Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1225-8962(Print)
ISSN : 2287-982X(Online)
Physical Therapy Korea Vol.19 No.2 pp.80-86

Comparison of Abdominal and Lumbar Multifidus Muscle Activity During Unilateral Hip Extension in Prone Position on the Floor and on a Round Foam Roll

Su-jung Kim1, Kyu-nam Park2, Sung-min Ha3, Oh-yun Kwon4, Hyun-sook Kim5

1­-3Dept. of Physical Therapy, The Graduate School, Yonsei University,
4Dept. of Physical Therapy, College of Health Science, Yonsei University,
Dept. of Ergonomic Therapy, The Graduate School of Health and Environment, Yonsei University,
5Dept. of Physical Therapy, Yeojoo Institute of Technology
This article was received March 11, 2012, was reviewed March 11, 2012, and was accepted March 20, 2012.


The purpose of this study was to compare the muscle activity of the abdominal and lumbar multifidusduring unilateral prone hip extension on the floor and on a round foam roll. Fifteen healthy participantswere recruited. They were instructed to perform a unilateral hip extension on the floor and on a roundfoam roll in the prone position. Surface electromyography (EMG) signals were recorded from bilaterallumbar multifidus (LM), external oblique (EO), and internal oblique (IO) muscles. A paired t-test wasused to compare muscle activity, with the level of significance set at α=.05. The results showed thatbilateral LM, EO, IO EMG activity during right-hip extension on a round foam roll was greater than thaton the floor, and EMG activity of bilateral LM, right EO, and left IO during left-hip extension on around foam roll was greater than that on the floor (p<.05). These findings suggest that the unilateralhip-extension exercise on a round foam roll can be used to activate the lumbar multifidus and abdominaloblique muscles and causes a different increasing pattern between the two lifting sides.


 Lumbar segmental stability is an important factor to treat low back pain (França et al, 2010; Freeman et al, 2010; Hides et al, 2001; Kumar, 2011). Rackwitz et al (2006) evaluated the effectiveness of lumbar segmental stabilizing exercise and demon-strated that it is more effective in reducing disability and pain than is medical management or general exercise for patients with acute or chronic low back pain (Moseley, 2002; O'Sullivan, 1997; Rackwitz et al, 2006).

 Previous studies have recommended that specific exercise such as abdominal hollowing, one-arm or-leg lift in four-point kneeling, and side bridge should be considered to increase segmental stability(França et al, 2010; Hides et al, 1996; Kolber and Beekhuizen, 2007; Kumar, 2011). Danneel et al (2001) compared the effects of three types of stabilization training on the cross-sectional area of the lumbar multifidus (LM) muscle in patients with chronic low back pain. The stabilization training in the first group consisted of daily living activities that were intended to activate LM and maintain physiological lordosis of the lumbar spine (Danneel et al, 2001). The same stabilization training was combined with an intensive lumbar extensor-strengthening program in the second and third group (Danneel et al, 2001). A back extensor-strengthening program consisted of one-leg extension in four-point kneeling and trunk lift or bilateral leg lifts to the greatest possible ex-tension of hips and spine in the prone position(Danneel et al, 2001). The second group performed concentric and eccentric extension of the leg or trunk(dynamic-resistance training) (Danneel et al, 2001). The third group performed a 5-second static con-traction between concentric and eccentric extension(dynamic-static resistance training) (Danneel et al, 2001). The results demonstrated that the dynam-ic-static resistance training was the most effective exercise to increase the cross-sectional area of LM muscle (Danneel et al, 2001). Excessive and repeated lumbar extension has a risk of low back pain and degenerative disk disease (Bennett et al, 2006; Harvey and Tanner, 1991; Watkins, 2002). For ex-ample, elite-level female gymnasts who perform re-peated vigorous lumbar hyperextension often have degenerative disk disease (Bennett et al, 2006).

 An unstable surface has been used to challenge exercise difficulty and increase muscle strength. Previous studies have demonstrated that abdominal muscle activity is increased on an unstable compared with stable surface (Imai et al, 2010; Kim et al, 2011). In contrast, Drake et al (2006) have shown that multifidus muscle activity is not significantly increased by back extension, contralateral arm and leg lift, and leg raise on a Swiss ball (unstable sur-face). A round foam roll is an unstable surface. Previous research has shown that the unilateral hip flexion performed on a round foam roll results in greater abdominal muscle activation than does the same exercise performed on a stable surface in the supine position (Kim et al, 2011). Unilateral hip flex-ion on a round foam roll causes rotation, so the ab-dominal muscle contracts to maintain balance in neutral position. We thought that unilateral hip ex-tension on a round foam roll would challenge ab-dominal and back muscle activities similarly without excessive lumbar extension. However, no study has investigated whether unilateral prone hip extension exercise on an unstable foam roll can effectively ac-tivate the LM and abdominal oblique muscles. Therefore, the purpose of this study was to compare the activity of the LM, external oblique (EO), and internal oblique (IO) muscles during unilateral prone hip extension on the floor and on a round foam roll.



 Fifteen healthy volunteers were recruited for this study (10 men, 5 women). All participants were free of low back pain, previous lumbar injury or surgery, spinal deformity, or neuromuscular or joint diseases in the lumbar area and lower extremities for 6months prior to the enrollment. Ely's test, which re-quires that the participant’s hip remain stationary until 120°, was negative for all participants. If the pelvis rises from the table during active knee flexion in a prone position, this is regarded as a positive sign of rectus femoris stiffness (Peeler and Anderson, 2008). Participant who had any regular training programs involving the back and abdominal muscles within the previous 3 months were excluded. All participants had right leg dominance, which was determined by kicking a soccer ball (Hoffman et al, 1998; Jacobs et al, 2005). The mean age, height, and weight of the participants are summarized in Table 1. Prior to the study, the principal investigator ex-plained all the procedures in detail to the participants and obtained their written informed consent.

Table 1. General characteristics of subjects (N=15)


 Surface electromyography (EMG) 

EMG data of bilateral EO, IO and LM muscles were collected using a Noraxon TeleMyo 2400 sys-tem1) and analyzed using Noraxon MyoResearch 1.16 XP software. The skin was shaved and then swab-bed with alcohol-soaked cotton before electrode placement to minimize skin resistance. Surface elec-trodes were attached at an interelectrode distance of 2 ㎝. LM electrodes were placed 3 ㎝ lateral to the spinous process at L5 (Colado et al, 2011; Hibbs et al, 2011). EO and IO electrodes were placed at the midpoint between the anterior-superior iliac spine (ASIS) and the ribs and at the midpoint between the anterior superior iliac crest and the symphysis pubis and proximal to the inguinal ligament, respectively(Cram et al, 1998; Cynn, 2010). The raw EMG signal was recorded at a sampling rate of 1000 ㎐. A bandpass filter of 20∼450 ㎐ was used to eliminate movement artifacts. The EMG signal was processed to the root mean square (RMS) using a window of 50 milliseconds. For normalization, the RMS of a 5-second maximal voluntary isometric contraction(MVIC) was measured three times for each muscle, as recommended by Dankaerts et al (2004). The average RMS of three measurements was used to determine the MVIC of each muscle.

1) Noraxon TeleMyo 2400T, Noraxon Inc., Scottsdale, AZ, U.S.A.


 Each participant was instructed to lie prone on ei-ther the floor or a round foam roll2) (15.2×91.4 ㎝). The two supporting surfaces were randomized by balloting number 1 for the floor and number 2 for the round foam roll. To avoid contact between the electrodes and the floor, two tables were arranged with a gap between them that ran from just above the umbilicus to below ASIS under the floor condition. A target bar was placed so that the par-ticipant’s thigh touched it at 10° extension of the hip joint with full extension of the knee joint (Figure 1 and 2). Participants attempted to keep their spines neutral on both supported surfaces during unilateral hip extension, and they were instructed to sustain the isometric contraction for 5 seconds. EMG data were collected when the participant maintained the test position without loss of balance and without lumbar rotation. Each participant completed three tri-als on the floor and three on the round foam roll. The average of each set of three trials was used for data analysis.

 2) Foam Therapy Rolls, Sammons Preston Rolyan, Bolingbrook, IL, U.S.A.

Figure 1. Right-hip extension on the floor.

Figure 2. Right-hip extension on a round foam roll.

Statistical Analysis

 A paired t-test was used to compare muscle ac-tivity during hip extension performed on the floor and on the round foam roll, with the level of sig-nificance set at α=.05. Statistical analysis was per-formed using PASW Statistics version 18.0 software.


 Bilateral LM, EO and IO muscle activity was greater on the round foam roll compared with that on the floor during left-hip extension (Table 2)(p<.05). During right-hip extension, muscle activity of the right EO, left IO and bilateral LM muscle was greater on the round foam roll compared with that on the floor (Table 3)(p<.05).

Table 2. Comparison of muscle activity during left hip extension (%MVIC)

Table 3. Comparison of muscle activity during right hip extension (%MVIC)


 We compared the muscle activity during unilateral hip extension on the floor and on a round foam roll. Previous research has revealed that the instability, relatively small contact area, and reduced somato-sensory input of a round foam roll require greater muscle activity to maintain lumbar stability (Kim et al, 2011). The results of the current study showed that the muscle activity on a round foam roll was greater than on the floor.

 EO and IO activity depended on lifting-leg side. During right-hip extension, ipsilateral EO and con-tralateral IO activity was significantly increased. This result is consistent with previous studies, which found that ipsilateral EO and contralateral IO activity were greater during right-leg extension in the four-point kneeling (Stevens et al, 2007) and su-pine positions (Kim et al, 2011). Janda has indicated that trunk muscle slings are necessary for facilitating reciprocal gait patterns between the upper and lower extremity as well as for rotational trunk stabilization(Page et al, 2009). The spiral sling is one of the an-terior trunk muscle slings, and the opposite sides of the EO and IO create a spiral sling and maintain trunk stability (Page et al, 2009). Therefore, the right(ipsilateral) EO and left (contralateral) IO increased during right-hip extension to counter rotational movement on the round foam roll.

 During left-hip extension, bilateral EO and IO ac-tivity was increased on the round foam roll com-pared with the floor. These results might be related to leg dominance. Sung and Kim (2011) reported that the spinal range of motion is significantly dif-ferent depending on the dominant hip motion. They suggest that the decreased axial trunk range of mo-tion with the dominant hip is related to stiffened passive structures of the hip joint. All participants in our study had right-leg dominance, and they were instructed to maintain a neutral spine position during hip movement, so more trunk rotation might be expected during left-hip extension. Additionally, the round foam roll was unstable and created rota-tion when participants lifted the leg (Kim et al, 2011). The trunk muscles on the round foam roll were challenged more during left-hip than right-hip extension. Thus, bilateral muscles were co-con-tracted to maintain trunk stability. This was con-sistent with previous findings demonstrating that the activity of all the abdominal muscles was greater on a round foam roll than on the floor during non-dominant leg lifting. However, to establish the relationship between leg dominance and trunk mus-cle activity during hip extension, further study is needed on participants who have left-leg dominance because all participants had right-leg dominance in our study.

 Participants performed unilateral hip extension without arm support in the prone position on the floor and on a round foam roll in our study. Compare with similar exercises in previous studies, LM muscle activity was approximately 20% MVIC during unilateral hip extension in four-point kneeling(Drake et al, 2006; Stevens et al, 2007) and approx-imately 60% MVIC during static lumbar extension with the trunk parallel to the floor and the pelvis supported by fixing the feet on the table (Colado et al, 2011). Drake et al (2006) suggested that the use of an exercise ball does not increase the challenge imposed on the musculoskeletal system of healthy young participants because abdominal and back mus-cle activity is unchanged or decreased on the ball during unilateral hip extension in four-point kneeling and static lumbar extension compared with the same maneuver on a mat (floor). Our results demonstrated that LM activity was approximately 27% MVIC on the floor and 43% MVIC on the round foam roll(Table 2 and 3). The round foam roll created a greater challenge for the LM muscle. Although LM activity on the round foam roll was less than that with static lumbar extension on the floor, the ex-ercise on a round foam roll had the advantage of preventing excessive lumbar extension during exercise. This result could be used to prescribe a gradual exercise protocol clinically.

 This study had some limitations. First, we used surface electrodes to collect LM activity data. Stokes et al (2003) have recommended that accurate meas-urement of multifidus muscle activity requires intra-muscular electrodes because surface electrodes placed over multifidus muscles are more sensitive to the adjacent longissimus muscles. We aimed to mini-mized crosstalk of adjacent muscles by collecting LM activity data at the L5 level. Second, our study in-volved healthy young participants, so the results cannot be generalized to other populations. Therefore, further studies should investigate symptomatic sub-jects and the general population.


 Unilateral hip extension on a round foam roll caused greater muscle recruitment than the same ex-ercise on the floor. A difference in muscle activity was noted between the two side during lifting, LM, EO and IO muscle activity increased bilaterally dur-ing left-hip extension on a round foam roll, and bi-lateral LM, ipsilateral EO, and contralateral IO mus-cle activity increased during right-hip extension. These findings suggest that the unilateral hip ex-tension exercise on a round foam roll can be used to activate the LM and abdominal oblique muscles.


1.Bennett DL, Nassar L, DeLano MC. Lumbar spine MRI in the elite-level female gymnast with low back pain. Skeletal Radiol. 2006;35(7):503-509.
2.Colado JC, Pablos C, Chulvi-Medrano I, et al. The progression of paraspinal muscle recruitment in-tensity in localized and global strength training exercises is not based on instability alone. Arch Phys Med Rehabil. 2011;92(11):1875-1883.
3.Cram JR, Kasman GS, Holtz J. Introduction to Surface Electromyography. Gaithersburg, MD, Aspen Publishers Inc, 1998:345.
4.Cynn HS. Effects of lumbar stabilization on the trunk and lower limb muscle activity and veloc-ity of the center of pressure during single leg standing. Phys Ther Kor. 2010;17(4):1-7.
5.Dankaerts W, O'Sullivan PB, Burnett AF, et al. Reliability of EMG measurements for trunk muscles during maximal and submaximal volun-tary isometric contractions in healthy controls and CLBP patients. J Electromyogr Kinesiol. 2004;14(3):333-342.
6.Danneels LA, Vanderstraeten GG, Cambier DC, et al. Effects of three different training modalities on the cross sectional area of the lumbar multifidus muscle in patients with chronic low back pain. Br J Sports Med. 2001;35(3):186-191.
7.Drake JD, Fischer SL, Brown SH, et al. Do exercise balls provide a training advantage for trunk ex-tensor exercises? A biomechanical evaluation. J Manipulative Physiol Ther. 2006;29(5):354-362.
8.França FR, Burke TN, Hanada ES, et al. Segmental stabilization and muscular strengthening in chronic low back pain: A comparative study. Clinics (Sao Paulo). 2010;65(10):1013-1017.
9.Freeman MD, Woodham MA, Woodham AW. The role of the lumbar multifidus in chronic low back pain: A review. PM R. 2010;2(2):142-146.
10.Harvey J, Tanner S. Low back pain in young athletes. A practical approach. Sports Med. 1991;12(6):394-406.
11.Hibbs AE, Thompson KG, French DN, et al. Peak and average rectified EMG measures: Which method of data reduction should be used for as-sessing core training exercises? J Electromyogr Kinesiol. 2011;21(1):102-111.
12.Hides JA, Jull GA, Richardson CA. Long-term ef-fects of specific stabilizing exercises for first-episode low back pain. Spine (Phila Pa 1976). 2001;26(11):E243-E258.
13.Hides JA, Richardson CA, Jull GA. Multifidus muscle recovery is not automatic after resolution of acute, first-episode low back pain. Spine (Phila Pa 1976). 1996;21(23):2763-2769.
14.Hoffman M, Schrader J, Applegate T, et al. Unilateral postural control of the functionally dominant and nondominant extremities of healthy subjects. J Athl Train. 1998;33(4):319-322.
15.Imai A, Kaneoka K, Okubo Y, et al. Trunk muscle activity during lumbar stabilization exercises on both a stable and unstable surface. J Orthop Sports Phys Ther. 2010;40(6):369-375.
16.Jacobs C, Uhl TL, Seeley M, et al. Strength and fa-tigability of the dominant and nondominant hip abductors. J Athl Train. 2005;40(3):203-206.
17.Kim SJ, Kwon OY, Yi CH, et al. Comparison of ab-dominal muscle activity during a single-legged hold in the hook-lying position on the floor and on a round foam roll. J Athl Train. 2011;46(4): 403-408.
18.Kolber MJ, Beekhuizen KS. Lumbar stabilization: An evidence-based approach for the athlete with low back pain. Strength Cond J. 2007;29(2):26-37.
19.Kumar SP. Efficacy of segmental stabilization ex-ercise for lumbar segmental instability in pa-tients with mechanical low back pain: A randomized placebo controlled crossover study. N Am J Med Sci. 2011;3(10):456-461.
20.Moseley L. Combined physiotherapy and education is efficacious for chronic low back pain. Aust J Physiother. 2002;48(4):297-302.
21.O'Sullivan PB, Phyty GD, Twomey LT, et al. Evaluation of specific stabilizing exercise in the treatment of chronic low back pain with radio-logic diagnosis of spondylolysis or spondylolisthesis. Spine (Phila Pa 1976). 1997;22(24):2959-2967.
22.Page P, Frank C, Lardner R. Assessment and Treatment of Muscle Imbalance: The Janda approach. Champaign, IL., Human Kinetics, 2009:33-36.
23.Peeler J, Anderson JE. Reliability of the Ely's test for assessing rectus femoris muscle flexibility and joint range of motion. J Orthop Res. 2008;26(6):793-799.
24.Rackwitz B, de Bie R, Limm H, et al. Segmental stabilizing exercises and low back pain. What is the evidence? A systematic review of random-ized controlled trials. Clin Rehabil. 2006;20(7): 553-567.
25.Stevens VK, Vleeming A, Bouche KG, et al. Electromyographic activity of trunk and hip muscles during stabilization exercises in four-point kneeling in healthy volunteers. Eur Spine J. 2007;16(5):711-718.
26.Stokes IA, Henry SM, Single RM. Surface EMG electrodes do not accurately record from lumbar multifidus muscles. Clin Biomech 2003;18(1):9-13.
27.Sung PS, Kim YH. Kinematic analysis of symmetric axial trunk rotation on dominant hip. J Rehabil Res Dev. 2011;48(8):1029-1036.
28.Watkins RG. Lumbar disc injury in the athlete. Clin Sports Med. 2002;21(1):147-165.