EMG ANALYSIS: A SUMMARY OF QUANTITATIVE AND QUALITATIVE METHODS
Authored by
Dr Michael A. Leitch & Mr Mark B. Wyatt
Conventionally, EMG traces are analysed using qualitative methods. This approach is highly dependent on the clinicians experience, skill and prior training, owing to the subjective nature of this particular technique. The corollary to this is that different examiners may reach different conclusions about the diagnosis when studying the same patient (Fuglsang et al., 1999; Johnsen et al., 1994). Nonetheless, qualitative analysis does allow for overall judgments to be made that are useful in distinguishing normal from abnormal motor units (Richfield et al., 1981) and in detecting certain patterns which correlate to specific diseases. However, it is widely accepted that qualitative methods of EMG examination are limited for many reasons and in practice, they are correlated with other methods, eg. nerve conduction studies and clinical assessment. One problem is that the examiner is able to view only a “few” motor units at a time. This means that low-level motor unit activation patterns (MUPs) (recruitment patterns) are difficult to interpret even for the most experienced clinicians (Farkas et al., 2010; Adel & Stashuk., 2013). At low levels of activation, potentially important EMG patterns may be visible to the examiner for periods of 50msec or less, thus challenging even the best of experts. It is even more difficult to visually analyse the recruitment patterns in certain disorders, particularly myopathy (Preston & Shapiro, 2013). Secondly, determining the morphology of a motor unit requires assessment of its rise time, fall time, duration, peak area and number of phases/turns. Achieving this comprehensively (eg, assessing rise-times) is well-nigh impossible with conventional, qualitative methods. During a live tracing the EMG examiner must infer all of this information, which can be extremely challenging and as aforementioned, requires significant knowledge and experience. This can be compared with attempting to report an X-ray film or CT scan in just a few seconds and not taking the longer time to necessary to analyse it reliably. Lastly, qualitative EMG analysis often places too much reliance on the presence of single abnormal potentials while the muscle “as a whole” may be fairly normal. This is much more accurately characterised at higher-levels of voluntary activation, which is very difficult when analysing qualitatively (Krarup, 2016).
“Quantitative” methods overcome most of these problems. These methods utilise statistical and objective methods to interpret pre-recorded EMG data and permit all recorded MUPs to be “replayed” and analysed at leisure. One of the most apparent advantages of this methodology is the ability to quantify individual MUP parameters (Nandedkar et al., 1996; Roeleveld et al., 1997; Boe et al., 2005). Several systematic techniques have been used to evaluate the parameters of a motor unit in normal and diseased states. These parameters include: duration (Katsis et al., 2006), amplitude (Moosa and Brown, 1972; Bergmans, 1971; Mischelevich, 1970), rise-time and turns (Nandedkar et al., 1986), phases (McComas et al., 1978) and discharge rates (Zalewska et al., 2004 & 2005; Sonoo, 2002). These parameters are essential in discriminating reliably between normal and diseased or denervated muscles. The quantity of “usable” MUPs detected in muscle is reliant on the position of the examiner’s needle, as the varying degrees to which a disorder may affect specific motor units in a single MUP are not clear indicators of the actual state and should not be used individually. This is why a minimum of 20 MUPs need to be sampled in any given muscle (Farkas et al., 2010; Stalberg et al., 1996). Each of the 20 attainable MUPs is vigilantly analysed and the individual parameters are computed (Parsaei, 2010). These values can give crucial information about the neurological state of a muscle (neurogenic, myopathic or normal). Another parameter that can be analysed and graphed using quantitative techniques, is the interference pattern (IP). This is the sum of many MUPs firing simultaneously during maximal voluntary contraction (Finsterer, 2001). This parameter is particularly important when determining if a muscle is pathologically myogenic. This is due to the fact that in myopathy the MUPs have much smaller peak amplitudes and tend to be recruited simultaneously rather than in a progressive manner, following the size principle (Henneman et al., 1957 & 1965; Milner-Brown et al., 1973). Automated quantitative analysis was thus developed (Nandedkar et al., 1986; Stashuk, 1999; Stalberg, 2003). Another element to consider when comparing quantitative to qualitative methods, is the time needed for analysis. As described above, qualitative analysis is performed in seconds and once complete, the results are instantaneous. In contrast, the quantitative method requires detailed analysis of the EMG trace by a highly trained neurophysiologist. As detailed above, at least 20 individual motor units with appropriate rise-times are collected and analysed. Typically, this takes approximately 5-10 minutes per muscle, after which the report can be completed. This increases the time required for each EMG study. Nonetheless, the increased cost of equipment and staff time is justifiable, as the additional benefits of quantitative vs. qualitative EMG are significant.
Many modern NCS/EMG systems offer fundamental advantages of live rise-time calculation and display in live-time, while the needle EMG study is being performed. This particular parameter (rise time) is critical for ensuring reliable EMG data as it reflects how close the examiner’s needle tip is from an isolated motor unit – the greater the rise time, the further the examiner is from that particular motor unit (Stalberg et al., 1986; Barkhaus & Nandedkar, 1995). Recent advances in EMG software and technology have therefore enabled much more comprehensive analysis of EMG signals compared with previous qualitative methods. Not only are these advances in technology important in the diagnosis of neurological and spinal pathologies but they are valuable for those undertaking training in EMG techniques, clinicians and scientists alike. The diverse features available with new, quantitative EMG technology are impressive and will almost certainly continue to improve in the future (Farkas et al., 2010).
References
Adel T, Stashuk D. Clinical Quantitative Electromyography. Electrodiagnosis in New Frontiers of Clinical Research 1, 2013.
Barkhaus PE, Nandedkar SD. On the selection of concentric needle EMG motor unit action potentials for analysis: is the rise time criteria too restrictive? Muscle & Nerve 19: 1554-1560, 1995.
Bergmans J. Computer-assisted on-line measurement of motor unit potential parameters in human electromyography: Electromyography 11: 161-81, 1971.
Boe SG, Stashuk D, Brown WF, Doherty TJ. Decomposition-based quantitative electromyography: Effect of force on motor unit potentials and motor unit number estimates. Muscle & Nerve 31: 356-73, 2005.
Farkas C, Hamilton-Wright A, Parsaei H, Stashuk D. A Review of Quantitative Electromyography: Critical Reviews in Biomedical Engineering 38: 467-85, 2010.
Finsterer J. EMG-interference pattern analysis: Journal of Electromyography and Kinesiology 11: 231-246, 2001.
Fuglsang-Frederiksen A, Johnsen B, Carvalho M DE, Fawcett PRW, Liguori R, Nix W, Schofield I, Vila A. Variation in diagnostic strategy of the EMG examination – a multicentre study. Clinical Neurophysiology 10: 1814-1824, 1999.
Henneman E, Somjen G, Carpenter DO. Functional significance of cell size in spinal motoneurones. J Neurophysiology 28: 560-580, 1965.
Henneman E. Relation between size of neurons and their susceptibility to discharge. Science126, 1345–1347, 1957.
Johnsen B, Fuglsang-Frederiksen A, Vingtoft S, Fawcett P, Liguori R, Nix W, Otte G, Proenca J, Schofield I, Sieben G, Veloso M, Vila A. Differences in the handling of the EMG examination at seven European laboratories. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section 93: 155-158, 1994.
Katsis C, Goletsis Y, Likas A, Fotiadis D, Sarmas I. A novel method for automated EMG decomposition and MUAP classification. Artifical Intelligence in Medicine 37: 5564, 2006. Krarup C, Boeckstyns M, Ibsen A, Moldovan M, Archibald S. Remodeling of motor units after nerve regeneration studied by quantitative electromyography. Clin Neurophysiol 127: 1675-1682, 2016.
McComas AJ, Sica RE. Automatic quantitative analysis of the electromyogram in partially denervated distal muscles: comparison with motor unit counting. J Neuro Sci 5: 377-83, 1978.
Milner-Brown H S, Stein RB, Yemm R. The orderly recruitment on human motor units during voluntary isometric contractions. J. Physiol. 230: 359–370, 1973.
Mishelevich DJ. On-line real-time digital computer separation of extracellular neuroelectric signals. IEEE Trans Biomed Eng 17: 147-50, 1970.
Moosa A, Brown BH. Quantitative electromyography: a new analogue technique for detecting. J Neurol Neurosurg Psychiatry 35: 216-20, 1972.
Nandedkar SD, Sanders DB, Stalberg EV. Simulation and analysis of the electromyographic interference pattern in normal muscle. Part I: Turns and amplitude measurements. Muscle & Nerve 9: 423-430, 1986.
Preston DC, Shapiro BE. Electromyography and Neuromuscular Disorders: 3rd edition,
2013.
Richfield EK, Cohen BA, Albers JW. Review of Quantitative and Automated Needle Electromyographic analyses. IEEE Transactions of Biomedical engineering 7: 28, 1981.
Roeleveld K, Stegeman DF, Falck B, Stalberg EV. Motor unit estimation: confrontation of surface EMG with macro EMG. Electroencephalography and Clinical Neurophysiology/Electromyography and Motor Control 105: 181-188, 1997.
Sonoo M. New attempts to quantify concentric needle electromyography. Muscle & Nerve 25: 98-102, 2002.
Stalberg E. Methods for the quantitation of conventional needle EMG. Handbook of Clinical Neurophysiology 2: 213-244, 2003.
Stalberg EV, Andreassen S, Falck B, Lang H, Rosenfalck A, Trojaborg W. Quantitative analysis of individual motor unit action potentials: A proposition for standardized terminology and criteria for measurement. J Clin Neurophysiol 3: 313-348, 1986.
Stalberg EV, Falck B. The role of electromyography in neurology. Electroencephalography Clin Neurophysiol 103: 579-98, 1997.
Stalberg E, Nandedkar SD, Sanders DB, Falck B. Quantitative Motor Unit Potential Analysis. Journal of Clinical Neurophysiology 5: 401-422, 1996.
Zalewska E, Hausmanowa- Petrusewizc I, Stalberg EV. Modeling studies on irregular motor unit potentials. Clin Neurophysiol 115: 543-56, 2004.
Zalewska E, Hausmanowa- Petrusewizc I. The SIIR index – a non-linear combination of waveform size and irregularity parameters for classification of motor unit potentials. Clin Neurophysiol 116: 957-64, 2005.
