doi: 10.2522/ptj.20060009Originally published online March 6, 2007

2007; 87:408-417.PHYS THER. Vicenzino and Paul HodgesDeborah Falla, Gwendolen Jull, Trevor Russell, BillWith Chronic Neck PainEffect of Neck Exercise on Sitting Posture in Patients

http://ptjournal.apta.org/content/87/4/408found online at: The online version of this article, along with updated information and services, can be

Online-Only Material 60009.DC1.html

http://ptjournal.apta.org/content/suppl/2007/03/30/ptj.200

Collections

Therapeutic Exercise     Injuries and Conditions: Neck    

in the following collection(s): This article, along with others on similar topics, appears

e-Letters

"Responses" in the online version of this article. "Submit a response" in the right-hand menu under

or click onhere To submit an e-Letter on this article, click

E-mail alerts to receive free e-mail alerts hereSign up

by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

Effect of Neck Exercise on SittingPosture in Patients With ChronicNeck PainDeborah Falla, Gwendolen Jull, Trevor Russell, Bill Vicenzino, Paul Hodges

Background and PurposePoor sitting posture has been implicated in the development and perpetuation ofneck pain symptoms. This study had 2 purposes: (1) to compare change in cervicaland thoracic posture during a distracting task between subjects with chronic neckpain and control subjects and (2) to compare the effects of 2 different neck exerciseregimens on the ability of people with neck pain to maintain an upright cervical andthoracic posture during this task.

SubjectsFifty-eight subjects with chronic, nonsevere neck pain and 10 control subjectsparticipated in the study.

MethodChange in cervical and thoracic posture from an upright posture was measured every2 minutes during a 10-minute computer task. Following baseline measurements, thesubjects with neck pain were randomized into one of two 6-week exercise interven-tion groups: a group that received training of the craniocervical flexor muscles or agroup that received endurance-strength training of the cervical flexor muscles. Theprimary outcomes following intervention were changes in the angle of cervical andthoracic posture during the computer task.

ResultsSubjects with neck pain demonstrated a change in cervical angle across the durationof the task (mean�4.4°; 95% confidence interval [CI]�3.3–5.4), consistent with amore forward head posture. No significant difference was observed for the change incervical angle across the duration of the task for the control group subjects(mean�2.2°; 95% CI�1.0–3.4). Following intervention, the craniocervical flexortraining group demonstrated a significant reduction in the change of cervical angleacross the duration of the computer task.

Discussion and ConclusionThis study showed that people with chronic neck pain demonstrate a reduced abilityto maintain an upright posture when distracted. Following intervention with anexercise program targeted at training the craniocervical flexor muscles, subjects withneck pain demonstrated an improved ability to maintain a neutral cervical postureduring prolonged sitting.

D Falla, PhD, BPhty(Hons), isNHMRC Research Fellow, Centerfor Sensory-Motor Interaction, De-partment of Health Science andTechnology, Aalborg University,Fredrik Bajers Vej 7D-3, DK-9220,Aalborg, Denmark, and Division ofPhysiotherapy, The University ofQueensland, Brisbane, Queens-land, Australia. Address all corre-spondence to Dr Falla at: deborahf@hst.aau.dk.

G Jull, PhD, MPhty, GradDipAdvManipTher, is Professor and Head,Division of Physiotherapy, TheUniversity of Queensland.

T Russell, PhD, BPhty, is Lecturer,Division of Physiotherapy, TheUniversity of Queensland.

B Vicenzino, PhD, MSc, GradDip-Phty(Sports), BPhty, is AssociateProfessor, Division of Physiother-apy, The University of Queensland.

P Hodges, PhD, BPhty, is Professorand NHMRC Senior Research Fel-low/Professorial Research Fellow,Division of Physiotherapy, The Uni-versity of Queensland.

[Falla D, Jull G, Russell T, et al. Ef-fect of neck exercise on sittingposture in patients with chronicneck pain. Phys Ther. 2007;87:408–417.]

© 2007 American Physical TherapyAssociation.

Research Report

For The Bottom Line:www.ptjournal.org

408 f Physical Therapy Volume 87 Number 4 April 2007 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

In an upright, neutral posture ofthe cervical spine, passive resis-tance to motion is minimal.1 Sup-

port of the cervical segments is pro-vided by the muscular sleeve formedby the longus colli muscle anteriorlyand the semispinalis cervicis andcervical multifidus muscles poste-riorly.2–5 In particular, the longuscolli muscle has a major posturalfunction in supporting and straight-ening the cervical lordosis.4 In addi-tion, the craniocervical region issupported by muscles that attach tothe cranium and span the uppercervical motion segments, such asthe longus capitis muscle anteriorlyand the subocciptal extensor, semi-spinalis, and splenius capitis musclesposteriorly.6

The importance of the deep mus-cles for the maintenance of cervicalposture was verified in a computermodel, which showed regions of lo-cal segmental instability if only thelarge superficial muscles of the neckwere simulated to produce move-ment, particularly in near-upright orneutral postures.7 Deep cervicalmuscle activity was required in syn-ergy with superficial muscle activityto stabilize the cervical segments,especially in functional mid-ranges ofthe cervical spine.

Recent studies have identified im-paired activation of the deep cervicalflexor muscles, the longus colli andlongus capitis, in people with neckpain.8,9 Given the role of the deepcervical flexor muscles in posturalsupport and the knowledge of im-paired activation of these muscles inpeople with neck pain, it is likelythat this patient population alsowould display deficits in the posturalendurance of these muscles. Indeed,evidence is emerging that suggeststhat people with neck pain drift intoa more forward head position whendistracted.10 This has been observeddespite a lack of postural differencesin people with neck pain in erect

sitting.11–14 Moreover, retraining thedeep cervical flexor muscles, whichhas been shown to decrease necksymptoms15,16 and increase the activa-tion of the deep cervical flexor mus-cles during performance of the clinicaltest of craniocervical flexion,16 mayimprove the ability to maintain an up-right posture of the cervical spine.

This study had 2 purposes: (1) to iden-tify whether people with neck paindemonstrate differences in their abilityto maintain an upright posture whendistracted by a computer task com-pared with a group of control subjectsand (2) to compare the effects of alow-load craniocervical flexion train-ing regimen against a conventionalneck flexor endurance-strength train-ing program on functional control ofhead and neck posture in peoplewith chronic neck pain. The low-loadcraniocervical flexion training regi-men was compared with a conven-tional strengthening regimen becauseit is not known whether such specifictraining of the deep cervical muscles isrequired in rehabilitation or if a moregeneral strengthening exercise of theneck flexor muscles would be suffi-cient to improve control of the cervi-cal postural position.

This study forms part of a series ofexperiments to investigate the mech-anisms of efficacy of cervical muscleretraining. The effect of both exer-cise regimens on measures of painand disability have been reported inour previous work.16,17

MethodSubjectsFifty-eight female subjects (meanage�37.9 years, SD�10.2 years) witha history of chronic, non-severe neckpain of greater than 3 months (X�7.9years, SD�6.4 years) participated inthis study. Subjects were recruited byadvertisements in the local press. Tobe included, the subjects had to score�15 (out of a possible 50) on the NeckDisability Index (NDI).18 An NDI score

�15 indicates mild to moderate neckpain.18

Subjects in this category were se-lected because previous studies in-vestigating motor control deficits inpeople with neck pain examined pa-tients with similar perceived painand disability scores. For example,reduced activation of the deep cervi-cal muscles has been observed inpeople with neck pain with an NDIscore of �15.9,19 Moreover, the av-erage NDI score of the patients in-cluded in this study is similar to pre-vious exercise trials.20,21 People withmore severe pain were excluded be-cause the endurance exercise regi-men may have increased the symp-toms of this group.

Subjects also had to have palpable cer-vical joint tenderness22 and demon-strate poor performance (unable toachieve 24 mm Hg) on the clinical testof craniocervical flexion as definedby Jull et al.23 Further details of thetest are presented in the “ExerciseRegimens” section. Subjects were ex-cluded if they had undergone cervicalspine surgery, reported any neuro-logical signs, or had participated in aneck exercise program in the past 12months.

The mean score of subjects on theNDI was 9.9 (out of a possible 50)(SD�3.1), and the average intensityof neck pain was 4.1�2.1 on a 10-cmnumerical rating scale (NRS) an-chored with “no pain” and “theworst possible pain imaginable.” Thesubjects with neck pain who par-ticipated in this study also formedpart of another study.17 The samplesize (26 per group plus a 10% drop-out allowance) was based on thedifference in fatigue of the cervicalmuscles between a group of sub-jects with neck pain and subjectswho were asymptomatic (mean dif-ference�0.65 Hz, SD�0.83, power�90%).24 Thus, the study had suffi-cient power to detect a difference in

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

April 2007 Volume 87 Number 4 Physical Therapy f 409 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

the ability of the exercise interven-tions to change parameters of mus-cle function, but was not designed tocompare the efficacy of the ap-proach to reduce pain and disability.

Ten volunteers (mean age�35.0years, SD�4.6) formed the controlgroup. The control group subjectswere recruited via local advertise-ments and were free of neck pain,had no past history of orthopedicdisorders affecting the neck, and hadno history of neurological disorders.

Experimental ProcedurePhase I. Subjects were positionedin front of the computer in sittingwith their knees in 90 degrees offlexion and their feet flat on theground. A plumb line was positionedin the background. The starting po-sition was standardized by placingthe subject in an upright posture,which was defined as a vertical pel-vic position (no anterior or posteriortilt) with the assumption of a lumbarlordosis and thoracic kyphosis.23 Sub-jects were asked to maintain the posi-tion while they were distracted byplaying the game of Solitaire on thecomputer for 10 minutes. Subjectsused the mouse with their domi-nant hand and the other handrested motionless on the desk infront of them.

Postural analysis. Cervical andthoracic posture was measuredthroughout the 10-minute computertask from a lateral photograph takenwith a digital camera (Canon DigitalIXUS, 1600 � 1200 pixels)* posi-tioned on a tripod at a distance of0.8 m. The axis of the lens of thecamera was placed orthogonal tothe sagittal plane of the patient at aheight that corresponded with theseventh cervical vertebra. Anatomi-cal markers were positioned on the

tragus of the ear and the spinousprocesses of the seventh cervicaland seventh thoracic vertebrae andwere fixed with double-sided medicaltape.

The digital technique used to quan-tify angular displacement in this studyhas been previously described.25,26

The technique has been shown toproduce reliable angular measure-ments (intraclass correlation coeffi-cient [ICC](2,2)�.93) and the crite-rion validity of the technique hasbeen established when compared tothe universal goniometer by a non-significant (F�0.02; df�1,5; P�.887)mean absolute difference (0.26°) be-tween the 2 measurement tech-niques.25 Using this technique, mea-sures of angular displacement in theshoulder, elbow, wrist, and knee jointshave demonstrated standard error ofmeasurement values of 0.83, 0.38,0.37 and 0.50 degree and a minimaldetectable change at the 90% confi-dence interval (CI)27 of 0.34, 0.23,0.17 and 0.23 degree, respectively(Russell et al, unpublished data).25

The angle of forward head posturewas measured from a line drawnfrom the tragus of the ear to theseventh cervical vertebra subtendedto the horizontal (Fig. 1, angle A).28

The software produced a horizontalline perpendicular to the verticalplumb line captured in the back-ground of the image. Thoracic pos-ture was calculated as the angle be-tween the horizontal line and a linedrawn between the seventh cervicalspinous process and the seventh tho-racic spinous process (Fig. 1, angleB). Changes in angles from an erectstarting posture (time 0) to the an-gles measured at 2-minute intervalsthroughout the 10-minute task werecalculated and expressed relative tothe angle at time 0.

Phase II. Following baseline mea-surements, the subjects with chronicneck pain were randomized into 1 of

2 exercise groups: a training regimenof the craniocervical flexor musclesor an endurance-strength trainingregimen for the cervical flexor mus-cles. The allocation sequence wasgenerated by an independent bodyand an independent investigator as-signed participants to their group.Figure 2 illustrates the progression ofsubjects through the exercise trial.Postural analysis during the com-puter task was performed at baselineand in the week immediately afterthe 6-week intervention period(week 7) for the patient group. Theresearcher taking the measurementswas blinded to subject group for theoutcome assessments and statisticalanalyses.

* Cannon Australia Pty Ltd, 1 Thomas Holt Dr,North Ryde, New South Wales, Australia,2113.

Figure 1.Cervical and thoracic postural parameters.Subjects were positioned in an uprightneutral posture. Anatomical markers werepositioned on the tragus of the ear, spi-nous process of the seventh cervical ver-tebra, and the spinous process of theseventh thoracic vertebra. The angle offorward head posture (A) was measuredfrom a line drawn from the tragus of theear to the seventh cervical vertebra sub-tended to the horizontal. Thoracic posturewas calculated as the angle between thehorizontal line and the line drawn fromthe seventh cervical spinous process tothe seventh thoracic spinous process (B).

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

410 f Physical Therapy Volume 87 Number 4 April 2007 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

Exercise RegimensThe exercise regimens were con-ducted over a 6-week period andsubjects in each group received per-sonal instruction and supervision byan experienced physical therapistonce per week for the duration ofthe trial. None of the exercise ses-sions were longer than 30 minutes.Subjects were asked not to receiveany other specific intervention fortheir neck pain; however, any medi-cation that a subject was currentlytaking was not withheld. All subjectswere supplied with an exercise diaryand requested to practice their re-spective regimen twice per day forthe duration of the trial. The exercise

occupied a period of no longer than10 to 20 minutes per day. The exer-cises were performed without anyprovocation of neck pain.

Craniocervical flexor training in-tervention. Training of the cranio-cervical flexor muscles followed theprotocol described by Jull et al.23 Theexercise targets the deep flexor mus-cles of the upper cervical region, thelongus capitis and longus colli mus-cles, rather than the superficial flexormuscles, the sternocleidomastoid andanterior scalene, which flex the neckbut not the head.19,29 In addition, theexercise is a low-load exercise in na-ture to more specifically train the deep

cervical flexors, rather than the neckflexors as a whole, which occurs in ahead lift exercise. The exercise used anair-filled pressure sensor (Stablizer),†

which was placed sub-occipitally tomonitor the subtle flattening of thecervical lordosis that occurs withthe contraction of the longus collimuscle.4

The subject was guided by the feed-back from the pressure sensor to se-quentially reach 5 pressure targets in2–mm Hg increments from a base-line of 20 mm Hg to the final level of30 mm Hg. Subjects were instructedto “gently nod their head as thoughthey were saying ‘yes’.” The physicaltherapist identified the target levelthat the subject could hold steadilyfor 10 seconds without resorting toretraction, without dominant use ofthe superficial neck flexor muscles,and without a quick, jerky cranio-cervical flexion movement.23 Contri-bution from the superficial muscleswas monitored by the physical ther-apist in all stages of the test usingobservation or palpation.

Training was commenced at the tar-get level that the subject could achievewith a correct movement of cranio-cervical flexion and without dominantuse or substitution by the superficialmuscles (sternocleidomastoid, hyoid,and anterior scalene muscles). The sub-jects were taught to perform a slowand controlled craniocervical flexionaction. They then trained to be able tosustain progressively increasing rangesof craniocervical flexion using feed-back from the pressure sensor, whichwas placed behind the neck. For eachtarget level, the contraction durationwas increased to 10 seconds, and thesubject trained to perform 10 repeti-tions. At this stage, the exercise wasprogressed to train at the next targetlevel.

† Chattanooga Group Inc, 4717 Adams Rd,Hixson, TN 37343.

Assessed for eligibility(n=147)

Enrollment

(n=58)

Excluded (n=89)NDI >15 (n=78)Other reasons (n=11)

Randomization

(n=58)

Allocated to CCF training

(n=29)

Allocated to endurance-

strength training

(n=29)

Followed up at week 7

(n=29)

Followed up at week 7

(n=29)

Figure 2.Progression of participants through the exercise trial. NDI�Neck Disability Index,CCF�craniocervical flexor.

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

April 2007 Volume 87 Number 4 Physical Therapy f 411 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

Endurance-strength training inter-vention. The endurance-strengthtraining regimen consisted of a pro-gressive resistance exercise programfor the neck flexors. The exercise wasperformed in supine position, withthe head supported in a comfortableresting position. Subjects were in-structed to lift up their head so thatcervical flexion occurred while main-taining a neutral upper cervical spineposition. The subjects slowly movedthe head and neck through as full arange of motion as possible withoutcausing discomfort or reproducingtheir symptoms.

This exercise regimen was a 2-stageprogram. The first stage was of 2weeks’ duration and the second wasof 4 weeks’ duration as recommend-ed30 for initiating a weight programin previously untrained individuals.In stage 1, the subjects performed12 to 15 repetitions with a weightthat they could lift 12 times (12-repetition maximum [RM]) on thefirst training session and progressedto 15 repetitions and maintained thislevel for the remainder of the 2-weekperiod.

In stage 2, the subjects performed 3sets of 15 repetitions of the initial12-RM load once per day. One-minute rest intervals were providedbetween sets. If repetitions wereeasily achieved, weighted sandbagswere applied to the patient’s fore-head in 0.5-kg increments. If the sub-ject was unable to perform repeti-tions of the head lift maneuver thenthe load on the neck flexors was re-duced by allowing the subject to per-form the task with the upper body(trunk and neck) inclined up fromthe horizontal so that the subjectcould perform the required repeti-tions of the movement.

Data AnalysisComparison between subjectswith neck pain and control sub-jects. Angle data were expressed

as a change from the starting angle ateach time interval throughout the 10-minute computer task. A repeated-measures general linear model wasused to identify whether change incervical and thoracic angles acrossthe duration of the task were differ-ent between the 2 subject groups.The independent variable was thesubject group (between-subjects fac-tor), and the within-subject factorwas the time interval of the task (5measurements).

Change in posture before and af-ter intervention for the exercisegroups. Paired sample t tests wereconducted to determine if NDI andNRS measurements were significantlydifferent before and after the interven-tion for both exercise groups, andindependent sample t tests were con-ducted to compare for group differ-ences. A repeated-measures generallinear model was used to comparebaseline cervical and thoracic anglesbetween the 2 intervention groupswith factors of group (craniocervicalflexor training and endurance-strengthtraining) and time (5 measurements).

For the preintervention to postinter-vention analysis, a repeated-measuresgeneral linear model was applied. Theindependent variables were the 2 in-tervention groups (between-subjectsfactor) and the within-subject factorwas the time interval of the task (5measurements). A polynomial or lin-ear trend was fitted to the time factorto explain the relative change in cer-vical and thoracic angle across the du-ration of the task. A value of P�.05was used as an indicator of statisticalsignificance.

ResultsComparison of Postural PositionBetween Subjects With Neck Painand Control SubjectsSubjects with neck pain demon-strated a significant, progressive in-crease in change of cervical anglefrom baseline throughout the 10-

minute computer task (F�19.3;df�1,56; P�.001; Fig. 3A). In con-trast, for the control subjects, therewas no evidence for a change in cer-vical angle over the 10-minute com-puter task (F�1.95; df�1,56; P�.17;Fig. 3A). Compared with the startingposition, the mean change in cervi-cal angle at 10 minutes was 4.4 de-grees (SD�4.1°, 95% CI�3.3–5.4)for the neck pain group and 2.2 de-grees (SD�1.6°, 95% CI�1.0–3.4)for the control group.

The subjects with neck pain also dem-onstrated a significant, progressive in-crease in change of thoracic anglefrom baseline across time (F�45.3;df�1,56; P�.001; Fig. 3B). Althoughless than the subjects with neck pain,the control subjects also demonstratedan increase for the change of thoracicangle (F�11.4; df�1,9; P�.01;Fig. 3B). Compared with the startingposition, the mean change in thoracicangle at 10 minutes was 8.2 degrees(SD�4.8°, 95% CI�6.9–9.5) for thesubjects with neck pain and 4.8 de-grees (SD�3.3°, 95% CI�2.4–7.1) forthe control subjects.

Changes in Cervical andThoracic Angle After ExerciseInterventionOf the 58 participants with neckpain who participated in the exer-cise interventions, none were lost tofollow up assessment. Subject de-scriptive data are presented in theTable. Baseline characteristics ofpain (NRS) and disability (NDI) werenot different between the 2 interven-tion groups (P�.05). In addition,preintervention cervical (F1�0.28,P�.05) and thoracic (F1�2.13, P�.05)angles were not significantly differentbetween the 2 intervention groups.All participants in the endurance-strength training group and cranio-cervical flexor training group receivedthe full 6 treatments. According to thepatient diaries, adherence to exercisewas 91.0% (SD�0.12%) for theendurance-strength training group

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

412 f Physical Therapy Volume 87 Number 4 April 2007 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

and 94.8% (SD�0.06%) for the cranio-cervical flexor training group. Nopatients reported any adverse events.

Both intervention groups demon-strated a reduction in average intensityof pain (craniocervical flexor training:�0.9�2.3, endurance-strength train-ing: �1.1�2.8), and NDI score (cranio-cervical flexion training: �3.5�4.8,endurance-strength training: �2.8�4.0). However, there was no differ-ence between groups for change inpain (NRS) or disability (NDI)(P�.05).

Following 6-weeks of intervention,the craniocervical flexor traininggroup demonstrated a significant re-

duction in the change of cervical angle(F�7.44; df�1,1,1; P�.01; Fig. 4)across the duration of the task whencompared with the endurance-strengthtraining group. In addition, both groupsimproved their ability to maintain anupright posture of the thoracic spine;however, there was no significant dif-ference between the 2 intervention groups(F�2.55; df�1,1,1; P�.05; Fig. 5).

DiscussionThe results of this study demonstratedthat subjects with chronic non-severeneck pain had a reduced ability to main-tain an upright neutral posture whendistracted by a computer task. More-over, exercise targeted at training thecraniocervical flexor muscles improved

the ability to maintain an uprightcervical posture during this task.

Comparison BetweenSubjects With Neck Pain andControl SubjectsIn support of previous findings,10

subjects with neck pain demon-strated a reduced ability to main-tain an upright posture during acomputer task. There was a subtleforward drift of the head of a mag-nitude of 4.4�4.1 degrees in asso-ciation with a subtle increase inthe thoracic flexion curve of8.2�4.8 degrees in subjects withneck pain. This may reflect im-paired endurance of the musclesthat would be required to control

Table.Baseline Characteristics for Patients With Chronic Neck Pain Randomized Into a Craniocervical Flexor Exercise Intervention or anEndurance-Strength Exercise Intervention

Craniocervical Flexor ExerciseIntervention (n�29)

Endurance-Strength ExerciseIntervention (n�29)

Mean�SD Median Range Mean�SD Median Range

Age 37.7�9.9 38.0 22.0–55.0 38.1�10.7 38.0 22.0–55.0

Length of neck pain history (y) 7.5�5.9 7.0 0.5–21.0 8.3�7.0 5.5 1.0–30.0

Neck pain intensity (0–10 cm) 3.6�2.0 3.4 0.7–7.1 4.7�2.0 4.5 1.8–9.0

Neck Disability Index (0–50) 9.8�3.3 10.0 2.0–14.0 10.4�3.4 10.0 3.0–15.0

Figure 3.Group comparisons for change in cervical and thoracic posture. Data (mean and standard deviation) are presented for change incervical posture (left) and change in thoracic posture (right) for patients with neck pain and for control subjects. Change in anglefrom an erect starting posture (time 0 [T0]) are expressed relative to the angle measured at 2-minute intervals (T2, T4, T6, T8, T10)throughout the 10-minute task.

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

April 2007 Volume 87 Number 4 Physical Therapy f 413 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

the postural position of the spine dur-ing sitting. In agreement with previousresearch, decreased endurance of thecraniocervical flexor muscles has beenobserved in patients with neck pain at20% of their maximal voluntary contrac-tion.31 Other factors such as reducedproprioception resulting in poor head

position awareness also may explainthe differences observed for thegroup with neck pain comparedwith the control group. Evidenceof reduced cervical kinestheticsense has been identified in bothpeople with idiopathic neck pain

and people with neck pain follow-ing a whiplash injury.32–34

A reduced ability to maintain an up-right posture of the cervical spinewhen distracted during sitting mightbe considered a measure of impair-ment in the postural supporting mus-

Figure 4.Group data for change in cervical posture following intervention in patients with neck pain. Preintervention and postintervention data(mean and standard deviation) are presented for change in cervical posture throughout the 10-minute computer task for thecraniocervical flexor (CCF) training group and endurance-strength training group. Change in angle from an erect starting posture(time 0 [T0]) are expressed relative to the angle measured at 2-minute intervals (T2, T4, T6, T8, T10) throughout the 10-minute task.

Figure 5.Group data for change in thoracic posture following intervention in patients with neck pain. Preintervention and postinterventiondata (mean and standard deviation) are presented for change in thoracic posture throughout the 10-minute computer task for thecraniocervical flexor (CCF) training group and the endurance-strength training group. Change in angle from an erect starting posture(time 0 [T0]) are expressed relative to the angle measured at 2-minute intervals (T2, T4, T6, T8, T10) throughout the 10-minute task.

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

414 f Physical Therapy Volume 87 Number 4 April 2007 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

cles during a functional task, an out-come that can be easily replicatedclinically.

A sustained forward flexion postureof the spine has been associatedwith increased cervical compressiveloading and a creep response in theconnective tissue.35,36 It would notbe unreasonable to consider that asustained forward head posture asso-ciated with prolonged sitting couldaggravate, if not initiate, neck pain.There is some evidence that haslinked prolonged static posture withincreased muscle loading and subse-quent risk for the development ofsymptoms in the upper body.37,38

Although both the subjects withneck pain and the control subjectsdemonstrated progressive change inthoracic posture throughout the 10-minute computer task, this changewas greater for the subjects withneck pain. The observation that tho-racic angle increased progressivelyin the control subjects throughoutthe task in the absence of a change incervical posture was somewhat sur-prising, but could indicate earlier fa-tigue in the trunk extensors than inthe neck muscles. Further investiga-tion of this finding is necessary.

Effect of Exercise on Control ofPosture During Sitting inSubjects With Neck PainFollowing a 6-week intervention witheither craniocervical flexor training orneck flexor endurance-strength train-ing, the participants with neck painimproved their ability to maintain anupright posture of the thoracic spineduring the 10-minute computer task.This improvement could be attributedto factors such as task familiarity orincreased postural awareness; how-ever, only the group that received thespecific craniocervical flexor trainingimproved their ability to maintain anupright position of the cervical spine.

Craniocervical flexor training involvesperforming and holding inner rangepositions of craniocervical flexion, theanatomical action of the deep cervicalflexor muscles. This training has beenshown to increase the activation ofthese muscles.16 The improved abilityto maintain an upright position of thecervical spine, which was observedfor the craniocervical flexor traininggroup, may reflect an improved en-durance of the deep cervical flexormuscles, which was identified dur-ing the functional task of sitting. Thisimprovement occurred even thoughthere was no exercise instructionon postural correction in sitting. Thisfinding supports our previous sug-gestion that inadequate control ofthe head in prolonged sitting may bea functional correlate of deep cervi-cal muscle impairment.

Moreover, craniocervical flexion di-rectly activates the deep cervicalflexor musculature,19,39 which havea relatively high density of musclespindles.2 Improved cervical kines-thetic sense following craniocervicalflexor training40 also may explain theimproved ability to maintain an up-right position of the cervical spine.

It is notable that the endurance-strength regimen did not influencepostural parameters of the cervicalspine. Although there is some evi-dence to suggest that an endurance-strength regimen for the neck flexormuscles reduces neck pain,17,41,42

improves strength,17,42 and reducesfatigue of the sternocleidomastoidand anterior scalene muscles,17 itdoes not appear to improve theability to maintain an uprightposture of the cervical spine in asitting task.

The maintenance of cervical posturalangle with the craniocervical flexortraining during the 10-minute dis-traction task reached statistical sig-nificance when compared with theendurance-strength regimen. Never-

theless, it can be questioned whetherthe subtle maintenance of posturalangles is clinically meaningful. Thisquestion cannot be answered directlyin this study. However, the magnitudeof change in cervical posture follow-ing craniocervical flexion training issimilar to the magnitude of differenceobserved between the subjects withneck pain and the control subjects inthe first phase of this study. Further-more, such subtlety in head driftwas also observed by Szeto et al43 intheir comparison of computer work-ers with and without neck pain. Theoutcomes of both studies suggest that,in sitting working postures, subtlechanges in posture over time, possiblyreflective of poor muscle control asproposed in this study, might be veryrelevant to the function of officeworkers with neck pain. The possibleassociations between functional work-ing postures and neck pain justifiesfurther research towards meeting thechallenge of prevention of neck painin office workers, which is recognizedas a significant contemporary problemin the workforce.44

Change in Pain andPerceived DisabilityFollowing 6 weeks of exercise, asignificant reduction in average in-tensity of pain (NRS) and perceiveddisability (NDI score) was identifiedfor both training groups. Althoughonly the craniocervical flexor traininggroup showed a significant improve-ment in their ability to maintain anupright position of the cervical spine,this was not associated with a greaterreduction in pain or perceived disabil-ity compared with the endurance-strength regimen group. However,because a sustained forward flexionposture of the spine has been associ-ated with compressive loading of thecervical tissues,35,36 improved cervicalposture during sitting may have an ad-ditional long-term benefit of reducingrecurrent episodes of neck pain. Thisis of particular relevance given the highrecurrence rate of neck pain.45 Further

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

April 2007 Volume 87 Number 4 Physical Therapy f 415 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

research is now warranted to examinewhether an improved ability to main-tain an upright position of the cervicalspine following specific exercise inter-vention is maintained in the long termand the effect that this may have onthe recurrence rate of neck pain.

Methodological ConsiderationsThis study used photographic analy-sis to describe change in cervical andthoracic posture using anatomicalmarkers. Although photographicanalysis has shown to be a reliabletool for quantifying change in cervi-cal angle,46 precise conclusionsabout the anatomical alignment ofthe spine as identified on radio-graphs cannot be inferred from vari-ation in surface measurement.47 De-spite this limitation, this studydemonstrates that postural analysisduring a common functional activityin sitting may provide a useful mea-sure to quantify postural changesduring tasks and to monitor the ef-fects of rehabilitation.

Only posture of the cervical and tho-racic spine were analyzed in thisstudy. In future studies, electromyo-graphy could be used concurrentlyto provide additional information onmuscle activation associated withthe observed postural changes.

Additional methodological aspectsmay include the duration of the com-puter task used in this study (10 min-utes). However, the duration of thetask was sufficient to demonstratedifferences between subjects withneck pain and control subjects.Finally, it must be noted that it is notknown whether the improvementsin postural endurance that were ob-served following 6-weeks of exerciseintervention would be maintained inthe long term. Additional research iswarranted to address these issues.

ConclusionSubjects with chronic neck paindemonstrated a reduced ability to

maintain an upright neutral posturewhen distracted by a computer task.Following intervention with an exer-cise program targeted at retrainingthe craniocervical flexor muscles,subjects with chronic neck paindemonstrated improved ability tomaintain a neutral cervical postureduring prolonged sitting. This mostlikely reflects an improvement in theendurance of the muscles that con-trol the postural position of the neckduring function.

Dr Falla, Dr Jull, Dr Vincenzino, and DrHodges provided concept/idea/research de-sign and fund procurement. Dr Falla, Dr Jull,Mr Russell, and Dr Hodges provided writing.Dr Falla provided data collection, and DrFalla, Mr Russell, and Dr Hodges provideddata analysis. Dr Falla and Dr Jull providedproject management. Dr Jull and Dr Hodgesprovided facilities/equipment. Mr Russellprovided consultation (including review ofmanuscript prior to submission). The authorsthank Amy Fagan from The University ofQueensland, Australia for assistance withdata collection.

Dr Falla is supported by the National Healthand Medical Research Council of Australia(ID 351678). This study was funded by agrant (ID 252771) received from the Na-tional Health and Medical Research Councilof Australia.

Ethical approval for the study was granted bythe Institutional Medical Research EthicsCommittee of The University of Queensland,and all procedures were conducted accord-ing to the Declaration of Helsinki.

This article was submitted January 9, 2006,and was accepted December 19, 2006.

DOI: 10.2522/ptj.20060009

References1 Oatis CA. Kinesiology: The Mechanics

and Pathomechanics of Human Move-ment. Philadelphia, Pa: Lippincott Wil-liams & Wilkins; 2004.

2 Boyd-Clark LC, Briggs CA, Galea MP. Mus-cle spindle distribution, morphology, anddensity in longus colli and multifidus mus-cles of the cervical spine. Spine. 2002;27:694–701.

3 Conley MS, Meyer RA, Bloomberg JJ, et al.Noninvasive analysis of human neck mus-cle function. Spine. 1995;20:2505–2512.

4 Mayoux-Benhamou MA, Revel M, Vallee C,et al. Longus colli has a postural functionon cervical curvature. Surg Radiol Anat.1994;16:367–371.

5 Vasavada AN, Li S, Delp SL. Influence ofmuscle morphometry and moment armson the moment-generating capacity ofhuman neck muscles. Spine. 1998;23:412–422.

6 Kettler A, Hartwig E, Schultheiss M, et al.Mechanically simulated muscle forcesstrongly stabilize intact and injured uppercervical spine specimens. J Biomech.2002;35:339–346.

7 Winters JM, Peles JD. Neck muscle activityand 3D head kinematics during quasistaticand dynamic tracking movements. In:Winters JM, Woo SLY, eds. Multiple Mus-cle Systems: Biomechanics and Move-ment Organization. New York, NY:Springer-Verlag; 1990:461–480.

8 Falla DL, Jull GA, Hodges PW. Patientswith neck pain demonstrate reduced elec-tromyographic activity of the deep cervi-cal flexor muscles during performance ofthe craniocervical flexion test. Spine.2004;29:2108–2114.

9 Falla D, Jull G, Hodges PW. Feedforwardactivity of the cervical flexor muscles dur-ing voluntary arm movements is delayed inchronic neck pain. Exp Brain Res. 2004;157:43–48.

10 Szeto GP, Straker LM, O’Sullivan PB. Acomparison of symptomatic and asymp-tomatic office workers performing monot-onous keyboard work, 2: neck andshoulder kinematics. Man Ther. 2005;10:281–291.

11 Grimmer K. The relationship between cer-vical resting posture and neck pain. Phys-iotherapy. 1996;82:45–51.

12 Hanten WP, Olson SL, Russell JL, et al. To-tal head excursion and resting head pos-ture: normal and patient comparisons.Arch Phys Med Rehabil. 2000;81:62–66.

13 Haughie LJ, Fiebert IM, Roach KE. Rela-tionship of forward head posture and cer-vical backward bending to neck pain.Journal of Manual and ManipulativeTherapy. 1995;3:91–97.

14 Treleaven J, Jull G, Atkinson L. Cervicalmusculoskeletal dysfunction in post-concussional headache. Cephalalgia.1994;14:273–279.

15 Jull G, Trott P, Potter H, et al. A random-ized controlled trial of exercise andmanipulative therapy for cervicogenicheadache. Spine. 2002;27:1835–1843.

16 Jull G, Falla D, Hodges P, et al. Cervicalflexor muscle retraining: physiologicalmechanisms of efficacy. Paper presentedat: 2nd International Conference on Move-ment Dysfunction; September 23–25,2005; Edinburgh, Scotland.

17 Falla D, Jull G, Hodges P, Vicenzino B. Anendurance-strength training regime is ef-fective in reducing myoelectric manifesta-tions of cervical flexor muscle fatigue infemales with chronic neck pain. Clin Neu-rophysiol. 2006;117:828–837.

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

416 f Physical Therapy Volume 87 Number 4 April 2007 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

18 Vernon H, Mior S. The Neck Disability In-dex: A study of reliability and validity.J Manipulative Physiol Ther. 1991;14:409–415.

19 Falla D, Bilenkij G, Jull G. Patients withchronic neck pain demonstrate alteredpatterns of muscle activation during per-formance of a functional upper limb task.Spine. 2004;29:1436–1440.

20 Ylinen J, Takala EP, Nykanen M, et al. Ac-tive neck muscle training in the treatmentof chronic neck pain in women: a random-ized controlled trial. JAMA. 2003;289:2509–2516.

21 Jull G, Barrett C, Magee R, Ho P. Furtherclinical clarification of the muscle dysfunc-tion in cervical headache. Cephalalgia.1999;19:179–185.

22 Jull G, Bogduk N, Marsland A. The accu-racy of manual diagnosis for cervical zyg-apophysial joint pain syndromes. Med JAust. 1988;148:233–236.

23 Jull G, Falla D, Treleaven J, et al. A thera-peutic exercise approach for cervical dis-orders. In: Boyling JD, Jull G, eds. Grieve’sModern Manual Therapy: The VertebralColumn. 3rd ed. Edinburgh, United King-dom: Elsevier; 2004.

24 Falla D, Rainoldi A, Merletti R, Jull G. Myo-electric manifestations of sternocleido-mastoid and anterior scalene musclefatigue in chronic neck pain patients.Clin Neurophysiol. 2003;114:488–495.

25 Russell TG, Jull GA, Wootton R. Can theInternet be used as a medium to evalu-ate knee angle? Man Ther. 2003;8:242–246.

26 Russell TG, Wootton R, Jull GA. Physicaloutcome measurements via the Internet: re-liability at two Internet speeds. J TelemedTelecare. 2002;8:50–52.

27 Haley SM, Fragala-Pinkham MA. Interpret-ing change scores of tests and measuresused in physical therapy. Phys Ther. 2006;86:735–743.

28 Watson DH, Trott PH. Cervical head-ache: an investigation of natural headposture and upper cervical flexor mus-cle performance. Cephalalgia. 1993;13:272–284.

29 Falla D, Jull G, O’Leary S, Dall’Alba P. Fur-ther evaluation of an EMG technique forassessment of the deep cervical flexormuscles. J Electromyogr Kinesiol. 2006;16:621–628.

30 McArdle WD, Katch FI, Katch VL. ExercisePhysiology: Energy, Nutrition, and Hu-man Performance. 4th ed. Baltimore, Md:Williams & Wilkins; 1996.

31 O’Leary S, Jull G, Kim M, Vicenzino B.Cranio-cervical flexor muscle impairmentat maximal, moderate, and low loads is afeature of neck pain. Man Ther. 2006 Jun12: Epub ahead of print.

32 Heikkila H, Astrom PG. Cervicocephlickinesthetic sensibility in patients withwhiplash injury. Scand J Rehabil Med.1996;28:133–138.

33 Kristjansson E, Dall’Alba P, Jull G. A studyof five cervicocephalic relocation tests inthree different subject groups. Clin Reha-bil. 2003;17:768–774.

34 Revel M, Andre-Deshays C, Minguet M.Cervicocephalic kinesthetic sensibility inpatients with cervical pain. Arch Phys MedRehabil. 1991;72:288–291.

35 Harms-Ringdahl K, Ekholm J, Schuldt K,et al. Load moments and myoelectric ac-tivity when the cervical spine is held in fullflexion and extension. Ergonomics. 1986;29:1539–1552.

36 Twomey L, Taylor J. Flexion creep defor-mation and hysteresis in the lumbar verte-bral column. Spine. 1982;7:116–122.

37 Aaras A, Fostervold KI, Ro O, et al. Pos-tural load during VDU work: a comparisonbetween various work postures. Ergo-nomics. 1997;40:1255–1268.

38 Schuldt K, Ekholm J, Harms-Ringdahl K,et al. Effects of changes in sitting workposture on static neck and shouldermuscle activity. Ergonomics. 1986;29:1525–1537.

39 Falla D, Jull G, Dall’Alba P, et al. An elec-tromyographic analysis of the deep cervi-cal flexor muscles during craniocervicalflexion. Phys Ther. 2003;83:899–906.

40 Jull G, Falla D, Treleaven J, et al. Retrainingcervical joint position sense: The effect oftwo exercise regimes. J Orthop Res. 2006Dec 1: Epub ahead of print.

41 Berg HE, Berggren G, Tesch PA. Dynamicneck strength training effect on pain andfunction. Arch Phys Med Rehabil. 1994;75:661–665.

42 Bronfort G, Evans R, Nelson B, et al. A ran-domized clinical trial of exercise and spinalmanipulation for patients with chronic neckpain. Spine. 2001;26:788–797.

43 Szeto GP, Straker L, Raine S. A field com-parison of neck and shoulder postures insymptomatic and asymptomatic officeworkers. Appl Ergon. 2002;33:75–84.

44 Blatter BM, Bongers PM. Duration of com-puter use and mouse use in relation tomusculoskeletal disorders of neck or up-per limb. International Journal of Indus-trial Ergonomics. 2002;30:295–306.

45 Gore DR, Sepic SB, Gardner GM, MurrayMP. Neck pain: a long term follow-up of205 patients. Spine. 1987;12:1–5.

46 Falla DL, Campbell CD, Fagan AE, et al.Relationship between cranio-cervical flex-ion range of motion and pressure changeduring the cranio-cervical flexion test.Man Ther. 2003;8:92–96.

47 Johnson GM. The correlation between sur-face measurement of head and neck pos-ture and the anatomic position of the up-per cervical vertebrae. Spine. 1998;23:921–927.

Neck Exercise and Sitting Posture in Patients With Chronic Neck Pain

April 2007 Volume 87 Number 4 Physical Therapy f 417 by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from

doi: 10.2522/ptj.20060009Originally published online March 6, 2007

2007; 87:408-417.PHYS THER. Vicenzino and Paul HodgesDeborah Falla, Gwendolen Jull, Trevor Russell, BillWith Chronic Neck PainEffect of Neck Exercise on Sitting Posture in Patients

References

http://ptjournal.apta.org/content/87/4/408#BIBLfor free at: This article cites 40 articles, 6 of which you can access

Cited by

http://ptjournal.apta.org/content/87/4/408#otherarticles

This article has been cited by 8 HighWire-hosted articles:

Information Subscription http://ptjournal.apta.org/subscriptions/

Permissions and Reprints http://ptjournal.apta.org/site/misc/terms.xhtml

Information for Authors http://ptjournal.apta.org/site/misc/ifora.xhtml

by guest on August 20, 2016http://ptjournal.apta.org/Downloaded from