The reliability of surface EMG recorded from the pelvic floor muscles

https://doi.org/10.1016/j.jneumeth.2009.05.027Get rights and content

Abstract

The neuromuscular function of the pelvic floor muscles (PFMs) is frequently evaluated using surface electrodes embedded on vaginal probes. The purpose of this study was to determine the between-trial and between-day reliability of EMG data recorded from the PFM using two different vaginal probes while subjects performed PFM maximum voluntary contractions and a coughing task. The Femiscanā„¢ and the Periformā„¢ vaginal probes were used to acquire EMG data while the subjects performed the tasks. Peak RMS amplitudes were computed for each instrument, task, and side of the pelvic floor using a sliding window technique. The between-trial reliability was evaluated using intraclass correlation coefficients (ICCs) and coefficients of variation (CV). Between-trial reliability was determined using ICCs, Pearson's correlation coefficients, computing the mean absolute difference between days, and calculating the standard error the measurement (SEM) for each instrument and task. EMG amplitude differences were detected between the left and right PFM (pĀ <Ā 0.05), therefore all of the analyses were performed separately for each side. Overall, between-trial reliability was fair to high for the Femiscanā„¢ (ICC(3,1)Ā =Ā 0.58ā€“0.98, CVĀ =Ā 8.5ā€“20.7%) and good to high for the Periformā„¢ (ICC(3,1)Ā =Ā 0.80ā€“0.98, CVĀ =Ā 9.6ā€“19.5%), however between-day reliability was generally poor for both vaginal probes (ICC(3,1)Ā =Ā 0.08ā€“0.84). The results suggest that although it is acceptable to use PFM surface EMG as a biofeedback tool for training purposes, it is not recommended for use to make between-subject comparisons or to use as an outcome measure between-days when evaluating PFM function.

Introduction

The use of surface electromyography (SEMG) to assess the neuromuscular function of the pelvic floor muscles (PFMs) using electrodes embedded on vaginal probes is widely used to increase our understanding of pelvic floor function (Aukee et al., 2003, Aukee et al., 2002, Madill and McLean, 2008, Santiesteban, 1988, Sapsford and Hodges, 2001, Sapsford et al., 2001, Smith et al., 2006, Smith et al., 2007a, Smith et al., 2007b, Smith et al., 2008, Thompson et al., 2006a, Thompson et al., 2006b, Thorp et al., 1991). When SEMG is used clinically as well as in research, the extent to which the measure is valid and reliable is fundamental to the interpretation of the results obtained. Despite the common use of SEMG to describe pelvic floor muscle function, the reliability of the different devices used has not been fully established.

Surface electrodes record the electrical potential of contracting muscles from the skin surface. The signal recorded using surface electrodes is the summation of electrical activity that is detected at the skinā€“electrode interface from the large number of motor units activated during a given contraction. Surface electrodes are readily available in most clinical and research settings. They are easy to apply, non-invasive, inexpensive, and are a practical way to record the composite electrical activity produced by the motor units within the detection region of the electrode. Valuable information can be extracted from the EMG data recorded using surface electrodes such as identifying whether a muscle is ā€œonā€ or ā€œoffā€, if there is an increase or decrease in activity during a particular task, and if fatigue recorded changes are seen in the amplitude and frequency content of a signal. When appropriate steps are taken to standardize procedures and tasks, the reliability of SEMG data recorded from selected skeletal muscles has been reported to be acceptable (Kadaba et al., 1985, Kadaba et al., 1989, Kollmitzer et al., 1999, Marshall and Murphy, 2003, Ollivier et al., 2005, Yang and Winter, 1983). The between-trial reliability of SEMG tends to be better than the between-day reliability (Kadaba et al., 1989, Kollmitzer et al., 1999).

Unlike the electrodes routinely used on other skeletal muscles, the deep location of the PFMSs and the moist environment of the vaginal lumen preclude the use of self-adhesive recessed electrodes, which are normally used for SEMG acquisition to minimize electrode movement during tasks and thus the effect of motion artifact. The development of vaginal probes with electrodes embedded on their surfaces created a convenient way to measure SEMG from the PFMs that is more specific than recording electrical potentials from the perineum, respects the dignity of the participants, and is less invasive than the use of fine-wire electrodes. The EMG data recorded using a vaginal probe has the potential to allow clinicians and researchers to extract valuable information when diagnosing and treating women with PFM dysfunction.

Good reliability of SEMG signals recorded from the pelvic floor has been reported in the literature both qualitatively (Deindl et al., 1993, Santiesteban, 1988) and quantitatively (Aukee et al., 2002, Glazer et al., 1999, Thompson et al., 2006b, Thorp et al., 1991). Each of these studies used vaginal probes with different geometric properties, had detection surfaces of varying sizes embedded on them, configured the electrodes differently, and had subjects perform variations of similar tasks making comparisons across studies next to impossible. A closer investigation of the reported reliability values from these studies reveals significant limitations in their interpretation. Between-day reliability issues have been expressed in the literature for other skeletal muscles (Kadaba et al., 1989, Komi and Buskirk, 1970, Viitasalo and Komi, 1975, Yang and Winter, 1983), given the additional challenges when recording EMG from the PFMS, the true reliability may have been masked due to the combination of statistical methods applied and the random and variable nature of SEMG data. The purpose of this study was to determine the between-trial and between-day (testā€“retest) reliability of two different surface EMG detection devices commonly used to record PFM EMG activity during voluntary and reflex activation of the PFM in standing and in supine using a variety of statistical analysis.

Section snippets

Subjects

This study was approved by the Queen's University Health Sciences Research Ethics Board (REB# REH-315-06) and all subjects provided written informed consent. Healthy nulliparous females between the ages of 18 and 40 years were recruited from the Queen's University population to participate in this study. Twelve subjects volunteered and participated in the study, but data from only 10 of the subjects was used in the analysis due to data quality issues discussed in Section 3. Potential subjects

Subjects

Ten healthy women (30.0Ā Ā±Ā 3.9 years) participated in the study. The subject demographic information is reported in Table 1. On physical examination elevation of the pelvic floor was observed during a voluntary contraction in all women. Using digital palpation to assess PFM strength, the group average was grade three out of five on the modified Oxford grading system. Two of the women reported performing PFM contractions as an exercise on a regular basis. The length of time between EMG testing

Discussion

Between-trial reliability was acceptable using the Femiscanā„¢ and Periformā„¢ probes to record PFM EMG, but the testā€“retest reliability was poor. Therefore, the results of this study suggest that if using either probe as a biofeedback device to train an individual to recruit and activate the PFMs on a given day it is possible to detect training-related changes in PFM activation that may result from motor learning during that session. Based on the results of this study the use of the Periformā„¢ or

Acknowledgements

Support for this work from Natural Science and Engineering Research Council of Canada and Canadian Institute of Health Research is gratefully acknowledged.

References (45)

  • K. Singh et al.

    Magnetic resonance imaging of normal levator and anatomy and function

    Obstet Gynecol

    (2002)
  • M.D. Smith et al.

    Disorders of breathing and continence have a stronger association with back pain than obesity and physical activity

    Aust J Physiother

    (2006)
  • R.E. Allen et al.

    Pelvic floor damage and childbirth: a neurophysiological study

    Br J Obstet Gynaecol

    (1990)
  • P. Aukee et al.

    Intravaginal surface EMG probe design test for urinary incontinence patients

    Acupunct Electro-Therap Res Int J

    (2002)
  • F. Azpiroz et al.

    The puborectalis muscle

    Neurogastroenterol Motil

    (2005)
  • J.V. Basmajian et al.

    Muscles alive: their functions revealed by electromyography

    (1985)
  • K. Bo et al.

    Pelvic floor muscle exercise for the treatment of female stress urinary incontinence. I. Reliability of vaginal pressure measurements of pelvic floor muscle strength

    Neurourol Urodynam

    (1990)
  • C. Dannecker et al.

    EMG-biofeedback assisted pelvic floor muscle training is an effective therapy of stress urinary or mixed incontinence: a 7-year experience with 390 patients

    Arch Gynecol Obstet

    (2005)
  • F.M. Deindl et al.

    Activity patterns of pubococcygeal muscles in nulliparous continent women

    Br J Urol

    (1993)
  • F.M. Deindl et al.

    Pelvic floor activity patterns: comparison of nulliparous continent and parous urinary stress incontinent women. A kinesiological EMG study

    Br J Urol

    (1994)
  • C. Dumoulin et al.

    Reliability of dynamometric measurements of the pelvic floor musculature

    Neurourol Urodynam

    (2004)
  • D. Farina et al.

    The extraction of neural strategies from the surface EMG

    J Appl Physiol

    (2004)
  • Cited by (100)

    View all citing articles on Scopus
    View full text