0148-396 X/86/1804-0483$02.00/0 NEUROSURGERY Copyright © 1986 by the Congress of Neurological Surgeons Vol. 18, No. 4, 1986 Printed in U.S.A. Complications after Subclavian-Cortical Middle Cerebral Artery Bypass Massimo Collice, M.D., Orazio Arena, M.D., and Maurizio Riva, M.D. Department of Neurosurgery (MC, OA) and Department of Neurology (MR), Ospedale Maggiore Niguarda Ca’ Granda, Milan, Italy Subclavian-cortical middle cerebral artery bypass by saphenous vein interposition is an uncommon recent revasculari- zation technique used for immediate volume flow to the brain or when donor arteries are unavailable or inadequate. We used this technique in four cases of symptomatic common carotid artery occlusion. In the last two cases, severe, reversible neurological complications occurred. We think that these complications are due to this particular technique, and more cautious indications than for traditional extra-intracranial arterial bypass are recommended. (Neurosurgery 18:483-486, 1986) Key words: Cerebrovascular ischemic disease, Common carotid artery occlusion, Subclavian-cortical middle cerebral artery bypass INTRODUCTION Since the introduction to clinical practice of the bypass graft between the superficial temporal artery (STA) and a cortical branch of the middle cerebral artery (MCA) (18), many vessel-to-vessel anastomoses have been described for both supratentorial and subtentorial revascularization (8, 11, 16). Increasing collateral circulation also has been obtained by a series of interposition grafts, usually when donor arteries were inadequate or unavailable (1, 4, 7, 10, 12, 14, 15, 17). One of the latest techniques reported is bypass between the subclavian artery and a cortical branch of the MCA (S-cMCA) by saphenous vein interposition. The technique was intro- duced in 1980 by Spetzler et al., who reported the results obtained in 11 patients (12). We have treated 4 patients using this method; after the last 2 procedures, serious, although reversible, neurological complications were observed. Because the type of complications and their severity differed from the general and our own experience with traditional bypass pro- cedures, we are reporting these 2 cases, although the causes of the complications are not fully understood. CASE REPORTS Case 1 The patient was a 53-year-old, right-handed man, a smoker, who had suffered from intermittent claudication for 2 years. On January 1, 1982, he had an ischemic stroke (aphasia and severe hemiparesis of the right limbs) and was admitted to our Neurological Department. He partially recovered over a l-month period. Sequential computed tomographic (CT) scans showed mild, diffuse corticosubcortical cerebral atrophy and the development of a large low density area in the white matter of the left hemisphere. Angiographic study of the aortocerebral circulation revealed occlusion of the left com- mon carotid artery (CCA) and 40% stenosis of the left verte- bral artery at its origin. The patient was discharged with antiplatelet therapy. Two months later, he had a myocardial infarction without sequelae. At the end of October 1982, the patient had several transient episodes of worsening of his residual aphasia. A regional perfusion study using a new brain-imaging agent, iodine-123-labeled trimethylpropane diamine (!3I-HIPDM), 483 and single photon emission computed tomography (5) showed large areas of reduced activity in the left MCA territory. According to our usual approach for these patients (2), on November 24, 1982, a S-cMCA bypass was performed using the saphenous vein after exploration of the carotid bifurcation had demonstrated old thrombosis of the internal carotid artery (ICA) and excluded the use of the origin of the external carotid artery (ECA) for a subclavian-ECA bypass. At operation, the flow rate at the open-ended graft was 160 ml/minute. The cortical artery was clamped for 25 minutes. The course of the operation was uneventful. When he awoke from anesthesia, the patient had partial motor seizures contralateral to the operated side, which subsequently generalized. The patient was comatose for 5 days; he had motor a reactivity and preservation of brain stem reflexes. When the patient awoke, global aphasia and right hemiplegia were present. Serial CT scans demonstrated infarction in the left parietal area, the site of the anastomosis (Fig. 1). An angiogram obtained 25 days after operation showed a patent anastomosis and peculiar features of the cerebral cir- culation, i.e., very early venous filling in the revascularized area (Fig. 2). Subsequently, the patient had a slow but pro- gressive recovery from his neurological deficits, and he was discharged in January 1983 with his clinical condition sub- stantially the same as at initial admission. Assessment of regional cerebral perfusion 3 months after operation showed the same picture as the preoperative study, i.e., large areas of decreased cerebral perfusion in the left hemisphere without any increase of activity when compared to the preoperative examination. Case 2 This patient was a 56-year-old man suffering from heart valvular disease and intermittent claudication for some years. In June 1981, he suddently experienced left hemiplegia, with subsequent partial recovery. At this time, a Doppler exami- nation showed right CCA occlusion. In November 1982, the patient was a candidate for cardiac surgery and therefore was sent to our department for surgical evaluation. A CT scan revealed a small, infarcted area deep in the parietal area. Angiographic study of the aortocerebral circulation showed right CCA occlusion and right vertebral occlusion. A regional perfusion study with '*I-HIPDM showed reduced activity in 484 COLLICE et al. Fic. |. Case 1. Preoperative and postoperative CT scans in a 53- year-old man with left CCA occlusion who underwent S-cMCA bypass. Lefi, preoperative CT scan shows a large, low density area in the white substance of the left hemisphere and mild diffuse cortico- subcortical cerebral atrophy. Right, postoperative CT scan shows a new ischemic infarction in the left parietal area, corresponding to the site of anastomosis. Fic. 2. Case 1. Postoperative selective left subclavian artery angio- gram shows the bypass flow in the saphenous vein prosthesis between the subclavian artery and a parietal branch of the MCA. Note the early filling of the superior sagittal sinus because of the very short circulation time. the right MCA territory. To protect the extracorporeal circu- lation, a right S-cMCA bypass was performed after explora- tion of the carotid bifurcation on December 12, 1982. The flow rate in the open-ended graft was over 500 ml/minute. Neurosurgery, Vol. 18, No. 4 The cortical artery was clamped for 20 minutes. After the anastomosis was completed, the clamp on the saphenous graft was released progressively during a 30-minute period. When the patient awoke from anesthesia, he had left hemiplegia and stupor. CT scan controls executed on the same day and on the 3rd postoperative day did not show any significant mod- ification when compared with the preoperative scan (Fig. 3). Another CT scan obtained 45 days after operation was appar- Fic. 3. Case 2. Preoperative (/efi) postoperative (right) CT scans in a 56-year-old patient affected by right CCA occlusion who under- went S-cMCA bypass grafting with saphenous vein interposition. No differences between the pre- and postoperative examination are shown, except for the poorer visualization in the postoperative scan of the cortical sulci of the right hemisphere. Fic. 4, Case 2, Anteroposterior postoperative selective right sub- clavian artery angiogram obtained 45 days after operation shows the stump (arrowhead) of the occluded S-cMCA bypass. April 1986 COMPLICATIONS AFTER S-cMCA BYPASS 485 ently identical to the preceding ones. Angiography 25 days after operation showed a patent bypass with filling of three cortical MCA areas. The clinical picture showed slow, but progressive improvement of both consciousness and focal defici months later, the neurological examination was the same as that at admission. Another angiogram, obtained 45 days after operation, showed occlusion of the graft at the origin with patency of the subclavian artery (Fig. 4). DISCUSSION Cerebral revascularization techniques that use vascular or synthetic prostheses have been developed to offer revascular- ization to patients with inadequate or unavailable donor arteries and to increase flow in the MCA territory more immediately than is possible with the traditional extracranial- intracranial bypass. In 1971, Lougheed et al. first reported a bypass procedure from the CCA to the intracranial ICA using a saphenous vein graft in a case of ICA occlusion after endarterectomy (7). The technical difficulties presented by the depth of the anastomotic site and the need to clamp a major cerebral artery prevented wide use of the technique. In 1979, Tew described the technique of venous interposition between the proximal STA or occipital artery and cortical branches of the MCA (17). We have seen few reports about this procedure, but probably it has been used successfully by many surgeons in patients with inadequate scalp arteries. Subsequently, a variety of more ambitious techniques for obtaining an immediate large volume flow have been re- ported, including graft interposition between major extracran- ial arteries, even the aortic arch, and cortical MCA branches. Except for some cases in which the radial artery or synthetic prostheses were used (1, 14), most surgeons have used the saphenous vein (4, 10, 12, 15). First, the CCA or ECA was used as the donor artery. Later, the subclavian artery was used; it has some definite advantages. It is straighter, so that turning of the head does not cause angulation or displacement of the graft, and clamping of a major artery to the brain is avoided. Moreover, the technique can be used in the presence of CCA occlusion. The results reported thus far seem to be encouraging, although the numbers are small. Among the 11 patients treated by Spetzler et al., | died of unrecognized subdural hematoma (12). In a series of 15 patients treated with long interposition grafts, Samson et al. reported a single perioper- ative death from cardiac arrhythmia and a postoperative subdural hematoma successfully evacuated (10). However, many aspects of this technique remain to be clarified, as was stressed by the authors who first used them. These include the size discrepancy between the venous graft and the cortical recipient artery, possible displacement of the graft along the subcutaneous tunnel, and the risk of postoperative hemor- thage and parenchymal damage (edema or hemorrhagic in- farction). The neurological complications observed in the 2 patients reported do not seem to be caused by the problems men- tioned. It seems unlikely that complications such as focal deficits, epileptic fits. and impairment of consciousness could be ascribed to the cortical artery clamping time; it was not longer than that for traditional STA-cMCA bypass. After routine STA-cMCA procedures, we have never seen such complications: among 200 procedures, we observed only | severe ischemic complication. The clinical picture in this patient was typical ischemic stroke; there was marked neuro- logical deficit without impairment of consciousness, and re- covery was poor. Several hypotensive episodes occurred dur- ing and after operation, and the bypass graft was shown to be closed by postoperative angiography. CT scan examination showed a new infarcted area. Other immediate postoperative neurological deficits were observed, but in general they were diffuse, slight, and lasted only a few days. Often they appeared in patients with transiently or permanently closed bypasses (unpublished data). In the two cases here reported, on the other hand, the neurological impairment was severe, although completely re- versible. The bypass grafts were functioning; in one case, the CT scan did not show any change when compared to the preoperative scan. Because the operation was performed 10 and 18 months, respectively, after stroke and | month after the last transient ischemic attack in Case 1, the blood-brain barrier should have been intact in these patients; therefore, it does not seem plausible to attribute the neurological impair- ment to a “normal perfusion pressure breakthrough” (13). Moreover, the CT scan findings do not support this hypoth- esis; no signs of hemorrhagic infarction or significant edema were seen. We have no conclusive explanation for these complications. They are probably ischemic, as suggested by the benign clinical evolution, the CT scan results, and the postoperative tomographic perfusion studies. We used tomo- graphic perfusion studies in a group of patients before and after traditional STA-cMCA bypass, and usually an increase in cerebral perfusion was observed postoperatively (3, 6). This increase was not found in Case 1. It seems reasonable that the immediate enormous volume flow of the bypass is involved in some way. The postoperative angiographic pattern in one of these patients is peculiar, especially considering the excep- tional reduction of the circulation time in the revascularized area (Fig. 2). Cerebral ischemia could have been produced by a sort of shunt flow or by venous hypertension, at least in this patient. The angiographic picture may be due to the epiphen- omenon of cerebral ischemia (9), but we never saw this angiographic pattern in postoperative imaging of more than 180 extra-intracranial bypass procedures. CONCLUSION Our impressions about this new procedure are that the size discrepancy between venous graft and cortical recipient artery is not a real technical problem. Contrary to our expectation, the revascularization evaluated by postoperative angiography has not been very different from that observed after traditional STA-cMCA anastomosis, although the small number of pa- tients treated does not allow a definitive judgment. The occlu- sion of a graft 45 days after operation emphasizes one of the most important problems of venous grafts, that of long term patency. The risk of serious complications due to the imme- diate large flow in a small artery is real. In our opinion, saphenous vein S-cMCA bypass grafting represents a new technique available to neurosurgeons in the field of cerebral revascularization, but more experience is necessary before it can be objectively evaluated. In the mean- time, we suggest using this procedure only when it is not possible to use more proven techniques and when there is very strong evidence that revascularization is necessary. Received for publication, July 7, 1985; accepted, November 15, 1985. Reprint requests: Massimo Collice, M.D., Department of Neuro- surgery, Ospedale Maggiore Niguarda Ca’ Granda. P.za Ospedale Maggiore Number 3, 20100 Milan, Italy. REFERENCES 1. Ausman JI, Nicoloff D. Chou S: Posterior fossa revascularization: Anastomosis of vertebral artery to posterior inferior cerebellar artery with interposed radial artery graft. Surg Neurol 9:28 1-285, 1978. 486 COLLICE et al. . Collice M. D'Angelo V. Arena O: Surgical treatment of common carotid artery occlusion. Neurosurgery 12:515-524, 1983. . Collice M. Fazio F. Fieschi C. Arena O. Possa M. Beduschi A: Angiographic findings and cross-sectional brain perfusion studies after extra-intracranial arterial bypass, in Peerless SJ. McCormick CW (eds): Microsurgery for Cerebral Ischemia. New York, Springer Verlag. 1980. pp 142-147. . Dorsch NCW: Aorta to middle cerebral artery bypass for ischemia following carotid ligation. Presented at the Fifth International Symposium on Microvascular Anastomoses for Cerebral Ische- mia, Wien, Austria, September 14-17. 1980. . Fazio F, Lenzi GL. Gerundini P, Collice M, Gilardi MP, Co- lombo R. Taddei G, Del Maschio A. Piacentini M, Kung HF, Blan M: Tomographic assessment of regional cerebral perfusion using intravenous I'?*-HIPDM and a rotating gammacamera. J Comput Assist Tomogr 8:911-921, 1984. . Fazio F, Lenzi GL. Gerundini B. Fieschi C, Collice M, Pozzilli C. Gilardi MP: Assessment of regional cerebral reperfusion with SPECT and I'?-HIPDM in patients with EC-IC bypass. Presented at the | 1th CBF Meeting of the International Society of Cerebral Blood Flow and Metabolism, Paris, France, June 20-24, 1983. . Lougheed WM, Marshall BM, Hunter M, Sandwith-Smyth H: Common carotid to intracranial internal carotid bypass venous graft: Technical note. J Neurosurg 34:114-118, 1971. . Nishikawa M, Hashi K, Shiguma M: Middle meningeal-middle cerebral artery anastomosis for cerebral ischemia. Surg Neurol 12:205-208. 1979. . Olsen TS, Lassen NA: Partial or total reperfusion of cerebral infarcts due to spontaneous recanalisation of occluded arteries: A common phenomenon in acute stroke. Presented at the 6th International Symposium on Microsurgical Anastomoses for Cer- ebral Ischemia, Kyoto, Japan. September 12-15, 1982. . Samson DS, Gewertz BL, Beyer CW, Hodosh RM: Saphenous vein interposition grafts in the microsurgical treatment of cerebral ischemia. Arch Surg 116:1578-1582, 1981. . Spetzler RF, Chater N: Occipital artery-middle cerebral artery anastomosis for cerebral artery occlusive disease. Surg Neurol 2:235-238, 1974. . Spetzler RF, Rhodes RS, Roski RA. Likavec MJ: Subclavian to middle cerebral artery saphenous vein bypass graft. J Neurosurg 53:465-469, 1980. . Spetzler RF. Wilson CB, Weinstein P, Mehdorn M, Townsend J. Telles D: Normal perfusion pressure breakthrough theory. Clin Neurosurg 25:651-672. 1978. . Story JL, Brown WE Jr, Eidelberg E: Cerebral revasculariza- tion—Proximal to distal middle cerebral artery bypass with a synthetic tube graft. Neurosurgery 3:61-65, 1978. . Story JL, Brown WE, Eidelberg E, Arom KV, Stewart JR: Cerebral revascularization—common carotid to distal middle cerebral artery bypass. Neurosurgery 2:131-134, 1978. . Sundt T. Piepgras C: Occipital to posterior inferior cerebellar artery bypass surgery. J Neurosurg 48:916-928. 1978. . Tew JM Jr: Techniques of supratentorial cerebral revasculariza- tion. Clin Neurosurg 26:330-345, 1979. . Yasargil MG: Anastomosis between the superficial temporal artery and a branch of the middle cerebral artery, in Yasargil MS (ed): Microsurgery Applied to Neurosurgery. Stuttgart, Georg Thieme Verlag. 1969, pp 105-115. COMMENT The authors present two cases of severe, although reversible, neurological complications after subclavian to cortical middle cel rebral artery bypass using the saphenous vein as an inter- Neurosurgery, Vol. 18, No. 4 position graft. As the authors point out, the first major series using this procedure was reported in 1980 by Spetzler and colleagues (5), and there was a subsequent communication by Samson et al. (4). The present report on complications in 2 of 4 long saphenous vein bypasses clearly points out the two major problems. First, is the decreased overall patency. In the recent series by Diaz et al., there were 5 occlusions in 17 patients with long grafts to a middle cerebral artery branch (1). Hodosh recently reviewed his 37 patients treated with saphenous vein to middle cerebral artery bypass; there was an overall patency rate of 84%, with 3 early and 3 delayed graft closures. Using short saphenous vein bypasses, the patency rate seems to be higher, as reported recently by Little (3) and by Diaz (1). The pathogenesis of the increased incidence of occlusion of saphenous vein bypasses may be more technical with regard to early occlusion and related to endothelial damage and internal hyperplasia with delayed closure. Cer- tainly, as we recently reported, this problem of endothelial hyperplasia is not unique to vein bypasses, as we clearly demonstrated in a patient with superficial temporal artery to middle cerebral artery bypass (2). The most interesting aspect of this report is the transient, but very severe neurological deficits that are implicated as being at least partially caused by the very high flow rates. One patient had a flow rate at the opened end of the graft of 160 ml/minute, and the other had an astounding flow of 500 ml/ minute. I certainly agree with the authors that placing such high flow rates into brain in patients with documented ische- mia (as evidenced by blood flow studies) almost certainly contributed to the postoperative neurological deficits. Indeed, in the series by Diaz, these high flow rates probably contrib- uted to the increased incidence of intracerebral hemorrhage in vein graft patients (1). Thus, if it is decided that some sort of revascularization is indicated, a decision that is going to be very difficult to make in view of the recent results of the international cooperative EC-IC study, I certainly agree with the authors’ conclusion that this procedure should only be used “when it is not possible to use more proven techniques and where there is very strong evidence that revascularization is necessary.” Edward A. Neuwelt, M.D. Portland, Oregon 1. Diaz FG, Pearce J, Ausman JI: Complications of cerebral revas- cularization with autogenous vein grafts. Neurosurgery 17:271- 276, 1985. . Gumerlock MK, Coull BM, Howieson J, Buchan C, Neuwelt EA: Late stenosis of a superficial temporal-middle cerebral artery bypass: Angiographic and histological findings. Neurosurgery 16:650-657, 1985. 3. Little JR, Furlan AJ, Bryerton B: Short vein grafts for cerebral revascularization. J Neurosurg 59:384-388, 1983. 4. Samson DS, Gewertz BL, Beyer CW Jr, Hodosh RM: Saphenous vein interposition grafts in the microsurgical treatment of cerebral ischemia. Arch Surg 116:1578-1582, 1981. 5. Spetzler RF, Rhodes RS. Roski RA, Likavec MJ: Subclavian to middle cerebral artery saphenous vein bypass graft. J Neurosurg 53:465-469, 1980. N