点击显示 收起
the Department of Neurology (T.K., G.D., B.P.-J., A.D., J.B.), Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland
Department of Diagnostic and Interventional Radiology (R.M.) and Department of Cardiovascular Surgery (P.R., L.V.S.), CHUV, Lausanne, Switzerland.
Abstract
Background and Purpose— Hyperperfusion syndrome (HS) after carotid endarterectomy (CEA) has been related to impaired cerebrovascular autoregulation in a chronically hypoperfused hemisphere. Our aim was to provide new insight into the pathophysiology of the HS using magnetic resonance imaging (MRI) studies with diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI).
Methods— Five out of 388 consecutive patients presented 2 to 7 days after CEA, partial seizures (n=5), focal deficits (n=5), and intracerebral hemorrhage (n=3). In 4 patients, using sequential examinations, we identified vasogenic or cytotoxic edema by DWI; we assessed relative interhemispheric difference (RID) of cerebral blood flow (CBF) by PWI; and we measured middle cerebral artery mean flow velocities (MCA Vm) by transcranial Doppler (TCD).
Results— None of the patients presented pathological DWI hyperintensities, consistent with the absence of acute ischemia or cytotoxic edema. In 2 patients, we found an MRI pattern of reversible vasogenic edema similar to that observed in the posterior leukoencephalopathy syndrome. Middle cerebral artery (MCA) mean flow velocities (Vm) were not abnormally increased at any time. PWI documented a 20% to 44% RID of CBF in favor of the ipsilateral to CEA hemisphere.
Conclusions— HS can occur in the presence of moderate relative hyperperfusion of the ipsilateral hemisphere. MCA Vm values may not accurately reflect RID of CBF over the cortical convexity. We suggest that the hemodynamic pathogenetic mechanisms of the HS are more complicated than hitherto believed and that they may be more accurately described by the term "reperfusion syndrome."
Key Words: carotid endarterectomy reperfusion ultrasonography, Doppler, transcranial
Introduction
Hyperperfusion syndrome (HS) after carotid endarterectomy (CEA) consists of a classical clinical triad: transient focal deficits associated with ipsilateral migraine-like headache, seizures, and intracerebral hemorrhage (ICH).1 The pathophysiology of the syndrome is still obscure, being probably related to the preoperative loss of vascular autoregulatory mechanisms in a chronically hypoperfused hemisphere. Although a degree of hyperperfusion seems to occur in most patients in the immediate postoperative period, florid HS develops in only a few.2–4 Reversible ipsilateral brain edema evolving in patients after CEA has only recently been recognized,5 and its pathophysiology is indeterminate. Hitherto, diffusion-weighted imaging (DWI) has been used in very few patients with HS after CEA,6 and perfusion-weighted imaging (PWI) has not been applied to the investigation of the HS. A study7 evaluated CEA with sequential PWI, but none of the patients had HS.
We now report 5 patients with HS, 4 of whom were systematically studied with sequential computed tomography (CT), standard magnetic resonance imaging (MRI), DWI, and PWI, and transcranial Doppler (TCD). We suggest that this post-CEA complication, at least in some patients, may result from interhemispheric asymmetries of perfusion and in the absence of striking absolute hyperperfusion.
Subjects and Methods
Patients
Three-hundred eighty-eight consecutive patients underwent CEA at our center by 2 experienced vascular surgeons. CEAs were routinely performed under loco-regional anesthesia without shunting. If neurological deterioration was observed during clamping of the carotid, shunting was performed (Figures 1–5 and Figure I available online only at http://www.strokeaha.org).
Data Acquisition and Analysis
TCD was performed using a Multi-Dop T2 instrument (DWL) with a 2-Mhz pulsed-wave transducer. The middle cerebral arteries (MCAs) were insonated using the temporal bone window at a depth of 50 to 60 mm, where the Doppler signal was mostly accurate and audible. The Pulsating Index (PI) quantifies the shape of the TCD waveform and is inversely correlated to cerebral vascular resistance. Using MCA mean flow velocity (Vm), we evaluated the relative interhemispheric difference (RID) as follows: RID (%)=100 x (Vm MCA ipsilateral to CEA – Vm MCA contralateral to CEA)/Vm MCA contralateral to CEA.
MRI was performed on a 1.5-T Magnetom Vision MRI unit (Siemens). After routine T1 (repetition time [TR]=510, echo time [TE]=12) and fast spin-echo T2 (TR=3800, TE=90) sequences, we obtained an echoplanar single-shot spin-echo DWI trace sequence (TR=5000 ms, TE=100 ms, b=1000 sec/mm2, 20 5-mm-thick slices with 1.5-mm gap, matrix size 128x128). PWI series was performed immediately afterward (T2 echoplanar gradient echo, TR=120 ms, TE=45 ms, 5 5-mm-thick slices with 1.5-mm gap, matrix size 128x128; and 40 acquisitions obtained every 2 seconds during intravenous administration of 0.2 mmol/kg Gd-DTPA; Omniscan, Nycomed Amersham), followed by a 15-mL saline flush at a rate of 5 mL/second into an antecubital vein with a power injector (Spectris MR Injection System; Medrad). Perfusion maps were constructed by a pixel-by-pixel analysis of the time series. The maximum slope of decrease, which is associated with the relative cerebral blood flow (rCBF), was computed. Regions of interest were chosen in comparable contralateral territories of the MCA and signal-to-time change curves were plotted. The RID of rCBF in the regions of interest was calculated according to the formula RID (%)=100 x (rCBF ipsilateral to CEA – rCBF contralateral to CEA)/rCBF contralateral to CEA. Values of rCBF in the specified regions of interest were measured in arbitrary MRI units.
Results
Two to 7 days postoperatively, focal neurological deficits developed in 5 patients, accompanied in all by partial seizures that were secondarily generalized in patients 3 and 4. Severe to mild headache ipsilateral to the operated carotid preceded the onset of symptoms by 3 hours to 6 days in patients 4 and 5. Five to 8 days post-CEA, a >2-cm in diameter ICH evolved in patients 1, 3, and 5, whereas cortical petechial hemorrhages evolved in patient 4. The study of patient 1 had been published before as a case report.8 All 5 patients had preoperative intra-arterial 4-vessel angiography and brain CT. Extracranial Doppler and TCD were performed within 1 week before surgery in all patients. Preoperative MRI was available in patients 1 through 3. Within 24 hours after the onset of symptoms of "hyperperfusion," we performed brain CT, electroencephalogram, extracranial Doppler, and TCD in all patients. Patients 1 through 4 had sequential MRI studies with DWI and magnetic resonance angiography (Table 1). Sequential PWI studies were performed in patients 1, 2, and 4. The baseline characteristics of the patients in whom HS developed compared with those who did not are presented in Table I (available at http://www.strokeaha.org).
Clinical Data of Patients 1 Through 4
Discussion
Hyperperfusion after CEA is classically defined as >100% increase in CBF compared with preoperative values.1,4,9 However, in a study using the intraoperative 133Xe technique,10 one-third of the patients who presented headache and seizures postoperatively had only modest CBF increases (27% to 36%). Using the same technique, another study9 could not document striking hyperperfusion in 5 of 14 patients in whom ICH developed after CEA. In a 133Xe study,11 half of the patients in whom ICH developed after CEA had <30% increase in CBF compared with preoperative values. Conversely, in recent prospective SPECT studies,4,12 all 4 patients who presented clinically HS showed >100% CBF increase.
To the best of our knowledge, this is the first PWI study in patients presenting symptoms of "hyperperfusion" after CEA. A previous MRI perfusion study7 found early bolus arrival time on the ipsilateral hemisphere in 4 of 7 patients who were asymptomatic after CEA, confirming the common occurrence of the reperfusion hyperemic response. In the present study, the relative hemispheric hyperperfusion ipsilateral to CEA in patients 1 to 4, as evidenced by PWI, was only moderate, ranging from 20% to 44%. Because PWI is not a quantitative method and preoperative PWI studies were not performed, we cannot draw conclusions about absolute CBF differences. However, these 4 patients had relatively unchanged and normal ipsilateral MCA Vm values early after the onset of symptoms. This observation raises the question whether the so-called hyperperfusion symptoms can occur in the absence of excessively increased absolute CBF values. Even if relative CBF increases are more important, the values of reference remain to be defined: preoperative and ipsilateral to the CEA, postoperative and contralateral to the CEA, or both
TCD has been extensively used to investigate the time course of cerebral hemodynamics during and after CEA3 but has a major limitation. Because TCD measures velocity and not flow, there exists a good correlation between alterations in measured velocity and changes in regional CBF, provided only that the diameter of the studied artery remains constant.13 However, changes in MCA diameter during the course of the HS are unknown. A single study14 has addressed the issue of regional CBF during CEA, correlating TCD-measured velocities with the intra-arterial 133Xe technique. The authors proposed that MCA Vm reflects hemodynamic events in the territory of the lenticulostriate arteries far better than in the cerebral cortex. Our results suggest that MCA Vm values either underestimate or overestimate the relative interhemispheric CBF differences over the cortical convexity.
Focal neurological deficits in patients who are initially well after CEA may result from brain ischemia caused by carotid occlusion or embolization from the endarterectomy site. DWI can differentiate cytotoxic from vasogenic edema. The former is characterized by DWI hyperintensity and a decline of the apparent diffusion coefficient (ADC) of water, whereas the latter is characterized by relative increase in the diffusion of water molecules and only a variable weak change in DWI signal intensity.6,15 In our study, the absence of pathological DWI hyperintensity in the hemisphere ipsilateral to the operated carotid, at least 6 hours after the onset of the hyperperfusion symptoms, suggests a nonsignificantly lowered ADC, consistent with the absence of acute cerebral infarction or cytotoxic edema. Furthermore, PWI did not demonstrate a perfusion deficit corresponding to the recent T2 hyperintensities, but on the contrary documented a mild to moderate relative hyperperfusion. Therefore, in patients 1 to 4, we cannot attribute the development of focal neurological deficits and seizures after CEA to cerebral ischemia. The extended subcortical T2 hyperintensities in patients 1 and 2, which implied white matter edema, were characterized by isointensity on DWI, suggesting normal or increased diffusion of water, that is, vasogenic edema in accordance with previous studies.6 Unfortunately, ADC maps were not technically available during the period of our study to document increased ADC in the edematous brain regions.
A DWI study of moderately hypertensive patients with posterior leukoencephalopathy syndrome (PLES) showed a similar to our study pattern of vasogenic edema.15 In patient 1, extracellular edema also affected the contralateral to CEA occipital white matter, documenting that an identical MRI pattern may be encountered in both the HS and the PLES, and confirming that CEA influences bilaterally cerebral hemodynamics.3,12 Some authors suggest that blood pressure (BP) elevations and hyperperfusion are interlinked and may act synergistically as triggers for a subsequent pathogenetic cascade.16 All our patients in whom HS developed had a history of hypertension, and all had higher than normal BP during the symptoms of hyperperfusion, although in patients 2 and 4 BP increases were not striking. It is known that cerebral hyperperfusion can occur even in patients with stable and normalized blood pressure after surgery;1,4,17 however, cerebral hyperperfusion in the presence of BP elevations is the rule. Continuous postoperative vigilance and meticulous BP control, even in normotensive patients, is of outmost importance because CBF is pressure-dependent during the period of postoperative hyperperfusion caused by defective cerebrovascular autoregulation.2,12 There is evidence that BP reduction can control symptoms and decrease MCA Vm ipsilateral to CEA, although the interaction between BP and MCA Vm is complex.2 In a series,2 there were patients with episodes of hypertension (>170 mm Hg systolic BP) but without symptoms of hyperperfusion and other patients with less elevated, yet abnormal BP values, in whom symptoms developed. Pharmacological control of BP did not influence MCA Vm in the former, but decreased ipsilateral MCA Vm in the latter. Strict BP control in selected patients using intra-operative TCD criteria has been associated with decreased rate of ICH after CEA.18 The "normal perfusion pressure breakthrough" hypothesis19 could explain the development of a bilateral (PLES) or unilateral (HS) "hypertensive encephalopathy" in the absence of severe hypertension. In a rat model of chronic cerebral hypoperfusion,20 it has been documented a >100% increase in capillary density, with some of these capillaries becoming structurally defective in their walls. These capillaries are mechanically weaker and consequently more vulnerable to breakthrough by the distending forces that occur on reperfusion. Furthermore, using the same experimental model, it has been shown that chronic hypoperfusion may be responsible for a reduction in the threshold for hypertensive breakthrough.21 Whether these changes occur in humans is unknown. However, if these were true, and in combination with the loss of the appropriate vasoconstrictive response, they would account for the post-CEA vasogenic edema even in patients without severe hypertension but with relative PWI-documented hyperperfusion.
Summary
After CEA, partial seizures, focal neurological deficits with or without headache, and intracerebral hematomas can occur in the presence of relative hyperperfusion of the ipsilateral hemisphere, ranging from 20% to 44%. In some patients, this is accompanied by TCD evidence not suggesting absolute hyperperfusion. MCA Vm values may not accurately reflect relative interhemispheric CBF differences over the cortical convexity. Strict postoperative BP control is of paramount importance. The edema associated with the HS is reversible and demonstrates a vasogenic MRI pattern similar to that of the PLES. Finally, we suggest that reperfusion, rather than hyperperfusion, is the term that may better-encompass the whole of the hemodynamic and neuroeffector pathogenetic mechanisms implicated in this rare but always intriguing post-CEA complication.
Acknowledgments
T.K. received a grant from the "Empeirikeion Foundation," Athens, Greece.
References
Sundt TM, Sharbrough FW, Piepgras DG, Kearns TP, Messick JM, O‘Fallon WM. Correlation of cerebral blood flow and electroencephalographic changes during carotid endarterectomy. With results of surgery and hemodynamics of cerebral ischemia. Mayo Clin Proc. 1981; 56: 533–543.
Jorgensen LG, Schroeder TV. Defective cerebrovascular autoregulation after carotid endarterectomy. Eur J Vasc Surg. 1993; 7: 370–379.
Araki CT, Babikian VL, Cantelmo NL, Johnson WC. Cerebrovascular hemodynamic changes associated with carotid endarterectomy. J Vasc Surg. 1991; 13: 854–860.
Ogasawara K, Yukawa H, Kobayashi M, Mikami C, Konno H, Terasaki K, Inoue T, Ogawa A. Prediction and monitoring of cerebral hyperperfusion after carotid endarterectomy by using single-photon emission computerized tomography scanning. J Neurosurg. 2003; 99: 504–510.
Breen JC, Caplan LR, DeWitt LD, Belkin M, Mackey WC, O’Donnell TP. Brain edema after carotid surgery. Neurology. 1996; 46: 175–181.
Schaefer PW, Gonzalez RG, Hunter G, Wang B, Koroshetz WJ, Schwamm LH. Diagnostic value of apparent diffusion coefficient hyperintensity in selected patients with acute neurologic deficits. J Neuroimaging. 2001; 11: 369–380.
Gilliard JH, Hardingham CR, Kirkpatrick PJ, Antoun NM, Freer CEL, Griffiths PD. Evaluation of carotid endarterectomy with sequential MR perfusion imaging: a preliminary report. AJNR. 1998; 19: 1747–1752.
Viana Baptista M, Maeder P, Dewarrat A, Bogousslavsky J. Conflicting images. Case report. Lancet. 1998; 351: 414.
Piepgras DG, Morgan MK, Sundt TM Jr., Yanagihara T, Mussman LM. Intracerebral hemorrhage after carotid endarterectomy. J Neurosurg. 1988; 68: 532–536.
Reigel MM, Hollier LH, Sundt TM, Piepgras DG, Sharbrough FW, Cherry KJ. Cerebral hyperperfusion syndrome: a cause of neurologic dysfunction after carotid endarterectomy. J Vasc Surg. 1987; 5: 628–634.
Schroeder T, Sillesen H, Boesen J, Laursen H, Sorensen P. Intracerebral haemorrhage after carotid endarterectomy. Eur J Vasc Surg. 1987; 1: 51–60.
Hosoda K, Kawaguchi T, Shibata Y, Kamei M, Kidoguchi K, Koyama J, Fujita S, Tamaki N. Cerebral vasoreactivity and internal carotid artery flow help to identify patients at risk for hyperperfusion after carotid endarterectomy. Stroke. 2001; 32: 1567–1573.
Bishop CC, Powell S, Rutt D, Browse NL. Transcranial Doppler measurement of middle cerebral artery blood flow velocity: a validation study. Stroke. 1986; 17: 913–915.
Halsey JH, McDowell HA, Gelmon S, Morawetz RB. Blood velocity in the middle cerebral artery and regional cerebral blood flow during carotid endarterectomy. Stroke. 1989; 20: 53–58.
Ay H, Buonanno FS, Schaefer PW, Le DA, Wang B, Gonzalez RG, Koroshetz WJ. Posterior leukoencephalopathy without severe hypertension: utility of diffusion-weighted MRI. Neurology. 1998; 51: 1369–1376.
Naylor AR, Evans J, Thompson MM, London NJ, Abbott RJ, Cherryman G, Bell PR. Seizures after carotid endarterectomy: hyperperfusion, dysautoregulation or hypertensive encephalopathy Eur J Vasc Endovasc Surg. 2003; 26: 39–44.
Ascher E, Markevich N, Schutzer RW, Kallakuri S, Jacob T, Hingorani AP. Cerebral hyperperfusion syndrome after carotid endarterectomy: predictive factors and hemodynamic changes. J Vasc Surg. 2003; 37: 769–777.
Dalman JE, Beenakkers ICM, Moll FL, Leusink JA, Ackerstaff RGA. Transcranial Doppler monitoring during carotid endarterectomy helps to identify patients at risk of postoperative hyperperfusion. Eur J Vasc Endovasc Surg. 1999; 18: 222–227.
Spetzler RF, Wilson CB, Weinstein P, Mehdorn M, Townsend J, Telles D. Normal perfusion pressure breakthrough theory. Clin Neurosurg. 1978; 25: 651–672.
Sekhon LHS, Morgan MK, Spence I. Normal perfusion pressure breakthrough: the role of capillaries. J Neurosurg. 1997; 86: 519–524.
Morgan MK, Johnston I, Besser M, Baines D. Cerebral arteriovenous malformations, steal, and the hypertensive breakthrough threshold. An experimental study in rats. J Neurosurg. 1987; 66: 563–567.