Ischemia Induced Hypervolemia and Inotropic Support for Acute Cerebral Arterial Insufficiency: An Underused Therapy B.J. Duke, M.D.,* R.E. Breeze, M.D.,* D. Rubenstein, M.D.,† B.I. Tranmer, M.D.,‡ and G.W. Kindt, M.D.* Divisions of *Neurosurgery and †Neuroradiology, University of Colorado Health Sciences Center, Denver, Colorado, and ‡Division of Neurosurgery, Albany Medical College, Albany, New York Duke BJ, Breeze RE, Rubenstein D, Tranmer BI, Kindt GW. Induced hypervolemia and inotropic support for acute cerebral arterial insufficiency: an underused therapy. Surg Neurol 1998;49:51–7. BACKGROUND Hypervolemia and induced systemic hypertension are generally considered the standard approach to the treatment of vasospasm. Despite evidence in favor of its efficacy, this therapy is used rarely in acute cerebrovascular occlusion. We present a case supporting this treatment paradigm. have an independent salutatory effect on vasospasm. Most neurosurgeons now use both hypertension and hypervolemia in the treatment of vasospasm. Hyperdynamic therapy is infrequently used to treat acute arterial occlusion, however. Recently we encountered angiographic evidence supporting the use of hypervolemia and beta-adrenergic inotropic support in the treatment of embolic stroke. CASE DESCRIPTION A patient developed aphasia and hemiplegia 8 h after carotid endarterectomy caused by embolic occlusion of the middle cerebral artery. Hyperdynamic/hypervolemic therapy was instituted. Serial angiograms filmed over the next 8 h demonstrated reperfusion of the hemisphere through collateral flow. The patient’s symptoms resolved. CONCLUSIONS We believe this case demonstrates the effectiveness of hypervolemia and inotropic support in the treatment of acute embolic stroke by inducing dilatation of the leptomeningeal collateral circulation. © 1998 by Elsevier Science Inc. KEY WORDS Cerebral ischemia, middle cerebral artery, volume expansion, cardiac output. or the past 20 years systemic hypertension has been used to treat acute cerebral ischemia secondary to cerebral vasospasm. More recently intravascular volume expansion has been found to F Address reprint requests to: Dr. Robert E. Breeze, UCHSC, 4200 E. Ninth Ave., Box C-307, Denver, CO 80262. Received August 14, 1996; accepted May 30, 1997. © 1998 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010 Case Report A 61-year-old white male presented with transient ischemic attacks while on medical therapy and was found to have a 50% stenosis of the left internal carotid artery (ICA) secondary to an ulcerated plaque [by North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria—the patient was not a NASCET enrollee]. He underwent carotid endarterectomy without difficulty (Figure 1). Eight hours postoperatively the patient suddenly became aphasic and hemiplegic. The patient was returned immediately to the operating room, the arterotomy re-opened, and ante- and retrograde flow demonstrated. An immediate angiogram revealed irregularity at the endarterectomy site but disclosed an obstruction at the bifurcation of the left ICA and proximal M1 with no orthograde flow in the middle cerebral artery (MCA) (Figure 2). A bolus of 125,000 units of urokinase was infused into the petrous portion of the left ICA and a continuous drip of 60,000 units per hour was begun through an indwelling catheter left in this location. A Swan-Ganz pulmonary arterial catheter was in0090-3019/98/$19.00 PII S0090-3019(97)00353-4 52 1 Surg Neurol 1998;49:51–7 Duke et al AP (A) and lateral (B) views of the pre-operative left common carotid angiogram show plaque at the bifurcation and a normal internal carotid artery bifurcation. serted. The baseline cardiac output (CO) was 4.5 liter/minute and the pulmonary arterial wedge pressure (PAWP) was 7 mm Hg. Volume resuscitation and a continuous dobutamine infusion were started. At 1 h after the urokinase bolus a second angiogram was obtained, which revealed minimal change in the M1 clot, but that demonstrated vessels in the left MCA territory beginning to fill via opercular collaterals primarily from the anterior cerebral artery with additional surface collateral perfusion from the left posterior cerebral artery (PCA) (Figure 3). At this point the patient remained aphasic and hemiplegic. A third angiogram performed 2.5 h after the urokinase bolus again revealed little change in the M1 clot but revealed further collateralization from the ACA and PCA (Figure 4). By this time the patient had received three units of packed red blood cells (1,500 cc) and 300 cc of 25% albumin (75 g), and was receiving 10 mcg/ kg/min of dobutamine by continuous infusion. The CO had increased to 9.7 liter/minute, the PAWP had increased to 16 mm Hg, and the mean arterial pressure remained unchanged from pretreatment levels (systolic pressure between 110 and 120). At this time the patient was naming objects, following commands with all extremities and had significantly improved motor function in the right upper and lower extremities. The M1 remained thrombosed (see Figure 3). The patient was treated with dobutamine (titrated between 5 and 20 mcg/kg/min to keep the CO greater than 8.0 liter/minute) and volume infusion (5% albumin infusion to keep PAWP greater than 12 mm Hg) for 3 days and was subsequently discharged with a subtle expressive aphasia and an intact motor exam. At the 6 months follow-up the patient had returned to work as a salesman, was no longer aphasic, and had a normal neurologic examination. Hemodynamic Therapy for Acute Ischemia The lateral view (A) of the initial postoperative left common carotid angiogram demonstrates a filling defect at the internal carotid artery bifurcation (open arrow) and no opacification of the proximal middle cerebral artery. The lateral view (B) in the venous phase demonstrates middle cerebral artery branches (arrowheads) that have filled by retrograde flow via pial collaterals from anterior cerebral artery branches. 2 Discussion Hypervolemic, hyperdynamic therapy has been extremely successful in the treatment of vasospasm after subarachnoid hemorrhage [1,4]. The application of this therapy to acute cerebrovascular occlusive disease remains controversial despite experimental and clinical evidence of its effectiveness. The rationale for the use of hypervolemic, hyperdynamic therapy for acute ischemic stroke is based on observations that in ischemic brain, local cerebral blood flow (CBF) becomes disautoregulated such that the CBF varies directly with the cardiac output [5,7,10]. Thus, the CBF to an ischemic area of brain can be increased by augmenting stroke volume by optimizing preload and Surg Neurol 1998;49:51–7 53 the inotropic state of the heart. The anatomy of the leptomeningeal collateral circulation has been well described in humans, and as seen in this case, provides a route for supplemental blood flow to enter the ischemic region [12]. Theoretically this should aid in decreasing the size of the final infarction and would be expected to improve outcome. Hypervolemic therapy for acute cerebrovascular insufficiency has been evaluated in several animal models and in clinical trials. It has been shown that colloid (Hetastarch) infusion in a primate model of MCA occlusion increases cerebral blood flow and improves EEG amplitudes in ischemic brain [11]. Albumin infusion after MCA occlusion in rats has been shown to decrease infarction size and associated edema [6]. Clinical trials in humans have focused not on hypervolemic treatment but on isovolemic hemodilution. The Scandinavian Stroke Study Group and the Italian Acute Stroke Study Group found no significant improvement in morbidity or decrease in mortality in patients treated with isovolemic hemodilution [3,8]. These studies emphasized decreasing the hematocrit by venesection with isovolemic colloid replacement and did not include hemodynamic criteria as therapeutic endpoints. Inasmuch as many stroke patients are hypervolemic on presentation [2], the patients in these studies may have been hemodiluted but under-resuscitated. The focus of these studies has been on the rheologic benefits of hemodilution and not on the effects of hypervolemia on stroke volume and cardiac output. Although it has been demonstrated that local cerebral blood flow varies inversely with blood viscosity [13], the clinical magnitude of this effect is unclear. A more recent randomized controlled trial at a single center using invasive hemodynamic monitoring and normovolemic crystalloid and albumin hemodilution showed a significant decrease in mortality at 3 months and an increase in functional status in the treatment group [2]. This study varied from the previous studies in that the treatment endpoint criteria included hemodynamic parameters and not hemodilutional parameters alone. In a previous report from this institution it was noted that intravascular volume expansion using invasive hemodynamic monitoring resulted in dramatic reversal of neurologic deficit in three of five consecutive patients with completed tomography and/or angiographic evidence of an acute MCA occlusion [9]. We believe that this case demonstrated the utility of combining hypervolemic and hyperdynamic (inotropic) therapy in the treatment of acute 54 Surg Neurol 1998;49:51–7 Duke et al Lateral views of the left common carotid angiogram performed 1 h after the angiogram in Figure 2 show the same middle cerebral branches (arrowheads) now fill in the capillary phase (A) and new branches (arrows) now fill via pial collaterals from posterior cerebral artery branches. Further retrograde filling of middle cerebral artery branches is seen in the venous phase (B). 3 cerebrovascular occlusion and that this therapy warrants further study. REFERENCES 1. Giannotta SL, McGillicuddy JE, Kindt GW. Diagnosis and treatment of postoperative cerebral vasospasm. Surg Neurol 1977;8:286 –90. 2. Goslinga H, Eijzenbach V, Heuvelmans JHA, ven der Laan de Vries E, Melis VMJ, Schmid-Schonbein H, Bezemer PD. Custom-tailored hemodilution with albumin and crystalloids in acute ischemic stroke. Stroke 1992;23:181– 8. 3. Italian Stroke Study Group. Hemodilution in acute stroke. Results of the Italian hemodilution trial. Lancet 1988;1:318 –21. 4. Kassell NF, Peerless SJ, Durward QJ, Beck DW, Drake CG, Adams HP. Treatment of ischemic deficits from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery 1982;11: 337– 43. 5. Levy ML, Rabb CH, Zelman V, Giannotta SL. Cardiac performance enhancement from dobutamine in patients refractory to hypervolemic therapy for cerebral vasospasm. J Neurosurg 1993;79:494 –9. 6. Matsui T, Sinyama H, Ansano T. Beneficial effect of prolonged administration of albumin on ischemic cerebral edema and infarction after occlusion of middle cerebral artery in rats. Neurosurgery 1993;33:293– 300. 7. Ohtaki M, Tranmer B. Role of hypervolemic hemodilution in focal cerebral ischemia of rats. Surg Neurol 1993;40:196 –206. 8. Scandinavian Stroke Study Group. Multicenter trial of hemodilution in acute ischemic stroke. I. Results in the total patient population. Stroke 1987;18:691–9. 9. Tranmer BI, Gross CE, Keller TS, Kindt GW. Acute middle cerebral artery occlusion: experience with volume expansion therapy. Neurosurgery 1986;18: 397– 401. 10. Tranmer BI, Keller TS, Kindt GW, Archer D. Loss of cerebral regulation during cardiac output variations in focal cerebral ischemia. J Neurosurg 1992;77:253–9. 11. Tranmer BI, Keller TS, Nagata K, Kindt GW, Adey GR. Blood volume expansion with hetastarch in acute ischaemic stroke: the effects on local cerebral blood flow and computer mapped EEG. Neurol Res 1986;8: 177– 82. 12. Vander Eeeken HM, Adams RD. The anatomy and the functional significance of the meningeal arterial anastomoses of the human brain. J Neuropath Exp Neurol 1953;12:132–57. 13. Wood JH, Fleischer AS. Observations during hypervolemic hemodilution of patients with acute focal cerebral ischemia. JAMA 1982;248:2999 –3004. COMMENTARY Duke et al present an interesting case report in which a patient who had an embolic phenomenon, presumably from the operative site, 8 h after carotid endarterectomy was demonstrated to have a documented middle cerebral artery stem embolus; he made a nice recovery with the institution of hypervolemic treatment. The most striking contribution of this single report is the beautiful angio- Hemodynamic Therapy for Acute Ischemia Surg Neurol 1998;49:51–7 55 Lateral views of the left common carotid angiogram performed 90 min after the angiogram in Figure 3 show that middle cerebral artery branches now fill in the arterial phase (A). Continued retrograde filling of some middle cerebral artery branches and orthograde filling of others (long arrows) is observed in the early (B), middle (C), and late (D) venous phase. 4 graphic documentation of collateral filling that increased with the institution of the hypertensive hypervolemic treatment. The omissions that weaken this report are the lack of computed tomography or magnetic resonance imaging data documenting the time course of the ischemia, the lack of any cerebral blood flow studies, which certainly would support the authors’ theory, and the lack of follow-up with either postoperative angiography or postoperative TCD or MRA to indicate the current status of the middle cerebral artery occlusion. In the absence of these post-therapy studies we are not certain whether the patient’s recovery was caused by a natural abundance of collateral circulation enhanced by the therapy, or from recanalization of the middle cerebral artery. What we can take from this single report, however, is that the institution of hypervolemic hypertensive therapy in the 56 Surg Neurol 1998;49:51–7 Duke et al acute, dysautoregulated hemisphere immediately after middle cerebral embolus was effective in tiding the patient over through the critical period for ischemia. I think this is the authors’ point and the technique that they are trying to support. The ultimate recovery of this particular patient could be attributable, as mentioned, either to the resumption of autoregulation with the presence of a natural rich collateral circulation or to recanalization of the middle cerebral tree. I have been faced with a similar clinical situation within the last year in which the patient immediately after carotid endarterectomy was found to have a contralateral brachial paresis and on angiography was demonstrated to have a single M2 branch occlusion. This was dissolved within 2 h by superselective infusion of urokinase. This patient was likewise treated with hypertensive hypervolemic therapy but did not make a complete neurologic recovery and was left with a mild paresis. In summary, this case demonstrates an interesting application of the hypertensive hypervolemic therapy we usually reserve for postsubarachnoid hemorrhage cerebral vasospasm. The strongest point of this report is the angiographic documentation of increased collateral filling with institution of therapy, and for this the authors are to be commended. Christopher M. Loftus, M.D. Division of Neurosurgery University of Iowa Hospital Iowa City, Iowa This paper is well-written and contains some fresh ideas on the therapy of the acute stage of brain ischemia. The article reports the case of a 61-yearold white male, who presented with transient ischemic attacks and a 50% stenosis of the left internal carotid artery. A few hours after endarterectomy the patient developed a severe hemiplegia on the right side and became aphasic. Attempts to dissolve the clot that occluded the origin of the middle cerebral artery, as shown by angiography, failed. The patient remained hemiplegic and aphasic a couple of hours after the occurrence of the embolus; he then received 3 units of packed red blood cells (1500 cc) and 300 cc of 25% albumin (75 g) and 10 mcg/kg/min of dobutamine by continuous infusion. As a consequence the cardiac output increased to 9.7 liters/min and the pulmonary artery wedge pressure increased to 16 mmHg, whereas the mean arterial pressure remained unchanged (systolic pressure between 110 and 120). Shortly after this therapy was instituted, the pa- tient started to name objects, followed commands with all extremities, and showed significantly improved motor function in the right upper and lower extremities. The treatment with dobutamine and volume infusion was continued for 3 days and after a couple of days the patient was discharged with a subtle expressive aphasia and an intact motor exam. Six months later he had returned to work as a salesman, was no longer aphasic, and had a normal neurologic examination. In the discussion, the authors explain that they based this type of therapy on the successful treatment of vasospasm after subarachnoid hemorrhage with hypervolemic-hyperdynamic therapy. Moreover, in a substantial number of animal experiments the usefulness of this type of therapy for the treatment of brain ischemia caused by an occlusion of the one of the major brain arteries is suggested. The authors provide some sensible arguments to support their treatment strategy. They try to explain why a number of clinical trials of isovolemic hemodilution in patients did not show significant improvement in morbidity or decrease in mortality. In these studies, the hematocrit was lowered, but hemodynamic criteria as therapeutic endpoints were not included. The authors suggest that the patients in these studies may have been hemodiluted, but under-resuscitated. The study by Goslinga et al that was recently published demonstrated the effectiveness of invasive hemodynamic monitoring, normovolemic crystalloid, and other means of hemodilution in patients with brain ischemia [1]. The authors suggest that the difference in outcome of this study compared with the previous ones is that in the study of Goslinga the treatment endpoint criteria included hemodynamic parameters and not hemodilutional parameters alone. In 1986, the authors published a paper in Neurosurgery in which they showed the dramatic reversal of neurological deficit in 3 of 5 consecutive patients with an MCA occlusion [1]. They now suggest that the combination of hypervolemic therapy and hyperdynamic (inotropic) therapy in the treatment of acute cerebrovascular occlusion seems to be superior. Although this paper is well-written and a number of important observations are made and a very sensible discussion of the topic is presented, there still remains a serious problem: the whole theory is based on the case history of one patient, and it cannot be ruled out that the neurologic improvement a couple of hours after the occlusion of the MCA was not induced by the hypervolemichyperdynamic therapy, but was spontaneous in nature. The increase in leptomeningeal collateral cir- Hemodynamic Therapy for Acute Ischemia culation may be unrelated to the therapy and may simply be an expression of ongoing collateral compensation for the occlusion of the MCA. It would have been much more convincing if the hemodynamic therapy was interrupted for a while and a relapse in the neurologic condition was noted. The authors do not explain why, after 3 days of hyperdynamic-hypervolemic therapy, the treatment could be stopped without a recurrence of neurologic deficits. They probably hold the opinion that the collaterals were now so well developed in the meantime, that the MCA territory was perfused sufficiently. Another possibility, of course, is that in the meantime the MCA became recanalized, and therefore it is a pity that no information is presented about the brain circulation after the moment when the hypervolemic and hyperdynamic therapy was stopped. Altogether I think this is an interesting paper concerning a rather controversial topic. Although some serious criticisms can be made, the possible clinical implications of this article are notable. Prof. Dr. C.A.F. Tulleken Department of Neurosurgery Academisch Ziekenhuis Utrecht Utrecht, The Netherlands REFERENCES 1. Goslinga H, Eijzenbach V, Heuvelmans JHA, ven der Laan de Bries E, Melis VMJ, Schmid-Schonbein H, Bezemer PD. Custom-tailored hemodilution with albumin and crystalloids in acute ischemic stroke. Stroke 1992; 23:181– 8. 2. Tranmer BI, Gross CE, Keller TS, Kindt GW. Acute middle cerebral artery occlusion: experience with volume expansion therapy. Neurosurgery 1986;18:397– 401. This is an important paper which, although it consists only of a single case report, is a powerful reminder of the possibilities of hypervolemic treatment for conditions other than delayed vasospasm. The results in the patient presented are impressive, considering the severe deficit present before treatment of a major arterial occlusion; the improvement is unlikely to have been attributable to recan- Surg Neurol 1998;49:51–7 57 alization at the point of occlusion, because the obstruction of M1 persisted for at least a few hours, and was still present after the patient had started to improve. The authors have had considerable anecdotal experience with this type of treatment for acute ischemia, and make the valid point that earlier studies using hemodilution alone in an attempt to improve cerebral blood flow by reducing blood viscosity were largely unsuccessful. Raising blood volume and cardiac output is an apparently more effective way of opening leptomeningeal collaterals to bypass an obstruction. Both experimental [3] and human [1] studies have suggested a similar mechanism to partly explain the efficacy of calcium antagonists in subarachnoid hemorrhage. The Amsterdam Stroke Study quoted by the authors (their Reference 2), in which attention was paid to pulmonary wedge pressure as well as hematocrit, showed significantly better outcomes in treated patients. Maintaining wedge pressure and cardiac output or index seems to be more important, and more related to outcome, than the simpler measures of systemic blood pressure and central venous pressure, both in this situation and in the management of vasospasm after aneurysm rupture [2]. Further study and controlled trials of this treatment would certainly appear justified on the evidence presented in this paper. N.W.C. Dorsch, FRCS, FRACS Department of Surgery Westmead Hospital Sydney, Australia REFERENCES 1. Auer LM, Oberbauer RW, Schalk HV. Human pial vascular reactions to intravenous nimodipine infusion during EC-IC bypass surgery. Stroke 1983;14:210 –3. 2. Finn SS, Stephensen SA, Miller CA, Drobnich L, Hunt WE. Observations on the perioperative management of aneurysmal subarachnoid hemorrhage. J Neurosurg 1986;65:48 – 62. 3. Takayasu M, Bassett JE, Dacey RG Jr. Effects of calcium antagonists on intracerebral penetrating arterioles in rats. J Neurosurg 1988;69:104 –9.