Association Between Changes in Electromyographic Signal Amplitude and Abdominal Muscle Thickness in Individuals With and Without Lumbopelvic Pain

Abstract

The muscles of the abdominal wall play a role in controlling motion of the spine, 1,20,21,24 and there is evidence of dysfunction of this musculature in persons with lumbopelvic pain (LPP). Proposed dysfunctions range from delayed onset 23 and seemingly diminished responses (smaller increase in thickness) of the transversus abdominis (TrA) 46,48 to increased electromyographic (EMG) signal amplitude and greater responses (larger increase in thickness) of the rectus abdominis (RA) and external and internal oblique (EO and IO) muscles.2,39,49 As these abnormalities have been linked to pain and dysfunction, there is interest in developing methods to accurately assess function of the abdominal musculature.Ultrasound imaging (USI) is a noninvasive and valid method to quantify abdominal muscle size 19 and is currently used in rehabilitation, in part, under the assumption that changes in muscle thickness reflect the extent of muscle contraction.42,46,50 USI has also been used for screening purposes 18 and as a source of real-time biofeedback, as it is presumed to be reflective of muscle activity during exercises commonly employed in rehabilitation programs. 15,38 Although some authors have stated that changes in muscle thickness measured with USI can be used as surrogate measures of muscle activity 11,34,42 and activation, 31,40,41,44 evidence T T STUDY DESIGN: Validation study. T T OBJECTIVES:To investigate the association between changes in electromyographic (EMG) signal amplitude and sonographic measures of muscle thickness of 4 abdominal muscles, during 2 clinical tests, in adults with and without lumbopelvic pain. T T BACKGROUND:There is a trend in rehabilitation to use ultrasound imaging (USI) to determine the extent of abdominal muscle contraction. However, the literature investigating the relationship between abdominal muscle thickness change and level of activation is inconclusive and has not included clinically relevant tasks. T T METHODS:Simultaneous recording from finewire EMG and USI was performed for 4 abdominal muscles, in 7 adults with lumbopelvic pain (mean  SD age, 29.7  12.0 years) and 7 adults without lumbopelvic pain (32.0  10.6 years), during an active straight leg raise (ASLR) test and an abdominal drawing-in maneuver (ADIM). Cross-correlation functions and linear regression analyses were used to describe the relationship between the 2 measures. Analyses of variance were used to compare individuals with and without lumbopelvic pain, with an alpha set at .05. T T RESULTS:Across all muscles, peak crosscorrelation values were low (ASLR, r = 0.28  0.09; ADIM, r = 0.35  0.11), and there was large variability in associated time lags (ASLR, τ = 0.69  2.56 seconds; ADIM, τ = 0.53  3.75 seconds). Regression analyses did not detect a systematic pattern of association between EMG signal amplitude and USI measurements, and analyses of variance revealed no differences between cohorts.

External abdominal oblique muscle ultrasonographic thickness changes is not an appropriate surrogate measure of electromyographic activity during isometric trunk contractions

“…In an attempt to better control the biomechanical demand of the EO muscle, we have used a specialized dynamometer combined with a direction-specific visual feedback. However, the strength and direction of the EMG/USI correlations were still variable across subjects, as observed previously [10,11,13] as well as more recently reported [16]. Finally, averaging measures across two EO sites did not really increase the proportion of statistically significant and positive correlations between EMG and USI measures.…”

“…The reader is referred to Brown and McGill [11] and Whittaker et al [16] for more detailed discussions on these factors. Interestingly, these authors introduced new ideas to reduce potential sources of error.…”

“…To date, three studies [10,11,16] found weak correlations (r correlation ranging from −0.22 to 0.30, depending on the task), most often statistically nonsignificant, between the two measures (EMG and USI) during trunk flexion contraction [10,11], abdominal drawing-in maneuver or straight leg raise [16], and brace, hollow and extension trunk efforts [11]. However, these studies used small samples of subjects (varying from 3 to 7 subjects) [10,11,16], which likely did not accurately represent inter-subject variations.…”

“…However, these studies used small samples of subjects (varying from 3 to 7 subjects) [10,11,16], which likely did not accurately represent inter-subject variations. Inter-subject variability is apparently of concern, as more recently shown with the consideration of time lag measures between EMG and USI measures [16].…”

“…However, these studies used small samples of subjects (varying from 3 to 7 subjects) [10,11,16], which likely did not accurately represent inter-subject variations. Inter-subject variability is apparently of concern, as more recently shown with the consideration of time lag measures between EMG and USI measures [16]. Moreover, none of these studies investigated a task involving the primary role of this muscle (axial rotation), which clearly appears to make a difference according to the results previously reported by John and Beith [13].…”