You are here

Article Text

Article menu
Neuromuscular
Research paper
Cortical thickness in ALS: towards a marker for upper motor neuron involvement
  1. Renée Walhout1,
  2. Henk-Jan Westeneng1,
  3. Esther Verstraete1,
  4. Jeroen Hendrikse2,
  5. Jan H Veldink1,
  6. Martijn P van den Heuvel3,
  7. Leonard H van den Berg1
  1. 1Department of Neurology, Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht, The Netherlands
  2. 2Department of Radiology, Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht, The Netherlands
  3. 3Department of Psychiatry, Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht, The Netherlands
  1. Correspondence to Leonard H van den Berg, Department of Neurology, G03.228, University Medical Center Utrecht, PO Box 85500, Utrecht 3508 GA, The Netherlands L.H.vandenBerg{at}umcutrecht.nl

Abstract

Objective Examine whether cortical thinning is a disease-specific phenomenon across the spectrum of motor neuron diseases in relation to upper motor neuron (UMN) involvement.

Methods 153 patients (112 amyotrophic lateral sclerosis (ALS), 19 patients with a clinical UMN phenotype, 22 with a lower motor neuron (LMN) phenotype), 60 healthy controls and 43 patients with an ALS mimic disorder were included for a cross-sectional cortical thickness analysis. Thirty-nine patients with ALS underwent a follow-up scan. T1-weighted images of the brain were acquired using a 3 T scanner. The relation between cortical thickness and clinical measures, and the longitudinal changes were examined.

Results Cortical thickness of the precentral gyrus (PCG) was significantly reduced in ALS (p=1.71×10−13) but not in mimic disorders (p=0.37) or patients with an LMN phenotype (p=0.37), as compared to the group of healthy controls. Compared to patients with ALS, patients with a UMN phenotype showed an even lower PCG cortical thickness (p=1.97×10−3). Bulbar scores and arm functional scores showed a significant association with cortical thickness of corresponding body regions of the motor homunculus. Longitudinal analysis revealed a decrease of cortical thickness in the left temporal lobe of patients with ALS (parahippocampal region p=0.007 and fusiform cortex p=0.001).

Conclusions PCG cortical thinning was found to be specific for motor neuron disease with clinical UMN involvement. Normal levels of cortical thickness in mimic disorders or LMN phenotypes suggest that cortical thinning reflects pathological changes related to UMN involvement. Progressive cortical thinning in the temporal lobe suggests recruitment of non-motor areas, over time.

  • ALS
  • MOTOR NEURON DISEASE
  • MRI

https://doi.org/10.1136/jnnp-2013-306839

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    The diagnosis of amyotrophic lateral sclerosis (ALS) requires evidence of upper motor neuron (UMN) and lower motor neuron (LMN) involvement.1 Evidence of UMN degeneration can be detected by physical examination; however, in the early phase of the disease or in cases with severe LMN loss, clinical UMN signs, such as hyper-reflexia or spasticity, may be difficult to determine objectively.2 This prevents neurologists from distinguishing ALS from the pure LMN phenotype, progressive muscular atrophy and other disorders that closely resemble its clinical presentation, so-called ALS mimic disorders,3 ,4 such as multifocal motor neuropathy5 or inclusion body myositis.6 Objective methodology to detect LMN loss has been developed using neurophysiological techniques,7 ,8 but equivalent methods for quantifying UMN involvement are currently limited.

    In the search for a UMN marker with MRI techniques, surface-based morphometry studies have indicated significant thinning of the primary motor cortex in patients with ALS.9––11 Change in grey matter architecture may be caused by motor neuron loss, although this has not been proven neuropathologically and may also be a secondary effect, due to LMN loss, muscle weakness or immobilisation.12 Inclusion of patients with ALS mimic disorders will give insight into whether cortical thinning is a disease-specific phenomenon. Furthermore, a large sample size of patients representing the full spectrum of motor neuron disease allows us to study the correspondence between the location of cortical atrophy on the motor cortex and the functionally impaired body region, which would also indicate that cortical thinning actually represents UMN loss.

    The overall aim of our study is to elucidate the association between cortical thickness measurements and clinical manifestations in patients with ALS and in appropriate controls, and to examine longitudinal changes in cortical thinning.

    Methods

    Participants

    A total of 153 patients with motor neuron disease (112 ALS, 19 patients with a ‘clinical UMN phenotype’ and 22 with a ‘clinical LMN phenotype’), were included in this study. These patients were recruited from the outpatient clinic of the neurology department at the University Medical Center of Utrecht. Demographic and clinical characteristics are listed in table 1. Patients with signs of UMN and LMN involvement were classified as having definite, probable or possible ALS using the revised El Escorial criteria after excluding other conditions.1 Patients with a ‘clinical UMN phenotype’ did not have any LMN findings on clinical examination or electromyography as described previously.13 Seven of these 19 patients had a disease duration longer than 4 years and were diagnosed as having primary lateral sclerosis.14 Patients with a ‘clinical LMN phenotype’ had no UMN findings on clinical examination.15 Two of these 22 patients had a disease duration of more than 4 years and were diagnosed with progressive muscular atrophy. We have purposely chosen to include patients with a ‘clinical LMN phenotype’ that were newly diagnosed to detect possible clinically undetectable UMN involvement in an early phase of the disease. Clinical characteristics, including site of disease onset and disease duration were recorded. Both sporadic and familial ALS cases were included in the study and their C9orf72 status was assessed.16 Patients with frontotemporal dementia were excluded from participation, as well as participants with a history of epilepsy, brain injury or psychiatric illnesses. To evaluate the functional status of our patients, the revised ALS Functional Rating Score (ALSFRS-R) was assessed in each patient. Disease progression rate was calculated using the ALSFRS-R score as follows: (48-ALSFRS-R)/disease duration (in months).

    View this table:
    Table 1

    Demographic and clinical characteristics of study participants

    In addition, data were acquired in 103 control subjects, including 60 healthy controls and 43 patients diagnosed with an ALS mimic disorder. The group of ALS mimic disorders included 14 patients with multifocal motor neuropathy, eight patients with inclusion body myositis and eight with spinobulbar muscular atrophy (Kennedy disease).17 Prior to diagnosing a mimic disorder, these patients had been referred because of a clinical suspicion of motor neuron disease. Demographic characteristics and diagnosis per individual patient are provided in the online supplementary data, table S1. There were no significant differences in age (Kruskall-Wallis, p=0.144) or male:female ratio (Pearson χ2, p=0.140) between patient groups and healthy controls.

    A follow-up scan was made of 39 patients with ALS who were able and willing to undergo a second scan. At follow-up an identical T1 scanning protocol (see for parameters below) was used. The median follow-up time was 5.5 months (range 3–12 months).

    The study was approved by the Medical Ethical Committee for research into humans of the University Medical Center Utrecht. Written informed consent was obtained from all participants, in line with the Declaration of Helsinki.

    Data acquisition

    High-resolution T1-weighted images were acquired using a 3 T Philips Achieve Medical Scanner. Acquisition parameters were: three-dimensional fast field echo using parallel imaging; repetition time/echo time=10/4.6 ms, flip angle 88 slice orientation sagittal, 0.75×0.75×0.8 mm voxel size, field of view (FOV)=160×240×240 mm and reconstruction matrix=200×320×320 covering whole brain.

    Data processing

    Cortical thickness was analysed with the Freesurfer image analysis suite (V.5.0.0, http://surfer.nmr.mgh.harvard.edu/). Cortical thickness was measured by computing the distance between the grey/white matter boundary and pial surface at each point (vertex) on the cortical mantle. The cortical surface was automatically parcellated into 68 anatomical regions to look for regional differences in cortical thickness. Second, all reconstructed individual data sets were resampled to an average anatomical surface and smoothed with a 10 mm full width at half maximum Gaussian kernel for a whole brain vertex-wise analysis. Image outputs after Freesurfer processing were visually inspected. Longitudinal changes in cortical thickness were assessed.18 ,19

    Data analysis

    Analysis I: region-of-interest comparison

    Based on previous observations,10 the precentral gyrus (PCG) was selected as region of interest to evaluate cortical thickness differences between groups. Region-wise comparison of cortical thickness was performed using general linear modelling included in the R software package for statistical computing (http://www.R-project.org, R 2.11.1 GUI 1.34). Cortical thickness values were corrected for age and average cortical thickness. Since there was no significant effect of handedness on cortical thickness, cortical thickness values were not corrected for handedness in the analysis.

    Analysis II: whole-brain analysis

    A whole-brain vertex-wise analysis was performed at the group level. To correct for multiple testing, we applied a cluster-wise correction, with a threshold of p=0.01 using Monte Carlo simulation (5000 iterations). This method was applied since it accounts for the degree of covariance between the large number of small neighbouring vertices, identifying significant contiguous clusters with an appropriate estimate of the size of these clusters.20

    Analysis III: association of cortical thickness with clinical measures in ALS

    The association of cortical thickness with clinical measures in the ALS patient group was assessed in both a region-wise and vertex-wise comparison. First, in the region-wise comparison, the associations of cortical thickness with clinical metrics (ALSFRS-R, subscores of bulbar, arm and leg function, disease progression rate) were analysed per cortical region in a linear model, including age and average cortical thickness as covariates. Regions showing a p<0.05 (false discovery rate (FDR) corrected) were considered significant. Second, a general linear model of cortical thickness was computed as a function of ALSFRS-R total scores, subscores and disease progression rate at every vertex on the surface, regressing out the effects of age and average cortical thickness. An exploratory threshold of p=0.01 (uncorrected for multiple testing) was used and resulting parametric maps were plotted on an inflated average cortical surface.

    Analysis IV: receiver operating characteristic

    A receiver operating characteristic (ROC) analysis was applied to explore the diagnostic test accuracy of cortical thickness measurements and to assign optimal cut-off values.

    Analysis V: longitudinal analysis

    A linear mixed effects model was used to assess the change of cortical thickness over time within patients with ALS,20 ,21 with ALSFRS-R, age at baseline, gender and disease duration included as covariates. Effects with a p value <0.05 (FDR corrected) were interpreted as statistically significant.

    Results

    Analysis I: region-of-interest comparison

    The mean cortical thickness of the PCG in patients with ALS (left=2.38 mm, right=2.36 mm) was revealed to be significantly lower compared to healthy controls (left=2.47 mm, right=2.45 mm), p=1.71×10−13 (left and right PCG averaged; figure 1A). As commonly reported in healthy controls,22 an asymmetry in overall levels of cortical thickness in the PCG was observed, with higher values in the left PCG, for both patients with ALS and healthy controls. This leftward asymmetry has been suggested to reflect functional lateralisation of limb dominance.22 The mean cortical thickness of the PCG in ALS mimics (left=2.45 mm, right=2.45 mm) appeared to be in the same range as normal controls (p=0.37) and was significantly different from patients with ALS (p=6.00×10−8). The observed effects were congruent with the results of cortical thickness analysis in the left and right hemisphere separately (ALS vs healthy controls: left p=3.13×10−10, right p=1.07×10−12; ALS vs mimics: left p=1.19×10−5, right p=3.09×10−8; mimics vs healthy controls: left p=0.19; right p=0.78). Patients with a clinical UMN phenotype had a significantly lower cortical thickness compared to patients with ALS (left=2.30 mm, right=2.29 mm, p=1.97×10−3), whereas cortical thickness in patients with a clinical LMN phenotype (left=2.46 mm, right=2.44 mm) was in the range of the healthy controls (p=0.37) and significantly different from patients with ALS (p=1.0×10−5). Within the mimic disorder group, there was no apparent clustering of any type of ALS mimic, as shown in figure 1B.

    Figure 1
    Figure 1

    Cortical thickness of the precentral gyrus in patients with motor neuron disease subtypes compared to mimic disorders and healthy controls. (A) Mean cortical thickness of the precentral gyrus (left and right averaged) was plotted per study group. Cortical thickness of patients with ALS was significantly reduced compared to patients with a clinical LMN phenotype, mimic disorders and healthy controls. Patients with a clinical UMN phenotype had a significantly lower cortical thickness compared to patients with ALS. The dashed horizontal line indicates the mean cortical thickness minus 2 SD of the control group (ie, controls and mimics combined). (B) Distribution of cortical thickness measurements according to the type of mimic disorder. There was no apparent clustering of any type of ALS mimic. UMN, upper motor neuron phenotype; ALS, amyotrophic lateral sclerosis; LMN, lower motor neuron phenotype; MIMIC, mimic disorder; CON, healthy control; MMN, multifocal motor neuropathy; SBMA, spinobulbar muscular atrophy (Kennedy disease); IBM, inclusion body myositis; SSMA, segmental spinal muscular atrophy.

    Considering the seven patients with the C9orf72 repeat expansion, mean cortical thickness of the PCG of these patients was similar compared to the other patients with ALS (left PCG=2.38 mm, right PCG=2.37 mm), and individual values were equally scattered within the range of the ALS patient group. However, this should be regarded as a preliminary finding since the group was relatively small.

    To examine whether patients with ALS with a low cortical thickness value were clinically different from the other patients with ALS, we divided the group of patients with ALS into two groups, with a threshold defined by the mean PCG cortical thickness minus 2 standard deviations (SD) from the control group (ie, controls and mimics combined) healthy controls. No clear clinical differences were found between patients with ALS with a cortical thickness value above versus below computed PCG cortical thickness threshold.

    Analysis II: whole-brain analysis

    Vertex-wise analysis confirmed cortical thinning of the PCG in patients with ALS, showing a decreased thickness compared to healthy controls and patients with a mimic disorder (figure 2). There were no differences in cortical thickness observed between healthy controls and patients with a mimic disorder. Figure 2 shows all significant results, there were no effects in the overlapping, non-visible regions (see also online supplementary data, figure S1). Supplemental figure S2 shows the results of the vertex-wise analysis with FDR correction.

    Figure 2
    Figure 2

    Regional differences in cortical thickness in a whole brain vertex-wise analysis. (A) The figure shows significant bilateral cortical thinning of the precentral gyrus in patients with amyotrophic lateral sclerosis (ALS) compared to healthy controls, regions with significant differences shown in red (p<0.01); (B) Results between patients with ALS and patients with a mimic disorder. Results corrected for multiple testing using Monte Carlo simulation cluster-wise correction, threshold of p=0.01. Note: the figure shows all significant results, there were no effects for non-visible regions (see also online supplementary data, figure S1).

    Analysis III: association of cortical thickness with clinical measures in ALS

    ALSFRS-R

    ALSFRS-R functional subscores of the bulbar region and of the arms showed a significant association with cortical thickness of corresponding body regions of the motor homunculus on the PCG (figure 3). Arm functional scores were positively associated with PCG cortical thickness, revealing significant effects in superior parts of the PCG (figure 3A). Bulbar functional scores showed a positive association of the PCG close to the lateral sulcus and temporal regions (p<0.01, uncorrected, figure 3B). No significant association between clinical leg scores and cortical thickness of the leg area, located at the medial section of the motor cortex, was observed.

    Figure 3
    Figure 3

    Vertex-wise analysis showing the association between cortical thickness and ALSFRS-R subscores in patients with amyotrophic lateral sclerosis. Positive correlations are shown in red (p<0.01, uncorrected). (A) Arm scores are positively associated with cortical thickness of the precentral gyrus, revealing significant effects in superior parts of the precentral gyrus. (B) Bulbar scores show a positive association with cortical thickness of the precentral gyrus close to the lateral sulcus and temporal regions.

    Progression rate and survival

    No significant effect of progression rate on cortical thickness was observed in the vertex-wise analysis. In the region-wise comparison, progression rate showed a marginal, negative association with the right inferior temporal gyrus (coefficient=0.07, p=0.025, corrected), indicating that a rapid clinical deterioration is related to progressive cortical thinning of this region. In addition, a positive association was found between progression rate and cortical thickness of the postcentral gyrus (left p=0.027; right p=0.025) and right paracentral gyrus (p=0.032). By the time of analysis, 59 of 112 patients with ALS were deceased. Mean survival was 27.7 months (median 25.8, SD 11.1, range 7.7–57.6 months). No clear association between PCG cortical thickness and survival was observed.

    Analysis IV: ROC

    To determine the accuracy of the use of cortical thickness in distinguishing patients with ALS from mimic disorders and healthy controls, an ROC analysis was performed. The sensitivity was 71.8% and the specificity was 75.9%; area under the ROC curve was 0.79 with an optimal cut-off value of 2.48.

    Analysis V: longitudinal analysis

    PCG cortical thickness showed no longitudinal changes (left: 11±24 µm/year; p=0.792. Right: −3±18 µm/year; p=0.901). Cortical thickness was found to decrease significantly over time in areas of the left temporal lobe (left parahippocampal (−103±28 µm/year; p=0.007, corrected) and fusiform cortex (−97±22 µm/year; p=0.001, corrected), online supplementary data, table S2). No significant association between cortical thickness changes and ALSFRS-R reduction was found (online supplementary data, table S3). A marginal association was observed between ALSFRS-R and PCG cortical thickness on both sides, but this effect did not reach correction for multiple testing (left: 5±2; uncorrected p=0.002; FDR corrected p=0.095. Right: 5±2; uncorrected p=0.003; FDR corrected p=0.095).

    Discussion

    In this study, we investigated the potential of cortical thickness as a marker for UMN degeneration in motor neuron diseases. We showed that cortical thickness of the primary motor regions was significantly reduced in motor neuron diseases with clinical UMN involvement. A second finding indicating that cortical thinning actually represents UMN involvement was the correspondence between the location of cortical atrophy on the motor cortex and the functionally impaired body region. During follow-up, patients with ALS showed cortical thinning in the temporal lobe of the left hemisphere, which might indicate that the disease process involves cortical thinning of non-motor brain regions over time instead of remaining confined to the primary motor system.

    It is sometimes challenging in clinical practice to differentiate patients with a motor neuron disease from patients with a mimic disorder. The diagnostic process could in these cases benefit from an objective marker for UMN involvement, similar to the use of electromyography for LMN involvement. Our comparison to mimics is an important step in the ongoing search for such an objective UMN marker, suggesting the observed cortical thinning in patients with ALS is not a secondary effect of muscle weakness and disuse, but rather a primary effect of disease. The results of our study show the potential of cortical thickness as a possible marker of UMN involvement in patients with ALS.

    The sample size allowed us to study clinical variables along the spectrum of ALS. Whole-brain analysis showed clear associations between functional disability of particular body regions and focal atrophy within corresponding regions on the motor cortex. These results support the notion that focal cortical changes correlate with clinical phenotypes based on the somatotopic representation on the motor cortex.23 Cortical atrophy may be related to UMN loss, and potentially reflect the loss of Betz cells within the cortex, as previously reported in postmortem studies,24 but a direct link has not been established. Our findings further strengthen the concept of cortical thinning as a reflection of clinical UMN involvement in motor neuron diseases.

    During follow-up, no progressive cortical thinning of the PCG was observed, supporting previous observations of longitudinal MR studies.10 Different explanations are possible; first, thinning of the primary motor cortex is an early event in ALS, occurring before distant non-motor regions become involved or second, lower levels of cortical thickness of motor regions form a risk factor for the development of ALS, potentially related to a genetic risk profile of the disorder. In parallel, the absence of progressive cortical thinning of the motor system tends to imply that progression of clinical motor symptoms in patients with ALS is predominantly a consequence of LMN degeneration rather than of UMN degeneration.

    Longitudinal analysis did reveal progressive cortical thinning over time in the left fusiform and parahippocampal cortex of the temporal lobe. Interestingly, the observed level of cortical thickness within the parahippocampal and fusiform cortex after 1 year of follow-up was found to be comparable to the absolute levels of cortical thinning of the precentral cortex at baseline (ca. 0.1 mm). This may indicate a comparable level of degeneration. Cortical atrophy of the right parahippocampal cortex has been noted before 25 and was also reported in a pathological study examining ALS.26 Furthermore, studies have shown decreased structural connectivity between the PCG and the left hippocampus,10 ,27 together with changes in functional connectivity between the primary sensorimotor cortex and the right parahippocampal and fusiform gyrus in patients with ALS.28 Progressive thinning of the temporal lobe in ALS might thus be linked to a connectivity-based degeneration over time of distant, non-motor areas. Future cross-modal studies combining functional and structural connectivity measurements,29–,31 and combining the assessment of cortical thickness changes with observed brain dysconnectivity effects29 ,32 would be of particular interest.

    Previous studies have reported UMN involvement in postmortem tissue of patients with a clinical LMN phenotype.33 ,34 However, no such effect could be replicated in our MR study, although in vivo neuroimaging studies allow for the examination of UMN involvement in a much earlier phase of the disease as compared to end-stage autopsy tissue. It would be of particular interest to study whether cortical thinning might eventually be observed during prospective follow-up studies in patients with clinical LMN phenotypes, and whether cortical thickness studies may serve as a (prognostic) marker for patients with progressive muscular atrophy. Future studies, including follow-up of larger patient groups with this phenotype will hopefully give us insight into these important questions.

    When effects were examined at the individual level, there was still a considerable overlap between patients with ALS, patients with a mimic disorder and healthy controls (see figure 1). The present results on cortical thickness should be viewed in light of other candidate imaging markers for UMN involvement in ALS. Diffusion-weighted imaging is currently regarded as a promising candidate marker,35 but a recent meta-analysis showed only a modest discriminatory capability of fractional anisotropy within the corticospinal tract, with a sensitivity of 0.65 and specificity of 0.67.36 ,37 Additionally, compared to the latter study, the present results on the diagnostic accuracy of cortical thickness appear promising, with an ROC analysis showing a sensitivity of 0.72 and specificity of 0.76. Combining different imaging techniques might improve the sensitivity and specificity for detecting UMN involvement in the individual patient.38 The international collaboration of the Neuroimaging Society in ALS (NISALS) provides the platform for such a multimodal approach, including cortical thickness measurements.39

    We conclude that cortical thinning of the primary motor cortex is present in motor neuron diseases with clinical UMN involvement, but not in LMN phenotypes or mimic disorders. Our findings thus indicate that cortical thinning reflects pathological changes related to UMN involvement.

    References

    1. ↵
      1. Brooks BR,
      2. Miller RG,
      3. Swash M,
      4. et al
      . El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord 2000;1:293–9.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Swash M
      . Why are upper motor neuron signs difficult to elicit in amyotrophic lateral sclerosis? J Neurol Neurosurg Psychiatry 2012;83:659–62.
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Visser J,
      2. van den Berg-Vos RM,
      3. Franssen H,
      4. et al
      . Mimic syndromes in sporadic cases of progressive spinal muscular atrophy. Neurology 2002;58:1593–6.
      OpenUrlCrossRef
    4. ↵
      1. Cellura E,
      2. Spataro R,
      3. Taiello AC,
      4. et al
      . Factors affecting the diagnostic delay in amyotrophic lateral sclerosis. Clin Neurol Neurosurg 2012;114:550–4.
      OpenUrlCrossRefPubMed
    5. ↵
      1. Vlam L,
      2. van der Pol WL,
      3. Cats EA,
      4. et al
      . Multifocal motor neuropathy: diagnosis, pathogenesis and treatment strategies. Nat Rev Neurol 2011;8:48–58.
      OpenUrlCrossRefPubMed
    6. ↵
      1. Dabby R,
      2. Lange DJ,
      3. Trojaborg W,
      4. et al
      . Inclusion body myositis mimicking motor neuron disease. Arch Neurol 2001;58:1253–6.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Neuwirth C,
      2. Nandedkar S,
      3. Stalberg E,
      4. et al
      . Motor unit number index (MUNIX): a novel neurophysiological technique to follow disease progression in amyotrophic lateral sclerosis. Muscle Nerve 2010;42:379–84.
      OpenUrlCrossRefPubMed
    8. ↵
      1. Shefner JM,
      2. Watson ML,
      3. Simionescu L,
      4. et al
      . Multipoint incremental motor unit number estimation as an outcome measure in ALS. Neurology 2011;77:235–41.
      OpenUrlCrossRef
    9. ↵
      1. Roccatagliata L,
      2. Bonzano L,
      3. Mancardi G,
      4. et al
      . Detection of motor cortex thinning and corticospinal tract involvement by quantitative MRI in amyotrophic lateral sclerosis. Amyotroph Lateral Scler 2009;10:47–52.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Verstraete E,
      2. Veldink JH,
      3. Hendrikse J,
      4. et al
      . Structural MRI reveals cortical thinning in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 2011;83:383–8.
      OpenUrlPubMed
    11. ↵
      1. Agosta F,
      2. Valsasina P,
      3. Riva N,
      4. et al
      . The cortical signature of amyotrophic lateral sclerosis. PLoS ONE 2012;7: e42816.
      OpenUrlCrossRefPubMed
    12. ↵
      1. Langer N,
      2. Hanggi J,
      3. Muller NA,
      4. et al
      . Effects of limb immobilization on brain plasticity. Neurology 2012;78:182–8.
      OpenUrlCrossRef
    13. ↵
      1. Brugman F,
      2. Veldink JH,
      3. Franssen H,
      4. et al
      . Differentiation of hereditary spastic paraparesis from primary lateral sclerosis in sporadic adult-onset upper motor neuron syndromes. Arch Neurol 2009;66:509–14.
      OpenUrlCrossRefPubMed
    14. ↵
      1. Gordon PH,
      2. Cheng B,
      3. Katz IB,
      4. et al
      . The natural history of primary lateral sclerosis. Neurology 2006;66:647–53.
      OpenUrlCrossRef
    15. ↵
      1. van den Berg-Vos RM,
      2. Visser J,
      3. Franssen H,
      4. et al
      . Sporadic lower motor neuron disease with adult onset: classification of subtypes. Brain 2003;126:1036–47.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. van Rheenen W,
      2. van Blitterswijk M,
      3. Huisman MH,
      4. et al
      . Hexanucleotide repeat expansions in C9ORF72 in the spectrum of motor neuron diseases. Neurology 2012;79:878–82.
      OpenUrlCrossRef
    17. ↵
      1. Parboosingh JS,
      2. Figlewicz DA,
      3. Krizus A,
      4. et al
      . Spinobulbar muscular atrophy can mimic ALS: the importance of genetic testing in male patients with atypical ALS. Neurology 1997;49:568–72.
      OpenUrlCrossRef
    18. ↵
      1. Reuter M,
      2. Fischl B
      . Avoiding asymmetry-induced bias in longitudinal image processing. Neuroimage 2011;57:19–21.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Reuter M,
      2. Rosas HD,
      3. Fischl B
      . Highly accurate inverse consistent registration: a robust approach. Neuroimage 2010;53:1181–96.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Hagler DJ Jr.,
      2. Saygin AP,
      3. Sereno MI
      . Smoothing and cluster thresholding for cortical surface-based group analysis of fMRI data. Neuroimage 2006;33:1093–103.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Bernal-Rusiel JL,
      2. Greve DN,
      3. Reuter M,
      4. et al
      . Statistical analysis of longitudinal neuroimage data with Linear Mixed Effects models. Neuroimage 2012;66C:249–60.
      OpenUrlPubMed
    22. ↵
      1. Luders E,
      2. Narr KL,
      3. Thompson PM,
      4. et al
      . Hemispheric asymmetries in cortical thickness. Cereb Cortex 2006;6:1232–8.
      OpenUrl
    23. ↵
      1. Bede P,
      2. Bokde A,
      3. Elamin M,
      4. et al
      . Grey matter correlates of clinical variables in amyotrophic lateral sclerosis (ALS): a neuroimaging study of ALS motor phenotype heterogeneity and cortical focality. J Neurol Neurosurg Psychiatry 2012;84:766–73.
      OpenUrlPubMed
    24. ↵
      1. Hirano A
      . Cytopathology of amyotrophic lateral sclerosis. Adv Neurol 1991;56:91–101.
      OpenUrlPubMed
    25. ↵
      1. Senda J,
      2. Kato S,
      3. Kaga T,
      4. et al
      . Progressive and widespread brain damage in ALS: MRI voxel-based morphometry and diffusion tensor imaging study. Amyotroph Lateral Scler 2011;12:59–69.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Takeda T,
      2. Uchihara T,
      3. Arai N,
      4. et al
      . Progression of hippocampal degeneration in amyotrophic lateral sclerosis with or without memory impairment: distinction from Alzheimer disease. Acta Neuropathol 2009;117:35–44.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Verstraete E,
      2. Veldink JH,
      3. van den Berg LH,
      4. et al
      . Structural brain network imaging shows expanding disconnection of the motor system in amyotrophic lateral sclerosis. Hum Brain Mapp 2013;35:1351–61.
      OpenUrlPubMed
    28. ↵
      1. Agosta F,
      2. Valsasina P,
      3. Absinta M,
      4. et al
      . Sensorimotor functional connectivity changes in amyotrophic lateral sclerosis. Cereb Cortex 2011;21:2291–8.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Verstraete E,
      2. van den Heuvel MP,
      3. Veldink JH,
      4. et al
      . Motor network degeneration in amyotrophic lateral sclerosis: a structural and functional connectivity study. PLoS ONE 2010;5:e13664.
      OpenUrlCrossRefPubMed
      1. van den Heuvel MP,
      2. Sporns O
      . An anatomical substrate for integration among functional networks in human cortex. J Neurosci 2013;33:14489–500.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Schmidt R,
      2. Verstraete E,
      3. Reus ,
      4. et al
      . Correlation between structural and functional connectivity impairment in amyotrophic lateral sclerosis. Human Brain Mapp 2014;35:4386–95.
    32. ↵
      1. van den Heuvel MP,
      2. Sporns O,
      3. Collin G,
      4. et al
      . Abnormal rich club organization and functional brain dynamics in schizophrenia. JAMA Psychiatry 2013;70: 783–92.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Brownell B,
      2. Oppenheimer DR,
      3. Hughes JT
      . The central nervous system in motor neurone disease. J Neurol Neurosurg Psychiatry 1970;33:338–57.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Ince PG,
      2. Evans J,
      3. Knopp M,
      4. et al
      . Corticospinal tract degeneration in the progressive muscular atrophy variant of ALS. Neurology 2003;60:1252–8.
      OpenUrlCrossRef
    35. ↵
      1. Turner MR,
      2. Agosta F,
      3. Bede P,
      4. et al
      . Neuroimaging in amyotrophic lateral sclerosis. Biomark Med 2012;6:319–37.
      OpenUrlCrossRefPubMed
    36. ↵
      1. Foerster BR,
      2. Dwamena BA,
      3. Petrou M,
      4. et al
      . Diagnostic accuracy using diffusion tensor imaging in the diagnosis of ALS: a meta-analysis. Acad Radiol 2012;19:1075–86.
      OpenUrlCrossRefPubMed
    37. ↵
      1. Foerster BR,
      2. Dwamena BA,
      3. Petrou M,
      4. et al
      . Diagnostic accuracy of diffusion tensor imaging in amyotrophic lateral sclerosis: a systematic review and individual patient data meta-analysis. Acad Radiol 2013;20:1099–106.
      OpenUrlCrossRefPubMed
    38. ↵
      1. Douaud G,
      2. Filippini N,
      3. Knight S,
      4. et al
      . Integration of structural and functional magnetic resonance imaging in amyotrophic lateral sclerosis. Brain 2011;134:3470–9.
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Turner MR,
      2. Grosskreutz J,
      3. Kassubek J,
      4. et al
      . Towards a neuroimaging biomarker for amyotrophic lateral sclerosis. Lancet Neurol 2011;10:400–3.
      OpenUrlCrossRefPubMedWeb of Science

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • RW, HJW and EV contributed equally. MPvdH and LHvdB contributed equally to this study.

    • Contributors RW, HJW, EV, MPvdH and LHvdB were involved in study concept and design. RW, HJW and EV were involved in acquisition of data. RW, HJW, EV, MPvdH and JHV took part in analysis and interpretation of data. RW, HJW and EV took part in drafting the manuscript. JHV, MPvdH, LHvdB were involved in critical revision of the manuscript for important intellectual content. RW, HJW and JHV were involved in statistical analysis. LHvdB and MPvdH obtained funding for the study. LHvdB supervised the study.

    • Funding JHV received support from Thierry Latran Foundation. MPvdH was supported by a VENI grant of the Netherlands Organization for Scientific Research (NWO), a grant of the ALS Foundation Netherlands, and a Fellowship of the Brain Center Rudolf Magnus, The Netherlands. LHvdB received travel grants and consultancy fees from Baxter and Biogen Idec; and receives research support from the Prinses Beatrix Fonds, ALS Foundation Netherlands, VSB Fonds, Adessium Foundation, and the European Community’s Health Seventh Framework Programme (FP7/2007–2013) under grant agreement n° 259867.

    • Competing interests LHvdB serves on scientific advisory boards for Prinses Beatrix Fonds, Thierry Latran Foundation, Cytokinetics and Biogen Idec; on the editorial boards of Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration and the Journal of Neurology, Neurosurgery and Psychiatry.

    • Ethics approval The study was approved by the Medical Ethical Committee for research into humans of the University Medical Center Utrecht.

    • Provenance and peer review Not commissioned; externally peer reviewed.