevated and an Esmarch bandage is wrapped around the leg from the foot to the tourniquet to exsanguinate the extremity. The tourniquet is then inflated to 400 mmHg. After the Esmarch bandage is removed, a 1 cm arteriotomy is performed in the exposed surface of the artery. The artery is not instrumented in any way. The vein is tailored to the *Aspen Labs, Englewood, California.
35
arteriotomy. The anastomosis is begun at the heel. The suture is usually run in a continuous fashion around five-eighths of the circumference of the arteriotomy. This allows careful placement of the sutures at the toe of the anastomosis. Then the other side of the suture is used to close the remaining three-eighths of the anastomosis. In very small or difficult arteries three or five mattress sutures are placed at the toe of the anastomosis to ensure that there is no narrowing. These are not tied down until the two running sutures are completed. In both instances the anastomoses are totally completed. If it is possible to perform the proximal anastomosis below the tourniquet, this is done. After the tourniquet is deflated, heparin solution is slowly infused into the graft to displace any air. If the proximal anastomosis is performed with the tourniquet up, heparinized saline is infused just prior to completing that anastomosis. After the graft is filled with heparinized saline and the air is displaced, the sutures are tied. The operation is then completed in the standard fashion. At present there do not appear to be any major limitations for application of this technique to anastomoses below the knee. Although not specifically used for in-situ bypasses in this series, the tourniquet can be easily applied to this technique. The in-situ operation is performed as usual since the distal anastomoses are done last. Two alterations in technique must be made. First~ to avoid any twisting in the graft a marker, such as longitudinal adventitial stitch, must be placed in the vein. This is essential because visualization and control of the distal vein are lost when the Esmarch wrap is applied. Second, just prior to completing the anastomosis, the tourniquet is released. This allows back-bleeding to clear the distal circulation and allows one to assess the pulsatility of the inflow. Over this two year period, 75 primary operations were performed for limb salvage using this technique. Four anastomoses were to the popliteal artery below the knee, 10 were to inframalleolar arteries, and 61 were to the tibial or peroneal artery. The proximal anastomosis was also performed using the tourniquet in 11 cases for a total of 86 anastomoses (Table I). During the same time period 18 secondary operations were performed on failing grafts. The technique used was similar to that previously described in that just enough of the anterior surface of the graft or artery was exposed to perform the procedure. There were 15 patch angioplasties and three distal extensions performed. Vigilant graft surveillance was performed using the Quantum Angiodynographt. The original scan
~Quantum Medical Systems, Issaquah, Washington.
A T R A U M A T I C VASCULAR ANASTOMOSES USING A TOURNIQUET
36
TABLE I.
Insertion
Number
Popliteal--below-knee Anterior tibial Posterior tibial Peroneal Inframalleolar
4 19 16 26 10
Total
75
Origin
Number
Tibial Popliteal--betow-knee Total
3 8 11
was performed usually within two weeks of operation. Scans were then performed every three months for the first 18 months and every six months thereafter. Two parameters were felt to be indicative of a failing graft. If the peak systolic flow velocity was less than 45 cm/sec as described by Bandyk [8], an arteriogram was obtained. Also, if there was a region in the graft where the peak systolic velocity accelerated three or more fold from the preceding segment, investigation was indicated. Arteriography confirmed correctable lesions in all failing grafts that were identified by these parameters. This was especially true when the peak systolic velocity fell below 45 cm/sec beyond an area identified with velocity acceleration.
ANNALS OF VASCULAR SURGERY
grafts occluded between the third and twelfth months. A cause for failure was not found in two because no further intervention was undertaken since the patient's minor amputations had healed. In the third patient urokinase infusion successfully reopened a graft that had occluded because of a stenotic valve. This was successfully repaired with a vein patch angioplasty. With close surveillance 13 graft-threatening stenoses (17.3% of grafts) were identified from three to 18 months postoperatively (Table iI). The incidence and timing of the stenoses are very similar to those reported by Bandyk [8]. The four anastomotic stenoses were caused by intimal hyperplasia. The nine stenoses identified within the body of the grafts were all venous valve stenoses. All of these stenoses were confirmed arteriographically. All of these 13 grafts were repaired with either vein patch angioplasties or short extension grafts. Of note is the fact that no stenoses occurred beyond the anastomoses at the usual site of clamp application. Again this lack of any stenoses beyond the distal anastomosis is identical to the results reported by Bandyk [8] and reflects the use of the tourniquet by his group. It is in contrast to the low, but real, incidence of clamp-induced stenoses reported by Whittemore [5] and also by Veith [9] (5%-15% of stenoses identified). No inflow stenoses or occlusions occurred under the area of tourniquet application if the artery was patent at the time of original operation.
DISCUSSION RESULTS This technique was successfully utilized to perform 104 anastomoses in 93 operations. All operations were performed for limb salvage (rest pain, ulceration or gangrene) or for salvage of failing distal bypasses. The average tourniquet time was 33 minutes (range 22-73 minutes). There was no tissue necrosis, hematoma, ecchymosis or nerve injury related to the use of the tourniquet. The tourniquet made the anastomoses easier to perform for three reasons. First, there were no bothersome clamps in the field that hindered the suturing. Second, since there was no clamp compressing the artery, a clearer view of the lumen was possible, allowing more precise placement of sutures. Subjectively, there was much less blood loss because, as in aneurysm surgery, a more limited dissection reduced the incidence of vein injury. The field was absolutely dry in all but one anastomosis. In this case a minute ooze was easily handled with periodic suction. No specific measurements were made of blood loss to specifically address this. There were four graft occlusions. One graft occluded in the first month due to poor outflow. Three
Tourniquets have been used by surgeons since the 16th century when Ambroise Par6 described their applicability when performing amputations. The term tourniquet was introduced in 1718 by Pelet when he described his innovation of a screw tourniquet which was felt to be an improvement over the simple strap and buckle tourniquet then in use. Lister in 1864 demonstrated the efficacy of elevation of the limb before application of the tourniquet, while Esmarch introduced an India rubber bandage for limb exsanguination in 1873. In 1904 Harvey Cushing introduced the pneumatic tourniquet with a pressure manometer [7]. Flatt in 1972 demonstrated that two hours of tourniquet time was safe in hand surgery [10], while in 1973 it was demonstrated that 500 mmHg pressure was safe in lower extremities [11]. In spite of all this work TABLE II.
Stenosis Proximal anastomosis Proximal vein Mid-vein Distal vein Distal anastomosis Total
Number 2 2 4 3 2 13
VOLUME 6 No 1 - 1992
A T R A U M A T I C VASCULAR A N A S T O M O S E S USING A TOURNIQUET
with tourniquets in different fields of surgery, however, there are only two papers describing the use of tourniquets in the field of vascular surgery [6,7]. Arterial damage by occluding clamps has been well described in the vascular surgical literature. In 1976 Slayback and colleagues showed that some degree of intimal injury is a virtual constant at the site of clamp application. They stressed that this was especially true in atherosclerotic arteries [3]. In t977 DePalma and coworkers showed that endothelial loss at the site of clamp application could persist for 13 months and served as a nidus for platelet adherence [ 12]. In 1981, Whittemore and associates brought attention to the fact that clamp injury could serve as a cause of graft failure [5]. Since then, numerous studies have examined and compared numerous vascular clamps and vessel loops for the degree of damage they cause to arteries. Whereas Silastic vessel loops have generally been shown to be the least traumatic form of occlusion, they can still cause intimal injury if pulled up too tightly [2]. All vascular clamps cause injury ranging from intireal imprinting through medial and intimal damage. With this wealth of knowledge it is surprising that the use of a pneumatic tourniquet for vascular anastomoses has been lost for the last decade. This study again demonstrates that the use of a tourniquet is very safe when performing anastomoses to arteries below the knee. Application to the mid thigh at a pressure of 400 mmHg totally occludes flow even when vessels are calcified and avoids any potential injury to the peroneal nerve. The use of the tourniquet facilitates the operation and minimizes dissection. Most importantly, this study demonstrates the total absence of lesions normally attributable to arterial clamping. The use of the tourniquet also is felt to be superior to other alternatives such as intraluminal occlusion devices or balloons for three reasons. First, some form of arterial control must be obtained in order to prevent blood loss while placing the occluding device. This is unnecessary with the tourniquet. Second, theoretically, the pressure applied to the intima from within could form a nidus for deposition of platelets or proliferation of smooth muscle cells, possibly causing future problems with stenosis or occlusion. Finally, the intraluminal devices do impair the surgeon's view of the anastomosis somewhat and tend to get in the way while
mmm
37
suturing. These problems do not occur with the use of the tourniquet. CONCLUSIONS In conclusion, the use of the pneumatic tourniquet while performing anastomoses to arteries below the knee is highly recommended. The tourniquet facilitates the operation and reduces blood loss. The tourniquet is not difficult to apply and reduces operating time, although it does add a small cost to the procedure. There are no recognized limitations to the use of the tourniquet for anastomoses performed below the knee. Finally, the use of the tourniquet eliminates the risk of arterial clamp injury and subsequent graft failure from its consequences. REFERENCES I. BUNT TJ, MANSH1P L, MOORE W. latrogenic vascular
injury during peripheral revascularizatiom J Vasc Sttr~ 1985:2:491-498. 2. MOORE W. MANSHIP L. BUNT TJ. Differential endothelial injury caused by vascular clamps and vessel loops. Am Sur~ 1985;5/:392-406. 3. SI,AYBACK JB. BOWEN WW, HENSHAW DB. Intimal injury from arterial clamps. Am J Surg 1976;132:183-188. 4. COELKO JCU, SIGEL B, FLANIGAN DP, et al. Arteriographic and ultrasonic evaluation of vascular clamp injuries using an in vitro human experimental model. Surg Gynecol Obstet 1982:155:506-512. 5. WHITTEMORE AD, CLOWES AW, COUCH NP, et al. Secondary femoropopliteal reconstruction. Ann Surg 1981: /93:35-42. 6, BERNHARD VM, BOREN CH, TOWNE JB. Pneumatic tourniquet as a substitute for vascular clamps in distal bypass surgery. Surgeo" 1980:87:709-713. 7. SCHEININ TM, FINDFORS O. Simplified repair of popliteal aneurysms. J Cardiovasc Surg 1979:20: 189-192. 8. BANDYK DF. Postoperative surveillance of femorodistal grafts: the application of echo-Doppler (duplex) ultrasonic scanning, In BERGAN J, YAO JS (eds). Reoperative Arterial Surgeo'. Orlando, FL: Grune and Stratton, 1986. 9, VEITH FJ, WEISER RK, GUPTA SK, et al. Diagnosis and management of failing lower extremity arterial reconstructions. J Cardiovasc Surg 1984:25:381-384. 10. FLATT AE. Tourniquet time in hand surgery. Arch Surg 1972;104:190-192. I I. The tourniquet; instrument or weapon? (Editorial) Can Med Assoc" J 1973;109:827. 12. DE PALMA RG, CHIDI CC, STEINFELD WV, et al. Pathogenesis and prevention of trauma-provoked atheromas. Surgery 1977:82:429-437.
Because of the occurrence of stenoses just beyond the distal anastomosis, lower leg exsanguination and the use of a mid-thigh tourniquet was adapted. From July 1, 1988 until June 30, 1990, 75 primary operations and 18 secondary operations on failing bypasses were performed using this technique. The primary operations were to the below-knee popliteai (4), anterior tibial (19), posterior tibial (16), peroneal (26), and inframaileolar arteries (10). The proximal anastomoses were also done while the tourniquet was inflated in 11 cases (three tibio-tibial and eight from the popliteal below the knee). Of the secondary procedures, 15 were patch angioplasties and three were distal extensions. The cuff was inflated to 400 mmHg from 22 to 73 minutes. This technique offers numerous advantages: only enough arterial surface is dissected for the anastomoses to be performed. Veins and arterial branches are left intact, thus decreasing the risk of bleeding. The artery is left in its bed to preserve its vasa vasorum. The lumen is not obliterated by clamps assuring the best anastomosis. Since there are no clamps in the field, suturing is less difficult. The tourniquet facilitates a faster, drier operation. No stenoses have been noted in the arteries just distal to the anastomoses on follow-up showing the value of atraumatic occlusion, especially on small vessels. No inflow stenoses occurred in the area under the tourniquet if the artery was patent at the time of operation. (Ann Vasc Surg 1992; 6:34-37). KEY WORDS:
Anastomoses; peripheral vascular disease; tourniquet.
shown that to be far from true [1,2]. Ultrasonographic and election microscopic examinations have demonstrated that all clamps injure arteries [3,4]. These injuries have ranged from endothelial imprinting to endothelial and medial injury. These injured areas attract platelets and fibrin and may cause immediate graft thrombosis or late stenoses in the inflow or outflow tract. The loss of the endothelium at the site of clamp injury was shown to persist at least 13 months in one study [3]. Silastic vessel loops have been shown to be the least injurious form of vessel occlusion. Studies have shown either no damage or only minimal intimal injury [2]. Loops have not been studied on either normal or diseased tibial vessels. Whittemore described localized arterial stenoses either just proximal or just distal to femoropopliteal
One of the keys to successful vascular operations is meticulous, atraumatic technique. This is especially true at the infrageniculate level where arteries are much smaller and frequently not normal. Many surgeons have advocated the use of microsurgical techniques. Vascular surgeons have always been trained in gentle, precise dissection with minimal tissue trauma. However, clamp technology has lagged behind. Although many manufacturers claim their vascular clamps are atraumatic, careful studies have From the Department of Surgery, Sewicktey Valley Hospital, Sewickley, Pennsylvania. Presented at the 16th Annual Meeting of the Peripheral Vascular Surgep:v Society, June 2, 1991, Boston, Massachusetts. Reprint requests: Paul E. Collier, MD, Suite 200, 301 Ohio River Boulevard, Sewickley, Pennsylvania 15143.
34
VOLUME 6 N o 1 - 1992
A TRA UMA TIC V A S C U L A R A N A S T O M O S E S USING A T O U R N I Q U E T
anastomoses and suggested "traumatic injury induced by a vascular clamp during initial surgery as the probable cause" [5]. Because of the occurrence of these arterial clamp injuries, Bernhard, Boren and Towne described the use of the pneumatic tourniquet as a substitute for vascular clamps [6]. They acknowledged that this had been previously described for repairing popliteal aneurysms [7]. Bernhard and associates however, did not provide any follow-up beyond the immediate postoperative period, nor have they subsequently reported it [6]. This technique seems to have been lost, and is never mentioned in reviews on techniques of arterial occlusion. In 1988 I switched from the use of Silastic vessel loops to the use of the pneumatic tourniquet when performing infrageniculate anastomoses. Intensive postoperative surveillance on all grafts was carried out with color Doppler imaging to ascertain whether there were adverse effects from the use of the pneumatic tourniquet, especially stenoses just beyond the anastomoses. This technique of atraumatic vascular occlusion and follow-up are the basis of this report.
MATERIAL AND METHODS From July 1, 1988 through June 30, 1990, 93 operations were performed using the pneumatic tourniquet. The saphenous vein was harvested in the usual fashion. All veins were either used in the reversed or the transposed, nonreversed fashion. The veins were placed in a chilled saline solution that contained 30 mg of papaverine and 5,000 units of heparin. The arteries were exposed utilizing standard incisions. Only enough of the anterior surface of the artery is exposed to perform the proposed anastomosis, leaving the artery in its bed with its branches and accompanying veins undisturbed. At this point, if the vein is to be used in the non-reversed fashion, the valves are cut using a Mills retrograde valvulotome. The patient is then given 5,000 units of heparin intravenously. At this point a laparotomy sponge moistened with antibiotic solution is placed in the vein harvest incision and several layers of cotton cast padding are wrapped around the mid- to distal thigh. A sterile, disposable tourniquet is wrapped snugly around this area (34 inch "Banana Cuff"*). The leg is then elevated and an Esmarch bandage is wrapped around the leg from the foot to the tourniquet to exsanguinate the extremity. The tourniquet is then inflated to 400 mmHg. After the Esmarch bandage is removed, a 1 cm arteriotomy is performed in the exposed surface of the artery. The artery is not instrumented in any way. The vein is tailored to the *Aspen Labs, Englewood, California.
35
arteriotomy. The anastomosis is begun at the heel. The suture is usually run in a continuous fashion around five-eighths of the circumference of the arteriotomy. This allows careful placement of the sutures at the toe of the anastomosis. Then the other side of the suture is used to close the remaining three-eighths of the anastomosis. In very small or difficult arteries three or five mattress sutures are placed at the toe of the anastomosis to ensure that there is no narrowing. These are not tied down until the two running sutures are completed. In both instances the anastomoses are totally completed. If it is possible to perform the proximal anastomosis below the tourniquet, this is done. After the tourniquet is deflated, heparin solution is slowly infused into the graft to displace any air. If the proximal anastomosis is performed with the tourniquet up, heparinized saline is infused just prior to completing that anastomosis. After the graft is filled with heparinized saline and the air is displaced, the sutures are tied. The operation is then completed in the standard fashion. At present there do not appear to be any major limitations for application of this technique to anastomoses below the knee. Although not specifically used for in-situ bypasses in this series, the tourniquet can be easily applied to this technique. The in-situ operation is performed as usual since the distal anastomoses are done last. Two alterations in technique must be made. First~ to avoid any twisting in the graft a marker, such as longitudinal adventitial stitch, must be placed in the vein. This is essential because visualization and control of the distal vein are lost when the Esmarch wrap is applied. Second, just prior to completing the anastomosis, the tourniquet is released. This allows back-bleeding to clear the distal circulation and allows one to assess the pulsatility of the inflow. Over this two year period, 75 primary operations were performed for limb salvage using this technique. Four anastomoses were to the popliteal artery below the knee, 10 were to inframalleolar arteries, and 61 were to the tibial or peroneal artery. The proximal anastomosis was also performed using the tourniquet in 11 cases for a total of 86 anastomoses (Table I). During the same time period 18 secondary operations were performed on failing grafts. The technique used was similar to that previously described in that just enough of the anterior surface of the graft or artery was exposed to perform the procedure. There were 15 patch angioplasties and three distal extensions performed. Vigilant graft surveillance was performed using the Quantum Angiodynographt. The original scan
~Quantum Medical Systems, Issaquah, Washington.
A T R A U M A T I C VASCULAR ANASTOMOSES USING A TOURNIQUET
36
TABLE I.
Insertion
Number
Popliteal--below-knee Anterior tibial Posterior tibial Peroneal Inframalleolar
4 19 16 26 10
Total
75
Origin
Number
Tibial Popliteal--betow-knee Total
3 8 11
was performed usually within two weeks of operation. Scans were then performed every three months for the first 18 months and every six months thereafter. Two parameters were felt to be indicative of a failing graft. If the peak systolic flow velocity was less than 45 cm/sec as described by Bandyk [8], an arteriogram was obtained. Also, if there was a region in the graft where the peak systolic velocity accelerated three or more fold from the preceding segment, investigation was indicated. Arteriography confirmed correctable lesions in all failing grafts that were identified by these parameters. This was especially true when the peak systolic velocity fell below 45 cm/sec beyond an area identified with velocity acceleration.
ANNALS OF VASCULAR SURGERY
grafts occluded between the third and twelfth months. A cause for failure was not found in two because no further intervention was undertaken since the patient's minor amputations had healed. In the third patient urokinase infusion successfully reopened a graft that had occluded because of a stenotic valve. This was successfully repaired with a vein patch angioplasty. With close surveillance 13 graft-threatening stenoses (17.3% of grafts) were identified from three to 18 months postoperatively (Table iI). The incidence and timing of the stenoses are very similar to those reported by Bandyk [8]. The four anastomotic stenoses were caused by intimal hyperplasia. The nine stenoses identified within the body of the grafts were all venous valve stenoses. All of these stenoses were confirmed arteriographically. All of these 13 grafts were repaired with either vein patch angioplasties or short extension grafts. Of note is the fact that no stenoses occurred beyond the anastomoses at the usual site of clamp application. Again this lack of any stenoses beyond the distal anastomosis is identical to the results reported by Bandyk [8] and reflects the use of the tourniquet by his group. It is in contrast to the low, but real, incidence of clamp-induced stenoses reported by Whittemore [5] and also by Veith [9] (5%-15% of stenoses identified). No inflow stenoses or occlusions occurred under the area of tourniquet application if the artery was patent at the time of original operation.
DISCUSSION RESULTS This technique was successfully utilized to perform 104 anastomoses in 93 operations. All operations were performed for limb salvage (rest pain, ulceration or gangrene) or for salvage of failing distal bypasses. The average tourniquet time was 33 minutes (range 22-73 minutes). There was no tissue necrosis, hematoma, ecchymosis or nerve injury related to the use of the tourniquet. The tourniquet made the anastomoses easier to perform for three reasons. First, there were no bothersome clamps in the field that hindered the suturing. Second, since there was no clamp compressing the artery, a clearer view of the lumen was possible, allowing more precise placement of sutures. Subjectively, there was much less blood loss because, as in aneurysm surgery, a more limited dissection reduced the incidence of vein injury. The field was absolutely dry in all but one anastomosis. In this case a minute ooze was easily handled with periodic suction. No specific measurements were made of blood loss to specifically address this. There were four graft occlusions. One graft occluded in the first month due to poor outflow. Three
Tourniquets have been used by surgeons since the 16th century when Ambroise Par6 described their applicability when performing amputations. The term tourniquet was introduced in 1718 by Pelet when he described his innovation of a screw tourniquet which was felt to be an improvement over the simple strap and buckle tourniquet then in use. Lister in 1864 demonstrated the efficacy of elevation of the limb before application of the tourniquet, while Esmarch introduced an India rubber bandage for limb exsanguination in 1873. In 1904 Harvey Cushing introduced the pneumatic tourniquet with a pressure manometer [7]. Flatt in 1972 demonstrated that two hours of tourniquet time was safe in hand surgery [10], while in 1973 it was demonstrated that 500 mmHg pressure was safe in lower extremities [11]. In spite of all this work TABLE II.
Stenosis Proximal anastomosis Proximal vein Mid-vein Distal vein Distal anastomosis Total
Number 2 2 4 3 2 13
VOLUME 6 No 1 - 1992
A T R A U M A T I C VASCULAR A N A S T O M O S E S USING A TOURNIQUET
with tourniquets in different fields of surgery, however, there are only two papers describing the use of tourniquets in the field of vascular surgery [6,7]. Arterial damage by occluding clamps has been well described in the vascular surgical literature. In 1976 Slayback and colleagues showed that some degree of intimal injury is a virtual constant at the site of clamp application. They stressed that this was especially true in atherosclerotic arteries [3]. In t977 DePalma and coworkers showed that endothelial loss at the site of clamp application could persist for 13 months and served as a nidus for platelet adherence [ 12]. In 1981, Whittemore and associates brought attention to the fact that clamp injury could serve as a cause of graft failure [5]. Since then, numerous studies have examined and compared numerous vascular clamps and vessel loops for the degree of damage they cause to arteries. Whereas Silastic vessel loops have generally been shown to be the least traumatic form of occlusion, they can still cause intimal injury if pulled up too tightly [2]. All vascular clamps cause injury ranging from intireal imprinting through medial and intimal damage. With this wealth of knowledge it is surprising that the use of a pneumatic tourniquet for vascular anastomoses has been lost for the last decade. This study again demonstrates that the use of a tourniquet is very safe when performing anastomoses to arteries below the knee. Application to the mid thigh at a pressure of 400 mmHg totally occludes flow even when vessels are calcified and avoids any potential injury to the peroneal nerve. The use of the tourniquet facilitates the operation and minimizes dissection. Most importantly, this study demonstrates the total absence of lesions normally attributable to arterial clamping. The use of the tourniquet also is felt to be superior to other alternatives such as intraluminal occlusion devices or balloons for three reasons. First, some form of arterial control must be obtained in order to prevent blood loss while placing the occluding device. This is unnecessary with the tourniquet. Second, theoretically, the pressure applied to the intima from within could form a nidus for deposition of platelets or proliferation of smooth muscle cells, possibly causing future problems with stenosis or occlusion. Finally, the intraluminal devices do impair the surgeon's view of the anastomosis somewhat and tend to get in the way while
mmm
37
suturing. These problems do not occur with the use of the tourniquet. CONCLUSIONS In conclusion, the use of the pneumatic tourniquet while performing anastomoses to arteries below the knee is highly recommended. The tourniquet facilitates the operation and reduces blood loss. The tourniquet is not difficult to apply and reduces operating time, although it does add a small cost to the procedure. There are no recognized limitations to the use of the tourniquet for anastomoses performed below the knee. Finally, the use of the tourniquet eliminates the risk of arterial clamp injury and subsequent graft failure from its consequences. REFERENCES I. BUNT TJ, MANSH1P L, MOORE W. latrogenic vascular
injury during peripheral revascularizatiom J Vasc Sttr~ 1985:2:491-498. 2. MOORE W. MANSHIP L. BUNT TJ. Differential endothelial injury caused by vascular clamps and vessel loops. Am Sur~ 1985;5/:392-406. 3. SI,AYBACK JB. BOWEN WW, HENSHAW DB. Intimal injury from arterial clamps. Am J Surg 1976;132:183-188. 4. COELKO JCU, SIGEL B, FLANIGAN DP, et al. Arteriographic and ultrasonic evaluation of vascular clamp injuries using an in vitro human experimental model. Surg Gynecol Obstet 1982:155:506-512. 5. WHITTEMORE AD, CLOWES AW, COUCH NP, et al. Secondary femoropopliteal reconstruction. Ann Surg 1981: /93:35-42. 6, BERNHARD VM, BOREN CH, TOWNE JB. Pneumatic tourniquet as a substitute for vascular clamps in distal bypass surgery. Surgeo" 1980:87:709-713. 7. SCHEININ TM, FINDFORS O. Simplified repair of popliteal aneurysms. J Cardiovasc Surg 1979:20: 189-192. 8. BANDYK DF. Postoperative surveillance of femorodistal grafts: the application of echo-Doppler (duplex) ultrasonic scanning, In BERGAN J, YAO JS (eds). Reoperative Arterial Surgeo'. Orlando, FL: Grune and Stratton, 1986. 9, VEITH FJ, WEISER RK, GUPTA SK, et al. Diagnosis and management of failing lower extremity arterial reconstructions. J Cardiovasc Surg 1984:25:381-384. 10. FLATT AE. Tourniquet time in hand surgery. Arch Surg 1972;104:190-192. I I. The tourniquet; instrument or weapon? (Editorial) Can Med Assoc" J 1973;109:827. 12. DE PALMA RG, CHIDI CC, STEINFELD WV, et al. Pathogenesis and prevention of trauma-provoked atheromas. Surgery 1977:82:429-437.
