J Neurosurg 64:309-312, 1986 Inadvertent middle cerebral artery embolism by a detachable balloon: management by embolectomy Case report KARL W. SWANN, M.D., ROBERTO C. HEROS, M.D., GERARD DEBRUN, M.D., AND CURT NELSON, M.D. Neurosurgery Service and Department of Radiology, Massachusetts General Hospital, Boston, Massachusetts, and Neurosurgical Service, Strong Memorial Hospital, Rochester, New York o, A case of middle cerebral artery embolism by a detachable intra-arterial balloon is presented. The balloon migrated after being detached in an effort to occlude the internal carotid artery proximal to an unclippable giant paraclinoid aneurysm. Volume expansion, induced hypertension, anticoagulation therapy, rapid middle cerebral artery embolectomy, and good collateral circulation are factors that may have contributed to the patient's complete recovery from hemiplegia. KEY WORDS giant aneurysm " detachable balloon T HE use of detachable balloons to occlude carotid cavernous fistulas and other lesions has been described by Serbinenko 9 and Debrun, et al. 3 Since the initial report by Welch, 1~ middle cerebral artery (MCA) embolectomy has been used in cases of naturally occurring or iatrogenic embolisms, with varying degrees of success. 6,7 This report details the successful management, by microsurgical embolectomy, of MCA occlusion by an embolizing balloon which was being used to occlude the internal carotid artery (ICA) proximal to an unclippable giant paraclinoid aneurysm. The surgical technique, medical management, and clinical course are described. Case Report This 53-year-old left-handed woman was discovered to have bilateral giant paraclinoid artery aneurysms during an evaluation for Cushing's syndrome. The aneurysm on the left was explored at another institution and was determined to be unclippable because of difficulty in exposing its proximal neck. The aneurysm was, therefore, wrapped and glued, but the patient's ipsilateral vision deteriorated to blindness over 4 years in spite of this. Vision in the right eye remained normal. On routine follow-up angiography, the right-sided anJ. Neurosurg. / Volume 64 / February, 1986 9 middle cerebral artery 9 embolectomy 9 eurysm was found to have grown significantly, and the patient was referred for an elective attempt at clipping. Operations. We performed a right pterional craniotomy and could not adequately expose the proximal neck of the aneurysm in spite of thoroughly drilling the anterior clinoid process. From the findings at surgery and further retrospective study of the arteriogram, it became apparent that the neck of the aneurysm was largely intracavernous, protruding intracranially in an "hourglass" fashion. Nine days later, after recovering fully from the craniotomy, the patient underwent attempted intra-arterial balloon occlusion of the right ICA proximal to the aneurysm. We did not consider that a preliminary bypass procedure was necessary in view of the adequacy of collateral circulation. Test occlusion for 10 minutes was tolerated without neurological deficit. Detachment of the balloon was achieved, but a few minutes later it was almost totally deflated and had migrated distally to the MCA trunk, which it totally occluded (Fig. 1). Simultaneously, the patient developed a left hemiplegia and dysarthria. An intravenous bolus of 500 cc 5% albumin was given, and an intravenous solution of 5% dextrose and 0.45% saline was administered at a rate of 125 cc/hr. A Neo-Synephrine drip (phenylephrine, 309 K. Swann, et al. FIG. |. Arteriograms, anteroposterior (left) and lateral (right) views, showing nonfilling of the right middle cerebral artery (MCA) distal to the balloon embolus. The radiopaque balloon is seen lodged in the proximal MCA. 10 me/250 cc 5% dextrose) was used to maintain a systolic blood pressure of 160 mm Hg. Dexamethasone, 10 gm, was given intravenously. The heparin used during the procedure was not reversed and the introducer catheter was left in the groin and sutured in place to prevent a hematoma. The patient was taken to the operating room where anesthesia was induced with a bolus of Pentothal (thiopental). In addition, 1000 U of intravenous heparin was given at this time. The previous pterional craniotomy was rapidly reopened. With use of a microscope, the MCA was followed distally by splitting the Sylvian fissure. The balloon was found lodged in the main trunk of the MCA, occluding several lenticulostriate vessels. Temporary low-pressure aneurysm clips were used to occlude the proximal MCA and its three distal divisions. An arteriotomy of about 1 cm in length was required since the balloon was long and relatively stiff. The balloon was deflated by puncturing and extracted with a microhook. Vigorous bleeding occurred from unclipped perforating arteries which had to be occluded individually with small temporary clips to allow precise closure of the arteriotomy. The arteriotomy was closed with interrupted 9-0 nylon sutures, and all clips were released. Flow appeared to be excellent and no areas of stenosis were apparent. At this time we gave the patient 10 mg of protamine sulfate to reverse anticoagulation at least partially. We estimate that flow was reestablished approximately 3 hours after the time of embolic occlusion of the MCA. Postoperative Course. Immediately after the operation the patient began to move her left side. She then 310 continued to improve gradually so that by 36 hours after surgery she had returned to her preoperative condition. In view of the difficulties encountered, a conservative course of action with periodic angiography, computerized tomography scanning, and visual examinations was elected. At the time of this writing, 1 year after the events described, the vision in the right eye remains normal. A recent arteriogram has shown that the aneurysm is unchanged in size and configuration, and that the previously embolectomized fight MCA is completely normal in appearance (Fig. 2). Discussion The use of detachable intra-arterial balloons in the management of carotid-cavernous fistulas and other lesions has become routine in some centers. Barrow, et aL,l have described potential complications of the technique, including cavernous sinus thrombosis, balloon rupture, neck hematoma from direct carotid artery puncture, and asymptomatic MCA occlusion by an embolus presumed to have been generated by the catheter system. Chalif, et al., 2 reported a patient in whom a detached balloon migrated distally, occluding the ICA at the bifurcation. This patient was managed successfully by embolectomy. Langford, et al., 6 reported a patient in whom a detached mini-balloon migrated from the ICA to the MCA, occluding the entire MCA main trunk as in our case. Their patient was also managed by embolectomy and recovered over a 2month period. The detachable balloon technique that we use consists of ligating a small latex cover over a small tubing J. Neurosurg. / Volume 6 4 / February, 1986 Detachable balloon MCA embolism FIG. 2. Arteriograms,oblique (left) and lateral (right) views,obtainedabout 1 yearafter embolectomyshowing the aneurysm unchanged in size. The middle cerebral artery is completelynormal. with a latex thread. With experience, this ligature is strong enough to avoid rapid leakage from the balloon after detachment. Rapid deflation of the balloon, as occurred in this case, is probably due to some degree of fissuring or rupture of the balloon. Also, whenever the carotid artery is permanently occluded, we detach a second balloon below the first one. If the first balloon becomes deflated after detachment of the second balloon, no harm results. But in this case, the first balloon burst and deflated before we had time to detach the second balloon. We meticulously examined the balloon after removal, but were unable to demonstrate any imperfection at the level of the ligature; however, the balloon had a scalpel slit made at the time of surgery in order to deflate it for easier removal. Several points with regard to the management of our patient are worthy of emphasis. Timing In spite of rapid recognition of the neurological deficit, expeditious preparation for surgery, and rapid reopening of a recent craniotomy, 3 hours elapsed from occurrence of the embolism to reestablishment of flow. Similarly, 2 89 hours expired prior to embolectomy in the patient reported by Chalif, et al., 2 and 5 hours in the patient reported by Langford, et al. 6 Some of this time is, by necessity, spent in the neuroradiology suite performing the arteriography necessary to demonstrate the site of embolic occlusion. Although experience in the laboratory has shown that clinical deficits are not solely dependent upon duration of ischemia, but also relate to the degree of ischemia, 5 rapid reestablishment of J. Neurosurg. / Volume 6 4 / February, 1986 flow is certainly necessary to yield optimal results. At present, it appears logistically difficult to reestablish flow by embolectomy sooner than about 3 hours from the time of occlusion. Collateral Circulation The degree of ischemia in embolic arterial occlusion is dependent largely on collateral flow. Meyer, et al., 7 found that, in patients with acute MCA occlusion, the degree of collateral circulation was a more important factor in determining outcome than the duration of arterial occlusion. Vigorous back-bleeding from small perforating arteries near the MCA bifurcation suggested that our patient's collateral circulation helped to maintain flow above a critical level, thereby allowing subsequent rapid recovery after restoration of normal flow. In the case reported by Langford, et al.,6 backflow from the perforating arteries was described as "poor," and this may have been one of the reasons why recovery in their patient was slower than in ours. It could be speculated that the collateral flow in our patient was good enough, and she may have recovered anyway, even without embolectomy. The fact that she remained completely hemiplegic from the time of the embolic event to the time of induction of anesthesia (a period of about 1 89hours), however, suggests that a substantial neurological deficit would have remained had flow not been reestablished. Volume expansion and induced hypertension are known to be of value in managing ischemic deficits due to a number of neurovascular lesions. 4'8 As soon as the cause of the neurological deterioration was discovered 311 K. Swann, et al. to be balloon embolism, our patient was given an intravenous bolus of 5% albumin solution, 0.45% normal saline was administered at a rapid rate, and a NeoSynephrine drip was utilized to elevate the systolic blood pressure. Barbiturates were not used until the time of induction of anesthesia because of the logistics of being in the neuroradiology suite and our choice of deferring endotracheal intubation until the patient was in the operating room ready for surgery. Langford, et al., 6 kept their patient in "barbiturate coma" for 24 hours after surgery; whether this helped in the ultimate recovery of their patient remains unanswered. Anticoagulation Therapy Anticoagulation therapy has been thought to prevent thromboembolic complications during intra-arterial catheter manipulation. ~ Therefore, our patient received 5000 U of heparin intravenously during the embolization procedure. Under ordinary circumstances, the anticoagulation agent used during these procedures is pharmacologically reversed with protamine sulfate. However, after recognition of the balloon embolism to the MCA, we did not reverse the heparin; an additional 1000 U of heparin was given intravenously during induction of anesthesia. We kept the patient anticoagulated throughout surgery until the time of flow reestablishment in an effort to prevent secondary thromboembolism and to facilitate microcirculation distal to the balloon occlusion. We realize that this was a risky decision based on "instinct" rather than on scientific data and, therefore, even though it may have helped in our case, we cannot recommend anticoagulation under these circumstances without reservations. If the patient is kept anticoagulated it is probably safer to leave the catheter introducer in the groin and remove it after surgery, when the anticoagulation effect is reversed, as we did in our case. Surgical Technique A standard pterional craniotomy had been used to explore the aneurysm during the initial clipping attempt. The approach proved very adequate to achieve exposure of the MCA. The balloon could be seen through the arterial wall, and an arteriotomy was made directly over it. The balloon was long enough to occupy almost the entire MCA trunk and, therefore, we had to place temporary clips proximally at the origin of the MCA and distally in the divisions, since obviously the clips could not be placed over the balloon. Puncturing the balloon with a scalpel reduced its size and aided its extraction from the lumen. We were surprised at the vigor of the "backflow" bleeding that occurred from the perforating arteries when the balloon was removed. We had not anticipated this and had not occluded those 312 vessels. We were then forced to place a t e m p o r a r y clip on each of these fragile arteries under less than optimal conditions. If faced with the same situation again, we would temporarily occlude each of these vessels under controlled conditions prior to arteriotomy. Recovery Resolution of symptoms began soon after the patient awakened from anesthesia, but the next morning (about 12 hours after embolectomy) she still had a moderately severe left hemiparesis. Over the next 24 hours, however, she recovered completely. We find it interesting that recovery occurred gradually rather than abruptly after flow reestablishment. This indicates that the degree of ischemia she suffered, although not severe enough to produce irreversible cell death, did result in significant, temporary impairment of neuronal metabolism. References I. Barrow DL, Heischer AS, Hoffman JC: Complications of detachable balloon catheter technique in the treatment of traumatic intracranial arteriovenous fistulas. J Neurosurg 56:396-403, 1982 2. Chalif DJ, Harem ES, Berenstein A, et al: Microsurgical removal of a balloon embolus to the internal carotid artery. Case report. J Neurosurg 48:112-116, 1983 3. Debrun G, Fox A, Drake C, et al: Giant unclippable aneurysms: treatment with detachable balloons. AJNR 2: 167-173, 1981 4. Farhat SM, Schneider RC: Observations on the effect of systemic blood pressure on the intracranial circulation in patients with cerebrovascular insufficiency. J Neurosurg 27:441-445, 1967 5. Jones TH, Morawetz RB, Crowell RM, et al: Thresholds of focal cerebral ischemia in awake monkeys. J Neurosurg 54:773-782, 1981 6. Langford KH, Vitek J J, Zeiger E: Migration of detachable mini-balloon from the ICA causing occlusion of the MCA. Case report. J Neurosurg 58:430-434, 1983 7. Meyer FB, Piepgras DG, Sundt TM Jr, et al: Emergency embolectomy for acute occlusion of the middle cerebral artery. J Neurosurg 62:639-647, 1985 8. Pritz MB, Giannotta SL, Kindt GW, et al: Treatment of patients with neurological deficits associated with cerebral vasospasm by intravascular volume expansion. Neurosurgery 3:364-368, 1978 9. Serbinenko FA: Balloon catheterization and occlusion of major cerebral vessels. J Neurosurg 41:125-145, 1974 10. Welch K: Excision of occlusive lesions of the middle cerebral artery. J Neurosurg 13:73-80, 1956 Manuscript received March 6, 1985. Accepted in final form July 2, 1985. Present address for Dr. Nelson: Neurosurgical Service, Strong Memorial Hospital, Rochester, New York. Address reprint requests to." Roberto C. Heros, M.D., Director of Cerebrovascular Surgery, Massachusetts General Hospital, Boston, Massachusetts 02114. J. Neurosurg. / Volume 64/ February, 1986