Journal of Electromyography and Kinesiology 23 (2013) 462–468

Contents lists available at SciVerse ScienceDirect

Journal of Electromyography and Kinesiology

journal homepage: www.elsevier .com/locate / je lek in

Effect of isometric horizontal abduction on pectoralis major and serratus anteriorEMG activity during three exercises in subjects with scapular winging

Kyung-Mi Park 1, Heon-Seock Cynn ⇑, Chung-Hwi Yi 2, Oh-Yun Kwon 3

Department of Physical Therapy, The Graduate School, Yonsei University, 1 Yonseidae-gil, Wonju, Gangwon-do, South Korea

a r t i c l e i n f o a b s t r a c t

Article history:Received 2 July 2012Received in revised form 17 November 2012Accepted 23 November 2012

Keywords:Scapular wingingPectoralis majorSerratus anterior

1050-6411/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.jelekin.2012.11.013

⇑ Corresponding author. Tel.: +82 33 760 2427; faxE-mail addresses: kyungmi87@hanmail.net (K.-M

(H.-S. Cynn), pteagle@yonsei.ac.kr (C.-H. Yi), kwonoy@1 Tel.: +82 33 760 2497; fax: +82 33 760 2496.2 Tel.: +82 33 760 2429; fax: +82 33 760 2496.3 Tel.: +82 33 760 2721; fax: +82 33 760 2496.

The aim of this study was to determine the effect of isometric horizontal abduction using Thera-Bandduring three exercises (forward flexion, scaption, and wall push-up plus) in subjects with scapular wing-ing by investigating the electromyographic (EMG) amplitude of the pectoralis major, serratus anteriorand the pectoralis major/serratus anterior activity ratio. Twenty-four males with scapular winging par-ticipated in this study. The subjects performed the forward flexion, scaption, and wall push-up plus withand without isometric horizontal abduction using Thera-Band. Surface EMG was used to collect the EMGdata of the pectoralis major and serratus anterior during the three exercises. Two-way repeated analysesof variance with two within-subject factors (isometric horizontal abduction condition and exercise type)were used to determine the statistical significance of pectoralis major and serratus anterior EMG activityand the pectoralis major/serratus anterior EMG activity ratio. Pectoralis major EMG activity was signifi-cantly lower during forward flexion and wall push-up plus with isometric horizontal abduction, and ser-ratus anterior EMG activity was significantly greater with isometric horizontal abduction. Additionally,the pectoralis major/serratus anterior activity ratio was significantly lower during the forward flexionand wall push-up plus with isometric horizontal abduction. The results of this study suggest that isomet-ric horizontal abduction using Thera-Band can be used as an effective method to facilitate the serratusanterior activity and to reduce excessive pectoralis major activity during exercises for activating serratusanterior.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

The serratus anterior is the primary muscle used to stabilize themedial border and inferior angle of the scapula and to preventwinging and anterior tilting of the scapula (Escamilla et al., 2009;Neumann, 2002). Accordingly, weakness, fatigue, and abnormalmuscle firing patterns of the serratus anterior can cause shoulderdysfunctions that deviate from the normal scapulohumeral rhythm(Madeleine et al., 2008), as well as glenohumeral and scapulotho-racic muscle imbalance (Cools et al., 2004; Hallström andKärrholm, 2006). Especially, scapular winging can result fromweakness of the serratus anterior itself without nerve involvement,even though the most common cause is injury of the long thoracicnerves that innervate the serratus anterior (Dvir and Berme, 1978;Martin and Fish, 2008). Therefore, the serratus anterior is a focus of

ll rights reserved.

: +82 33 760 2496.. Park), cynn@yonsei.ac.kr

yonsei.ac.kr (O.-Y. Kwon).

therapeutic exercise protocols for the prevention and rehabilita-tion of scapular winging.

Many previous studies have investigated diverse exercises todetermine the most effective exercise to elicit serratus anterioractivity including open and closed kinetic chain exercises (Coolset al., 2007; Decker et al., 1999; Hardwick et al., 2006; Ludewiget al., 2004; Martins et al., 2008; McClure et al., 2004). Of these di-verse exercises, forward flexion has been routinely included inshoulder rehabilitation programs (Ellenbecker, 2006; McCannet al., 1993) and is reported to induce serratus anterior activitygreater than 40% of maximal isometric voluntary contraction(Wattanaprakornkul et al., 2011). Furthermore, the scaption andpush-up plus were found to consistently elicit serratus anterioractivity greater than 20% of maximal isometric voluntary contrac-tion (Decker et al., 1999). Thus, these exercises have been recom-mended for increasing serratus anterior activation. Decker et al.(1999) also reported that the plus phase of push-up plus inducesthe highest average serratus anterior activation as compared to anumber of other exercises used in the study. Especially, the wallpush-up plus is commonly advocated because this modifiedpush-up plus is believed to be a less demanding exercise for clini-cal use, and it is generally recommended to patients who cannot

K.-M. Park et al. / Journal of Electromyography and Kinesiology 23 (2013) 462–468 463

perform the standard push-up plus (Ludewig et al., 2004). For thisreason, we decided to include forward flexion, scaption (Coolset al., 2007), and the wall push-up plus (Ludewig et al., 2004) tobe examined in this study.

Previous investigators reported muscle activity of the trapeziusand serratus anterior during a variety of shoulder exercises de-signed to explore not only the actual activity of the muscle but alsothe relative muscle activity ratio among synergist muscles (Coolset al., 2007; Ludewig et al., 2004; Martins et al., 2008). Specifically,a high ratio of the upper trapezius to serratus anterior wasproposed as a contributing factor in abnormal scapular motion(Ludewig and Cook, 2000; Sharmann, 2001). Also, other previousinvestigators have reported that the pectoralis major can be acti-vated simultaneously with the serratus anterior during activescapular exercises (Decker et al., 1999; Hiengkaew et al., 2003;Hintermeister et al., 1998; Kim et al., 2010). Additionally, greateractivation of the pectoralis major compared with the serratus ante-rior can be observed in subjects with scapular winging during theplus phase of the push-up plus (Kim et al., 2010). Accordingly, it ispossible that the pectoralis major may compensate for a weak ser-ratus anterior because the pectoralis major may act as a synergist(Decker et al., 1999; Hiengkaew et al., 2003; Hintermeister et al.,1998). However, the excessive pectoralis major activation may leadto glenohumeral and scapulothoracic pathologies such as shoulderanterior joint translations or decreased compression forces on theglenoid (Konrad et al., 2006; Labriola et al., 2005; McMahon et al.,2003).

Reciprocal inhibition can reduce unwanted muscle activationby contracting the antagonist (Kisner and Colby, 2002). With re-gard to pectoralis major, it can be inhibited when the shoulder isabducted and rotated externally. Thus, isometric horizontal abduc-tion of the shoulder can be used to reduce activity of the pectoralismajor during exercises that activate the serratus anterior. How-ever, no previous studies have investigated a strategy to increaseactivation of the serratus anterior and decrease activation of thepectoralis major for subjects with scapular winging. Therefore,the purpose of this study was to determine the effect of isometrichorizontal abduction using Thera-Band during three exercises (for-ward flexion, scaption, and wall push-up plus) in subjects withscapular winging by investigating the EMG amplitude in pectoralismajor and serratus anterior and the pectoralis major/serratus ante-rior activity ratio. We hypothesized that isometric horizontalabduction using Thera-Band during three exercises will changethe pectoralis major/serratus anterior muscle activity ratio byreducing pectoralis major EMG activity and/or increasing serratusanterior EMG activity.

2. Methods

2.1. Subjects

Twenty-four males with scapular winging participated in thisstudy (age = 24.5 ± 2.4 years, heigh = 174.6 ± 5.5 cm, weight =70.8 ± 9.3 kg, amount of scapular winging = 22.1 ± 1.9 mm). Scapu-lar winging was confirmed by measuring the distance between thethoracic wall and the inferior angle of the scapula using a scapulo-meter. The distance equal to or greater than 2 cm was defined asscapular winging (Weon et al., 2011). Weon et al. (2011) demon-strated the test–retest reliability of the scapulometer for measur-ing scapular winging. The interclass correlation coefficient was0.97 (95% confidence interval: 0.87–0.99), and the standard errorof the measurement was 0.1 cm.

The exclusion criteria for study participation were a history orclinical exam revealing pain or dysfunction that substantiallylimited shoulder motion or led to gross instability of the shoulder

during daily activities, signs and symptoms of cervical pain, adhe-sive capsulitis, thoracic outlet syndrome, or a current complaint ofnumbness or tingling in the upper extremity. Additional exclusioncriteria included a history of shoulder injury or surgery, participa-tion in overhead sports at a competitive level, and upper limbstrength training for more than 5 h per week. Prior to participation,subjects read and signed an informed consent form. The investiga-tion was approved by Yonsei University Wonju Campus HumanStudies Committee.

2.2. EMG recording and data processing

Surface EMG data were collected using a Noraxon TeleMyo2400 system (Noraxon, Inc., Scottsdale, AZ, USA) and analyzedusing Noraxon MyoResearch 1.06 XP software. The EMG signalswere amplified, band-pass filtered (10 and 450 Hz), and notch fil-tered (60 Hz, 120 Hz) before being digitally recorded at 1000 Hzand processed into the root mean square. Data were collected fromthe pectoralis major and serratus anterior on the scapular wingingside. After shaving and scrubbing the skin with alcohol, disposableAg/AgCl surface electrodes were placed on each muscle at the stan-dardized sites (Criswell, 2010). The electrodes of the pectoralis ma-jor (sternal fiber) were located on the chest wall horizontal fromthe arising muscle mass (approximately 2 cm out from the axillaryfold). The serratus anterior electrodes were attached just below theaxillary area, at the level of the inferior tip of the scapula, and med-ial to the latissimus dorsi. Two electrodes were placed approxi-mately 20 mm apart in the direction of the muscle fibers, and aground electrode was placed over the ipsilateral clavicle.

Reference voluntary contractions (RVCs) were collected to nor-malize the EMG data of the pectoralis major and the serratus ante-rior, because the normalization using a maximal voluntaryisometric contraction could increase the risk of injury or pain inthe shoulder joint while applying the maximal manual resistanceto the humerus. To collect RVC data, while wearing a 6.5 kg sand-bag on their forearm, subjects flexed their shoulder at 90� in thescapular plane (Park and Yoo, 2011). Subjects maintained this pos-ture for 5 s without moving their center of mass. The middle 3-s ofthe 5-s contraction was used for data analysis, and a 1-min restwas given between trials. The mean value of three trials for eachmuscle activity was taken as the RVC. The EMG signals collectedduring the three exercises were expressed as a percentage of thecalculated root mean square of RVC (%RVC). To calculate the pecto-ralis major/serratus anterior ratio, the normalized pectoralis majoramplitude was divided by the normalized serratus anterioramplitude.

2.3. Procedures

Subjects were uniformly instructed by a primary investigator onstandardized position of three exercises and how to perform eachexercise. Subjects were then allowed to familiarize for approxi-mately 20 min until the proper motion and timing were achieved.For data collection, subjects performed three trials of each exercisewith 1-min rest period between trials. The order of the recording ofthree exercises was randomized using the random number gener-ator in Microsoft Excel (Microsoft Corp., Redmond, WA, USA). Trialsduring which the subject failed to maintain the standardized posi-tion and holding period were excluded from the data analysis(three trials were discarded because 80% of peak pressure wasnot sustained during push-up plus). The mean value of three trialsof each exercise was used for data analysis. Each subject first per-formed three exercises without isometric horizontal abduction andthen completed with isometric horizontal abduction. A 10-min restperiod was given between the two conditions to avoid fatigue.

464 K.-M. Park et al. / Journal of Electromyography and Kinesiology 23 (2013) 462–468

The three exercises with isometric horizontal abduction weresimilar to the exercises without isometric horizontal abduction ex-cept that Thera-Band was applied around both wrists providingresistance to shoulder horizontal abduction. The tensile load ofthe Thera-Band was determined when the subject was able to per-form more than 10 repetitions while maintaining consistent met-ronome speed (Decker et al., 1999).

2.3.1. Forward flexionSubjects stood with their backs flat against a wall, knees slightly

bent and feet positioned shoulder-width apart (Fig. 1). Two target

Fig. 1. Ending position of three exercises with and without IHA (A: forward flexion, B: scabduction).

bars (a base, a vertical pole, and a height-adjustable horizontal bar)were positioned at the side of the subjects bilaterally to control theplane of flexion and amount of flexion. A universal goniometer wasused to measure the level of 130� shoulder flexion in each subjectto determine the height of the horizontal bars. Subjects were in-structed to lift both arms with their elbows extended and the fore-arm and wrist in a neutral position by brushing the vertical polewith the dorsal side of the forearm to maintain the shoulders inthe sagittal plane. Subjects then maintained forward flexion pos-ture for 5 s when the radial border of the wrist slightly touchedthe horizontal bar which was placed at 130� shoulder flexion.

aption, C: wall push-up plus, 1: without IHA, 2: with IHA; IHA: isometric horizontal

K.-M. Park et al. / Journal of Electromyography and Kinesiology 23 (2013) 462–468 465

2.3.2. ScaptionSubjects stood with their backs flat against a wall, knees slightly

bent and feet positioned shoulder-width apart (Fig. 1). Two targetbars (a base, a vertical pole, and a height-adjustable horizontal bar)were positioned at the plane of the scapula (30� anterior to thefrontal plane) of the subjects bilaterally to control the plane ofthe scapula and 130� shoulder flexion. A universal goniometerwas used to measure the level of 130� shoulder flexion at the planeof the scapula in each subject to determine the height of the hori-zontal bars. Subjects were then instructed to lift both arms withthe elbows extended and the forearm and wrist in a neutral posi-tion by brushing the vertical pole with the dorsal side of their fore-arm to maintain the shoulders in the scapular plane. Subjects thenmaintained this scaption posture for 5 s when the radial border ofthe wrist slightly touched the horizontal bar at 130� shoulderflexion.

2.3.3. Wall push-up plusSubjects stood facing a wall at arm’s length away with feet posi-

tioned parallel and shoulder-width apart on the floor. Subjectswere instructed to maintain shoulders flexed at 90�, to exend theelbows and to make a fist with neutral wrist in both hands(Fig. 1). Target bars (a base, a vertical pole, and a height-adjustablehorizontal bar) were positioned in front of the subjects. The heightof the horizontal bar was set at the level of the manubrium of thesternum for each subject. The end tip of the horizontal bar touchedthe manubrium of the sternum and prevented forward movementof the trunk during the exercise. Before the wall push-up plus, thepeak pressure was measured by pressure biofeedback units placedat the level of 90� shoulder flexion on the wall to control theamount of pushing force subjects exerted toward the wall duringthe exercise. Then, subjects were asked to push the center of pres-sure biofeedback units with their knuckles to control the protrac-tion of bilateral scapulae without forward movement of thetrunk, sustaining 80% of their maximal pressure. Trials duringwhich the subject failed to maintain 80% of their maximal pressurewere excluded in the data analysis. Each subject maintained thisposture for 5 s.

2.4. Statistical analysis

Two-way repeated analyses of variance (ANOVA) with twowithin-subject factors (isometric horizontal abduction condition:with and without isometric horizontal abduction; exercise type:forward flexion, scaption, and wall push-up plus) were used to as-sess the statistical significance of pectoralis major and serratusanterior EMG activity and the pectoralis major/serratus anteriorEMG activity ratio. The level of significance was set at a = 0.05. Ifa significant isometric horizontal abduction condition � exercisetype interaction was not revealed from the two-way repeatedANOVA, the main effects of isometric horizontal abduction condi-tion and exercise type were determined. If a significant main effectof exercise type was observed, the Bonferroni correction was used(a = 0.05/3 = 0.017). If a significant isometric horizontal abductioncondition � exercise type interaction was found, a pair-wise com-parison with Bonferroni correction was used to determine the sim-ple effect (a = 0.05/6 = 0.008). All statistical analyses wereperformed using PASW Statistics 18 (SPSS, Chicago, IL, USA).

Fig. 2. Comparison of PM EMG activity during the serratus anterior activationexercises with and without IHA (PM: pectoralis major, IHA: isometric horizontalabduction). ��p < 0.008, significant simple effect.

3. Results

3.1. Pectoralis major

The pectoralis major activity during three exercises with andwithout isometric horizontal abduction is shown in Fig. 2. A signif-

icant condition � exercise type interaction was observed(F2.22 = 11.444, p < .05). The pectoralis major activity was signifi-cantly lower in the forward flexion and wall push-up plus exerciseswith isometric horizontal abduction compared to without isomet-ric horizontal abduction (forward flexion: 30.9 ± 8.4 %RVC with iso-metric horizontal abduction, 35.4 ± 9.9 %RVC without isometrichorizontal abduction; scaption: 27.2 ± 9.1 %RVC with isometrichorizontal abduction, 29.1 ± 8.6 %RVC without isometric horizontalabduction; wall push-up plus: 52.1 ± 14.5 %RVC with isometrichorizontal abduction, 97.3 ± 18.4 %RVC without isometric horizon-tal abduction). The pectoralis major activity was significantlygreater in the wall push-up plus compared with forward flexionand scaption regardless of the presence or absence of isometrichorizontal abduction.

3.2. Serratus anterior

The serratus anterior EMG activity during three exercises withand without isometric horizontal abduction is shown in Fig. 3.No significant isometric horizontal abduction condition � exercisetype interaction was found (F2.22 = 1.780, p = .192). A significantmain effect of isometric horizontal abduction was observed(F1.23 = 25.843, p < .05), but no significant main effect of exercisetype was found (F2.22 = 1.653, p = .203). The serratus anterioractivity was significantly greater with than without isometric hor-izontal abduction (55.1 ± 6.2 %RVC with isometric horizontalabduction and 47.4 ± 5.4 %RVC without isometric horizontalabduction).

3.3. Pectoralis major/serratus anterior activity ratio

The pectoralis major/serratus anterior ratios during the threeexercises with and without isometric horizontal abduction areshown in Fig. 4. A significant isometric horizontal abduction condi-tion � exercise type interaction was observed (F2.22 = 10.560,p < .05), with the pectoralis major/serratus anterior ratio signifi-cantly greater in the forward flexion and wall push-up plus withisometric horizontal abduction compared to without isometrichorizontal abduction. The pectoralis major/serratus anterior ratiowas significantly greater in the wall push-up plus than in the for-ward flexion and scaption regardless of the presence or absence ofisometric horizontal abduction (forward flexion: 0.7 ± 0.3 with iso-metric horizontal abduction, 1.3 ± 0.5 without isometric horizontalabduction; scaption: 0.7 ± 0.3 with isometric horizontal abduction,0.84 ± 0.4 without isometric horizontal abduction; wall push-up

Fig. 3. Comparison of SA EMG activity during the SA activation exercises with andwithout IHA (SA: serratus anterior, IHA: isometric horizontal abduction). �p < 0.05,significant difference in main effect for IHA condition.

Fig. 4. Comparison of the PM/SA EMG activity ratio during the SA activationexercises with and without IHA (PM: pectoralis major, SA: serratus anterior, IHA:isometric horizontal abduction). ��p < 0.008. significant simple effect.

466 K.-M. Park et al. / Journal of Electromyography and Kinesiology 23 (2013) 462–468

plus: 1.1 ± 0.5 with isometric horizontal abduction, 2.3 ± 0.9 with-out isometric horizontal abduction).

4. Discussion

The purpose of this study was to investigate whether pectoralismajor and serratus anterior EMG activity and the pectoralis major/serratus anterior activity ratio could be changed with isometrichorizontal abduction during the three exercises in subjects withscapular winging. To our knowledge, this is the first study to inves-tigate this strategy to increase serratus anterior activity and de-crease activation of the pectoralis major in subjects with scapularwinging during exercises.

Our results demonstrated significantly lower pectoralis majoractivity during the forward flexion and wall push-up plus with iso-metric horizontal abduction compared to without isometric hori-zontal abduction. This finding can be explained by the reciprocalinhibition mechanism. The mechanism of reciprocal inhibition isdue to afferent impulses from agonist muscle spindles stimulatingan inhibitory interneuron in the spinal cord, which causes inhibi-tion of the alpha motor neuron to the antagonist muscle (Leonard,1997; Stuart, 2007). Thus, isometric horizontal abduction mayhave induced contraction of the agonist muscle (horizontal abduc-tors) by providing resistance while simultaneously relaxing andinhibiting the antagonist muscle (pectoralis major).

In this study, the isometric horizontal abduction had no signif-icant effect on pectoralis major activity during the scaption. This

finding can also be explained by the reciprocal inhibition mecha-nism (Leonard, 1997; Stuart, 2007). During scaption, the arms werepositioned in the scapular plane (i.e., 30� anterior to the frontalplane) activating the shoulder horizontal abductor including pos-terior deltoid, supraspinatus, infraspinatus, and teres minor, andthus inhibiting pectoralis major activation. Considering the compa-rable pectoralis major EMG amplitudes between scaption withoutisometric horizontal abduction and forward flexion with isometrichorizontal abduction, positioning the arms in the scapular planemay have inherently reduced excessive pectoralis major activityduring the exercise.

Our data indicated that pectoralis major activity during the wallpush-up plus was significantly greater than that during forwardflexion and scaption. This finding may be attributed to the differingnature of weight bearing on the upper extremity. Forward flexionand scaption are open kinematic chain motions, whereas the wallpush-up plus is classified as a closed kinematic chain exercise, asthe distal segment was fixed (Ludewig et al., 2004). Uhl et al.(2003) also reported that greater demand during weight bearingexercises increases pectoralis major activation. Thus, pectoralismajor activity is increased during the wall push-up plus comparedwith that during forward flexion and scaption.

Our study found that serratus anterior activity was significantlyincreased during three exercises with isometric horizontal abduc-tion compared to without isometric horizontal abduction. The in-crease in serratus anterior EMG activity during the exercises maybe associated with reduced activation of the pectoralis major. Pre-vious studies have suggested that synergistic muscles that worktogether can influence each other through movement (Chance-Lar-sen et al., 2010; Page et al., 2009), and that surface EMG contentswere changed in the presence of altered afferent information tomaintain force (Farina et al., 2008; Madeleine et al., 2006). Specif-ically, when the movement is performed in the same range of mo-tion, decreased activity of one muscle can give rise to increasedactivity of another muscle to achieve the same range of motion(Jonkers et al., 2003; Oh et al., 2007). In the present study, becausepectoralis major activity was reduced during the exercises by add-ing isometric horizontal abduction, it might be assumed that rela-tively greater serratus anterior activation was required to performthe same exercises.

The results of this study were supportive of the hypothesis thatthe use of isometric horizontal abduction would decrease the pec-toralis major/serratus anterior EMG activity ratio during exercises.A lower pectoralis major/serratus anterior ratio signified that theserratus anterior activation increased and the pectoralis majoractivation decreased. All of three exercises that included isometrichorizontal abduction demonstrated pectoralis major/serratus ante-rior ratios below 1, except the wall push-up plus. If higher pecto-ralis major/serratus anterior ratios are of concern when choosinga particular exercise program for increasing serratus anterior acti-vation, instructing the individuals to perform the exercises withisometric horizontal abduction might be necessary. For patientswith scapular winging, an exercise that demonstrates low pecto-ralis major activity would be an important component of rehabili-tation to facilitate selective serratus anterior strengthening andreduction of the risk of the glenohumeral joint instability. There-fore, the wall push-up plus would be least indicated, as pectoralismajor activation was nearly twice that of the serratus anterior de-spite the use of isometric horizontal abduction.

However, limitations of this study should be noted. Other mus-cles beyond the pectoralis major and serratus anterior that couldcontribute to scapular control were not considered in this investi-gation. This was because the major interest in this study was theactivity of the pectoralis major and serratus anterior musclesduring three exercises with and without isometric horizontalabduction. Further studies are warranted to examine the muscle

K.-M. Park et al. / Journal of Electromyography and Kinesiology 23 (2013) 462–468 467

activity of other agonist and antagonist muscles which can affectscapulothoracic joint would strengthen the results of this study.Another limitation is that the restricted age range of our subjectpopulation (21–29 years of age) affects generalizability outside ofthis age group. Third, the EMG data were collected during isomet-ric holding contraction phase in this study. The changes in pecto-ralis major and serratus anterior activity with isometrichorizontal abduction during dynamic contraction (ascending anddescending phases) of three exercises should be investigated infurther investigation. Fourth, the order of two conditions (withvs. without isometric horizontal abduction) was not randomized.Further studies should consider randomizing two conditions in or-der to observe the effect of isometric horizontal abduction duringexercises more precisely. However, we believe that the lack of ran-domization did not interfere in our results, as we considered thatperforming exercises without isometric horizontal abduction firstcan prevent the results in this study from being affected by poten-tial carry over effect or learning effect.

5. Conclusions

This study examined the effects of isometric horizontal abduc-tion using Thera-Band on the EMG amplitude of the pectoralis ma-jor and serratus anterior and the pectoralis major/serratus anteriorEMG activity ratio during forward flexion, scaption, and wall push-up plus in subjects with scapular winging. Our results showed thatadding isometric horizontal abduction using Thera-Band signifi-cantly decreased pectoralis major activity during forward flexionand wall push-up plus, significantly increased serratus anterioractivity, and significantly decreased the pectoralis major/serratusanterior EMG activity ratio during forward flexion and wall push-up plus. Therefore, isometric horizontal abduction using Thera-Band can be implemented as an effective method for inducinggreater serratus anterior activation while preventing excessivepectoralis major activation during three exercises.

References

Chance-Larsen K, Littlewood C, Garth A. Prone hip extension with lower abdominalhollowing improves the relative timing of gluteus maximus activation inrelation to biceps femoris. Man Ther 2010;15(1):61–5.

Cools AM, Witvrouw EE, Declercq GA, Vanderstraeten GG, Cambier DC. Evaluationof isokinetic force production and associated muscle activity in the scapularrotators during a protraction–retraction movement in overhead athletes withimpingement symptoms. Br J Sports Med 2004;38(1):64–8.

Cools AM, Dewitte V, Lanszweert F, Notebaert D, Roets A, Soetens B, et al.Rehabilitation of scapular muscle balance: which exercises to prescribe? Am JSports Med 2007;35(10):1744–51.

Criswell E. Cram’s introduction to surface electromyography. 2nded. Massachusetts: Jones and Bartlett Publishers; 2010.

Decker MJ, Hintermeister RA, Faber KJ, Hawkins RJ. Serratus anterior muscle activityduring selected rehabilitation exercises. Am J Sports Med 1999;27(6):784–91.

Dvir Z, Berme N. The shoulder complex in elevation of the arm: a mechanismapproach. J Biomech 1978;11(5):219–25.

Ellenbecker TS. Shoulder rehabilitation: non-operative treatment. NY: ThiemeMedical Publishers, Inc.; 2006.

Escamilla RF, Yamashiro K, Paulos L, Andrews JR. Shoulder muscle activity andfunction in common shoulder rehabilitation exercises. Sports Med2009;39(8):663–85.

Farina D, Leclerc F, Arendt-Nielsen L, Buttelli O, Madeleine P. The change in spatialdistribution of upper trapezius muscle activity is correlated to contractionduration. J Electromyogr Kinesiol 2008;18(1):16–25.

Hallström E, Kärrholm J. Shoulder kinematics in 25 patients with impingement and12 controls. Clin Orthop Relat Res 2006;448:22–7.

Hardwick DH, Beebe JA, McDonnell MK, Lang CE. A comparison of serratus anteriormuscle activation during a wall slide exercise and other traditional exercises. JOrthop Sports Phys Ther 2006;36(12):903–10.

Hiengkaew V, Wichaiwong K, Chaiyakul S, Deesin A. Concerning the pectoralismajor in active reaching exercise. Electromyogr Clin Neurophysiol2003;43(3):157–63.

Hintermeister RA, Lange GW, Schultheis JM, Bey MJ, Hawkins RJ. Electromyographicactivity and applied load during shoulder rehabilitation exercises using elasticresistance. Am J Sports Med 1998;26(2):210–20.

Jonkers I, Stewart C, Spaepen A. The complementary role of the plantar flexors,hamstrings and gluteus maximus in the control of stance limb stability duringgait. Gait Posture 2003;17(3):264–72.

Kim B, Gong W, Lee S. The effect of push-up plus exercise with visual biofeedbackon the activity of shoulder stabilizer muscles for winged scapula. J Phys Ther Sci2010;22(4):355–8.

Kisner C, Colby LA. Therapeutic exercise: foundations and techniques. 4thed. Philadelphia: F.A. Davis Co.; 2002.

Konrad GG, Jolly JT, Labriola JE, McMahon PJ, Debski RE. Thoracohumeral muscleactivity alters glenohumeral joint biomechanics during active abduction. JOrthop Res 2006;24(4):748–56.

Labriola JE, Lee TQ, Debski RE, McMahon PJ. Stability and instability of theglenohumeral joint: the role of shoulder muscles. J Shoulder Elbow Surg2005;14(1 Suppl. S):32S–8S.

Leonard CT. The neuroscience of human movement: principles of reflex action andmotor control. 1st ed. St. Louis: Mosby; 1997.

Ludewig PM, Cook TM. Alterations in shoulder kinematics and associated muscleactivity in people with symptoms of shoulder impingement. Phys Ther2000;80(3):276–91.

Ludewig PM, Hoff MS, Osowski EE, Meschke SA, Rundquist PJ. Relative balance ofserratus anterior and upper trapezius muscle activity during push-up exercises.Am J Sports Med 2004;32(2):484–93.

Madeleine P, Leclerc F, Arendt-Nielsen L, Ravier P, Farina D. Experimental musclepain changes the spatial distribution of upper trapezius muscle activity duringsustained contraction. Clin Neurophysiol 2006;117(11):2436–45.

Madeleine P, Mathiassen SE, Arendt-Nielsen L. Changes in the degree of motorvariability associated with experimental and chronic neck-shoulder pain duringa standardized repetitive arm movement. Exp Brain Res 2008;185(4):689–98.

Martin RM, Fish DE. Scapular winging: anatomical review, diagnosis, and treatment.Curr Rev Musculoskelet Med 2008;1:1–11.

Martins J, Tucci HT, Andrade R, Araújo RC, Bevilaqua-Grossi D, Oliveira AS.Electromyographic amplitude ratio of serratus anterior and upper trapeziusmuscles during modified push-ups and bench press exercises. J Strength CondRes 2008;22(2):477–84.

McCann PD, Wootten ME, Kadaba MP, Bigliani LU. A kinematic andelectromyographic study of shoulder rehabilitation exercises. Clin OrthopRelat Res 1993;288:179–88.

McClure PW, Bialker J, Neff N, Williams G, Karduna A. Shoulder function and 3-dimensional kinematics in people with shoulder impingement syndrome beforeand after a 6-week exercise program. Phys Ther 2004;84:832–48.

McMahon PJ, Chow S, Sciaroni L, Yang BY, Lee TQ. A novel cadaveric model foranterior–inferior shoulder dislocation using forcible apprehension positioning. JRehabil Res Dev 2003;40(4):349–59.

Neumann DA. Kinesiology of the musculoskeletal system: foundations for physicalrehabilitation. 1st ed. St. Louis: Mosby; 2002.

Oh JS, Cynn HS, Won JH, Kwon OY, Yi CH. Effects of performing an abdominaldrawing-in maneuver during prone hip extension exercises on hip and backextensor muscle activity and amount of anterior pelvic tilt. J Orthop Sports PhysTher 2007;37(6):320–4.

Page P, Frank CC, Lardner R. Assessment and treatment of muscle imbalance: theJanda approach. 1st ed. Illinois: Human Kinetics; 2009.

Park SY, Yoo WG. Differential activation of parts of the serratus anterior muscleduring push-up variations on stable and unstable bases of support. JElectromyogr Kinesiol 2011;21(5):861–7.

Sharmann S. Diagnosis and treatment of movement impairment syndromes. 1sted. St. Louis: Mosby; 2001.

Stuart I. Human physiology: mechanism of contraction and neural control. 10thed. New York: McGraw-Hill; 2007.

Uhl TL, Carver TJ, Mattacola CG, Mair SD, Nitz AJ. Shoulder musculature activationduring upper extremity weight-bearing exercise. J Orthop Sports Phys Ther2003;33(3):109–17.

Wattanaprakornkul D, Cathers I, Halaki M, Ginn KA. The rotator cuff muscles have adirection specific recruitment pattern during shoulder flexion and extensionexercises. J Sci Med Sport 2011;14(5):376–82.

Weon JH, Kwon OY, Cynn HS, Lee WH, Kim TH, Yi CH. Real-time visual feedback canbe used to activate scapular upward rotators in people with scapular winging:an experimental study. J Physiother 2011;57(2):101–7.

Kyung-Mi Park received her B.S. degree in PhysicalTherapy from Daejeon University in 2010, and M.A.degree in Physical Therapy from Yonsei University in2012. She is a member of the Korean Physical Ther-apy Association. Her research interests are the clini-cal biomechanics associated with therapeutic optionsfor enhancing a variety of musculoskeletal problems,including the scapular winging, patellofemoral painsyndrome, and pes planus.

468 K.-M. Park et al. / Journal of Electromyography and Kinesiology 23 (2013) 462–468

Heon-Seock Cynn is an associate professor in theDepartment of Physical Therapy at the College ofHealth Science of Yonsei University. He received B.S.degree in Physical Therapy from Yonsei University,M.A. degree in Physical Therapy from New YorkUniversity, and Ph.D. degree in Physical Therapy fromYonsei University. He was a full time lecturer of SeoulHealth College and an associate professor of HanseoUniversity. His research interests are identification ofetiologic factors, classification, and interventionapproaches for movement disorders and musculo-skeletal diseases.

Chung-Hwi Yi received his Ph.D. degree in PhysicalTherapy from Yonsei University in 1990. He joinedthe Department of Rehabilitation Therapy of YonseiUniversity in 1993. He was a president of The KoreanAcademy of University Trained Physical Therapists.From 1993 onwards he has been employed as aprofessor in the Department of Physical Therapy atthe College of Health Science of Yonsei University.His research interests include motion, posture anal-ysis, and the development of outcome measures forevaluating disability.

Oh-Yun Kwon is a professor in the Department ofPhysical Therapy at the College of Health Science ofYonsei University. He received his B.S. degree inPhysical Therapy and M.P.H. degree from YonseiUniversity in 1986 and 1992 respectively, and Ph.D.degree from Keimyung University in 1998. He hadresearch experience in Program in Physical Therapyat Washington University in St Louis as a Post Doc-toral Fellow. He is a director in Lab of Kinetic Ergocisebased on Movement Analysis (KEMA). He is inter-ested in the mechanisms of movement impairment,movement analysis, and prevention and manage-ment of the work related musculoskeletal pain syn-drome.