EFFECT OF SEATED THORACIC MANIPULATION ON

CHANGES IN SCAPULAR KINEMATICS AND SCAPULOHUMERAL

RHYTHM IN YOUNG ASYMPTOMATIC PARTICIPANTS:A RANDOMIZED STUDY

Dayana P. Rosa, PT, a Francisco Alburquerque-Sendín, PhD,b Tania F. Salvini, PhD,c andPaula R. Camargo, PT, PhDc

a Physical Thethodist Univeb Professor, Dalamanca, Salamc Professor, D

ity of São CarloSubmit reque

rofessor, Departão Carlos, Rodoarlos, SP, BraziPaper submitt

013; accepted J0161-4754/$3Copyright ©http://dx.doi.o

ABSTRACT

Objective: The purpose of this study was to evaluate the immediate effects of seated thoracic manipulation onscapulothoracic kinematics and scapulohumeral rhythm during arm flexion in young asymptomatic participants.Methods: A convenience sample of 42 young asymptomatic participants was randomly divided in 2 groups:manipulation and sham group. Measurements were taken before and after the intervention. All participants completedthe Disabilities of the Arm, Shoulder, and Hand questionnaire to assess pain and physical function. The manipulationgroup received the manipulation (high velocity, low amplitude), which was performed by a physical therapist with thepatient in the seated position and with the arms crossed over the chest and hands passed over the shoulders. For thesham group, the same procedure was performed, with the exception that the high-velocity thrust was not applied.Three-dimensional (3D) kinematic data were collected with the participants in a relaxed standing position using a 3Delectromagnetic tracking system. All participants performed 3 repetitions of arm flexion before and after manipulation.Results: There were no differences (P = .79) in Disabilities of the Arm, Shoulder, and Hand scores when themanipulation (3.37 ± 3.72) was compared with the sham group (3.68 ± 4.27). The 3-way analysis of variance showedno significant interaction among group, angle, and time differences for the outcomes (scapulothoracic internal/externalrotation [F = 0.43; P = .82], upward/downward rotation [F = 0.08; P = .99], tilt [F = 0.23; P = .94], andscapulohumeral rhythm [F = 4; P = .86]). The intragroup effect was small for the outcomes measured in both groups.Conclusions: Thoracic manipulation in the seated position did not affect scapulohumeral rhythm and 3D scapularkinematics during arm flexion in young asymptomatic participants. (J Manipulative Physiol Ther 2013;xx:1-9)

Key Indexing Terms: Rehabilitation; Shoulder; Spine; Scapula; Manipulation; Musculoskeletal Manipulations

Thoracic manipulation is defined as a high-velocity/low-amplitude movement or “thrust” directed at anysegment of the thoracic spine1 and has been widely

used to treat different musculoskeletal conditions such asneck, shoulder, and back pain.1-3 Recent studies have eval-

erapist, Physical Therapy Graduate Program,rsity of Piracicaba, Piracicaba, SP, Brazil.epartment of Physical Therapy, University ofanca, Spain.

epartment of Physical Therapy, Federal Univer-s, São Carlos, SP, Brazil.sts for reprints to: Paula R. Camargo, PT, PhD,ment of Physical Therapy, Federal University ofvia Washington Luis, km 235, 13565-905 Sãol (e-mail: paularezendecamargo@gmail.com).ed September 5, 2012; in revised form July 23,uly 25, 2013.6.002013 by National University of Health Sciences.rg/10.1016/j.jmpt.2013.07.006

M

S

s

PSC

2

uated the immediate effects of spinal manipulation in asymp-tomatic participants showing no consensus on the results.4-9

These studies bring different aspects about possible effects ofspinalmanipulation: reduction of the α-motoneuronal activity(muscle tone) with no alteration of the pressure painthresholds9 and increase of phasic perineal contraction andbasal perineal tonus5 after manipulation at the sacroiliac joint,changes in the nocioceptive afferent system7 and in pressurepain threshold4 after thoracic manipulation, and no changesin spine stiffness6 and in lumbar range of motion8 afterthoracic and lumbar manipulation, respectively. The absenceof consensus in the literature may be caused by the hetero-geneity of the intervention and assessment procedures. It issuggested that more studies on the motor component shouldbe done.10,11

Theodoridis and Ruston12 have considered that the bio-mechanical relationship between the arm and the thoracicspine is important to clinical practice because of thecontribution of the thoracic spine to shoulder movements.This relationship leads us to the term regional

2 Journal of ManipulativeRosa et alMonth 2013Thoracic Manipulation and Scapular Kinematics

interdependence, which refers to the concept that seeminglyunrelated impairments in a remote anatomical region maycontribute to, or be associated with, the patient's primarycomplaint.13 The clinical rationale for using thoracicmanipulation to treat the shoulder in this example is basedon this theory. Nevertheless, the dimension of thisrelationship in specific samples of participants with andwithout shoulder pain is not known.

Investigations have shown that thoracic manipulationmay be effective in reducing pain and increasing rangeof motion of the shoulder.14-17 In addition, thoracicmanipulation in asymptomatic participants was demon-strated to increase the strength of the lower trapezius,18

which is strongly associated with the scapular motion.19

Because appropriate motion of the scapula duringelevation of the arm is one of the most importantaspects to be considered in shoulder dysfunctions, itwould be interesting to know if thoracic manipulationhas any effect on scapular kinematics and scapulohum-eral rhythm. There is only one study in the literaturethat evaluated the effects of thoracic manipulation onscapular kinematics and muscle activation in participantswith shoulder pain.17 The authors suggested thatthe decreased pain and improved function of theshoulder after the thoracic manipulation are not likelyassociated with changes in scapular kinematics orchanges in muscle activation. This finding shows thatdifferent pathways for sensory and motor changes mayexist in the shoulder after the application of thoracicmanipulation. However, the absence of a control groupand blinding were limitations from the previous study.

Mechanisms to explain how thoracic manipulation canbe associated with improved shoulder pain and motion arenot well defined. The consequences of applying a thoracicmanipulation to elicit changes in shoulder motion have notbeen adequately proven yet, neither in patients nor inhealthy participants. As such, there is a need of research thatevaluates the motor component and helps to better under-stand the mechanisms by which thoracic manipulationmight induce the changes usually described. Specifically,there is a lack of studies on thoracic manipulation and3-dimensional (3D) shoulder kinematics.

It is important to evaluate asymptomatic participants inan attempt to eliminate factors (pain, for example) that canhave influence on the results avoiding the confusion ofcause and effect and also adverse effects that might adviseagainst the application of the thoracic manipulation. Basedon the neurophysiologic effects of the spine manipulationalready demonstrated by some studies,20,21 it was hypoth-esized that the thoracic manipulation would increase sca-pular upward rotation and posterior tilt. Therefore, thepurpose of this study was to evaluate the immediate effectsof thoracic manipulation on 3D scapulothoracic kinematicsand scapulohumeral rhythm during arm flexion in youngasymptomatic participants.

METHODS

This was a controlled laboratory study of immediateeffects (before/after intervention) that took place at Uni-versidade Federal de São Carlos. Volunteers were recruitedfrom December 2009 to January 2011 at the University andcommunity. A convenience sample of 42 asymptomaticpatients participated in this study. They were randomlydivided into 2 groups: manipulation group and sham group.The last author was responsible for the randomization, whichwas performed by using www.randomization.com. Alloca-tion of each participant was only known by the time ofevaluation. Figure 1 shows the flow diagram of progress ofboth groups in the study. Table 1 shows the descriptive datafor both groups. All participants were recruited by directcontact and screened by the first author.

The participants were included if they had no history ofshoulder or cervical pathology and if they had range ofmotion for shoulder elevation next to 150° as evaluated byvisual observation. Exclusion criteria included the follow-ing: pregnancy, ligamentous laxity based on positiveSulcus22 and Apprehension23 tests; positive impingementtests based on Hawkins and Kennedy,24 Neer,25 and Jobeand Moynes,26 or pain during external rotation with thearm in 90° of elevation in the coronal plane; history ofclavicle, scapula, or humerus fracture; systemic illnesses;apprehension of being manipulated; tape allergy; and bodymass index greater than 28 kg/m2 (the amount of subcu-taneous tissue can compromise the quality of the motiondata because surface sensors are used to track the bones).

This project was approved by the Ethical Committeein Human Research of Universidade Federal de SãoCarlos (protocol number 331/2010). The participants gavetheir written and informed consent agreement to partic-ipate in this study, which was conducted according to theHelsinki Statement.

Disabilities of the Arm, Shoulder, and Hand QuestionnaireThe Disabilities of the Arm, Shoulder, and Hand

(DASH) self-assessment questionnaire contains 30 ques-tions to measure physical function and symptoms of theupper limbs of the participants. Each question has 5 pos-sible responses, ranging from “no difficulty” to “unable toperform activity,” and is scored on a 1- to 5-point ratingscale. The questionnaire score is calculated by applying anestablished formula in which the maximum score is 100,which indicates the worst possible condition.27 Thisquestionnaire has been shown to be a reliable, valid, andresponsive measure.28,29

Three-Dimensional Scapular KinematicsFor 3D scapular kinematics, data capture and analysis

were completed using Flock of Birds (miniBird; AscensionTechnology Corporation, Burlington, VT) hardware

Table 1. Descriptive data of both groups

Manipulation group (n = 21) Sham group (n = 21) P

Sex 11 women; 10 men 11 women; 10 men 1.00Age (y), mean ± SD 23.81 ± 3.75 23.95 ± 3.2 .89Body mass (kg), mean ± SD 68.31 ± 14.6 68.38 ± 11.23 .49Height (cm), mean ± SD 173 ± 1.0 171 ± 0.9 .63Evaluated side 13 dominants;

8 nondominants10 dominants; 11 nondominants .35

Assessed for eligibility (n = 49)

Excluded (n = 4)Not meeting inclusion criteria (n = 4)

Analysed (n=21)Excluded from analysis (no cavitation) (n = 3)

Allocated to manipulation group (n = 24)Received allocated intervention (n = 24)

Allocated to sham group (n = 21)Received allocated intervention (n = 21)

Analysed (n = 21)

Excluded from analysis (n = 0)

Randomized (n = 45)

Fig 1. Flow diagram of the progress of participants in the study.

3Rosa et alJournal of Manipulative and Physiological TherapeuticsScapular Kinematics and Scapulohumeral RhythmVolume xx, Number x

integrated with MotionMonitor (Ascension TechnologyCorporation) software. The Flock of Birds is a direct currentelectromagnetic tracking device able to locate multiplesensors relative to a source transmitter. The 3D position andorientation of each sensor can be tracked simultaneously atsampling rates of 30 to 144 Hz. The sensors are small andlightweight (1.8 × 0.8 × 0.8 cm). In a metal-freeenvironment up to a 76-cm distance from the transmitter,the root mean square accuracy of the system is 0.5° fororientation and 0.18 cm for position, as reported by themanufacturer. One of the sensors is attached to a stylus withknown offsets to digitize anatomical landmarks for buildingthe joint coordinate systems.

The electromagnetic sensors were attached with double-sided adhesive tape to the sternum, to the acromion of thescapula, and to a thermoplastic cuff secured to the distalhumerus to track humeral motion. These surface sensorplacements have been previously used.30-33 The participantstood with the arms relaxed at the side in a neutral positionwith the transmitter directly behind the shoulder testedwhile bony landmarks on the thorax, scapula, and humeruswere palpated and digitized to allow transformation of thesensor data to local anatomically based coordinate systems.Thorax landmarks included the sternal notch, C7 spinous

process, T8 spinous process, and xyphoid process. Scapularlandmarks included the root of the spine, posterolateralacromion, and the inferior angle of the scapula. Humeruslandmarks included the lateral and the medial epicondyles.The center of the humeral head was estimated by movingthe arm passively through short arcs (b45°) to define thepivot point.34

Data ReductionLocal coordinate systems were established for the trunk,

clavicle, scapula, and humerus using the digitized landmarksfollowing the International Society of Biomechanics–recommended protocol.35 The z-axis pointed laterally; thex-axis, anteriorly; and the y-axis, superiorly. The YXZsequence was used to describe scapular motions relativeto the trunk. For the scapula, the rotations were described inthe order of internal/external rotation, upward/downwardrotation, and anterior/posterior tilt. The humeral positionwith reference to the trunk was determined using the YX′Y″sequence. The first rotation defines the plane of elevation,the second defines the humeral elevation angle, and the thirddefines internal/external rotation. The humeral position withreference to the scapula was determined using the XZYsequence. The first rotation defines glenohumeral elevation,

4 Journal of ManipulativeRosa et alMonth 2013Thoracic Manipulation and Scapular Kinematics

the second defines the plane of glenohumeral elevation, andthe third defines internal/external rotation.

Scapulohumeral RhythmThe ratio of glenohumeral elevation relative to scapular

upward rotation was determined by calculating the slope ofthe linear regression line using scapular upward rotation asthe X value and glenohumeral elevation as the Y value, asproposed by Braman et al.36 The ratio was calculated from30° to 120° of humerothoracic elevation and at 30°increments (30°-60°, 60°-90°, 90°-120°).

Fig 2. Therapist and participant position for the manipulationprocedure. (Color version of figure is available online.)

InterventionsThe thoracic manipulation used in this study has been

previously described in other studies37-39 and was per-formed by a physical therapist with more than 10 years ofexperience. The participant was seated with the armscrossed over the chest and hands passed over the shoulders.The therapist placed his upper chest at the level of theparticipant's middle thoracic spine and grasped theparticipant's elbows (Fig 2). The participant was instructedto take a deep breath in and let it out while a gentle flexionof the thoracic spine was introduced until slight tensionwas felt in the tissues at the contact point between thetherapist's chest and participant's back. Then, a distractionthrust manipulation in a superior/posterior direction wasapplied.40 Cavitation was expected to occur during themanipulation procedure. If no popping was heard on thefirst attempt, the therapist repositioned the participant andperformed a second manipulation. A maximum of 2attempts were performed on each participant, as previouslydescribed.38 Three participants were excluded from thestudy because no cavitation occurred after the secondattempt. The sham group received the same procedure, andthe participant was positioned the same as the manipulationgroup, with the exception that the high-velocity thrustwas not performed in these participants. No participantsfrom both groups had any adverse events with the mani-pulation technique.

Participants were given incomplete information aboutthe purpose of the study to control any effects that theirexpectations about the results would cause. Same instruc-tions were given to participants in the manipulation andsham groups before the testing session. Examiner andtherapist were not the same person. The examiner wasresponsible for managing the computer to register thekinematic data, and the therapist applied the assigned inter-vention. Because the tool used to track the kinematics wasan electromagnetic system, the data were not collecteddirectly by the examiner but had to undergo computerprocessing, decreasing the examiner's ability to influencethe outcomes.

ProceduresKinematic data collection was done with the participants

in a relaxed standing position. Kinematic motion analysisinvolved selecting scapular data at humerothoracic eleva-tion angles of initial, 30°, 60°, 90°, 120°, and maximumelevation. Participants were asked to maintain light fin-gertip contact with a flat planar surface to keep positioningof the arm in the sagittal plane (Fig 3). They were alsoinstructed to keep their hand with their thumb pointingtoward the ceiling when tested at each humeral elevation.Three repetitions were performed before and immediatelyafter the intervention (manipulation or sham). Participantswere asked to elevate their arm from the rest positionthrough their full range of motion at a speed such that ittook around 3 seconds to elevate their arm and 3 secondsto lower it. All measurements were taken by the primaryinvestigator. The tested side, dominant or nondominant,was randomly chosen. The sensors were not removed orreplaced between repetitions, or for the intervention(manipulation or sham). This method has been previouslyestablished to have good repeatability within the sametesting session.41

Statistical AnalysisThe results were analyzed using the SPSS statistical

package (16.0 version; SPSS, Chicago, IL). The χ2 and theStudent t test were conducted to determine if the 2 groupsdiffered on the demographic characteristics. Normality test

Fig 3. Participant during kinematics evaluation. They were askedto maintain light fingertip contact with a flat planar surface tokeep positioning of the arm in the sagittal plane. (Color version offigure is available online.)

Fig 4. Mean and SE values for scapular internal rotation acrosshumerothoracic elevation angles for both manipulation and shamgroups before and after the intervention. (Color version of figureis available online.)

5Rosa et alJournal of Manipulative and Physiological TherapeuticsScapular Kinematics and Scapulohumeral RhythmVolume xx, Number x

(Kolmogorov-Smirnov) was conducted for the dependentvariables, which presented normal behavior (P N .05). Anindependent t test was used to analyze the DASH scoresbetween both groups. For each dependent variable (scapularinternal rotation, upward rotation, posterior tilt), a 3-wayrepeated-measures analysis of variance was used to test formain effects of group (manipulation and sham), time(preintervention and postintervention), and angle of armelevation (initial, 30°, 60°, 90°, 120°, and maximum), or forinteractions of time and group or of time, group and angle,separately. For scapulohumeral rhythm, a 3-way repeated-measures analysis of variance was also used to test maineffects of group, time and interval (30°-120°, 30°-60°, 60°-90°, 90°-120°), or for interactions. Tukey post hoc analyseswere performed to investigate significant interactions. AP value of less than .05 was considered significant.

Intragroup effect sizes for each scapular rotation acrosshumerothoracic elevation angles and for scapulohumeralrhythm at each interval were calculated using Cohen dcoefficient42 for both manipulation and sham groups. Aneffect size greater than 0.8 was considered large; around0.5, moderate; and less than 0.2, small.

RESULTS

The DASH QuestionnaireThere were no differences (P = .79) in DASH scores

when the manipulation group (3.37 ± 3.72) was compared

with the sham group (3.68 ± 4.27). Also, no adverse eventswere reported from the study participants.

Three-Dimensional Scapular KinematicsFor scapular internal rotation, there was no significant

interaction among group, angle, and time (P = .91, df = 5,F = 0.29) for elevation of the arm (Fig 4). There was also nointeraction between group and time during elevation (P =.06, df = 1, F = 3.59) of the arm. The average differencesbetween preintervention and postintervention for bothmanipulation and sham groups were 1.14° and 2.27°,respectively, during elevation of the arm.

For scapular upward rotation, there was no significantinteraction among group, angle, and time (P = .97, df = 5,F = 0.17) for elevation of the arm (Fig 5) and also betweengroup and time (P = .98, df = 1, F = 0.001). The averagedifferences between preintervention and postinterventionfor both manipulation and sham groups were 1.2° and 1.1°,respectively, during elevation of the arm.

For scapular posterior tilt, there was no significantinteraction among group, angle, and time (P = .97, df = 5,F = 0.16) for elevation of the arm (Fig 6). There was also nointeraction between group and time (P = .06, df = 1, F =3.55) during elevation of the arm. The average differencesbetween preintervention and postintervention for bothmanipulation and sham groups were 0.09° and 0.66°,respectively, during elevation of the arm.

The intragroup effect size for all scapular rotationsacross humerothoracic elevation angles for both groups wassmall (Table 2).

Fig 5. Mean and SE values for scapular upward rotation acrosshumerothoracic elevation angles for both manipulation and shamgroups before and after the intervention. (Color version of figureis available online.)

Fig 6. Mean and SE values for scapular posterior tilt acrosshumerothoracic elevation angles for both manipulation and shamgroups before and after the intervention. (Color version of figureis available online.)

6 Journal of ManipulativeRosa et alMonth 2013Thoracic Manipulation and Scapular Kinematics

Scapulohumeral RhythmThere was no significant interaction among group,

interval, and time (P = .86, df = 0.31, F = 4) and betweengroup and time (P = .25, df = 1, F = 1.32) for scapu-lohumeral rhythm during elevation of the arm. Table 3shows the glenohumeral/scapulothoracic ratios for armelevation for both manipulation and sham groups. Theintragroup effect size was small for all intervals evaluated(Table 4).

DISCUSSION

This study brings new information about the effects ofthoracic manipulation on scapular kinematics and scapulo-humeral rhythm. The results suggest that the seated thoracicmanipulation technique used in this study does not haveimmediate effect on scapular kinematics during elevationof the arm in young asymptomatic participants. The lackof effect was demonstrated by both inferential analysis andeffect size.

A recent investigation demonstrated a small but sig-nificant increase in middle trapezius muscle and a decreasein scapular upward rotation after the thoracic manipulationin participants with shoulder pain.17 However, this fact didnot explain the improvement in pain and function presentedby them because the change in kinematics and muscleactivation was not clinically important. This finding sup-ports a dichotomy between sensory and motor effects ofthe thoracic manipulation. Cleland et al18 have suggestedthat thoracic manipulation can increase lower trapezius

muscle strength in asymptomatic participants owing to theactivation of the mechanoreceptors around the manipulatedjoint. Considering the controversial results from the pre-vious studies and that the lower trapezius is an importantscapular upward rotator,19 we would expect that scapularkinematics could change after the manipulation. However,changes in scapular kinematics and scapulohumeral rhythmwere not found in this study.

Although other studies have demonstrated the effects ofspinal manipulation in asymptomatic participants,4-9 thescapular kinematics was not measured in these past studies.As such, some aspects are important to be discussed toexplain the lack of effects demonstrated in this investiga-tion. Despite the absence of symptoms in our participants,we considered the concept of “regional interdependence.”Arecent study43 has shown that manipulation with thrustcan increase muscle activity in muscles adjacent to themanipulated segment. However, Campbell and Snodgrass6

suggested that the mechanical effects that might occurduring manipulation is specific to the level manipulated andnot to other levels. This fact can indicate that the intra-articular dynamic where manipulation is applied is morelikely to benefit and that adjacent regions may not beaffected. Although it is known that there is a functionalrelationship between the shoulder and the spine during armelevation, the thoracic manipulation may not influence theadjacent regions enough to change the scapular muscleactivation and kinematics, at least in asymptomatic parti-cipants, as observed in the present study.

Pickar21 has suggested that the application of manipu-lation to a specific restricted segment may facilitate the

Table 4. Intragroup effect size (Cohen d coefficient) forscapulohumeral rhythm for both manipulation and sham groups

Manipulation group(n = 21)

Sham group(n = 21)

30°-120° 0.02 0.130°-60° −0.03 −0.0360°-90° 0.01 0.00590°-120° −0.02 0.03

Cohen d coefficient was calculated as follows: drepeated measures = (mean 1−mean 2)/(pooled SD/SE of the difference).

Table 2. Intragroup effect size (Cohen d coefficient) for scapular internal rotation, upward rotation, and posterior tilt acrosshumerothoracic elevation angles for both manipulation and sham groups

Manipulation group (n = 21) Sham group (n = 21)

Scapular internalrotation

Scapular upwardrotation

Scapular posteriortilt

Scapular internalrotation

Scapular upwardrotation

Scapular posteriortilt

Initialposition

0.14 −0.11 −0.02 0.33 −0.15 0.16

30° 0.15 −0.10 0.001 0.35 −0.16 0.1460° 0.13 −0.11 −0.01 0.39 −0.05 0.1690° 0.08 −0.12 −0.01 0.35 −0.19 0.07120° 0.06 −0.17 0.02 0.31 −0.18 0.03Maximum 0.27 −0.15 −0.05 0.24 −0.16 0.01

Cohen d coefficient was calculated as follows: drepeated measures = (mean 1 − mean 2)/(pooled SD/SE of the difference).

Table 3. Glenohumeral/scapulothoracic ratios at 30° increments during elevation of the arm throughout the arc of motion for bothmanipulation and sham groups before and after the intervention

Manipulation group (n = 21) Sham group (n = 21)

Before After Before After

30°-120° 2.24 ± 0.19 (1.84-2.63) 2.41 ± 0.25 (1.88-2.95) 2.99 ± 0.49 (1.96-4.03) 3.61 ± 0.91 (1.70-5.52)30°-60° 2.34 ± 0.19 (1.93-2.76) 2.13 ± 0.16 (1.79-2.49) 2.40 ± 0.20 (1.97-2.82) 2.14 ± 0.17 (1.77-2.51)60°-90° 1.88 ± 0.17 (1.51-2.25) 1.96 ± 0.24 (1.46-2.46) 2.11 ± 0.15 (1.78-2.44) 2.14 ± 0.18 (1.77-2.51)90°-120° 1.56 ± 0.21 (1.11-2.00) 1.44 ± 0.18 (1.06-1.81) 1.78 ± 0.30 (1.15-2.41) 2.02 ± 0.32 (1.34-2.69)

Results are mean ± SE (95% confidence interval).

7Rosa et alJournal of Manipulative and Physiological TherapeuticsScapular Kinematics and Scapulohumeral RhythmVolume xx, Number x

mechanoreceptors activity resulting in decreased neural in-hibition and increased muscle activation. The evaluation ofspecific segmental restrictions was not done in the presentstudy, and this may have contributed for not findingdifferences in the scapular kinematics between premanipu-lation and postmanipulation because possible differencescould be related to alteration in scapular muscle activity. It isimportant to consider that the thoracic spines of youngparticipants are relatively mobile, resulting in smallerchanges after manipulation in this population (any possiblebiomechanical effect would not be enough to changescapular kinematics).

It is well known that full range of motion during elevationof the arm necessitates motion of scapula. Studies haveshown that excessive thoracic flexion can decrease shoulderrange of motion, change scapular kinematics, and decreasescapular muscle strength.44-46 Because postural evaluationwas not performed in the current study, it is possible thatsome of our participants had thoracic rectification contrib-uting for lack of differences in the scapular kinematics whencomparing premanipulation and postmanipulation.

The findings of this investigation contribute to theliterature about manipulative therapy. Despite the fact thatno effects on scapular kinematics and scapulohumeralrhythm were shown, the manipulation was not harmful inyoung asymptomatic participants. Also, no adverse effectswere reported, suggesting that the manual therapy used inthis study was relatively safe. Conversely, the findings ofthis study suggest that seated thoracic manipulation doesnot cause pathologic motion of the scapula. Researchersshould continue investigating the effects of the manipula-

tive therapy in people with shoulder dysfunction because itsmechanisms are not well understood, despite of the high usein the clinical practice. These investigations are necessary tobetter understand and describe the concept of regionalinterdependence when treating shoulder pain disorders withthoracic manipulation.

Study LimitationsThe present study has some limitations. Because only the

immediate effects of a specific type of manipulation wereevaluated, the results cannot be generalized to long-termeffects and to other manipulation techniques. The manip-ulation technique used in this research was only one typeand performed by only one person. As such, we cannotassert if another practitioner or another type of thoracicmanipulation technique would have the same effect, as, forexample, manipulation with the patient supine or prone,which are usually more used because it can be more specific

8 Journal of ManipulativeRosa et alMonth 2013Thoracic Manipulation and Scapular Kinematics

to a certain segment.6-47 The absence of a random samplingcould have introduced selection bias in the studied sample.

This study used young asymptomatic participants withouthistory of shoulder or cervical pathology and who had goodrange of shoulder motion. Therefore, our study cannot beextrapolated to participants with shoulder pain or dysfunctionand to the general population. Because the participants had noshoulder dysfunction, the expected result of an effectivetherapy would likely be small. As such, future researchanalyzing long-term effects and using participants withshoulder pain and dysfunction would be important. Theinclusion of a no-treatment group would also be interesting aswell as the performance of other types of manipulationtechniques to determine which technique is the most effectiveand if the thoracic manipulation can influence scapularkinematics. It should also be addressed that the assessor wasnot completely blinded to the allocation group of theparticipants. Also, it is possible that the sham manipulationmay have had some therapeutic effect because the patient waspositioned, and thus, some movement was applied. Futurestudies should take these limitations in to consideration.

CONCLUSION

The results of the present study suggest that middlethoracic manipulation in the seated position does notinfluence scapulohumeral rhythm and scapular kinematicsduring arm flexion in young asymptomatic participants.Furthermore, the thoracic manipulation does not provokeharmful consequences in asymptomatic participants.

FUNDING SOURCES AND POTENTIAL CONFLICTS OF INTEREST

Financial assistance of this study was provided byFundação de Amparo à Pesquisa do Estado de São Paulo(2010/18124-0, 2010/18439-0). No conflicts of interestwere reported for this study.

Practical Applications● Immediate effects on 3D scapular kinematicswere not observed after seated thoracic manip-ulation in young asymptomatic participants.

● Studies with participants with shoulder pain orkyphosis may be necessary to conclude onpossible effects of thoracic manipulation onscapular kinematics during arm elevation.

CONTRIBUTORSHIP INFORMATIONConcept development (provided idea for the research):DPR, FA, TFS, PRC.

Design (planned the methods to generate the results):DPR, FA, TFS, PRC.

Supervision (provided oversight, responsible for orga-nization and implementation, writing of the manuscript):DPR, FA, TFS, PRC.

Data collection/processing (responsible for experiments,patient management, organization, or reporting data):DPR, FA, TFS, PRC.

Analysis/interpretation (responsible for statistical analy-sis, evaluation, and presentation of the results): DPR,FA, TFS, PRC.

Literature search (performed the literature search): DPR,FA, TFS, PRC.

Writing (responsible for writing a substantive part of themanuscript): DPR, FA, TFS, PRC.

Critical review (revised the manuscript for intellectualcontent; this does not relate to spelling and grammarchecking): DPR, FA, TFS, PRC.

REFERENCES

1. Walser RF, Meserve BB, Boucher TR. The effectiveness ofthoracic spine manipulation for the management of musculo-skeletal conditions: a systematic review and meta-analysis ofrandomized clinical trials. J Man Manipulative Ther 2009;17:237-46.

2. Ho C-YC, Sole G, Munn J. The effectiveness of manualtherapy in the management of musculoskeletal disorders of theshoulder: a systematic review. Man Ther 2009;14:463-74.

3. Cross KM, Kuenze C, Grindstaff T, Hertel J. Thoracic spinethrust manipulation improves pain, range of motion, and self-reported function in patients with mechanical neck pain: asystematic review. J Orthop Sports Phys Ther 2011;41:633-42.

4. Fernández-de-las-Peñas C, Alonso-Blanco C, Cleland JA,Rodríguez-Blanco C, Alburquerque-Sendín. Changes inpressure pain thresholds over C5-C6 zygapophyseal jointafter a cervicothoracic junction manipulation in healthysubjects. J Manipulative Physiol Ther 2008;31:332-7.

5. de Almeida BS, Sabatino JH, Giraldo PC. Effects of high-velocity, low-amplitude spinal manipulation on strength andthe basal tonus of female pelvic floor muscles. J ManipulativePhysiol Ther 2010;33:109-16.

6. Campbell BD, Snodgrass SJ. The effects of thoracicmanipulation on posteroanterior spinal stiffness. J OrthopSports Phys Ther 2010;40:685-93.

7. Bishop MD, Beneciuk JM, George SZ. Immediate reductionin temporal sensory summation after thoracic spine manipu-lation. Spine J 2011;11:440-6.

8. Stamos-Papastamos N, Petty NJ, Williams JM. Changes inbending stiffness and lumbar spine range of movementfollowing lumbar mobilization and manipulation. J Manipu-lative Physiol Ther 2011;34:46-53.

9. Orakifar N, Kamali F, Pirouzi S, Jamshidi F. Sacroiliac jointmanipulation attenuates alpha-motoneuron activity in healthywomen: a quasi-experimental study. Arch Phys Med Rehabil2012;93:56-61.

10. Schmid A, Brunner F, Wright A, Bachmann LM. Paradigmshift in manual therapy? Evidence for a central nervous systemcomponent in the response to passive cervical joint mobili-zation. Man Ther 2008:387-96.

9Rosa et alJournal of Manipulative and Physiological TherapeuticsScapular Kinematics and Scapulohumeral RhythmVolume xx, Number x

11. Gross A, Miller J, D'Sylva J, et al. Manipulation or mobili-sation for neck pain: a Cochrane review. Man Ther 2010;15(4):315-33.

12. Theodoridis D, Ruston S. The effect of shoulder movementson thoracic spine 3D motion. Clin Biomech 2002;17:418-21.

13. Wainner RS, Whitman JM, Cleland JA, Flynn TW. Regionalinterdependence: a musculoskeletal examination modelwhose time has come. J Orthop Sports Phys Ther 2007;37:658-60.

14. Winters JC, Sobel JS, Groenier KH, Arendzen HJ, Meyboom-de Jong B. Comparison of physiotherapy, manipulation, andcorticosteroid injection for treating shoulder complaints ingeneral practice: randomized, single-blind study. BMJ 1997;314:1320-5.

15. Bergman GJD, Winters JC, Groenier KH, et al. Manipulativetherapy in addition to usual medical care for patients withshoulder dysfunction and pain: a randomized, controlled trial.Ann Intern Med 2004;141:432-9.

16. Strunce JB, Walker MJ, Boyles RE, Young BA. Theimmediate effects of thoracic spine and rib manipulation onsubjects with primary complaints of shoulder pain. J ManManipulative Ther 2009;17:230-6.

17. Muth S, Barbe MF, Lauer R, McClure PW. The effects ofthoracic spine manipulation in subjects with signs of rotator cufftendinopathy. J Orthop Sports Phys Ther 2012;42:1005-16.

18. Cleland J, Selleck B, Stowell T, et al. Short-term effects ofthoracic manipulation on lower trapezius muscle strength.J Man Manipulative Ther 2004;12:82-90.

19. Phadke V, Camargo PR, Ludewig PM. Scapular and rotatorcuff muscle activity during arm elevation: a review of normalfunction and alterations with shoulder impingement. Rev BrasFisioter 2009;13:1-9.

20. Maigne JY, Vautraver P. Mechanism of action of spinalmanipulative therapy. Joint Bone Spine 2003;70:336-41.

21. Pickar JG. Neurophysiological effects of spinal manipulation.Spine J 2002;2:357-71.

22. Neer CS, Foster CR. Inferior capsular shift for involuntaryinferior and multidirectional instability of the shoulder. Apreliminary report. J Bone Joint Surg Am 1980;62:897-908.

23. Rowe CR, Zarins B. Recurrent transient subluxation of theshoulder. J Bone Joint Surg Am 1981;63:863-72.

24. Hawkins RJ, Kennedy JC. Impingement syndrome in athletes.Am J Sports Med 1980;8:151-8.

25. Neer CS. Anterior acromioplasty for the chronic impingementsyndrome in the shoulder: a preliminary report. J Bone JointSurg Am 1972;54:41-50.

26. Jobe FW, Moynes DR. Delineation of diagnostic criteria and arehabilitation program for rotator cuff injuries. Am J SportsMed 1982;10:336-9.

27. Hudak PL, Amadio PC, Bombardier C. Development of anupper extremity outcome measure: the DASH. Am J Ind Med1996;26:602-6.

28. GummessonC, Atroshi I, Ekdahl C. The disabilities of the arm,shoulder and hand (DASH) outcome questionnaire: longitu-dinal construct validity and measuring self-rated health changeafter surgery. BMC Musculoskelet Disord 2003;16:11.

29. Orfale AG, Araújo PM, Ferraz MB, Natour J. Translation intoBrazilian Portuguese, cultural adaptation and evaluation of thereliability of the Disabilities of the Arm, Shoulder and HandQuestionnaire. Braz J Med Biol Res 2005;38:293-302.

30. Ludewig PM, Cook TM. Alterations in shoulder kinematicsand associated muscle activity in people with symptoms ofshoulder impingement. Phys Ther 2000;80:276-91.

31. Borstad JD, Ludewig PM. Comparison of scapular kinematicsbetween elevation and lowering of the arm in the scapularplane. Clin Biomech 2002;17:650-9.

32. McClure PW, Michener LA, Karduna AR. Shoulder function and3-dimensional scapular kinematics in people with and withoutshoulder impingement syndrome. Phys Ther 2006;86:1075-90.

33. Teece RM, Lunden JB, Lloyd AS, Kaiser AP, Cieminski CJ,Ludewig PM. Three-dimensional acromioclavicular jointmotions during elevation of the arm. J Orthop Sports PhysTher 2008;38:181-90.

34. An KN, Korineck SL, Kilpela T, Edis S. Kinematic and kineticanalysis of push-up exercise. Biomed Sci Instrum 1990;26:53-7.

35. Wu G, van der Helm FC, Veeger HE, et al. ISB recommen-dation on definitions of joint coordinate systems of variousjoints for the reporting of human joint motion—part II:shoulder, elbow, wrist and hand. J Biomech 2005;38:981-92.

36. Braman JP, Engel SC, LaPrade R, Ludewig PM. In vivoassessment of scapulohumeral rhythm during unconstrainedoverhead reaching in asymptomatic subjects. J ShoulderElbow Surg 2009;18:960-7.

37. Fernández-De-Las-Peñas C, Cleland JA, Huijbregts P, Palo-meque-Del-Cerro L, González-Iglesias J. Repeated applicationsof thoracic spine thrust manipulation do not lead to tolerance inpatients presenting with acute mechanical neck pain: asecondary analysis. J Man Manipulative Ther 2009;17:154-62.

38. González-Iglesias J, Fernández-de-las-Peñas C, Cleland JA,Alburquerque-Sendín F, Palomeque-del-Cerro L, Méndez-Sánchez R. Inclusion of thoracic spine thrust manipulationinto an electro-therapy/thermal program for the managementof patients with acute mechanical neck pain: a randomizedclinical trial. Man Ther 2009;4:306-13.

39. González-Iglesias J, Fernández-de-las-Peñas C, Cleland JA,Gutiérrez-Vega Mdel R. Thoracic spine manipulation for themanagement of patients with neck pain: a randomized clinicaltrial. J Orthop Sports Phys Ther 2009;39:20-7.

40. Gibbons P, Tehan P. Manipulation of the spine, thorax andpelvis: an osteopathic perspective. 2nd ed. London, UK:Churchill Livingstone; 2006.

41. Thigpen CA, Gross MT, Karas SG, Garrett WE, Yu B. Therepeatability of scapular rotations across three planes ofhumeral elevation. Res Sports Med 2005;13:181-98.

42. Cohen J. Statistical power analysis for the behaviouralsciences. Hillsdale: Lawrence Erlbaum Associates; 1988.

43. de Camargo VM, Alburquerque-Sendín F, Bérzin F, StefanelliVC, de Souza DP, Fernández-de-las-Peñas C. Immediateeffects on electromyographic activity and pressure painthresholds after a cervical manipulation in mechanical neckpain: a randomized controlled trial. J Manipulative PhysiolTher 2011;34:211-20.

44. Kebaetse M, McClure P, Pratt NA. Thoracic position effect onshoulder range of motion, strength, and three-dimensionalscapular kinematics. Arch Phys Med Rehabil 1999;80:945-50.

45. Finley MA, Lee RY. Effect of sitting posture on 3-dimensionalscapular kinematics measured by skin-mounted electromagnet-ic tracking sensors. Arch Phys Med Rehabil 2003;84:563-8.

46. Bullock MP, Foster NE, Wright CC. Shoulder impingement:the effect of sitting posture on shoulder pain and range ofmotion. Man Ther 2005;10:28-37.

47. Ross JK, Bereznick DE, McGill SM. Determining cavitationlocation during lumbar and thoracic spinal manipulation: isspinal manipulation accurate and specific? Spine (Phila Pa1976) 2004;29:1452-7.