Archives of Physical Medicine and Rehabilitation
Volume 89, Issue 9 , Pages 1795-1802, September 2008

Intraobserver Reliability of Angular and Linear Measurements of Scapular Position in Subjects With and Without Symptoms

  • Jeremy S. Lewis, PhD

      Affiliations

    • Therapy Department, Chelsea and Westminster Hospital, London, UK
    • Physiotherapy Department, St George's Hospital, London, UK
    • University of London, London, UK
    • Corresponding Author InformationReprint requests to Jeremy S. Lewis, PhD, Therapy Department, Chelsea and Westminster Hospital, 369 Fulham Rd, London, SW10 9NH UK
    • ,
  • Rachel E. Valentine, BSc

      Affiliations

    • Therapy Department, Chelsea and Westminster Hospital, London, UK

Article Outline

Abstract 

Lewis JS, Valentine RE. Intraobserver reliability of angular and linear measurements of scapular position in subjects with and without symptoms.

Objective

To assess intraobserver reliability of angular and linear clinical measurements of scapular position.

Design

Test-retest analyses.

Setting

Outpatient department in National Health Service teaching hospital in the United Kingdom.

Participants

Subjects (n=45) without symptoms (21 men, 24 women; age range, 23–56y) and 45 subjects (22 men, 23 women; age range, 19–84y) with shoulder symptoms (defined as pain in the C5-6 dermatome reproduced by shoulder movement and not reproduced with cervical movement).

Interventions

Not applicable.

Main Outcome Measures

Intraclass correlation coefficient (ICC) models 2,1 and 2,3, 95% confidence intervals (CIs), and SE of measurements for 68% confidence and 2 SEs of measurement (for the 95% CI) for the bilateral angular measurements of scapular rotation and tilt, and the bilateral linear measurements of lateral scapular displacement (protraction) and vertical displacement (elevation).

Results

For subjects without symptoms, ICC2,3 results ranged from .75 to .98. The 2 SE results for the angular movements ranged from 1.8° to 2.4° and from 0.4 to 1.0cm for the direct linear measurements. Subjects with symptoms: ICC2,3 results ranged from .61 to .98. The 2 SE results for the angular movements ranged from 1.4° to 2° and from 0.6 to 1cm for the direct linear measurements.

Conclusions

Repeated-measure (ICC2,3) results were more reliable than single-measure (ICC2,1) results. Very good to excellent intraobserver reliability was demonstrated for the angular and linear measurements of interest in both shoulders of subjects with and without symptoms. The 2 SE results provide guidance about the error associated with the individual measurements and will assist the clinician determining whether a change in the static position of the scapula has occurred as a result of intervention or over time. (National Research Register identifier N0060148286.)

Key Words: Posture, Rehabilitation, Scapula, Shoulder

List of Abbreviations: CI, confidence interval, ICC, intraclass correlation coefficient

 

CONTROVERSY EXISTS ABOUT the relevance of posture and muscle imbalance in the etiology of shoulder and upper-quadrant pathology. Particularly, poor scapular posture is a mechanism purported in the pathogenesis of a number of shoulder pathologies.1, 2, 3 Although this is yet to be proven, many clinicians believe that the clinical assessment of scapular position is an essential part of the clinical examination of the shoulder.3, 4, 5, 6 Sahrmann3 has described 2 conditions associated with an abnormal scapular posture, scapular downward rotation syndrome and scapular tilting syndrome, which result in pain and a loss of shoulder function. Scapular asymmetry has been attributed to other shoulder pathologies including impingement syndrome and instability.2, 7 Smith et al8 have demonstrated that in comparison with the natural resting position of the scapula, both retraction and protraction resulted in significantly less shoulder isometric strength. It has also been reported that a downwardly rotated, anteriorly tilted scapula will result in a decreased acromion-humeral distance, resulting in a loss of glenohumeral elevation and subacromial impingement syndrome.1, 9

Numerous techniques to measure the position of the scapula have been proposed. These have included observation,6 palpation,10, 11 radiography,12, 13, 14 goniometry,15 tape measurements,16, 17 and photography.10, 18, 19 These techniques have allowed for 2-dimensional analysis of scapular position. Three-dimensional analysis has been reported using sophisticated radiologic techniques,20, 21 electromechanical devices,22, 23 and electromagnetic equipment.24, 25 Three-dimensional investigations generally require expensive equipment and considerable set-up times and often necessitate specialized software for data interpretation.25 Set-up times, financial concerns, and environmental issues generally preclude 3-dimensional measurements in the clinical setting. As such, clinicians require a battery of clinical tests that produce reliable side-to-side comparisons of scapular position. Reliable multiplanar clinical tests would allow clinicians to measure side-to-side differences and changes in position during the course of clinical intervention. For clinical researchers, reliable tests would facilitate and inform clinical investigations of the relationship of scapular position and pathology. No research that we know of has been published that has investigated the longitudinal impact of scapular position in an asymptomatic population over time. As such, reliable tests for both subjects with and without symptoms are required.

Published studies investigating the reliability of clinically available methods to measure scapular position have generally not included all the commonly described angular and linear positions of the scapula and have not always included the SE of measurement together with the ICC measurements and 95% CIs. Table 1 details published reliability statistics for clinical techniques to measure scapular position.

Table 1. Review of Studies Investigating the Reliability of Clinical Measurements of Scapular Position
StudyMeasurementSubjectsMethodPositionIntrarater ReliabilityInterrater ReliabilityClinical Comment
DiVeta et al16LSD (protraction)AS(n=60)T3 SP to posterior acromial angleStandingICC1,1=.94NTSEM and 95% CI not reported
Neiers and Worrell36LSD (protraction)AS(n=60)T3 SP to posterior acromial angleStandingICC1,1 d=.80, d SEM=9mmNT95% CI not reported
Gibson et al37LSD (protraction)AS(n=32)T3 SP to posterior acromial angleStandingICC (MNS) d=.93, nd=.94, d SEM=6mmICC (MNS) d=.91, nd=.92, d SEM=7mm95% CI not reported
T'Jonck et al38LSD (protraction)AS(n=17)T3 SP to posterior acromial angleSittingICC (MNS) d=.90, nd=.89, d SEM=4mmICC (MNS) d=.45, nd=.72, d SEM=10mm95% CI not reported
Sobush et al39LSD (protraction)AS(n=15)Tx SP to root of spine of scapulaStandingNTICC2,1 d=.80, nd=.66SEM and 95% CI not reported
Peterson et al40LSD (protraction)AS(n=49)T3 SP to medial scapular borderStandingICC2,1=.91 (on left side)NTSEM and 95% CI not reported
DiVeta et al16NSPAS(n=60)T3 SP to AA/RSp to AAStandingICC1,1=.78 (reported by Neiers and Worrell36)NTTested only in AS subjects
Neiers and Worrell36NSPAS(n=60)T3 SP to AA/RSp to AAStandingICC1,1=.34NTTested only in AS subjects
Greenfield et al30NSPAS(n=8)T3 SP to AA/RSp to AAStandingICC (MNS)=.97ICC (MNS)=.96Tested only in AS subjects
T'Jonck et al38NSPAS(n=17)T3 SP to AA/RSp to AASittingICC (MNS)=.87ICC (MNS)=.27Tested only in AS subjects
Gibson et al37Kibler position 1AS(n=32)T8 SP to IAS, arm by sideStandingICC2,1 d=.93, nd=.92, SEM=6mmICC2,1 d=.67, nd=.69, SEM=10mm95% CI not reported
Gibson et al37Kibler position 2AS(n=32)T8 SP to IAS, hand on hipStandingICC2,1 d=.90, nd=.88, SEM=7mmICC2,1 d=.52, nd=.53, SEM=12mm95% CI not reported
Gibson et al37Kibler position 3AS(n=32)T8 SP to IAS, shoulder 90° abductionStandingICC2,1 d=.85, nd=.90, SEM=8mmICC2,1 d=.28, nd=.18, SEM=17mm95% CI not reported
Sobush et al39Kibler position 1AS(n=15)Tx SP to inferior angle scapulaStandingNTICC2,1 d=0.77, nd=.86SEM and 95% CI not reported
T'Jonck et al38Kibler position 1AS(n=17)Tx SP to IAS, arm by sideSittingICC (MNS) d=.93, nd=.83, d SEM=4mmICC (MNS) d=.78, nd=.72, d SEM=6mm95% CI not reported
T'Jonck et al38Kibler position 2AS(n=17)Tx SP to IAS, hand on hipSittingICC (MNS) d=.90, nd=.80, d SEM=4mmICC (MNS) d=.87, nd=.78, d SEM=5mm95% CI not reported
T'Jonck et al38Kibler position 3AS(n=17)Tx SP to IAS, shoulder 90° abductionSittingICC (MNS) d=.96, nd=.93, d SEM=5mmICC (MNS) d=.89, nd=.90, d SEM=8mm95% CI not reported
Odom et al41Kibler position 1AS(n=26)Tx SP to IAS, arm by sideStandingICC1,1=.75; 95% CI, .61−.85; SEM=6mmICC1,1=.67; 95% CI, .25–.85; SEM=8mmBilateral measurements pooled
Odom et al41Kibler position 2AS(n=26)Tx SP to IAS, hand on hipStandingICC1,1=.77; 95% CI, .60−.86; SEM=6mmICC1,1=.43; 95% CI, −.29 to .75; SEM=11mmBilateral measurements pooled
Odom et al41Kibler position 3AS(n=26)Tx SP to IAS, shoulder 90° abductionStandingICC1,1=.80; 95% CI, .65−.88; SEM=8mmICC1,1=.74; 95% CI, .41–.88; SEM=12mmBilateral measurements pooled
Odom et al41Kibler position 1S(n=20)Tx SP to IAS, arm by sideStandingICC1,1=.52; 95% CI, .10−.74; SEM=8mmICC1,1=.79; 95% CI, .46–.91; SEM=8mmBilateral measurements pooled
Odom et al41Kibler position 2S(n=20)Tx SP to IAS, hand on hipStandingICC1,1=.66; 95% CI, .36−.82; SEM=6mmICC1,1=.45; 95% CI, −.38 to .78; SEM=8mmBilateral measurements pooled
Odom et al41Kibler position 3S(n=20)Tx SP to IAS, shoulder 90° abductionStandingICC1,1=.62; 95% CI, .27−.79; SEM=9mmICC1,1=.57; 95% CI, −.23 to .85; SEM=11mmBilateral measurements pooled
McKenna et al42Kibler position 1AS(n=15)T7/8 SP to IAS, arm by sideStandingNTICC2,1 d=.65, nd=.74, d SEM=6mmd 95% CI, .38−.85
McKenna et al42Kibler position 2AS(n=15)T7/8 SP to IAS, hand on hipStandingNTICC2,1 d=.79, nd=.82, d SEM=6mmd 95% CI, .60−.92
McKenna et al42Kibler position 3AS(n=15)T7/8 SP to IAS, shoulder 90° abductionStandingNTICC2,1 d=.48, nd=.20, d SEM=12mmd 95% CI, .16−.75
McKenna et al42Shoulder flexionAS(n=15)T7/8 SP to IAS, full shoulder flexionStandingNTICC2,1 d=.57, nd=.71, d SEM=12mmd 95% CI, .27−.81
Greenfield et al30Scapular rotationAS(n=8)Tan SPIAS to IAS/SPRSp to SP IASStandingICC (MNS)=.97ICC (MNS)=.97SEM and 95% CI not reported
Sobush et al39Scapular rotationAS(n=15)Tan ([SPIAS to IAS] – [SPRSp to RSp])/SPRSp to SPIASStandingNTICC2,1 d=.64, nd=.84SEM and 95% CI not reported
Johnson et al34Scapular rotation
AS(n=23)

S(n=16)

Modified digital inclinometer on spine of scapula (at rest, 60°, 90°, 120°)StandingICC3,1 range, .89−.96; SEM range, 2.0°−2.8°; 95% CI, .80−.98NTMeasurement on d side more reliable than on nd side. SEM comparable at all levels
Watson et al35Scapular rotationS(n=26)Inclinometer on spine of scapula (at 5 stages of humeral elevation from rest to end of range)StandingICC (MNS) range, .81−.94; SEM range, 1.7°−5.2°; 95% CI, 0.67−1.00NTAt rest, most reliable 135°, least reliable
Plafcan et al43Inferior angle (RtSaFP)AS(n=11)Scapular measurement instrumentSittingICC (MNS) range, .67−.87; SEM range, 4−8mmICC (MNS) range, .68−.81; SEM range, 6−8mm95% CI not reported
Inferior angle (RtSaFP)AS(n=11)Scapular measurement instrumentReachICC (MNS) range, .60−.91; SEM range, 4−8mmICC (MNS) range, .48−.66; SEM range, 8−10mm95% CI not reported
Spine of scapula (RtSaFP)AS(n=11)Scapular measurement instrumentSittingICC (MNS) range, .68−.86; SEM range, 4−8mmICC (MNS) range, .61−.68; SEM range, 7−8mm95% CI not reported
Spine of scapula (RtSaFP)AS(n=11)Scapular measurement instrumentReachICC (MNS) range, .51−.95; SEM range, 4−8mmICC (MNS) range, .29−.86; SEM range, 5−10mm95% CI not reported

Abbreviations: AA, posterior acromial angle; AS, asymptomatic; d, dominant; IAS, inferior angle of scapula; Kibler position 1, arms relaxed by the side; Kibler position 2, hands on hips, thumb facing posteriorly, with thumb and index finger placed on lateral iliac crest; Kibler position 3, with the shoulder in maximum internal rotation with arms abducted to 90° in the coronal plane; LSD (protraction), lateral scapular displacement—protraction; MNS, model not stated; nd, nondominant; NSP, normalized scapular position; NT, not tested; reach, reaching with the hand to the opposite shoulder blade; RSp, root of the spine of the scapula; RtSaFP, relative to sagittal and frontal planes; S, symptomatic, SEM, SE of measurement; SP, spinous process; SPIAS to IAS, distance from the thoracic spinous process at the level of the inferior angle of the scapula to the inferior angle of the scapula; SPRSp to RSp, distance from the spinous process at the root of the spine of the scapula to the root of the spine of the scapula; SPRSp to SPIAS, distance between the thoracic spinous process at the level of the root of the thoracic spinous process at the level of the inferior angle of the scapula; Tan, tangent; Tx, thoracic.

The need for reliable quantitative measurements is highlighted by research that has demonstrated visual measurements of scapular position lack adequate and acceptable clinical reliability.5 As such, the aim of this investigation was to determine the intratester reliability of a series of static linear and angular positions of the scapula during natural posture in subjects with and without shoulder symptoms.

Back to Article Outline

Methods 

Participants 

Subjects with symptoms were recruited through the orthopedic and physical therapy outpatient department in the teaching hospital where the study was conducted. Symptoms included pain in the C5-6 dermatome at rest or reproduced on shoulder movement and not influenced by cervical movement. Because this study involved the reliability of measuring scapular position, subjects with acute and chronic conditions with varying durations and severities were included. Subjects with symptoms were assessed by their treating physical therapist for pain and loss of normative movement in the shoulder. Subjects without symptoms were recruited through personal and public advertisements. Permission to conduct this study was granted by the local research ethics committee. All subjects signed informed consent documents and were aware of all their rights, including the right to withdraw from the study at any stage of the investigation. Potential participants were approached until the required number was recruited.

Inclusion and Exclusion Criteria 

Participants for this investigation included men and women over the age of 18 years. For subjects with symptoms, inclusion criteria were acute (<6wk), subacute (6–12wk), and chronic (>3mo) pain conditions, and/or restriction of movement arising from the area of the shoulder (C5-6 dermatome, not reproduced with movement of the cervical spine). For the participants without symptoms, inclusion criteria were no lumbar, thoracic, cervical, shoulder, or upper-limb symptoms. For both groups, exclusion criteria were an inability to communicate fully in English; cardiac, respiratory, kidney, or circulatory problems; systemic disease; diabetes; and pregnancy. For subjects without symptoms, additional exclusion criteria were a history of fractures, treatment, or surgery to the lumbar spine, thoracic spine, cervical spine, and upper limbs.

Procedure 

Measurements were made with the patient in standing. To facilitate a natural posture, subjects were requested to swing their arms gently backward and forward 3 times by their sides and stop in a position that felt natural and comfortable to them; to flex and extend their head 3 times gently and stop in a position that felt natural and comfortable to them; and to take 3 breaths and adopt a position that felt natural and comfortable to them. These identical instructions were given to each subject prior to each data collection period. Once the subject's natural posture had been achieved, self-adhesive markers 6mm in diameter were placed directly on the skin over the following landmarks bilaterally: the posterior aspect of the acromion (tubercle at lateral end of spine of scapula), the root of the spine of the scapula, the thoracic spinous process at the level of with the root of the spine of the scapula, the inferior angle of the scapula, the spinous process corresponding with the inferior angle of the scapula, and the spinous process corresponding with the level of the iliac crests. The examiner has had approximately 20 years of clinical experience and works as a consultant physiotherapist in the National Health Service in the United Kingdom. To avoid measurement confusion, 2 sets of differently colored markers were used, 1 set for the landmarks on the left side of the body and 1 set for the right side. The anatomic reference points used as locations to make measurements are detailed in figure 1.

  • View full-size image.
  • Fig 1. 

    Anatomic reference points. A, Spinous process corresponding with the root of the spine of the scapula; B, root of the spine of the scapula; C, lateral end of the spine of the scapula; D, spinous process corresponding with the inferior angle of the scapula; E, inferior angle of the scapula; F, T12 spinous process.

Data Collection 

Postural data were collected from each subject on 2 occasions, separated by a 30-minute interval. Random number tables were used to determine the side to be tested first. The investigator made the postural measurements using the center of the adhesive marks as the reference points and verbally relayed them to an assistant who entered all data measurements. To ensure recall bias would not influence the results, the postural measurements for any 1 subject were staggered so that the first set of postural measurements made in subject 1 was followed by the first set of postural measurements in subject 2. The adhesive marks were removed at the end of each data collection period. Half an hour after this, the second set of measurements was made in subject 1, and after this, the second set of measurements in subject 2. The 30-minute interval further ensured that any redness from the area of the removed markers would have disappeared. It was determined that postural measurements would be made by 1 investigator because this method most closely resembles the method used in the clinical environment. Scapular measurements were made in the same order for each subject. For each subject, the order for measuring left and right shoulders was reversed for the second data collection period.

Measurements of Scapular Position 

This investigation involved measuring a series of linear and angular measurements. With respect to figure 1, linear measurements were made using a standard nonstretch fiberglass tape measure bilaterally between the following landmarks: AB, AC, BC, DE, FE, and FD. Angular measurements were made bilaterally with a commercially available gravity-dependent inclinometera between the points BC and BE.

Power Analysis 

This study formed part of a series of studies aiming to investigate measurement reliability as well as relationships of posture for the shoulder and upper body. Walter et al26 have provided estimates for sample size requirements for reliability studies using ICCs. For a true p0 (the minimally acceptable level of reliability) of 0.7 against an alternative p1 (alternative level of reliability) of 0.9, based on a 5% significance level and a power of 80% (β=.20) for 2 raters, or 2 time points, 19 subjects are required.26, 27 Forty-five subjects were recruited into each group (N=90). This number of subjects was considered adequate to determine the intrarater reliability for measuring the linear and angular measurements of interest in this study.

Statistical Analysis 

The reliability of the measurements was analyzed using ICC, 95% CI, and SE of measurement. The descriptive statistics, ICC model 2, 95% CI, and the statistics were analyzed using SPSSb software. Portney and Watkins28 have described 6 different equations for calculating ICC and have argued that model 2 should be used when wishing to generalize confidently the findings of a reliability trial of a particular method of measurement to equally trained clinicians, and model 3 should be selected when an investigator is interested in establishing the reliability of a measurement procedure for 1 specific data collection experience without the intention to generalize the findings to equally trained clinicians.28(p562) ICC model 2 was used in the current analysis. The analysis of reliability involved determining the reliability of (1) the first measurement (ICC2,1) and (2) the mean of the 3 measurements (ICC2,3). Using SPSS software, ICC2,1 was analyzed by selecting the options 2-way random, single measure, and absolute agreement, and ICC2,3 was analyzed by selecting 2-way random, average measure, and absolute agreement. Portney and Watkins28 suggested that ICC values above .75 are indicative of good reliability, and those below .75 should be considered as poor to moderate, and in addition, reliability should exceed .90 to ensure reasonable validity. Clinicians may be 68% certain that the true measurement value lies within ±1 SE of measurement from the clinical measurement. Two SEs of measurement provide the clinician with 95% confidence.

Back to Article Outline

Results 

Forty-five subjects without symptoms (21 men [47%]; 24 women [53%]; combined mean age, 32y; range, 23–56y; mean height, 1.7m; range, 1.6–1.9m; mean weight, 70.4kg; range, 50–110kg) were recruited. Forty-five subjects with symptoms were recruited (22 men [49%]; 23 women [51%]; combined mean age, 43y; range, 19–84y; mean height, 1.7m; range, 1.5–1.9m; mean weight, 71.4kg; range, 49–90kg). Diagnostic categories for the subjects with symptoms included nonspecific shoulder pain (n=21), rotator cuff tendinopathy (n=12), frozen shoulder (n=2), acromioclavicular joint pathology (n=2), glenohumeral instability (n=2), stable humeral fractures (n=1), stable clavicular fractures (n=4), and stable scapular fractures (n=1).

For the subjects without symptoms, all ICC2,3 results exceeded .75. With the exception of the AC/BC ratio on the right for subjects without symptoms, all the ICC2,1 results exceeded .75. With the exception of the AC/BC ratio on the left and right and the left DE measurement, all the ICC2,3 results exceeded the 0.9 threshold. The 2 SEs of measurement results for all the measurements ranged from 0.40 to 1.00cm for the linear measurements and from 1.8° to 2.4° for the angular measurements. Table 2 details the reliability statistics for the angular and linear measurements for the subjects without symptoms.

Table 2. Reliability Statistics for the Subjects Without Symptoms
MeasurementMean ± SDRangeICC2,1 (95% CI)ICC2,3 (95% CI)SEM From ICC2,32 SEM From ICC2,3
Angular measurements
Left rotation−3.6°±3.9°−13.0°to6.0°.90(.82–.94).95(.90–.97)0.9°1.8°
Right rotation−2.9°±4.1°−12.0°to7.0°.84(.72–.91).92(.83–.96)1.2°2.4°
Left tilt12.8°±5.6°−4.0°to24.0°.94(.88–.97).97(.94–.98)1.0°2.0°
Right tilt14.9°±6.2°−3.0°to27.0°.95(.95–.99).98(.95–.99)0.9°1.8°
Linear measurements
AC left21.3±1.7cm18.0to25.7cm.95(.92–.98).98(.96–.99)0.3cm0.6cm
AC right21.8±2.1cm18.8to26.5cm.89(.80–.94).94(.89–.97)0.5cm1.0cm
BC left13.8±1.2cm10.8to16.0cm.89(.82–.94).95(.90–.97)0.3cm0.6cm
BC right13.6±1.2cm11.2to16.0cm.90(.82–.94).95(.90–.97)0.3cm0.6cm
AB left7.6±1.0cm5.5to10.6cm.90(.83–.95).95(.91–.97)0.2cm0.4cm
AB right8.5±1.2cm5.9to10.9cm.97(.95–.98).98(.97–.99)0.2cm0.4cm
DE left9.0±1.3cm6.3to12.3cm.79(.66–.88).88(.79–.94)0.5cm1.0cm
DE right9.5±1.4cm6.7to13.3cm.87(.73–.94).93(.84–.97)0.4cm0.8cm
FE left25.6±2.4cm19.2to30.2cm.93(.88–.96).97(.94–.98)0.4cm0.8cm
FE right25.1±2.4cm18.4to30.3cm.92(.85–.95).96(.92–.98)0.5cm1.0cm
FD left23.3±2.3cm16.6to28.5cm.91(.84–.95).95(.91–.97)0.5cm1.0cm
FD right23.1±2.3cm16.8to28.5cm.90(.83–.95).95(.91–.97)0.5cm1.0cm
AC/BC leftNANA.81(.69–.89).89(.81–.94)NANA
AC/BC rightNANA.61(.38–.76).76(.55–.87)NANA

Abbreviations: NA, not applicable; SEM, SE of measurement.

For the subjects with symptoms, the reliability for all the angular measurements exceeded 0.9 for both the ICC2,1 and ICC2,3 results. The 2 SEs of measurement results ranged from 1.4° to 2.0°. With the exception of the AC/BC ratios for the left and right sides, all the ICC2,3 results for the linear measurements exceeded .90. The 2 SEs of measurement results for these measurements ranged from 0.60 to 1.00cm. The bilateral pairs of measurements that exceeded the .90 threshold for the ICC2,1 results were the AC and FE measurements. Table 3 details the reliability statistics for the angular and linear measurements for the subjects with symptoms.

Table 3. Reliability Statistics for the Subjects With Symptoms
MeasurementMean ± SDRangeICC2,1 (95% CI)ICC2,3 (95% CI)SEM From ICC2,32 SEM From ICC2,3
Angular measurements
Left rotation−3.9°±4.6°−11.0°to7.0°.92(.86–.95).96(.92–.98)0.9°1.8°
Right rotation−3.6°±4.1°−13.0°to8.0°.94(.89–.97).97(.95–.98)0.7°1.4°
Left tilt12.8°±5.3°−5.0°to26.5°.94(.89–.97).97(.94–.98)1.0°2.0°
Right tilt14.5°±5.6°−3.0°to25.0°.95(.93–.98).98(.96–.99)0.8°1.6°
Linear measurements
AC left21.2±1.7cm17.6to25.3cm.94(.90–.97).97(.95–.98)0.3cm0.6cm
AC right21.8±1.9cm17.7to26.0cm.94(.90–.97).97(.95–.98)0.3cm0.6cm
BC left13.7±1.1cm11.0to17.1cm.89(.80–.94).94(.89–.97)0.3cm0.6cm
BC right13.6±1.1cm10.9to16.4cm.83(.71–.90).91(.83–.95)0.3cm0.6cm
AB left7.6±1.0cm5.8to10.1cm.83(.72–.91).91(.83–.95)0.3cm0.6cm
AB right8.3±1.2cm5.2to10.8cm.90(.83–.94).95(.90–.97)0.3cm0.6cm
DE left9.0±1.4cm6.5to12.7cm.89(.81–.94).93(.89–.97)0.4cm0.8cm
DE right9.6±1.4cm6.1to12.5cm.92(.86–.95).96(.92–.98)0.3cm0.6cm
FE left25.3±2.0cm20.8to29.2cm.94(.89–.97).97(.94–.98)0.4cm0.8cm
FE right25.0±2.0cm20.7to30.2cm.92(.86–.96).96(.93–.98)0.4cm0.8cm
FD left23.1±1.9cm18.8to27.6cm.91(.84–.95).95(.92–.97)0.5cm1.0cm
FD right23.0±2.0cm18.7to27.5cm.89(.81–.94).94(.89–.98)0.5cm1.0cm
AC/BC leftNANA.69(.51–.82).82(.68–.91)NANA
AC/BC rightNANA.61(.38–.76).75(.55–.86)NANA

Abbreviations: NA, not applicable; SEM, SE of measurement.

Back to Article Outline

Discussion 

Although assessment of scapular position is considered to be an important component of the clinical examination of the shoulder, conflicting clinical and research findings1, 3, 16, 17, 18, 19, 29, 30, 31 suggest further investigations are necessary. To date, no study known to us has investigated the reliability of a series of static angular and linear clinical measurements of the scapula. In addition, we are unaware of any study that has measured scapular tilt using a simple clinical procedure. The findings of this investigation suggest that the series of tests used in this investigation demonstrate acceptable reliability to measure the static position of the scapula for subjects with and without symptoms. The associated 2 SEs of measurements provide clinicians with 95% confidence that the true angular or linear measurement lies between 2 SEs of measurements from the measurement made. This would allow clinicians to determine whether a real change has occurred to the static resting position of the scapula in subjects with or without symptoms as a consequence of intervention or time. For clinicians interested in measuring the static position of the scapula, the following protocol is suggested: measurements AC, AB, and DE for the lateral linear displacement (protraction, retraction) of the scapula. The normalized scapular protraction measurement (AC/AB) described by DiVeta et al16 is not recommended. The distance DF may be measured for an appreciation of the amount of scapular elevation above the T12 spinous process. Although this measurement involves 2 separate measurements (ie, DE, DF), which compound the error, it is clinically more relevant than the direct measurement FE. This is because the distance FE would change if the measurement DE increased or decreased over time even if no real increase or decrease in scapular elevation and/or depression had occurred. In addition, the angular measurements of scapular tilt and rotation may be used by clinicians requiring a more quantitative assessment of angular scapular position. The bilateral assessment of the angular and linear position of the scapula as described in this investigation takes approximately 5 minutes.

In this investigation, the mean of 3 measurements (ICC2,3) was more reliable than 1 measurement (ICC2,1), and we recommend that clinicians using these methods measure the required distances on 3 occasions and use the mean measurement in their calculations.

Kibler7 argued that a difference in the distance between the inferior angle of the scapula and the corresponding thoracic spinous process with the arm by the side (distance DE) of greater than 1cm indicated substantial asymmetry and the presence of shoulder dysfunction. In a later publication, Kibler2 reported that the threshold of asymmetry was 1.5cm. The 2 SEs of measurement results from this study suggest that this distance may be measured reliably within the 1.5cm threshold. Kibler2 also reported that there should not be greater than 1.5cm asymmetry in this measurement when the subjects' hands are placed on the hips and when the shoulder is abducted to 90°. Kibler2, 7 did not measure the DE distance in asymptomatic subjects, assuming scapular symmetry to be the norm. Koslow et al32 questioned the validity of Kibler's2, 7 claim and reported that approximately 73% of asymptomatic athletes demonstrated asymmetry of this measurement in 1 or more of the 3 positions described by Kibler.2, 7 These equivocal findings together with the beliefs pertaining to scapular posture demonstrate the pressing need for ongoing research in this area to understand better whether a correlation between posture and symptoms exists. This confusion is highlighted by findings reported by McClure et al,33 who reported that although subjects with subacromial impingement syndrome improved after a 6-week exercise program, there was no change in scapular position or kinematics.

The results of the current investigation appear to be very comparable to those reported by Johnson34 and Watson35 and colleagues in the measurement of scapular rotation. Table 4 compares the reliability results for scapular rotation at rest for the current study and for those of Johnson34 and Watson.35

Table 4. Comparison of Studies Using Direct Clinical Measurements of Scapular Rotation
StudySubjectsICC (95% CI)SEM (deg)2 SEM (deg)
Johnson et al34AS (n=23), S (n=16)ICC3,1
Groups mixed for reliability studyDominant arm, .91 (.82–.95)2.04.0
Nondominant arm, .91 (.82–.95)2.04.0
Watson et al35S (n=26)ICC (MNS)
Painful or most painful shoulder examinedPainful arm, .94 (.90–.99)1.73.4
Current studyAS (n=45), S (n=45)ICC2,3
Both shoulders tested in both groups (ie, symptomatic and asymptomatic side of subjects with symptoms)Asymptomatic subjects
Left arm, .95 (.90–.97)0.91.8
Right arm, .92 (.82–.95)1.22.4
Symptomatic subjects
Left arm, .96 (.92–.98)0.91.8
Right arm, .97 (.95–.98)0.71.4

Abbreviations: AS, asymptomatic; MNS, model not stated; S, symptomatic; SEM, SE of measurement.

The method employed in this study and those of Johnson34 and Watson35 indirectly measures scapular rotation through the scapular spine. There is no certainty how this measurement correlates with the sagittal plane angle of the glenoid fossa. There is also no certainty that the bony architecture of the spine is consistent between subjects or is symmetrical side-to-side, and it is possible that asymmetrical muscle activity between the dominant and nondominant shoulders may lead to differences in the development of the scapular spine. However, this measurement may be of benefit in determining changes in overall scapular rotation over time or as a consequence of a treatment program. It would be advantageous for future research to establish the relationship between the angulation of the glenoid fossa and the scapular spine.

Study Limitations 

There are a number of limitations of this study. The first relates to the static measurement of the scapula in standing with the arm by the side. Other studies using sophisticated electromagnetic and electromechanical devices23, 25 have detected differences in the position of the scapula between subjects with and without symptoms with the shoulder in different angles of elevation. Future research is warranted to determine whether the current method is reliable at higher ranges of shoulder elevation. However, taking 1 or 3 measurements at 90°, 120°, and full elevation, as other studies have, would lead to the confounding issue of requiring subjects to maintain static positions with isometric muscle contractions for prolonged periods against the normalized pattern of continuous shoulder movement. Another limitation is that only intratester reliability has been assessed in this study. Although this is similar to other investigations,16, 35 future research should aim to establish the interrater reliability of the techniques used in the current investigation.

Back to Article Outline

Conclusions 

Poor scapular posture and movement have been suggested as mechanisms in the pathogenesis of a number of shoulder pathologies. Although this is yet to be proven, many clinicians believe that the clinical assessment of scapular position is an essential part of the clinical examination of the shoulder. The findings of the current study provide clinicians with clinically accessible measurements to determine the static and angular position of the scapula (when the arm is by the side) in subjects with and without symptoms.

Suppliers

Back to Article Outline

Acknowledgment 

We thank the members of the physiotherapy outpatient department at Chelsea and Westminster Hospital for their support of this investigation.

Back to Article Outline

References 

  1. Grimsby O, Gray J. Interrelation of the spine to the shoulder girdle. In:  Donatelli R editors. Physical therapy of the shoulder. 3rd ed.. New York: Churchill Livingstone; 1997;p. 95–129
  2. Kibler WB. The role of the scapula in athletic shoulder function. Am J Sports Med. 1998;26:325–337
  3. Sahrmann S. Diagnosis and treatment of movement impairment syndromes. London: Mosby; 2002;
  4. Mottram SL. Dynamic stability of the scapula. Man Ther. 1997;2:123–131
  5. Kibler WB, Uhl TL, Maddux JW, Brooks PV, Zeller B, McMullen J. Qualitative clinical evaluation of scapular dysfunction: a reliability study. J Shoulder Elbow Surg. 2002;11:550–556
  6. Kendall F, McCreary E, Provance P. Muscle testing and function. 4th ed.. Baltimore: Williams & Wilkins; 1993;
  7. Kibler W. Role of the scapula in the overhead throwing motion. Contemp Orthop. 1991;22:525–532
  8. Smith J, Kotajarvi BR, Padgett DJ, Eischen JJ. Effect of scapular protraction and retraction on isometric shoulder elevation strength. Arch Phys Med Rehabil. 2002;83:367–370
  9. Gray J, Grimsby O. Interrelationship of the spine, rib cage, and shoulder. In:  Donatelli R editors. Physical therapy of the shoulder. 4th ed.. Edinburgh: Churchill Livingston; 2004;p. 133–185
  10. Bagg SD, Forrest WJ. A biomechanical analysis of scapular rotation during arm abduction in the scapular plane. Am J Phys Med Rehabil. 1988;67:238–245
  11. Lewis J, Green A, Reichard Z, Wright C. Scapular position: the validity of skin surface palpation. Man Ther. 2002;7:26–30
  12. Freedman L, Munro R. Abduction of the arm in the scapular plane: scapular and gleno-humeral movements: a roentgenographic study. J Bone Joint Surg Am. 1966;48:1503–1510
  13. Poppen NK, Walker PS. Forces at the glenohumeral joint in abduction. Clin Orthop Relat Res. 1978;(135):165–170Sep
  14. Michiels I, Grevenstein J. Kinematics of shoulder abduction in the scapular plane: on the influence of abduction velocity and external load. Clin Biomech. 1995;10:137–143(Bristol, Avon)
  15. Doody SG, Freedman L, Waterland JC. Shoulder movements during abduction in the scapular plane. Arch Phys Med Rehabil. 1970;51:595–604
  16. DiVeta J, Walker ML, Skibinski B. Relationship between performance of selected scapular muscles and scapular abduction in standing subjects. Phys Ther. 1990;70:470–476discussion 6-9
  17. Lewis JS, Wright C, Green A. Subacromial impingement syndrome: the effect of changing posture on shoulder range of movement. J Orthop Sports Phys Ther. 2005;35:72–87
  18. Raine S, Twomey LT. Head and shoulder posture variations in 160 asymptomatic women and men. Arch Phys Med Rehabil. 1997;78:1215–1223
  19. Lewis JS, Green A, Wright C. Subacromial impingement syndrome: the role of posture and muscle imbalance. J Shoulder Elbow Surg. 2005;14:385–392
  20. Boyd EA, Goudreau L, O'Riain MD, Grinnell DM, Torrance GM, Gaylard A. A radiological measure of shoulder subluxation in hemiplegia: its reliability and validity. Arch Phys Med Rehabil. 1993;74:188–193
  21. Prevost R, Arsenault AB, Dutil E, Drouin G. Rotation of the scapula and shoulder subluxation in hemiplegia. Arch Phys Med Rehabil. 1987;68:786–790
  22. Kebaetse M, McClure P, Pratt NA. Thoracic position effect on shoulder range of motion, strength, and three-dimensional scapular kinematics. Arch Phys Med Rehabil. 1999;80:945–950
  23. Lukasiewicz AC, McClure P, Michener L, Pratt N, Sennett B. Comparison of 3-dimensional scapular position and orientation between subjects with and without shoulder impingement. J Orthop Sports Phys Ther. 1999;29:574–583discussion 84-6
  24. Culham EG, Noce RR, Bagg SD. Shoulder complex position and glenohumeral subluxation in hemiplegia. Arch Phys Med Rehabil. 1995;76:857–864
  25. Ludewig PM, Cook TM. Alterations in shoulder kinematics and associated muscle activity in people with symptoms of shoulder impingement. Phys Ther. 2000;80:276–291
  26. Walter SD, Eliasziw M, Donner A. Sample size and optimal designs for reliability studies. Stat Med. 1998;17:101–110
  27. Sim J, Wright C. Research in health care: concepts, designs and methods. Cheltenham: Stanley Thornes; 2000;
  28. Portney L, Watkins M. Foundations of clinical research: applications to practice. 2nd ed.. Upper Saddle River: Prentice Hall Health; 2000;
  29. Ayub E. Posture and the upper quarter. In:  Donatelli R editors. Physical therapy of the shoulder. 2nd ed.. Melbourne: Churchill Livingstone; 1991;p. 81–90
  30. Greenfield B, Catlin PA, Coats PW, Green E, McDonald JJ, North C. Posture in patients with shoulder overuse injuries and healthy individuals. J Orthop Sports Phys Ther. 1995;21:287–295
  31. Lewis JS, Valentine RE. The pectoralis minor length test: a study of the intra-rater reliability and diagnostic accuracy in subjects with and without shoulder symptoms. BMC Musculoskelet Disord. 2007;8:64
  32. Koslow PA, Prosser LA, Strony GA, Suchecki SL, Mattingly GE. Specificity of the lateral scapular slide test in asymptomatic competitive athletes. J Orthop Sports Phys Ther. 2003;33:331–336
  33. McClure PW, Michener LA, Sennett BJ, Karduna AR. Direct 3-dimensional measurement of scapular kinematics during dynamic movements in vivo. J Shoulder Elbow Surg. 2001;10:269–277
  34. Johnson MP, McClure PW, Karduna AR. New method to assess scapular upward rotation in subjects with shoulder pathology. J Orthop Sports Phys Ther. 2001;31:81–89
  35. Watson L, Balster SM, Finch C, Dalziel R. Measurement of scapula upward rotation: a reliable clinical procedure. Br J Sports Med. 2005;39:599–603
  36. Neiers L, Worrell TW. Assessment of scapular position. J Sport Rehabil. 1993;2:20–25
  37. Gibson MH, Goebel GV, Jordan TM, Kegerreis S, Worrell TW. A reliability study of measurement techniques to determine static scapular position. J Orthop Sports Phys Ther. 1995;21:100–106
  38. T'Jonck L, Lysens R, Grasse G. Measurements of scapular position and rotation: a reliability study. Physiother Res Int. 1996;1:148–158
  39. Sobush DC, Simoneau GG, Dietz KE, Levene JA, Grossman RE, Smith WB. The Lennie test for measuring scapular position in healthy young adult females: a reliability and validity study. J Orthop Sports Phys Ther. 1996;23:39–50
  40. Peterson DE, Blankenship KR, Robb JB, et al. Investigation of the validity and reliability of four objective techniques for measuring forward shoulder posture. J Orthop Sports Phys Ther. 1997;25:34–42
  41. Odom CJ, Taylor AB, Hurd CE, Denegar CR. Measurement of scapular asymmetry and assessment of shoulder dysfunction using the Lateral Scapular Slide Test: a reliability and validity study. Phys Ther. 2001;81:799–809
  42. McKenna L, Cunningham J, Straker L. Inter-tester reliability of scapular position in junior elite swimmers. Phys Ther Sport. 2004;5:146–155
  43. Plafcan DM, Canavan PK, Sebastianelli WJ, Swope KM. Reliability of a new instrument to measure scapular position. J Man Manipulative Ther. 2000;8:183–192
  • a Isomed Inc, 975 SE Sandy Blvd, Portland, OR 97214.
  • b Version 14; SPSS Ltd, St Andrews House, West St, Woking, Surrey, GU21 6EB, UK.

 Supported by the Westminster Medical School Research Trust and the Chelsea and Westminster Healthcare National Health Service Trust Charity.

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.

PII: S0003-9993(08)00403-6

doi:10.1016/j.apmr.2008.01.028

Archives of Physical Medicine and Rehabilitation
Volume 89, Issue 9 , Pages 1795-1802, September 2008

Article Tools

* Image Usage & Resolution