- Email: [email protected]
Changes in regional brain volume three months after stroke
Journal of the Neurological Sciences 322 (2012) 122–128
Contents lists available at SciVerse ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Changes in regional brain volume three months after stroke Amy Brodtmann a, b,⁎, Heath Pardoe a, b, Qi Li a, Renee Lichter a, b, Leif Ostergaard c, Toby Cumming a, b a b c
Florey Neuroscience Institutes, Melbourne Brain Centre, 245 Burgundy Street, Heidelberg, 3081, Australia University of Melbourne, Parkville, 3050, Australia Aarhus University, Copenhagen, Denmark
a r t i c l e
i n f o
Article history: Received 13 January 2012 Received in revised form 13 June 2012 Accepted 10 July 2012 Available online 1 August 2012 Keywords: Cortical thickness Post-stroke atrophy Remodeling
a b s t r a c t Introduction: Little is known about changes in regional brain volume after stroke. We investigated cortical thickness changes over 3 months in a group of stroke patients compared with controls. Material and methods: Patients with acute hemispheric stroke were studied within 3 h of stroke onset and serially over 3 months. We compared the acute and 3 month scans with independently acquired control images. High resolution isotropic T1 images were analyzed using FreeSurfer V5.0, comparing regional average cortical thickness, hippocampal and thalamic volumes. Stroke patient results were analyzed separately for ipsilesional and contralesional regions, whereas control results were averaged across hemisphere. Percentage change scores between the two time points were computed for each participant, and paired sample t-tests were used to assess significant change. Results: 12 stroke patients (9 men, 7 left-hemispheric, mean age = 65.1 years) and 10 control participants (5 men, mean age = 67.2 years) were included. There were no significant differences between the 2 time points in global or regional average cortical thickness, or hippocampal and thalamic volume estimates for control subjects. Regional variability in patient data was demonstrated, particularly cortical thickness increases in contralesional paracentral, superior frontal and insular regions, areas known to be activated in functional MRI studies of motor recovery. A significant reduction in thalamic volume was also found, most apparent ipsilesionally. Conclusions: Post-stroke changes in regional cortical thickness are demonstrable even over short time-frames. Contralesional cortical thickness increases may represent compensatory mechanisms. Significant reductions in thalamic volume may represent evidence of early post-stroke atrophy. Further studies are required to confirm and extend these preliminary results. © 2012 Elsevier B.V. All rights reserved.
1. Introduction 1.1. Post-stroke reorganization Significant dynamic reorganization of distributed networks is well described after stroke, especially in the functional neuroimaging literature [1–3]. This reorganization has been described in the visual [4], language [5], attention and sensory networks [6], but the majority of researchers have focused on motor recovery and associated motor cortical regions due to the importance of motor recovery to functional independence [7]. Activation patterns are usually characterized by early utilization of homologous regions in the contralesional (intact) hemisphere (contralesional recruitment) [1–3], followed by expansion of cortical representation of the damaged area of the cortex into adjacent areas (peri-infarct), with subsequent activation of other cortical and ⁎ Corresponding author at: Florey Neuroscience Institutes, Melbourne Brain Centre, 245 Burgundy Street, Heidelberg 3084, Australia. Tel.: + 61 3 9035 7004; fax: + 61 3 9035 7404. E-mail address: [email protected] (A. Brodtmann). 0022-510X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2012.07.019
sub-cortical regions distant to the lesion [2,8–10]. There is strong evidence that restitution of activation to the ipsilesional hemisphere is associated with a better functional outcome, particularly in the chronic post-stroke phases [2,3,11]. Most of these studies have used positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) techniques to document changes, but more recently methods such as diffusion tensor imaging and connectivity analyses have been used [12–16]. 1.2. Cortical thickness changes after stroke Cortical thickness estimates have been used to document dynamic changes in the human brain, both in normal learning and in pathological states, particularly neurodegenerative processes such as Alzheimer's disease (AD) and frontotemporal dementia. Researchers have reported significant regional volume increases in people intensively learning new tasks, such as London cab drivers learning the maps of London [17,18], melody recognition in expert musicians [19], and students acquiring expertise in a new language [20]. Surprisingly, cortical thickness analyses have not been used to chart cortical plasticity after stroke,
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
123
despite the fact that there is strong evidence that such functional changes are taking place. There is now some evidence that regional volumes may decline in the long-term after stroke, particularly in patients who subsequently develop cognitive decline [21,22].
significant psychiatric disease, had no significant carotid artery stenosis on carotid duplex Doppler ultrasound, and no stroke on MRI.
1.3. Atrophy and cognitive decline
Patients: 12 patients were scanned. 10 patients had 3D T1-weighted whole brain FSPGR images acquired on a GE Signa Excite 3 T MRI scanner with the following acquisition parameters: echo time TE= 3 ms, repetition time TR= 625 ms, flip angle=20°, voxel resolution = 0.9375 mm in plane, slice thickness= 1.3 mm. Two stroke patients were scanned on a GE Genesis Signa 1.5 T MRI scanner: TE= 4.2 ms, TR= 830 ms, flip angle = 20°, voxel resolution = 0.9375 mm in plane, slice thickness= 1.5 mm. Controls: 3D T1-weighted whole brain MPRAGE images were acquired on a Siemens TRIO MRI scanner with the following acquisition parameters: echo time TE = 2.55 ms, inversion time TI = 900 ms, repetition time TR = 1900 ms, flip angle = 9°, voxel resolution = 1 mm isotropic.
Recent research has indicated that consistent patterns of regional cortical thickness change are strongly associated with AD [23–25]. There is now a large body of work examining the association between brain volume changes and disease diagnosis and progression in many dementia syndromes. There is evidence from animal models that cortical thickness changes occur in the post-stroke period. Using a rat stroke model, Karl et al. found decreases in cortical thickness, volume, and neural density extending far beyond the stroke infarct [26]. There is evidence in humans that some brain regions—such as hippocampi and thalami—exhibit disproportionate atrophy after stroke [27,28]. Cognitive impairment and dementia are common after stroke [29], with vascular dementia accounting for about one-fifth of all dementia cases [30]. Yet we still know very little about whether brain volume loss—a hallmark of dementia—occurs after stroke, and whether such atrophy is related to cognitive decline. Magnetic resonance imaging (MRI) markers of structural brain aging (such as lower total brain volume, hippocampal volume or increasing white matter hyperintensity load) and performance on neuropsychological tests of memory and executive function are powerful predictors of dementia in the general population [31–34]. In contrast to pathologically-based autopsy studies, using brain volume measures to investigate the association between stroke and dementia permits a longitudinal approach. 1.4. Current study Dynamic remodeling after stroke may lead to changes in cortical thickness, but these have not been demonstrated. Minimal evidence hints at some post-stroke regional atrophy, but this is also unclear [21,22,28]. Despite the hundreds of longitudinal stroke studies using MRI, there have been no reports of cortical thickness analyses using high-resolution regional estimates such as FreeSurfer [35,36]. We performed an exploratory analysis of cortical thickness changes in order to address the feasibility of volume comparisons between the hyperacute post-stroke periods and more chronic time points. We investigated cortical thickness changes over a 3 month period in both a group of stroke patients and healthy controls of a comparable age. We compared the regional volumes of patients studied acutely with their scans performed at 3 months post-stroke. We hypothesized that there would be a decline in the regional volume of the hippocampi and thalami of stroke patients, but did not expect significant cortical thickness changes in either stroke or control participants over this short time frame. 2. Materials and methods 2.1. Participants Patients were prospectively recruited into a thrombolysis study using MRI imaging in the acute and chronic post-stroke periods to establish recanalization rates (author L.O.). They were included if: they presented with a first-ever acute middle cerebral artery (MCA) territory stroke, were able to be studied with MRI within 2 h of stroke onset, were previously independent without cognitive decline, and were able to be scanned over a 3 month period. Patients were studied within 2 h and serially over 3 months (at study inclusion, 3 h, 24 h, 1 week, 6 weeks and 12 weeks). We compared the 2 hour and 3 month scans with independently acquired control images, also taken 3 months apart. Healthy control participants were included if: they were aged 60–90 years, had no prior neurological or
2.2. Imaging
2.3. Analysis The structural scans were processed using FreeSurfer V 5.0 with default processing settings. Processed images were visually inspected and skull stripping and white matter edits were made where appropriate. Cortical thickness measures were averaged at the lobar level using the inbuilt FreeSurfer cortical parcellation procedure (“aparc”). FreeSurfer produces cortical thickness maps of 35 cortical regions (measured in millimeters) as well as volume estimates of structures, including the thalami (in mm3). We compared average cortical thickness in each of the cortical regions as well as hippocampal and thalamic volumes. For stroke patients, results were divided into ipsilesional and contralesional regions, in order to compare stroke hemisphere with non-stroke hemisphere. To identify whether stroke severity was related to brain volume change, correlations were computed between acute National Institutes of Health Stroke Score (NIHSS) score and volume measures. For controls, cortical thickness and volume measures were averaged across left and right hemispheres. Paired-sample t-tests were used to establish whether the measures of brain volume changed significantly over 3 months. Given the differences in regional thickness between individuals, particularly with increasing age, we elected to use change scores rather than absolute values. This was done to minimize the problems associated with cortical thickness differences that occur as part of normal aging [25,37]. To account for individual variability in baseline volumes, baseline and 3 month data were used to compute percentage change scores for each participant. Group differences in mean percentage change score were analyzed using independent t-tests. The significance level of α = 0.05 was not adjusted for multiple testing, and results from this preliminary study should be interpreted with this in mind. Individual images were inspected to check the accuracy of the automated segmentation. Given the very acute time frame of the first imaging session, the stroke site was barely visible, as ischemic infarction in the hyperacute setting is not associated with the accumulation of significant edema. The presence of edema becomes problematic in the 24–72 hour period, hence the choice of the 2 hour scan for the comparison. We performed the analyses with the lesions unmasked, particularly given only 2 were cortical. Qualitative inspection of the cortical thickness results showed that the modeling of the gray/ white matter and outer gray matter surfaces exterior to the lesion was unaffected by the presence of the lesion. 3. Results Twelve stroke patients (9 men, 7 left-hemispheric, mean age = 65.1 years, range 45–74 years) were included. Ten strokes were
124
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
subcortical, and 2 cortical. Five patients had no vessel occlusion on acute and follow-up imaging, and 7 had minimal or incomplete recanalization on the 3 month imaging. Mean NIHSSs at presentation was 9.7 ± 3.9 (range 3–16), and at 3 months was 2 ± 2.7 (range 0–9). Patients were scanned on average 98 min after symptom onset (range 50–143 min). Ten control participants (5 men, mean age = 67.2 years) were included. Edema had fully resolved by 3 months, but involution had occurred. Only 2 of the strokes involved cortical structures, and FreeSurfer correctly excluded these regions in its automated segmentation. Cortical thickness estimation was only performed on extant cortex. Within-group changes over time are outlined in Table 1. Stroke patients exhibited a significant increase in contralesional cortical thickness and significant reductions in the contralesional thalamus and hippocampus, as well as a significant reduction in the ipsilesional thalamus (see Fig. 1). A positive correlation between acute NIHSS and change in contralesional cortical thickness (r = 0.65, p = 0.021) indicated that more severe stroke was linked with greater cortical thickness increase. Data on percentage change in volume across time are detailed in the table and shown in Fig. 1. When compared directly with control results, there was a significant increase in the contralesional cortical thickness [t(20) = − 2.59, p = 0.018] and a significant decrease in the ipsilesional thalamus [t(15)= 2.49, p = 0.025]. Percentage change in FreeSurfer's 35 different cortical thickness regions is illustrated in Figs. 2–4. Contralesionally, significant increases were found in the superior parietal (p=0.049), middle temporal (p=0.031), temporal pole (p=0.038), paracentral (p = 0.005), insula (p = 0.005), precentral (p = 0.033) and superior frontal (p = 0.006)—see Figs. 2 and 5. Ipsilesionally, significant increases were found in the caudal middle frontal (p = 0.012) and rostral anterior cingulate (p = 0.002)—see Figs. 3 and 5. In controls, there were no regions with significant cortical thickness or volume change over time—see Fig. 4. 4. Discussion We found that hippocampal and thalamic volumes do decline in the 3 months post-stroke, although only the thalamic volume decreases were significant over such a short time-frame. These reductions in thalamic volume may represent evidence of early post-stroke atrophy. There was evidence that the volumetric changes are more marked in some patients, raising the possibility of using these volume changes to track individual decline. Interestingly, significant regional increases in cortical thickness were also apparent, most apparent contralesionally, but also evident in the ipsilesional anterior cingulate gyrus. These findings correlate with remodeling seen in fMRI and PET studies post-stroke, and confirm that post-stroke changes in regional cortical thickness are demonstrable even over short time-frames. These contralesional cortical thickness increases may represent compensatory mechanisms, but correlation with
Fig. 1. Percentage change in cortical thickness and regional brain volume between baseline and 3 months in stroke patients' contralesional (“contra”, dark gray) and ipsilesional (“ipsi”, light gray) hemispheres and controls (white, average both hemispheres): standard error bars are shown; *p b 0.05, **p b 0.01.
individual motor recovery was not performed given the small study numbers. Longitudinal brain volume changes have not been thoroughly examined following stroke. This is surprising, given the wealth of knowledge that exists on changes in brain volume over time in other conditions, such as schizophrenia [38] and dementia [39]. The few reported cross-sectional samples have yielded conflicting results. In 2000, Pohjasvaara et al. found an association between medial temporal lobe atrophy and dementia in 337 patients scanned 3 months after their first stroke [40]. In the Sydney Stroke Study, manual hippocampal volume estimates in 90 cerebrovascular patients (15 TIA) scanned 3–6 months after their event were not different from a group of control subjects [41]. Stebbins et al. grouped 91 ischemic stroke patients into those with no cognitive impairment and those with some impairment, and found greater gray matter volume reductions in the cognitively impaired group, predominantly thalamic [28]. Kraemer et al. identified delayed shrinkage of brain tissue that was not confined to peri-infarct regions in 10 stroke patients scanned between 1 and 4 years after stroke [21]. Remote changes were detected in the ipsilesional white matter and subcortical structures (striatum, thalamus). Observed atrophy occurred remotely in other vascular territories (e.g., the ipsilateral thalamus, supplied by posterior cerebral artery, in patients with MCA infarcts). Recently Nitkunan et al. found a significant difference in brain volume between the patients with small vessel disease and control subjects when compared across 2 time-points [22]. They corroborated the hippocampal and thalamic
Table 1 Change in cortical thickness and regional volume between baseline and 3 months. Cortical thickness in mm; thalamic and hippocampal volume in mm3; significant p-values in bold. Baseline mean value
Stroke—contralesional
Stroke—ipsilesional
Controls
Cortical thickness Thalamus Hippocampus Cortical thickness Thalamus Hippocampus Cortical thickness Thalamus Hippocampus
2.49 6770 4769 2.51 6613 4662 2.51 6169 3765
3 month mean value
2.60 6395 4556 2.54 6027 4584 2.52 6095 3786
Mean change
0.10 −375 −213 0.03 −586 −77 0.01 −74 21
SD
0.09 564 320 0.09 659 462 0.05 244 169
95% CI Lower
Upper
0.04 −733 −416 −0.03 −1005 −371 −0.02 −249 −100
0.16 −17 −9 0.09 −168 216 0.05 101 142
t
p
3.80 −2.30 −2.30 1.25 −3.08 −0.58 0.86 −0.96 0.40
0.003 0.042 0.042 0.24 0.010 0.57 0.41 0.36 0.70
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
125
Fig. 2. Stroke patients: percentage change (y-axis) in cortical thickness regions between baseline and 3 months in contralesional hemisphere (standard error bars are shown; *p b 0.05, **p b 0.01).
volume reductions, and suggested that these volume reductions are apparent very early after ischemic infarction. These conflicting results may have arisen from analyses at different post-stroke periods, the use of different field strengths for image acquisition, the naturally disparate and heterogeneous nature of the stroke population, and the different image analysis methods. Automated hippocampal volume analysis is known to systematically overestimate hippocampal volumes when compared to manual estimation methods [42,43]; images acquired in 1.5 T scanners may have poorer gray–white matter contrast to enable accurate surface modeling required for cortical thickness analyses [44]. Kraemer et al. and Stebbins et al. used modified optimized voxel-based morphometric methods [45]; in the latter paper the lesion site was masked,
and the authors subsequently assessed regional differences in gray matter volume, doing voxel-by-voxel comparisons for selected regions [21,28]. Nitkunan et al. used SIENA from the FSL suite (Structural Image Evaluation, using Normalization, of Atrophy; www.fmrib.ox.ac. uk/fsl) [22]. Pohjasvaara et al. [40] rated brain atrophy from 0 to 3 (none, mild, moderate, severe) according to Scheltens et al. [46]. Sachdev et al. performed manual hippocampal estimations [41]. None of these researchers used surface-based cortical thickness estimates, such as FreeSurfer, which has been found to be reliably sensitive to regional gray matter loss in other disease states [47,48], and has also been demonstrated to have good performance stability in both simulated and real MR brain data sets [49]. The inclusion or exclusion of the hippocampal and thalamic gray matter in the global average thickness
Fig. 3. Stroke patients: percentage change (y-axis) in cortical thickness regions between baseline and 3 months in ipsilesional hemisphere (standard error bars are shown; *p b 0.05, **p b 0.01).
126
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
Fig. 4. Control participants: percentage change (y-axis) in cortical thickness regions between baseline and 3 months (standard error bars are shown).
would significantly skew these estimates, given that the hippocampi and thalami have been demonstrated to decline in regional analyses. Clearly, there needs to be more work in evaluating and assessing different analysis methods (e.g., voxel-based versus surface-based, etc.) in the stroke population. In addition, scanning at disparate time-points would have had very significant effects. More incident atrophy would be expected 4 years after stroke than 1—particularly given that these patients often have co-existing small vessel disease with probable associated accelerated atrophy [22]—and extensive post-stroke remodeling would be expected at a 3 month time-point when compared to 6 months, when the majority of reorganization will have occurred [50]. The contralesional increases in cortical thickness are also interesting, particularly given that these are areas known to be activated
Fig. 5. Regional cortical thickness maps. Areas of significant cortical thickness increase overlaid onto FreeSurfer generated average brain. A = contralesional cortical thickness increases; B = ipsilesional cortical thickness increases. Regions are labeled via FreeSurfer anatomical convention. Some regions are numbered for identification: 1 = rostral anterior cingulate, 2 = middle frontal gyrus, 3 = superior frontal gyrus, 4 = precentral gyrus, 5 = superior parietal lobule, 6 = middle temporal gyrus, 7 = mesial paracentral regions including anterior cingulate cortex.
in functional MRI studies of motor recovery [2,3,7–10,14]. The ipsilesional increase in anterior cingulate thickness is also of note, as cingulate regions are known to be involved in motor learning and recruited with tasks of increasing complexity [51], as well as being important in recovery after stroke affecting the motor regions [52]. It is possible that the thicknesses measured at the three month post-stroke stage represent a normalization following the acute phase. This “pseudo-normalization” has been seen in other disease states, such as hippocampal volume increases followed by subsequent atrophy after temporal lobe seizures [53], and the initial regional volume increases that precede atrophy in schizophrenia. However, most of the changes were contralesional, and the acute scan was done less than 2 h from symptom onset, before the development of significant edema. We designed the study to assess the feasibility of volume comparisons between the acute post-stroke periods and more chronic time-points. Clearly, there are small numbers in this study, and correlation with individual patient recovery scores or cognitive measures was not performed given that only 12 stroke patients were examined. There is still no agreement on how lesions—such as strokes—affect the cortical thickness measures. Stroke patients and control participants, while prospectively recruited over a similar time period, scanned using the same field strength and analyzed centrally, were scanned at 2 different centers, but results from the same scanner were used for the within subject analyses. Qualitative inspection of the cortical thickness results showed that the modeling of the gray/white matter and outer gray matter surfaces exterior to the lesion was unaffected by the presence of the lesion. We are in the process of performing further analyses with lesions masked and unmasked, as well as comparing regional volume estimate techniques (cortical thickness versus voxel-based methods) in order to establish the best methods for longitudinal regional volume measures after stroke. 5. Conclusions Hippocampal and thalamic volumes do decline in the 3 months post-stroke. There is significant individual variability, suggesting larger numbers and longer time-frames are needed. Significant
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
regional increases in cortical thickness are also evident, most apparent contralesionally. Post-stroke changes in the regional cortical thickness are demonstrable even over short time-frames. Contralesional cortical thickness increases may represent compensatory mechanisms. Significant reductions in thalamic volume may represent evidence of early post-stroke atrophy. Further studies are required to evaluate the longitudinal nature of these preliminary findings. Conflict of interest The authors have no conflict of interest to declare. Acknowledgments This work was supported by grants from the Brain Foundation, the JO & JR Wicking Trust, and by the Sidney and Fiona Myer Family Foundation. The Florey Neuroscience Institutes acknowledges the strong support from the Victorian Government and in particular the funding from the Operational Infrastructure Support Grant. Enormous thanks to Soren Christensen for assistance with accessing the stroke scans. References [1] Rehme AK, Fink GR, von Cramon DY, Grefkes C. The role of the contralesional motor cortex for motor recovery in the early days after stroke assessed with longitudinal FMRI. Cereb Cortex Apr 2011;21(4):756-68. [2] Calautti C, Jones PS, Naccarato M, Sharma N, Day DJ, Bullmore ET, et al. The relationship between motor deficit and primary motor cortex hemispheric activation balance after stroke: longitudinal fMRI study. J Neurol Neurosurg Psychiatry Jul 2010;81(7):788-92. [3] Calautti C, Leroy F, Guincestre JY, Baron JC. Displacement of primary sensorimotor cortex activation after subcortical stroke: a longitudinal PET study with clinical correlation. Neuroimage Aug 2003;19(4):1650-4. [4] Brodtmann A, Puce A, Darby D, Donnan G. Serial functional imaging poststroke reveals visual cortex reorganization. Neurorehabil Neural Repair Feb 2009;23(2): 150-9. [5] Fridriksson J, Richardson JD, Fillmore P, Cai B. Left hemisphere plasticity and aphasia recovery. Neuroimage Apr 2 2012;60(2):854-63. [6] Corbetta M, Shulman GL. Spatial neglect and attention networks. Annu Rev Neurosci 2011;34:569-99. [7] Calautti C, Baron JC. Functional neuroimaging studies of motor recovery after stroke in adults: a review. Stroke Jun 2003;34(6):1553-66. [8] Marshall RS, Zarahn E, Alon L, Minzer B, Lazar RM, Krakauer JW. Early imaging correlates of subsequent motor recovery after stroke. Ann Neurol May 2009;65(5): 596-602. [9] Askim T, Indredavik B, Vangberg T, Haberg A. Motor network changes associated with successful motor skill relearning after acute ischemic stroke: a longitudinal functional magnetic resonance imaging study. Neurorehabil Neural Repair Mar– Apr 2009;23(3):295-304. [10] Calautti C, Naccarato M, Jones PS, Sharma N, Day DD, Carpenter AT, et al. The relationship between motor deficit and hemisphere activation balance after stroke: a 3 T fMRI study. Neuroimage Jan 1 2007;34(1):322-31. [11] Ward NS, Newton JM, Swayne OB, Lee L, Thompson AJ, Greenwood RJ, et al. Motor system activation after subcortical stroke depends on corticospinal system integrity. Brain Mar 2006;129(Pt 3):809-19. [12] Rehme AK, Eickhoff SB, Rottschy C, Fink GR, Grefkes C. Activation likelihood estimation meta-analysis of motor-related neural activity after stroke. Neuroimage Feb 1 2011;59(3):2771-82. [13] Wang LE, Tittgemeyer M, Imperati D, Diekhoff S, Ameli M, Fink GR, et al. Degeneration of corpus callosum and recovery of motor function after stroke: a multimodal magnetic resonance imaging study. Hum Brain Mapp Oct 22 2011 [Electronic publication ahead of print]. [14] Grefkes C, Fink GR. Reorganization of cerebral networks after stroke: new insights from neuroimaging with connectivity approaches. Brain May 2011;134(Pt 5): 1264-76. [15] Rehme AK, Eickhoff SB, Wang LE, Fink GR, Grefkes C. Dynamic causal modeling of cortical activity from the acute to the chronic stage after stroke. Neuroimage Apr 1 2011;55(3):1147-58. [16] Park CH, Chang WH, Ohn SH, Kim ST, Bang OY, Pascual-Leone A, et al. Longitudinal changes of resting-state functional connectivity during motor recovery after stroke. Stroke May 2011;42(5):1357-62. [17] Maguire EA, Woollett K, Spiers HJ. London taxi drivers and bus drivers: a structural MRI and neuropsychological analysis. Hippocampus 2006;16(12):1091-101. [18] Maguire EA, Gadian DG, Johnsrude IS, Good CD, Ashburner J, Frackowiak RS, et al. Navigation-related structural change in the hippocampi of taxi drivers. Proc Natl Acad Sci U S A Apr 11 2000;97(8):4398-403.
127
[19] Foster NE, Zatorre RJ. Cortical structure predicts success in performing musical transformation judgments. Neuroimage Oct 15 2010;53(1):26-36. [20] Stein M, Federspiel A, Koenig T, Wirth M, Strik W, Wiest R, et al. Structural plasticity in the language system related to increased second language proficiency. Cortex Apr 2012;48(4):458-65. [21] Kraemer M, Schormann T, Hagemann G, Qi B, Witte OW, Seitz RJ. Delayed shrinkage of the brain after ischemic stroke: preliminary observations with voxel-guided morphometry. J Neuroimaging Jul 2004;14(3):265-72. [22] Nitkunan A, Lanfranconi S, Charlton RA, Barrick TR, Markus HS. Brain atrophy and cerebral small vessel disease: a prospective follow-up study. Stroke Jan 2011;42(1):133-8. [23] Du AT, Schuff N, Kramer JH, Rosen HJ, Gorno-Tempini ML, Rankin K, et al. Different regional patterns of cortical thinning in Alzheimer's disease and frontotemporal dementia. Brain Apr 2007;130(Pt 4):1159-66. [24] Julkunen V, Niskanen E, Koikkalainen J, Herukka SK, Pihlajamäki M, Hallikainen M, et al. Differences in cortical thickness in healthy controls, subjects with mild cognitive impairment, and Alzheimer's disease patients a longitudinal study. J Alzheimers Dis 2010;21(4):1141-51. [25] Salat DH, Buckner RL, Snyder AZ, Greve DN, Desikan RS, Busa E, et al. Thinning of the cerebral cortex in aging. Cereb Cortex Jul 2004;14(7):721-30. [26] Karl JM, Alaverdashvili M, Cross AR, Whishaw IQ. Thinning, movement, and volume loss of residual cortical tissue occurs after stroke in the adult rat as identified by histological and magnetic resonance imaging analysis. Neuroscience Sep 29 2010;170(1):123-37. [27] Pohjasvaara T, Mantyla R, Salonen O, Aronen HJ, Ylikoski R, Hietanen M, et al. How complex interactions of ischemic brain infarcts, white matter lesions, and atrophy relate to poststroke dementia. Arch Neurol Sep 2000;57(9):1295-300. [28] Stebbins GT, Nyenhuis DL, Wang C, Cox JL, Freels S, Bangen K, et al. Gray matter atrophy in patients with ischemic stroke with cognitive impairment. Stroke Mar 2008;39(3):785-93. [29] Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol Nov 2009;8(11):1006-18. [30] Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, et al. Global prevalence of dementia: a Delphi consensus study. Lancet Dec 17 2005;366(9503):2112-7. [31] Jack Jr CR, Lowe VJ, Weigand SD, Wiste HJ, Senjem ML, Knopman DS, et al. Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer's disease: implications for sequence of pathological events in Alzheimer's disease. Brain May 2009;132(Pt 5):1355-65. [32] Jack Jr CR, Shiung MM, Weigand SD, O'Brien PC, Gunter JL, Boeve BF, et al. Brain atrophy rates predict subsequent clinical conversion in normal elderly and amnestic MCI. Neurology Oct 25 2005;65(8):1227-31. [33] Seshadri S, Wolf PA, Beiser A, Elias MF, Au R, Kase CS, et al. Stroke risk profile, brain volume, and cognitive function: the Framingham Offspring Study. Neurology Nov 9 2004;63(9):1591-9. [34] Elias MF, Beiser A, Wolf PA, Au R, White RF, D'Agostino RB. The preclinical phase of Alzheimer disease: a 22-year prospective study of the Framingham Cohort. Arch Neurol Jun 2000;57(6):808-13. [35] Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis. II: inflation, flattening, and a surface-based coordinate system. Neuroimage Feb 1999;9(2):195-207. [36] Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage Feb 1999;9(2):179-94. [37] Fjell AM, Westlye LT, Amlien I, Espeseth T, Reinvang I, Raz N, et al. High consistency of regional cortical thinning in aging across multiple samples. Cereb Cortex Sep 2009;19(9):2001-12. [38] Pantelis C, Velakoulis D, McGorry PD, Wood SJ, Suckling J, Phillips LJ, et al. Neuroanatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet Jan 25 2003;361(9354):281-8. [39] Weiner MW, Aisen PS, Jack Jr CR, Jagust WJ, Trojanowski JQ, Shaw L, et al. The Alzheimer's disease neuroimaging initiative: progress report and future plans. Alzheimers Dement May 2010;6(3):202-11 [e7]. [40] Pohjasvaara T, Mantyla R, Salonen O, Aronen HJ, Ylikoski R, Hietanen M, et al. MRI correlates of dementia after first clinical ischemic stroke. J Neurol Sci Dec 1 2000;181(1–2):111-7. [41] Sachdev PS, Chen X, Joscelyne A, Wen W, Altendorf A, Brodaty H. Hippocampal size and dementia in stroke patients: the Sydney stroke study. J Neurol Sci Sep 15 2007;260(1–2):71-7. [42] Konrad C, Ukas T, Nebel C, Arolt V, Toga AW, Narr KL. Defining the human hippocampus in cerebral magnetic resonance images—an overview of current segmentation protocols. Neuroimage Oct 1 2009;47(4):1185-95. [43] Tae WS, Kim SS, Lee KU, Nam EC, Kim KW. Validation of hippocampal volumes measured using a manual method and two automated methods (FreeSurfer and IBASPM) in chronic major depressive disorder. Neuroradiology Jul 2008;50(7): 569-81. [44] Jovicich J, Czanner S, Han X, Salat D, van der Kouwe A, Quinn B, et al. MRI-derived measurements of human subcortical, ventricular and intracranial brain volumes: reliability effects of scan sessions, acquisition sequences, data analyses, scanner upgrade, scanner vendors and field strengths. Neuroimage May 15 2009;46(1): 177-92. [45] Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS. A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage Jul 2001;14(1 Pt 1):21-36. [46] Scheltens P, Leys D, Barkhof F, Huglo D, Weinstein HC, Vermersch P, et al. Atrophy of medial temporal lobes on MRI in “probable” Alzheimer's disease and normal ageing: diagnostic value and neuropsychological correlates. J Neurol Neurosurg Psychiatry Oct 1992;55(10):967-72.
128
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
[47] Dewey J, Hana G, Russell T, Price J, McCaffrey D, Harezlak J, et al. Reliability and validity of MRI-based automated volumetry software relative to auto-assisted manual measurement of subcortical structures in HIV-infected patients from a multisite study. Neuroimage Jul 15 2010;51(4):1334-44. [48] Rohrer JD, Warren JD, Modat M, Ridgway GR, Douiri A, Rossor MN, et al. Patterns of cortical thinning in the language variants of frontotemporal lobar degeneration. Neurology May 5 2009;72(18):1562-9. [49] Klauschen F, Goldman A, Barra V, Meyer-Lindenberg A, Lundervold A. Evaluation of automated brain MR image segmentation and volumetry methods. Hum Brain Mapp Apr 2009;30(4):1310-27.
[50] Murphy TH, Corbett D. Plasticity during stroke recovery: from synapse to behaviour. Nat Rev Neurosci Dec 2009;10(12):861-72. [51] Seidler RD, Noll DC. Neuroanatomical correlates of motor acquisition and motor transfer. J Neurophysiol Apr 2008;99(4):1836-45. [52] Ward NS. Mechanisms underlying recovery of motor function after stroke. Postgrad Med J Aug 2005;81(958):510-4. [53] Van Paesschen W. Qualitative and quantitative imaging of the hippocampus in mesial temporal lobe epilepsy with hippocampal sclerosis. Neuroimaging Clin N Am Aug 2004;14(3):373-400 [vii].
Contents lists available at SciVerse ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Changes in regional brain volume three months after stroke Amy Brodtmann a, b,⁎, Heath Pardoe a, b, Qi Li a, Renee Lichter a, b, Leif Ostergaard c, Toby Cumming a, b a b c
Florey Neuroscience Institutes, Melbourne Brain Centre, 245 Burgundy Street, Heidelberg, 3081, Australia University of Melbourne, Parkville, 3050, Australia Aarhus University, Copenhagen, Denmark
a r t i c l e
i n f o
Article history: Received 13 January 2012 Received in revised form 13 June 2012 Accepted 10 July 2012 Available online 1 August 2012 Keywords: Cortical thickness Post-stroke atrophy Remodeling
a b s t r a c t Introduction: Little is known about changes in regional brain volume after stroke. We investigated cortical thickness changes over 3 months in a group of stroke patients compared with controls. Material and methods: Patients with acute hemispheric stroke were studied within 3 h of stroke onset and serially over 3 months. We compared the acute and 3 month scans with independently acquired control images. High resolution isotropic T1 images were analyzed using FreeSurfer V5.0, comparing regional average cortical thickness, hippocampal and thalamic volumes. Stroke patient results were analyzed separately for ipsilesional and contralesional regions, whereas control results were averaged across hemisphere. Percentage change scores between the two time points were computed for each participant, and paired sample t-tests were used to assess significant change. Results: 12 stroke patients (9 men, 7 left-hemispheric, mean age = 65.1 years) and 10 control participants (5 men, mean age = 67.2 years) were included. There were no significant differences between the 2 time points in global or regional average cortical thickness, or hippocampal and thalamic volume estimates for control subjects. Regional variability in patient data was demonstrated, particularly cortical thickness increases in contralesional paracentral, superior frontal and insular regions, areas known to be activated in functional MRI studies of motor recovery. A significant reduction in thalamic volume was also found, most apparent ipsilesionally. Conclusions: Post-stroke changes in regional cortical thickness are demonstrable even over short time-frames. Contralesional cortical thickness increases may represent compensatory mechanisms. Significant reductions in thalamic volume may represent evidence of early post-stroke atrophy. Further studies are required to confirm and extend these preliminary results. © 2012 Elsevier B.V. All rights reserved.
1. Introduction 1.1. Post-stroke reorganization Significant dynamic reorganization of distributed networks is well described after stroke, especially in the functional neuroimaging literature [1–3]. This reorganization has been described in the visual [4], language [5], attention and sensory networks [6], but the majority of researchers have focused on motor recovery and associated motor cortical regions due to the importance of motor recovery to functional independence [7]. Activation patterns are usually characterized by early utilization of homologous regions in the contralesional (intact) hemisphere (contralesional recruitment) [1–3], followed by expansion of cortical representation of the damaged area of the cortex into adjacent areas (peri-infarct), with subsequent activation of other cortical and ⁎ Corresponding author at: Florey Neuroscience Institutes, Melbourne Brain Centre, 245 Burgundy Street, Heidelberg 3084, Australia. Tel.: + 61 3 9035 7004; fax: + 61 3 9035 7404. E-mail address: [email protected] (A. Brodtmann). 0022-510X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2012.07.019
sub-cortical regions distant to the lesion [2,8–10]. There is strong evidence that restitution of activation to the ipsilesional hemisphere is associated with a better functional outcome, particularly in the chronic post-stroke phases [2,3,11]. Most of these studies have used positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) techniques to document changes, but more recently methods such as diffusion tensor imaging and connectivity analyses have been used [12–16]. 1.2. Cortical thickness changes after stroke Cortical thickness estimates have been used to document dynamic changes in the human brain, both in normal learning and in pathological states, particularly neurodegenerative processes such as Alzheimer's disease (AD) and frontotemporal dementia. Researchers have reported significant regional volume increases in people intensively learning new tasks, such as London cab drivers learning the maps of London [17,18], melody recognition in expert musicians [19], and students acquiring expertise in a new language [20]. Surprisingly, cortical thickness analyses have not been used to chart cortical plasticity after stroke,
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
123
despite the fact that there is strong evidence that such functional changes are taking place. There is now some evidence that regional volumes may decline in the long-term after stroke, particularly in patients who subsequently develop cognitive decline [21,22].
significant psychiatric disease, had no significant carotid artery stenosis on carotid duplex Doppler ultrasound, and no stroke on MRI.
1.3. Atrophy and cognitive decline
Patients: 12 patients were scanned. 10 patients had 3D T1-weighted whole brain FSPGR images acquired on a GE Signa Excite 3 T MRI scanner with the following acquisition parameters: echo time TE= 3 ms, repetition time TR= 625 ms, flip angle=20°, voxel resolution = 0.9375 mm in plane, slice thickness= 1.3 mm. Two stroke patients were scanned on a GE Genesis Signa 1.5 T MRI scanner: TE= 4.2 ms, TR= 830 ms, flip angle = 20°, voxel resolution = 0.9375 mm in plane, slice thickness= 1.5 mm. Controls: 3D T1-weighted whole brain MPRAGE images were acquired on a Siemens TRIO MRI scanner with the following acquisition parameters: echo time TE = 2.55 ms, inversion time TI = 900 ms, repetition time TR = 1900 ms, flip angle = 9°, voxel resolution = 1 mm isotropic.
Recent research has indicated that consistent patterns of regional cortical thickness change are strongly associated with AD [23–25]. There is now a large body of work examining the association between brain volume changes and disease diagnosis and progression in many dementia syndromes. There is evidence from animal models that cortical thickness changes occur in the post-stroke period. Using a rat stroke model, Karl et al. found decreases in cortical thickness, volume, and neural density extending far beyond the stroke infarct [26]. There is evidence in humans that some brain regions—such as hippocampi and thalami—exhibit disproportionate atrophy after stroke [27,28]. Cognitive impairment and dementia are common after stroke [29], with vascular dementia accounting for about one-fifth of all dementia cases [30]. Yet we still know very little about whether brain volume loss—a hallmark of dementia—occurs after stroke, and whether such atrophy is related to cognitive decline. Magnetic resonance imaging (MRI) markers of structural brain aging (such as lower total brain volume, hippocampal volume or increasing white matter hyperintensity load) and performance on neuropsychological tests of memory and executive function are powerful predictors of dementia in the general population [31–34]. In contrast to pathologically-based autopsy studies, using brain volume measures to investigate the association between stroke and dementia permits a longitudinal approach. 1.4. Current study Dynamic remodeling after stroke may lead to changes in cortical thickness, but these have not been demonstrated. Minimal evidence hints at some post-stroke regional atrophy, but this is also unclear [21,22,28]. Despite the hundreds of longitudinal stroke studies using MRI, there have been no reports of cortical thickness analyses using high-resolution regional estimates such as FreeSurfer [35,36]. We performed an exploratory analysis of cortical thickness changes in order to address the feasibility of volume comparisons between the hyperacute post-stroke periods and more chronic time points. We investigated cortical thickness changes over a 3 month period in both a group of stroke patients and healthy controls of a comparable age. We compared the regional volumes of patients studied acutely with their scans performed at 3 months post-stroke. We hypothesized that there would be a decline in the regional volume of the hippocampi and thalami of stroke patients, but did not expect significant cortical thickness changes in either stroke or control participants over this short time frame. 2. Materials and methods 2.1. Participants Patients were prospectively recruited into a thrombolysis study using MRI imaging in the acute and chronic post-stroke periods to establish recanalization rates (author L.O.). They were included if: they presented with a first-ever acute middle cerebral artery (MCA) territory stroke, were able to be studied with MRI within 2 h of stroke onset, were previously independent without cognitive decline, and were able to be scanned over a 3 month period. Patients were studied within 2 h and serially over 3 months (at study inclusion, 3 h, 24 h, 1 week, 6 weeks and 12 weeks). We compared the 2 hour and 3 month scans with independently acquired control images, also taken 3 months apart. Healthy control participants were included if: they were aged 60–90 years, had no prior neurological or
2.2. Imaging
2.3. Analysis The structural scans were processed using FreeSurfer V 5.0 with default processing settings. Processed images were visually inspected and skull stripping and white matter edits were made where appropriate. Cortical thickness measures were averaged at the lobar level using the inbuilt FreeSurfer cortical parcellation procedure (“aparc”). FreeSurfer produces cortical thickness maps of 35 cortical regions (measured in millimeters) as well as volume estimates of structures, including the thalami (in mm3). We compared average cortical thickness in each of the cortical regions as well as hippocampal and thalamic volumes. For stroke patients, results were divided into ipsilesional and contralesional regions, in order to compare stroke hemisphere with non-stroke hemisphere. To identify whether stroke severity was related to brain volume change, correlations were computed between acute National Institutes of Health Stroke Score (NIHSS) score and volume measures. For controls, cortical thickness and volume measures were averaged across left and right hemispheres. Paired-sample t-tests were used to establish whether the measures of brain volume changed significantly over 3 months. Given the differences in regional thickness between individuals, particularly with increasing age, we elected to use change scores rather than absolute values. This was done to minimize the problems associated with cortical thickness differences that occur as part of normal aging [25,37]. To account for individual variability in baseline volumes, baseline and 3 month data were used to compute percentage change scores for each participant. Group differences in mean percentage change score were analyzed using independent t-tests. The significance level of α = 0.05 was not adjusted for multiple testing, and results from this preliminary study should be interpreted with this in mind. Individual images were inspected to check the accuracy of the automated segmentation. Given the very acute time frame of the first imaging session, the stroke site was barely visible, as ischemic infarction in the hyperacute setting is not associated with the accumulation of significant edema. The presence of edema becomes problematic in the 24–72 hour period, hence the choice of the 2 hour scan for the comparison. We performed the analyses with the lesions unmasked, particularly given only 2 were cortical. Qualitative inspection of the cortical thickness results showed that the modeling of the gray/ white matter and outer gray matter surfaces exterior to the lesion was unaffected by the presence of the lesion. 3. Results Twelve stroke patients (9 men, 7 left-hemispheric, mean age = 65.1 years, range 45–74 years) were included. Ten strokes were
124
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
subcortical, and 2 cortical. Five patients had no vessel occlusion on acute and follow-up imaging, and 7 had minimal or incomplete recanalization on the 3 month imaging. Mean NIHSSs at presentation was 9.7 ± 3.9 (range 3–16), and at 3 months was 2 ± 2.7 (range 0–9). Patients were scanned on average 98 min after symptom onset (range 50–143 min). Ten control participants (5 men, mean age = 67.2 years) were included. Edema had fully resolved by 3 months, but involution had occurred. Only 2 of the strokes involved cortical structures, and FreeSurfer correctly excluded these regions in its automated segmentation. Cortical thickness estimation was only performed on extant cortex. Within-group changes over time are outlined in Table 1. Stroke patients exhibited a significant increase in contralesional cortical thickness and significant reductions in the contralesional thalamus and hippocampus, as well as a significant reduction in the ipsilesional thalamus (see Fig. 1). A positive correlation between acute NIHSS and change in contralesional cortical thickness (r = 0.65, p = 0.021) indicated that more severe stroke was linked with greater cortical thickness increase. Data on percentage change in volume across time are detailed in the table and shown in Fig. 1. When compared directly with control results, there was a significant increase in the contralesional cortical thickness [t(20) = − 2.59, p = 0.018] and a significant decrease in the ipsilesional thalamus [t(15)= 2.49, p = 0.025]. Percentage change in FreeSurfer's 35 different cortical thickness regions is illustrated in Figs. 2–4. Contralesionally, significant increases were found in the superior parietal (p=0.049), middle temporal (p=0.031), temporal pole (p=0.038), paracentral (p = 0.005), insula (p = 0.005), precentral (p = 0.033) and superior frontal (p = 0.006)—see Figs. 2 and 5. Ipsilesionally, significant increases were found in the caudal middle frontal (p = 0.012) and rostral anterior cingulate (p = 0.002)—see Figs. 3 and 5. In controls, there were no regions with significant cortical thickness or volume change over time—see Fig. 4. 4. Discussion We found that hippocampal and thalamic volumes do decline in the 3 months post-stroke, although only the thalamic volume decreases were significant over such a short time-frame. These reductions in thalamic volume may represent evidence of early post-stroke atrophy. There was evidence that the volumetric changes are more marked in some patients, raising the possibility of using these volume changes to track individual decline. Interestingly, significant regional increases in cortical thickness were also apparent, most apparent contralesionally, but also evident in the ipsilesional anterior cingulate gyrus. These findings correlate with remodeling seen in fMRI and PET studies post-stroke, and confirm that post-stroke changes in regional cortical thickness are demonstrable even over short time-frames. These contralesional cortical thickness increases may represent compensatory mechanisms, but correlation with
Fig. 1. Percentage change in cortical thickness and regional brain volume between baseline and 3 months in stroke patients' contralesional (“contra”, dark gray) and ipsilesional (“ipsi”, light gray) hemispheres and controls (white, average both hemispheres): standard error bars are shown; *p b 0.05, **p b 0.01.
individual motor recovery was not performed given the small study numbers. Longitudinal brain volume changes have not been thoroughly examined following stroke. This is surprising, given the wealth of knowledge that exists on changes in brain volume over time in other conditions, such as schizophrenia [38] and dementia [39]. The few reported cross-sectional samples have yielded conflicting results. In 2000, Pohjasvaara et al. found an association between medial temporal lobe atrophy and dementia in 337 patients scanned 3 months after their first stroke [40]. In the Sydney Stroke Study, manual hippocampal volume estimates in 90 cerebrovascular patients (15 TIA) scanned 3–6 months after their event were not different from a group of control subjects [41]. Stebbins et al. grouped 91 ischemic stroke patients into those with no cognitive impairment and those with some impairment, and found greater gray matter volume reductions in the cognitively impaired group, predominantly thalamic [28]. Kraemer et al. identified delayed shrinkage of brain tissue that was not confined to peri-infarct regions in 10 stroke patients scanned between 1 and 4 years after stroke [21]. Remote changes were detected in the ipsilesional white matter and subcortical structures (striatum, thalamus). Observed atrophy occurred remotely in other vascular territories (e.g., the ipsilateral thalamus, supplied by posterior cerebral artery, in patients with MCA infarcts). Recently Nitkunan et al. found a significant difference in brain volume between the patients with small vessel disease and control subjects when compared across 2 time-points [22]. They corroborated the hippocampal and thalamic
Table 1 Change in cortical thickness and regional volume between baseline and 3 months. Cortical thickness in mm; thalamic and hippocampal volume in mm3; significant p-values in bold. Baseline mean value
Stroke—contralesional
Stroke—ipsilesional
Controls
Cortical thickness Thalamus Hippocampus Cortical thickness Thalamus Hippocampus Cortical thickness Thalamus Hippocampus
2.49 6770 4769 2.51 6613 4662 2.51 6169 3765
3 month mean value
2.60 6395 4556 2.54 6027 4584 2.52 6095 3786
Mean change
0.10 −375 −213 0.03 −586 −77 0.01 −74 21
SD
0.09 564 320 0.09 659 462 0.05 244 169
95% CI Lower
Upper
0.04 −733 −416 −0.03 −1005 −371 −0.02 −249 −100
0.16 −17 −9 0.09 −168 216 0.05 101 142
t
p
3.80 −2.30 −2.30 1.25 −3.08 −0.58 0.86 −0.96 0.40
0.003 0.042 0.042 0.24 0.010 0.57 0.41 0.36 0.70
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
125
Fig. 2. Stroke patients: percentage change (y-axis) in cortical thickness regions between baseline and 3 months in contralesional hemisphere (standard error bars are shown; *p b 0.05, **p b 0.01).
volume reductions, and suggested that these volume reductions are apparent very early after ischemic infarction. These conflicting results may have arisen from analyses at different post-stroke periods, the use of different field strengths for image acquisition, the naturally disparate and heterogeneous nature of the stroke population, and the different image analysis methods. Automated hippocampal volume analysis is known to systematically overestimate hippocampal volumes when compared to manual estimation methods [42,43]; images acquired in 1.5 T scanners may have poorer gray–white matter contrast to enable accurate surface modeling required for cortical thickness analyses [44]. Kraemer et al. and Stebbins et al. used modified optimized voxel-based morphometric methods [45]; in the latter paper the lesion site was masked,
and the authors subsequently assessed regional differences in gray matter volume, doing voxel-by-voxel comparisons for selected regions [21,28]. Nitkunan et al. used SIENA from the FSL suite (Structural Image Evaluation, using Normalization, of Atrophy; www.fmrib.ox.ac. uk/fsl) [22]. Pohjasvaara et al. [40] rated brain atrophy from 0 to 3 (none, mild, moderate, severe) according to Scheltens et al. [46]. Sachdev et al. performed manual hippocampal estimations [41]. None of these researchers used surface-based cortical thickness estimates, such as FreeSurfer, which has been found to be reliably sensitive to regional gray matter loss in other disease states [47,48], and has also been demonstrated to have good performance stability in both simulated and real MR brain data sets [49]. The inclusion or exclusion of the hippocampal and thalamic gray matter in the global average thickness
Fig. 3. Stroke patients: percentage change (y-axis) in cortical thickness regions between baseline and 3 months in ipsilesional hemisphere (standard error bars are shown; *p b 0.05, **p b 0.01).
126
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
Fig. 4. Control participants: percentage change (y-axis) in cortical thickness regions between baseline and 3 months (standard error bars are shown).
would significantly skew these estimates, given that the hippocampi and thalami have been demonstrated to decline in regional analyses. Clearly, there needs to be more work in evaluating and assessing different analysis methods (e.g., voxel-based versus surface-based, etc.) in the stroke population. In addition, scanning at disparate time-points would have had very significant effects. More incident atrophy would be expected 4 years after stroke than 1—particularly given that these patients often have co-existing small vessel disease with probable associated accelerated atrophy [22]—and extensive post-stroke remodeling would be expected at a 3 month time-point when compared to 6 months, when the majority of reorganization will have occurred [50]. The contralesional increases in cortical thickness are also interesting, particularly given that these are areas known to be activated
Fig. 5. Regional cortical thickness maps. Areas of significant cortical thickness increase overlaid onto FreeSurfer generated average brain. A = contralesional cortical thickness increases; B = ipsilesional cortical thickness increases. Regions are labeled via FreeSurfer anatomical convention. Some regions are numbered for identification: 1 = rostral anterior cingulate, 2 = middle frontal gyrus, 3 = superior frontal gyrus, 4 = precentral gyrus, 5 = superior parietal lobule, 6 = middle temporal gyrus, 7 = mesial paracentral regions including anterior cingulate cortex.
in functional MRI studies of motor recovery [2,3,7–10,14]. The ipsilesional increase in anterior cingulate thickness is also of note, as cingulate regions are known to be involved in motor learning and recruited with tasks of increasing complexity [51], as well as being important in recovery after stroke affecting the motor regions [52]. It is possible that the thicknesses measured at the three month post-stroke stage represent a normalization following the acute phase. This “pseudo-normalization” has been seen in other disease states, such as hippocampal volume increases followed by subsequent atrophy after temporal lobe seizures [53], and the initial regional volume increases that precede atrophy in schizophrenia. However, most of the changes were contralesional, and the acute scan was done less than 2 h from symptom onset, before the development of significant edema. We designed the study to assess the feasibility of volume comparisons between the acute post-stroke periods and more chronic time-points. Clearly, there are small numbers in this study, and correlation with individual patient recovery scores or cognitive measures was not performed given that only 12 stroke patients were examined. There is still no agreement on how lesions—such as strokes—affect the cortical thickness measures. Stroke patients and control participants, while prospectively recruited over a similar time period, scanned using the same field strength and analyzed centrally, were scanned at 2 different centers, but results from the same scanner were used for the within subject analyses. Qualitative inspection of the cortical thickness results showed that the modeling of the gray/white matter and outer gray matter surfaces exterior to the lesion was unaffected by the presence of the lesion. We are in the process of performing further analyses with lesions masked and unmasked, as well as comparing regional volume estimate techniques (cortical thickness versus voxel-based methods) in order to establish the best methods for longitudinal regional volume measures after stroke. 5. Conclusions Hippocampal and thalamic volumes do decline in the 3 months post-stroke. There is significant individual variability, suggesting larger numbers and longer time-frames are needed. Significant
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
regional increases in cortical thickness are also evident, most apparent contralesionally. Post-stroke changes in the regional cortical thickness are demonstrable even over short time-frames. Contralesional cortical thickness increases may represent compensatory mechanisms. Significant reductions in thalamic volume may represent evidence of early post-stroke atrophy. Further studies are required to evaluate the longitudinal nature of these preliminary findings. Conflict of interest The authors have no conflict of interest to declare. Acknowledgments This work was supported by grants from the Brain Foundation, the JO & JR Wicking Trust, and by the Sidney and Fiona Myer Family Foundation. The Florey Neuroscience Institutes acknowledges the strong support from the Victorian Government and in particular the funding from the Operational Infrastructure Support Grant. Enormous thanks to Soren Christensen for assistance with accessing the stroke scans. References [1] Rehme AK, Fink GR, von Cramon DY, Grefkes C. The role of the contralesional motor cortex for motor recovery in the early days after stroke assessed with longitudinal FMRI. Cereb Cortex Apr 2011;21(4):756-68. [2] Calautti C, Jones PS, Naccarato M, Sharma N, Day DJ, Bullmore ET, et al. The relationship between motor deficit and primary motor cortex hemispheric activation balance after stroke: longitudinal fMRI study. J Neurol Neurosurg Psychiatry Jul 2010;81(7):788-92. [3] Calautti C, Leroy F, Guincestre JY, Baron JC. Displacement of primary sensorimotor cortex activation after subcortical stroke: a longitudinal PET study with clinical correlation. Neuroimage Aug 2003;19(4):1650-4. [4] Brodtmann A, Puce A, Darby D, Donnan G. Serial functional imaging poststroke reveals visual cortex reorganization. Neurorehabil Neural Repair Feb 2009;23(2): 150-9. [5] Fridriksson J, Richardson JD, Fillmore P, Cai B. Left hemisphere plasticity and aphasia recovery. Neuroimage Apr 2 2012;60(2):854-63. [6] Corbetta M, Shulman GL. Spatial neglect and attention networks. Annu Rev Neurosci 2011;34:569-99. [7] Calautti C, Baron JC. Functional neuroimaging studies of motor recovery after stroke in adults: a review. Stroke Jun 2003;34(6):1553-66. [8] Marshall RS, Zarahn E, Alon L, Minzer B, Lazar RM, Krakauer JW. Early imaging correlates of subsequent motor recovery after stroke. Ann Neurol May 2009;65(5): 596-602. [9] Askim T, Indredavik B, Vangberg T, Haberg A. Motor network changes associated with successful motor skill relearning after acute ischemic stroke: a longitudinal functional magnetic resonance imaging study. Neurorehabil Neural Repair Mar– Apr 2009;23(3):295-304. [10] Calautti C, Naccarato M, Jones PS, Sharma N, Day DD, Carpenter AT, et al. The relationship between motor deficit and hemisphere activation balance after stroke: a 3 T fMRI study. Neuroimage Jan 1 2007;34(1):322-31. [11] Ward NS, Newton JM, Swayne OB, Lee L, Thompson AJ, Greenwood RJ, et al. Motor system activation after subcortical stroke depends on corticospinal system integrity. Brain Mar 2006;129(Pt 3):809-19. [12] Rehme AK, Eickhoff SB, Rottschy C, Fink GR, Grefkes C. Activation likelihood estimation meta-analysis of motor-related neural activity after stroke. Neuroimage Feb 1 2011;59(3):2771-82. [13] Wang LE, Tittgemeyer M, Imperati D, Diekhoff S, Ameli M, Fink GR, et al. Degeneration of corpus callosum and recovery of motor function after stroke: a multimodal magnetic resonance imaging study. Hum Brain Mapp Oct 22 2011 [Electronic publication ahead of print]. [14] Grefkes C, Fink GR. Reorganization of cerebral networks after stroke: new insights from neuroimaging with connectivity approaches. Brain May 2011;134(Pt 5): 1264-76. [15] Rehme AK, Eickhoff SB, Wang LE, Fink GR, Grefkes C. Dynamic causal modeling of cortical activity from the acute to the chronic stage after stroke. Neuroimage Apr 1 2011;55(3):1147-58. [16] Park CH, Chang WH, Ohn SH, Kim ST, Bang OY, Pascual-Leone A, et al. Longitudinal changes of resting-state functional connectivity during motor recovery after stroke. Stroke May 2011;42(5):1357-62. [17] Maguire EA, Woollett K, Spiers HJ. London taxi drivers and bus drivers: a structural MRI and neuropsychological analysis. Hippocampus 2006;16(12):1091-101. [18] Maguire EA, Gadian DG, Johnsrude IS, Good CD, Ashburner J, Frackowiak RS, et al. Navigation-related structural change in the hippocampi of taxi drivers. Proc Natl Acad Sci U S A Apr 11 2000;97(8):4398-403.
127
[19] Foster NE, Zatorre RJ. Cortical structure predicts success in performing musical transformation judgments. Neuroimage Oct 15 2010;53(1):26-36. [20] Stein M, Federspiel A, Koenig T, Wirth M, Strik W, Wiest R, et al. Structural plasticity in the language system related to increased second language proficiency. Cortex Apr 2012;48(4):458-65. [21] Kraemer M, Schormann T, Hagemann G, Qi B, Witte OW, Seitz RJ. Delayed shrinkage of the brain after ischemic stroke: preliminary observations with voxel-guided morphometry. J Neuroimaging Jul 2004;14(3):265-72. [22] Nitkunan A, Lanfranconi S, Charlton RA, Barrick TR, Markus HS. Brain atrophy and cerebral small vessel disease: a prospective follow-up study. Stroke Jan 2011;42(1):133-8. [23] Du AT, Schuff N, Kramer JH, Rosen HJ, Gorno-Tempini ML, Rankin K, et al. Different regional patterns of cortical thinning in Alzheimer's disease and frontotemporal dementia. Brain Apr 2007;130(Pt 4):1159-66. [24] Julkunen V, Niskanen E, Koikkalainen J, Herukka SK, Pihlajamäki M, Hallikainen M, et al. Differences in cortical thickness in healthy controls, subjects with mild cognitive impairment, and Alzheimer's disease patients a longitudinal study. J Alzheimers Dis 2010;21(4):1141-51. [25] Salat DH, Buckner RL, Snyder AZ, Greve DN, Desikan RS, Busa E, et al. Thinning of the cerebral cortex in aging. Cereb Cortex Jul 2004;14(7):721-30. [26] Karl JM, Alaverdashvili M, Cross AR, Whishaw IQ. Thinning, movement, and volume loss of residual cortical tissue occurs after stroke in the adult rat as identified by histological and magnetic resonance imaging analysis. Neuroscience Sep 29 2010;170(1):123-37. [27] Pohjasvaara T, Mantyla R, Salonen O, Aronen HJ, Ylikoski R, Hietanen M, et al. How complex interactions of ischemic brain infarcts, white matter lesions, and atrophy relate to poststroke dementia. Arch Neurol Sep 2000;57(9):1295-300. [28] Stebbins GT, Nyenhuis DL, Wang C, Cox JL, Freels S, Bangen K, et al. Gray matter atrophy in patients with ischemic stroke with cognitive impairment. Stroke Mar 2008;39(3):785-93. [29] Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol Nov 2009;8(11):1006-18. [30] Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, et al. Global prevalence of dementia: a Delphi consensus study. Lancet Dec 17 2005;366(9503):2112-7. [31] Jack Jr CR, Lowe VJ, Weigand SD, Wiste HJ, Senjem ML, Knopman DS, et al. Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer's disease: implications for sequence of pathological events in Alzheimer's disease. Brain May 2009;132(Pt 5):1355-65. [32] Jack Jr CR, Shiung MM, Weigand SD, O'Brien PC, Gunter JL, Boeve BF, et al. Brain atrophy rates predict subsequent clinical conversion in normal elderly and amnestic MCI. Neurology Oct 25 2005;65(8):1227-31. [33] Seshadri S, Wolf PA, Beiser A, Elias MF, Au R, Kase CS, et al. Stroke risk profile, brain volume, and cognitive function: the Framingham Offspring Study. Neurology Nov 9 2004;63(9):1591-9. [34] Elias MF, Beiser A, Wolf PA, Au R, White RF, D'Agostino RB. The preclinical phase of Alzheimer disease: a 22-year prospective study of the Framingham Cohort. Arch Neurol Jun 2000;57(6):808-13. [35] Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis. II: inflation, flattening, and a surface-based coordinate system. Neuroimage Feb 1999;9(2):195-207. [36] Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage Feb 1999;9(2):179-94. [37] Fjell AM, Westlye LT, Amlien I, Espeseth T, Reinvang I, Raz N, et al. High consistency of regional cortical thinning in aging across multiple samples. Cereb Cortex Sep 2009;19(9):2001-12. [38] Pantelis C, Velakoulis D, McGorry PD, Wood SJ, Suckling J, Phillips LJ, et al. Neuroanatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet Jan 25 2003;361(9354):281-8. [39] Weiner MW, Aisen PS, Jack Jr CR, Jagust WJ, Trojanowski JQ, Shaw L, et al. The Alzheimer's disease neuroimaging initiative: progress report and future plans. Alzheimers Dement May 2010;6(3):202-11 [e7]. [40] Pohjasvaara T, Mantyla R, Salonen O, Aronen HJ, Ylikoski R, Hietanen M, et al. MRI correlates of dementia after first clinical ischemic stroke. J Neurol Sci Dec 1 2000;181(1–2):111-7. [41] Sachdev PS, Chen X, Joscelyne A, Wen W, Altendorf A, Brodaty H. Hippocampal size and dementia in stroke patients: the Sydney stroke study. J Neurol Sci Sep 15 2007;260(1–2):71-7. [42] Konrad C, Ukas T, Nebel C, Arolt V, Toga AW, Narr KL. Defining the human hippocampus in cerebral magnetic resonance images—an overview of current segmentation protocols. Neuroimage Oct 1 2009;47(4):1185-95. [43] Tae WS, Kim SS, Lee KU, Nam EC, Kim KW. Validation of hippocampal volumes measured using a manual method and two automated methods (FreeSurfer and IBASPM) in chronic major depressive disorder. Neuroradiology Jul 2008;50(7): 569-81. [44] Jovicich J, Czanner S, Han X, Salat D, van der Kouwe A, Quinn B, et al. MRI-derived measurements of human subcortical, ventricular and intracranial brain volumes: reliability effects of scan sessions, acquisition sequences, data analyses, scanner upgrade, scanner vendors and field strengths. Neuroimage May 15 2009;46(1): 177-92. [45] Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS. A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage Jul 2001;14(1 Pt 1):21-36. [46] Scheltens P, Leys D, Barkhof F, Huglo D, Weinstein HC, Vermersch P, et al. Atrophy of medial temporal lobes on MRI in “probable” Alzheimer's disease and normal ageing: diagnostic value and neuropsychological correlates. J Neurol Neurosurg Psychiatry Oct 1992;55(10):967-72.
128
A. Brodtmann et al. / Journal of the Neurological Sciences 322 (2012) 122–128
[47] Dewey J, Hana G, Russell T, Price J, McCaffrey D, Harezlak J, et al. Reliability and validity of MRI-based automated volumetry software relative to auto-assisted manual measurement of subcortical structures in HIV-infected patients from a multisite study. Neuroimage Jul 15 2010;51(4):1334-44. [48] Rohrer JD, Warren JD, Modat M, Ridgway GR, Douiri A, Rossor MN, et al. Patterns of cortical thinning in the language variants of frontotemporal lobar degeneration. Neurology May 5 2009;72(18):1562-9. [49] Klauschen F, Goldman A, Barra V, Meyer-Lindenberg A, Lundervold A. Evaluation of automated brain MR image segmentation and volumetry methods. Hum Brain Mapp Apr 2009;30(4):1310-27.
[50] Murphy TH, Corbett D. Plasticity during stroke recovery: from synapse to behaviour. Nat Rev Neurosci Dec 2009;10(12):861-72. [51] Seidler RD, Noll DC. Neuroanatomical correlates of motor acquisition and motor transfer. J Neurophysiol Apr 2008;99(4):1836-45. [52] Ward NS. Mechanisms underlying recovery of motor function after stroke. Postgrad Med J Aug 2005;81(958):510-4. [53] Van Paesschen W. Qualitative and quantitative imaging of the hippocampus in mesial temporal lobe epilepsy with hippocampal sclerosis. Neuroimaging Clin N Am Aug 2004;14(3):373-400 [vii].