вход по аккаунту


Dynamic [18F]fluorodeoxyglucose positron emission tomography and hypometabolic zones in seizures Reduced capillary influx.

код для вставкиСкачать
Dynamic ["F] Fluorodeoxyglucose Positron
Emission Tomography and Hypometabolic
Zones in Seizures: Reduced Capillary Influx
Eain M. Cornford, PhD,*t$ Manyee N. Gee, PhD,$ Barbara E. Swartz, MD, PhD,*t$
Mark A. Mandelkern, MD, PhD,§S William H. Blahd, MD,$ Elliot M. Landaw, MD, PhD,"
and Antonio V. Delgado-Escueta, MD*:'F$
We performed dynamic ['8F]fluorodeoxyglucose (["FJFDG) positron emission tomographic (PET) analyses in 8 patients.
Rate constants of influx (K,*),efflux (k2*),phosphorylation (k3*),and dephosphorylation (k4*)were derived for the
regions of interest (ROIs), which included (1) the hypometabolic epileptogenic regions and (2) the homologous regions
in the contralateral hemispheres. In addition, the four constants were determined from at least one clearly defined
(control) ROI from the same plane and its homologous contralateral ROI. Influx (K,*)in the epileptogenic region was
reduced in comparison with the contralateral ROI. Reductions in influx (K,*),which averaged 18 f 13% (mean -t SD),
["FIFDG phosphorylation (k3*)(25 -t 20%), and brain glucose utilization rates (26 f 10%) were observed in the
epileptogenic region. Reductions in efflux were not statistically significant (k2*= 13 f 28%) but were comparable in
magnitude to the average reduction in K,*. No ipsilateral versus contralateral differences were seen for any rate constants
measured outside the epileptogenic region. Influx correlated highly with phosphorylation in the epileptogenic region.
Our data suggest that the hypometabolic epileptogenic focus seen in ["FJFDG-PET studies is also a region of reduced
blood-brain barrier glucose transporter activity and that reductions in phosphorylation are proportional to reductions in
["JFDG influx.
Cornford EM, Gee MN, Swartz BE, Mandelkern MA, Blahd WH, Landaw EM, Delgado-Escueta AV. Dynamic
['*F] fluorodeoxyglucose positron emission tomography and hypometabolic zones in seizures:
reduced capillary influx. Ann Neurol 1998;43:801-808
After noninvasive methods were developed for the
measurement of local cerebral glucose utilization rates
in humans,' positron emission tomography (PET)
measurements of glucose metabolism using ['8F]flu~rodeoxyglucose ([18F]FDG) became a valuable diagnostic
tool in the presurgical evaluation of patients with intractable
Interictal epileptogenic zones are
identified as areas of hypometabolism in ["FIFDGPET studies of patients with complex partial seizures.'-'
Although it is known that this interictal
zone of hypometabolism becomes hypermetabolic ictally,2,8,11,12 reduced energy requirement in the interictal hypometabolic areas has been attributed to neuronal loss and g l i ~ s i s . ' ~ - The
' ~ physiology of blood flow
and glucose metabolism is also altered in these zones."
Areas of [ "IFDG-PET hypometabolism often extend
far beyond the electrographically defined epileptogenic
zones. These areas sometimes extend from the medial
and lateral temporal lobe into cortical and subcortical
regions of the contiguous frontal and parietal lobes (of
the ipsilateral cerebral hemisphere). Although observed
in several studies," the reason(s) for this hypometabolism is not fully understood.
The blood-brain barrier (BBB) glucose transporter
activity may be acutely up-regulated in animal models
of seizures; this is inferred because the glucose utilization rate determined in convulsing rat brain exceeds
the BBB glucose transporter maximal velocity measured in vivo. l' Glucose transporter activity is abundant in capillaries from a human seizure resection,18
and increased transporter activity is also seen in animal
models. l 9 These observations collectively support the
possibility that there is a change in BBB glucose transporter activity in the hypometabolic zone of the human
seizure focus, in contrast to the recent findings of Fink
and colleagues.20 However, a significant reduction in
the fractional glucose extraction within the region of
hypometabolism was reported."
From the Departments of *Neurology and "Biomathematics, and
+Brain Research Institute, UCLA School of Medicine, and $California Comprehensive Epilepsy Program and Neurology Service, Research Service, and §Department of Nuclear Medicine, Veterans Affairs West L o s Angeles Medical Center, Los Angeles; and SPhysics
Department, University of California at Irvine, Irvine, CA.
Received Sep 18, 1997, and in revised form Oct 20, 1997, and Jan
12, 1998. Accepted for publication Jan 13, 1998.
Address correspondence to Dr Cornford, Neurology Service W127,
VA West Los Angeles Medical Center, L o s Angeles, CA 90073.
Copyright 0 1998 by the American Neurological Association 801
Table. Summary of Putients Enrolled in Dynamic [I8F]FDG-PET And&
Seizure Focus Based on EEG and
Neuroimaging (l), CCTV-EEG
Age at Plasma
Patient Age (yr)/
(2), Intracranial CCTV-EEG
Handedness Sex Onset Glucosea or ECOG and Surgery (3)
Magnetic Resonance
Imaging Findings
Left anteriotemporal lobe (1, 2, 3)
Left mesiotemporal
arachnoid cyst
Gliosis, left parietoLeft parietoccipital region (1, 2)
occipital area
Left temporal lobe (1, 2)
Right temporal lobe (1, 2, 3)
None; ventricles
somewhat enlarged
Left temporal lobe (1, 2, 3)
Left anterior temporal encephalomalacia
Cortex, left temporal lobe (1, 2)
None in cortex; left
mesial hippocampal sclerosis
Left frontotemporal
Left temporal lobe (1)
Left frontotemporal region (1, 2, 3) Left frontal lobe
deformity; and left
mesial temporal
CPS, auras
CPS, tonic
clonic seizures
Partial seizures
CPS, suppressed
after surgery
CPS, seizure-free
after surgery
Perisylvian CPS
"Plasma glucose is expressed as milligrams per 100 milliliters, measured from arterial plasma at the time of [ 1 8 F ] f l u ~ r ~ d e ~ x y g l positron
emission tomographic (['8F]FDG-PET) examination.
EEG = electroencephalography; CCTV = closed circuit television; ECOG = electrocorticography; L-h = left handed; R-h = right handed;
A = ambidextrous; CPS = complex partial seizures.
T h e present study therefore tested the hypothesis
that (in patients with a previously defined ["FIFDGPET asymmetry) the interictal zone of hypometabolism was coincidentally a site of altered BBB glucose
transporter activity. W e report herein on dynamic
[ "F] FDG-PET estimations of influx, efflux, phosphorylation, and dephosphorylation rate constants' in patients being evaluated for surgical treatment of intractable seizures.
Patients and Methods
Eight patients, 4 women and 4 men, ranging in age from 22
to 43 years (mean age, 31 years) were enrolled in the present
study (Table). Interictal electroencephalography (EEG),
computed tomographic and magnetic resonance imaging
(MRI) scans, a neuropsychometric battery, and the modified
Wada test were part of their evaluation." No estimates of
cerebral blood flow (CBF) were obtained.
["FIFDG was synthesized by the methods of Hamacher
and associates?1 Dynamic PET scans were performed with
an ECAT I11 (CTI, Knoxville, TN) scanner, operated with a
slice thickness of 22 mm. Resolution of the tomograph was
approximately 5 mm full width half maximum (FWHM)
within the plane of the image, and 11 mm FWHM axially.
Each patient was positioned so that the plane of the tomograph was parallel to the canthomeatal line, with ears unplugged and eyes open. The room was quiet, lights were
dimmed, behavior was maintained with a video camera, and
microphones provided verbal communication between the
Annals of Neurology Vol 43
No 6 June 1998
patient and operator. The morning after an overnight fast,
dynamic PET scans (of -1 hour) were performed after the
injection of 7 to 10 mCi ["FIFDG (0.1 mCi [lsF]FDG/kg
body weight; sterilized and pyrogen tested) into the antecubital vein. The radiochemical purity was determined by liquid chromatography to be greater than 97% and specific activity was 2 to 10 mCi/mg ["FIFDG. Each patient had a
prior nondynamic ["IFDG-PET that identified a zone of
hypometabolism, and the head had been immobilized with
an individual mold constructed from fast-setting foaming
plastic. Individual molds had been saved for reestablishing
positional parameters and controlling head movement in
subsequent scanning examinations. Patients were positioned
so that the brain slice identified at least one predefined hypometabolic epileptogenic region.
A computed attenuation correction was applied to the
scan data. Regions of interest (ROIs) were drawn manually
and l8F activities calculated in localized areas of the slice"
by comparison of the planes with corresponding MRI images
and with the atlases of Matsui and hi ran^^^ and Talairach
and co-worker~.~~
Sequential (radial artery) blood samples
were drawn to delineate the plasma concentration curve.
Brain radioactivity was measured at 2-minute intervals,
and plasma glucose levels (see Table 1) were determined in
each patient. The ["FIFDG concentrations in each plasma
sample were also determined. The mathematical model, and
derivation of the rate constants, is described in detail by
Huang and collaborators.' This model forms the basis for
["FIFDG-PET studies, and assumes that rate constants for
transport of ["FIFDG from plasma to brain = K," and
brain-to-blood efflux = k2*, the rate constant for ["FIFDG
phosphorylation = k3*, and k4* = the rate constant for dephosphorylation. Although Sasaki and colleagues25describe a
method in which k4* is assumed to be zero, we performed
the more conservative analysis, scanning our patients for
more than 45 minutes.26
The time-dependent arterial plasma concentration [C,*(t)]
of ["FIFDG was measured throughout the scan. This, and
the regional cerebral activity [C,'(t)] measured from the scan,
was used to estimate the rate constants (Pvalues) by solving
the convolution integral of Equation 3 in the study by
Huang and colleagues.' Equation 6 was used to calculate cerebral glucose metabolic rates in the regions studied, and we
used the lumped constant (0.418) of Huang and colleagues.'
Regression estimates of rate constants for influx (Kl*),efflux
(k2*), phosphorylation (k3*), and dephosphorylation (k4*)
were derived for the ROI, which contained the (epileptogenic) zone of hypometabolism as well as the homologous
contralateral ROI. In addition, two to four clearly defined
control ROIs from the same plane, and their corresponding
contralateral ROIs, were also analyzed and the four constants
estimated for these regions. In 2 of the 8 patients examined,
more than one hypometabolic zone was identified, so the
four rate constants were estimated in two different regions
(for ipsilateral and contralateral comparisons). Values were
averaged in these patients (with more than one epileptogenic
focus) for statistical analyses. Analyses of treatment and control hemispheres compared K,*, &*, k3*, and k** values in
seizure and contralateral ROIs. For each parameter, paired
analysis of the percent difference between the ipsilateral
and contralateral ROI was computed by [I - (ipsilateral/
contralateral ratio)] X 100%. Statistical significance of differences in comparison with zero was assessed by the Wilcoxon signed-rank test.
The seizure foci were well characterized in all 8 patients enrolled in this study; in addition to a prior MRI
and ["F] FDG-PET scans, television-assisted EEG
monitoring had been performed in 7 of the subjects
(see Table 1). From the dynamic PET measurements,
estimates of the rate constants of ["FIFDG influx, efflux, phosphorylation, and dephosphorylation were determined for each patient. Figure 1 shows these rate
constants in each epileptogenic focus in comparison
with its contralateral ROI, ie, the region in the contralateral hemisphere homologous to the epileptogenic
focus. Asymptotic coefficients of variation of the re19% for k2*,
gression estimates averaged 9% for Kl*,
13% for k,*, and 14% for k4*. Therefore intrapatient
estimation variability was a small fraction of observed
interpatient variability. Observed patient-to-patient coefficients of variation (ie, the standard deviation of the
observed rate constants across individuals divided by
the mean) ranged from 29 to 49% for the various rate
In all 8 patients enrolled in this study, the influx rate
constant (K,*)
was significantly reduced (mean -t- SD,
18 2 13%; p = 0.008) in the seizure foci compared
with the opposite hemispheres (see Fig IA). Efflux rate
constants (k,* values) were also comparably reduced on
average (13 ? 28%; see Fig lB), and although these
changes were not significantly different from zero,
there was no significant change in the ratio (K,*)l(k,*)
between the epileptogenic foci and their contralateral
ROIs. As expected from prior ["FIFDG scans, phosphorylation rate constants (k3* values) were reduced
(25 2 20%; p = 0.008) in the seizure foci (see Fig
1C). Brain glucose utilization rates were also significantly reduced (26 ? 10%) in the epileptogenic region. The estimated dephosphorylation rate constants
(k4* values) were nominally decreased, but the reductions in k4* (3 2 26%) could not be distinguished
statistically from zero ( p > 0.30; see Fig 1D). One of
the study patients (Patient 85) experienced a seizure 25
minutes after injection of the isotope, and the experimental procedure was abruptly terminated. When data
from this subject are excluded from the analysis, differences in both Kl*and k,* (in the seizure focus vs the
homologous contralateral region of interest) remain significantly ( p < 0.05) different. The average reduction
in Kl* in our study group increases from 17.7 2
12.5% to 24.6 2 13.5%; the average reduction in k,*
also increases from 24.7 2 20.5% to 42.7 t 21.1%.
In addition, the mean reductions in k2* and k4* do not
become statistically significant with the removal of data
from this patient.
For the 2 patients who exhibited more than one seizure focus, intrapatient variability in Kl*was similar to
the interpatient variability shown in Figure 1. In Patient 24, K,*was reduced 11.2% in one ROI versus
17.6% in the other. In Patient 25, the reductions in
K,*were 10.8% and 36% for the two ROIs. It is also
noteworthy that Patients 90, 39, and 34 differed from
the others in that MRI analyses indicated there was no
structural lesion at the PET ROI (see Table 1). However, the reductions in rate constants seen in these patients were similar to those seen in the other subjects
(see Fig l), suggesting that the observed alterations in
K,* are not attributable to anatomical effects.
In the same brain slice, other ROIs (ie, nonepileptogenic ROIs) of the ipsilateral hemisphere were compared with their respective contralateral ROIs. The
mean percent changes in the rate constants were as follows: K,* = -0.9 2 4.6% (Fig 2A), k,* = -10.4 -C
19.5% (not shown), k,* = -2.1 2 10.8% (see Fig
2B), and k4* = - 11.1 i- 16.6% (not shown), respectively. For all ROIs beyond the epileptogenic region,
no ipsilateral versus contralateral difference could be
distinguished statistically from zero ( p > 0.05). In addition, patients without MRI-defined structural lesions
(see Table 1) exhibited a pattern similar to the other
Our observed reductions in K,*and k2* suggest that
Cornford et al: Dynamic ["FIFDG-PET and Hypometabolic Zones in Seizures 803
K, : Seizure Focus lpsilateral vs. Contralateral
0.18 T
Seizure FOCUS
K2 :Seizure Focus lpsilateral vs. Contralateral
I E Contralateral
Seizure FOCUS
Patient Number
Patient Number
K3 : Seizure Focus lpsilateral vs. Contralateral
K, : Seizure Focus lpsilateral vs. Contralateral
Seizure FOCUS J
E Contralateral
Seizure Focus
'c 0.008
E 0.05
u" 0.04
Patient Number
Patient Number
Fig 1. A comparison of influx rate constants (A), .flux, rate constants (B), [18Flfluorodeo~gl~~ose
(["FIFDG) phophorylation rates
and dephosphorylation rates (0)obtained ftom dynamic [I8F]FDG positron emission tomographic examinations o f complex
partial seizure patients. The lightly shaded bars represent rate constants measured within the epileptogenic region of interest (ROI),
and the darker bars represent rate constants measured in the homologous ROI contralateral to the epileptogenic site. In all 8 patients examined, there is a comparative reduction in K, * (influx,) and k3* (conversion of["F]FDG to [18F]FDG-6-pho~hate);
these reductions were statistically sign$cant (p < 0.01). Increases in the K,*/k2* ratio (4 ? 36%) were not signij5cant. Mean
reductions in k2* (13 ? 28%) and k4* (3 2 26%) alro could not be distinguished statistically porn zero (p > 0.30).
the epileptogenic focus that is visualized as a zone of
reduced metabolism in ["FIFDG-PET studies is also a
region of reduced BBB glucose transporter activity. In
the epileptogenic focus, reductions in influx (K,*) were
also highly correlated with reductions in phosphorylation (k3*),as indicated in Figure 3. For all 8 patients,
the reductions in K,* and ',k were highly correlated
( p = 0.002). As indicated in Figure 3, the 1 patient
who deviated most from the linear trend (represented
by an encircled data point) experienced a seizure during the PET analysis. Even when data from this subject
are excluded from the analysis, the K,* and k3* rate
constants remain highly correlated ( p < 0.04).
Annals of Neurology Vol 43
No 6 June 1998
All estimated rate constants (Kl*, k2*, k3*, and k4*)
derived in the present study (see Figs 1 and 2) are consistent with previous estimates for the human
brain.25,27-3 1 In a dynamic ["FIFDG-PET study of
patients with the syndrome of myoclonus epilepsy and
ragged-red fibers ( M E W ) , it was reported that k3*
was reduced, but Kl* and k2* values were normal.32
Fink and associates" reported the same pattern in their
analysis of 13 temporal lobe epilepsy patients. To our
knowledge, these represent the only prior dynamic
["FIFDG examinations of patients with a seizure disorder. It is noteworthy that Fink and associates2' were
K, : Non-Seizure lpsilateralvs. Contralateral
[ w Contralateral
y' 0 0 8
Patient Number
K3 : Non-Seizure lpsilateralvs. Contralateral
Patient Number
Fig 2. We compared in$m rate constants @), eflux rate constants (not shown), [18F]jluorodeoxyglucose ([18F]FDG)phosphotylation rates (B), and dephosphoylation rates (not shown)
obtained j o m dynamic ri8F]FDG positron emission tomographic examinations of complex partial seizure patients. The
lightly shaded bars represent mean rate constants measured
within the regions of interest (ROIs) beyond the epileptogenic
site (but j o m the ipsilateral hemisphere, within the same sectional plane), and the darkened bars represent average rate
constants measured in the ROIs on the contralateral hemisphere. Note the consistency between ispilateral and contralatera1 rate constants (in contrast to Fig I), despite individual
variations. Across all 8 patients examined, changes in Kl *
(injlux), k," (eflm), k3* @hosphorylation),and k," (dephospho ylation) could not be distinguished statistically j?om zero.
unable to detect a significant reduction in CBF to the
epileptogenic region (vs contralateral hemisphere); but
reductions were seen in 6 of 13 patients. They suggested that variables such as testing methods and
patient selection criteria possibly contributed to their
inability to define significantly reduced CBF. Furthermore, Fink and associates" included 2 epilepsy patients who did not exhibit a localization-related hypo-
Cornford et
Fig 3. Correlation between the ipsilateral versus contralateral
reduction of [18F]~uorodeoxygluco~e
(["FIFDG) injlm (K, *)
and a reduction in phosphorylation of [18F]FDG (k,y within
the seizure focus of patients with complex partial seizures during dynamic [18]FDG positron emission tomographic analyses.
For all 8 patients, the reductions in K, * and k3* were highly
correlated (correlation coeficient = 0.88, p value = 0.002;
Spearman rank correlation coeficient = 0.905, p = 0.002).
In one instance (encircled data point), the patient experienced
a seizure 25 minutes a8er the injection of J"F]FDG (and the
procedure was terminated). In this situation, the resulting imprecision in estimation of k3* may explain the large deviation
j o r n the least-squares line. When data )om this subject are
excluded j o m the analyses, the K," and k," rate constants
remain highly correlated (the correlation coeficient = 0.74,
p = 0.037; Spearman rank correlation = 0.857, p =
0.014). The data collectively sugest that in vivo, there is a
reduction in blood-brain barrier glucose transporter activity
(in addition to an apparent reduction in phosphoylation) in
the epileptogenic site.
metabolism, whereas we only examined patients known
to exhibit hypometabolism. In addition, Fink and associates2' observed a significant reduction in the relative extraction of glucose in seizure focus ( p < 0.05),
although it is not clear that this is due solely to reductions in k3*.
In hepatocellular carcinoma, dynamic ['*F] FDGPET can distinguish lower from higher grade tumor
carcinomas on the basis of altered rate
prior studies of human brain tumors, K,* and k2* values were similar or higher than those of contralateral
gray matter in n e u r o ~ y t o m a sand
~ ~ elevated in gliomas28.34 as well as in a case of recurrent m e n i n g i ~ m a . ~ ~
In Alzheimer's disease (AD) a 25% reduction in frontal and temporal cortical K,* was seen (compared
with age-matched controls), and phosphorylation (k3*)
was reduced in all brain region^.^' Thus, dynamic
['*F]FDG-PET studies have demonstrated that in
other central nervous system diseases, the influx and
phosphorylation rates may be altered; but reduced influx and phosphorylation was only seen in AD. Jagust
and collaborators2' also noted that glucose transporter
analyses in brain microvessels from AD cases showed
Dynamic ["FIFDG-PET and Hypometabolic Zones in Seizures 805
reduced glucose transporter activity. We observed a
tory attempts by brain homeostatic mechanisms to redown-regulation of both influx and phosphorylation in
strict glucose entry to a specific locus, such as an
complex partial seizure foci. Qualitative changes in the
epileptogenic site, apparently not only regulate endobrain capillary glucose transporter are seen in immunothelial glucose entry to the anatomical lesion, but also
cytochemical postmortem analyses of
but quanto adjacent (normally hnctioning) cells within the
neuropil. We presume that this mechanism is operative
titative reductions in capillary Glut 1 glucose transin those complex partial seizure patients who have deporter protein were recently described in complex
monstrable zones of reduced ["FIFDG trapping surpartial seizures.35
rounding the epileptogenic foci.
One of the most persistent problems, in nearly two
decades of PET research, is the unknown pathophysiIt has been postulated35 of the two different glucose
transporter configurations of capillaries seen in seizure
ology of interictal hypometabolism in localizationresections, that low glucose transporter density capillarrelated e p i l e p s i e ~ .Several
different theories were reies may restrict substrate availability and provide some
cently re~iewed,~'but the possible alteration of BBB
anticonvulsant effect through maintenance of the hyglucose transporter activity was not considered. Within
pometabolic state. The high glucose transporter density
the epileptogenic focus there is an up-regulation of glucapillaries are presumed to assume an important funccose utilization and glucose transport ictally; postiction during ictal events, ensuring that sufficient subtally, a rolonged depression in glucose utilization is
strate is supplied to the seizure focus to prevent cell
~een'~*~'alongwith reduced transporter activity (the
death.35 In vitro studies show increased Glut1 glucose
present study). Our findings suggest that in response to
transporter protects against seizure-induced neuron
the repeated seizure events, there is an accommodation
The observed 15% reduction in baseline plasma
of glucose transporter activity, which is seen as a lowglucose levels seen with the ketogenic diet4' and examered basal state, and thus a zone of reduced influx in
inations of status epilepticus in fasting41 indirectly sugdynamic [18F]FDG-PET scans. Influx of glucose into
normal brain is linked to metabolism, and typically the
gest that reduced glucose availability may provide some
maximal transport rate is twice that of u t i l i ~ a t i o n . ~ ~ , ~anticonvulsant
The redundancy, or excess, in transport rate ensures an
The present investigation also appears to provide a
rational explanation for the consistent observation that
adequate supply of substrate. It is hypothesized that in
the epileptogenic region of complex partial seizure pathe zone of hypometabolism seen in complex partial
seizure patients both surrounds, and also extends far
tients, this redundancy in glucose availability is combeyond, the anatomical lesion.12,36 Diaschisis was propromised. In complex partial seizure resections, two
posed to explain reduced metabolic activity in areas beconfigurations of brain capillaries were recently seen
yond the epileptogenic focus.36 We postulate that, althat display widely different, high and low, numbers of
ternatively, the area of reduced transporter activity
glucose transporter epitopes. About 25% of all the capextends into endothelial cells well beyond the anatomillary profiles exhibit a 10-fold reduction in the numical lesion, and thus reduced metabolism in areas beber of Glut 1 glucose transporter proteins.35
yond the seizure focus might not always be seen in
The reduced influx that we observed in the present
conjunction with remote insult and neuronal loss.
study suggests that the brain attempts to compensate
Analyses of capillary glucose transporter numbers, comfor the extremes of glucose transport and phosphorylaparing the most actively spiking regions of the resection seen in ictal events, by restricting substrate availtion and nonspiking regions, have demonstrated that
ability in the interictal period. Reduced transporter acalthough gliosis and extravasation of serum proteins is
t i v i g 5 would tend to cause similar reductions in both
greater in the actively spiking regions, patterns of alK,* and kz*, as we observed. Although only the reduction in Kl* was statistically significant, failure to distered glucose transporter densities are unchanged.35
tinguish k,* differences from zero may simply reflect
Both the high and low glucose transporter capillaries
are seen throughout the spiking and nonspiking rethe larger variation of the kz* regression estimates. Begions of the resected brain.35
cause the total surface area of brain cell membranes is
In the human epileptogenic focus, an uncoupling is
considerably greater than the surface area of brain capobserved between cerebral glucose metabolism esriillary membrane^,'^ rate limitation of glucose transport
mated from [18F]FDG-PET and CBF measured with
occurs at the BBB membranes, and not at the mem16
branes of cells in the neuropil. If the glucose influx rate
150 water.
In the epileptogenic zones studied,
[18F]FDG-measured metabolism was reduced an avis interictally reduced in the epileptogenic region to a
erage of I 1Yo relative to the contralateral region,
degree that transport is only slightly greater than utiliwhereas CBF rates in the contralateral region dropped
zation rate, this would have a profound effect on ictal
events. Namely, with the rapid increase in metabolism
by only 3 to 6%," although 7 to 12% reductions
that accompanies a seizure, glucose transport would behave been reported in other studies.*' Our [18F]FDG
come the rate-limiting step in metabolism. Autoregulastudy provides a possible explanation for this finding.
Annals of Neurology Vol 43
No 6 June 1998
[”FIFDG has a higher affinity (ie, a lower halfsaturation constant, K,) for the BBB Glut1 glucose transporter than does the natural substrate
D-glucose,” and thus less [18F]FDG than D-glucose
will cross the BBB in the seizure focus (where reduced
influx is seen). The lumped constant assumes plasma/
brain levels to be constant throughout the brain, but
reduced glucose influx presumably could cause some
drop in brain glucose within the zone of hy ometabo h m . Microdialysis studies in animals4324gindicate
that extracellular brain glucose levels can rapidly
change in localized regions. Changes in the lumped
constant are predicted when alterations in the brainto-plasma glucose ratios occur.45 Uncoupling in the
epileptogenic temporal cortex between CBF and
glucose metabolism observed by Gaillard and coworkers“ may be a function of reduced transporter
activity and brain glucose levels (and the lumped constant) within the epileptogenic focus.45 In addition,
two populations of capillaries with differing crosssectional areas have been observed in seizure resections, and these morphological changes presumably
contribute to altered regional flow rates.35
Our observation of reduced BBB influx also poses
the question that if homeostatic mechanisms attempt
to regulate glucose entry into the seizure focus, there
may be situations in the treatment of status epilepticus where clinical management might be augmented
through regulation of glucose availability to the seizure
focus. In focal brain ischemia, it is known that the perifocal “penumbra” region is viable for some extended
period of time and can be rescued from damage with
adequate pharmacological treatment.46 Because an oversupply of glucose may lead to lactate accumulation in
ischemia, and minimal levels are required to maintain
cell viability in the normal brain, the clinical challenge
is to balance these divergent requirements. As more is
learned about the role of BBB glucose transporter
modulations in pathophysiological situations, future
treatment of both brain ischemia and status epilepticus
may include attempts to modify glucose entry into selected brain regions.
Thia study was supported by N I H grants NS-25554,NS-21908
(AVE1. and CA-16042(EML), and in part by the Veterans Administration.
We gratefully acknowledge the assistance of Joan Spellman, Valeri
Nenov, and Nayda Quinones; and we thank Dr William M.
Pardridge for criticisms and suggestions with this study.
1. Huang SC, Phelps ME, Hoffman GJ, et al. Non-invasive determination of local cerebral metabolic rate of glucose in man.
Am J Physiol 1980;238:E69-E82
2. Engel J Jr, Kuhl DE, Phelps ME, Crandall PH. Comparative
localization of the epileptic foci in partial epilepsy by I’CT and
EEG. Ann Neurol 1982;12:529-537
3. Theodore WH, Donvort R, Holmes M , et al. Neuroimaging in
epilepsy: comparison of PET, MRI and CT. Adv Epileptol
4. Phelps ME. Surgical treatment of intractable neonatal onset
seizures: the role of positron emission tomography. Neurology
5. Chugani HT, Shields W D , Shewmon DA,
et al. Infantile
spasms: I. PET identifies focal cortical dysgenesis in cryptogenic
cases for surgical treatment. Ann Neurol 1990;27:406-413
6. Abou-Khalil BW, Siege1 GJ, Sackellares JC, et al. Positron
emission tomography studies of cerebral glucose metabolism in
chronic partial epilepsy. Ann Neurol 1987;22:480-486
7.Chee MWL, Morris H H , Antar MA, et al. Presurgical evaluation of temporal lobe epilepsy using interictal spikes and
positron emission tomography. Arch Neurol 1993;50:45-48
8. Engel J Jr. The use of positron emission tomographic scanning
in epilepsy. Ann Neurol 1984;15(suppl):S180-Sl91
9. Henry TR, Mazziotta JC, Engel J Jr. Interictal metabolic anatomy of mesial temporal lobe epilepsy. Arch Neurol 1993;50:
10. Swartz BE, Halgren E, Delgado-Escueta AV, et al. Neuroimaging in patients with seizures of probable frontal lobe origin.
Epilepsia 1989;30:547-558
11. Theodore WH, Newmark ME, Sato S, et al. [‘8F]Fluorodeoxyglucose positron emission tomography in refractory partial seizures. Ann Neurol 1983;14:429-437
12. Duncan R. Epilepsy, cerebral blood flow rate and cerebral metabolic rate. Cerebrovasc Brain Metab Rev 1992;4:105-121
13. Engel J Jr, Brown WJ, Kuhl DE, et al. Pathological finding
underlying focal temporal lobe hypometabolism in partial epilepsy. Ann Neurol 1982;12:518-528
14. Hajek M, Antonini A, Leender KL, Wieser HG. Mesiobasal
versus lateral temporal epilepsy: metabolic differences in the
temporal lobe shown by interictal ”F-FDG positron emission
tomography. Neurology 1983;43:79-86
15. Sperling MR, Wilson G, Engel J, et al. Magnetic resonance
imaging in intractable partial epilepsy: correlative studies. Ann
Neurol 1986;20:57-62
16. Gaillard WD, Fazilat S, White S, et al. Interictal metabolism
and blood flow are uncoupled in temporal lobe cortex of patients with complex partial epilepsy. Neurology 1995;45:1841-
17. Pardridge WM. Brain metabolism: a perspective from the
blood-brain barrier. Physiol Rev 1983;63:1481-1535
18. Cornford EM, Hyman S, Swartz BE. The human brain
GLUT1 glucose transporter: ultrastructural localization to the
blood-brain barrier endothelia. J Cereb Blood Flow Metab
1994;14:106-1 12
19. Gronlund KM,Gerhart DZ, Leino RL, et al. Chronic seizures
increase glucose transporter abundance in rat brain. J Neuropathol Exp Neurol 1996;55:832-840
20. Fink GR, Pawlik G, Stefan H, et al. Temporal lobe epilepsy:
evidence for interictal uncoupling of blood flow and glucose
metabolism in temporomesial structures. J Neurol Sci 1996;
21. Hamacher K,Coenen H H , Stocklin G. Efficient stereospecific
synthesis of no-carrier-added 2-[iXF]-fluoro-2-deoxy-~-glucose
using aminopolyether supported nucleophilic substitution.
J Nucl Med 1986;27:235-238
22. Nenov VI, Halgren E, Mandelkern MA, Smith ME. Human
brain metabolic responses to familiarity during lexical decision.
Hum Brain Mapping 1994;1:249-268
23. Matsui T, Hirano A. An atlas of the human brain for computerized tomography. Tokyo: Igaku-Shoin, 1978
Cornford et al: D y n a m i c [ i 8 F ] F D G - P E T a n d Hypometabolic Zones i n Seizures
24. Talairach J, Szikla G, l‘urnoux 1’. Atlas d’anatomie stereotazique den telencephale. Paris: Masson, 1967
25. Sasaki H , Kanno I, Murakami M, et al. Tomographic mapping
of rate constants i n the fluorodeoxyglucose model using dynamic positron emission tomography. J Cereb Blood Flow
Metab 1986;6:447- 454
26. Phelps ME. Huang SC, Hoffman EJ, et al. Tomographic measurement of local cerebral glucose metabolic rate in humans
with (F-l8)2-fluoro-2-deoxy-D-glucose: validation of the
method. Ann Neurol 1979;6:371-388
27. Hawkins RA,Phelps ME, Huang SC. Effects of temporal sampling, glucose metabolic rates and disruptions of the blood
brain barrier o n the FDG model with and without a vascular
compartment: studies in human brain tumors with PET.
J Cereb Blood Flow Metab 1986;6:170-183
28. Yamaguchi T,Sasaki H , Ogawa T, et al. Relation hetween tissue nature and (“I.:) fluorodeoxyglucose kinetics evaluated by
dynamic positron emission tomography in human brain tumors. Acra Radio1 Suppl (Stockh) 1986;369:415-418
29. Jagust WJ, Seab JI’, Huesman RH, et al. Diminished glucose
transport in Alzheimer’s disease: dynamic PET studies. J Cereb
Blood Flow Metab 1991;11:323-330
30. Shioya H,Mineura K, Sasajima T , et al. Kinetics o f glucose
metabolism in cenrral neurocyromas. Brain Nerve 1995;47:
31. Shioya H,Mineura K, Sasajima T , et al. Kinetic analysis of
glucose metabolism in meningiomas-comparison
with malignant gliomas. Brain Nerve 1995;47:549-556
32. Bcrkovic SF, Carpenter S, Evans A, et al. Myoclonus epilepsy
and ragged red fibres (MERIIF). 1. A clinical, pathological, biochemical, magnetic resonance spectrographic and positron emission tomographic study. Brain 1989;112:1231-1260
33. Torizuka T, Tamaki N, Inokuma T, et al. In vivo assessment of
glucose metabolism in hepatocellular carcinoma with FDGPET. J Nucl Med 1995;36:1811-1817
34. Shioya H,Mineura K, Sasajima T, et al. Longitudinal analysis
of glucose metabolism in recurrent meningioma. Brain Nerve
Annals of Neurology
Vol 43
No 6 June 1998
35. Cornford EM, Hynian S, Cornford ME, et al. Intsrictal seizure
resections show two configurations of endothelial Glut1 glucose
transporter in the human blood brain barrier. J Cereb Blood
Flow Metab 1998;18:26-42
Henry TR. Functional neuroimaging with positron emission tomography. Epilepsia 1996;37:1141-1154
Crenier JE,Cunningham VJ, Seville MP. Relationships between extraction and metabolism of glucose, blood flow, and
tissue blood volume in regions o f rat brain. J Cereb Blood Flow
Metab 1983;3:2?1-302
Hawkins RA,Mans AM, Davis DW, et al. Glucose availability
to individual cerebral structures is correlated to metabolism.
J Neurochem 1983;40:1013-1018
Lawrence MW,H o DY, Dash R, Sapolsky RM. Herpes simplex
virus vectors overcxpressing the glucose transporter gene protect
against seizure-induced neuron loss. Proc Natl Acad Sci USA
Nordli DR,De Vivo DC. The ketogenic diet revisited: back to
the future. Epilcpsia 1997;38:743-749
Blennow G, Folbcrgrova J, Nilsson 13, Siesjo BK. Effects of
bicuculline-induced seizures on cerebral metabolism and circulation of rats rcndered hypoglycemic by starvation. A n n Neurol
Theodore WH, Leiderman D, Gaillard W, et al. The effect of
naloxone on cerebral blood flow and glucose metabolism in patients with complex partial seizures. Epilepsy Res 1993;16:
Fellows LK, Boutelle MG, Fillenz M. Extracellular brain glucose levels reflect local neuronal activity: a microdialysis study
in awake, freely moving rats. J Neurochcm 1992;59:2141-2147
Hi1 Y,Wilson GS. Rapid changes in local extracellular rat brain
glucose observed with an in vivo glucose sensor. J Neurochem
1997;GS:1745-1 752
Pardridge WM,Crane PI>, Oldendorf WH. On “lumped constant” nomograms. J Neurochem 199239:1775-1776
Siesjo BK, Katsura KI, Zhao Q, et al. Mechanisms of secondary
brain damage in global and focal ischemia: a speculative hypothesis. J Neurotrauma 1995;5:943-956
Без категории
Размер файла
863 Кб
18f, influx, fluorodeoxyglucose, hypometabolism, tomography, emissions, seizure, zone, dynamics, positron, reduced, capillary
Пожаловаться на содержимое документа