Reliability of a modified motor unit number index (MUNIX) technique

Abstract

Introduction

The purpose of this study was to examine the relative and absolute between-day reliability of the motor unit number index (MUNIX).

Methods

Young, healthy adults (n = 19) attended two testing sessions separated by 4-weeks where their maximal pinch-grip strength, MUNIX, and motor unit size index (MUSIX) were assessed in the abductor pollicis brevis muscle. Reliability was assessed by intraclass correlation coefficients (ICC), coefficient of variation (CV) and limits of agreement (LOA).

Results

No mean differences were observed for MUNIX or MUSIX. The CV for the MUNIX and MUSIX measures were between 13.5% and 17.5%. The ICC for both measures were moderate to moderately-high (0.73–0.76), The LOA for both indicated a homoscedastic relationship.

Discussion

Our findings indicate moderate to moderately-high reliability for both MUNIX and MUSIX. Future work is needed to ensure both measures are reliable in other muscles and cohorts, and further investigations are required to examine the validity of MUNIX.

Introduction

A motor unit is defined as an alpha motor neuron and all of the muscle fibers it innervates (Sherrington, 1925). Quantification of motor unit number has long been of clinical and scientific significance as it relates to monitoring disease progression and/or assessing the effects of pharmacologic and behavioral interventions on motor unit numbers (Bromberg, 2013). Over the past four decades a number of techniques have been developed to estimate motor unit number in vivo (Doherty et al., 1995, Nandedkar et al., 2004, Rashidipour and Chan, 2008). The earliest of these techniques was the motor unit number estimation (MUNE) technique introduced in 1971 (McComas et al., 1971), and since this time a number of modifications to the MUNE technique have been developed and implemented (Brown et al., 1988, Daube, 1995, Doherty and Brown, 1994, Doherty and Stashuk, 2003, Kadrie et al., 1976, Shefner et al., 2011, Wang and Delwaide, 1995). The MUNE methods involve estimates of single motor unit action potential size, using either incremental electrical nerve stimulation, spike triggered averaging, or multipoint stimulation techniques. While these methods are generally considered the standard for in vivo motor unit number quantification they are not without shortcomings. For instance, the methods can be time consuming for both patient and examiner, and physically uncomfortable for the patient (e.g., the high number of electrical stimuli and/or insertion of a needle electrode into a muscle can be painful) (Rashidipour and Chan, 2008). As such, there has been a demand from both scientists and clinicians to develop non-invasive, easy to implement, and highly tolerable alternative techniques to obtain an in vivo estimate of motor unit number (Bromberg, 2013).

In 2004, Nandedkar and colleagues proposed a novel neurophysiological technique (the motor unit number index, or MUNIX) to derive an index associated with the number of motor units in a muscle (Nandedkar et al., 2004). The MUNIX is derived from the maximum compound muscle fiber action potential (CMAP, or M-wave) observed in response to supramaximal electrical stimulation and voluntary surface electromyogram (EMG) recordings associated with a series of submaximal muscle contractions. The MUNIX technique is non-invasive, quick (i.e., it requires ∼10 min to derive a MUNIX value per muscle), and easy to implement (i.e., the instrumentation and technical expertise required are readily available in most clinical and research settings). Additionally, it is considered quite tolerable to most (e.g., only a few electrical stimuli and muscle contractions are required). In general, the MUNIX is considered a value that is proportional to the motor unit numbers in a muscle, as opposed to representing an absolute number of motor units in a muscle (Nandedkar et al., 2004, Nandedkar et al., 2010, Nandedkar et al., 2011). In recent years, MUNIX has been used to quantify motoneuron loss in Amyotrophic Lateral Sclerosis (ALS), and it has been observed to be lower in (i) older adults when compared to younger adults (Neuwirth et al., 2011b) and (ii) paretic muscle when compared to contralateral muscles of stroke survivors (Li et al., 2011).

Only a few studies have examined the reliability of the MUNIX in healthy individuals where no changes in motor unit numbers are expected to occur over time (Ahn et al., 2010, Furtula et al., 2013, Nandedkar et al., 2011, Neuwirth et al., 2011a, Sandberg et al., 2011). The majority of these studies were poorly controlled in that within- and between-day test–retest data were pooled to derive measures of MUNIX stability (Ahn et al., 2010, Furtula et al., 2013, Sandberg et al., 2011, Neuwirth et al., 2011a). In fact, to our knowledge only one study has reported the between-day reliability of the MUNIX, and this study simply noted that the mean MUNIX values were similar across testing sessions occurring up to one-year apart in a small number of healthy subjects (n = 6–8; range: 154–162) (Nandedkar et al., 2011). Accordingly, the purpose of the present study was to comprehensively examine the relative and absolute reliability of the MUNIX in young, healthy individuals when assessed on two occasions separated by 4-weeks using a modified version of the original technique. Additionally, we also examined the relative and absolute reliability of the motor unit number size index (MUSIX), which is derived by dividing the maximum CMAP amplitude by the MUNIX. The MUSIX measurement, when interpreted alongside MUNIX, can provide more insight into the spinal motorneuron or motor unit pool. A well-developed simulation designed to parse out the subtleties of the technique noted that reduced MUNIX values without changes in MUSIX may be the result of muscle atrophy or motor unit number loss with incomplete reinnervation. On the contrary, a reduction in MUNIX and increase in MUSIX provides evidence of spinal motor neuron loss (Li et al., 2012).

We should note that our modified MUNIX technique is different than that classically employed as we controlled the contraction intensity by providing individuals visual feedback of their exerted force relative to their target force. Unlike the original technique, where surface interference patterns (SIPs) were obtained from contractions that were not controlled per se, we standardized the technique presented herein in an attempt to better control for the potential error introduced by subjective perception of contraction intensity. As such, these findings should not be interpreted or extrapolated to reflect the reliability of the originally described MUNIX technique, as it is possible that the differences in the two techniques could result in different levels of reliability.