ejercicios basados en la evidencia escapulotoracicos y glh

journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 105

[ CLINICAL COMMENTARY ]

1Coordinator of Rehabilitation Research and Education, Department of Orthopedic Surgery, Division of Sports Medicine, Massachusetts General Hospital, Boston, MA; RehabilitationCoordinator/Assistant Athletic Trainer, Boston Red Sox Baseball Club, Boston, MA. 2Professor, Department of Physical Therapy, California State University, Sacramento, Sacramento,CA. 3Clinical Director, Champion Sports Medicine, Director of Rehabilitative Research, American Sports Medicine Institute, Birmingham, AL. Address correspondence to Dr MichaelM. Reinold, Rehabilitation Coordinator/Assistant Athletic Trainer, Boston Red Sox Baseball Club, Fenway Park, 4 Yawkey Way, Boston, MA 02215. Email: [email protected]

MICHAEL M. REINOLD, PT, DPT, ATC, CSCS¹ PT, PhD, CSCS, FACSM² PT, DPT³

Current Concepts in the Scientificand Clinical Rationale Behind

Exercises for Glenohumeral andScapulothoracic Musculature

The biomechanical analysis of rehabilitation exercises hasgained recent attention. As our knowledge of specific musclebiomechanics and function has increased, we have seena gradual progression towards more scientifically based

rehabilitation exercises. Several investigators have sought to describecommon rehabilitation exercises using kinematics, kinetics, andelectromyographic (EMG) data in an attempt to better understand theimplications of each exercise on the soft tissues of the glenohumeraland scapulothoracic joints. Advances in the understanding of

advantageous rehabilitation programs.The purpose of this paper is to provide

an overview of the biomechanical andclinical implications associated with therehabilitation of the glenohumeral andscapulothoracic joints. We will reviewthe function and biomechanics of eachmuscle, with specific emphasis on manycommonly performed rehabilitation ex-ercises. The goal of this is to provide theclinician with a thorough overview of theavailable information to develop safe,potentially effective, and appropriate ex-ercise programs for injury rehabilitationand prevention.

The rotator cuff has been shownto be a substantial dynamic stabiliz-er of the glenohumeral joint in mul-

tiple shoulder positions.49,88 Appropriaterehabilitation progression and strength-ening of the rotator cuff muscles areimportant to provide appropriate forceto help elevate and move the arm, com-press and center the humeral head withinthe glenoid fossa during shoulder move-ments (providing dynamic stability), andprovide a counterforce to humeral headsuperior translation resulting from del-

the biomechanical factors of rehabilita-tion have led to the enhancement ofrehabilitation programs that seekto facilitate recovery, while plac-ing minimal strain on specifichealing structures.

Though the fields of orthope-dics and sports medicine have evolved

to emphasize the necessity of evidence-based practice, few studies have been

conducted to determine the efficacyof specific shoulder rehabilitationexercises. Thus, knowledge ofanatomy, biomechanics, and func-

tion of specific musculature is criti-cal in an attempt to develop the most

The biomechanical analysis of re-habilitation exercises has led to more scientificallybased rehabilitation programs. Several investiga-tors have sought to quantify the biomechanics andelectromyographic data of common rehabilitationexercises in an attempt to fully understand theirclinical indications and usefulness. Furthermore,the effect of pathology on normal shoulder bio-mechanics has been documented. It is importantto consider the anatomical, biomechanical, andclinical implications when designing exercise

programs. The purpose of this paper is to providethe clinician with a thorough overview of the avail-able literature relevant to develop safe, effective,and appropriate exercise programs for injuryrehabilitation and prevention of the glenohumeraland scapulothoracic joints.

Level 5. J Orthop SportsPhys Ther 2009; 39(2):105-117. doi:10.2519/jospt.2009.2835

electromyography, infraspinatus,serratus anterior, supraspinatus, trapezius

106 | february 2009 | volume 39 | number 2 | journal of orthopaedic & sports physical therapy

[ CLINICAL COMMENTARY ]toid activity (minimizing subacromialimpingement).6,8,9,21,34,50,60,65,74 In addi-tion, rotator cuff muscles are frequentlytreated either conservatively or surgicallysecondary to injuries.

Exercise designed to strengthen themuscles of the rotator cuff are often pre-scribed to patients with pathologies suchas subacromial impingement. Duringscapular plane abduction in healthy sub-jects, the humeral head translates 1 to 3mm in the superior direction from 0° to30° of abduction, slightly inferiorly from30° to 60° of abduction, and in the su-perior or inferior direction during 60° to90° of abduction.26,50,67 Other data dem-onstrate that, during passive scapularplane abduction, the humeral head trans-lates superiorly 0.6 to 1.8 mm between0° to 150°.25,26 But during active scapularplane abduction the humeral head re-mains nearly centered in the glenoid fos-sa throughout the range of movement.26

These data illustrate the importance ofrotator cuff strength and muscle balanceto resist humeral head superior transla-tion and help center the humeral headwithin the glenoid fossa during shoulderelevation.74 With rotator cuff pathology,altered kinematics and muscle activ-ity are present,31 and superior humeralhead translation increases and subacro-mial space decreases.24 Moreover, duringscapular plane shoulder abduction from30° to 90°, infraspinatus and subscapu-laris activity was found to be significantlyless in individuals with subacromial im-pingement compared to those withoutimpingement.68

Subjects with shoulder laxity and in-stability have also been shown to havealtered kinematics and firing patterns ofthe rotator cuff.7,35,45,46,55,64,72 Compared tohealthy subjects, patients with general-ized joint laxity demonstrated increasedsubscapularis activity during internalrotation (IR) exercise and decreased su-praspinatus and subscapularis activityduring external rotation (ER) exercise.7,43

Compared to healthy subjects, those withanterior instability exhibited less su-praspinatus activity between 30° to 60°

of shoulder elevation during abductionand scaption exercises.59

These EMG data clearly illustrateaberrant muscle-firing patterns in in-dividuals with shoulder pathology. It isoften the goal of rehabilitation special-ists to prescribe exercises to normalize orprevent these abnormal firing patterns.Proper selection of exercises to activatemuscle function for each muscle of therotator cuff should be considered duringrehabilitation.

The supraspinatus compresses, abducts,and generates a small ER torque to theglenohumeral joint. Supraspinatus activi-ty increases as resistance increases duringabduction/scaption movements, peakingat 30° to 60° of elevation for any givenresistance. At lower elevation angles, su-praspinatus activity increases, providingadditional humeral head compressionwithin the glenoid fossa to counter thehumeral head superior translation occur-ring with contraction of the deltoid.1 Dueto a decreasing moment arm with abduc-tion, the supraspinatus is a more effectiveabductor in the scapular plane at smallerabduction angles.34,50,65

Relatively high supraspinatus activityhas been measured in several commonrotator cuff exercises3,5,17,33,54,63,70,75,79,87

and in several exercises that are notcommonly thought of as rotator cuff ex-ercises, such as standing forward scap-ular punch, rowing exercises, push-upexercises, and 2-hand overhead medi-cine ball throws.13,17,32,81 These resultssuggest the importance of the rotatorcuff in providing dynamic glenohumeralstability by centering the humeral headwithin the glenoid fossa during all up-per extremity functional movements.This is an important concept for theclinician to understand. The muscle’sability to generate abduction torque inthe scapular plane appears to be great-est with the shoulder in neutral rotationor in slight IR or ER.50,65 This biome-chanical advantage has led to the devel-opment of exercises in the plane of the

scapula to specifically strengthen thesupraspinatus.38

Jobe38 was the first to recommend ele-vation in the scapular plane (30° anteriorto the frontal plane) with glenohumer-al IR, or the “empty can” exercise, tostrengthen the supraspinatus muscle.Other authors37,40,66,69,70,77 have suggestedthe “full can” position, or elevation in thescapular plane with glenohumeral ER,to best strengthen and test the supraspi-natus muscle. Furthermore, comparedto the empty can exercise, Blackburn5

reported significantly greater supraspi-natus activity during prone horizontalabduction at 100° with full ER, or pronefull can, position. The results of studiescomparing these exercises provide in-consistent results due to methodologicallimitations, including lack of statisticalanalysis,38,79 lack of data for all 3 exercis-es,40,54,79,87 and absence of data on deltoidmuscle activity.87

Recently, Reinold et al69 comprehen-sively evaluated the EMG signal of thesupraspinatus and deltoid musculatureduring the full can, empty can, andprone full can exercises in an attemptto clarify the muscular activation duringthese exercises. The results showed thatall 3 exercises provide a similar amountof supraspinatus activity ranging from62% to 67% of maximal voluntary iso-metric contraction (MVIC). However,the full can exercise demonstrated asignificantly lower amount of middleand posterior deltoid activity comparedto the 2 other exercises. This is clinicallysignificant when trying to strengthenthe supraspinatus while simultaneouslyminimizing potentially disadvanta-geous superior sheer force due to del-toid activity.

In patients with shoulder pain,weakness of the rotator cuff, or inef-ficient dynamic stabilization, it is theauthors’ opinion that activities thatproduce higher levels of deltoid activ-ity in relation to supraspinatus activity,such as the empty can and prone fullcan exercise, may be detrimental. Thisis due to the increased amount of supe-


rior humeral head migration that maybe observed when the rotator cuff doesnot adequately compress the humeruswithin the glenoid fossa to counteract,or oppose, the superior pull of the del-toid ( ).61 Poppen and Walker66

have shown that the empty can exerciseresults in a greater superior-orientatedforce vector than the full can exercise( ). This superior humeral headmigration may result in subacromialimpingement, subdeltoid bursa trauma,

bursal thickening, and may result intendon degeneration and eventual fail-ure.21 Clinically, superior humeral headmigration may be disadvantageous topatients with rotator cuff pathology ora deficiency in glenohumeral dynamicstabilization that are symptomatic. Thismay partially explain why the empty canposition often elicits a certain amountof pain and discomfort in patients.

In addition to the altered ratio of su-praspinatus to deltoid muscle activity,

there are several reasons why the fullcan exercise may be preferred over theempty can exercise during rehabilitationand supraspinatus testing. Anatomically,the IR of the humerus during the emptycan exercise does not allow the greatertuberosity to clear from under the acro-mion during arm elevation, which mayincrease subacromial impingement riskbecause of decreased subacromial spacewidth.15,23,71

Biomechanically, shoulder abductionperformed in extreme IR progressivelydecreases the abduction moment arm ofthe supraspinatus from 0° to 90° of ab-duction.50 A diminished mechanical ad-vantage may result in the supraspinatusneeding to generate more force, thus in-creasing the tensile stresses in the injuredor healing tendon. This may also makethe exercise more challenging for patientswith weakness, facilitating compensatorymovements such as a shoulder “shrug.”

Scapular kinematics are also differentbetween these exercises, with scapularIR, or “winging” (which occurs in thetransverse plane with the scapular me-dial border moving posterior away fromthe trunk) and anterior tilt (which occursin the sagittal plane with the scapular in-ferior angle moving posterior away fromthe trunk) being greater with the emptycan compared to the full can exercise.78

This occurs in part because IR of thehumerus in the empty can position ten-sions both the posteroinferior capsule ofthe glenohumeral joint and the rotatorcuff (primarily the infraspinatus). Ten-sion in these structures contributes toanterior tilt and IR of the scapula, whichcontribute to scapular protraction. Thisis clinically important because scapularprotraction has been shown to decreasethe width of the subacromial space, in-creasing the risk of subacromial impinge-ment.76 In contrast, scapular retractionhas been shown to both increase sub-acromial space width76 and increase su-praspinatus strength potential (enhancedmechanical advantage), when comparedto a more protracted position.41 Thesedata also emphasize the importance of

Direction of the magnitude of the resultant force vector for different glenohumeral joint positions as afunction of different muscle activity, (A) deltoid activity, (B) rotator cuff activity, (C) combined deltoid and rotatorcuff activity. Reprinted with permission from Morrey et al.61

The position of the resultant force vector of the rotator cuff and deltoid for different positions of armelevation with (N) neutral rotation, (I) internal rotation, and (X) external rotation. Reprinted with permission fromPoppen and Walker.66


[ CLINICAL COMMENTARY ]until it is about 1.3 cm at 60° abduction.65

These data imply that the infraspinatus isa more effective external rotator at lowershoulder abduction angles. The teres mi-nor has a relatively constant ER momentarm (approximately 2.1 cm) and the abil-ity to generate torque throughout shoul-der abduction movement, which impliesthat shoulder abduction angle does notaffect the effectiveness of the teres minorto generate ER torque.65

Several studies have been designed totest the results of the model; but, as instudies on the supraspinatus, variationsin experimental methodology have result-ed in conflicting results and controversyin exercise selection.3,5,17,19,27,33,44,54,63,70,77,79,81

Several exercises have been recommend-ed based on EMG data, including shoul-der ER in the side-lying,3,70,79 standing,27,70

or prone3,70 positions performed at 0°,3,70

45°,27,70 and 90°3,70 of abduction. Anotherexercise that has been shown to generatea high EMG signal of the infraspinatusand teres minor is prone horizontal ab-duction with ER.5,79

Reinold et al70 analyzed several dif-ferent exercises commonly used tostrengthen the shoulder external rota-tors to determine the most effectiveexercise and position to recruit muscleactivity of the posterior rotator cuff. Theauthors report that the exercise that elic-ited the most combined EMG signal forthe infraspinatus and teres minor wasshoulder ER in side-lying (infraspina-tus, 62% maximal voluntary isometriccontraction [MVIC]; teres minor, 67%MVIC), followed closely by standing ERin the scapular plane at 45° of abduction(infraspinatus, 53% MVIC; teres minor,55% MVIC), and finally prone ER in the90° abducted position (infraspinatus,50% MVIC; teres minor, 48% MVIC).

Exercises in the 90° abducted posi-tion are often incorporated to simulatethe position and strain on the shoulderduring overhead activities such as throw-ing. This position produced moderateactivity of the external rotators but alsoincreased activity of the deltoid and su-praspinatus. It appears that the amount

of infraspinatus and teres minor activityprogressively decreases as the shouldermoves into an abducted position, whileactivity of the supraspinatus and deltoidincreases. This suggests that as the armmoves into a position of increased vulner-ability away from the body, the supraspi-natus and deltoid are active to assist inthe ER movement, while providing somedegree of glenohumeral stability throughmuscular contraction.

While standing ER exercises per-formed at 90° of shoulder abduction mayhave a functional advantage over exercis-es performed at 0° of shoulder abductionor performed in the scapular plane, dueto the close replication in sporting activi-ties, the combination of shoulder abduc-tion and ER places strain on the shouldercapsule, particularly the anterior band ofthe inferior glenohumeral ligament.30,85,86

The clinician must carefully consider thiswhen designing programs for patientswith capsulolabral pathology.

Side-lying ER may be the optimal ex-ercise to strengthen the external rotatorsbased on the previously mentioned stud-ies. The inclusion of this exercise shouldbe considered in all exercise programsattempting to increase ER strength ordecrease capsular strain.

Theoretically, ER performed at 0° ofshoulder abduction with a towel roll be-tween the rib cage and the arm providesboth the low capsular strain and also agood balance between the muscles thatexternally rotate the arm and the musclesthat adduct the arm to hold the towel.Our clinical experience has shown thatadding a towel roll to the ER exerciseprovides assistance to the patient by en-suring that proper technique is observedwithout muscle substitution. Reinold etal70 report that adding a towel roll to theexercise consistently exhibited a tendencytowards higher activity of the posteriorrotator cuff muscles as well. An increaseof 20% to 25% in EMG signal of the in-fraspinatus and teres minor was notedwhen using the towel roll compared tono towel roll.

What is not readily apparent is the

strengthening the scapular retractors andmaintaining a scapular retracted postureduring shoulder exercises. The authorsroutinely instruct patients to emphasizean upright posture and a retracted posi-tion of the scapula during all shoulderand scapula strengthening exercises.

Thus, the full can exercise appears tobe the most advantageous exercise whilethe empty can exercise is not commonlyrecommended. The prone full can exer-cise warrants further consideration be-cause the exercise results in greater EMGsignal of the posterior deltoid than themiddle deltoid, which may result in lesssuperior sheer force. The prone full canexercise may also be beneficial because ofscapular muscle recruitment.

The infraspinatus and teres minor com-prise the posterior cuff, which providesglenohumeral compression and resistssuperior and anterior humeral headtranslation by exerting an inferoposteri-or force on the humeral head.74 The pos-terior cuff muscles provide glenohumeralER, which functionally helps clear thegreater tuberosity from under the cora-coacromial arch during overhead move-ments, thus minimizing subacromialimpingement.

Based on 3-D biomechanical shouldermodels, the maximum predicted isomet-ric infraspinatus force was 723 N for ERat 90° of abduction and 909 N for ER at0° of abduction.34 The maximum predict-ed teres minor force was much less thanfor the infraspinatus during maximumER at both 90° (111 N) and 0° abduction(159 N).34 The effectiveness of the musclesof the posterior rotator cuff to externallyrotate the arm depends on glenohumeralposition. The superior, middle, and infe-rior heads of the infraspinatus have theirlargest ER moment arm (approximate-ly 2.2 cm) and generate their greatesttorque at 0° abduction.65 As the abduc-tion angle increases, the moment arms ofthe inferior and middle heads stay rela-tively constant, while the moment arm ofthe superior head progressively decreases


significant role of the infraspinatus asa shoulder abductor in the scapularplane.34,50,65 From 3-D biomechanicalshoulder models, predicted infraspina-tus force during maximum isometriceffort scapular plane abduction (90°position) was 205 N, nearly twice thepredicted force from the supraspinatusin this position.34 Liu et al50 reportedthat in scapular plane abduction withneutral rotation the infraspinatus has anabductor moment arm that was small at0° abduction, but increased to 1 cm at15° abduction, and remained fairly con-stant throughout increasing abductionangles. Moreover, infraspinatus activityincreases as resistance increases, peakingat 30° to 60° for any given resistance.1 Asresistance increases, infraspinatus activ-ity increases to help generate a highershoulder scapular abduction torque, and,at lower elevation angles, infraspinatusactivity increases to resist superior hu-meral head translation due to the actionof the deltoid.74

In contrast to the infraspinatus, theteres minor generates a weak shoulderadductor torque due to its relativelylower attachments to the scapula and hu-merus.34,50,65 A 3-D biomechanical mod-el of the shoulder reveals that the teresminor does not generate scapular planeabduction torque when it contracts, but,rather, generates an adduction torqueand 94 N of force during maximum effortscapular plane adduction.34 In addition,Otis et al65 reported that the adductormoment arm of the teres minor was ap-proximately 0.2 cm at 45° of IR and ap-proximately 0.1 cm at 45° of ER. Thesedata imply that the teres minor is a weakadductor of the humerus, regardless ofthe rotational position of the humerus.In addition, because of its posterior posi-tion at the shoulder, it also helps gener-ate a weak horizontal abduction torque.Therefore, although its activity is simi-lar to the infraspinatus during ER, it ishypothesized that the teres minor wouldnot be as active as the infraspinatus dur-ing scapular abduction, abduction, andflexion movements, but would show ac-

tivity similar to that of the infraspina-tus during horizontal abduction. Thishypothesis is supported by EMG andmagnetic resonance imaging data, whichshow that teres minor activity duringflexion, abduction, and scapular abduc-tion is drastically less than infraspinatusactivity.1,3,5,54,77,79 Even though the teresminor generates an adduction torque, itis active during these different elevation-type movements, as it likely acts to en-hance joint stability by resisting superiorhumeral head translation and providinghumeral head compression within theglenoid fossa.74 This is especially likelythe case at lower shoulder abductionangles and when abduction and scapu-lar abduction movements are performedagainst greater resistance.1 In contrast tothe movements of shoulder abduction,scapular abduction, and flexion, teres mi-nor activity is much higher during pronehorizontal abduction at 100° abductionwith ER, exhibiting similar activity as theinfraspinatus.5,54,70,77,79

The subscapularis provides glenohumer-al compression, IR, and anterior stabilityof the shoulder. From 3-D biomechanicalshoulder models, predicted subscapu-laris force during maximum effort IRwas 1725 N at 90° abduction and 1297N at 0° abduction.34 Its superior, middle,and inferior heads all have their larg-est IR moment arm (approximately 2.5cm) and torque generation at 0° abduc-tion.65 As the abduction angle increases,the moment arms of the inferior andmiddle heads stay relatively constant,while the moment arm of the superiorhead progressively decreases until it isabout 1.3 cm at 60° abduction.65 Thesedata imply that the upper portion of thesubscapularis muscle (innervated by theupper subscapularis nerve) may be amore effective internal rotator at lowerabduction angles compared to higher ab-duction angles. However, there is no sig-nificant difference in upper subscapularisactivity among IR exercises performed at0°, 45°, or 90° abduction.17,39 Abduction

angle does not appear to affect the abilityof the lower subscapularis (innervated bythe lower subscapularis nerve) to gener-ate IR torque.65 However, lower sub-scapularis muscle activity is affected byabduction angle, where some EMG datashow significantly greater activity withIR at 0° abduction compared to IR at 90°abduction,17 while EMG data of anotherstudy show greater activity with IR ex-ercise performed at 90° compared to 0°abduction.39 Performing IR at 0° abduc-tion produces similar amounts of upperand lower subscapularis activity.17,28,39

Although biomechanical data remaininconclusive as to which position to per-form IR exercises (0° versus 90° abduc-tion), during IR at 0° abduction the actionof the subscapularis is assisted by severallarge muscles, such as the pectoralis ma-jor, latissimus dorsi, and teres major.17

Clinically, this may allow for compensa-tion of larger muscles during the exercisein the presence of subscapularis weak-ness. Decker et al17 demonstrated that IRat 90° abduction produced less pectoralismajor activity compared to 0° abduction.The authors’ findings revealed that pecto-ralis major and latissimus dorsi activityincreased when performing IR exercisesin an adducted position or while mov-ing into an adducted position during theexercise. Thus, IR at 90° abduction maybe performed if attempting to strengthenthe subscapularis while minimizing larg-er muscle group activity.

The subscapularis is active in numer-ous shoulder exercises other than specificIR of the shoulder. Decker et al17 reportedhigh subscapularis activity during thepush-up with plus and dynamic-hug ex-ercises. These authors also described an-other exercise that consistently producedhigh levels of subscapularis activity, whichthey called the “diagonal exercise” (3). Relatively high subscapularis activ-ity has been measured while performingside-lying shoulder abduction, standingshoulder extension from 90° to 0°, mili-tary press, D2 diagonal proprioceptiveneuromuscular facilitation (PNF) patternflexion and extension, and PNF scapular



clock, depression, elevation, protraction,and retraction movements.17,33,44,63,75,79

The subscapularis also generatesan abduction torque during arm eleva-tion.50,65 From 3-D biomechanical shoul-der models, predicted subscapularis forceduring maximum effort scapular planeabduction at 90° was 283 N, approxi-mately 2.5 times the predicted force forthe supraspinatus in this position.34 Thiswas similar to that of the infraspinatus,highlighting the theoretical force couplethat the 2 muscles provide to center thehumeral head within the glenoid fossaduring abduction. Liu et al50 reportedthat in scapular plane abduction withneutral rotation the subscapularis had apeak abductor moment arm of 1 cm at 0°abduction, which slowly decreased to 0cm at 60° abduction. Moreover, the ab-ductor moment arm of the subscapularisgenerally decreased as abduction was per-formed with greater shoulder IR,50 suchas performing the empty can exercise. Incontrast, the abductor moment arm ofthe subscapularis generally increased asabduction was performed with greatershoulder ER, similar to performing thefull can exercise.

Otis et al65 reported that the superior,middle, and inferior heads of the sub-

scapularis all have an abductor momentarm (greatest for the superior head andleast for the inferior head) that varies as afunction of humeral rotation. The lengthsof the moment arm for the 3 muscle headsare approximately 0.4 to 2.2 cm at 45° ofER, 0.4 to 1.4 cm in neutral rotation, and0.4 to 0.5 cm at 45° of IR. These data sug-gest that the subscapularis is most effec-tive as a scapular plane abductor with theshoulder in ER and least effective withthe shoulder in IR. Therefore, the simul-taneous activation of the subscapularisand infraspinatus during arm elevationgenerates both an abductor moment andan inferiorly directed force to the humer-al head to resist superior humeral headtranslation.74 In addition, a simultane-ous activation neutralizes the IR and ERtorques these muscles generate, furtherenhancing joint stability.

DELTOID

The deltoid plays an importantrole in shoulder biomechanics andduring glenohumeral and scapu-

lothoracic exercises. Extensive researchhas been conducted on deltoid activityduring upper extremity weight-liftingexercises, such as bench press, dumb-

bell flys, military press, and push-ups.4,13,16,19,44,57,63,79,81,83

The abductor moment arm is ap-proximately 0 cm for the anterior del-toid and 1.4 cm for the middle deltoidwhen the shoulder is in 0° abductionand neutral rotation in the scapularplane.50,65 The magnitude of these mo-ment arms progressively increases withshoulder abduction, such that, by 60°of abduction, they are approximately1.5 to 2 cm for the anterior deltoid and2.7 to 3.2 cm for the middle deltoid.From 0° to 40°of abduction the momentarms for the anterior and middle del-toids are less than the moment arms forthe supraspinatus, subscapularis, andinfraspinatus.50,65 These data suggestthat the anterior and middle deltoidare not effective shoulder abductors atlow abduction angles and the shoulderin neutral rotation, especially the ante-rior deltoid. This is in contrast to thesupraspinatus and to a lesser extent theinfraspinatus and subscapularis, whichare more effective shoulder abductorsat low abduction angles. These biome-chanical data are consistent with EMGdata, in which anterior and middle del-toid activity generally peaks between60° to 90° of abduction in the scapularplane, while supraspinatus, infraspina-tus, and subscapularis activity generallypeaks between 30° and 60° of shoulderabduction in the scapular plane.1

The abductor moment arm for theanterior deltoid changes considerablywith humeral rotation, increasing withER and decreasing with IR.50 At 60° ERand 0° abduction, a position similar tothe beginning of the full can exercise, theanterior deltoid moment arm is 1.5 cm(compared to 0 cm in neutral rotation),which makes the anterior deltoid an ef-fective abductor even at small abductionangles.50 By 60° abduction with ER, itsmoment arm increased to approximately2.5 cm (compared to approximately 1.5to 2 cm in neutral rotation).50 In con-trast, at 60° IR at 0° abduction, a po-sition similar to the beginning of theempty can exercise, its moment arm was

Diagnonal exercise for the subscapularis begins in shoulder external rotation at 90° abduction in thecoronal plane (A) and internal rotation and horizontal adduction are performed simultaneously (B), similar to atennis swing.


0 cm (the same as with neutral rotation),which suggests that in this positionthe anterior deltoid is not an effectiveabductor.50

It has been reported that, given a peakisometric abduction torque of 25 N·m at0° abduction and neutral rotation, up to35% to 65% of this torque may be gener-ated by the middle deltoid, 30% by thesubscapularis, 25% by the supraspinatus,10% by the infraspinatus, 2% by the an-terior deltoid, and 0% by the posteriordeltoid.50 Interestingly, the rotator cuffprovides a significant contribution to theabduction torque. The ineffectivenessof the anterior and posterior deltoids togenerate abduction torque with neutralrotation may appear surprising.50,65 How-ever, it is important to understand thatthe low abduction torque for the anteriordeltoid does not mean that this muscle isonly minimally active. In fact, because theanterior deltoid has an abductor momentarm near 0 cm, the muscle could be veryactive and generate very high force butvery little torque (in 0° abduction thisforce attempts to translate the humeralhead superiorly).

The aforementioned torque data arecomplemented and supported by muscleforce data from Hughes and An.34 Theseauthors reported predicted forces fromthe deltoid and rotator cuff during maxi-mum effort abduction with the arm 90°abducted and in neutral rotation. Poste-rior deltoid and teres minor forces wereonly 2 N and 0 N, respectively, whichfurther demonstrates the ineffectivenessof these muscles as shoulder abductors.In contrast, middle deltoid force was thehighest at 434 N, which suggests a highcontribution of this muscle during abduc-tion. The anterior deltoid generated thesecond highest force of 323 N. This mayappear surprising given the low abductortorque for this muscle reported above,but it should be re-emphasized that forceand torque are not the same, and that theshoulder was positioned at 90° abductionin the study by Hughes and An,34 in con-trast to 0° abduction in the study by Liuet al.50 As previously mentioned, the mo-

ment arm of the anterior deltoid progres-sively increases as abduction increases,and it becomes a more effective abduc-tor. It is also important to rememberthat muscle force is generated not only togenerate joint torque, but also to providestabilization, such as joint compression.Also of interest is the 608-N force that,collectively, the subscapularis (283 N),infraspinatus (205 N), and supraspinatus(117 N) generate. These larges forces aregenerated not only to abduct the shoul-der but also to compress and stabilizethe joint, and neutralize the superiorlydirected force generated by the deltoid atlower abduction angles.

It should also be noted that deltoidmuscle force in different shoulder posi-tions may also affect shoulder stability.All 3 heads of the deltoid generate a forcethat increases shoulder stability at 60°abduction in the scapular plane (helps tostabilize the humeral head in the glenoidfossa) but decreases shoulder stability at60° abduction in the frontal plane (tendsto translate the humeral head anterior).48

These data provide evidence for the useof scapular abduction exercises instead ofabduction exercises for individuals withanterior instability.

Thus, it appears that the 3 heads ofthe deltoid have different roles duringupper extremity movements and, there-fore, different implications for exerciseselection. The middle deltoid may havethe most significant impact on superiorhumeral head migration, and exerciseswith high levels of middle deltoid activity(as well as anterior deltoid activity), suchas the empty can exercise, should likelybe minimized for most patients. Con-versely, high levels of posterior deltoidactivity may not be as disadvantageousas high levels of middle or anterior del-toid activity. It does not appear that theposterior deltoid has a significant role inproviding abduction or superior humeralhead migration. Thus, exercises such asthe prone full can, which generates highlevels of rotator cuff and posterior deltoidactivity, may be both safe and effective forrotator cuff strengthening.

The primary muscles that con-trol scapular movements include thetrapezius, serratus anterior, levator

scapulae, rhomboids, and pectoralis mi-nor. Appropriate scapular muscle strengthand balance are important because thescapula and humerus move together incoordination during arm movement,referred to as scapulohumeral rhythm.During humeral elevation, the scapulaupwardly rotates in the frontal plane,rotating approximately 1° for every 2°of humeral elevation until 120° humeralelevation, and thereafter rotates approxi-mately 1° for every 1° humeral elevationuntil maximal arm elevation, achievingat least 45° to 55° of upward rotation.52,58

During humeral elevation, in addition toscapular upward rotation, the scapula alsonormally tilts posteriorly approximately20° to 40° in the sagittal plane and exter-nally rotates approximately 15° to 35° inthe transverse plane.52,58

When the normal 3-D scapular move-ments are disrupted by abnormal scapularmuscle-firing patterns, fatigue, or injury,it has been hypothesized that the shouldercomplex functions less efficiently, leadingto injuries to the shoulder, including theglenohumeral joint.10,11,12,18,58,76,80,82 Duringarm elevation in the scapular plane, in-dividuals with subacromial impingementexhibit decreased scapular upward rota-tion, increased scapular IR (winging) andanterior tilt, and decreased subacromialspace width, compared to those withoutsubacromial impingement.24,51 Alteredscapular muscle activity is commonly as-sociated with impingement syndrome.For example, upper and lower trapeziusactivity increased and serratus anterioractivity decreased in individuals with im-pingement as compared to those withoutimpingement.51 Therefore, it is importantto include the scapulothoracic muscula-ture in the rehabilitation of patients withshoulder pathology.42

The serratus anterior works with the

raul

Resaltado



pectoralis minor to protract the scapulaand with the upper and lower trapeziusto upwardly rotate the scapula. The ser-ratus anterior is an important muscle be-cause it contributes to all components ofnormal 3-D scapular movements duringarm elevation, which includes upwardrotation, posterior tilt, and external ro-tation.52,58 The serratus anterior is alsoimportant in athletics, such as duringoverhead throwing, to accelerate thescapula during the acceleration phase ofthrowing. The serratus anterior also helpsstabilize the medial border and inferiorangle of the scapula, preventing scapularIR (winging) and anterior tilt.

Several exercises elicit high serratusanterior activity, such as D1 and D2 di-agonal PNF pattern flexion, D2 diagonalPNF pattern extension, supine scapularprotraction, supine upward scapularpunch, military press, push-up plus, gle-nohumeral IR and ER at 90° abduction,and shoulder flexion, abduction, and

scaption with ER above 120°.16,20,32,62,63

Serratus anterior activity tends to increasein a somewhat linear fashion with arm el-evation.2,20,29,52,62 However, increasing armelevation increases subacromial impinge-ment risk,15,71 and arm elevation at lowerabduction angles also generates relativelyhigh serratus anterior activity.20

It is interesting that performingshoulder IR and ER at 90° of abductiongenerates relatively high serratus ante-rior activity, because these exercises areusually thought to primarily work rotatorcuff muscles.20,63 However, during IR andER at 90° abduction the serratus ante-rior helps stabilize the scapula. It shouldbe noted that the rotator cuff musclesalso act to move the scapula (where theyoriginate) in addition to the humerus.For example, the force exerted by the su-praspinatus at the supraspinous fossa hasthe ability to downwardly rotate the scap-ula if this force is not counterbalanced bythe scapulothoracic musculature.

Not surprising is high serratus ante-rior activity generated during a push-upexercise. When performing the stan-dard push-up, push-up on knees, andwall push-up, serratus anterior activityis greater when full scapular protrac-

tion occurs after the elbows fully extend(push-up plus).53 Moreover, serratusanterior activity was lowest in the wallpush-up plus, exhibited moderate activi-ty during the push-up plus on knees, andrelatively high activity during the stan-dard push-up plus.16,53 Compared to thestandard push-up, performing a push-up plus with the feet elevated producedsignificantly greater serratus anterioractivity.47 These findings demonstratethat serratus anterior activity increasesas the positional (gravitational) chal-lenge increases.

Decker et al16 compared several com-mon exercises designed to recruit the ser-ratus anterior. The authors identified thatthe 3 exercises that produced the great-est serratus anterior EMG signal were thepush-up with a plus, dynamic hug (4), and punch exercises (similar to a jab-bing protraction motion).

Ekstrom20 also looked at the activityof the serratus anterior during commonexercises. His data indicated that theserratus anterior is more active whenperforming a movement that simultane-ously creates scapular upward rotationand protraction, as with the serratus an-terior punch performed at 120° of abduc-

Dynamic hug exercise for the serratusanterior begins with the elbows in approximately 45°of flexion, the shoulder abducted 60° and internallyrotated 45° (A). The humerus is then horizontallyadducted by following an arc movement similar toa hugging action, until full shoulder protraction isreached (B).

Bilateral serratus anterior punch to 120° abduction begins with hands by the side (A) before extendingelbows and elevating shoulders up to 120° of elevation and full protraction (B).


tion and during a diagonal exercise thatincorporated protraction with shoulderflexion, horizontal adduction, and exter-nal rotation. It appears that the punch ex-ercise can be enhanced by starting at 0°abduction and extending the elbow, whileelevating and protracting the shoulder( ).

Hardwick et al29 compared the wallpush-up plus, full can, and a wall slideexercise. The wall slide begins by slightlyleaning against the wall with the ulnarborder of the forearms in contact withthe wall, elbows flexed 90°, and shoul-ders abducted 90° in the scapular plane.From this position the arms slide up thewall in the scapular plane, while leaninginto the wall. Interestingly, the wall slideproduce similar serratus anterior activitycompared to scapular abduction above120° abduction with no resistance. Oneadvantage of the wall slide compared toscapular abduction is that, anecdotally,patients report that the wall slide is lesspainful to perform.29 This may be be-cause during the wall slide the upper ex-tremities are supported against the wall,making it easier to perform while also as-sisting with compression of the humeralhead within the glenoid. Thus, this maybe an effective exercise to perform dur-ing the earlier protective phases of somerehabilitation programs.

General functions of the trapezius includescapular upward rotation and elevationfor the upper trapezius, retraction for themiddle trapezius, and upward rotationand depression for the lower trapezius.In addition, the inferomedial-directedfibers of the lower trapezius may alsocontribute to posterior tilt and externalrotation of the scapula during arm eleva-tion,52 which decreases subacromial im-pingement risk24,51 and makes the lowertrapezius an important area of focus inrehabilitation. Relatively high uppertrapezius activity occurs in the shouldershrug, prone rowing, prone horizontalabduction at 90° and 135° of abductionwith ER and IR, D1 diagonal PNF pat-

tern flexion, standing scapular dynamichug, PNF scapular clock, military press,2-hand overhead medicine ball throw,and scapular abduction and abductionbelow 80°, at 90°, and above 120° withER.13,16,20,62,75 During scapular abduction,upper trapezius activity progressively in-creases from 0° to 60°, remains relativelyconstant from 60° to 120°, and contin-ues to progressively increase from 120°to 180°.2

Relatively high middle trapezius ac-tivity occurs with shoulder shrug, pronerowing, and prone horizontal abductionat 90° and 135° abduction with ER andIR.20,62 Some authors have reported rela-tively high middle trapezius activity dur-ing scapular abduction at 90° and above120°,2,16,20 while authors of another studyshowed low EMG signal amplitude of themiddle trapezius during this exercise.62

Relatively high lower trapezius activityoccurs in the prone rowing, prone hori-zontal abduction at 90° and 135° abduc-tion with ER and IR, prone and standingER at 90° abduction, D2 diagonal PNFpattern flexion and extension, PNF scap-ular clock, standing high scapular rows,and scapular abduction, flexion, and ab-duction below 80° and above 120° withER.20,62,63,75 Lower trapezius activity tendsto be relatively low at angles less than 90°of scapular abduction, abduction, andflexion, and then increases exponentially

from 90° to 180°.2,20,29,62,75,84 Significantlygreater lower trapezius activity has beenreported during the prone ER at 90° ab-duction exercise compared to the emptycan exercise.3 As previously mentioned,the lower trapezius is an extremely im-portant muscle in shoulder function dueto its role in scapular upward rotation,external rotation, and posterior tilt.

Ekstrom et al20 reported that the great-est EMG signal amplitude of the lowertrapezius occurred during the prone fullcan, prone ER at 90°, and prone horizon-tal abduction at 90° with ER exercises.Based on these results, it appears thatthe prone full can exercise should not beperformed at a set degree of abduction,but should be individualized based on thealignment of the lower trapezius fibers( ). In the authors’ experience, thisis typically around 120° of abduction butmay fluctuate, depending on the specificpatient and body type.

It is often clinically beneficial to en-hance the ratio of lower trapezius-to-up-per trapezius strength.11 In the opinion ofthe authors, poor posture and muscle im-balance often seen in patients with a va-riety of shoulder pathologies is often theresult of poor muscle balance betweenthe upper and lower trapezius, with theupper trapezius being more dominant.McCabe et al56 report that bilateral ERat 0° abduction resulted in the greatestlower trapezius-upper trapezius ratiocompared to several other similar trape-zius exercises ( ). Cools et al11 alsoidentified side-lying ER and prone hori-zontal abduction at 90° abduction andER as 2 beneficial exercises to enhancethe ratio of lower trapezius to upper tra-pezius activity.

Both the rhomboids and levator scap-ulae function as scapular retractors,downward rotators, and elevators. Ex-ercises used to strengthen rotator cuffand scapulothoracic musculature arealso effective in eliciting activity of therhomboids and levator scapulae. Rela-tively high rhomboid activity has been

The proper alignment of the upperextremity during the prone horizontal abductionexercise with external rotation. Note how the upperextremity is aligned with the muscle fiber orientationof the lower trapezius.



reported during D2 diagonal PNF pat-tern flexion and extension, standingshoulder ER at 0° and 90° abduction,standing shoulder IR at 90° abduction,standing shoulder extension from 90°to 0°, prone shoulder horizontal abduc-tion at 90° abduction with IR, scapularabduction, abduction, and shoulder flex-ion above 120° with ER, prone rowing,and standing high, mid, and low scapu-lar rows.62,63 Relatively high rhomboidsand levator scapulae activity has beenreported with scapular abduction above120° with ER, prone horizontal abduc-tion at 90° abduction with ER and IR,prone rowing, and prone extension at90° flexion.62 Therefore, the prone ex-tension exercise may be performed inaddition to many of the previously men-tioned exercises for other scapulotho-racic muscles. Other specific exercisesto activate the rhomboids and levatorscapulae muscles are not often neces-sary to perform.

The preceding review can be usedto identify appropriate rehabilitationexercises for specific muscles. Based

on the reported studies and the collectiveexperience of the authors, we recommendthat exercises should be selected based onthe appropriate anatomical, biomechani-cal, and clinical implications. We haveidentified a set of exercises that the cur-rent authors use clinically for rehabilita-tion and injury prevention (TABLE). Theseexercises have been selected based on theresults of the numerous studies previous-ly cited and take into consideration theseimplications for each exercise described.Furthermore, the authors encourage theclinician to carefully consider emphasiz-ing posture and scapular retraction dur-ing the performance of glenohumeral andscapulothoracic exercises.

A common recommendation in reha-bilitation is to limit the amount of weightused during glenohumeral and scapu-lothoracic exercises to assure that the ap-propriate muscles are being utilized andnot larger compensatory muscles. Tworecent studies have analyzed this theoryand appear to prove the recommenda-tion inaccurate and not necessary. Alpertet al7 studied the rotator cuff and deltoidmuscles during scapular plane elevationand noted that EMG signal amplitudeof the smaller rotator cuff muscles andlarger deltoid muscles increased linearlyin relation to the amount of weight used.This finding is consistent with that ofDark et al,14 who showed similar resultsfor the rotator cuff, deltoid, pectoralis,and latissimus dorsi during ER and IRat 0° abduction. Thus, it appears thatlarger muscle groups do not overpowersmaller groups, such as the rotator cuff.Weight selection should be based on theindividual goals and performance of eachpatient. It does not appear necessary tolimit the amount of weight performedduring these rotator cuff exercises.

As our understanding of the anatomi-cal and biomechanical implications asso-ciated with exercise selection continues

to grow, we are seeing advances in exer-cise selection and the integration of thewhole-body kinetic-chain approach tostrengthening and rehabilitating injuries.This may involve strengthening multiplejoints simultaneously and during move-ment patterns that mimic athletic andfunctional daily activities of living. Theauthors often employ these techniqueswhen our patients improve in strengthyet continue to have symptoms duringactivities. In addition, we often attemptto further challenge our patients by per-forming many of the recommended exer-cise on various unstable surfaces (such asfoam or physioballs), with altered basesof support (such as sitting, standing, orsingle-leg balancing), in an attempt torecruit whole-body muscle patterns thatinteract together to perform active rangeof motion while stabilizing other areas ofthe body. We believe that these conceptsare important to consider in addition tostraight-plane, isolated movements ofspecific muscle groups, and that strength,posture, balance, and neuromuscularcontrol are all vital components to anyinjury prevention of rehabilitation pro-gram. Future research on the validity ofthese techniques is needed to justify theiruse. We believe that this is the next stepin the evolution of research on the clini-cal and biomechanical implications ofexercise selection for the glenohumeraland scapulothoracic musculature.

Athorough understanding ofthe biomechanical factors as-sociated with normal shoulder

movement, as well as during commonlyperformed exercises, is necessary tosafely and effectively design appropriateprograms. We have reviewed the normalbiomechanics of the glenohumeral andscapulothoracic muscles during func-tional activities, common exercises, andin the presence of pathology. These find-ings can be used by the clinician to designappropriate rehabilitation and injuryprevention programs.

Bilateral external rotation for infraspinatusand lower trapezius strengthening involves graspingexercise tubing with both hands and externallyrotating. Emphasis should be placed on providingscapular retraction and posterior tilting.


TABLERecommended Exercises for Glenohumeral and Scapulothoracic Muscles

Based on Anatomical, Biomechanical, and Clinical Implications

Abbreviations: EMG, electromyography; ER, external rotation; IR, internal rotation.

Supraspinatus 1. Full can 1. Enhances scapular position andsubacromial space

1. Decreased deltoid involvementcompared to empty can

1. Minimizes chance of superior humeral head migration bydeltoid overpowering supraspinatus

2. Prone full can 2. Enhances scapular position andsubacromial space

2. High posterior deltoid activitywith similar supraspinatus activity

2. High supraspinatus activity and also good exercise forlower trapezius

Infraspinatusand teresminor

1. Side-lying ER 1. Position of shoulder stability,minimal capsular strain

1. Increased moment arm ofmuscle at 0° abduction.Greatest EMG activity

1. Most effective exercise in recruiting infraspinatus activity.Good when cautious with static stability

2. Prone ER at 90°abduction

2. Challenging position for stability,higher capsular strain

2. High EMG activity 2. Strengthens in a challenging position for shoulder stability.Also good exercise for lower trapezius

3. ER with towel roll 3. Allows for proper form withoutcompensation

3. Increased EMG activity withaddition of towel, also incorpo-rates adductors

3. Enhances muscle recruitment and synergy with adductors

Subscapularis 1. IR at 0° abduction 1. Position of shoulder stability 1. Similar subscapularis activitybetween 0° and 90° abduction

1. Effective exercise, good when cautious with static stability

2. IR at 90° abduction 2. Position of shoulder instability 2. Enhances scapular position andsubacromial space. Lesspectoralis activity

2. Strengthens in a challenging position for shoulder stability

3. IR diagonal exercise 3. Replicates more functional activity 3. High EMG activity 3. Effective strengthening in a functional movement pattern

Serratus anterior 1. Push-up with plus 1. Easy position to produceresistance against protraction

1. High EMG activity 1. Effective exercise to provide resistance against protraction,also good exercise for subscapularis

2. Dynamic hug 2. Performed below 90° abduction 2. High EMG activity 2. Easily perform in patients with difficulty elevating arms orperforming push-up. Also good exercise for subscapularis

3. Serratus punch 120° 3. Combines protraction withupward rotation

3. High EMG activity 3. Good dynamic activity to combine upward rotation andprotraction function

Lower trapezius 1. Prone full can 1. Can properly align exercise withmuscle fibers

1. High EMG activity 1. Effective exercise, also good exercise for supraspinatus

2. Prone ER at 90°abduction

2. Prone exercise below 90°abduction

2. High EMG activity 2. Effective exercise, also good exercise for infraspinatus andteres minor

3. Prone horizontalabduction at 90°abduction with ER


3. Good ratio of lower to uppertrapezius activity

3. Effective exercise, also good exercise for middle trapezius

4. Bilateral ER 4. Scapular control without armelevation

4. Good ratio of lower to uppertrapezius activity

4. Effective exercise, also good for infraspinatus and teres minor

Middle trapezius 1. Prone row 1. Prone exercise below 90°abduction

1. High EMG activity 1. Effective exercise, good ratios of upper, middle, and lowertrapezius activity



2. High EMG activity 2. Effective exercise, also good exercise for lower trapezius

Upper trapezius 1. Shrug 1. Scapular control without armelevation

1. High EMG activity 1. Effective exercise

2. Prone row 2. Prone exercise below 90°abduction

2. High EMG activity 2. Good ratios of upper, middle, and lower trapezius activity



3. High EMG activity 3. Effective exercise, also good exercise for lower trapezius

Rhomboids andlevator scapulae

1. Prone row 1. Prone exercise below 90°abduction

1. High EMG activity 1. Effective exercise, good ratios of upper, middle, and lowertrapezius activity



2. High EMG activity 2. Effective exercise, also good for lower and middle trapezius

3. Prone extension with ER 3. Prone exercise below 90° abduction 3. High EMG activity 3. Effective exercise, unique movement to enhance scapular control



1. Alpert SW, Pink MM, Jobe FW, McMahon PJ, Ma-thiyakom W. Electromyographic analysis of deltoidand rotator cuff function under varying loads andspeeds. J Shoulder Elbow Surg. 2000;9:47-58.

2. Bagg SD, Forrest WJ. Electromyographic study ofthe scapular rotators during arm abduction in thescapular plane. Am J Phys Med. 1986;65:111-124.

3. Ballantyne BT, O’Hare SJ, Paschall JL, Et al.Electromyographic activity of selected shouldermuscles in commonly used therapeutic exercises.Phys Ther. 1993;73:668-682.

4. Barnett C, Kippers V, Turner P. Effects of variationsof the pench press exercise on the EMG activityof five shoulder muscles. J Strength Cond Res.1995;9:222-227.Blackburn TA, McLeod WD, White B, Wofford L.EMG analysis of posterior rotator cuff exercises.Athl Train J Natl Athl Train Assoc. 1990;25:40-45.Brossmann J, Preidler KW, Pedowitz RA, WhiteLM, Trudell D, Resnick D. Shoulder impingementsyndrome: influence of shoulder position on rota-tor cuff impingement—an anatomic study. AJRAm J Roentgenol. 1996;167:1511-1515.Brostrom LA, Kronberg M, Nemeth G. Muscleactivity during shoulder dislocation. Acta OrthopScand. 1989;60:639-641.

8. Burke WS, Vangsness CT, Powers CM. Strengthen-ing the supraspinatus: a clinical and biomechani-cal review. Clin Orthop Relat Res. 2002;292-298.

9. Cain PR, Mutschler TA, Fu FH, Lee SK. Anteriorstability of the glenohumeral joint. A dynamicmodel. Am J Sports Med. 1987;15:144-148.

10. Cools AM, Declercq GA, Cambier DC, Mahieu NN,Witvrouw EE. Trapezius activity and intramuscularbalance during isokinetic exercise in overheadathletes with impingement symptoms. ScandJ Med Sci Sports. 2007;17:25-33. http://dx.doi.org/10.1111/j.1600-0838.2006.00570.x

11. Cools AM, Dewitte V, Lanszweert F, et al.Rehabilitation of scapular muscle balance:which exercises to prescribe? Am J SportsMed. 2007;35:1744-1751. http://dx.doi.org/10.1177/0363546507303560

12. Cools AM, Witvrouw EE, Declercq GA, DanneelsLA, Cambier DC. Scapular muscle recruitmentpatterns: trapezius muscle latency with and with-out impingement symptoms. Am J Sports Med.2003;31:542-549.

13. Cordasco FA, Wolfe IN, Wootten ME, Bigliani LU.An electromyographic analysis of the shoulderduring a medicine ball rehabilitation program. AmJ Sports Med. 1996;24:386-392.

14. Dark A, Ginn KA, Halaki M. Shoulder musclerecruitment patterns during commonly used rota-tor cuff exercises: an electromyographic study.Phys Ther. 2007;87:1039-1046. http://dx.doi.org/10.2522/ptj.20060068De Wilde L, Plasschaert F, Berghs B, Van HoeckeM, Verstraete K, Verdonk R. Quantified measure-ment of subacromial impingement. J ShoulderElbow Surg. 2003;12:346-349. http://dx.doi.

org/10.1016/mse.2003.S1058274603000387Decker MJ, Hintermeister RA, Faber KJ, HawkinsRJ. Serratus anterior muscle activity during se-lected rehabilitation exercises. Am J Sports Med.1999;27:784-791.Decker MJ, Tokish JM, Ellis HB, Torry MR, HawkinsRJ. Subscapularis muscle activity during selectedrehabilitation exercises. Am J Sports Med.2003;31:126-134.

18. Ebaugh DD, McClure PW, Karduna AR. Scapu-lothoracic and glenohumeral kinematics followingan external rotation fatigue protocol. J OrthopSports Phys Ther. 2006;36:557-571. http://dx.doi.org/10.2519/jospt.2006.2189

19. Ekholm J, Arborelius UP, Hillered L, Ortqvist A.Shoulder muscle EMG and resisting momentduring diagonal exercise movements resisted byweight-and-pulley-circuit. Scand J Rehabil Med.1978;10:179-185.

20. Ekstrom RA, Donatelli RA, Soderberg GL. Surfaceelectromyographic analysis of exercises for thetrapezius and serratus anterior muscles. J OrthopSports Phys Ther. 2003;33:247-258.

21. Flatow EL, Soslowsky LJ, Ticker JB, et al. Excur-sion of the rotator cuff under the acromion. Pat-terns of subacromial contact. Am J Sports Med.1994;22:779-788.

22. Glousman R, Jobe F, Tibone J, Moynes D, Antonel-li D, Perry J. Dynamic electromyographic analysisof the throwing shoulder with glenohumeral insta-bility. J Bone Joint Surg Am. 1988;70:220-226.

23. Graichen H, Bonel H, Stammberger T, EnglmeierKH, Reiser M, Eckstein F. Subacromial spacewidth changes during abduction and rotation--a 3-D MR imaging study. Surg Radiol Anat.1999;21:59-64.

24. Graichen H, Bonel H, Stammberger T, et al. Three-dimensional analysis of the width of the subacro-mial space in healthy subjects and patients withimpingement syndrome. AJR Am J Roentgenol.1999;172:1081-1086.Graichen H, Hinterwimmer S, von Eisenhart-Rothe R, Vogl T, Englmeier KH, Eckstein F. Effectof abducting and adducting muscle activity onglenohumeral translation, scapular kinematicsand subacromial space width in vivo. J Biomech.2005;38:755-760. http://dx.doi.org/10.1016/j.jbiomech.2004.05.020Graichen H, Stammberger T, Bonel H, Karl-HansE, Reiser M, Eckstein F. Glenohumeral transla-tion during active and passive elevation of theshoulder: a 3D open-MRI study. J Biomech.2000;33:609-613.Greenfield BH, Donatelli R, Wooden MJ, WilkesJ. Isokinetic evaluation of shoulder rotationalstrength between the plane of scapula and thefrontal plane. Am J Sports Med. 1990;18:124-128.

28. Greis PE, Kuhn JE, Schultheis J, Hintermeister R,Hawkins R. Validation of the lift-off test and analy-sis of subscapularis activity during maximal inter-nal rotation. Am J Sports Med. 1996;24:589-593.

29. Hardwick DH, Beebe JA, McDonnell MK, LangCE. A comparison of serratus anterior muscleactivation during a wall slide exercise and othertraditional exercises. J Orthop Sports Phys Ther.

2006;36:903-910. http://dx.doi.org/10.2519/jospt.2006.2306

30. Harryman DT, 2nd, Sidles JA, Clark JM, McQuadeKJ, Gibb TD, Matsen FA, 3rd. Translation ofthe humeral head on the glenoid with passiveglenohumeral motion. J Bone Joint Surg Am.1990;72:1334-1343.

31. Hess SA, Richardson C, Darnell R, Friis P, Lisle D,Myers P. Timing of rotator cuff activation duringshoulder external rotation in throwers with andwithout symptoms of pain. J Orthop Sports PhysTher. 2005;35:812-820. http://dx.doi.org/10.2519/jospt.2005.2134

32. Hintermeister RA, Lange GW, Schultheis JM, BeyMJ, Hawkins RJ. Electromyographic activity andapplied load during shoulder rehabilitation exer-cises using elastic resistance. Am J Sports Med.1998;26:210-220.

33. Horrigan JM, Shellock FG, Mink JH, DeutschAL. Magnetic resonance imaging evaluation ofmuscle usage associated with three exercises forrotator cuff rehabilitation. Med Sci Sports Exerc.1999;31:1361-1366.

34. Hughes RE, An KN. Force analysis of rotator cuffmuscles. Clin Orthop Relat Res. 1996;75-83.

Illyes A, Kiss RM. Electromyographic analysis in pa-tients with multidirectional shoulder instability duringpull, forward punch, elevation and overhead throw.Knee Surg Sports Traumatol Arthrosc. 2007;15:624-631. http://dx.doi.org/10.1007/s00167-006-0163-1Itoi E, Berglund LJ, Grabowski JJ, et al. Tensileproperties of the supraspinatus tendon. J OrthopRes. 1995;13:578-584. http://dx.doi.org/10.1002/jor.1100130413Itoi E, Kido T, Sano A, Urayama M, Sato K. Whichis more useful, the “full can test” or the “emptycan test,” in detecting the torn supraspinatustendon? Am J Sports Med. 1999;27:65-68.

38. Jobe FW, Moynes DR. Delineation of diagnosticcriteria and a rehabilitation program for rotatorcuff injuries. Am J Sports Med. 1982;10:336-339.

39. Kadaba MP, Cole A, Wootten ME, et al. Intramus-cular wire electromyography of the subscapularis.J Orthop Res. 1992;10:394-397. http://dx.doi.org/10.1002/jor.1100100312

40. Kelly BT, Kadrmas WR, Speer KP. The manualmuscle examination for rotator cuff strength. Anelectromyographic investigation. Am J SportsMed. 1996;24:581-588.

41. Kibler WB, Sciascia A, Dome D. Evaluationof apparent and absolute supraspinatusstrength in patients with shoulder injury us-ing the scapular retraction test. Am J SportsMed. 2006;34:1643-1647. http://dx.doi.org/10.1177/0363546506288728

42. Konrad GG, Jolly JT, Labriola JE, McMahon PJ,Debski RE. Thoracohumeral muscle activityalters glenohumeral joint biomechanics duringactive abduction. J Orthop Res. 2006;24:748-756.http://dx.doi.org/10.1002/jor.20062

43. Kronberg M, Brostrom LA, Nemeth G. Differencesin shoulder muscle activity between patients withgeneralized joint laxity and normal controls. ClinOrthop Relat Res. 1991;181-192.

44. Kronberg M, Nemeth G, Brostrom LA. Muscle ac-


@ WWW.JOSPT.ORG

J. Comparative electromyographic analysis ofshoulder muscles during planar motions: anteriorglenohumeral instability versus normal. J Shoul-der Elbow Surg. 1996;5:118-123.Meskers CG, van der Helm FC, Rozing PM. Thesize of the supraspinatus outlet during elevationof the arm in the frontal and sagittal plane: a3-D model study. Clin Biomech (Bristol, Avon).2002;17:257-266.

Morrey BF, Itoi E, An KN. Biomechanics of the shoul-der. In: Rockwood CA, Matsen FA, 3rd, eds. TheShoulder. Philadelphia: Saunders; 1998:233-276.Moseley JB, Jr., Jobe FW, Pink M, Perry J, TiboneJ. EMG analysis of the scapular muscles duringa shoulder rehabilitation program. Am J SportsMed. 1992;20:128-134.Myers JB, Pasquale MR, Laudner KG, Sell TC,Bradley JP, Lephart SM. On-the-field resistance-tubing exercises for throwers: an electromyo-graphic analysis. J Athl Train. 2005;40:15-22.Ogston JB, Ludewig PM. Differences in 3-dimen-sional shoulder kinematics between persons withmultidirectional instability and asymptomaticcontrols. Am J Sports Med. 2007;35:1361-1370.http://dx.doi.org/10.1177/0363546507300820

Otis JC, Jiang CC, Wickiewicz TL, Peterson MG, War-ren RF, Santner TJ. Changes in the moment arms ofthe rotator cuff and deltoid muscles with abductionand rotation. J Bone Joint Surg Am. 1994;76:667-676.Poppen NK, Walker PS. Forces at the gle-nohumeral joint in abduction. Clin Orthop RelatRes. 1978;165-170.Poppen NK, Walker PS. Normal and abnormalmotion of the shoulder. J Bone Joint Surg Am.1976;58:195-201.Reddy AS, Mohr KJ, Pink MM, Jobe FW. Electro-myographic analysis of the deltoid and rotator cuffmuscles in persons with subacromial impinge-ment. J Shoulder Elbow Surg. 2000;9:519-523.Reinold MM, Macrina LC, Wilk KE, et al. Electro-myographic analysis of the supraspinatus anddeltoid muscles during 3 common rehabilitationexercises. J Athl Train. 2007;42:464-469.Reinold MM, Wilk KE, Fleisig GS, et al. Elec-tromyographic analysis of the rotator cuff anddeltoid musculature during common shoulderexternal rotation exercises. J Orthop SportsPhys Ther. 2004;34:385-394. http://dx.doi.org/10.2519/jospt.2004.0665Roberts CS, Davila JN, Hushek SG, Tillett ED, Cor-rigan TM. Magnetic resonance imaging analysisof the subacromial space in the impingement signpositions. J Shoulder Elbow Surg. 2002;11:595-599. http://dx.doi.org/10.1067/mse.2002.127095Santos MJ, Belangero WD, Almeida GL. The effectof joint instability on latency and recruitment or-der of the shoulder muscles. J Electromyogr Kine-siol. 2007;17:167-175. http://dx.doi.org/10.1016/j.jelekin.2006.01.010Scovazzo ML, Browne A, Pink M, Jobe FW, Kerri-gan J. The painful shoulder during freestyle swim-ming. An electromyographic cinematographicanalysis of twelve muscles. Am J Sports Med.1991;19:577-582.Sharkey NA, Marder RA. The rotator cuff opposes

tivity and coordination in the normal shoulder. Anelectromyographic study. Clin Orthop Relat Res.1990;76-85.Labriola JE, Jolly JT, McMahon PJ, Debski RE. Ac-tive stability of the glenohumeral joint decreasesin the apprehension position. Clin Biomech(Bristol, Avon). 2004;19:801-809. http://dx.doi.org/10.1016/j.clinbiomech.2004.05.008Labriola JE, Lee TQ, Debski RE, McMahon PJ. Sta-bility and instability of the glenohumeral joint: therole of shoulder muscles. J Shoulder Elbow Surg.2005;14:32S-38S. http://dx.doi.org/10.1016/j.jse.2004.09.014Lear LJ, Gross MT. An electromyographical analy-sis of the scapular stabilizing synergists during apush-up progression. J Orthop Sports Phys Ther.1998;28:146-157.

48. Lee SB, An KN. Dynamic glenohumeral stabilityprovided by three heads of the deltoid muscle.Clin Orthop Relat Res. 2002;40-47.

49. Lee SB, Kim KJ, O’Driscoll SW, Morrey BF, AnKN. Dynamic glenohumeral stability providedby the rotator cuff muscles in the mid-range andend-range of motion. A study in cadavera. J BoneJoint Surg Am. 2000;82:849-857.Liu J, Hughes RE, Smutz WP, Niebur G, Nan-AnK. Roles of deltoid and rotator cuff muscles inshoulder elevation. Clin Biomech (Bristol, Avon).1997;12:32-38.Ludewig PM, Cook TM. Alterations in shoulderkinematics and associated muscle activity inpeople with symptoms of shoulder impingement.Phys Ther. 2000;80:276-291.Ludewig PM, Cook TM, Nawoczenski DA. Three-dimensional scapular orientation and muscle ac-tivity at selected positions of humeral elevation. JOrthop Sports Phys Ther. 1996;24:57-65.Ludewig PM, Hoff MS, Osowski EE, MeschkeSA, Rundquist PJ. Relative balance of serratusanterior and upper trapezius muscle activityduring push-up exercises. Am J Sports Med.2004;32:484-493.Malanga GA, Jenp YN, Growney ES, An KN. EMGanalysis of shoulder positioning in testing andstrengthening the supraspinatus. Med Sci SportsExerc. 1996;28:661-664.Matias R, Pascoal AG. The unstable shoulder inarm elevation: a three-dimensional and electro-myographic study in subjects with glenohumeralinstability. Clin Biomech (Bristol, Avon). 2006;21Suppl 1:S52-58. http://dx.doi.org/10.1016/j.clinbiomech.2005.09.014

McCabe RA. Surface electromyographic analysis ofthe lower trapezius muscle during exercises performedbelow ninety degrees of shoulder elevation in healthysubjects. N Am J Sports Phys Ther. 2007;2:34-43.McCaw ST, Friday JJ. A comparison of muscleactivity between a free weight and machine benchpress. J Strength Cond Res. 1994;8:259-264.McClure PW, Michener LA, Sennett BJ, KardunaAR. Direct 3-dimensional measurement ofscapular kinematics during dynamic movementsin vivo. J Shoulder Elbow Surg. 2001;10:269-277.http://dx.doi.org/10.1067/mse.2001.112954McMahon PJ, Jobe FW, Pink MM, Brault JR, Perry

superior translation of the humeral head. Am JSports Med. 1995;23:270-275.Smith J, Dahm DL, Kaufman KR, et al. Electro-myographic activity in the immobilized shouldergirdle musculature during scapulothoracic exer-cises. Arch Phys Med Rehabil. 2006;87:923-927.http://dx.doi.org/10.1016/j.apmr.2006.03.013Solem-Bertoft E, Thuomas KA, Westerberg CE.The influence of scapular retraction and protrac-tion on the width of the subacromial space. AnMRI study. Clin Orthop Relat Res. 1993;99-103.Takeda Y, Kashiwaguchi S, Endo K, Matsuura T, SasaT. The most effective exercise for strengthening thesupraspinatus muscle: evaluation by magnetic reso-nance imaging. Am J Sports Med. 2002;30:374-381.Thigpen CA, Padua DA, Morgan N, Kreps C,Karas SG. Scapular kinematics during supraspi-natus rehabilitation exercise: a comparison offull-can versus empty-can techniques. Am JSports Med. 2006;34:644-652. http://dx.doi.org/10.1177/0363546505281797Townsend H, Jobe FW, Pink M, Perry J. Elec-tromyographic analysis of the glenohumeralmuscles during a baseball rehabilitation program.Am J Sports Med. 1991;19:264-272.

80. Tsai NT, McClure PW, Karduna AR. Effects of mus-cle fatigue on 3-dimensional scapular kinematics.Arch Phys Med Rehabil. 2003;84:1000-1005.

81. Uhl TL, Carver TJ, Mattacola CG, Mair SD, Nitz AJ.Shoulder musculature activation during upper ex-tremity weight-bearing exercise. J Orthop SportsPhys Ther. 2003;33:109-117.

82. Warner JJ, Micheli LJ, Arslanian LE, KennedyJ, Kennedy R. Patterns of flexibility, laxity, andstrength in normal shoulders and shoulders withinstability and impingement. Am J Sports Med.1990;18:366-375.

83. Welsch EA, Bird M, Mayhew JL. Electromyograph-ic activity of the pectoralis major and anteriordeltoid muscles during three upper-body lifts.J Strength Cond Res. 2005;19:449-452. http://dx.doi.org/10.1519/14513.1

84. Wiedenbauer MM, Mortensen OA. An electromyo-graphic study of the trapezius muscle. Am J PhysMed. 1952;31:363-372.Wilk KE, Andrews JR, Arrigo CA. The physical ex-amination of the glenohumeral joint: emphasis onthe stabilizing structures. J Orthop Sports PhysTher. 1997;25:380-389.Wilk KE, Arrigo CA, Andrews JR. Currentconcepts: the stabilizing structures of theglenohumeral joint. J Orthop Sports Phys Ther.1997;25:364-379.Worrell TW, Corey BJ, York SL, Santiestaban J. Ananalysis of supraspinatus EMG activity and shoul-der isometric force development. Med Sci SportsExerc. 1992;24:744-748.

88. Wuelker N, Korell M, Thren K. Dynamic gle-nohumeral joint stability. J Shoulder Elbow Surg.1998;7:43-52.

ejercicios basados en la evidencia escapulotoracicos y glh

Documents

to90 of abduction

to30 of abduction

rotator cuff pathology

duringscapular plane

active scapularplane

rotator cuff muscles

division of sports medicine

champion sports medicine