MUNIX: Reproducibility and clinical correlations in Amyotrophic Lateral Sclerosis

Accepted Manuscript MUNIX: reproducibility and clinical correlations in Amyotrophic Lateral Sclerosis Marcio Luiz Escorcio-Bezerra, Agessandro Abrahao, Isac de Castro, Marco Antonio Troccoli Chieia, Lyamara Apostolico de Azevedo, Denise Spinola Pinheiro, Nadia Iandoli de Oliveira Braga, Acary Souza Bulle de Oliveira, Gilberto Mastrocola Manzano PII: DOI: Reference:

S1388-2457(16)30441-2 http://dx.doi.org/10.1016/j.clinph.2016.06.011 CLINPH 2007876

To appear in:

Clinical Neurophysiology

Accepted Date:

12 June 2016

Please cite this article as: Escorcio-Bezerra, M.L., Abrahao, A., de Castro, I., Chieia, M.A.T., de Azevedo, L.A., Pinheiro, D.S., de Oliveira Braga, N.I., de Oliveira, A.S.B., Manzano, G.M., MUNIX: reproducibility and clinical correlations in Amyotrophic Lateral Sclerosis, Clinical Neurophysiology (2016), doi: http://dx.doi.org/10.1016/ j.clinph.2016.06.011

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

MUNIX: reproducibility and clinical correlations in Amyotrophic Lateral Sclerosis

Marcio Luiz Escorcio-Bezerra1, MD, MSc; Agessandro Abrahao1, MD, MSc; Isac de Castro 2, PhD; Marco Antonio Troccoli Chieia1, MD, MSc; Lyamara Apostolico de Azevedo1, MD, PhD; Denise Spinola Pinheiro 1,MD, PhD; Nadia Iandoli de Oliveira Braga1, MD, PhD; Acary Souza Bulle de Oliveira1, MD, PhD; Gilberto Mastrocola Manzano1, MD, PhD

1 – Department of Neurology, Universidade Federal de São Paulo, SP, Brazil. 2 - Department of Medicine, Division of Molecular Medicine, University of São Paulo School of Medicine (FMUSP)

Correspondence to: Marcio Luiz Escorcio-Bezerra, MD, MSc. Department of Neurology, Escola Paulista de Medicina, Universidade Federal São Paulo, Rua Pedro de Toledo, 650, 04039-002. São Paulo, SP, Brazil E-mail: [email protected]

2 Highlights •

MUNIX showed greater test-retest variability in ALS patients than healthy controls.

•

MUNIX detected denervation cross-sectionally in ALS patients.

•

Decrease of MUNIX probably occurs in ALS at an earlier stage than decrease in CMAP and muscle strength.

Abstract

Objective: To study the reproducibility, diagnostic yield to detect denervation, and clinical correlations of the Motor Unit Number Index (MUNIX) in subjects with Amyotrophic Lateral Sclerosis (ALS). Methods: MUNIX evaluation was performed in three muscles twice on the same day to assess reproducibility. Cutoff values for the MUNIX were based on data from 51 healthy subjects (Controls) to evaluate the sensitivity of the technique to detect denervation in 30 subjects with ALS. Results: The method had good reproducibility. The variability was greater in the ALS group. In 23 ALS subjects (77%), low MUNIX values were detected. Most of the muscles with low MUNIX had also low compound muscle action potential (CMAP) and strength, but these parameters were normal in 9% of muscles. According to ROC curve analysis, MUNIX was generally accurate (AUC = 0.9504) for discriminating between healthy individuals and subjects with at least one denervated muscle. Conclusions: MUNIX variability was higher in the ALS group. The method showed good diagnostic performance for the detection of denervation in a sample of patients with ALS. Significance: This study demonstrated that in addition to being a quantitative tool MUNIX can detect denervation in subjects with ALS.

Keywords: MUNIX; ALS; MUNE; reproducibility, clinical correlation, denervation.

3 1. Introduction Motor Unit Number Index (MUNIX) analysis is a technique, developed by Nanadedkar et al. (Nandedkar et al., 2004), that provides an index directly correlated with the number of motor units. Thus, unlike the motor unit estimation (MUNE) methods, it does not directly count or estimate the number of functioning motor neurons (Bromberg et al., 2007). Compared to existing MUNE methods, MUNIX is faster and easier, which can potentially represent a practical advantage. For a given muscle, MUNIX requires the compound muscle action potential (CMAP) and progressive grades of voluntary muscle contractions recorded in surface electromyography (EMG) epochs, the surface interference pattern (SIP). Using the CMAP and SIPs an “ideal case motor unit count” is obtained, which corresponds to the MUNIX in a certain SIP area obtained from a power regression that is part of the mathematical model explained in detail elsewhere (Nandedkar et al., 2004, 2010). MUNIX is a potential biomarker for denervating conditions, such as Amyotrophic Lateral Sclerosis (ALS), a devastating disease characterized by progressive upper and/or lower motor neuron loss that ultimately results in paralysis and respiratory failure. In ALS, following up lower motor neuron loss is key to studying disease progression and to understanding the response to new drugs in clinical trials. Adequate reproducibility is of paramount importance for a biomarker of motor unit loss. MUNIX reproducibility was previously studied in small samples of healthy subjects and patients with ALS in the abductor pollicis brevis (APB) and abductor digiti minimi (ADM) muscles. There was a good correlation between

4 tests performed twice on the same day, i.e., high intraclass correlation coefficient (ICC) values in healthy subjects. ALS subjects had better ICC values than controls but higher coefficient of variation (COV) values (Ahn et al., 2010; Nandedkar et al., 2011). A multicenter study of healthy individuals evaluated MUNIX reproducibility through ICC and found that this method had good reproducibility in the more experienced centers (Neuwirth et al., 2011). However, this study did not evaluate individual variability (measured by the COV). Prospective studies showed a progressive decrease in MUNIX values in small groups of ALS patients, indicating a decrease in the motor unit population (Nandedkar et al., 2010; Neuwirth et al., 2010). Furtula et al. demonstrated that the progressive reduction in MUNIX was comparable to the reduction of incremental stimulation MUNE (IS-MUNE) in a prospective study assessing the ADM muscle in 13 patients with ALS [Furtula et al., 2013]. MUNIX also had similar performance to high-density MUNE (HD-MUNE) (Boekestein et al., 2012). In addition to being a quantitative tool for longitudinal follow-up of motor unit loss, under certain conditions, MUNIX can potentially detect denervation in a cross-sectional manner. Furtula et al. studied the ADM muscle of 13 ALS patients performing the MUNIX in order to discriminate between normal and ALS disease status (Furtula et al., 2013). The authors found that the technique had good sensitivity and specificity when compared with IS-MUNE. Further studies with larger sample sizes and muscles from lower extremities may provide relevant data. In this study, we performed the MUNIX in subjects with and without ALS in muscles from lower and upper extremities: the tibialis anterior (TA), APB and

5 ADM muscles. We compared these two groups of individuals and established cutoff values in order to study the diagnostic yield for detection of denervation in a group of subjects with ALS. This approach has not previously been performed in multiple muscles or in larger groups of ALS patients. Using a set of muscles instead of a single muscle is likely to more accurately reflect the disease progression and to allow the evaluation of different ALS variants with different sites of symptom onset. We also correlated the neurophysiological data with the available clinical data. Lastly, we studied the reproducibility of the MUNIX, through the ICC and the COV for measures performed twice on the same day.

2. Methods 2.1 Subjects Healthy individuals and patients diagnosed with laboratory-supported probable, clinically possible, probable or definite ALS according to the World Federation of Neurology Revised El Escorial criteria (Brooks et al., 2000) from the Neuromuscular Diseases Center, Universidade Federal de Sao Paulo (UNIFESP, Brazil) were prospectively recruited from June 2013 to June 2015. The UNIFESP Research Ethics Board approved this study and we obtained written informed consent from all participants. Healthy subjects were interviewed and those with any symptoms were excluded from the study. Patients underwent a review of symptoms and neurologic examination, along with chart review. Muscle strength was scored according to the Medical Research Council (MRC) scale.

6 2.2 MUNIX procedure All patients and controls were positioned in a supine and comfortable position and studies were performed with a commercially available electromyography instrument (brand: Neurosoft; model: Neuro-MEP-Micro) using disposable pregelled surface electrodes with the same dimensions (10x30 mm) for all tests. Skin temperature was kept above 32°C in the evaluated limb (a heat pack was used when necessary). Subjects with and without ALS underwent the MUNIX procedure twice on the same day (Test- Retest), performed by the same Clinical Neurophysiology Boardcertified neurologist. The test and retest were performed with an interval of at least 30 minutes and no more than 6 hours. After the first tests, all electrodes and marks were completely removed so that there was no trace of the initial placement before the retest. MUNIX was performed on the TA, APB and ADM muscles on the right side of healthy subjects and on the strongest side of subjects with ALS (total of 90 muscles studied in ALS subjects). The compound muscle action potential (CMAP) was recorded using standard motor nerve conduction techniques. TA muscle study: the recording electrode (G1) was placed over the muscle belly, the reference electrode (G2) was placed lateral to the knee, and the stimulus was applied at the lateral popliteal fossa. APB muscle: G1 was placed over the muscle belly, G2 was placed over the first metacarpophalangeal joint and the stimulus was delivered at the wrist (between tendons to the flexor carpi radialis and palmaris longus). ADM muscle: G1 was

7 placed over the muscle belly, G2 was placed at the fifth metacarpophalangeal joint, and the stimulus was applied at the wrist adjacent to the flexor carpi ulnaris. We paid special attention to obtaining the CMAP with the highest amplitude (baseline to negative peak) in every sample. Then, without moving the electrode, the SIP was recorded. In this step, the patient was instructed to exert an isometric contraction at varying levels of effort. For each MUNIX, we collected seven progressive isometric efforts against resistance provided by the operator. Each SIP epoch was recorded at an increasing force level from minimum to maximum. The operator relied on acoustic and visual feedback from the electromyograph to help measure the levels of effort. Values of CMAP and SIPs from each muscle were analyzed in a standard spreadsheet application (Microsoft Excel), based on a mathematical model described by Nandedkar and colleagues (Nandedkar et al., 2004, 2010). Recordings with low SIP amplitude (<200µV) were rejected, so that SIP area from all recordings was > 20mV/ms. CMAP and SIPs were recorded using a bandpass filter setting of 5-10000 Hz. Each SIP epoch was 500 ms long. 2.3 Measures The reproducibility was studied through the ICC and COV. Considering that COV is a variability measure of individual subjects, not a group measure like the ICC, we also obtained mean COV values (COVm) for all individuals, as well as the standard deviation (SD) and range for each muscle. To determine the most accurate lower limit of normal (LLN) of MUNIX able to discriminate between normal and clinically possible ALS disease status, we used Receiver Operating Characteristics (ROC) curve analysis (Metz,1978). The ROC

8 curve was designed using different cut-off values calculated with the mean MUNIX -0.5 SD to -2.5 SD for each of the three muscles tested in the controls aged older than 50 years. An individual from the ALS group was considered as true positive whenever there was at least one muscle (among the three tested) with MUNIX below the cut-off. The MUNIX megascore is the sum of MUNIX values from all muscles and was calculated for the ALS group and age-matched controls. 2.4 Statistical Analysis All results are expressed as the mean ± SD. Comparisons between groups were performed using the Mann–Whitney test. The Spearman rank correlation was used to analyze the correlations. Differences were considered significantly different at p < 0.05. The Shapiro-Wilk and Lilliefors tests were used to assess normal distribution. We used the Wilcoxon signed-rank test to calculate the significance of the percent difference between the decreases in the MUNIX and CMAP.

3. Results Fifty-one controls (31 females) with a mean age of 42.3±16.4 years (range: 2475) were enrolled in this study. CMAP and MUNIX of the controls showed a normal distribution according to normality tests (Shapiro-Wilk, p>0.44; Lilliefors p>0.20). MUNIX values in healthy individuals over 50 years of age (n=18) were lower than those in the group below 50 years of age (n= 33). This difference was significant for the APB muscle and non-significant for the TA and ADM muscles

9 (Table 1). The MUNIX megascore mean value was 550 ± 112 for health subjects over 50 years of age. Thirty patients with ALS (18 females) with a mean age of 62.3 ± 8.5 years (range: 50-83) and mean disease duration (DD) of 19 ± 11 months were included. The site of symptom onset, El Escorial classification, muscle strength on the strongest side and MUNIX results of each patient are presented in Table 3. DD did not correlate with MUNIX values (data not shown). For the TA, APB, and ADM muscles, the mean CMAP and mean MUNIX were significantly lower in the individuals with ALS than in the age-matched controls (above 50 years of age; p < 0.05). The relative decreases in the mean MUNIX were significantly greater than the relative decrease in the mean CMAP in all three muscles tested (p<0.001) (Table 2). Motor unit size index (MUSIX) values differed significantly between age-matched healthy subjects and individuals with ALS for all three muscles (P<0.01; Table 2). The MUNIX megascore mean value was 290 ± 115 in subjects with ALS. MUNIX Test-Retest showed satisfactory intra-operator reproducibility in tests performed twice on the same day in subjects with and without ALS, as indicated by an ICC greater than 0.7 for all muscles. COVm was higher in ALS subjects than in healthy subjects, indicating a greater variability for the MUNIX in the ALS group (Table 4). In the ALS group, the range of COV values was larger than in controls for all muscles. In addition, COV values higher than the maximum observed among the controls were observed in three ALS patients for the APB, in five patients for the ADM, and in six patients for the TA. The standard deviation was also higher for the ALS group than for healthy subjects for all muscles.

10 The ROC analysis had an area under the curve (AUC) of 0.9504 and the best diagnostic performance to discriminate between normal and possible ALS disease status was mean MUNIX - 1.5 SD of the age-matched controls (0.83 sensitivity and 0.83 specificity) (Figure 1). However, for the purpose of this study, we opted to define the LLN as the mean -2 SD (Table 2), which corresponded to high specificity (1.0) and good sensitivity (0.77). The LLN of the MUNIX megascore from controls over 50 years of age (mean-2SD) was 326. This cut-off detected abnormality in 20 ALS individuals (0.66 sensitivity). Among the 90 muscles tested from the ALS group (3 muscles in 30 individuals), 44 had MUNIX values below the LLN. In this group of 44 muscles, 15 muscles had normal strength. Among the 15 muscles with low MUNIX and normal strength, 4 also had normal CMAP, and 11 had CMAP values below the LLN (Table 3; Figure 2). Additionally, 1 individual had low MUNIX, normal CMAP values and reduced strength. On the other hand, most of the muscles with low MUNIX (28 muscles) also had low CMAP and reduced strength (Figure 2).

4. Discussion In this study, we found that, in addition to being a reproducible quantitative tool for disorders that lead to denervation, MUNIX is sensitive for indicating loss of motor neurons in patients who meet criteria for at least possible ALS according to the Revisited El Escorial criteria (Brooks et al., 2000).

11 Reproducibility is a fundamental aspect of any biomarker of disease progression. Our study showed that MUNIX had ICC values > 0.7 in controls, indicating a good correlation among measures. Although ALS individuals had even better ICC values for every muscle, the individual variability, tested using the COV, was greater in this group of individuals (Table 4). A greater individual MUNIX variability in ALS individuals has also been found in previous studies (Ahn et al., 2010; Nandedkar et al., 2011). The individual variability is more significant than the ICC for the purpose of an eventual follow-up study as it is a reflection of the precision of the method. The reproducibility in our study was similar to that found using MUNE techniques (Furtula et al., 2013; Boekestein et al., 2012). Mean MUNIX values were lower in the group of healthy subjects over 50 years of age than in the younger group (Table 1), similar to the findings of previous studies (Nandedkar et al. 2004; Neuwirth 2011), probably due to the motor unit loss in normal aging. The ALS group had lower MUNIX values than age-matched controls (over 50 years). The MUNIX reduction in ALS individuals was proportionally greater than the reduction in CMAP for the same muscles (Table 2), consistent with previous reports (Ahn et al., 2010; Nandedkar et al., 2010). Most of the muscles of the subjects with ALS presenting with low MUNIX also had reduced CMAP and strength, which is to be expected for certain degrees of denervation. However, in a small subset of the muscles, low MUNIX values were accompanied by normal CMAP and strength (Figure 2). This suggests that MUNIX decreases earlier than the CMAP and has the potential to detect denervation in a cross-sectional manner irrespective of its characteristics as a tool for tracking motor unit loss.

12 Indeed, MUNIX showed good sensitivity to detect denervation in a group of patients with ALS. It was reduced in at least one muscle in 77% (23/30) of the ALS patients. The AUC (0.9504) of the ROC curve analysis (Figure 1) indicated an excellent diagnostic accuracy for the detection of denervation in the ALS group. To test the sensitivity of the method in detecting denervation, we assumed that all subjects classified as having ALS had at least one muscle with some degree of motor unit loss at the time of the evaluation. Considering that motor neuron disease spreads contiguously (Ravits and La Spada, 2009) and depending on the site of symptom onset, the most affected muscle would be in the upper or lower extremity. Hence, evaluating muscles from upper and lower extremities allows a broader evaluation, including different ALS subtypes. Note that we performed this technique on the strongest side and included ALS individuals with bulbar onset and few limb symptoms. The purpose of this approach was to evaluate muscles with normal strength and CMAP. Had we chosen the most affected muscles there, would be probably a higher sensitivity and a larger number of affected muscles. In addition, we could also have performed MUNIX analysis on the bulbar segment (Ahn et al., 2015) in the orbicularis oculi muscle, adding sensitivity to the evaluation. The MUNIX megascore detected less abnormality (20 individuals/0.66 sensitivity) than the evaluation of the muscles separately, as we described above. This was probably because the megascore considers the sum of all three muscles. Thus, a subtle abnormality present in only one muscle may be diluted by the inclusion of the other muscles.

13 Conclusions regarding the diagnostic yield of MUNIX for the detection of denervation must be placed within the proper context. The evaluation was performed on a selected population of individuals classified as at least possible ALS; therefore, we do not know how it would perform for other denervating disorders or myopathies. Additionally, a more appropriate gold standard for denervation would be needle electromyography, with quantitative analysis ideally performed on the same day as the MUNIX evaluation. On the other hand, the use of El Escorial criteria to assume the presence of denervation could be justified because motor neuron disease is known for having a long preclinical phase (Swash and Ingram, 1988), and denervation on EMG is expected to precede initial signs and symptoms. Notably, the same controls used to set cutoff values for MUNIX and CMAP were used to build the ROC curve, which could be regarded as a limitation of the study.

5. Conclusion Despite better ICC values in subjects with ALS than in healthy subjects, the individual variability was greater in the motor neuron disease group, a factor that should be considered in eventual follow-up assessments. This study was the first to investigate the potential of the MUNIX to detect motor neuron loss crosssectionally in a set of muscles from ALS patients. In addition to its properties as follow-up tool, this technique had a good diagnostic performance for detecting denervation in this group of individuals. Considering that MUNIX is relatively easy and fast to perform, the inclusion of muscles from different body segments allows a broader evaluation of the motor unit loss and different ALS variants.

14

Conflict of interest

None of the authors have potential conflicts of interest to be disclosed.

Financial disclosure We have nothing to disclose.

15 References Ahn SW, Kim SH, Kim JE, Kim SH, Park KS, Sung JJ, et al. Reproducibility of the motor unit number index (MUNIX) in normal controls and amyotrophic lateral sclerosis patients. Muscle Nerve 2010;42:808-13. Ahn SW, Kim KW, Kim JE, Shin JY, Kim DG, Lee KW, et al. Motor unit number index (MUNIX) in the orbiculari oculi muscle of healthy subjects. Muscle Nerve 2015;51:197-200. Bromberg MB. Updating motor unit number estimation (MUNE). Clin Neurophysiol 2007;118:1-8. Boekestein WA, Schelhaas HJ, van Putten MJAM, Stegeman DF, Zwarts MJ, van Dijk JP. Motor unit number index (MUNIX) versus motor unit number estimation (MUNE): a direct comparison in a longitudinal study of ALS patients. Clin Neurophysiol 2012;123:1644-9. Brooks BR, Miller RG, Swash M, Munsat TL; World Federation of Neurology Research Group on Motor Neuron Diseases. El Escorial revisited: revisited criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord 2000;1:293-9. Furtula J, Johnsen B, Christensen PB, Pugdahl K, Bisgaard C, Christensen MK, et al. MUNIX and incremental stimulation MUNE in ALS patients and control subjects. Clin Neurophysiol 2013;124:610-8. Metz CE. Basic principles of ROC analysis. Semin Nucl Med 1978;8:283-98. Nandedkar SD, Nandedkar DS, Barkhaus PE, Stalberg EV. Motor unit number index (MUNIX). IEEE Trans Biomed Eng 2004;51:2209-11.

16 Nandedkar SD, Barkaus PE, Stalberg EV. Motor unit number index (MUNIX): principle, method, and findings in health subjects and in patients with motor neuron disease. Muscle Nerve 2010;42:798-807. Nandedkar SD, Barkhaus PE, Stalberg EV. Reproducibility of MUNIX in patients with amyotrophic lateral sclerosis. Muscle Nerve 2011;44:919-22. Neuwirth C, Nandedkar SD, Stalberg EV, Weber M. Motor unit number index (MUNIX): a novel neurophysiological technique to follow disease progression in amyotrophic lateral sclerosis. Muscle Nerve 2010;42:37984. Neuwirth C, Nandekar SD, Stalberg EV, Barkaus PE, Carvalho M, Furtula J, et al. Motor Unit Number Index (MUNIX): a novel neurophysiological marker for neuromuscular disorders; test-retest reliability in healthy volunteers. Clin Neurophysiol 2011;122:1867-72. Ravits JM, La Spada AR. ALS motor phenotype heterogeneity, focality, and spread: deconstructing motor neuron degeneration. Neurology 2009;73:805-11. Swash M, Ingram DA. Preclinical and subclinical events in motor neuron disease. J Neurol Neurosurg Psychiatr 1988;51:165-68.

17

ABBREVIATIONS ADM = abductor digiti minimi ALS = amyotrophic lateral sclerosis APB = abductor pollicis brevis AUC = area under the curve CMAP = compound muscle action potential COV = coefficient of variation COVm = mean COV values DD = disease duration HD-MUNE = high-density MUNE ICC = intraclass correlation coefficient LLN = lower limit of normal MRC = Medical Research Council MUNE = motor unit number estimation MUNIX = Motor unit number index MUSIX = Motor unit size index ROC = Receiver Operating Characteristics SD = standard deviation SIP = surface interference patterns TA = tibialis anterior

18 Figure Legends

Figure 1: Receiver operating characteristic (ROC) curve analysis was performed using different cut-off values for the three muscles for discriminating between healthy individuals and those with at least one muscle denervated. Area under the curve (AUC) was 0.9504.

Figure 2: In this graph we show the characteristics of the muscles from the ALS patients who presented reduced MUNIX values. In some muscles, the reduction in MUNIX values occurred before the CMAP or muscle strength was affected.

19 Table 1: Neurophysiological data from controls under and over 50 years of age.

TA

Age

Healthy subjects under 50 (n=33) 31.2±5.5

Healthy subjects over 50 (n=18) 62.6±7.4

Female

14

4

CMAP

7.8 (± 2.5)

7.4 (± 1.2)

MUNIX

175 (± 44)

166 (± 29)*

CMAP

12.4 (± 3.8)

9.8 (± 2.2)

MUNIX

254 (± 70)

189 (± 58)*

CMAP

10.6 (± 2)

10.5 (± 2.1)

197 (± 41)

195 (± 44)*

(mV)

APB

(mV)

ADM

(mV) MUNIX

Mean values and standard deviation in ALS and control individuals. n = number of patients; the results are expressed as the mean ± SD. mV = millivolts; TA= tibialis anterior; APB = abductor pollicis brevis; ADM= abductor digiti minimi; * = Values were significantly different for the APB muscle (P<0.01) and non significant for the TA and ADM muscles (p>0.05).

20 Table 2: Neurophysiological data and demographics of ALS patients and healthy subjects over 50 years old (age-matched). Demographics

ALS (n=30)

Age

Healthy controls (n=18) 62.6(±7.4)

Female

4

20

DD (months) Neurophysiological data TA

RR (%)

62.2(±8.5)

19.28 (±11.53)

CMAP

7.4 (±1.2)

(mV)

(LLN=5)

MUNIX

166 (±29)

5 (±1.9)*

32

101 (±44)*

39

+*

(LLN=108)

APB

MUSIX

45 (±4.6)

56 (±19.7)**

CMAP

9.8 (±2.2)

6 (±2.8)*

39

(mV)

(LLN=5.3)

MUNIX

189 (±58)

91 (±49)*

52+*

(LLN=73)

ADM

MUSIX

54 (±9.7)

72 (±23.3)**

CMAP

10.5 (±2.1)

7.1 (±2)*

32

(mV)

(LLN=6.3)

MUNIX

195 (±44)

113 (±45)*

42

+*

(LLN=108) MUSIX

54 (±7.3)

67 (±18.2)**

Mean values and standard deviation in ALS and control individuals. n = number of patients; the results are expressed as the mean ± SD. ALS = amyotrophic lateral sclerosis; mV = millivolt; TA= tibialis anterior; APB = abductor pollicis brevis; ADM= abductor digiti minimi; DD= disease duration; LLN = lower limit of normal; RR = relative percent reduction in CMAP or MUNIX in ALS individuals in relation to age-matched controls (>50 years of age); * = MUNIX and CMAP values were significantly different in ALS patients compared with agematched controls (P< 0.001); ** = MUSIX values were significantly different between age-matched controls and subjects with ALS for all three muscles (P<0.01). +* = RR% was more pronounced for MUNIX values than for CMAP and this difference was statistically significant (P<0.001).

21 Table 3: MUNIX, CMAP and clinical data in 30 ALS patients. TA SSO

CMAP

APB MUNIX

MRC

CMAP

ADM

El Escorial

MRC

MUNIX

MRC

CMAP

MUNIX

Prob Lab Prob Lab Poss ALS Prob Lab Prob Lab

5 5 5 5 5

6.6 4.4 6 7.1 5.8

112 84 125 166 133

4+ 5 5 5 5

6.9 4.4 10.8 9.7 7.3

92 54 144 185 128

4+ 5 5 5 5

8.1 6.4 9.8 9 7.9

150 108 210 195 158

Poss ALS Definite Prob Lab Possible Probable Possible Prob Lab Prob Lab Definite Prob Lab

5 2 5 5 5 5 5 5 5 5

5.3 5.4 6.3 4.7 6 4.9 3.6 8.3 3.6 7.8

110 115 122 93 116 92 46 178 64 173

5 45 44 5 4 5 5 5

6.2 4.7 8.1 4.5 3.7 6.1 3.1 6.9 6.1 4.9

93 34 124 46 71 110 50 141 109 99

5 45 4 4 5 4 5 5 5

7.8 5 5.8 7.9 4.1 5.9 8.4 6.8 7.5 9

88* 42 82 78 64 119 149 128 134 157

Prob ALS Definite Probable

5 3 5

6.1 1.7 4.3

133 45 84

4+ 4+ 4+

9.7 4.5 2.5

158 47 22

4+ 4+ 4-

9 5.6 8.5

155 87 158

Poss ALS Definite Prob Lab

2 5 5

4.1 6.3

95 120

44+ 4

1.3 7.4 3.5

28 128 28

4 4+ 4

4.7 5.7 4.8

79 111 59

Definite Definite Definite Prob Lab Poss ALS Possible Prob Lab Definite Definite

3 5 3 1 5 3 2 3 1

3.4 8.8 2.2 3.7 2.3 1.5 -

29 168 40 89 61 21 -

5 4 445 5 5 5 4+

6.2 1.9 3.2 5.3 12.8 8.8 8.5 8.3 3.5

63* 25 29 93 193 123 106 134 55

5 4 5 45 5 5 5 4+

5.7 5.1 6.5 5.5 12.5 7.1 7.8 10.2 3.8

72 54 74* 96 186 81* 132 113 59

Bulbar

Bulb+ UE

UE

LE

CMAP = compound motor action potential; TA = tibialis anterior; APB = abdutor pollicis brevis; ADM = abductor digiti minimi; - = not obtainable; mV = millivolt; MRC = Medical Research Council muscle strength scale; Bulbar = bulbar onset of symptoms; UE = upper extremities; LE = lower extremities; Bulb+UE = Bulbar onset simultaneously with upper extremities; El Escorial = El Escorial revisited criteria; Possible = clinically possible ALS; Prob Lab = clinically probable – laboratory-supported ALS; Probable = clinically probable ALS; Definite = clinically definite ALS. SSO = site of symptom onset; * = muscles with low MUNIX and normal CMAP and strength.

22

Table 4: MUNIX Test-Retest reproducibility in subjects with and without ALS.

MUNIX reproducibility Healthy

Test

Retest

subjects

COVm

ICC

(SD±) (Range)

TA

172 (40)

177 (43)

13.8

0.81

(±6.4) (8.2-22.1)

APB

231 (73)

233 (67)

15.7

0.83

(±8.5) (1.5-38.6) ADM

197 (42)

202 (46)

11.5

0.71

(±7.8) (4.2-32.3)

Subjects with ALS

TA

101 (44)

108 (51)

16.3

0.93

(±12.8) (2.3-66.8)

APB

91 (49)

103 (60)

23.7

0.89

(±16) (2.6-69.9)

ADM

113 (45)

123 (41)

21.5

0.81

(±15.1) (1.5-61.3) n = number of patients; the results are expressed as the mean ± SD; SD = standard deviation; COVm = average values of the coefficient of variation; ICC = intraclass correlation coefficient.

23

24