You are on page 1of 11

See discussions, stats, and author profiles for this publication at: https://www.researchgate.


Abdominal muscle recruitment during a range
of voluntary exercises

Article in Manual Therapy · June 2005
DOI: 10.1016/j.math.2004.08.011 · Source: PubMed


132 303

4 authors:

Donna M Urquhart Paul W Hodges
Monash University (Australia) University of Queensland


Trevor J Allen Ian Story
Monash University (Australia) Deakin University


Some of the authors of this publication are also working on these related projects:

Psychological Features and Their Relationship to Movement-Based Subgroups in People Living with
Low Back Pain View project

The Newcastle FRED Intervention for Low Back Pain Study View project

All content following this page was uploaded by Paul W Hodges on 09 March 2017.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.

. Commercial Rd.: +61 3 9903 0590. However. r 2004 Elsevier TrA is emphasized initially. For example. 1999) and dysfunction of this muscle occurs in people with LBP (Hodges and Richardson. Abdominal muscles. Corresponding author. 1998. However. Storyb a Department of Epidemiology and Preventive Medicine. recruitment of the abdom. Australia Received 5 March 2003. Victoria. 1992. 1972.1016/j. 1998. Low back pain 1.. Central and Eastern Clinical School.. Introduction et al. All rights reserved. 1990). OE was predominately recruited.. Inward movement of the lower abdominal wall in supine produced greater activity of TrA compared to OI. abdominal bracing..M.08. Australia b School of Physiotherapy..b. Urquhart).. and inward movement of the lower and upper abdominal wall. Trevor J.see front matter r 2004 Elsevier Ltd. control have not been clearly defined. Alfred Hospital. The University of Melbourne. New South Wales. Queensland. This et al. Goldman et al. The University of Queensland.d. Vezina and Hubley. O’Sullivan et al.. Australia c Prince of Wales Medical Research Institute.. Hodges. received in revised form 11 August 2004. Allison muscles in the early stages of rehabilitation. Tel. Richardson et al. 1990). few studies have directly investigated the activity of all the abdominal muscles or the recruitment of regions of the abdominal muscles during these manoeuvres. Jull et al. Regions of TrA were recruited differentially and an inverse relationship between lumbopelvic motion and TrA electromyography (EMG) was found. with minimal activity of the superficial abdominal 1980. Monash University. 1981. approach is based on evidence that activity of TrA Kozey.011 .au are considered to be important for the restoration of (D. normal function and progression involves strategies for 1356-689X/$ . Strohl et al. doi:10. pelvic tilting.. recruitment varies between regions of TrA. Urquharta. Although recruitment of fax: +61 3 9903 0556. pelvis and abdomen during inward movement of the lower abdominal wall. all of the trunk muscles E-mail address: donna. OE and RA. Transversus Original article Abdominal muscle recruitment during a range of voluntary exercises Donna M.2004.urquhart@med. 1960. Melbourne. Victoria 3004. intramuscular EMG studies are inconclusive (Carman Hodges et al. Allenb. obliquus internus (OI) and externus abdominis (OE). Keywords: Exercises. and movement of the lumbar spine.. All rights reserved. Paul W. De Troyer et al. This study indicates that inward movement of the lower abdominal wall in supine produces the most independent activity of TrA relative to the other abdominal muscles. ARTICLE IN PRESS Manual Therapy 10 (2005) 144–153 www. During posterior pelvic tilting. middle OI was most active and with abdominal bracing. 1998. 1996b. Most studies have A contemporary approach for low back pain (LBP) used surface electromyography (EMG) to investigate involves recruitment of transversus abdominis (TrA) these techniques (Partridge and Walters. Kennedy. accepted 27 August 2004 Abstract Various exercises are used to retrain the abdominal muscles in the management of low back pain and other musculoskeletal disorders. 1987. three different A diverse range of exercises is used clinically to retrain recruitment patterns were reported when six subjects the trunk muscles. and observation of abdominal and lumbopelvic motion may assist in evaluation of exercise performance. This study examined the activity of different regions of transversus abdominis (TrA). 2000) and the results of the small number of contributes to spinal control (Cresswell et al. 1990. Ian H. 2001).elsevier.math. Australia d Department of Physiotherapy. 1995.. and rectus abdominis (RA). were instructed to ‘‘pull in’’ their abdominal wall (De inal muscles during exercises that aim to restore motor Troyer et al.

. OI and obliquus externus An additional consideration is that there are differ. 174(9) cm. and image (Fig. 1999). are related to declaration of Helsinki. Subjects were al. O’Sullivan abdominal. A ground electrode rotation to lower and middle fibres (Urquhart et al. Abdominal bracing wire electrodes inserted into three regions of the (lateral flaring of the abdominal wall) (Kennedy.. Vezina and stainless steel wire (75 mm) (A-M Systems Inc. However. Mountain View. All procedures (Jull et al.1. 1972). halfway between the TrA are also active with the opposite direction of trunk umbilicus and the pubic symphysis. 2001). neurological or respiratory et al. and weight of 30(4) years. direct EMG measures of TrA motor control (Hodges et al.. and middle and lower fascicles that Pairs of surface EMG electrodes (Ag/AgCl discs. Electrodes were inserted into tilting are not clearly defined in people with or without the upper region of TrA (adjacent to the 8th rib). with 1 mm of Teflon removed from regarding the specific patterns of abdominal muscle the ends. their recruitment has UK). A marker was placed on the spinous . ARTICLE IN PRESS D. Methods video camera (Sony DCR TRV20. The aims of this study were to investigate recruitment of regions of the abdominal muscles during 2. Cresswell et al. no study has directly investigated 2. Electrodes abdominal pressure and by reduction of the lumbar were fabricated from two strands of Teflon-coated lordosis. condition. spinal and pelvic motion. Subjects A diffuse light source. 1998). 1980) abdominal wall under the guidance of real-time ultra- and posterior pelvic tilting have been proposed to sound imaging (5 MHz curved array transducer) improve lumbopelvic control by elevation of intra. Electromyography TrA activity during this. However. and the lower region of TrA and obliquus internus abdominis (OI) (Askar. / Manual Therapy 10 (2005) 144–153 145 re-education of the whole muscle system (Richardson et 68(15) kg. gastrointestinal.. 1999. The data was exported and analysed using Matlab not been comprehensively investigated during voluntary 6 (release 12. Surface naire (Baecke et al. such as observation of A video motion analysis system was used to quantify abdominal. the LBP (Vezina et al.. middle region of TrA.70  38 mm) and the tips bent back concluded that muscle activation patterns during pelvic 1–2 mm to form hooks. 2004). The efficacy of this method has been excluded if they had a history of low back or leg pain established in randomized control trials with acute and that affected function in the preceding 2 years. Five subjects had performed EMG studies indicate that activity of the superficial the exercises previously and all subjects were involved in abdominal muscles is minimal during this manoeuvre another study (Urquhart et al. cage are horizontal. or any chronic LBP patients (Hides et al. Natick. The technique involves inward move.. there is controversy Washington. abdominis (OE) (midway between the iliac crest and ences in the morphology and recruitment of regions of inferior border of the rib cage). CA). placed under the subject’s abdomen. Urquhart et al. using a Power1401 data acquisition system and Spike2 Although these reports suggest regional differences in software (Cambridge Electronic Design. Cambridge. Hubley-Kozey. 1990.c). Data were captured with a digital 2. was placed on the iliac crest. MA. with a mean (SD) highlight the edge of the abdominal wall in the video age. Tokyo. 1980. Hodges et al. (128XP/4. EMG data were bandpass 2004). The electrodes were threaded into a hypoder- recruitment during these exercises. 2. 1995). Fibres of upper over rectus abdominis (RA). 2004). participated in the study. Video motion analysis exercises used in LBP management.M. Japan).. exercises.. and a black background were used to Seven subjects (4 male. 1999).2. positioned 2 m away and perpendicular to the subject. Urquhart et al. Rizk. 2000). displacement of the upper.. 1 cm fuse with the thoracolumbar fascia and the iliac crest are diameter and 2 cm inter-electrode distance) were placed inferomedial (Urquhart et al. 1992. 1996. 1996a)... respectively (Kennedy. USA). Acuson. (ASIS)) (De Troyer et al.3.. Everett... 1982). and activity of lower and upper fibres of OI vary filtered between 50 Hz and 1 kHz and sampled at 2 kHz during posterior pelvic tilting (Carman et al. and to determine if common clinical techniques. 1980. middle and lower regions of tion of patterns of abdominal muscle recruitment. A recent review mic needle (0. 1). Urquhart et al. TrA and OI (adjacent to the anterior superior iliac spine 1977. 2001. activity of the abdominal muscles. All subjects had an ‘average’ activity level. as ment of the lower abdominal wall without movement of determined by the habitual physical activity question- the spine or pelvis (Richardson et al. Upper fascicles of TrA that attach to the rib Hodges and Richardson. 2004). and indirect measurements of TrA were approved by the institutional research ethics activity with a pressure cuff under the abdomen to committee and conducted in accordance with the indicate movement of the abdominal wall.. assist differentia. or other exercise approaches.. 1997. 1997b. 3 female). the abdominal wall and movement of the lumbar spine and pelvis in prone. Other exercise strategies have also been argued to be Recordings of EMG were made using bipolar fine- beneficial in LBP management. MathWorks. USA)... height.

Burden and Bartlett. and spine and to provide muscle stiffness required for joint control pelvic motion was measured with NIH Image (National (Hoffer and Andreassen. 1999).M. A black background was used for post-hoc analysis. maximal efforts have been considered to provide more mean- ingful values for interpretation (Andersson et al. A marker was placed on the spinous process of the L3 vertebrae and the left ASIS. The peak in neutral and the hips were flexed to 451. MD. inward movement of the lower and pelvis. USA) to measure sistent with evidence that suggests low effort is sufficient displacement of the abdominal wall.. and a maximal supports placed underneath the xiphisternum and pubic valsalva and forced expiratory manoeuvre were per- symphysis (Fig. Video data were (posterior rotation of the pelvis). Urquhart et al.. Bethesda. to perform four manoeuvres using standard measurement of linear displacement of the lumbar spine instructions (Table 1). A trigger was activated by the subject to signal when 2. each task was performed with ‘‘mild’’ were calibrated to an object of known dimensions filmed effort. The motion parameters were found to be tion and verbal and tactile feedback until they were able accurate and repeatable over a 24-h interval (ICC[2. was performed as an alternative task for EMG normal- Subjects were trained by physiotherapists. A submaximal isometric manoeuvre lumbar spine.. which is con- (WaveMetrics Inc. This allowed the edge of the anterior formed in supine for normalization of RA. Statistical analysis Fig. respectively (Hodges et al. OI.. which is equivalent to a rating of 2 on the Borg in the same plane as the abdominal wall.99) (Urquhart. ARTICLE IN PRESS 146 D. ipsilat- Subjects were positioned in prone with raised eral and contralateral trunk rotation. Three repetitions =0. 1). 1999). Lake Oswego. and the motion of the spine and pelvis in the vertical and horizontal planes was also determined for these periods. The border of the upper and middle abdominal wall (Richardson et al. ASIS EMG activity recorded during the maximal and light marker trigger source black background submaximal tasks was used to normalize the RMS (A) EMG amplitude. and in compare activity between exercise tasks and between supine with their hips flexed to 451 (B). abdominal abdominal regions (lower border of the rib cage). 1998). OE and abdominal wall to be visible. experi. 1999). An Contemporary exercise interventions focus on low level edge detection program was written using Igor Pro contractions (Richardson et al. were performed and the order of tasks was randomized.. CA). Resolution was scale (Borg. and bracing (flaring of the lateral and anterior abdominal the middle and lower abdominal regions (upper border wall) (Kennedy.1] to perform the manoeuvres correctly. To examine the association and a light source was placed inferior to the abdominal wall. Duncan’s multiple-range test was used the abdominal regions were marked. Procedure they were relaxed (baseline) and had performed the task. and borders of muscles/regions. and combined inward transferred to computer and edited using iMovie editing movement of the lower and upper abdominal wall. 2002). / Manual Therapy 10 (2005) 144–153 process of the L3 vertebrae and the left ASIS to allow muscles. Institute of Health. spine 2. Subjects were trained with instruc- 0. 1999). Subjects were positioned in prone with A two-way repeated-measures ANOVA was used to supports underneath the xiphisternum and pubic symphysis (A). subjects were positioned in supine with similar for normalization. Thus. Although reduced variance has been reported with normalization of surface EMG recordings to a submaximal task (Allison et al. Experimental set-up. In separate activity of each muscle across these tasks was selected trials. software (Apple Computer. USA). 1981. Inc.5.. 1982). 1998. The mean displacement of the upper. The spine was positioned TrA EMG. between EMG activity of the abdominal muscles and ..5 mm. Cupertino. middle and lower regions of the abdominal wall. 1980). Cholewicki and McGill. electrode insertion (B) sites 2. 1. Distances 1996).6. Maximum voluntary isometric trunk flexion. Data processing abdominal regional marker markers The root mean square (RMS) EMG amplitude was LED calculated for 2 s at baseline and for 2 s during the manoeuvre (at the time indicated by the trigger).4. ization and involved elevation of both legs so that the enced in exercise prescription for the abdominal heels were 5 cm from the supporting surface. 1965. hip and knee positions. posterior pelvic tilting of the iliac crest) were also identified.

02 3.06 0.78 S 0.05 3.07). lower/upper—inward movement of the lower and upper abdominal wall.02 1. The Fmax statistic each exercise was used to investigate differences in variance between the mean RMS EMG for each muscle normalized to a There were differences in recruitment between the maximal and submaximal task (Winer et al. Lower (supine)—inward movement of the lower abdominal wall in supine.67 S 0. 3.07 0. Gently and slowly draw in your lower abdomen below your navel without moving your upper stomach.009).3%. Gently and slowly swell out your waist without drawing your abdomen inwards or moving your back or pelvis.21 S 0. Table 2 Standard deviation data for the RMS EMG amplitude of the abdominal muscles normalized to maximal (Mx) and submaximal (SMx) isometric voluntary contractions and results of the Fmax test (F) for comparison of the variance between these normalization techniques Muscle/region Abdominal exercise Lower (supine) Pelvic tilting Bracing Lower (prone) Lower/upper Mx SMx F Mx SMx F Mx SMx F Mx SMx F Mx SMx F LTrA 0. the intramus. 2A).004 0. During inward movement of the lower abdominal Prior to analysis of the abdominal tasks. Therefore.9%..01 1. back or pelvis. In addition.001). EMG normalization lower and upper abdominal wall (P40. 1. bracing—abdominal bracing. There submaximal procedure for all muscles except RA (Table were regional differences in TrA recruitment.02 NS 0.15 S 0.02 0.03 2.29 S 0.02 0.45 S 0. In contrast.04 NS 0. Comparison of abdominal muscle recruitment for moment correlations were calculated.03 0.02 0.005 0.26 S 0.001) (Fig. Gently and slowly rock your pelvis backwards.03 3.05 0.03 0.02 NS 0. TrA EMG was 70%.20 S 0.43 S MOI 0. abdominal muscles during inward movement of the Statistical significance was set at 0. spinal and pelvic motion. Gently and slowly draw in your lower and upper abdomen without abdominal wall moving your back or pelvis.02 0. The standard deviations for the RMS EMG of lower RMS EMG amplitude of the upper region was and middle TrA were up to 180 times greater compared approximately half that of the lower and middle regions to the maximal normalization.02).06 0. Inward movement of the lower and upper Breathe in and out.02 0.05 0.005 0.02 0.21 S 0.05) and combined inward movement of the 3. no difference in OI EMG was .02 0. OI EMG was less than lower cular EMG data was normalized to EMG activity and middle TrA (Po0. abdominal bracing and pelvic tilting (Po0.8%) was also observed for one subject.02 S 0. In contrast.03 0. but similar to RA and OE recorded during the maximal manoeuvre. pelvic tilting—posterior tilting of the pelvis.03 0. ARTICLE IN PRESS D.11 S 0. S—significant (Po0.47 S OE 0.04 0.25 S 0.05.67 S 0.21 S 0.03 1. RMS EMG amplitude with normalization to the 0. Posterior pelvic tilting Breathe in and out.05 NS 0. respectively compared.001 0. the maximal wall in supine. / Manual Therapy 10 (2005) 144–153 147 Table 1 Standardized instructions used for the voluntary exercises Exercise Instructionsa Inward movement of the lower abdominal wall Breathe in and out. abdominal.01 0. There was greater variability in the mean (Po0.09 NS 0.59 S 0.05 S 0.05). 1991).M.01 0.00 S 0.06 NS 0.01).02 3. M—middle.04 0.2. lower (prone)—inward movement of the lower abdominal in prone. Pearson product. Minimal activity of OI. (P40.37 S 0. (Po0.11 S 0. Mean 2).003 0.13 S 0.00 S MTrA 0. lower abdominal wall in supine.19 S LOI 0.01 0. NS—non-significant. U—upper. a Subjects were also instructed to perform each exercise with ‘mild’ effort (a rating of 2 on the Borg scale).1. Abdominal bracing Breathe in and out.01 1.10 NS L—lower.01 S 0.68 S 0. 100% and 65% and submaximal EMG normalization methods were greater than that of OI.78 S UTrA 0.07 NS RA 0. Urquhart et al. OE and RA (1. Results with inward movement of the lower abdominal wall in prone (P40. no difference between the abdominal muscles was observed 3. OE and RA.

Mean regions of TrA. Similarities in activation of the lower and middle regions of TrA. contrast with differences in activation of the upper region of the muscle. and upper TrA. indicating reduction in 0:07. OE and RA during inward movement of the lower abdominal wall (supine (lower supine) and prone (lower prone)). ARTICLE IN PRESS 148 D.05). and although there OI had greater activity compared to RA (P ¼ 0:03) and was a trend for differences in the EMG activity of upper TrA (P ¼ 0:01). OE EMG was greater than bracing (P ¼ 0:09). Note the greater and more independent activity of TrA in supine compared to prone.M. The standard deviations are large indicating variability in abdominal muscle recruitment between subjects. 2. In contrast. There was also similar activity from baseline.05. and RA (Po0. lower OI. there was no .  Po0. Mean (SD) RMS EMG of lower and middle TrA and OI. identified between regions of the muscle (P ¼ 0:3). There When subjects tilted their pelvis posteriorly. middle was minimal activity of upper TrA. / Manual Therapy 10 (2005) 144–153 Fig. posterior tilting of the pelvis (pelvic tilting) and combined inward movement of the lower and upper abdominal wall (lower/upper). this was not significant (lower TrA: P ¼ OE RMS EMG was negative. bracing. activity of the lower and middle OI during abdominal With abdominal bracing. RMS EMG amplitude of abdominal muscles/regions during different exercise conditions (normalized to a maximal voluntary contraction). that of upper TrA. middle TrA: P ¼ 0:051). Urquhart et al.

and the lower and upper abdominal wall in the upper and lower abdominal wall was greater than prone (P40. abdominal bracing. 3).0 Mean abdominal * * 2. there was a trend towards abdominal bracing and inward movement of the lower greater TrA activity compared to the other abdominal abdominal wall in prone (Po0.05). lower and middle OI. 5).001) (Fig. / Manual Therapy 10 (2005) 144–153 149 difference between the abdominal muscles during pared to the other abdominal manoeuvres (Po0.001) tion between movement of the lumbar spine and pelvis (Fig.002. 3. RA and upper TrA was similar there was a positive correlation between OE EMG and between exercises and between the supine and prone lumbopelvic motion.05. 4. there was a low to positions. with the exception of posterior pelvic tilting. * 6. two exercises did not differ in abdominal motion (P ¼ 0:5).05).001). 2B).3. Lower and middle TrA EMG was greater (r ¼ 0:9). In contrast. However. inward movement of the lower abdominal wall and abdominal displacement with inward movement of (P40. pelvic tilting.001).0 12 1.1 0 1.0 * abdominal displacement 0. Urquhart et al. Po0. and a significant negative correlation between during inward movement of the lower abdominal wall in lumbopelvic motion and TrA EMG (as a proportion of supine than other tasks (Po0.0 (B) lower lower/upper bracing pelvic tilting 0 Fig. pelvis and spine during posterior pelvic tilting. Abdominal (B) lower lower/upper bracing pelvic tilting displacement expressed as the mean (SD) of absolute movement (A) and the mean expressed as a proportion of the total abdominal Fig. but not between the upper.0 0.4 0. 4). Although was greater during abdominal bracing than the other there was no significant correlation between displace- techniques (except pelvic tilting) (Po0. and combined inward movement of the lower and upper). Greater 7. The later muscles.001). and combined movement of the lower abdominal wall in prone (lower). ARTICLE IN PRESS D.05. Po0. 3.6 Regional proportion of middle Pelvic displacement (mm) 0. .0 * abdominal motion occurred during pelvic tilting com. middle and * lower abdominal regions (P ¼ 0:1) (Fig. spine and pelvis Abdominal wall displacement differed between tasks (Po0.0 Spinal displacement (mm) 5.0 0. Movement of the abdominal wall. Note greater movement of the exercise conditions. inward movement of the lower and upper abdominal wall (lower/ abdominal bracing.3 3. pelvic tilting.0 0. Po0. moderate correlation between movement of the abdom- inal wall and TrA EMG (r ¼ 0:4.09). Note the differences in abdominal displacement between the upper abdominal wall (lower/upper).  Po0.0 4.5 lower 4. 5). Comparison of abdominal muscle recruitment Lumbar spine and pelvic motion was minimal and did between exercises not differ between tasks. Displacement of regions of the abdominal wall.2 2. There was a high correla- differed between the exercise conditions (Po0. OE EMG total activity) was found (r ¼ 0:6) (Fig.4. Activity of ment of the lumbopelvic region and OI and RA EMG.M.0 20 * displacement (mm) 16 3. In addition. and OE occurred (Po0.0 8 0 4 (A) lower lower/upper bracing pelvic tilting 0 (A) lower lower/upper bracing pelvic tilting * * upper 5. movement (B) during inward movement of the lower abdominal wall Movement of the lumbar spine (A) and pelvis (B) during inward in prone (lower). 3.05).05) (Fig. in which greater spine and pelvic motion Recruitment of lower and middle TrA. Mean (SD) displacement of the lumbar spine and pelvis.

1999). 1997b. muscle activity has been argued to be an important feature of inward movement of the lower abdominal 4.. supine.4 The results suggest that recruitment of TrA with minimal activity of other abdominal muscles may be best achieved during inward movement of the lower 0. In this study mean EMG activity of these muscles was considerably less than that of TrA. there were common feature of the interventions. and pelvic motion (z axis). Notably. at relatively consistent with previous intramuscular EMG least with training during this task. including supine there were distinct patterns of abdominal muscle (O’Sullivan et al. abdominal displacement ment of LBP. recruitment between exercise tasks. O’Sullivan et al. Although this number is independently from the other abdominal muscles. Inward movement of the lower abdominal wall in (increase) supine 0.. minimal activity of OI.2. 1995). was greater groups have reported improvements in pain and when abdominal movement was performed without pelvic motion. the great. The results are also t (d consistent with an exercise approach for the manage- Fig. First. Although this contrasts with a previous study (Allison et al.. Urquhart et al.5 0.3 (in l di 0. further research is required to when lumbopelvic motion was limited. 1997b. Goldman et al. it is important to consider that this may There was no difference between OI. Second. Methodological issues wall. 5. A three-dimensional graph depicting the TrA to be independent of the other abdominal muscles relationship between EMG activity of all regions of TrA (as a (Richardson et al... the slightly greater activity of OI may contractions.8 4.c) and standing (Hides et al. the 1.. These results determine whether TrA activity can be changed with have important implications for selection of exercise this intervention. which involves retraining the activity of and lumbopelvic motion. 1996.2 differences may be explained by the use of surface EMG in that investigation. OE and RA limit the statistical power of the study. proportion of the total abdominal muscle activity) (y axis). activity of TrA that the same manoeuvre examined in the current study was greater relative to the other abdominal muscles was implemented. maximal abdominal displacement (x axis). These outcomes have been 4. 1996) in the early stages of rehabilitation.. but not in prone. due to the invasive nature of the study only suggests that it may be possible to activate TrA almost seven subjects were recruited.7 mi manoeuvre (Strohl et al. function with exercise interventions that involve inward movement of the lower abdomen (Hides et al. Variability in the present study was less have reached significance with a greater number of . O’Sullivan et al. ARTICLE IN PRESS 150 D. 1981. 1990). relative to the other abdominal muscles. Discussion hypothesized to result from improved motor control of TrA (and multifidus). OE and RA in one subject Firstly. techniques. investigations. De na 3. These findings agree with reports that Ab TrA is most consistently active during a ‘‘belly in’’ do 0. training in a variety of positions. Two methodological issues require consideration..0 ic mo e) abdominal muscle activity occurs during this task (Jull e) m lv s en 0. data in during inward movement of the lower abdominal wall in this study were normalized to maximal voluntary supine. positions and strategies for assessment and Activation of TrA with minimal superficial abdominal retraining of abdominal muscle function.3 Troyer et al. Although it is unlikely that the improvements were dent on body position. Richardson et al. 1987.. Finally. and that minimal superficial cre spl tion as ace 6. Second. 1996. In addition. As the results of regional differences in the recruitment of TrA.1.M.. / Manual Therapy 10 (2005) 144–153 when data were normalized to maximal manoeuvres rather than submaximal tasks. EMG Three randomised control trials of different sub- activity of TrA.. 1995.c). However. Third. this is the evident in supine. 1998).. 1997b. with the present study suggest that the ability to activate TrA greater activity of the lower and middle regions of TrA may vary between positions and it cannot be confirmed compared to the upper region. and during est and most independent activity of TrA was recorded functional activities as exercise retraining was pro- with inward movement of the lower abdominal wall in gressed (Hides et al. TrA EMG activity 0.c). with differential activity of TrA solely due to changes in TrA function. Association between EMG activity..1 Pe ecrea et al. abdominal muscle activity was depen.7 abdominal wall. Each of these studies involved This study presents several important findings.

activation of TrA is more independent if there is no pelvis or spinal motion (Richardson et al. ARTICLE IN PRESS D. This is consistent with previous Unlike supine.. 1995).4. Richardson et al. It is important to note that activity of lower OI representation has been shown to vary with the relative followed a similar pattern to that of TrA. In contrast to OI and RA. 1999). This is a novel finding. However. 1999). (Vezina and Hubley-Kozey.g. This may be due to the small size of the displacement. Urquhart et al.6. trunk rotation (Urquhart et al. 1965. Identification of greater OE activity than the other This finding is consistent with clinical observations abdominal muscles with abdominal bracing differs from (Richardson et al. task..5. However. 1999). compared to the anterolateral abdominals (Richardson OE and RA (as a proportion of total abdominal muscle et al. Vezina and Hubley-Kozey. Comparison of abdominal muscle recruitment have identified regional differences during voluntary between exercises manoeuvres. This is would not be appropriate if the aim of the exercise is to consistent with clinical hypotheses and indicates that preferentially activate TrA or OI. Although this technique is widely 4. This is consistent (drawing your navel up and in towards your spine) with previous studies that report differences in abdom.. other studies have found intramuscular EMG but did not record from TrA. or reflex. which may influence move. These 2000)...7. RA (Vezina and Hubley-Kozey. 1998). In greater RA activity compared to the anterolateral contrast. 1997a. this may have been due to insufficient differentiation of abdominal muscle activity in prone is statistical power that resulted from the small number not consistent with the use of this position for of subjects used in this invasive study. activity of TrA and OE 4. surface EMG studies have reported greater abdominals (Richardson et al. there 4. Dowd (1992) reported similar findings using posterior pelvic tilt. Differences mediated activity of the superficial muscles in response between studies may be due to variation in the level of to stretch.. 1972). / Manual Therapy 10 (2005) 144–153 151 subjects. 2004) and repeti- tive limb movements (Hodges et al. no studies 4. with greater activity during prone inward movement of the abdominal wall and pelvic tilting.. they also provide evidence that body position wall in supine. In addition. 1994).. electrode placement or EMG normalization ment during inward movement of the lower abdominal technique. However. an individual’s internal body effort. Although these tion of the superficial muscle activity. Abdominal bracing was trend towards greater movement of the lower region during inward movement of the lower abdominal wall. activity of inal muscle activity between positions (Carman et al. The absence of exercises. lumbar spine and pelvic movement referenced and the position used in this study differs in several characteristics to the clinical test (e. there was no variation in the displacement future clinical and laboratory work. and no difference between muscles (Allison activity) was found to vary linearly with the amplitude et al. In addition.. abdominal Although abdominal wall movement differed between support). previous reports which indicate greater RA activity Recruitment of TrA and the combined activity of OI. al. Posterior pelvic tilt was also a trend for TrA EMG to be related to abdominal wall movement. between regions of the abdominal wall.. may contribute to differences in abdominal muscle vity of upper and lower/middle TrA varies during recruitment. there was no difference in OI activity between the ment performance (Gurfinkel. 1995). evaluation of TrA activity in clinical practice (Richard- son et al..3.. reported to vary between these manoeuvres. 1999). 1992) has also been greater gravitational demand in prone. 1992). and greater activity of OI and/or OE relative to RA (O’Sullivan et activity of OE than RA (Vezina and Hubley-Kozey. similar activity of OI and RA has differences may be explained by cross-talk from deeper been observed during this manoeuvre (Flint and and adjacent muscles. Carman et al. varying results may have been due to differences in the There were regional differences in TrA recruit. Partridge and Walters (1960) reports of a relationship between pressure change (as reported greater activity of OI than RA and OE with measured with an air-filled cuff) associated with inward . 2000). assessment in supine may be more optimal for the tasks.. the results suggest that bracing of lumbar spine and pelvic displacement. There 4. Abdominal. Inward movement of the lower abdominal wall in differed between the tasks. respectively.M. 2000) and the ‘oblique 1972. possibly resulting in overestima. Gudgell. Although acti. This may be due to the abdominals’ (Richardson et al. Although position of body segments. However. This agrees with previous Similar to our data. there was no differentiation in reports of greater OE EMG activity during posterior abdominal muscle activity with inward movement of tilting of the pelvis compared to ‘abdominal hollowing’ the lower abdominal wall in prone. 2000).

21(22): 2640–50. Electromyographic study of the patients. or posterior tilting Dowd C.. Richardson CA. Neurophysiology 1981. these results indicate study of the abdominal muscles during postural and respiratory that abdominal muscle recruitment may be influenced manoeuvres. A motor Acknowledgments control evaluation of transversus abdominis. Scandina- vian Journal of Rehabilitation Medicine 1998. movement (Hodges et al. 1996a). Van Gansbeke D. Section F. Physiology and General Biology indicating that assessment and re-education of abdom. Soviet may be improved in supine compared to prone. it would be Transversus abdominis muscle function in humans. 1992. by patient positioning. Silver JR. Bartlett R.11(1):46–56. with limited lumbopelvic motion. first-episode low back pain. Godfrey P. Grundstrom H. Experimental Brain Research 2001. task may also be identified. Furthermore. Richardson CA. motion of the abdominal wall and lumbopelvic 29–37.8(1):51–7. Annals of the Royal College of Surgeons abdominal activity when abdominal movement occurs of England 1977. An electromyographic recruitment strategy. abdominal wall in the rehabilitation of TrA in LBP Carman DJ.14(5):377–81. National Health and Medical Research Council. For cular activity during exercise. Cresswell A. Jull GA. Journal of Neurology. Medical evidence to validate inward movement of the lower Engineering and Physics 1999. Ma Z. Hodges PW.21(23):2763–9. Evaluation of the relation- with LBP and to develop improved strategies for ship between laboratory and clinical tests of transversus abdominis restoration of motor control. Medicine and This study has implications for abdominal muscle Science in Sports and Exercise 1982. Spine 1996b. Lehr RP. Journal of Electromyography and Kinesiology 1998. Mechanical stability of the in vivo lumbar spine: implications for injury and chronic low back pain. Clinical implications 936–42. retraining in clinical practice. recorded as EMG onsets associated with arm training exercises for abdominal and hip flexor muscles.11(1):1–15. Normalisation of EMG amplitude: an evaluation and comparison of old and new methods. The American Journal of Clinical Nutrition 1982. Spine 1996. Neurosurgery and Psychiatry 1987. pelvic tilting is likely Cholewicki J. function. Thorstensson A.59(4):313–21. Frijters JER. Biggs NL. Andreassen S.M. Gudgell J. determine whether similar strategies are used by people Hodges PW. Journal of important to discourage movement of the upper abdo. Acta Physiologica Scandinavica 1992.36(5): 4. Surgical anatomy of the aponeurotic expansions of the likely to represent a greater proportion of total anterior abdominal wall. Preparatory trunk motion discussions on methodology. The results provide further Burden A. Flint MM.51(3):113–29. Journal of Spinal Disorders 1998. Physiotherapy Research International 1996a. Psychophysical bases of perceived exertion. Urquhart et al. In abdominal pressure and patterns of abdominal intra-muscular addition. Burema J. Hodges PW. Masters Thesis. The Hodges PW. bracing of the abdominal wall.36(1): instance. Clinical to produce greater activity of middle OI relative to Biomechanics 1996.8. For instance. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. However. siotherapy Association (Victorian Branch) and the Experimental Brain Research 1997. Hodges PW. men. lower OI and RA. This work was supported by the Australian Phy. To activate TrA indepen- De Troyer A.1(1): 30–40. incorrect strategies used to mimic the required activity in man. The findings may also assist in selection of anterolateral abdominal musculature utilising indwelling electro- exercises for assessment and retraining of the other des. 1994. further research is required to automatic after resolution of acute. An electromyographic study of the abdominal muscles in of the pelvis. Robinson G. Thorstensson A. upper TrA and RA. Thus. Borg GA.114(2):362–70. Hodges PW. .21(4):247–57. Observations on intra- with less activity of upper TrA. EMG signal amplitude Hoffer J. Inefficient muscular stabilization of the lumbar spine associated with low back pain. Feedforward contraction of transversus abdominis is not influenced by the direction of arm movement. Millar AB. The Research Quarterly 1965. dently from the other abdominal muscles.124(1):69–79. Jull GA.50(7):866–9. Ninane V. University of South These results also emphasize the importance of Australia. Regulation of soleus muscle stiffness in assessment during abdominal bracing and hollowing. Electromyographic study of abdominal mus- observation for assessment of muscle function. Multifidus muscle recovery is not considered.68(3):1010–6. TrA is more Askar OM. Blanton PL. Adelaide. Richardson CA. Richardson CA. Thorstensson A. ARTICLE IN PRESS 152 D. abdominal muscles. Changes in motor planning of feedforward postural References responses of the trunk muscles in low back pain. McGill SM. and function of Andersson EA.144(4): 409–18. Richardson CA. region may assist in the determination of the muscle Goldman JM.45(2):267–85. inal muscle function in a range of positions should be Hides JA. various exercise positions. Experimental Brain Research 1999. accompanies rapid upper limb movement. American Journal of Physical Medicine 1972.30(3):175–83. Differential recruitment of TrA Gurfinkel VS. and abdominal bracing recruits OE Cresswell AG.141(2):261–6. Journal of premammillary cats: intrinsic and reflex components. Delayed postural contraction of authors wish to thank Lorimer Moseley for his transversus abdominis in low back pain associated with movement assistance with data collection and Beryl Kennedy for of the lower limb. Applied Physiology 1990. Relative EMG levels in TrA.7:59–89. Baecke JAH. / Manual Therapy 10 (2005) 144–153 displacement of the abdominal wall. The mechanisms of postural regulation in man. Gorini M. Estenne M. Allison GT. Scientific Reviews.

Australian Journal of 1995. 1995. School of Physiotherapy. Richardson BA. radiologic diagnosis of spondylolysis or spondylolisthesis. p. Hubley-Kozey CL. Toppenberg R.38(2):105–12. Banzett RB. 1997b. Winer BJ. Twomey L. View publication stats .6(4):191–201. Australian Journal of Physiotherapy 1990.39(12):791–800. Allison GT. PhD Thesis. Participation of the abdominal muscles in the treatment of low back pain.M. Walters CE. A dysfunction of the deep Australia. Adelaide.81(10):1370–9. Archives of Physical Medicine eight abdominal exercises for their ability to provide stabilisation and Rehabilitation 2000. Physiotherapy 1980. Hamilton CA. Richardson CA. Loring SH. 3rd ed. Richardson CA. the abdominal muscles: Implications for control and movement of O’Sullivan PB. London: Churchill Livingstone. 2002. Hides JA. exercise intervention. Briggs CA. Sinclair J. Regional recruitment of O’Sullivan PB. Evaluation of specific transversus abdominis in trunk rotation. Australia.51(6):1471–6. 35.22(24):2959–67. The Physical Therapy Review 1960. Regional 1997a. Hodges PW. Specific stabilising in humans.27(2):114–24. for the lumbar spine. Muscle-bracing technique for utilising intra-abdominal of World Confederation of Physical Therapists. Hubley-Kozey CL. exercise in the treatment of chronic low back pain with a clinical Urquhart DM. Altered abdominal muscle the lumbar spine and pelvis. Brown DR. and clinical approach. / Manual Therapy 10 (2005) 144–153 153 Jull GA. Muscle activation in therapeutic Richardson CA. Twelfth Biennial Conference. European Spine Journal stabilizing exercise in the treatment of chronic low back pain with 2004. pressure to stabilise the lumbar spine. New York: McGraw-Hill. Michels KM. Twomey L. Washington DC. Therapeutic exercise Kennedy B. Journal of Applied Physiology 1981. Spine Urquhart. Allison GT. 139. Knox J. Twomey LT.43(2):91–8. Egan DA.11(1):102–6. Allison G. Twomey L. Australian Journal of Physiotherapy Strohl KP. Australia. Gold Coast. activation patterns associated with the pelvic tilt exercise used in Partridge MJ. recruitment in patients with chronic back pain following a specific The University of Melbourne.131(3):373–85. Miller K. An electromyographic Manipulative Therapy 1998. Jull G. p. Barker PJ. Jull GA. Ng JK-F. Australia. abdominal muscles exists in low back pain patients. Journal of Anatomy 1980. In: Proceedings of Ninth Biennial Australian Journal of Physiotherapy 1992. O’Sullivan PB. Urquhart DM. Regional variation in morphology and recruitment of 1997c. 22–5. Hodges PW. ARTICLE IN PRESS D. 932. Toppenberg R. chronic low back pain. Hodges PW. 2001. In: Proceedings Kennedy B. Regional and radiological diagnosis of lumbar segmental ‘instability’. mental design. Comerford M. Techniques for Towards the validation of a clinical test for the deep abdominal active lumbar stabilisation for spinal protection: A pilot study. The Journal of Manual and various movements of the trunk in man. Journal of Orthopaedic and Sports Physical Vezina MJ. Richardson CA. DM. Statistical principles in experi- 36:6–11. Kosch PC. In: Proceedings of Manipulative Physiotherapists Association of Proceedings of the Musculoskeletal Physiotherapy Australia Australia Tenth Biennial Conference. Mead J. muscles in back pain patients. study. 1999. Jull G. p. Allison G. Jull GA. Melbourne. Taylor JR. Hodges PW. Rizk NN. Story IH. Conference of the Manipulative Physiotherapists Association of Richardson CA. Physiotherapy 1965. A review of the muscle Therapy 1998. In: morphology of transversus abdominis and internal oblique. 1991. differences in abdominal muscle activity during various maneuvers O’Sullivan PB. Story IH. An initial evaluation of exercises to improve trunk stability. Vezina MJ. An Australian programme for management of back for spinal segmental stabilization in low back pain: Scientific basis problems. Urquhart et al.66(4):108–11. A new description of the anterior abdominal wall in man Altered patterns of abdominal muscle activation in patients with and mammals. p.