The superiority of synthetic arterial grafts over autologous veins in carotid-subclavian bypass

Transcription

1 The superiority of synthetic arterial grafts over autologous veins in carotid-subclavian bypass Stanley Ziomek, M.D., William J. Quifiones-Baldrich, M.D., Ronald W. Busuttil, M.D., Ph.D., J. Dennis Baker, M.D., Herbert I. Machleder, M.D., and Wesley S. Moore, M.D., Los Angeles, Calif. From May 1964 to June 1983, 36 carotid-subclavian bypasses were done in 36 patients who had symptomatic lesions at the origin of the common carotid and/or subclavian arteries at the Center for Health Sciences of the University of California, Los Angeles. Ages ranged from 28 to 82 years (mean, 58 years). Eighteen bypasses were done with prosthetic grafts, 13 done with autogenous vein, and five were transpositions with primary anastomosis of the subclavian and carotid arteries. Follow-up was available on all patients and ranged from 9 to 156 months (mean, 51.5 months). The graft patency rate at 5 years determined by actuarial methods and documented by clinical examination, noninvasive evaluation, and/or arteriography was 94.1% for prosthetic grafts and 58.3% for vein grafts (p <.1). The 5-year cerebrovasoalar accident (CVA) rate for patients with carotid-subclavian bypass done with prosthetic grafts was 6% in contrast to 39% for those with vein grafts (p <.545). All reconstructions done by transposition and primary anastomosis remain patent and there have been no late CVAs. We conclude that prosthetic grafts are the arterial substitute of choice in carotid-subclavian bypass. Transposition and primary anastomosis between the carotid and subclavian artery, when technically feasible, may be preferable to the use of free grafts in carotid-subclavian reconstruction. (J VASe SURG 1986; 3:14-5.) Extrathoracic revascularization has become the standard treatment for patients with symptomatic branch occlusions of the aortic arch. Carotid-subclavian bypass has been used successfully to revascularize the common carotid or subclavian artery. 1'2,a'4 Excellent long-term results have been reported when this bypass has been used to treat patients with subclavian steal, ipsilateral arm symptoms, or symptoms referable to a stenosis or occlusion of the common carotid origin.l,s The technique of the operation has been well described. 6 The options include a bypass graft between the carotid and subclavian arteries with either autologous vein or prosthetic material or transposition with primary anastomosis between the origin of the occluded artery and the inflow artery, which creates an autologous reconstruction without the use of a free graft. To our knowledge no report has been From the Section of Vascular Surgery, Department of Surgery, University of California, Los Angeles Medical Center. Reprint requests: Wesley S. Moore, M.D., Section of Vascular Surgery, Department of Surgery/General, CHS 72-16, School of Medicine, University of California at Los Angeles, 1833 LeConte Ave., Los Angeles, CA published that specifically compares the results of these alternatives. We describe our accumulated experience with 36 carotid-subclavian bypasses in which, on the basis of the surgeon's preference, all three choices were used. The early and long-term results of bypasses done wi( autologous vein, prosthetic graft, and primary anastomosis in the carotid-subclavian position c~,nstitute the focus of this report. PATIENTS AND METHODS From May 1964 to June 1983, 36 patients with symptomatic occlusive disease of one or more branches of the aortic arch underwent extrathoracic carotid-subclavian revascularization. Nineteen patients were men and 17 were women; their ages ranged from 28 to 82 years (mean, 58 years). The hospital records of all 36 patients were reviewed retrospectively to determine the presenting clinical manifestations, the extent and location of the lesions on arch angiography, details of the operation, and immediate postoperative results. Long-term results were obtained from records of the follow-up visits and from examination of patients in the clinic. Fol-

2 Volume 3 Number 1 January 1986 Deleterious effects of vein grafts in carotid-subclavian bypasses 141 low-up ranged from 9 to 156 months with a mean interval of 51.5 months (4.3 years). Causes of the lesions were atherosclerosis in 35 patients and penetrating trauma resulting in thrombosis in one patient (Table I). In 12 (33%) patients more than one vessel was involved whereas only a singl e lesion was found in 24 (66%) patients. Complete occlusion of the affected artery was seen in 2 of 36 patients (55.5%). The innominate artery was involved in five patients (13.8%), the left subclavian artery in 22 patients (61.1%), right subclavian artery in four patients (11.1%), and the left common carotid artery at the origin in five patients (13.8%). Six patients (16.6%) had associated significant stenosis of the internal carotid artery and three patients [8.3%) had an associated total occlusion. All 36 patients had symptoms at the time of operatic~n. Seventeen patients (47.2%) experienced vertebrobasllar symptoms such as dizziness, drop attacks, or vertigo. Nine patients (25%) had transient ischemic attacks referable to the carotid circulation such as amaurosis fugax, contralateral paresthesia, and/or weakness, and 1 patients (27.7%) experienced both anterior and posterior circulation symptoms. Five patients had suffered a cerebrovascular accident (CVA) with complete recovery. No patients had previous extra-anatomic bypasses. One patient had undergone a previous contralateral carotid endarterectomy. Of 36 carotid-subclavian reconstructions 13 patients (36.1%) had autogenous vein grafts (1 saphenous, two cephalic, and one external jugular vein). Eighteen patients (5%) had revascularization with prosthetic grafts (12 double-velour Dacron wafts, five polytetrafluoroethylene (PTFE), and one bovine graft), and five patients (13.8%) underwent prima~end-to-side carotid-subclavian artery anastomoses. Five patients with an innominate artery lesion underwent a left subclavian-to-right carotid bypass (one vein, four prosthetic grafts). Seventeen of 22 patients with left subclavian artery disease underwent left carotid-subclavian bypass (seven vein, 1 prosthetic grafts), and five patients underwent left carotid-subclavian transposition with primary anastomosis. All four patients with right subclavian artery disease underwent right carotid-subclavian bypasses (two vein, two Dacron grafts), and five patients with common Carotid occlusions underwent subclaviancarotid bypass (three vein, two Dacron grafts). Ipsilateral internal carotid endarterectomy was done concurrently in six patients (16.6%). Because the patients who received various grafts Table I. Distribution of lesions Left SCA occlusion 11 Right SCA occlusion 3 Left SCA occlusion plus left ICA 5 stenosis Right SCA occlusion plus right ICA 1 stenosis Left SCA stenosis 5 Left SCA stenosis plus VA occlusion 1 IA stenosis 3 IA stenosis plus left CCA occlusion 2 Left CCA stenosis 1 Left CCA occlusion 1 Left CCA stenosis plus left ICA oc- 1 clusion Left CCA plus right ICA occlusions 1 Left CCA plus left ICA occlusions 1 SCA = subclavian artery; ICA = internal carotid artery; VA = vertebral artery; IA = innominate artery; CCA = common carotid artery. were not randomized, potential patient differences that could affect the thrombosis and CVA rates were analyzed. No significant differences were noted in factors of age, sex, length of follow-up, previous history of CVA, or distribution of lesions (Table II). The early and late results of the 36 patients who underwent carotid-subclavian bypass were analyzed according to the graft material used. Early follow-up was considered to be from the time of operation to 1 month and results thereafter are presented as late follow-up. Results were subjected to statistical analysis by the construction of 5-year actuarial life-table curves that specifically focused on the CVA-free interval and patcncy rate of each group of patients. The log rank test, or Mantel-Haenszel test was used to analyze the actuarial life-table curves to detect whether real differences exist overall between patients with prosthetic grafts, vein grafts, or primary anastomosis. RESULTS No statistically significant difference was observed between vein grafts, prosthetic grafts, or transposition with primary end-to-side anastomosis in the early results (Table III). Symptoms of cerebrovascular insufficiency were completely relieved in 88.8% of patients with prosthetic grafts, 84.6% of patients with vein grafts, and in all patients who had primary anastomosis. One postoperative CVA occurred in each of the prosthetic graft and vein graft groups. There were no postoperative CVAs in patients undergoing transposition with primary anastomosis. These differences were not statistically significant (p <.1) by the Fishcr exact test. One postopcrative death occurred after occlusion of a vein graft that

3 142 Ziomek et al. Journal of VASCULAR SURGERY Table II. Composition of treatment groups Prosthetic grafts Vein grafts Primary anastomosis Follow-up (mo) Mean Range Median Quartile 19, 81 18, , 6 Age (yr) Range Median Qualxile 52, 24 56, 78 54, 68 Sex M F Presenting symptoms Hemispheric Nonhemispheric Hypertension (No. of patients) 85 (1/18) 88 (8/13) 86 (3/5) Previous CVAs (No. of 11.1% (2/18) 15.4% (2/13) 2.% (1/5) patients) Lesions treated Left SCA stenosis/occlusion Right SCA stenosis/occlusion 2 2 IA stenosis/occlusion 4 1 CCA stenosis/occlusion 2 3 For legend see Table I. Table III. Early follow-up (3 days after operation) Symptoms completely Symptoms relieved decreased No. of CVAs No. of deaths Patency Prosthetic grafts 16/18 (88.8%) 2/18 (5.5%) 1/18 (5.5%) 17/18 (94.4%) Vein grafts 11/13 (84.6%) 1/13 (7.6%) 1/13 (7.6%) 1/13 (7.6%) 12/13 (92.3%) Primary anastomoses 5/5 (1%) 5/5 (1%) Table IV. Late follow-up (1 month to 13 years; mean interval, 4.3 years) Patients without Symptoms No. of symptoms return No. of CVAs deaths Patency Prosthetic grafts 15/18 (83.3%) 2/18 (1i%) 1/18 (5.5%) 2 16/17 (94.1%)t Vein grafts 6/12 (5%) 2/12 (16.7%) 4/12 (33.3%) 2 3/12 (25%) 7/12 (5~.3%)t Primary anastomoses 4/5 (8%) 1/5 (2%) 5/5 (k,d%) *p <.545. tp <.1. resulted in an ipsilateral cerebral infarction. Early thrombosis of carotid-subclavian bypass occurred in one additional patient. This patient had a prosthetic graft that occluded on the tenth postoperative day; this occlusion was considered the result of technical error. The patient refused reoperation despite recurrence of symptoms. In a mean follow-up of 4.3 years, a significant difference in patients with prosthetic compared with vein grafts was noted (Table IV). Eighty percent of patients with primary anastomosis and 83.3% 6f patients with prosthetic grafts remained without symp- toms in contrast to only 5% of patients with vein grafts who remained symptom free. Four of 12 patients (33.3%) with vein grafts suffered late CVAs compared with 1 of 18 patients (5.5%) with prosthetic grafts. All thrombosed vein grafts were autogenous saphenous veins. Life-table analysis of these results (Fig. 1) shows that at the end of 5 years the prosthetic graft patency rate is 91% compared with the vein graft patency rate of 57%. Log rank testing shows the difference between these two curves is statistically significant (p <.1). Life-table analysis of the CVA rate (Fig. 2), at the end of 5 years, shows

4 Volume 3 Number 1 January 1986 Deleterious effects of vein grafts in carotid-subclavian bypasses " ~ I O = : = * PERCENTAGE OF GRAFTS PATENT 6 57* 4-2- STROKE FREE PAT I ENTS 4O 2 61" I I I I I I / I I Monlhs Fig. 1. Five-year life-table analysis of graft patency. Closed circles, prosthetic graft; open circles, vein graft; open squares, primary anastomosis, p <.1. that 94% of patients with prosthetic grafts remained free q/~ CVA compared with 61% of patients with vein grafts. Log rank testing shows the difference between these two curves approaches statistical significance (p <.545). All patients in whom a carotid-subclavian transposition with primary anastomosis was performed have remained free of CVA and their reconstructions remained patent at the time of this report. One patient had ipsilateral amaurosis fugax but refused angiography. The reconstruction was shown to be patent by physical examination and noninvasive studies. Seventeen of the 36 patients who underwent carotid-subclavian bypasses had the carotid artery used as the donor vessel and the subclavian artery as the recipient vessel. Four bypasses became thrombosed (23.5%) and three patients suffered three ipsilateral CVAs (17.6%). Nineteen patients had the subclavian ~tery used as the donor and the carotid artery as the recipient vessel. Four of these bypasses became thrombo~ed (21%) and those four patients suffered ipsitatera/cvas (21%). There is no statistical significance between these two subgroups. Two patients in the subgroup of five patients who had subclavian bypasses done to treat common carotid artery occlusions had CVAs before operation that were resolved; none suffered CVA during follow-up. Six patients underwent ipsilateral carotid endarterectomy concomitant with carotid-subclavian bypass. One patient experienced recurrence of symptoms but none experienced CVA during follow-up. One reoperation (5.5%) was done in a patient who had a prosthetic graft because the distal anastomosis of the graft became stenotic, which caused recurrence of symptoms. The patient underwent a carotid-to-carotid bypass with a Dacron graft and had complete relief of symptoms. During follow-up I I I I I I t I I Monlhs Fig. 2. Five-year life-table analysis of CVA-free rate. Closed circles, prosthetic grafts; open circles, vein grafts; open squares, primary anastomosis, p <.545. five reoperations were needed in the surviving 12 patients (41.6%) who had carotid-subclavian bypasses with autologous vein. Two patients underwent angioplasty of the distal anastomosis, and one patient had excision of the vein graft and stenotic subclavian artery with primary anastomosis of the subclavian and axillary artery. Two patients had "redo" carotid-subclavian bypasses with prosthetic grafts. Since 198 eight of nine carotid-subclavian bypasses done at UCLA have been done with prosthetic grafts; five were PTFE and three were Dacron. All PTFE grafts have remained patent and the patients have remained without symptoms with an average follow-up of 31/2 years. One of three patients with Dacron grafts suffered recurrence of symptoms with a patent graft as documented by angiography. Four deaths occurred during follow-up. One patient died on the tenth day after operation of a massive CVA on the side of an occluded vein graft. Three patients died of myocardial infarctions in late followup. All three of these patients had vein grafts and died at 39, 47, and 7 months, respectively. Each operative report was reviewed and the length of time required to perform each operation was analyzed. The average time for carotid-subclavian bypass revascularization with prosthetic graft was 2 hours and 39 minutes, compared with 3 hours and 45 minutes for vein graft. No differences were noted in surgical or anticoagulation technique. All patients in this study received aspirin and dipyridamole postoperatively. DISCUSSION Numerous reports have documented the effectiveness of carotid-subclavian bypass in the relief of cerebrovascular and/or ipsilateral arm symptoms in

5 144 Ziomek et al. Journal of VASCULAR SURGERY Table V. Carotid-subclavian bypass patency according to type of graft used as reported in the literature Prosthetic grafts No. of Long-term patency Author (Year) patients Follow-up No. (%) ~ No. Vein grafts Long-term patency (%) Diethrich et al.1 (1967) mo-14 yr /125 (97.6) Moore, Malone, Goldstone 6 (1976) 15 3 mo-8 yr 9 9/9 (1) Wylie, Effeney 2 (1979) 21 4 mo-8 yr 4 4/4 (1) Raithel 3 (198) 37 Mean, 3 yr 37 36/37 (97.2) Thompson, Read, Campbell mo-17 yr 64 61/64 (95.3) DePalma, Broadbent 9 (1981) 14 3 mo-12 yr 14 14/14 (1) Vogt et al) (1982) 37 2 mo-15 yr 1 1/1 (1) Current study 3_.66 6 too-9 yr /17 (94.1)I: Total 349 Mean 1.7 yr /271 (97.%) ~Includes grafts removed because of infection. tone patient had an early graft occlusion (on tenth postoperative day) because of technical error. ~:One patient died in postoperative period. 6 4/6 (66.6) 17 12/17 (7.5) 36 3/36 (83.3) 13 7/12 (58.3):~ 72 53/71 (74.6%) patients with significant stenosis or occlusion at the origin of the subclavian or common carotid artery. 1-6 The initial concern about the creation of a carotid steal has disappeared as it is clear from experimental and clinical data that a steal will only occur in the presence of significant stenosis in the inflow artery. 7 In contrast to the direct intrathoracic approach, extrathoracic carotid-subclavian bypass has lower morbidity and mortality. The durability of the extrathoracic revascularization of branch occlusions of the aortic arch, specifically carotid-subclavian bypass, has been well documented by others and our series adds to that experience (Table V). We reported a graft patency rate of 71% at 5 years for extra-anatomic brachiocephalic reconstructions in a previous report from this institution. 1~ However, differences related to graft material used were not analyzed. The indication for operation and use of concomitant carotid endarterectomy were compared between the group of patients with vein grafts and prosthetic grafts. One difference was noted; four of five patients with subclavian-to-contralateral carotid bypass were done with prosthetic grafts and all five of these reconstructions have remained patent. Concomitant carotid endarterectomy had no deleterious effect on long-term patency of the carotid-subclavian graft or on the CVA rate at 5 years. One of the largest series reported in the literature on carotid-subclavian bypass consists of 125 cases, all done with Dacron grafts) In that series, two grafts occluded, one at 4 months and the other during longterm follow-up, for an overall thrombosis rate of 1.6%; follow-up was from 9 months to 14 years. One additional graft was removed because of infec- tion, which resulted in a total of three graft-related problems (2.4%) on long-term follow-up. The largest published series of carotid-subclavian bypass done with autogenous vein consists of 36 reconstructions with a follow-up ranging from 2 months to 15 years) Two (5.5%) postoperative CVAs occurred in this series with one (2.7%) early thrombosis. In that series there have been four thromboses, one aneurysmal dilatation that required replacement, and one stenosis that required revision, for a total of six (16.6%) graft-related problems. Although those results appear better than our experience, they are inferior to those achieved with prosthetic grafts in our own series. The combined experience of seven series of carotid-subclavian bypass with Dacron grafts, including our own, yields a total of 272 cases with an over# ~ average follow-up of 1.7 years) 3'6,81 Five thromboses (1.8%) and four (1.4%) graft infect&gns have been reported among these patients, for ~a total of nine (3.3 %) graft-related problems. In three of these series combined with our own, a total of 72 carotidsubclavian bypasses with autogenous vein are reported. 2,6,1 Within these series combinations, 18 (25.3%) thromboses occurred during the same follow-up period. These combined results are similar to ours, in which long-term patency of the Dacron bypass was superior to those with autogenous vein. Several possibilitics serve to explain thc poor performance experienced with the use of autologous vein in carotid-subclavian reconstructions. The caliber of the vein may be a limiting factor in this position. The mobility of the neck in several axes may cause kinking and shortening of the graft. The inhcrent stiffness of the prosthetic grafts may represent

6 Volume 3 Number 1 January 1986 Deleterious effects of vein grafts in carotid-subclavian bypasses 145 av advantage in this situation. In our experience the duration of the operation was reduced with the use of prosthetic grafts, which represents an additional benefit for this group of patients. An attractive alternative to the use of free grafts for the revascularization of the common carotid or subclavian artery is the transposition of the occluded or stenotic vessel with primary end-to-side anastomosis. Although this procedure is suitable only for patients with lesions at the origin of the vessel, this feature occurred in the majority of patients in this group. We have performed five carotid-subclavian transpositions through a cervical approach; followup averaged 5 months. All patients were relieved of their initial symptoms and all reconstructions are pa- : nt at the time of this study. Our experience is shared by others. 4 In r',ummary, our own results and a review of the literature lead us to conclude that prosthetic grafts fire superior to vein grafts for long-term patency in carotid-subclavian bypasses. The use of autologous vein should be avoided unless specific circumstances exist (i.e., infection). Although the number of carotid-subclavian primary anastomoses is very small in this study, brachiocephalic revascularization without a graft when feasible may yield the best overall long-term patency. We extend our sincere appreciation to Candace L. Vescera, R.N., whose assistance during the preparation of this manuscript was most valuable. REFERENCES 1. Diethrich EB, Garrett EH, Amersio J, Crawford ES, EI-Bayar M, DeBakey ME. Occlusive disease of the common carotid and subclavian arteries treated by carotid-subclavian bypass. Am J Surg 1967; 114: Wylie EJ, Effeney DJ. Surgery of the aortic branches and vertebral arteries. Surg Clin North Am 1979; 59: Raithel D. Our experience of surgery for irmominate and subclavian lesions. J Cardiovasc Surg (Totino) 198; 21: Edwards WH, Wright RS. Current concepts in the management of arteriosclerotic lesions of the subclavian and vertebral arteries. Ann Surg 1972; 175: Crawford ES, DeBakey ME, Morris GC, Howell IF. Surgical treatment of occlusion of the innominate, common carotid, and subclavian arteries: A 1-year experience. Surgery 1969; 65: ;' 6. Moore WS, Malone JM, Goldstone J. Extrathoracic repair of branch occlusions of the aortic arch. Am J Surg 1976; 132: Lord RS, Ehrenfeld WK. Carotid-subclavian bypass: A hemodynamic study. Surgery 1969; 66: Thompson BW, Read RC, Campbell GS. Operative correction of proximal blocks of the subclavian or innominate arteries. J Cardiovasc Surg (Torino) 198; 21: DePalma RG, Broadbent RV. Management of occlusive disease of the subclavian and innominate arteries. Am J Surg 1981; 142: Vogt DP, Hertzer NR, O'Hara PJ, Beven EG. Brachiocephalic arterial reconstruction. Ann Surg 1982; 196: Livesay JJ, Atkinson JB, Baker JD, Busuttil RW, Barker WJ, Machleder HI. Late resuks of extra-anatomic bypass. Arch Surg 1979; 114:126-7.