ABSTRACT: Template-operated motor unit potential (MUP) analysis has made quantitative electromyography (EMG) feasible, even in busy laboratories, but validation of this approach is still necessary. In the present study, the utility of multi-MUP analysis was assessed in patients with a molecular genetic diagnosis of facioscapulohumeral muscular dystrophy (FSHD). Manual assessment of muscle strength and concentric-needle EMG of the biceps brachii and vastus lateralis muscles were performed. The sensitivity for diagnosing myopathy (mean values and outliers) was tested for eight MUP parameters and four of their combinations. The group comprised 31 patients. Elbow flexion and knee extension strength was normal in 45% and 52% of patients, respectively. The most sensitive MUP parameter was thickness, followed by duration. A combination of three MUP parameters (thickness, amplitude, and duration/area) was needed for maximal sensitivity. The study demonstrated a high sensitivity of multi-MUP analysis in FSHD. Myopathic abnormalities were demonstrated in all weak biceps brachii muscles, and in 77% of biceps brachii muscles with normal strength. Muscle Nerve 34: 451– 456, 2006
SENSITIVITY OF MOTOR UNIT POTENTIAL ANALYSIS IN FACIOSCAPULOHUMERAL MUSCULAR DYSTROPHY SIMON PODNAR, MD, DSc, and JANEZ ZIDAR, MD, DSc Institute of Clinical Neurophysiology, Division of Neurology, University Medical Center, SI-1525 Ljubljana, Slovenia Accepted 15 May 2006
From its beginning in the 1950s, quantitative electromyographic (EMG) analysis was a slow and cumbersome procedure. As a consequence it was only rarely performed in the routine diagnostic evaluation of patients with neuromuscular disorders. The situation changed dramatically with the appearance of template-operated motor unit potential (MUP) analysis. Using this approach, quantitative EMG became feasible, even in busy laboratories.13,24 However, validation of this approach is still required. In our previous studies we validated in detail multiMUP analysis for neuropathic conditions in the external anal sphincter (EAS) muscle.15,17–19 For other muscles and conditions such validation has not yet been performed. In the present study, we examined the utility of multi-MUP analysis in a specific myopathic disorder. Abbreviations: EAS, external anal sphincter; EMG, electromyography; FSHD, facioscapulohumeral muscular dystrophy; MRC, Medical Research Council; MUP, motor unit potential Key words: biceps brachii; facioscapulohumeral muscular dystrophy; motor unit potential; quantitative electromyography; sensitivity; vastus lateralis Correspondence to: S. Podnar; e-mail:
[email protected] © 2006 Wiley Periodicals, Inc. Published online 7 July 2006 in Wiley InterScience (www.interscience.wiley. com). DOI 10.1002/mus.20613
Quantitative EMG in FSHD
The sensitivity of individual MUP parameters and the cumulative sensitivity of several combinations of MUP parameters were calculated in the biceps brachii and vastus lateralis muscles of patients with a genetic diagnosis of facioscapulohumeral muscular dystrophy (FSHD). MATERIALS AND METHODS
All patients with a clinical diagnosis of FSHD in our register of neuromuscular disorders received a written invitation to undergo molecular genetic testing. In addition, all patients who were clinically diagnosed with FSHD during the study (2003—2005) were included. The pretest diagnostic evaluation included history, clinical examination, and quantitative EMG. The study was approved by the National Ethics Committee of Slovenia, and all subjects provided informed consent. During genetic testing, DNA digestion with EcoRI and BlnI restriction enzymes was followed by Southern blot analysis using the genomic hybridization probe p13E11.28 The diagnosis of FSHD was confirmed by detection of short fragments (⬍35 kb) within the FSHD locus on chromosome 4q35.26 Only patients with positive genetic testing were included in the study.
MUSCLE & NERVE
October 2006
451
On clinical examination the strength levels of elbow flexion and knee extension were assessed using the Medical Research Council (MRC) scale. Quantitative EMG of the biceps brachii and vastus lateralis muscles was performed using a standard concentric EMG needle, and a commercially available EMG system (Keypoint; Medtronic Functional Diagnostics, Skovlunde, Denmark) with standard settings (filters: 5 Hz to 10 kHz). Right-side muscles were examined in all patients, without regard to asymmetry in muscle wasting or strength. EMG activity was observed during rest to detect fibrillation potentials or positive sharp waves, following which we attempted to obtain at least 20 different MUPs from each muscle using the template-operated multi-MUP analysis.24 On applying this analysis during a constant level of EMG activity, the operator initiated the computerized acquisition of the previous (last) 4.8 s of the signal. From the acquired signal, MUPs were then automatically extracted, quantitatively described, and sorted into one of six classes, with each class representing consecutive discharges of a particular MUP. MUP classes were averaged and presented according to the number of MUP discharges (in descending order) during the period of analysis.19,24 Analysis was performed at slight voluntary activation at which 3–5 MUPs were sampled. Duration cursors remained unchanged; MUPs with an unsteady baseline were deleted. MUP amplitude, duration, spike duration, area, number of phases, and number of turns were measured. In addition, MUP thickness (area/amplitude)10 and size index (2 * log amplitude ⫹ area/amplitude)21 were calculated. Only muscles with at least 20 MUPs sampled were included in further analyses. MUP parameters from patients’ muscles were compared to the 5th and 95th percentile reference limits for mean values, and to the 5th and 95th percentile of 10th percentile “outlier” reference limits.16,23 Reference limits were obtained from a control population included in an international multicentric compilation of normative MUP data (34 biceps brachii and 46 vastus lateralis subjects). Normative data obtained from the majority of these subjects have been published previously.3 To declare the muscle abnormal for individual MUP parameters, either the mean value or at least 3 (out of 20) individual MUPs (outliers) had to be outside the normative range.23 Changes were designated “myopathic” if MUP parameter values (except number of phases and number of turns) were below the normative range. For biceps brachii and vastus lateralis muscles, MUP parameters were added consecutively in a de-
452
Quantitative EMG in FSHD
creasing order of sensitivity to obtain the sensitivity of all eight MUP parameters. Several patterns of MUP parameters were then tested: (a) MUP area, duration, and number of turns17; (b) MUP thickness, duration, and number of phases; (c) combination of MUP parameters providing identical sensitivity to all MUP parameters; and (d) all eight MUP parameters. Sensitivity when at least one, at least two, and at least three of six diagnostic criteria [3 MUP parameters ⫻ 2 statistics (mean values, outliers) ⫽ 6 diagnostic criteria] were below (above for number of phases and turns) the appropriate normative range15 were then calculated. The sensitivity levels of MUP analyses in the biceps brachii and vastus lateralis muscles with normal strength (grade 5 on the MRC scale) of elbow flexion and knee extension, respectively, and in those muscles with reduced strength (grade ⬍5), were also calculated. For biceps brachii and vastus lateralis muscles, Spearman’s correlation coefficient (r) between muscle strength (MRC grades of elbow flexion and knee extension) and quantitative EMG findings [overall EMG grading (definitely, probably, possibly abnormal, and normal),15 and individual MUP parameters in biceps brachii and vastus lateralis] was calculated. RESULTS
Of the 38 eligible patients, 7 were excluded because genetic studies were either negative (in 5) or not available (in 2). Quantitative EMG was performed in the remaining 31 patients (17 men), aged 22–77 years (median, 43 years), with a molecular genetic diagnosis of FSHD. Muscle strength of elbow flexion was normal (grade 5) in 45%, and of knee extension in 52% of patients. Fibrillation potentials or positive sharp waves were found in 2 (6%) biceps brachii and in 3 (10%) vastus lateralis muscles. MUP analysis was completed in the biceps brachii in 29 and in the vastus lateralis in 30 patients. Due to insufficient muscle bulk MUP analysis could not be completed in either muscle in 1 patient, and in the biceps brachii muscle in another patient (only 13 MUPs could be sampled). It was often difficult to obtain an appropriately low level of MUP recruitment, as EMG pattern became dense even on minimal muscle activation. The sensitivity of individual MUP parameters is presented in Table 1, and the cumulative sensitivity of 4 MUP parameter combinations in Table 2. In both the biceps brachii and vastus lateralis muscles, three MUP parameters were needed to obtain the same sensitivity as all eight MUP parameters (i.e., biceps
MUSCLE & NERVE
October 2006
Table 1. Sensitivity (%) of mean values and outliers for individual MUP parameter values below and above the confidence limits. Biceps brachii Mean values
Outliers
Vastus lateralis Combined
Mean values
Outliers
Combined
MUP parameter
⬍5th percentile
⬎95th percentile
⬍5th percentile
⬎95th percentile
⬍5th percentile
⬎95th percentile
⬍5th percentile
⬎95th percentile
⬍5th percentile
⬎95th percentile
⬍5th percentile
⬎95th percentile
Thickness Duration Size index Spike duration Area Amplitude Phases Turns
55 31 35
0 0 0
59 55 28
0 3 0
69 55 38
0 3 0
43 40 33
3 7 0
43 10 37
0 7 0
50 40 43
3 7 0
30 31 14 0 7
0 0 14 10 0
44 24 31 0 3
4 3 17 10 0
48 31 35 0 10
4 3 17 14 0
13 20 10 7 0
4 0 0 3 0
25 40 33 7 0
4 0 0 7 0
29 40 33 10 0
4 0 0 7 0
Parameters are presented according to their sensitivity (from highest to lowest). Data were obtained in 29 biceps brachii and 30 vastus lateralis muscles of patients with a genetic diagnosis of facioscapulohumeral muscular dystrophy. Nonparametric normative limits were used.
brachii, and no correlation was found in vastus lateralis.
brachii, 90%; vastus lateralis, 60%). In the biceps brachii these MUP parameters were thickness (69%), duration (increased sensitivity for 14% ⫽ ⫹14%) and amplitude (⫹7%), and in the vastus lateralis they were thickness (50%), area (⫹7%), and amplitude (⫹3%). In the biceps brachii muscles with normal/reduced strength of elbow flexion, the sensitivity of MUP analysis using the same three MUP parameters was 77%/100%, 31%/81%, and 15%/ 63%, and in the vastus lateralis muscles with normal/reduced strength of knee extension the sensitivity was 53%/67%, 33%/67%, and 27%/47% when at least one, at least two, and at least three of six diagnostic criteria were below the appropriate normative range (or above it for number of phases and turns). Using Spearman’s correlation coefficient (r), in biceps brachii we found a significant (P ⬍ 0.01) correlation between elbow flexion MRC grades and overall EMG grading (r ⫽ 0.71). In addition, MUP parameters of duration (r ⫽ 0.68), area (r ⫽ 0.64), thickness (r ⫽ 0.65), and size index (r ⫽ 0.68) also correlated significantly (P ⬍ 0.01) with MRC grades. In contrast, no correlation was found for the remaining MUP parameters (amplitude, number of phases, number of turns, and spike duration) in biceps
DISCUSSION
Although the present study was not designed primarily to estimate the overall sensitivity of multi-MUP analysis to diagnose myopathy, it demonstrated a high sensitivity of template-operated quantitative MUP analysis in patients with a genetic diagnosis of FSHD. All biceps brachii muscles with reduced strength of elbow flexion on manual testing showed myopathic features. Furthermore, the sensitivity of MUP analysis in biceps brachii muscles with completely normal strength on manual testing was 77%. The sensitivity of MUP analysis in the vastus lateralis was lower, which accords with the less frequent involvement of this muscle in FSHD.14 MUP parameters correlated with muscle strength only in the biceps brachii, not in the vastus lateralis. The overall sensitivity of the analysis could be further increased if a larger number, and only clinically affected muscles, were studied. At least for FSHD, selective muscle involvement is characteristic, with the biceps brachii and particularly the vastus lateralis usually not being the first and most affected muscles.14 Fur-
Table 2. Sensitivity (%) of MUP analysis for detection of myopathic abnormalities in the biceps brachii (BB) and vastus lateralis (VL) using four sets of MUP parameters, and a different number of diagnostic criteria to declare a muscle abnormal. One criterion Parameters Area, duration, number of turns Thickness, duration, number of phases Thickness, amplitude, duration (BB)/area (VL) All eight MUP parameters
Two criteria
Three criteria
BB
VL
BB
VL
BB
VL
55% 83% 90% 90%
57% 50% 60% 60%
38% 52% 59% 59%
27% 43% 50% 57%
28% 41% 41% 48%
23% 37% 37% 50%
Data were obtained in 29 biceps brachii and 30 vastus lateralis muscles of patients with a genetic diagnosis of facioscapulohumeral muscular dystrophy.
Quantitative EMG in FSHD
MUSCLE & NERVE
October 2006
453
thermore, the disease process in FSHD is typically very asymmetric,14 so the sensitivity of MUP analysis would also be improved if the more affected side were examined, although we examined only the right side. Only two previous studies have provided data on the sensitivity of quantitative EMG in FSHD patients, but neither used multi-MUP or similar templateoperated MUP analysis.7,8 Sensitivity levels for MUP analysis in the study by Buchthal and Kamieniecka (85%) and our study were similar, despite the different inclusion criteria, muscle selection, and EMG criteria used.7 A similar sensitivity was also found in the study by Fawcett et al. (92%), who evaluated only MUP duration and polyphasia.8 These two MUP parameters yielded a sensitivity of only 59% in our study. It is possible that patients with a clinical, but not a genetic, diagnosis of FSHD included in these two studies were more severely affected, but no data on muscle strength of tested muscles were provided in these studies. The study by Buchthal and Kamieniecka using manual MUP analysis was also significant for the finding of a higher sensitivity of quantitative EMG analysis (sensitivity in 188 patients with a clinical diagnosis of myopathy was 87%) than muscle biopsy (sensitivity in FSHD was 62%, and in all patients with myopathy was 79%).7 The value of quantitative EMG over routine qualitative EMG in myopathy was demonstrated in another important study performed in patients with inclusion-body myositis prior to muscle biopsy.5 Whereas quantitative EMG (using manual MUP analysis) demonstrated myopathic changes in all 13 patients (sensitivity 100%), equally experienced electromyographers diagnosed myopathy in only 1 of 8 patients (sensitivity 13%) on routine qualitative EMG.5 When there is clinical suspicion of myopathy, EMG is usually performed for substantiation of the diagnosis before proceeding with supplementary procedures such as muscle biopsy or genetic testing. Because of the high sensitivity of genetic testing in FSHD (positive in 90 –95% of cases),26 however, genetic analyses might be a logical step when clinical examination reveals characteristic findings. Genetic testing with a negative result does not exclude the diagnosis of FSHD, because the disease may not be caused by the deletion but rather by other mutations affecting function of the D4Z4 sequence.26 In our study, genetic testing was negative in only 5 of 36 patients. The positive predictive value of clinical examination combined with positive MUP analysis was thus very high (86%). MUP analysis also proved to
454
Quantitative EMG in FSHD
be more sensitive than other laboratory tests; serum creatine kinase was abnormal in only 33% of patients with normal elbow flexion. Furthermore, quantitative EMG was shown previously to be more sensitive than muscle biopsy.7 The value of quantitative EMG in other myopathies with a less distinctive clinical phenotype and less useful or available genetic testing might be much higher than in FSHD. Our quantitative EMG study was performed on a group of patients with a genetic diagnosis of a specific myopathy. The template-operated sampling technique was used previously only in smaller studies of patients with acquired myopathies.2,13 Most contemporary EMG systems have the capability for some form of template-operated MUP analysis, which is fast and thus clinically useful.13,24 The most sensitive MUP parameter in both of the examined muscles was thickness,10 followed by duration,6,7 size index,21 spike duration, area, amplitude, number of phases, and number of turns (Table 1). MUP thickness was developed specifically to increase the sensitivity of MUP analysis in myopathies,10 and our study confirmed its value, as also demonstrated in a study using single-MUP analysis.25 Simple and polyphasic MUPs were not separated when determining duration because this would have made analysis more difficult and required sampling of a larger number of MUPs. Separation was suggested by Buchthal,7 and confirmed by others,5,12,27 to increase the sensitivity of MUP analysis in myopathies. Barkhaus et al. also found that the sensitivity of thickness was higher than duration, although they measured duration only for simple MUPs.1 During routine qualitative MUP analysis, the duration of the main negative MUP spikes is usually observed rather than MUP duration itself. Although duration of the negative spike at first sight corresponds to MUP thickness (measurement of the duration of a triangle defined by MUP area and amplitude),10 we found the sensitivity of spike duration to be lower than the sensitivity of thickness. It is not clear whether this is also due to its persistent relation to amplitude or the inclusion of late components in the measurement of spike duration.22 In this study, area and size index were less sensitive than in the diagnosis of neuropathy.17–19 The low sensitivity of the number of turns is in even sharper contrast to our previous studies of neuropathic conditions in the anal sphincter muscle,17,18 and was even lower than the number of phases, which had a very low sensitivity in neuropathy.17,18 This can probably be at least partially explained by the different amplitude thresholds of the two parameters (number of turns, 50 V; number of phases, 20 V). Thus, Stewart et al., who used a
MUSCLE & NERVE
October 2006
25-V criterion for number of turns, found that its sensitivity was surpassed only by MUP thickness in patients with different myopathies.25 As in neuropathy,15,18 outliers demonstrated a somewhat higher sensitivity than mean values in the present study of myopathy. The application of both mean values and outliers increased the sensitivity quite substantially in some MUP parameters (Table 1). The sensitivity was further increased by the concomitant use of several MUP parameters. The sensitivities obtained with three MUP parameters (MUP thickness, duration, and number of phases), which were previously claimed to be most useful in myopathies,7,10 were higher than those obtained with the three MUP parameters (MUP area, duration, and number of turns) most useful in neuropathic sphincter muscles.17,18 The present study thus confirmed our previous notion that no uniform pattern of MUP parameters is optimal in all muscles and conditions.15,17 In the present study, in both the biceps brachii and vastus lateralis muscles, we were able to identify a set of three MUP parameters (thickness and amplitude in both muscles, duration in the biceps brachii, and area or size index in the vastus lateralis) that provided the same sensitivity as all eight MUP parameters. This points to an even higher correlation between MUP parameters in myopathy than neuropathy; in neuropathy, six MUP parameters were needed to obtain the same sensitivity as all eight MUP parameters.18,20 However, when more than one diagnostic criterion was needed to declare a muscle pathological, all MUP parameters were more sensitive than each of these three combinations of MUP parameters (Table 2), but this would probably result in a low specificity of MUP analysis. In contrast to some previous reports,9 we found no “neuropathic abnormalities” in FSHD. Those studies probably misinterpreted the increased MUP amplitude as a sign specific for neuropathy. Because MUP amplitude is dependent on only a few or even the single nearest muscle fiber,11 increased MUP amplitude can result from muscle fiber hypertrophy,2 which is a compensatory mechanism that often occurs in myopathy after muscle fiber loss.4 The present study thus demonstrated a high sensitivity of multi-MUP analysis in the biceps brachii and vastus lateralis muscles of patients with a genetic diagnosis of FSHD, particularly when clinically involved muscles were tested. New MUP parameters, particularly MUP thickness, seem to significantly increase the sensitivity of MUP analysis. However, further studies on specificity are needed to firmly estab-
Quantitative EMG in FSHD
lish the position of template-operated MUP analysis in this and other myopathies. The authors thank Professor Erik Stålberg, Uppsala, Sweden, for data obtained from normative subjects; Professor Reginald Bittner, Vienna, Austria, for genetic testing of most of our patients; and Dr. Dianne Jones for language review. The study was supported by the Republic of Slovenia Research Agency (Grant No. J3 7899).
REFERENCES 1. Barkhaus PE, Nandedkar SD, Sanders DB. Quantitative EMG in inflammatory myopathy. Muscle Nerve 1990;13:247–253. 2. Barkhaus PE, Periquet MI, Nandedkar SD. Quantitative electrophysiologic studies in sporadic inclusion body myositis. Muscle Nerve 1999;22:480 – 487. 3. Bischoff C, Stalberg E, Falck B, Eeg-Olofsson KE. Reference values of motor unit action potentials obtained with multiMUAP analysis. Muscle Nerve 1994;17:842– 851. 4. Bradley WG, Jones MZ, Mussini JM, Fawcett PR. Becker-type muscular dystrophy. Muscle Nerve 1978;1:111–132. 5. Brannagan TH, Hays AP, Lange DJ, Trojaborg W. The role of quantitative electromyography in inclusion body myositis. J Neurol Neurosurg Psychiatry 1997;63:776 –779. 6. Buchthal F. An introduction to electromyography. Oslo: J.W. Capelen; 1957. 7. Buchthal F, Kamieniecka Z. The diagnostic yield of quantified electromyography and quantified muscle biopsy in neuromuscular disorders. Muscle Nerve 1982;5:265–280. 8. Fawcett PRW, Kennett RP, Schofield IS, Johnson MA. A neurophysiological investigation of the facio-scapulo-humeral (FSH) syndrome. J Neurol Sci 1990;98(suppl):195. 9. Furukawa T, Toyokura Y. Chronic spinal muscular atrophy of facioscapulohumeral type. J Med Genet 1976;13:285–289. 10. Nandedkar SD, Barkhaus PE, Sanders DB, Stalberg EV. Analysis of amplitude and area of concentric needle EMG motor unit action potentials. Electroencephalogr Clin Neurophysiol 1988;69:561–567. 11. Nandedkar SD, Sanders DB, Stalberg EV, Andreassen S. Simulation of concentric needle EMG motor unit action potentials. Muscle Nerve 1988;11:151–159. 12. Nandedkar SD, Sanders DB. Simulation of myopathic motor unit action potentials. Muscle Nerve 1989;12:197–202. 13. Nandedkar SD, Barkhaus PE, Charles A. Multi-motor unit action potential analysis (MMA). Muscle Nerve 1995;18:1155– 1166. 14. Padberg GW, Lunt PW, Koch M, Fardeau M. Diagnostic criteria for facioscapulohumeral muscular dystrophy. Neuromuscul Disord 1991;1:231–234. 15. Podnar S. Criteria for neuropathic abnormality in quantitative anal sphincter electromyography. Muscle Nerve 2004;30: 596 – 601. 16. Podnar S. Comparison of different outlier criteria in quantitative anal sphincter electromyography. Clin Neurophysiol 2005;116:1840 –1845. 17. Podnar S, Mrkaic M. Predictive power of motor unit potential parameters in anal sphincter electromyography. Muscle Nerve 2002;26:389 –394. 18. Podnar S, Vodusek DB. Standardization of anal sphincter electromyography: utility of motor unit potential parameters. Muscle Nerve 2001;24:946 –951. 19. Podnar S, Vodusek DB, Stalberg E. Comparison of quantitative techniques in anal sphincter electromyography. Muscle Nerve 2002;25:83–92. 20. Podnar S, Zalewska E, Hausmanowa-Petrusewicz I. Evaluation of the complexity of motor unit potentials in anal sphincter electromyography. Clin Neurophysiol 2005;116:948 –956.
MUSCLE & NERVE
October 2006
455
21. Sonoo M, Stalberg E. The ability of MUP parameters to discriminate between normal and neurogenic MUPs in concentric EMG: analysis of the MUP “thickness” and the proposal of “size index.” Electroencephalogr Clin Neurophysiol 1993;89:291–303. 22. Stalberg E, Andreassen S, Falck B, Lang H, Rosenfalck A, Trojaborg W. Quantitative analysis of individual motor unit potentials: a proposition for standardized terminology and criteria for measurement. J Clin Neurophysiol 1986;3:313– 348. 23. Stalberg E, Bischoff C, Falck B. Outliers, a way to detect abnormality in quantitative EMG. Muscle Nerve 1994;17:392– 399. 24. Stalberg E, Falck B, Sonoo M, Stalberg S, Astrom M. Multi-MUP EMG analysis—a two year experience in daily clinical work. Electroencephalogr Clin Neurophysiol 1995;97:145–154.
456
Quantitative EMG in FSHD
25. Stewart CR, Nandedkar SD, Massey JM, Gilchrist JM, Barkhaus PE, Sanders DB. Evaluation of an automatic method of measuring features of motor unit action potentials. Muscle Nerve 1989;12:141–148. 26. Tupler R, Gabellini D. Molecular basis of facioscapulohumeral muscular dystrophy. Cell Mol Life Sci 2004;61:557– 566. 27. Uncini A, Lange DJ, Lovelace RE, Solomon M, Hays AP. Long-duration polyphasic motor unit potentials in myopathies: a quantitative study with pathological correlation. Muscle Nerve 1990;13:263–267. 28. Wijmenga C, Hewitt JE, Sandkuijl LA, Clark LN, Wright TJ, Dauwerse HG, et al. Chromosome 4q DNA rearrangements associated with facioscapulohumeral muscular dystrophy. Nat Genet 1992;2:26 –30.
MUSCLE & NERVE
October 2006
