C51 STATISTICAL MOTOR UNIT NUMBER ESTIMATION AND ALS TRIALS: THE EFFECT OF MOTOR UNIT INSTABILITY
Shefner JM1, Jillapalli D1 & Cudkowicz ME2
1SUNY Upstate Medical University, Syracuse, and 2Massachusetts General Hospital, Boston, USA
E‐mail address for correspondence: [email protected]
Background: Motor unit number estimation (MUNE) remains the most attractive physiological surrogate marker to monitor disease progression in ALS. The statistical method of MUNE has a number of potential advantages for use in trials, including a well defined protocol, high test‐retest reliability, and clear relationship to survival when employed by experienced neurophysiologists. In two multicenter, double‐blind, placebo‐controlled trials of ALS, statistical MUNE displayed good reliability and declined linearly with time. However, single motor unit potential (SMUP) amplitude was unchanged over 6 and 12 months, despite modifications made to the method between trials to identify the expected increase in motor unit size over time. We suspected that the reason motor units did not increase over time was amplitude instability of single motor units, based on previous studies showing excess variability in motor units from ALS patients.
Objectives: To determine whether the previously measured increase in motor unit response variability present in patients with ALS is sufficient to confound MUNE using the statistical method.
Methods: Using data from prior studies on motor unit variability, submaximal compound motor action potentials (CMAPs) were modelled using a constant number of motor units of different amplitudes. It was assumed that repeated stimulation resulted in the same motor units being activated to every stimulus. The variance of the CMAPs was calculated, based on the variances of the individual SMUPs. From that variance, the statistical method of MUNE was used to generate an estimate of the SMUP amplitude.
Results: Statistical MUNE calculates SMUP amplitude based on the variability of CMAP amplitude in response to a constant intensity stimulus, assuming that this variability is due completely to a variable number of motor units responding. Our model requires that the same sample of motor units fire to every stimulus, so that the correct estimate of SMUP amplitude is 0. However, the intrinsic SMUP variability present in ALS motor units resulted in SMUP estimates of 48–70 uV, well within the range of normal motor unit sizes.
Discussions and conclusions: In ALS patients, the increased variability of motor units renders the statistical technique of MUNE invalid. It is likely that in prior clinical trials, the failure of SMUP amplitude to increase with time was a function of unstable single motor units being mistakenly identified as multiple motor units firing in a statistical fashion. While motor unit instability poses a problem for all methods of MUNE, the statistical method is particularly vulnerable, as it is the variability of the response to a repeated stimulus that forms the basis for determination of SMUP amplitude.
C52 EVIDENCE FOR CORTICAL DYSFUNCTION IN ALS: RESULTS OF AN MEG STUDY
Boyajian RA1, Amo C1, Otis SM1, Romine JS1 & Smith RA2
1Scripps Clinic, La Jolla, and 2Center for Neurologic Study, La Jolla, USA
E‐mail address for correspondence: [email protected]
Background: The motor areas of the brain have long been known to be involved in ALS but the timing of onset and the magnitude of cortical involvement have only recently been studied. Magnetoencephalography (MEG) is a sensitive technique that can detect changes of cortical electrophysiology and map sources of abnormal slow wave activity.
Objectives: The intent of this study is to determine whether MEG could detect evidence of altered cortical function in patients with ALS. A secondary objective, having documented abnormal cortical function, focuses on the timing and anatomic extent of these abnormalities.
Methods: Non‐invasive MEG was obtained in patients with clinically definite ALS and in control subjects of similar age. Whole head 148 channel MEG of spontaneous brain activity is recorded for 15 min in awake state. Abnormal slow wave activity (focal, high amplitude, 1–7 Hz) is identified visually. The slow wave dipole source solutions are generated from a single equivalent current dipole model. Qualifying dipole solutions meeting preselected confidence volume and additional statistical stringency measures are counted and then mapped to anatomic brain regions by co‐registration with the subject's own or reference brain MRI.
Results: Focal, high amplitude, slow wave activity is present in 6/6 ALS subjects and 0/6 control subjects. The abnormal waveforms are 3–6 Hz with a persistent intermittent pattern. The slow wave source generators localize to the frontal lobes in all ALS subjects, and to varying degrees in the temporal lobes. Frontal slow wave source generators span the supplementary motor and premotor areas, the anterior to mid‐cingulate bilaterally, operculum, and insula. This multifocal bi‐hemispheric abnormal cortical electrophysiology is also present at the presentation of early stage (short duration) disease.
Conclusions: The finding of abnormal slow wave activity in a representative population of ALS patients suggests early and widespread involvement of the cortex. In all patients, frontal involvement is seen and to varying degrees concomitant temporal involvement. These observations need to be extended to a larger group of patients. These findings are consistent with the results obtained with PET scanning using a benzodiazepine ligand. Both MEG and PET scanning could have broad research and clinical application. For example, MEG might be useful for making the diagnosis of ALS when patients exhibit only lower motor neuron findings at the time of presentation.
C53 IMAGING IN PATIENTS WITH PURE AND COMPLICATED HEREDITARY SPASTIC PARAPARESIS: 3D MRI MORPHOMETRY OF THE BRAIN AND UPPER SPINAL CORD PLANIMETRY
Kassubek J & Sperfeld A‐D
Department of Neurology, University of Ulm, Ulm, Germany
E‐mail address for correspondence: [email protected]
Hereditary Spastic Paraparesis (HSP) encompasses a group of inherited neurodegenerative disorders of heterogeneous phenotype that share the principal clinical feature of progressive weakness and spasticity predominantly affecting the lower limbs. Conventionally, they are categorized into pure forms (pHSP) and complicated forms (cHSP), the latter characterized by the presence of additional major clinical features.
In the present study, 33 HSP patients suffering from pHSP (n = 22) or cHSP (n = 11) were investigated using different magnetic resonance imaging (MRI) techniques. MRI data of the brain, including a volume‐rendering 3‐D data set, and of the upper spinal cord were acquired. Besides standard visual assessment, the diameters of the cervical and thoracic myelon were measured by planimetric techniques. Cerebral 3D MRI was analyzed by use of the observer‐independent whole‐brain based technique of voxel‐based morphometry (VBM), both in comparison with an age‐matched normal database and with respect to a comparison between pHSP and cHSP. Moreover, relative global brain volumes were analyzed by the automated brain parenchymal fraction (BPF) method.
Despite the heterogeneity of the cHSP phenotypes, the standardized BPF analysis showed characteristic patterns of the global cerebral alterations, both with respect to BPF itself and to the grey and white matter sub‐fractions. VBM demonstrated regional grey and white matter volume decreases (atrophy) in both HSP subtypes with the most robust alterations located in the corpus callosum. The upper spinal cord was significantly decreased in diameter with respect to controls. Our results demonstrate the involvement of extrapyramidal cerebral and spinal CNS structures in a sizeable group of pHSP/cHSP patients and give insight into pathophysiological basics and, thus, might help to improve our understanding of the characteristics of this heterogeneous disorder.
C54 DEVELOPMENT OF UPPER MOTOR NEURON INVOLVEMENT IN ALS: A LONGITUDINAL MRI STUDY USING DIFFUSION TENSOR IMAGING AND VOXEL‐BASED MORPHOMETRY
Grosskreutz J, Emmerich L, Keil C, Dengler R & Peschel T
Medical School Hannover, Hannover, Germany
E‐mail address for correspondence: [email protected]
Background: Extent and distribution of upper motor neuron (UMN) involvement in ALS has been described in post‐mortem brains. Little is known about the extent and development of UMN degeneration in vivo, and it is unclear whether cortical atrophy is present at all early in the course of ALS and how it develops over time. Region of interest based analyses have suggested neuronal loss in central regions, and signal abnormalities were found in the corticospinal tract (CST) by computerised analyses in diffusion tensor imaging (DTI). Detection of cortical atrophy in the sensorimotor and frontal cortex as well as reduction of the anisotropy in the CST has been demonstrated recently by our group using voxel‐based morphometry (VBM) and DTI analysed on a voxel‐by‐voxel basis.
Objective: The aim of the present MRI study was to describe the development of UMN involvement in ALS patients. We used voxel‐based morphometry and Diffusion Tensor Imaging to detect region‐specific changes in grey matter volumes (VBM) and differences in fractional anisotropy (FA) of white matter tracts (DTI).
Methods: High‐resolution anatomical MRI and DTI (15 directions) were performed in 21 patients with ALS at baseline and in 13 six months later, and 23 age‐matched controls on a 1.5 T GE‐scanner. Images were analysed on a voxel‐by voxel basis using SPM2. Group comparisons of regional grey matter and FA and were made using ANCOVA with the global mean voxel value as confounding factor. Longitudinal assessments were carried out using paired t‐tests.
Results: Compared with controls, ALS patients at baseline showed a relative decrease in GMV in the pre‐ and post‐central gyrus, inferior parietal lobe and the middle frontal gyrus. In addition, there was reduced FA in the CST bilaterally. At follow‐up after six months, the relative decrease in GMV had progressed to cingulate areas and the cerebellum. In DTI, FA had decreased in the genu of the internal capsule and CST descending into the brainstem bilaterally.
Conclusion: The present study confirms our previous findings of largely central atrophy extending to frontal areas and FA reductions in the CST in a second cohort of patients. On follow‐up examination, the atrophy had spread to larger areas of the sensorimotor cortex and included cingulate and cerebellar structures which is in line with recent histopathological findings. In addition, CST abnormalities extended into the brainstem. Our results consistently show grey matter atrophy in sensorimotor, parietal and frontal regions which increases during the course of ALS. In groups of patients, voxel‐based morphometry and DTI may be used as a monitoring tool in future clinical trials.
C55 QUANTITATIVE EVALUATION OF FLAIR‐MR IMAGES CORRELATES TO SPASTICITY SCORE BUT NOT TO CORTICAL EXCITABILITY
Hecht MJ1, Schirmer K1, Vetter K1, Schmid A2 & Neundörfer B1
1University of Erlangen, Department of Neurology, Erlangen, and 2University of Erlangen, Institute of Radiology, Erlangen, Germany
E‐mail address for correspondence: [email protected]
Background: Degeneration of the upper motor neuron is often clinically masked by the degeneration of the lower motor neuron in ALS. MR imaging of the brain is commonly used to exclude diseases mimicking ALS. However, quantification of the subcortical precentral gyrus FLAIR‐signal revealed a significant hyperintensity in ALS patients which increased in the follow‐up examination (1,2).
Objectives: The aim of our study was to compare quantified FLAIR‐MR images with a standardized clinical examination of upper motor neuron and cortical excitability.
Methods: In 30 ALS patients (16 males, 14 females, 59.6±9.6 years, probable or definite ALS) we performed axial FLAIR‐MR images and calculated the ‘contrast to noise ratio’ (CNR) of regions of interest within both subcortical precentral gyri (1). In addition, all patients where clinically investigated using the ‘spasticity score’. Cortical inhibition and facilitation was determined using paired transcranial stimulation.
Results: The CNR of the right and left subcortical precentral gyrus correlated significantly to the contralateral spasticity score (CNR right – spasticity left, r = 0.437, p = 0.016; CNR left – spasticity right, r = 0.488, p = 0.006; CNR mean – spasticity bilateral, r = 0.613, p = 0.005). We found no correlation of the subcortical precentral CNR to cortical inhibition (r = −0.216, p = 0.375). Short interval cortical inhibition showed only a trend to the spasticity score (r = 0.421, p = 0.057, n = 21).
Discussion and conclusion: Cortical excitability changes seem to occur independently of morphological changes demonstrated by FLAIR‐MR images. This finding might be due to the complex changes of cortical excitability in ALS. For clinical purposes however, the high correlation of the subcortical precentral CNR with the spasticity score demonstrates a possible clinical value. CNR measurement might be useful as an objective marker of upper motor neuron changes in clinical trials.
References
- Hecht MJ, Fellner F. Fellner C, et al. J Neurol Sci. 2001;186:37.
- Hecht MJ, Fellner F, Fellner C, et al. J Neurol Sci. 2002;199:59.
C56 IMPAIRED CARDIOVAGAL AND INCREASED VASOMOTOR RESPONSES IN EARLY‐STAGE AMYOTROPHIC LATERAL SCLEROSIS
Gutierrez J1, Mustelier R1, Lestayo Z1, Zaldivar T1, Voustianiouk A2 & Kaufmann H2
1Institute of Neurology, Havana, Cuba, and2Mount Sinai Hospital, New York, USA
E‐mail address for correspondence: [email protected]
Background & objectives: Previous studies have indicated autonomic nervous system (ANS) involvement in amyotrophic lateral sclerosis (ALS).
Subjects and methods: ECG and continuous tonometric blood pressure were monitored in 25 ALS patients (age 56.2±14.5 years, time from onset 11.2 months) and 42 age‐matched normal controls (53.16±13.14 years) during paced breathing (six breaths per min), Valsalva manoeuvre, and upon active standing. Expiratory/Inspiratory (E/I), Valsalva and 30:15 ratios were calculated. Sensitivity of the baroreflex was assessed on the basis of Valsalva manoeuvre by calculating the following latencies: tachycardia (TL), bradycardia (BL), blood pressure recovery (BpL) and overshoot (OvL) latencies (1).
Results: No differences in resting heart rate and BP were found between groups. The E/I ratio was significantly lower in the patients (1.19±0.12 vs. 1.38±0.15, p = 0.0003) but Valsalva and 30:15 ratios were similar in both groups. Valsalva latencies TL (2.50±0.53 vs. 3.50±1.3, p = 0.002) and BpL (2.50±0.53 vs. 2.83±1.11, p = 0.02) were significantly shorter in patients, as was the duration of early phase II (5.50±1.7 vs. 7.58±1.32, p = 0.03). BL and OvL were not different between groups. By the end of the phase II of the Valsalva manoeuvre, systolic BP was significantly higher in patients than in controls (142.4±25.9 vs. 126.5±17.9, p = 0.001).
Conclusions: These results demonstrate mild parasympathetic impairment and hyperactive peripheral sympathetic activity in patients with ALS. These findings suggest a combined central and peripheral involvement of the ANS and could be useful in the differentiation of ALS from other clinically similar disorders.
References
- Ferrer et al. Neurology 1991;41:1462–6.