low back pain: clinimetric properties of the trendelenburg test, active straight leg raise test, and...

LOW BACK PAIN: CLINIMETRIC PROPERTIES OF THE

TRENDELENBURG TEST, ACTIVE STRAIGHT LEG RAISE

TEST, AND BREATHING PATTERN DURING ACTIVE

STRAIGHT LEG RAISING

Nathalie A. Roussel, MT, PT,a Jo Nijs, PhD,b Steven Truijen, PhD,c Line Smeuninx, PT,d andGaetane Stassijns, MD, PhDe

ABSTRACT

270

a Teacher, Divisment of Health ScResearch Fellow,itation, AntwerpUniversity of Antw

b Assistant ProfeDepartment of HealAssistant ProfessoEducation and Phys

c Assistant ProfeDepartment of Heal

d Physiotherapisletal PhysiotherapCollege Antwerp,

e Assistant ProfRehabilitation, AnUniversity of AntwSubmit requests

Campus HIKE-DAertselaerstraat 31(e-mail: N.RousselPaper submitted

12, 2006; accepted0161-4754/$32.Copyright D 20doi:10.1016/j.jm

Objective: Classification of patients with low back pain (LBP) into subgroups is important as considerable variability

exists in the LBP population. Clinical applicable, reliable, and valid tests to differentiate patients with LBP are therefore

necessary. The purpose of this study is to examine the reliability, internal consistency, and clinical importance of 3 clinical

tests that analyze motor control mechanisms of the lumbopelvic region in patients with nonspecific LBP.

Methods: Thirty-six patients with chronic nonspecific LBP volunteered for the study (cross-sectional design). The patientswere examined by 2 assessors who were blinded to the results of each other. The following tests were performed: the

Trendelenburg test, the active straight leg raise (ASLR) test, and the ASLR with visual inspection of the breathing pattern.

Results: The test-retest reliability coefficients (j) were greater than 0.75 for the Trendelenburg score and greater than 0.70for the ASLR. The interobserver reliability coefficients were greater than 0.39 for the assessment of the breathing pattern

during the ASLR. The Cronbach a coefficient for internal consistency of the Trendelenburg and ASLR tests was greater than

.73. No significant associations were found between the outcome of the tests and self-reported pain severity or disability.

Conclusions: These data provide evidence favoring the test-retest reliability of the Trendelenburg and ASLR tests in

patients with LBP. The internal consistency of the outcome of these tests was high for both assessors, suggesting that these

tests assess the same dimension. The interobserver reliability of the assessment of the breathing pattern was fair to moderate.

Further research regarding the interobserver reliability, clinical importance, validity, and responsiveness of the Trendelenburg

test, ASLR test, and breathing pattern during these tests is required. (J Manipulative Physiol Ther 2007;30:270-278)

Key Indexing Terms: Low Back Pain; Reproducibility of Results; Respiration; Observation; Physical Examination

ion of Musculoskeletal Physiotherapy, Depart-iences, University College Antwerp, Belgium;Department of Physical Medicine and ReUniversity Hospital; Faculty of Mederp, Belgium.ssor, Division of Musculoskeletal Physiotth Sciences, University College Antwerp, Ber, Spinal Research Group, Faculty of Piotherapy, Vrije Universiteit Brussel, Belgiussor, Division of Musculoskeletal Physiotth Sciences, University College Antwerp, Bet, Assistant Professor, Division of Muscuy, Department of Health Sciences, UniBelgium.essor, Department of Physical Medicintwerp University Hospital; Faculty of Meerp, Belgium.for reprints to: Nathalie A. Roussel, Mepartement G, Hogeschool Antwerpen, 2170 Merksem, [email protected]).September 7, 2006; in revised form DecJanuary 2, 2007.

0007 by National University of Health Sciept.2007.03.001

The disability caused by low back pain (LBP) leads

to a high socioeconomic impact. Despite the

growing research addressing assessment and treat-

ment strategies, a considerable percentage of patients

continue to be affected by lower back problems. Today,

LBP is perceived as a multidimensional problem.1 Pathoa-

natomical, physical, neurophysiological, psychological, and

social factors could influence the disorder.2 Classification of

the LBP population into subgroups seems necessary to

determine the dominant factor(s) in each patient.2,3 It is

therefore important to have reliable and valid outcome

measures. Changes in motor control2,4,5 and altered move-

ment patterns2,6 in patients with LBP have been reported in

the literature. Although there is some agreement about

changes in trunk muscle function (eg, impaired deep intrinsic

trunk muscle activity), considerable variability exists in

patients with LBP.7 This has led to the need to differentiate

patients with LBP. Simple, reliable, and valid tests that could

be used in clinical settings are therefore necessary. The active

straight leg raise (ASLR) and standing hip flexion (Trende-

lenburg) are 2 clinical tests that assess the ability of the

lumbopelvic region to transfer loads between the trunk and

habil-icine,

herapy,lgium;hysicalm.herapy,lgium.loske-versity

e anddicine,

T, PT,, Van

ember

nces.

Roussel et alJournal of Manipulative and Physiological Therapeutics

Motor Control in Low Back Pain PatientsVolume 30, Number 4271

the legs. These 2 tests have been extensively analyzed in

carefully selected subgroups of patients, such as patients with

sacroiliac joint (SIJ) pain8 and patients with posterior pelvic

pain since pregnancy (PPPP).9,10 Transferring load through

the lumbopelvic region is a dynamic process, requiring both

articular stability (form closure) and optimal neuromuscular

stability (force closure).11 Patients with SIJ disorders8,11 and

PPPP12 exhibit changes in movement patterns and in force

closure during these 2 tests. Changes in movement patterns

and in motor control have been found in patients with

nonspecific LBP as well.2-6 Therefore, we hypothesized that

the ASLR and the Trendelenburg tests could also be useful in

patients with nonspecific LBP, as these tests examine the

whole lumbopelvic region.

MOTOR CONTROL ISSUES

Mens et al9 used the ASLR test to assess the ability of the

pelvic girdle to transfer loads between lumbosacral spine and

legs in patients with PPPP.8 Healthy subjects with optimal

functioning lumbopelvic region should be able to rise an

extended leg from the supine position (non–weight bearing)

without effort12 and without movement of the pelvis.13 Both

local and global muscles should activate properly to stabilize

the lumbopelvic region during this test.13 Mens et al12 found

that patients with PPPP showed a decreased ability to raise the

leg. In some patients, an anterior rotation of the ilium on the

side of the raised leg was observed during the ASLR.

Application of a belt reduced the impairment in nearly all

patients.12 In patients with SIJ pain, altered kinematics of the

diaphragm and the pelvic floor have been found during the

performance of the ASLR compared with healthy subjects.8

Manual compression of the pelvis to enhance the pelvic

stability during ASLR normalized the results. Therefore, the

authors suggested that the observed changes in motor control

strategies in patients with SIJ pain are a compensatory strategy

of the neuromuscular system to enhance force closure ormotor

control of the pelvis when stability has been comprised.8 The

test-retest repeatability of the ASLR, measured with an

intraclass correlation in women with PPPP, was 0.83. The

correlation between the scores of a patient and a blinded

observer was 0.77.9 In addition, the test was able to

discriminate betweenwomenwith PPPP and healthy subjects.9

Data addressing the external validity (r = 0.70with the Quebec

Back Pain Disability Scale) of the ASLR have been provided

as well.10 We are unaware of data addressing the reliability or

validity of the ASLR in patients with nonspecific LBP.

The standing hip flexion test has originally been developed

by Trendelenburg to assess dysfunctioning of the hip joint.

Hardcastle and Nade14 defined a standard method of perform-

ing the Trendelenburg test in patients with neurologic disorders

or with mechanical disorders of the hip and spine (scoliosis,

ankylosing spondylitis, and iliac crest defect after spinal

fusion).14 The Trendelenburg test has not been examined in

patients with nonspecific LBP. On the other hand, standing hip

flexion (Gillet or Stork test) is often used in patients with SIJ

pain15 and in patients with LBP. Both the Trendelenburg test

and the Gillet test examine the ability of the lower back, pelvis,

and hip to transfer load unilaterally in a weight-bearing

position.13 During unilateral weight bearing, a self-locking

mechanism of the pelvis, consisting of nutation of the sacrum

and posterior rotation of the ilium, should occur to optimize

lumbopelvic stability.13 Incontrast, anterior rotationof the ilium

of the weight bearing leg has been found with 3-dimensional

motion analysis in patients with SIJ pain.11 This can be

considered as failed load transfer through the lumbopelvic

region, as anterior rotation of the ilium is believed to be a less

stable position for the SIJ.13 Moreover, altered lumbopelvic

muscle recruitment has been found in patients with SIJ pain

during unilateral weight bearing.16 Hungerford et al16 found a

delayed onset of the electromyographic activity of the obliquus

internus, multifidus, and gluteus maximus muscles in patients

with SIJ pain compared with healthy subjects. On the other

hand, the onset of biceps femoris electromyographic activity

was earlier.16 These changes in muscle recruitment could be

responsible for the failed load transfer.

Despite the relative similarity of the performance of both

tests, the method used for the scoring of these clinical tests is

different. In the Gillet test, movement of the ilium in relation

to the sacrum is palpated and recorded. The reliability of the

Gillet test, assessed clinically with palpation, however,

remains questionable.15 The Trendelenburg test is considered

negative if the pelvis of the nonstance side can be elevated as

high as hip abduction on the stance side will allow and

providing the patient can maintain this posture for 30 seconds

without symptoms.14 According to Hardcaste and Nade,14 a

positive Trendelenburg test occurs when the pelvis drops on

the nonstance side. This is associated with hip adduction of

the weight-bearing leg and a compensatory scoliosis convex

to the nonstance side.14

Motor control of the lumbopelvic region is essential in both

the ASLR and Trendelenburg tests despite the fact that the first

one is a non–weight-bearing test and the second a weight-

bearing test. These tests have been examined in patients with

SIJ pain and in women with PPPP but not in patients with

nonspecific LBP. As impaired motor control has been

documented in patients with nonspecific LBP, it motivated

the authors to perform these tests in patients with nonspecific

LBP. The question arises whether these tests are associated

with each other and assess the same underlying dimension, that

is, force-transducing mechanisms of the lumbopelvic region.

RESPIRATION AND MOTOR CONTROL

Respiratory movements represent a perturbation to pos-

ture.17 There has been discrepancy regarding the degree to

which postural disturbances may be compensated in healthy

people.18 At least partial compensation occurs, but consid-

erable variation exists between individuals.17 A little body

sway due to postural disturbance, measured by ground

272 Journal of Manipulative and Physiological TherapeuticsRoussel et al

May 2007Motor Control in Low Back Pain Patients

reaction forces, is found in healthy people.19 Angular

movements of the trunk and lower limbs may indeed

compensate for the postural disturbance in healthy subjects,

as these small movements are in phase with respiration.18

Conversely, according to Hamaoui et al,20 respiration

represents a greater disturbing effect on body balance in

patients with LBP. These results have been confirmed by

Grimstone and Hodges18 who found a greater displacement

of the center of pressure related to respiration in patients with

LBP. This increased body sway with breathing in patients

with LBP could not be attributed to reduced trunk move-

ments,21 but it has been proposed as an indication of changes

in coordination by the central nervous system.19 In addition,

the breathing pattern seems to be compromised in patients

with LBP. As the diaphragm is involved in postural control of

the trunk,22,23 it is not surprising that the breathing pattern

could be changed in patients with LBP.Moreover, O’Sullivan

et al8 found altered breathing patterns and kinematics of the

diaphragm in patients with SIJ pain compared with healthy

controls. The patients with SIJ exhibit a decrease in

diaphragmatic excursion during the ASLR, with breath hold

(a complete loss of diaphragmatic motion) in 7 of 13 patients.

These breathing patterns normalized when the patients

performed the test with manual compression of the pelvis,

enhancing the pelvis stability. The authors therefore sug-

gested the diaphragm was recruited predominantly to

generate and control intra-abdominal pressure during the

ASLR instead of recruitment for its respiratory function in

addition to its postural function.8 Further information about

the influence of LBP on respiratory patterns during the ASLR

and other clinical tests assessing motor and postural control

(eg, Trendelenburg test) is lacking.

STUDY AIMS

The aim of this study was to examine the following: (1)

the test-retest reliability and the internal consistency of

2 clinical tests for the assessment of the ability of the

lumbopelvic region to transfer loads between lumbosacral

spine and legs; and (2) the interobserver reliability of the

assessment of the breathing pattern both at rest and during

the ASLR in patients with nonspecific LBP. To examine the

clinical importance of the tests in these patients, we

searched for associations between the outcome of the tests

and self-reported pain severity and disability.

METHODS

Subject Recruitment and Research DesignA sample of 36 patients with LBP was recruited from a

private practice of physiotherapy and from 2 hospital

outpatient physical therapy divisions. Inclusion criteria were

an age range of 18 to 65 years, a diagnosis of chronic

nonspecific LBP made by a physician, referral by a physician

for physiotherapy, and having Dutch/English as a native

language. Chronic LBP is defined as pain that persists longer

than 3 months. Patients with specific underlying pathology as

a cause of LBP (eg, tumor, known disk derangement, trauma,

infection, diagnosed inflammatory joint disease, spinal

stenosis, spondylolysis, or spondylolisthesis) were excluded.

Also, patients with a history of spinal fracture, severe

degenerative change, severe scoliosis, osteoporosis, obesity,

radicular signs, malignancies, and metabolic or rheumato-

logic diseases were excluded from the sample population. In

addition, women who were pregnant or up until 1 year

postnatal were excluded from participation. Finally, patients

with a history of spinal surgery were not included in the

study. Patients were recruited by their treating physiothera-

pists, which were not involved in the clinical examination

during the study. The treating physiotherapists checked the

previously mentioned specific causes for LBP with the

patient and the physician, using a standardized checklist. To

ascertain that no subjects with specific LBP entered the study,

the study investigators reevaluated the inclusion and exclu-

sion criteria after performance and scoring of the clinical

tests. The Human Research Ethics Committee of the

University Hospital of Antwerp approved the study, and

written informed consent was obtained from all participants

before testing.

Before participation, all subjects received verbal informa-

tion addressing the study. Next, an information booklet was

handed out to the participants. The patients were instructed to

read it vigilantly and to ask for additional explanation if

necessary. Demographic information, such as age, weight, and

height, was recorded by the time of testing. They were then

asked to complete a visual analogue scale (VAS) for pain

severity, and a number of self-reported questionnaires, as

described below. Afterward, the patient was evaluated by 1 of

the 2 examiners. At the time the study took place, one

investigator was a holder of a bachelor degree in physical

therapy and completed a master’s program in physical therapy.

The other investigator is a manual therapist with 4 years of

clinical experience. In 2 separate 2-hour training sessions

before data collection, the investigators were trained in

performing the tests. An expert with 7 years of both clinical

and research experience in the field of manual therapy and

respiratory physiotherapywas leading these sessions. Awritten

protocol was used to standardize the examination method. The

examiners subsequently examined 10 patients (pilot study)

before data collection. These assessments were recorded and

the video was discussed afterward with the expert and the

2 examiners. A consensus protocol was made to evaluate the

breathing pattern (described under the Clinical Tests section).

The investigator performed the following tests: the

Trendelenburg test, the ASLR, and the ASLR with assess-

ment of the breathing pattern. Also, the breathing pattern at

rest position was assessed. After the first test session a rest

period of 10 minutes was respected. The patients were asked

to complete the questionnaires during this time interval.

Then, the patient was examined by the other investigator.

Both investigators were blinded to each others’ scores and

Table 1. Descriptive statistics and interobserver reliability of the outcome on the Trendelenburg and the ASLR, 2 clinical tests for theassessment of the load transfer of the trunk to the legs

Test n Assessor 1 [median (P25, P75)] Assessor 2 [median (P25, P75)] j P

Trendelenburg score, left (0-5) 36 2 (1, 2) 2 (1, 2) 0.83 b.001

Trendelenburg score, right (0-5) 36 2 (1, 2) 1.5 (1, 2) 0.75 b.001

ASLR score, left (0-5) 36 1.5 (1, 2) 2 (1, 2) 0.70 b.001

ASLR score, right (0-5) 36 2 (1, 2) 2 (1, 2) 0.71 b.001

P, Percentile; j, weighted j value (quadratic).



the patients’ medical history. The order of the tests was

randomly assigned to avoid order effects. The order of

investigators was balanced so that each investigator tested

the same number of patients as first investigator.

Self-Reported MeasuresSelf-reported scales are often used in patients with LBP

mainly to evaluate the psychosocial aspect of the pain and

functional outcomes related to back pain.24 In the present

study, several questionnaires were used to assess potential

confounders for LBP-related disability. Indeed, psycholog-

ical aspects of pain such as kinesiophobia, depressive

symptoms, and catastrophic thoughts are known to be of

prognostic value in LBP-related disability. Therefore, the

Dutch versions of the following questionnaires were used in

this study: Oswestry Disability Low Back Pain Question-

naire, VAS, Beck Depression Inventory (BDI), Tampa Scale

for Kinesiophobia, Pain Catastrophizing Scale (PCS), Pain

Coping Inventory (PCI), Pain Vigilance and Awareness

Questionnaire (PVAQ), and Baecke Questionnaire for

Habitual Physical Activity. In addition, a self-established

questionnaire aimed at assessing participation in professio-

nal activities and sports was used.

The Oswestry Disability Low Back Pain Questionnaire

(ODQ) aims at assessing disability in patients with LBP. The

ODQ is a self-administered questionnaire consisting of

10 items, graded from 0 to 5. The test-retest repeatability

varied between 0.83 and 0.99.25 Data supporting the content

validity and the responsiveness of the ODQ have been

reported as well.26

The VAS (100 mm) was used for the assessment of

lumbar pain severity. The VAS pain score is believed to be

reliable,27-29 valid,29 and sensitive to change.28

The psychometric properties of the BDI,30 Tampa Scale for

Kinesiophobia,31,32 PCS,33-35 PCI,36,37 Pain Vigilance and

Awareness Questionnaire,38-40 and the Baecke Questionnaire

for Habitual Physical Activity41-43 have been described.

Clinical TestsThe thoracoabdominal motion during breathing was

assessed with subjects in the standing position. The patient

was instructed to gently breathe in and out. Movements of the

thorax and the abdomen during respiration were inspected and

palpated during spontaneous, quiet breathing. Next, the patient

was asked to take a deep breath. The thoracoabdominalmotion

during deep breathing was further observed and palpated. A

normal breathing pattern was defined as an outward move-

ment of the abdomen combined with an elevation of the thorax

during inspiration and the reverse pattern during expiration. A

paradoxical breathing pattern was defined as an inward

motion of the abdomen during inspiration. A third breathing

patternwith predominant rib cage expansion during inspiration

was observed during a pilot study (10 patients). Therefore,

3 different breathing patterns were assessed. Both the para-

doxical breathing pattern and the breathing pattern with

predominant rib expansion were considered as asynchronous

breathing motions.44

The ASLR was performed as described by Mens et al9 to

assess the ability of the pelvic girdle to transfer loads between

the lumbopelvic region and the legs. The patient was

positioned supine and was instructed to raise a straight leg

20 cm above the table.9 The patient was then asked to score

the perceived effort to perform the test on a 6-point scale: not

difficult at all, minimally difficult, somewhat difficult, fairly

difficult, very difficult, or unable to perform.9 In addition,

during the performance of the ASLR, expanding of the rib

cage and abdomen during inspiration was assessed with

visual inspection. The patients were asked to hold the

extended straight leg during 20 seconds to be able to detect

changes in breathing patterns. If the patient was able to

continue to breathe normally, the test was deemed negative.

Holding the breath during the ASLR was considered as

asynchronous breathing motion. Also, if the patient changed

the breathing pattern during the test, even without a real

breath hold, the results were scored as asynchronous breath-

ing motion. To avoid influencing patients’ breathing patterns,

they were not told that the breathing pattern was assessed

while performing ASLR. Therefore, the thoracoabdominal

motion during breathing was only assessed with visual

inspection and not with palpation.

The Trendelenburg test was developed to assess the

function of the hip joint. The test was performed in the

standing position, based on the description provided by

Hardcastle and Nade.14 The patient was asked to flex one hip

to 308 and to lift the pelvis of the nonstance side above the

transiliac line. This position was maintained for 30 seconds.

The pelvis should not tilt or rotate as the weight is shifted to

the supporting leg. The patient was allowed to touch the table

with one finger to correct for potential balance problems. If

the patient was not able to hold the test position, or if the

Table 2. Interobserver reliability of the breathing pattern assessedat rest and during the ASLR

Test n j P

Visual inspection of the breathing

pattern assessed at rest

36 0.42 b.01

Palpation of the breathing

pattern assessed at rest

36 0.47 b.01

Breathing pattern assessed

during the ASLR, left

36 0.47 b.01

Breathing pattern assessed

during the ASLR, right

36 0.39 .02

j, Weighted j value (quadratic).

Table 3. Correlations between the outcome of the breathingpattern assessed at rest and during the ASLR (n = 36)

Palpation at

rest [q ( P)] VAS [q ( P)]

Visual inspection at rest 0.822 (b.001) 0.341 (.042)

Palpation at rest 1 0.466 (.004)

q, Spearman correlation coefficient.



pelvis of the nonstance side could not be elevated above the

transiliac line, then the test was scored positive by the

examiner. In addition, after performing the test, the patient

was asked to score the perceived effort to perform the test.

This scoring method, which was found to be reliable for the

scoring of the ASLR in patients with PPPP,9 was reused to

assess the Trendelenburg test in the present study.

Statistical AnalysisAll data were analyzed using SPSS 12.0 for Windows

(SPSS Inc, Chicago, Ill). The power analysis determined

that between 30 and 43 subjects were necessary to establish

statistical significance at a power of .80 and a P value of .05

and .01, respectively. Appropriate descriptive statistics were

used. For assessing the interobserver reliability of the

clinical tests, the weighted j (quadratic) was used. For

interpretation of the reliability, the standards proposed by

Landis and Koch45 for strength of agreement for the jcoefficients were used (b0, poor; 0.01-0.20, slight; 0.21-

0.40, fair; 0.41-0.60, moderate; 0.61-0.80, substantial; 0.81-

1, almost perfect agreement). The Cronbach a coefficient

was calculated to examine the internal consistency of the

outcome of the clinical tests. A 1-sample Kolmogorov-

Smirnov goodness-of-fit test was used to examine whether

the variables were normally distributed. A Spearman

correlation coefficient was used to analyze correlations

between the clinical tests, and between the clinical tests and

the questionnaires. The significance level was set at .05.

RESULTS

Demographic DataTwenty-one women (58%) and 15 (42%) men with

nonspecific LBP volunteered for this study. The mean age

of the patients was 37.4 F 11.6 years (range, 21-62 years).

At the time the study took place, the mean VAS score was

22.4 F 17.5 mm (range, 0-69 mm). The duration of the

symptoms was 84.0 F 76.9 months (range, 3-288 months).

The descriptive statistics of the tests’ outcome for both

examiners are presented in Table 1. The variables Trende-

lenburg test, ASLR, breathing pattern, and breathing pattern

during ASLR were not normally distributed (Kolmogorov-

Smirnov test, P b .05; data not shown).

ReliabilityThe test-retest reliability of the patients’ score for the

Trendelenburg test, measured with a weighted j, was 0.83for the left side and 0.75 for the right side. The test-retest

reliability of the patients’ score for the ASLR, measured

with a weighted j, was 0.71 for the left side and 0.70 for theright side. Table 1 provides an overview of the results of the

tests and the outcome of the reliability analysis.

The visual inspection of the breathing pattern at rest

showed an interobserver reliability of 0.42. The palpation of

the breathing pattern demonstrated a similar result (j = .46).

The assessment of the breathing pattern during the ASLR

showed an interobserver reliability coefficient (j) of 0.47 forthe left leg and 0.38 for the right leg. The data are shown in

Table 2. Significant correlations were found between the

visual inspection of the breathing pattern and the palpation of

the breathing pattern assessed at rest (Spearman correlation

coefficients are 0.82, P b .001) (Table 3).

Internal ConsistencyThe Cronbach a coefficient was computed to examine

the internal consistency of the outcome of the clinical tests.

The Cronbach a coefficient for internal consistency for the

Trendelenburg and ASLR tests was .80 for assessor 1 and

.73 for assessor 2.

Clinical ImportanceThe Spearman correlation analysis searched for associ-

ations between the self-reported measures (VAS and ODQ)

and the outcome of the tests for the assessment of the force-

transducing mechanisms of the lumbopelvic region (Tren-

delenburg test and ASLR). Using the data of either assessor

1 or assessor 2, we found no significant associations (P N

.01). In contrast, a significant association between pain

severity (VAS) and the ODQ total score (Pearson r = 0.53,

P b .001) was found.

Four patients exhibited complete breath holding during

ASLR. In 18 other patients, a change in breathing pattern was

assessed by both examiners while performing ASLR. For

examiner 2, a significant correlation was found between the

assessment of the breathing pattern assessed both with visual



inspection and with palpation, and pain severity (q = 0.34 and

0.47, respectively P b .05). The results are shown in Table 3.

The assessment of the breathing pattern during the ASLR

correlated significantly with the BDI (q = 0.43, P = .009), with

the PCI—worrying subscale (q = 0.52, P = .001), with the

ODQ (q = 0.34, P = .04), and with the PCS—magnification

subscale (q = 0.42, P = .011). No additional statistically

significant associations were observed (data not shown).

DISCUSSION

Patients with LBP have been the object of research during

the last decades. In recent years, the research focus has shifted

to movement and motor control impairments.2-7,46 The

knowledge that considerable variability exists in patients

with LBP7 has led to the need to differentiate patients with

nonspecific LBP and to classify them into subgroups. Simple,

reliable, and valid tests that could be used in clinical settings

are therefore necessary. The present study examined some of

the clinimetric properties of clinical tests, potentially of use

for identifying subgroups in the LBP population.

The results of this study showed that the score given by the

patient (ranging from 0 to 5) appears to be highly reliable

(almost perfect agreement for the left leg). To the knowledge

of the authors, the Trendelenburg test has not been evaluated

before with this 6-point Likert scale for perceived effort. In

addition, the patients’ score for the ASLR appears to be

reliable in patients with nonspecific LBP. Good test-retest

reliability for the ASLR has already been found by Mens

et al.9 There are several differences between the study Mens

performed and the present study. Firstly, only women were

analyzed in the study of Mens. Secondly, the reliability was

assessed in patients with a variety of pregnancy-related

lumbopelvic pain. In the present study, only patients with

chronic nonspecific LBP were assessed, whereas patients

with pregnancy-related pain were excluded. To date, these

populations (women with pregnancy-related lumbopelvic

pain and patients, both men and women, with chronic

nonspecific LBP) are considered to be different. In the

present study, the test-retest reliability was analyzed for each

leg separately. A j value was therefore used, as the results

were not normally distributed. At first sight, it may seem

logical to find a high test-retest reliability, as the patient’s

scores are used to determine the reliability. During the

interval between the 2 test sessions, the patients were asked to

fill in a battery of questionnaires. It seems therefore unlikely

that the patients remembered all the test scores. Moreover,

pain and fatigue could have influenced the test results and

limited the patient’s second performance. All the tests were

performed during 20 (ASLR andASLRwith inspection of the

breathing pattern) or 30 seconds (Trendelenburg). Kankaan-

paa et al47 found a relationship between electromyographic

analysis (objective estimate of fatigue) and the unmodified

Borg score for perceived effort (subjective estimate of

fatigue). Patients with LBP experience fatigue faster than

healthy subjects.48 Therefore, a higher score for perceived

effort could have been expected for the second performance

of the patient. However, based on the results of the present

study, neither pain nor fatigue seems to significantly affect the

test-reliability of the patient’s scores.

Based on our results, it appears that the interobserver

reliability of the assessment of the breathing pattern is fair to

moderate. Neither the visual inspection nor the palpation of the

thoracoabdominal motion during breathing, as assessed in this

study, could be used in a clinical setting. As a result, the visual

assessment of the breathing pattern during the ASLR is not

reliable enough to differentiate patients. The repeatability of

pulmonary physical examination seems low.44 Particularly, the

visual assessment of the thoracoabdominal expansion shows

high interobserver variation. Gjlrup et al49 found a j value of

0.14 for the assessment of decreased breath movements.49 In a

study by Perkins et al,50 medical students were asked to

evaluate simulated breathing patterns (normal, abnormal, and

absent). They concluded that medical students were unable to

identify normal breathing from abnormal breathing. Despite a

different setting (simulated situation of an unconscious

patient), it can be concluded that clinical assessment of the

breathing pattern is difficult. To the knowledge of the authors,

this is the first study addressing the clinical assessment of the

breathing pattern in patients with nonspecific LBP.

The ASLR is a clinical test measuring effective load

transfer between the lumbopelvic region and lower limbs.13

The Trendelenburg test was originally described to assess

hip function and has not been previously assessed in

patients with nonspecific LBP. The internal consistency of

the Trendelenburg and ASLR tests was high for both

assessors despite the fact that the ASLR can be considered

as a non–weight-bearing test and the Trendelenburg test as a

weight-bearing test. This suggests that both tests assess the

same underlying dimension. As the ASLR has already been

validated in women with PPPP, it is tempting to speculate

that both tests address impaired load transfer between the

lumbopelvic region and the legs. Future studies are required

to confirm this hypothesis and to examine the possible

neuromuscular alterations between patients with LBP and

healthy people during the performance of these 2 tests.

Mens et al9 found the best balance between sensitivity

and specificity when the scores of 1 to 10 are considered as

positive and 0 as negative. If this cutoff score is used for the

interpretation of the ASLR data of the present study, only

3 of 36 patients presented a negative test result. Thus, nearly

all patients with nonspecific LBP experienced fatigue when

raising one leg in the supine position. If the same cutoff was

used for interpreting the Trendelenburg data, identical

results were found (ie, only 3 of 36 patients had a negative

test result). No SIJ instability was diagnosed in any of the

study participants and postpartum patients with LBP were

not allowed to participate in this study. The effort needed to

perform this test could be because of a motor control

impairment of the lumbopelvic region. While performing



the test, rotation of the pelvis during the ASLR performance

was frequently observed. This could be a sign of insufficient

lumbopelvic stability to keep the pelvis level while raising

one leg. Further study is required to address this issue.

No correlations were found between the outcome of the

clinical tests for the assessment of the force-transducing

mechanisms of the lumbopelvic region (Trendelenburg test

and ASLR) and the self-reported measures (VAS and ODQ).

Correlations between pain, physical impairment, and dis-

ability are generally low in patients with LBP.51 Notwith-

standing, the ASLR correlated well with the Quebeck Back

Pain Disability Scale in patients with PPPP.10 Waddel et al51

found 8 tests that successfully discriminated patients with

LBP from healthy subjects and that were related to self-

reported disability in activities of daily living. Bilateral

active straight leg raise is one of these tests.51 Also, Stuge et

al52 found a close association between changes in pain/

disability and the ASLR performance. The fact that no

associations were found in the present study with a cross-

sectional design does not mean that the tests are without

clinical importance. For studying the clinical importance of

the outcome of tests, comparative (examining differences in

the outcome of tests between patients with LBP and

asymptomatic subjects) and prospective studies (examining

whether the outcome of tests are of prognostic value for

patients with LBP) are warranted.

In the present study, 61% (22/36) of the patients exhibited a

change in breathing pattern observed by both examiners

during ASLR performance. For examiner 2, a significant

association was found between the assessment of the breath-

ing pattern at rest and the actual level of pain. Also, the

breathing pattern during the ASLR correlated with the PCI,

BDI, and PCS. The breathing pattern appeared to have some

clinical importance to patients with LBP. Despite the fact that

the method to assess the breathing pattern used in this study

did not show sufficient reliability, further research to assess the

breathing pattern in patients with LBP remains imperative.

Practical Applications

! Classification of the LBP population into sub-

groups is important. Clinically applicable, reliable,

and valid tests that can be used in clinical settings

are therefore necessary.

! The Trendelenburg and the ASLR are 2 reliable

tests and have a high internal consistency. Further

research is needed to examine the responsiveness

and the validity of these tests.

! The evaluation of the breathing pattern at rest and

during the ASLR, as performed in the present study,

is not reliable enough to be used in clinical settings.

Further research regarding the relationship between

changes in breathing patterns during motor control

tests in patients with LBP is required.

Methodological ConsiderationsThe results of this study should be interpreted in the light

of several methodological issues. Firstly, no specific sacroil-

iac joint assessment was made by the investigators for the

exclusion of comorbid SIJ disorders. However, comorbid SIJ

disorders are generally not regarded as exclusionary criteria

for nonspecific LBP. Secondly, great effort wasmade to select

patients with nonspecific LBP solely. Therefore, the Euro-

pean Guidelines for the Management of Chronic Non-

Specific Low Back Pain were used.53 The patients were

selected by their treating physiotherapists. The physiothera-

pists checked the specific causes for LBP with the patient and

the physician, using a standardized checklist. To ascertain

that no subjects with specific LBP entered the study, the study

investigators reevaluated the inclusion and exclusion criteria

after performance and scoring of the clinical tests. Unfortu-

nately, there were no imaging data available for all patients, as

these are only recommended by the COST Action B13

guidelines if a specific cause of LBP is suspected. Thirdly,

variation in training and clinical experience among examiners

could account for the observed discrepancies, especially for

the assessment of the breathing pattern. Fourthly, no control

group was used and therefore other attributes, such as

sensitivity and specificity, remain to be examined, as well

as validity and responsiveness.

CONCLUSION

These data provide evidence favoring the test-retest

reliability of both the ASLR and Trendelenburg tests. The

internal consistency of the outcome of the tests was high for

both assessors, suggesting that these tests assess the same

dimension, that is, the force-transducing mechanisms of the

lumbopelvic region. Based on these results, it appears that

the interobserver reliability of the assessment of the breath-

ing pattern is fair to moderate. Further research regarding

the interobserver reliability, clinical importance, validity,

and responsiveness of the Trendelenburg test, ASLR test,

and breathing pattern during these tests is required.

ACKNOWLEDGMENT

The authors thank the physical therapists and patients for

participating in the study. Special thanks to Shona Bouchery

for editing the manuscript. This study was financially

supported by a PhD grant (bMotor control of the lumbo-

pelvic region in dancers and patients with low back painQ)supplied by Hogeschool Antwerpen, Merksem, Belgium

(G806). 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 in which the authors are associated.



REFERENCES

1. Waddel G. The back pain revolution. Churchill Livingstone,2004.

2. O’Sullivan P. Diagnosis and classification of chronic low backpain disorders: maladaptive movement and motor controlimpairments as underlying mechanism. Man Ther 2005;10:242-55.

3. Dankaerts W, O’Sullivan PB, Burnett A, Straker L, Skouen J.Towards a clinical validation of a classification method for nonspecific chronic low back pain patients with motor controlimpairment. Man Ther 2006;11:28-39.

4. Silfies SP, Squillante D, Maurer P, Westcott S, Karduna A.Trunk muscle recruitment patterns in specific chronic low backpain populations. Clin Biomech 2005;20:465-73.

5. Richardson CA, Snijders CJ, Hides JA, Damen L, Pas MS,Storm J. The relation between the transversus abdominismuscles, sacroiliac joint mechanics, and low back pain. Spine2002;27:399-405.

6. Shum GLK, Crosbie J, Lee RYW. Effect of low back painon the kinematics and joint coordination of the lumbar spineand hip during sit-to-stand and stand-to-sit. Spine 2005;30:1998-2004.

7. Hodges PW, Moseley GL. Pain and motor control of thelumbopelvic region: effect and possible mechanisms. J Electro-myogr Kines 2003;13:361-70.

8. O’Sullivan PB, Beales DJ, Beetham JA, Cripps J, Graf F, Lin I,et al. Altered motor control strategies in subjects with sacroiliacjoint pain during the active straight-leg-raise test. Spine 2002;27:E1-E8.

9. Mens JM, Vleeming A, Snijders CJ, Koes B, Stam H.Reliability and validity of the active straight leg raise testin posterior pelvic pain since pregnancy. Spine 2001;26:1167-71.

10. Mens JMA, Vleeming A, Snijders C, Koes B, Stam H. Validityof the active straight leg raise test for measuring diseaseseverity in patients with posterior pelvic pain after pregnancy.Spine 2002;27:196-200.

11. Hungerford B, Gilleard W, Lee D. Altered patterns of pelvicbone motion determined in subjects with posterior pelvic painusing skin markers. Clin Biomech 2004;19:456-64.

12. Mens JM, Vleeming A, Snijders CJ, Stam HJ, Ginai AZ. Theactive straight leg raising test and mobility of the pelvic joints.Eur Spine J 1999;8:468-73.

13. Lee D. The pelvic girdle. An approach to the examination andtreatment of the lumbopelvic-hip region. Churchill Livingstone,2004.

14. Hardcastle P, Nade S. The significance of the Trendelenburgtest. J Bone Joint Surg 1985;67:741-6.

15. van der Wurff P, Hagmeijer RH, Meyne W. Clinical tests of thesacroiliac joint. A systemic methodological review. Part 1:reliability. Man Ther 2000;5:30-6.

16. Hungerford B, Gilleard W, Hodges P. Evidence of alteredlumbopelvic muscle recruitment in the presence of sacroiliacjoint pain. Spine 2003;28:1593-600.

17. Hodges PW, Gurfinkel VS, Brumagne S, Smith T, Cordo P.Coexistence of stability and mobility in postural control:evidence from postural compensation for respiration. ExpBrain Res 2002;144:293-302.

18. Grimstone SK, Hodges PW. Impaired postural compensationfor respiration in people with recurrent low back pain. ExpBrain Res 2003;151:218-24.

19. Smith M, Coppieters MW, Hodges PW. Effect of experimen-tally induced low back pain on postural sway with breathing.Exp Brain Res 2005;166:109-17.

20. Hamaoui A, Do MC, Poupard L, Bouisset S. Does respirationperturb body balance more in chronic low back pain subjectsthan in healthy subjects? Clin Biomech 2002;17:548-50.

21. Hamaoui A, Do MC, Bouisset S. Postural sway increase in lowback pain subjects is not related to reduced spine range ofmotion. Neurosci Lett 2004;357:135-8.

22. Hodges PW, Butler JE, McKenzie DK, Gandevia SC.Contraction of the human diaphragm during rapid posturaladjustments. J Physiol (London) 1997;505:539-48.

23. Hodges PW, Gandevia SC. Activation of the human diaphragmduring a repetitive postural task. J Physiol (London) 2000;522(Part 1):165-75.

24. Kopec JA. Measuring functional outcomes in persons withback pain: a review of back-specific questionnaires. Spine2000;25:3110-4.

25. Beurskens AJ, de Vet HC, Kfke AJ, van der Heijden G,Knipschild P. Measuring the functional status of patients withlow back pain. Assessment of the quality of four disease-specific questionnaires. Spine 1995;20:1017-28.

26. Beurskens AJ, de Vet HC, Kfke AJ. Responsiveness offunctional status in low back pain: a comparison of differentinstruments. Pain 1996;65:71-6.

27. Harms-Ringdahl K, Carlsson AM, Ekholm J, Raustorp A,Svensson T, Toresson H. Pain assessment with differentintensity scales in response to loading of joint structures. Pain1986;27:401-11.

28. Jensen MP, Karoly P, Braver S. The measurement of clinical painintensity: a comparison of six methods. Pain 1986;27:117-26.

29. Ogon M, Krismer M, Sfllner W, Kantner-Rumplmair W,Lampe A. Chronic low back pain measurement with visualanalogue scales in different settings. Pain 1996;64:425-8.

30. Osman A, Kopper BA, Barrios F, Gutierrez P, Bagge C.Reliability and validity of the Beck depression inventory–IIwith adolescent psychiatric inpatients. Psychol Assess 2004;16:120-32.

31. Vlaeyen JW, Kole-Snijders AM, Boeren RG, Van Heek H.Fear of movement/(re)injury in chronic low back pain and itsrelation to behavioral performance. Pain 1995;62:363-72.

32. Swinkels-Meewisse EJCM, Swinkels RAHM, Verbeek ALM,Vlaeyen J, Oostendorp R. Psychometric properties of theTampa Scale for kinesiophobia and the fear-avoidance beliefsquestionnaire in acute low back pain. Man Ther 2003;8:29-36.

33. Sullivan MJ, Bishop S, Pivik J. The pain catastrophizing scale:development and validation. Psychol Assess 1995;7:524-32.

34. Sullivan MJ, Stanish W, Waite H, Sullivan M, Tripp D.Catastrophizing, pain, and disability in patients with soft-tissueinjuries. Pain 1998;77:253-60.

35. Crombez G, Vlaeyen J, Heuts P, Lysens R. Pain-relatedfear is more disabling than pain itself: evidence on the role ofpain-related fear in chronic back pain disability. Pain 1999;80:329-39.

36. Kraaimaat FW, Bakker A, Evers AWM. Pijncoping-strategieJnbij chronische pijnpatiJnten: de ontwikkeling van de Pijn-Coping-Inventarisatielijst. Gedragstherapie 1997;30:185-201.

37. Kraaimaat FW, Evers AWM. Pain-coping strategies in chronicpain patients: psychometric characteristics of the Pain-CopingInventory (PCI). Int J Behav Med 2003;10:343-63.

38. McCracken L. dAttentionT to pain in persons with chronic pain:a behavioral approach. Behav Ther 1997;28:271-84.

39. Roelofs J, Peters ML, Muris P, Vlaeyen J. Dutch version of thePain Vigilance and Awareness Questionnaire: validity andreliability in a pain-free population. Behav Res Ther 2002;40:1081-90.

40. Roelofs J, Peters ML, McCracken L, Vlaeyen J. The PainVigilance and Awareness Questionnaire (PVAQ): further



psychometric evaluation in fibromyalgia and other chronicpain syndromes. Pain 2003;101:299-306.

41. Baecke JA, Burema J, Frijters JE. A short questionnaire for themeasurement of habitual physical activity in epidemiologicalstudies. Am J Clin Nutr 1982;36:936-42.

42. Pols MA, Peeters PH, Bueno-De-Mesquita HB, et al. Validityand repeatability of a modified Baecke questionnaire onphysical activity. Int J Epidemiol 1995;24:381-8.

43. Richardson MT, Ainsworth BE, Wu HC, Jacobs D, Leon A.Ability of the Atherosclerosis Risk in Communities (ARIC)/Baecke Questionnaire to assess leisure-time physical activity.Int J Epidemiol 1995;24:685-93.

44. Maitre B, Similowski T, Derenne JP. Physical examination ofthe adult patient with respiratory diseases: inspection andpalpation. Eur Respir J 1995;8:1584-93.

45. Landis J, Koch G. The measurement of observer agreement forcategorical data. Biometrics 1977;33:159-74.

46. Cholewicki J, Silfies SP, Shah RA, et al. Delayed trunk musclereflex responses increase the risk of low back injuries. Spine2005;30:2614-20.

47. Kankaanpaa M, Taimela S, Webber CL, Airaksinen O,Hanninen O. Lumbar paraspinal muscle fatigability in repet-

itive isoinertial loading: EMG spectral indices, Borg scale andendurance time. Eur J Appl Physiol Occup Physiol 1997;76:236-42.

48. Taimela S, Kankaanpaa M, Airaksinen O. A submaximal backextension endurance test utilising subjective perception of lowback fatigue. Scand J Rehabil Med 1998;30:107-12.

49. Gjlrup T, Bugge PM, Jensen AM. Interobserver variation inassessment of respiratory signs. Physicians’ guesses as tointerobserver variation. Acta Med Scand 1984;216:61-6.

50. Perkins GD, Stephenson B, Hulme J, Monsieurs K. Birming-ham assessment of breathing study (BABS). Resuscitation2005;64:109-13.

51. Waddell G, Somerville D, Henderson I, Newton M. Objectiveclinical evaluation of physical impairment in chronic low backpain. Spine 1992;17:617-28.

52. Stuge B, Laerum E, Kirkesola G, Vollestad N. The efficacy of atreatment program focusing on specific stabilizing exercises forpelvic girdle pain after pregnancy: a randomized controlledtrial. Spine 2004;29:351-9.

53. Airaksinen O, Brox J-I, Cedraschi C, et al. Chapter 4.European guidelines for the management of chronic non-specific low back pain. Eur Spine J 2006;15:S192-S300.

low back pain: clinimetric properties of the trendelenburg test, active straight leg raise test, and...

Documents