Brain, Vol. 124, No. 8, 1601-1609,
August 2001
© 2001 Oxford University Press
Networks mediating the clinical effects of pallidal brain stimulation for Parkinson's disease
A PET study of resting-state glucose metabolism
1 Center for Neurosciences, North Shore-Long Island Jewish Research Institute, Manhasset, 2 Department of Neurology, New York University School of Medicine, 3 Department of Psychiatry, Columbia College of Physicians and Surgeons, New York, 4 Department of Neurology, Oregon Health Science University, Portland, Oregon, 5 University of Miami Medical Center, Miami, Florida, USA and 6 The Toronto Western Hospital, Toronto, Ontario, Canada
Correspondence to:
Dr D. Eidelberg, Center for Neurosciences, North Shore-Long Island Jewish Research Institute, 350 Community Drive, Manhasset, NY 11030, USA E-mail: david1{at}nshs.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Employing [18F]fluorodeoxyglucose (FDG) and PET, we have found previously that stereotaxic ablation of the internal globus pallidus (GPi) for Parkinson's disease causes resting metabolic changes in brain regions remote from the lesion site. In this study we determined whether similar metabolic changes occur in Parkinson's disease patients treated with deep brain stimulation (DBS) of the GPi. We studied seven Parkinson's disease patients with FDG-PET to measure resting regional cerebral glucose utilization on and off GPi stimulation. We used statistical parametric mapping to identify significant changes in regional brain metabolism that occurred with this intervention. We also quantified stimulation-related changes in the expression of a specific abnormal Parkinson's disease-related pattern of metabolic covariation (PDRP) that had been identified in earlier FDG-PET studies. Metabolic changes with DBS were correlated with clinical improvement as measured by changes in Unified Parkinson's Disease Rating Scale (UPDRS) motor ratings off medication. GPi DBS improved UPDRS motor ratings (36%, P < 0.001) and significantly increased regional glucose metabolism in the premotor cortex ipsilateral to stimulation and in the cerebellum bilaterally. GPi DBS also resulted in a significant (P < 0.01) decline in PDRP activity ipsilateral to stimulation, which correlated significantly with clinical improvement in UPDRS motor ratings (P < 0.03). Clinical improvement with GPi DBS is associated with reduced expression of an abnormal Parkinson's disease-related metabolic network involving elements of the cortico-striato-pallido-thalamocortical and the cerebello-cortical motor loops.
Parkinson's disease; deep brain stimulation; PET; [18F]fluorodeoxyglucose; brain networks
BA = Brodmann area; CSPTC = cortico-striato-pallido-thalamocortical; DBS = deep-brain stimulation; FDG = [18F]fluorodeoxyglucose; GPi = globus pallidus internus; PDRP = Parkinson's disease-related metabolic covariance pattern; PMC = premotor cortex; PVP = posterior ventral pallidotomy; rCBF = regional cerebral blood flow; rCMRGlc = regional cerebral metabolic rate for glucose; SMA = supplementary motor area; SPM = Statistical Parametric Mapping; TPR = topographic profile rating; UPDRS = Unified Parkinson's Disease Rating Scale
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
High-frequency deep-brain stimulation (DBS) of the internal globus pallidus (GPi) has been employed as a reversible means of treating Parkinson's disease (Gross et al., 1997
; Pahwa et al., 1997; Tronnier et al., 1997). Although it has been proposed that GPi DBS achieves its therapeutic effect in a manner similar to stereotaxic posterior ventral pallidotomy (PVP) (Starr et al., 1998; Brown et al., 1999), other mechanisms of action have been proposed for this intervention (Montgomery and Baker, 2000). PET studies of motor activation in PVP using 15O-labelled water (H215O) have revealed that this procedure enhances regional cerebral blood flow (rCBF) activation responses in selected motor regions (Grafton et al., 1995; Samuel et al., 1997). By contrast, the results of activation studies performed in conjunction with GPi DBS have been inconsistent (Limousin et al., 1997; Fukuda et al., 2001). Limousin and colleagues (Limousin et al., 1997) described an enhancement of activation responses during joystick movement in the supplementary motor area (SMA), cingulate cortex and dorsolateral prefrontal cortex following stimulation of the subthalamic nucleus but not the pallidum. By contrast, we have recently detected increases in rCBF in primary and auxiliary motor regions during GPi DBS in Parkinson's disease patients scanned during the performance of a kinematically controlled motor execution task (Fukuda et al., 2001). The inconsistency of these PET activation findings may be attributed to differences in experimental design and the demands of the task that the subject performs. The results of H215O-PET studies of GPi DBS patients scanned in the absence of movement have also been inconsistent. Davis and colleagues reported resting increases in the SMA and basal ganglia with pallidal stimulation (Davis et al., 1997). However, comparable rCBF changes were not detected in two other PET studies of GPi DBS performed using H215O techniques (Limousin et al., 1997; Fukuda et al., 2001).
Resting state measurements of regional glucose utilization obtained using [18F]fluorodeoxyglucose (FDG) and PET are another method of localizing the effects of stereotaxic surgical procedures on brain function. In previous FDG-PET studies we have found that Parkinson's disease is associated with the presence of a specific abnormal metabolic brain network (Eidelberg et al., 1996). This Parkinson's disease-related pattern (PDRP) is characterized by lentiform, thalamic and brainstem hypermetabolism covarying with metabolic reductions in the lateral premotor cortex (PMC) and SMA. PDRP activity has been found to be increased in multiple populations of unmedicated Parkinson's disease patients (Moeller et al., 1999) and correlates consistently with advancing motor disability and disease duration (Eidelberg et al., 1990, 1994, 1995; Moeller and Eidelberg, 1997). Indeed, using FDG-PET we have found that the clinical effects of PVP are highly correlated with operative changes in the expression of Parkinson's disease-related metabolic brain networks (Eidelberg et al., 1996). Whether GPi DBS produces its therapeutic effect through an analogous modulation of motor system networks is unknown.
In the present study, we used quantitative FDG-PET to evaluate Parkinson's disease patients in the resting condition on and off pallidal stimulation. The results were used to determine (i) whether GPi DBS induces changes in brain metabolism in the absence of movement; (ii) if so, do the regional metabolic changes induced by GPi DBS correlate with clinical outcome; and (iii) whether pallidal stimulation reduces abnormal PDRP network activity in a manner comparable to ablative pallidotomy.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Patients
We studied seven patients with advanced Parkinson's disease [four men and three women; age (mean ± SD) 50.0 ± 9.6 years, disease duration 12.0 ± 5.6 years]. The clinical characteristics of these patients appear in Table 1
. A diagnosis of Parkinson's disease was made if the patient had `pure' parkinsonism without a history of known causative factors, such as encephalitis and neuroleptic treatment, and did not have dementia, supranuclear gaze abnormalities or ataxia. All patients demonstrated a convincing response to levodopa at the time of diagnosis. Family histories were negative for neurodegenerative illnesses. Patients were referred for GPi DBS because of (i) bradykinesia and rigidity-predominant parkinsonism, and (ii) severe response fluctuations, disabling on-state dyskinesias, or both. Four patients (Patients 1–4) underwent unilateral GPi DBS implantation (three on the left side, one on the right side); the remaining three patients (Patients 5–7) underwent staged bilateral surgery. The details of DBS placement have been described previously (Gross et al., 1997; Pahwa et al., 1997). The target position was confirmed following surgery by MRI. [H215O-PET activation data from six of these patients (Patients 1–3 and 5–7) have been reported previously (Fukuda et al., 2001). These patients were stimulated in the left pallidum, contralateral to the performing dominant right hand. Patient 4, with unilateral stimulation of the right pallidum, was not available for the earlier motor activation study.]
|
Study design
The PET studies were performed on two consecutive days as described previously (Fukuda et al., 2001
). On one of the two testing days, DBS was deployed for at least 12 h prior to PET imaging; on the other testing day, stimulation was off for at least 12 h prior to the imaging experiments. The order of the testing days was randomized. All antiparkinsonian medications were withheld for at least 12 h before each day of testing. Stimulation parameters for the DBS ON condition were selected to achieve maximal improvement in the motor portion of the Unified Parkinson's Disease Rating Scale (UPDRS items 19–31) (Fahn and Elton, 1987) without inducing dyskinesia. The stimulation parameters of these patients appear in Table 2. The more severely affected patients with bilateral DBS experienced sustained limb discomfort following medication washout, even during the deployment of one of the two stimulators. We avoided treating symptoms with adjunctive levodopa because of its demonstrated effects on regional metabolism and network activity (Feigin et al., 2000). In order for these patients to tolerate both days of imaging off-medication and to minimize their discomfort, we elected to deploy both stimulators in the DBS ON imaging studies (Fukuda et al., 2001). Raters blinded to the imaging data assessed the patients in both DBS conditions to obtain ON and OFF motor UPDRS measures. Clinical improvement, as measured objectively by the change in motor UPDRS ratings between stimulation conditions [i.e. (ON – OFF)/OFF x100% (Eidelberg et al., 1996; Kazumata et al., 1997)], was correlated with the changes in regional cerebral metabolism that were produced by the intervention. The FDG-PET scans in the two DBS conditions were also used to assess the effects of stimulation on PDRP network activity.
|
PET
All patients fasted overnight prior to FDG-PET scanning. PET studies were performed in 3D mode using the GE Advance tomograph (Milwaukee, Wis., USA) at North Shore University Hospital, Manhasset, NY (Dhawan et al., 1998
). This eight-ring bismuth germanate scanner provided 35 two-dimensional image planes with an axial field of view of 14.5 cm and transaxial resolution of 4.2 mm (full width half maximum) in all directions. Ethical permission for these studies was obtained from the Institutional Review Board of North Shore University Hospital. Written consent was obtained from each subject following detailed explanation of the procedures.
In each PET session (DBS OFF and DBS ON), patients were positioned in the scanner using the Laitinen stereoadapter (Laitinen et al., 1985) (Sandstrom Trade and Technology, Welland, Ontario, Canada) with 3D laser alignment with reference to the orbitomeatal line. To minimize head movement during studies (Hariz and Eriksson, 1986), we used identical stereoadapter and laser settings in each scan. A cylindrical tube filled with 68Ge was placed in the field of view to provide internal calibration for each slice. All FDG-PET studies were performed with the subject's eyes open in a dimly lit room and minimal auditory stimulation. The time course of plasma 18F radioactivity was determined by sampling radial arterial blood. Global and regional cerebral metabolic rates for glucose (rCMRGlc) were calculated on a pixel-by-pixel basis, as described by us previously (Takikawa et al., 1993; Eidelberg et al., 1994, 1997).
Data analysis
Data processing was performed using SPM99 (Wellcome Department of Cognitive Neurology, London, UK) implemented in MATLAB (Mathworks, Sherborn, Mass., USA). The scans from each subject were aligned, non-linearly warped into Talairach space (Talairach and Tournoux, 1988) and scaled proportionately. The CMRGlc images were normalized for the grand mean and were smoothed with an isotropic Gaussian kernel (full width half maximum 10 mm for all directions) to allow for interindividual gyral variation and to improve the signal : noise ratio. Images from Patient 4 were reversed such that the stimulated hemisphere appeared on the left. In this way, we designated the left hemisphere as being stimulated, i.e. having a DBS effect in each of the seven patients. In statistical parametric mapping (SPM) we analysed only the left cerebral hemisphere, both cerebellar hemispheres and the brainstem. (The right cerebral hemisphere was excluded in this analysis because of its heterogeneous stimulation effects.)
Effects of GPi DBS on regional brain metabolism
rCMRGlc measurements obtained in the DBS ON and OFF conditions were compared on a voxel-by-voxel basis using the paired t-test option in SPM99. On the basis of previous FDG-PET data (Eidelberg et al., 1994, 1996, 1997), we expected GPi DBS to alter rCMRGlc within brain regions known to be associated with the pallidum and its output. We further hypothesized that GPi DBS will also alter metabolic activity in individual components of cortico-striato-pallido-thalamocortical (CSPTC) loops (Wichmann and DeLong, 1996) and in cerebello-cortical pathways (Middleton and Strick, 2001) that are functioning abnormally in parkinsonism. To test these hypotheses, we assessed stimulation-induced changes in glucose metabolism within the set of voxels that was known to be abnormal in Parkinson's disease through prior FDG-PET studies. This was accomplished by creating a mask (Fukuda et al., 2001) defined by significant between-group metabolic differences in an independent population comprising Parkinson's disease patients and age-matched normal volunteer subjects. We used FDG-PET to scan 38 moderately advanced Parkinson's disease patients (OFF-state Hoehn and Yahr score >3.0, age 58.2 ± 7.7 years) and 31 normal subjects (age 53.9 ± 13.7 years); patient scans were conducted 12 h after the cessation of antiparkinsonian medications. The mask was created from the t map of the between-group contrast thresholded at P < 0.001 (t > 3.21). The mask (Fig. 1) sampled metabolic changes in the basal ganglia, thalamus, pons and cerebellum as well as in the premotor, prefrontal, anterior cingulate and posterior parietal cortical regions. DBS effects on glucose metabolism in these areas were considered part of the distributed network of regions altered in Parkinson's disease (Eidelberg et al., 1994) and not independent of each other. Thus, changes within the mask were treated as hypothesis-testing and significant when the P value was thresholded at P < 0.01, uncorrected for multiple independent comparisons. Effects of GPi DBS outside the mask were considered to be hypothesis-generating if significant for P < 0.001, uncorrected for multiple comparisons; these effects were considered significant if they survived correction for multiple comparisons at P = 0.05.
|
We also used SPM to correlate the clinical effects of stimulation with ON–OFF differences in rCMRGlc. This was accomplished by correlating changes in UPDRS motor ratings with the rCMRGlc difference scans across the patient cohort. Correlations were performed on a voxel-by-voxel basis. As an exploratory analysis, hypothesis-driven searches were conducted within the mask and considered significant for P < 0.01, uncorrected for multiple comparisons.
Effect of GPi DBS on Parkinson's disease network activity
The mean effects of stimulation on local brain metabolism evident with SPM do not take into account the functional connectivity that exists between multiple brain regions constituting a spatially distributed neural network. We have found that regional covariance analysis can be used to identify meaningful changes in network activity following localized stereotaxic interventions (Eidelberg et al., 1996). In this study we examined the notion that GPi DBS can reduce the activity of the pathological PDRP network selectively and that this change correlates with the clinical response. Specifically, we considered the possibility that PDRP expression is significantly reduced in stimulated hemispheres but not in unstimulated hemispheres. In order to do this, we employed the topographic profile rating (TPR) algorithm (Eidelberg et al., 1995; Moeller et al., 1996; Moeller and Eidelberg, 1997) to quantify hemispheric network expression in individual patients (i.e. PDRP subject scores) in both stimulation conditions. TPR computations were performed using standardized regions of interest, as described previously (Eidelberg et al., 1995, 1997; Moeller and Eidelberg, 1997; Moeller et al., 1999). FDG-PET images acquired with DBS ON and OFF were aligned so that region-of-interest coordinates were identical for the two treatment conditions (Eidelberg et al., 1998). In the present study the TPR operation was performed separately on the rCMRGlc data from each hemisphere using the left/right means of the PDRP region weights on each of the regions of interest (Eidelberg et al., 1994). All TPR computations were performed in an automated fashion, blind to subject, hemisphere, stimulation condition and clinical response.
Network activity changes with intervention were computed for each subject and hemisphere and expressed as 
PDRP, the ON – OFF difference in the computed PDRP subject scores. In the patients with bilateral stimulation (n = 3), changes in PDRP (PDRP) in the two hemispheres were not considered to be independent; PDRP values for these subjects were averaged across hemispheres. The significance of interventional changes in PDRP network activity was assessed by comparing DBS ON and OFF PDRP subject scores using paired Student's t tests. These comparisons were performed separately for the stimulated and unstimulated hemispheres and considered significant for P < 0.05, two-tailed. Additionally, we correlated UPDRS measures of clinical improvement with stimulated hemisphere PDRP values. Correlations were assessed by computing Pearson product moment correlation coefficients and considered significant for P < 0.05. SPM and TPR analyses were conducted on PCs running Windows NT. Statistical comparisons and correlation analyses were performed using JMP software (SAS Institute, Cary, NC, USA) for PC.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Effects of GPi DBS on regional brain metabolism
GPi DBS improved off-medication UPDRS motor ratings by 36.2% (range 32.1–50.3%, P < 0.001). SPM analysis of the FDG-PET data revealed significant regional increases in rCMRGlc with stimulation (Table 3
and Fig. 2). Hypothesis-driven searches within the mask revealed significant (P < 0.01) metabolic increases with DBS in the PMC [Brodmann area (BA) 6] of the stimulated (left) hemisphere (Fig. 2A, top) and in the cerebellum bilaterally (Fig. 2A, middle and bottom). Scatterplots of adjusted rCMRGlc in these regions are presented for each of the two DBS conditions in Fig. 2B. Regional metabolic decrements subsequent to stimulation were not present within the mask. No significant alterations in regional glucose utilization were identified outside the mask.
|
|
Improvement in the UPDRS motor ratings with stimulation correlated significantly with rCMRGlc increases in the left PMC (BA 6: x = –28, y = –2, z = 52 mm; Zmax = 4.98, P < 0.001 corrected). The voxels correlating with clinical improvement were spatially proximal to those in which stimulation significantly enhanced resting state glucose utilization (Fig. 2A,
top). Indeed, the correlation between the rCMRGlc changes and improvement in the UPDRS motor ratings was significant (r2 = 0.73, P < 0.01) at the local maximum of the mean effect of DBS on resting state metabolism (x = –14, y = 22, z = 60 mm) (Table 3). No other areas of significant correlation were identified. In no region did the improvement in UPDRS motor ratings with stimulation correlate with concomitant metabolic decrements.
Effect of GPi DBS on Parkinson's disease network activity
To assess the effect of GPi DBS on metabolic network activity, we quantified the changes in hemispheric PDRP expression that were induced by this intervention in each patient. On an individual subject basis, network activity declined (PDRP < 0) consistently in the stimulated hemispheres (Fig. 3A, left) and in two of the four unstimulated hemispheres (Fig. 3A, right). For the group as a whole, the mean reduction in PDRP activity with stimulation was highly significant (P < 0.01); no significant mean change was evident contralateral to stimulation (P = 0.6). Changes in PDRP expression in the stimulated hemisphere correlated significantly with clinical improvement in UPDRS motor ratings (r2 = 0.66, P < 0.03) (Fig. 3B). The greatest clinical benefit was associated with the largest decline in resting-state PDRP network activity.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
In this study, we found that GPi stimulation enhanced resting-state glucose utilization in the PMC and in both cerebellar hemispheres. This suggests that this intervention may be associated with functional changes in particular nodes within CSPTC circuits and related cerebello-cortical pathways (Wichmann and DeLong, 1996
; Middleton and Strick, 2001). To assess the effects of DBS on the functional modulation of these networks, we quantified the expression of the PDRP, a previously validated abnormal disease-related metabolic covariance pattern, in each of the two stimulation conditions. We found that GPi stimulation reduced PDRP network activity and that this metabolic change correlated significantly with the clinical improvement that occurred during this intervention. These findings suggest that GPi DBS can improve parkinsonism by suppressing the abnormal resting metabolic activity of CSPTC and can affect cerebello-cortical motor circuits.
Effect of GPi DBS on regional brain metabolism
We detected a significant metabolic increase with GPi DBS in the PMC (BA 6) in the stimulated cerebral hemisphere. This metabolic increase in the resting state can be understood as a consequence of the enhanced cortical afferent synaptic activity from the ventral thalamus that occurred following a reduction in pallidothalamic inhibitory output with stimulation. In this respect, these findings are in agreement with our previous resting-state FDG-PET studies before and after stereotaxic pallidal ablation (Eidelberg et al., 1996). The GPi neurones that project via the thalamus to the PMC are located ventrolaterally, whereas those projecting to the SMA are in the mid-dorsal GPi and those influencing the primary motor cortex (BA 4) are in an intermediate location (Middleton and Strick, 2001). From our results it appears that, in the resting condition, surgical interventions such as GPi DBS and PVP exert their primary metabolic effects on the circuit originating from the ventral GPi to PMC. Nonetheless, pallidal output pathways to the primary motor cortex, the SMA, and even the dorsolateral prefrontal cortex, may also be affected (Eidelberg et al., 1996). This is especially true of PVP, in which irreversible destruction of large areas within the GPi can occur, thereby affecting the functional activity of multiple pallidal output channels (Lombardi et al., 2000). The enhancement of PMC activity by GPi DBS is clinically relevant in that metabolic changes in this area correlate significantly with UPDRS measures of treatment efficacy. The spatial proximity of the voxels which increase metabolically with DBS to those which correlate with clinical improvement during this intervention support the conclusion that therapeutic benefit is linked to enhanced neural function in this region. Nonetheless, other pallidal output channels influencing the primary motor cortex and SMA may be affected by GPi DBS, depending on technical factors such as electrode placement, the specific contacts deployed and the individual stimulation parameters. The effect of these variables on regional brain metabolism, either singly or in combination, is currently unknown.
We found that GPi DBS also enhanced resting state metabolic activity in both cerebellar hemispheres. Although the GPi does not influence cerebellar activity directly, metabolic changes in this region may have occurred indirectly via the pedunculopontine nucleus, an important target of pallidal output to the brainstem (Parent and Hazrati, 1995; Eidelberg et al., 1997; Pahapill and Lozano, 2000). These pallidofugal projections may influence cerebellar function, providing a link between CSPTC and cerebello-cortical pathways. Alternatively, the metabolic increases noted in the cerebellum may have been secondary to increased afferent activity from the PMC.
Using SPM analysis, we did not identify significant stimulation-mediated metabolic declines in the globus pallidus or thalamus, despite the presence of focal increases in the metabolism of the PMC and cerebellum. These findings are in agreement with our previous findings with PVP (Eidelberg et al., 1996), in which such metabolic decrements were also not detected. We attribute this to the relatively small volume of tissue influenced locally by the stimulating electrode and the likelihood of metabolic effects of low magnitude within the target region.
Our FDG-PET findings of significant resting state metabolic change with GPi DBS are consistent with an earlier H215O-PET study of this intervention (Davis et al., 1997). In that investigation, GPi DBS gave rise to significant rCBF increases in the SMA and the putamen ipsilateral to stimulation. We note that the reported locus of rCBF change in the SMA was quite lateral (x = –16 mm), bordering on the PMC. Given differences in the spatial resolution of the two tomographs used in these resting-state studies, it is unlikely that significant differences exist in the localization of the cortical motor region maximally affected by pallidal stimulation. The results of these PET studies suggest that GPi DBS can influence the resting activity of motor pathway elements in regions spatially removed from the site of stimulation.
Effect of GPi DBS on Parkinson's disease network activity
In addition to assessing the effects of GPi DBS on local glucose utilization, we used a covariance approach to determine the impact of this intervention on Parkinson's disease-related metabolic network activity. Because of its reproducibility across multiple patient populations and its consistent correlation with independent disease severity measures, the specific disease-related regional metabolic covariance pattern known as the PDRP (Moeller et al., 1999) was selected as the network imaging marker for this study. By prospectively quantifying DBS-mediated changes in its expression in both cerebral hemispheres, we were able to test the notion that abnormal Parkinson's disease-related network activity is suppressed by pallidal stimulation. In support of this hypothesis, we found that DBS gave rise to a consistent decline in network activity in stimulated but not unstimulated hemispheres. We did note, however, that bilateral declines in PDRP scores were present in two of the four patients with unilateral stimulation (Fig. 3A, right). This raises the possibility of interaction between the two hemispheres at either a thalamic or a cortical level (Parent and Hazrati, 1995; Eidelberg et al., 1996). By contrast, the effects of bilateral stimulation on hemispheric network activity were not augmentative. Indeed, the largest intervention-mediated reductions in PDRP activity were encountered in unilateral DBS patients. A formal within-subject comparison of network modulation during unilateral and bilateral stimulation will be needed to assess the impact of hemispheric interactions on the response to treatment.
The clinical relevance of the network findings is supported by the observation that DBS-mediated changes in PDRP expression in the stimulated hemisphere correlate significantly with objective measures of treatment response. Patients with a relatively high degree of PDRP suppression were found to have the greatest clinical improvement subsequent to this intervention. Additionally, we found that the total motor UPDRS improvement correlated significantly with reductions in PDRP activity in the stimulated hemisphere, in accordance with our previous observation of bilateral clinical benefit following unilateral pallidal ablation (Eidelberg et al., 1996; Fazzini et al., 1997).
In summary, these findings suggest that GPi DBS achieves its therapeutic benefit in Parkinson's disease patients by reducing the resting activity of the abnormal disease-related metabolic network. The suppression of an interfering pathological brain network in the rest condition may enhance motor performance as well as increasing activation responses in elements of the motor CSPTC loop (Fukuda et al., 2001). Our FDG-PET data also suggest that in parkinsonism GPi DBS operates in a manner analogous to that proposed for PVP. Indeed, the magnitude of PDRP, the decline in PDRP activity with treatment, did not differ in the two interventions (P = 0.45, t test). It is likely that, by reducing inhibitory pallidal output to the thalamus and brainstem, both interventions enhance activation responses in the cortical motor regions and cerebellum (Samuel et al., 1997; Eidelberg et al., 1999). However, the two procedures are by no means identical and may give rise to distinctive patterns of metabolic change. Because of the small number of GPi DBS patients available to us for PET study, our analysis was limited to the measurement of treatment-mediated changes in the activity of the previously validated PDRP network. Nonetheless, with more subjects, novel DBS-related covariance patterns may be identified through network analysis in a manner analogous to our prior study of PVP (Eidelberg et al., 1996). Specific brain networks associated with GPi DBS and PVP can then be compared to identify similarities and differences in the patterns of metabolic change produced by the two interventions. Indeed, the functional topography of interventional change subsequent to GPi DBS may not be limited to features referable to the suppression of pallidal outflow. Other mechanisms, such as noise reduction and the synchronization of pallidal output (Montgomery and Baker, 2000), will have to be considered in the interpretation of these PET findings.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
The authors wish to thank Dr Thomas Chaly for radiochemistry support and Ms Christine Edwards for editorial assistance. We acknowledge the valuable technical support provided by Mr Claude Margouleff and Dr Abdel Belakhleff. This work was supported by NIH RO1 NS 35069. M.F. was supported by the Veola T. Kerr Fellowship of the Parkinson Disease Foundation. D.E. was supported by the American Parkinson Disease Association and NIH K24 NS 02101.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Brown RG, Dowsey PL, Brown P, Jahanshahi M, Pollak P, Benabid AL, et al. Impact of deep brain stimulation on upper limb akinesia in Parkinson's disease. Ann Neurol 1999; 45: 473–88.[ISI][Medline]
Davis KD, Taub E, Houle S, Lang AE, Dostrovsky JO, Tasker RR, et al. Globus pallidus stimulation activates the cortical motor system during alleviation of parkinsonian symptoms. Nat Med 1997; 3: 671–4.[ISI][Medline]
Dhawan V, Kazumata K, Robeson W, Belakhlef A, Margouleff C, Chaly T, et al. 1998. Quantitative brain PET: comparison of 2D and 3D acquisitions on the GE Advance Scanner. Clin Positron Imaging 1998; 1: 135–44.[Medline]
Eidelberg D, Moeller JR, Dhawan V, Sidtis JJ, Ginos JZ, Strother SC, et al. The metabolic anatomy of Parkinson's disease: complementary [18F]fluorodeoxyglucose and [18F]fluorodopa positron emission tomographic studies. Mov Disord 1990; 5: 203–13.[ISI][Medline]
Eidelberg D, Moeller JR, Dhawan V, Spetsieris P, Takikawa S, Ishikawa T, et al. The metabolic topography of parkinsonism. J Cereb Blood Flow Metab 1994; 14: 783–801.[ISI][Medline]
Eidelberg D, Moeller JR, Ishikawa T, Dhawan V, Spetsieris P, Chaly T, et al. Assessment of disease severity in parkinsonism with fluorine-18-fluorodeoxyglucose and PET. J Nucl Med 1995; 36: 378–83.
Eidelberg D, Moeller JR, Ishikawa T, Dhawan V, Spetsieris P, Silbersweig D, et al. Regional metabolic correlates of surgical outcome following unilateral pallidotomy for Parkinson's disease. Ann Neurol 1996; 39: 450–9.[ISI][Medline]
Eidelberg D, Moeller JR, Kazumata K, Antonini A, Sterio D, Dhawan V, et al. Metabolic correlates of pallidal neuronal activity in Parkinson's disease. Brain 1997; 120: 1315–24.
Eidelberg D, Moeller JR, Antonini A, Kazumata K, Nakamura T, Dhawan V, et al. Functional brain networks in DYT1 dystonia. Ann Neurol 1998; 44: 303–12.[ISI][Medline]
Eidelberg D, Nakamura T, Mentis MJ, Antonini A, Dhawan V, Ghilardi MF, et al. Brain activation responses with internal pallidal stimulation in Parkinson's disease [abstract]. Neurology 1999; 52: A176.
Fahn S, Elton R. Unified Parkinson's Disease Rating Scale. In: Fahn S, Marsden C, Calne D, Goldstein M, editors. Recent developments in Parkinson's disease. New York: Macmillan; 1987. p. 293–304.
Fazzini E, Dogali M, Sterio D, Eidelberg D, Beric A. Stereotactic pallidotomy for Parkinson's disease: a long-term follow-up of unilateral pallidotomy. Neurology 1997; 48: 1273–7.[Abstract]
Feigin A, Mentis MJ, Dhawan, Margouleff C, Eidelberg D. Levodopa infusion reduces the expression of a Parkinson's disease metabolic brain network. FDG PET study [abstract]. Neurology 2000; 54 (Suppl 3): A113–4.
Fukuda M, Mentis M, Ghilardi MF, Dhawan V, Antonini A, Hammerstad J, et al. Functional correlates of pallidal stimulation for Parkinson's disease. Ann Neurol 2001; 49: 155–64.[ISI][Medline]
Grafton ST, Waters C, Sutton J, Lew MF, Couldwell W. Pallidotomy increases activity of motor association cortex in Parkinson's disease: a positron emission tomographic study. Ann Neurol 1995; 37: 776–83.[ISI][Medline]
Gross C, Rougier A, Guehl D, Boraud T, Julien J, Bioulac B. High-frequency stimulation of the globus pallidus internalis in Parkinson's disease: a study of seven cases. J Neurosurg 1997; 87: 491–8.[ISI][Medline]
Hariz MI, Eriksson AT. Reproducibility of repeated mountings of a noninvasive CT/MRI stereoadapter. Appl Neurophysiol 1986; 49: 336–47.[ISI][Medline]
Kazumata K, Antonini A, Dhawan V, Moeller JR, Alterman RL, Kelly P, et al. Preoperative indicators of clinical outcome following stereotaxic pallidotomy. Neurology 1997; 49: 1083–90.
Laitinen LV, Liliequist B, Fagerlund M, Eriksson AT. An adapter for computed tomography-guided stereotaxis. Surg Neurol 1985; 23: 559–66.[ISI][Medline]
Limousin P, Greene J, Pollak P, Rothwell J, Benabid AL, Frackowiak R. Changes in cerebral activity pattern due to subthalamic nucleus or internal pallidum stimulation in Parkinson's disease. Ann Neurol 1997; 42: 283–91.[ISI][Medline]
Lombardi WJ, Gross RE, Trepanier LL, Lang AE, Lozano AM, Saint-Cyr JA. Relationship of lesion location to cognitive outcome following microelectrode-guided pallidotomy for Parkinson's disease: support for the existence of cognitive circuits in the human pallidum. Brain 2000; 123: 746–58.
Middleton FA, Strick PL. Cerebellar projections to the prefrontal cortex of the primate. J Neurosci 2001; 21: 700–12.
Moeller JR, Eidelberg D. Divergent expression of regional metabolic topographies in Parkinson's disease and normal ageing. Brain 1997; 120: 2197–206.
Moeller JR, Ishikawa T, Dhawan V, Spetsieris P, Mandel F, Alexander GE, et al. The metabolic topography of normal aging. J Cereb Blood Flow Metab 1996; 16: 385–98.[ISI][Medline]
Moeller JR, Nakamura T, Mentis M, Dhawan V, Spetsieris P, Antonini A, et al. Reproducibility of regional metabolic covariance patterns: comparison of four populations. J Nucl Med 1999; 40: 1264–9.
Montgomery EB Jr, Baker KB. Mechanisms of deep brain stimulation and future technical developments. [Review]. Neurol Res 2000; 22: 259–66.[ISI][Medline]
Pahapill PA, Lozano AM. The pedunculopontine nucleus and Parkinson's disease. [Review]. Brain 2000; 123: 1767–83.
Pahwa R, Wilkinson S, Smith D, Lyons K, Miyawaki E, Koller WC. High-frequency stimulation of the globus pallidus for the treatment of Parkinson's disease. Neurology 1997; 49: 249–53.
Parent A, Hazrati LN. Functional anatomy of the basal ganglia. I. The cortico-basal ganglia thalamo-cortical loop. [Review]. Brain Res Brain Res Rev 1995; 20: 91–127.[Medline]
Samuel M, Ceballos-Baumann AO, Turjanski N, Boecker H, Gorospe A, Linazasoro G, et al. Pallidotomy in Parkinson's disease increases supplementary motor area and prefrontal activation during performance of volitional movements: an H2(15)O PET study. Brain 1997; 120: 1301–13.
Schmahmann JD, Doyon J, McDonald D, Holmes C, Lavoie K, Hurwitz AS, et al. Three-dimensional MRI atlas of the human cerebellum in proportional stereotaxic space. Neuroimage 1999; 10: 233–60.[ISI][Medline]
Starr PA, Vitek JL, Bakay RA. Ablative surgery and deep brain stimulation for Parkinson's disease. [Review]. Neurosurgery 1998; 43: 989–1015.[ISI][Medline]
Takikawa S, Dhawan V, Spetsieris P, Robeson W, Chaly T, Dahl R, et al. Noninvasive quantitative fluorodeoxyglucose PET studies with an estimated input function derived from a population-based arterial blood curve. Radiology 1993; 188: 131–6.
Talairach J and Tournoux P. Coplanar stereotaxic atlas of the human brain. Stuttgart: Thieme; 1988.
Tronnier VM, Fogel W, Kronenbuerger M, Steinvorth S. Pallidal stimulation: an alternative to pallidotomy? J Neurosurg 1997; 87: 700–5.[ISI][Medline]
Wichmann T, DeLong MR. Functional and pathophysiological models of the basal ganglia. [Review]. Curr Opin Neurobiol 1996; 6: 751–8.[ISI][Medline]
Received September 25, 2000. Revised February 9, 2001. Accepted April 18, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Hilker, J. Voges, T. Weber, L. W. Kracht, J. Roggendorf, S. Baudrexel, M. Hoevels, V. Sturm, and W. D. Heiss STN-DBS activates the target area in Parkinson disease: An FDG-PET study Neurology, September 2, 2008; 71(10): 708 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hirano, K. Asanuma, Y. Ma, C. Tang, A. Feigin, V. Dhawan, M. Carbon, and D. Eidelberg Dissociation of Metabolic and Neurovascular Responses to Levodopa in the Treatment of Parkinson's Disease J. Neurosci., April 16, 2008; 28(16): 4201 - 4209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Feigin, M. G. Kaplitt, C. Tang, T. Lin, P. Mattis, V. Dhawan, M. J. During, and D. Eidelberg Modulation of metabolic brain networks after subthalamic gene therapy for Parkinson's disease PNAS, December 4, 2007; 104(49): 19559 - 19564. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Asanuma, C. Tang, Y. Ma, V. Dhawan, P. Mattis, C. Edwards, M. G. Kaplitt, A. Feigin, and D. Eidelberg Network modulation in the treatment of Parkinson's disease Brain, October 1, 2006; 129(10): 2667 - 2678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Halliday, V. Macdonald, and J. M. Henderson A comparison of degeneration in motor thalamus and cortex between progressive supranuclear palsy and Parkinson's disease Brain, October 1, 2005; 128(10): 2272 - 2280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sestini, S. Ramat, A. R. Formiconi, F. Ammannati, S. Sorbi, and A. Pupi Brain Networks Underlying the Clinical Effects of Long-Term Subthalamic Stimulation for Parkinson's Disease: A 4-Year Follow-up Study with rCBF SPECT J. Nucl. Med., September 1, 2005; 46(9): 1444 - 1454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Habeck, J. W. Krakauer, C. Ghez, H. A. Sackeim, D. Eidelberg, Y. Stern, and J. R. Moeller A New Approach to Spatial Covariance Modeling of Functional Brain Imaging Data: Ordinal Trend Analysis Neural Comput., July 1, 2005; 17(7): 1602 - 1645. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Feigin, M. F. Ghilardi, M. Carbon, C. Edwards, M. Fukuda, V. Dhawan, C. Margouleff, C. Ghez, and D. Eidelberg Effects of levodopa on motor sequence learning in Parkinson's disease Neurology, June 10, 2003; 60(11): 1744 - 1749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Feigin, M. F. Ghilardi, M. Fukuda, M. J. Mentis, V. Dhawan, A. Barnes, C. P. Ghez, and D. Eidelberg Effects of levodopa infusion on motor activation responses in Parkinson's disease Neurology, July 23, 2002; 59(2): 220 - 226. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Feigin, M. Fukuda, V. Dhawan, S. Przedborski, V. Jackson-Lewis, M.J. Mentis, J.R. Moeller, and D. Eidelberg Metabolic correlates of levodopa response in Parkinson's disease Neurology, December 11, 2001; 57(11): 2083 - 2088. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||