Ultrasound in Med. & Biol., Vol. 34, No. 8, pp. 1209–1216, 2008Copyright © 2008 World Federation for Ultrasound in Medicine & Biology

Printed in the USA. All rights reserved0301-5629/08/$–see front matter

doi:10.1016/j.ultrasmedbio.2008.01.003

● Original Contribution

RELIABILITY OF MEASURING SCIATIC AND TIBIAL NERVEMOVEMENT WITH DIAGNOSTIC ULTRASOUND DURING A NEURAL

MOBILISATION TECHNIQUE

RICHARD ELLIS,* WAYNE HING,* ANDREW DILLEY,† and PETER MCNAIR**Health and Rehabilitation Research Centre, School of Physiotherapy, AUT University, Auckland, New Zealand;

and †Division of Clinical Investigation, Brighton and Sussex Medical School, University of Sussex, Falmer,Brighton, United Kingdom

(Received 24 October 2007; revised 16 December 2007; in final form 10 January 2008)

Abstract—Diagnostic ultrasound provides a technique whereby real-time, in vivo analysis of peripheral nervemovement is possible. This study measured sciatic nerve movement during a “slider” neural mobilisation technique(ankle dorsiflexion/plantar flexion and cervical extension/flexion). Transverse and longitudinal movement was assessedfrom still ultrasound images and video sequences by using frame-by-frame cross-correlation software. Sciatic nervemovement was recorded in the transverse and longitudinal planes. For transverse movement, at the posteriormidthigh (PMT) the mean value of lateral sciatic nerve movement was 3.54 mm (standard error of measurement[SEM] � 1.18 mm) compared with anterior-posterior/vertical (AP) movement of 1.61 mm (SEM � 0.78 mm). At thepopliteal crease (PC) scanning location, lateral movement was 6.62 mm (SEM � 1.10 mm) compared with APmovement of 3.26 mm (SEM � 0.99 mm). Mean longitudinal sciatic nerve movement at the PMT was 3.47 mm (SEM� 0.79 mm; n � 27) compared with the PC of 5.22 mm (SEM � 0.05 mm; n � 3). The reliability of ultrasoundmeasurement of transverse sciatic nerve movement was fair to excellent (Intraclass correlation coefficient [ICC] �0.39–0.76) compared with excellent (ICC � 0.75) for analysis of longitudinal movement. Diagnostic ultrasoundpresents a reliable, noninvasive, real-time, in vivo method for analysis of sciatic nerve movement. (E-mail:richard.ellis@aut.ac.nz) © 2008 World Federation for Ultrasound in Medicine & Biology.

Key Words: Ultrasound, Sciatic nerve, Neurodynamics, Neural mobilisation, Frame-by-frame cross-correlation,

Reliability.

INTRODUCTION

Diagnostic ultrasound presents an important tool in thestudy of neurodynamics. Neurodynamics is a term thatrefers to the integrated biomechanical and physiologicalfunctioning of the nervous system (Butler 2000; Shack-lock 1995, 2005), i.e., the ability for a peripheral nerve toslide and stretch to accommodate changes in its bedlength during joint movements.

Neural mobilisation is a clinical technique that at-tempts to, amongst other goals, restore normal neurody-namics in conditions where nerve sliding may be af-fected. Neural mobilisation exercises are well docu-mented within the literature (Cleland et al. 2007;Coppieters et al. 2004; Coppieters and Butler 2007;Herrington 2006; Scrimshaw and Maher 2001; Shack-

Address correspondence to: Richard Ellis, School of Physiother-apy and Health and Rehabilitation Research Centre, AUT University,

Private Bag 92006, Auckland 1020, New Zealand. E-mail: richard.ellis@aut.ac.nz

1209

lock 2005; Totten and Hunter 1991). A large number ofstudies examining neural mobilisation have used cadav-ers (Beith et al. 1995; Coppieters and Butler 2007; La-Ban et al. 1989; Szabo et al. 1994; Wright et al. 1996).The dissection and removal of certain tissues, requiredfor the examination of neural mobilization, and embalm-ing techniques in cadavers may alter neural mechanicscompared with in vivo. Diagnostic ultrasound, therefore,offers a potentially portable and noninvasive method forthe assessment of peripheral nerve movement. Diagnos-tic ultrasound has a distinct advantage over other staticimaging methods (i.e., magnetic resonance imaging)with its ability to measure soft tissue movement both inreal time and in vivo (Chiou et al. 2003; Hashimoto et al.1999; Jeffery 2003; Martinoli et al. 2000).

Several research groups have developed techniques toutilise real-time ultrasound imaging to assess both trans-verse (Greening et al. 2001) and longitudinal peripheralnerve movement in vivo (Dilley et al. 2001; Hough et al.

2000b). Methods for measuring longitudinal movement

inCom

1210 Ultrasound in Medicine and Biology Volume 34, Number 8, 2008

have utilised Doppler ultrasound (Hough et al. 2000a,2000b) as well as frame-by-frame cross-correlation analysisof high-frequency ultrasound image sequences (Dilley et al.2001, 2003). Most of the previous studies that have useddiagnostic ultrasound to measure peripheral nerve move-ment have examined nerves in the upper limb (Dilley et al.2003, 2007; Erel et al. 2003; Greening et al. 2001, 2005;Hough et al. 2000a, 2000b, 2007). In contrast, the presentstudy uses diagnostic ultrasound to measure movement ofthe sciatic nerve in response to a neural mobilization exer-cise, consisting of cervical extension combined with ankledorsiflexion. Calculations of longitudinal nerve motionwere performed using frame-by-frame cross-correlationanalysis. The key aims of the present study were to assessthe reliability of measuring sciatic nerve movement, both inlongitudinal and transverse planes, using diagnostic ultra-sound digital caliper measurement and frame-by-framecross-correlation analysis.

MATERIALS AND METHODS

SubjectsTwenty-seven subjects (14 females, 13 males), with

an age range between 18 and 38 y (average � 22.82 y,SD 4.61), were included in this study. Informed consentto participate in this study was given by all subjects.Subjects met the inclusion criteria if they were healthyindividuals between the ages of 18 to 40 and did not havea history of significant/major trauma or surgery to thelumbar, hip, buttock (gluteal) or hamstring (posteriorthigh) regions, symptoms consistent with sciatic nerveimpairment (i.e., paraesthesias, weakness etc.) or a pos-itive slump test (a test which determines dysfunction ofthe sciatic nerve and its associated branches) as de-scribed by Butler (2000)).

Equipment and Procedures

Subject Set-up. A Kinom dynamometer (Kinetic

Fig. 1. “Slider” neural mobilisation sequence, from StarK

Communicator, Chattex Corp., Chattaneoga, TN, USA)

system was used to provide a consistent, fixed and sup-ported subject position. Subjects were seated, with theseat back upright (resting against an additional backsupport) and the pelvis fixed using a lap seatbelt. Thisset-up provided a consistent and stable sitting positionfor all subjects (Fig. 1).

In all subjects, the right leg was imaged. The midaspect of the right calf was supported using a woodensupport arm. The hip and knee joints were held at 90°and 50° flexion, respectively. The angle of each jointposition was determined with the use of a hand-held,manual goniometer (Fig. 1).

Diagnostic Ultrasound Parameters and ImagingAn experienced sonographer performed all ultra-

sound scans. The sonographer was blinded to all ultra-sound measurements taken.

B-mode real-time ultrasound scanning was per-formed using a Philips HD11 (Philips Medical SystemsCo., Eindhoven, The Netherlands) ultrasound machinewith a 5 to 12 MHz, 55 mm, linear array transducer. Thesciatic nerve was scanned in both the transverse andlongitudinal planes at two separate locations, the poste-rior midthigh (PMT) and popliteal crease (PC).

Transverse movement. A static ultrasound image(Fig. 2) was taken at the Start position of the neuralmobilization exercise (Fig. 1). At this position, the neckwas in full flexion (i.e., chin on the chest) with the anklejoint in full plantar flexion. Within the ultrasound image,digital markers were placed at the lateral-medial andanterior-to-posterior (AP) extremities/boundaries of thevisualised nerve. Measurements of the location of thenerve were recorded.

During the neural mobilisation, active extension ofthe neck, to a maximum tolerable extension (e.g., 40 to70° from neutral), was performed simultaneously withpassive dorsiflexion of the ankle until maximum tolera-

p position. Also shows standardised subject position in.

t to Sto

ble dorsiflexion (e.g., 20 to 40° from neutral). At this

sverse

Reliability of measuring sciatic and tibial nerve movement ● R. ELLIS et al. 1211

Stop position, a second static ultrasound image wastaken. Digital markers were again placed at the lateral-medial and AP extremities of the visualised nerve.

Digital calipers were used to measure the vertical/anterior-posterior distance (AP movement) between theposterior markers on the Start and the Stop positions. Thesame procedure was also used to measure the distancebetween the right side lateral markers on the Start to theStop position to represent the medial-lateral distance(lateral movement) (Fig. 2).

A total of three successive measurements weretaken at 1 min intervals. All measurements were taken bythe same sonographer.

Longitudinal Movement. For longitudinal analysis,it was not possible to use the same method of digitalcaliper placement, as was used for the transverse mea-surements, since no clearly defined landmarks are avail-able from which to record movement from a Start to aStop position within the longitudinal plane. Therefore, anultrasound cine-loop of the nerve in longitudinal planewas recorded from the Start to the Stop position. Thevideo loop was captured over a 2 second period at 40frames per second. A total of three successive measure-ments were recorded at 1-min intervals. This video foot-age could then be analysed off-line in digital format.

Frame-by-Frame Cross-Correlation Algorithm and Cal-culation Software

Each video loop of the longitudinal nerve move-ment was converted to digital format (bitmaps) and

Fig. 2. Location of sciatic nerve, in tran

analysed off-line using a method of frame-by-frame

cross-correlation analysis that was developed in Mat-lab (MathWorks, Natick, MA, USA) by Dilley et al.(2001). This method employs a cross-correlation al-gorithm to determine relative movement between suc-cessive frames in a sequence of ultrasound images(Dilley et al. 2001). From the initial frame of thesequence, three contiguous rectangular regions-of-in-terest (ROI), of varied dimensions, were selectedwithin the sciatic nerve. During the analysis, the pro-gram compares the gray-scale values from the ROIsbetween adjacent frames of the image sequence. In thecompared frame, the coordinates of the ROI are offsetalong the horizontal image plane a pixel at a timewithin a predetermined range. A correlation coeffi-cient is calculated for each individual pixel shift. Thepeak of a quadratic equation fitted to the maximumthree correlation coefficients is equivalent to the pixelshift/movement between adjacent frames (Dilley et al.2001). Pixel shift measurements for the nerve wereoffset against (subtracted from) pixel shifts measure-ments within the same ultrasound field, from station-ary structures (i.e., subcutaneous layers, bone etc.).This method allows for any slight movement of theultrasound transducer to be eliminated from the anal-ysis.

Statistical AnalysisTwo-way mixed, intraclass correlation coefficients

(ICC), with 95% confidence intervals, were calculated,using SPSS (version 14.0) statistical analysis software,

plane, at the Start and Stop positions.

as an indication of the reliability of both transverse and

rior; SE7) asid

1212 Ultrasound in Medicine and Biology Volume 34, Number 8, 2008

longitudinal sciatic nerve movement analysis at the PMTand PC. Bland-Altman plots have been used to providegraphical representation of some of the key reliabilityfindings (Bland and Altman 1986).

As a measure of combined transverse nerve move-ments, basic trigonometry (Pythagoras’ theorem) wasused to calculate the diagonal distance of nerve move-ment, or hypotenuse distance, between the Start and Stoppositions. This particular movement was also quantifiedin terms of the direction of the movement (i.e., supero-medial, superolateral, inferomedial, inferolateral).

Paired t tests were utilised to explore statisticaldifferences of sciatic nerve measurements between dif-ferent movement directions and also between differentscanning locations.

EthicsEthics approval was sought and approved by the

Auckland University of Technology Ethics Committee(AUTEC).

Table 1. Amount of sciatic nerve movement (mm

Trans

AP

Scanning location Range Mean SEM

Posterior midthigh (PMT) 0.02 - 4.19 1.61 0.62Popliteal fossa/crease (PC) 3.00 - 8.28 3.26 0.85

mm � millimeter; n � number of subjects; AP � anterior-to-posteAll figures representative of data collected from 27 subjects (n � 2

0

1

2

3

4

5

6

7

8

9

Ner

ve m

ovem

ent (

mm

)

Fig. 3. Comparison of sciatic nerve movement (transver

each scanning location (posterior midthigh [PMT] an

RESULTS

Transverse Ultrasound Imaging of Lateral and Anterior-to-Posterior Sciatic Nerve Movement

Data representing the transverse sciatic nerve move-ments measured at the PMT and PC are presented inTable 1 and Fig. 3.

At the PMT, the range of lateral movement follow-ing the neural mobilisation, was 0.01–12.30 mm, with amean value of 3.54 mm (standard error of measurement[SEM] �1.18 mm). This compares with AP movementwhich ranged from 0.02 to 4.19 mm with a mean valueof 1.61 mm (SEM � 0.78 mm). Paired t tests concludeda statistically significant difference between lateral andAP measurements at the PMT (p � 0.05).

With respect to the direction of the nerve move-ment, in 78% of scans at the PMT the nerve movedsuperiorly (i.e., toward the skin) and medially, 9%moved superior-laterally, 11% inferior (i.e., away fromthe skin) and medially and 2% inferior-lateral (Fig. 4).

ured for each direction at each scanning location

ion of sciatic nerve movement

LongitudinalLateral

ange Mean SEM Range Mean SEM

- 12.30 3.54 1.01 0.45 - 8.61 3.47 0.75- 10.90 6.62 1.38 0.71 - 7.60 5.22 0.48

M � standard error of measurement.e from those in Italics (n � 3).

Lateral PMTAP PMTLongitudinal PMTLateral PCAP PCLongitudinal PC

eral and anterior-to-posterior (AP)] and longitudinal) at

) meas

Direct

verse

R

0.010.31

se [lat

d popliteal crease [PC]) (SEM bars included).

rior-me

Reliability of measuring sciatic and tibial nerve movement ● R. ELLIS et al. 1213

At the PC scanning location, the range of lateralmovement, during the neural mobilisation, was 0.31–10.90 mm, with a mean value of 6.62 mm (SEM � 1.10mm). This compares with AP movement which rangedfrom 3.00–8.28 mm with a mean value of 3.26 mm(SEM � 0.99 mm). Paired t tests concluded a statisti-cally significant difference between these measurementsat the PC (p � 0.05).

There was a significant reduction in lateral movementat the PMT (mean � 3.54 mm) compared with the PC(mean � 6.62 mm; p � 0.05). There was also a significantreduction in AP movement at the PMT (mean � 1.61 mm)compared with the PC (mean � 3.26 mm; p � 0.05).

6-11-

Infe

rior (

deep

) to

supe

rior (

supe

rfic

ial)

mov

emen

t (m

m)

Lateral to Fig. 4. Direction of transverse nerve movement at th

superior-lateral; IM � infe

-11 -6

Lateral to

Infe

rior

(dee

p) to

sup

erio

r (s

uper

ficia

l) m

ovem

ent

(mm

)

Fig. 5. Direction of transverse nerve movement at th

superior-lateral; IM � inferior-me

With respect to the direction of the nerve move-ment, in 100% of scans at the PC, the nerve movedsuperiorly (i.e., toward the skin) and medially (Fig. 5).

Measurements of transverse sciatic nerve move-ment were possible across all 27 subjects.

Longitudinal Ultrasound Imaging of Sciatic NerveMovement

Data representing the longitudinal sciatic nervemovement measured at the PMT and PC are presented inTable 1 and Fig. 3.

At the PMT scanning location, the range of longi-tudinal movement during the neural mobilisation was

94

l movement (mm)erior midthigh (PMT). SM � superior-medial, SL �dial; IL � inferior-lateral.

4 9

movement (mm)

erior midthigh (PMT). SM � superior-medial; SL �

-11

-6

-1

4

9

1-

mediae post

-11

-6

-1

4

9

-1

medial

e post

dial; IL � inferior-lateral.

ce inte7) asid

1214 Ultrasound in Medicine and Biology Volume 34, Number 8, 2008

0.45–8.61 mm, with a mean value of 3.47 mm (SEM �0.79 mm; n � 27). This compares with longitudinalmovement measured at the PC, which ranged from 0.71–7.60 mm with a mean value of 5.22 mm (SEM � 0.05mm; n � 3). There was a statistically significant differ-ence between the two different scanning locations (PMTand PC) for longitudinal movement (p � 0.05).

With respect to the direction of the nerve move-ment, in 100% of scans at both the PMT and PC, thenerve moved distally.

Reliability of Diagnostic Ultrasound Assessment of Sci-atic Nerve Movement

The reliability of sciatic nerve movement recordedat the PMT resulted in an ICC, across three trials for eachof the 27 subjects, of 0.76 for lateral sciatic movement,compared with 0.39 for AP movement (Table 2). Bland-Altman plots representing these analyses (comparing thedifference between the results, plotted against their av-erage) are shown in Fig. 6 and Fig. 7, respectively.

The reliability of sciatic nerve movement recorded

Table 2. Intraclass correlation coefficients (ICC) of teeach sca

Tran

AP

Scanning location ICC 95% CI

Posterior midthigh (PMT) 0.392 0.154 - 0.625Popliteal fossa/crease

(PC) 0.560 0.335 - 0.750

n � number of subjects; AP � anterior-to-posterior; CI � confidenAll figures representative of data collected from 27 subjects (n � 2

Fig. 6. Bland-Altman graph (difference vs. average) for alltrials measuring anterior-to-posterior (AP) sciatic nerve move-ment at the posterior midthigh (PMT). Bolded line indicates

limits of agreement (95%).

at the PC resulted in an ICC of 0.70 for lateral sciaticmovement, compared with 0.56 for AP movement(Table 2).

The reliability of longitudinal sciatic nerve move-ment recorded at the PMT (n � 27) resulted in an ICC of0.75 (see Fig. 7). However, because data could only berecorded from three subjects (n � 3) for longitudinalmovement at the PC, it was not appropriate to calculatean ICC on such a small subject number (see Table 2).

DISCUSSION

Diagnostic ultrasound assessment of peripheralnerve movement is useful to establish the extent ofneural movement in both the normal and pathologicsituation. In the present study, sciatic nerve movementwas measured using diagnostic ultrasound, a techniquethat has mainly been used to examine nerve movement inthe upper limb (i.e., median and ulnar nerves). This studysought to determine the reliability of this technique formeasuring sciatic nerve movement and to make com-ment of sciatic nerve movement during neural mobilisa-tion.

Intraclass correlation coefficients were determinedas a measure of reliability (Table 2). Generally, an ICCbelow 0.40 represents poor reliability, 0.40–0.75 repre-sents fair reliability and above 0.75 represents excellentreliability (Fleiss 1986).

The reliability of the lateral component of the trans-verse nerve movement was considered excellent at thePMT (ICC � 0.76). However, the reliability of the lateralcomponent at the PC and the AP component of both thePMT and PC was considered to be fair (ICC � 0.39 to0.69).

Diagnostic ultrasound has been described as opera-tor dependant (Beekman and Visser 2003, 2004; Chiou etal. 2003; Martinoli et al. 2000; Peer et al. 2002) which

st reliability for each direction of nerve movement atlocation

ction of sciatic nerve movement

Longitudinal

Lateral

ICC 95% CI ICC 95% CI

.758 0.600 - 0.871 0.753 0.593 - 0.868

.697 0.510 - 0.837 0.970 0.732 - 0.999

rval.e from those in Italics (n � 3).

st-retenning

Dire

sverse

0

0

may account for the fair reliability result. Error can occur

Reliability of measuring sciatic and tibial nerve movement ● R. ELLIS et al. 1215

during placement of the digital markers. It is, therefore,important for the sonographer to be consistent with theplacement of the digital markers and calipers betweentrials. Blinding of the sonographer from the measure-ments themselves can help eliminate any measurementbias.

Although potential movement of the ultrasoundtransducer was taken into account during the analysis oflongitudinal movements (see Frame-by-frame cross-cor-relation algorithm and calculation software), it remains apossibility that any movement of the transducer over theskin may have influenced the levels of reliability whichwere calculated during assessment of transverse move-ment.

The reliability of longitudinal nerve movement wasconsidered excellent at the PMT (ICC � 0.75). At thePC, however, reliability analysis was performed in onlythree subjects. In the remaining 24 subjects, it was notpossible to obtain sequences of longitudinal nerve move-ment, since the nerve moved beyond the field of theultrasound image during the neural mobilization exer-cise, i.e., the nerve was no longer in line with the trans-ducer. Interestingly, a similar movement out of the imageplane was not seen at the PMT. The observed movementof the sciatic nerve out of the image plane during mea-surement of longitudinal sliding at the PC, which is alsoconsistent with an increase in transverse movement at thePC compared with the PMT, may indicate a region ofhigh compliance behind the knee (i.e., a wavy nervecourse). In the upper limb, there are regions of highcompliance across major joints in both the median andulnar nerve that allow for large changes in their bedlength during joint rotations (Dilley et al. 2003; Dilley etal. 2007). Through the shoulder region for example, boththe median and ulnar nerves follow a wavy course, whichaccommodates an increase in nerve bed length duringshoulder abduction. The presence of a high compliant

Fig. 7. Bland-Altman graph (difference vs. average) for alltrials measuring longitudinal sciatic nerve movement at theposterior midthigh (PMT). Bolded line indicates limits of

agreement (95%).

region behind the knee is consistent with unloading of

the sciatic nerve with the knee in a flexed position (i.e.,the knee was fixed in 50 degrees of flexion during theneural mobilization).

In the present study, movement of the sciatic nervewas generated with two simultaneous movements,namely ankle plantar flexion-dorsiflexion and cervicalflexion-extension. The results demonstrated sciatic nervemovement in different directional patterns dependingupon the site and location. At the PMT (i.e., the mostproximal site), in a majority of instances, distal longitu-dinal nerve movement was coupled with superior-medialtransverse movement (78%). In the minority of trials,movement was seen in other transverse directions (9%superior-lateral, 11%, inferior-medial and 2% inferior-lateral) (Fig. 4). This is in contrast to the trials recordedat the PC where in 100% of instances distal longitudinalmovement of the sciatic/tibial nerve was coupled withsuperior-medial transverse movement (Fig. 5). This find-ing could be explained by the differing mechanical in-terface at each location. At the PC, within the poplitealfossa, there is much more relative space for the nerve tomove more freely (consistent with an area of high com-pliance) compared with the tighter mechanical interfaceafforded by the hamstring musculature at the PMT.

Limitations of the StudyThis study utilised an ultrasound method which

enabled measurement from two-dimensional (2D) im-ages. This method does not allow analysis of simulta-neous movement within multiple planes. This poses alimitation not only of this study, but of all studies using2D imaging which purport to accurately measure trueand pure longitudinal nerve movement using diagnosticultrasound. It has been suggested that the use of three(3D) and four-dimensional (4D) diagnostic ultrasoundscanner may present an opportunity to measure real-timesoft-tissue movement in multiple planes, thereby, givinga more accurate representation of actual movement(Deng et al. 2000).

CONCLUSION

The study concluded that using diagnostic ultra-sound assessment and frame-by-frame cross-correlationanalysis to quantify sciatic nerve movement, it is possi-ble to provide excellent reliability for measurement oflongitudinal sciatic nerve movement during neural mo-bilisation. The reliability of the transverse movement ofthe sciatic nerve was fair irrespective of location ordirection.

The results of this study are clinically important toexpand into research that can allow quantitative mea-surement and comparison of in vivo, real-time sciatic

nerve movement in both normal and symptomatic pop-

1216 Ultrasound in Medicine and Biology Volume 34, Number 8, 2008

ulations. Clinically there is also a capacity to now moreaccurately conclude the benefit of neurodynamic assess-ment procedures and the therapeutic use of neural mo-bilisation.

Acknowledgments—The authors acknowledge the New Zealand Soci-ety of Physiotherapist, Scholarship Fund and the Maurice and PhyllisPaykel Trust Scholarship in Health Science. They also thank Mr. S.Allen, Sonographer, Horizon Scanning, AUT, Akoranga Dr., North-cote, Auckland.

REFERENCES

Beekman R, Visser LH. Sonography in the diagnosis of carpal tunnelsyndrome: A critical review of the literature. Muscle Nerve 2003;27:26–33.

Beekman R, Visser LH. High-resolution sonography of the peripheralnervous system - a review of the literature. Eur J Neurol 2004;11:305–314.

Beith ID, Richards PR, Robins EJ. Changes in length of peripheralnerve beds of the lower limb during movements of adjacent joints.J Anatomy 1995;185:700–701.

Bland J, Altman D. Statistical methods for assessing agreement be-tween two methods of clinical measurements. Lancet 1986;8:307–310.

Butler DS. The sensitive nervous system. Adelaide: Noigroup Publi-cations, 2000.

Chiou H-J, Chou Y-H, Chiou S-Y, Liu J-B, Chang C-Y. Peripheralnerve lesions: Role of high-resolution US. Radiographics 2003;23:e15 (Epub 2003).

Cleland JA, Childs JD, Palmer JA, Eberhart S. Slump stretching in themanagement of non-radicular low back pain: A pilot clinical trial.Manual Ther 2007;11:279–286.

Coppieters MW, Bartholomeeusen KE, Stappaerts KH. Incorporatingnerve-gliding techniques in the conservative treatment of cubitaltunnel syndrome. J Manipulative Physiol Ther 2004;27:560–568.

Coppieters MW, Butler DS. Do “sliders” slide and “tensioners” ten-sion? An analysis of neurodynamic techniques and considerationsregarding their application. Manual Ther 2007 (in press).

Deng J, Newton NM, Hall-Craggs MA, Shirley RA, Linney AD, LeesWR, et al. Novel technique for three-dimensional visualisation andquantification of deformable, moving soft-tissue body parts. Lancet2000;356:127–131.

Dilley A, Greening J, Lynn B, Leary R, Morris V. The use of cross-correlation analysis between high-frequency ultrasound images oflongitudinal median nerve movement. Ultrasound Med Biol 2001;27:1211–1218.

Dilley A, Lynn B, Greening J, DeLeon N. Quantitative in vivo studiesof median nerve sliding in response to wrist, elbow, shoulder and

neck movements. Clin Biomech 2003;18:899–907.

Dilley A, Summerhayes C, Lynn B. An in vivo investigation of ulnarnerve sliding during upper limb movements. Clin Biomech 2007;22:774–779.

Erel E, Dilley A, Greening J, Morris V, Cohen B, Lynn B. Longitudinalsliding of the median nerve in patients with carpal tunnel syndrome.J Hand Surg 2003;28B:439–443.

Fleiss JL. The design and analysis of clinical experiments. New York:John Wiley and Sons, 1986.

Greening J, Dilley A, Lynn B. In vivo study of nerve movement andmechanosensitivity of the median nerve in whiplash and nonspe-cific arm pain patients. Pain 2005;115:248–253.

Greening J, Lynn B, Leary R, Warren L, O’Higgins P, Hall-Craggs M.The use of ultrasound imaging to demonstrate reduced movementof the median nerve during wrist flexion in patients with non-specific arm pain. J Hand Surg 2001;26B:401–406.

Hashimoto BE, Kramer DJ, Wiitala L. Applications of musculoskeletalsonography. J Clin Ultrasound 1999;27:293–318.

Herrington L. Effect of different neurodynamic mobilization tech-niques on knee extension range of motion in the slump position. JManual Manipulative Ther 2006;14:101–107.

Hough AD, Moore AP, Jones MP. Measuring longitudinal nerve mo-tion using ultrasonography. Manual Ther 2000a;5:173–180.

Hough AD, Moore AP, Jones MP. Peripheral nerve motion measure-ment with spectral doppler sonography: A reliability study. J HandSurg 2000b;25B:585–589.

Hough AD, Moore AP, Jones MP. Reduced longitudinal excursion ofthe median nerve in carpal tunnel syndrome. Arch Phys MedRehabil 2007;88:569–576.

Jeffery R. Diagnostic ultrasound imaging and its role within musculo-skeletal medicine. Division of Rehabilitation and Occupation Stud-ies. Auckland: Auckland University of Technology, 2003.

LaBan M, MacKenzie J, Zemenick G. Anatomic observations in carpaltunnel syndrome as they relate to the tethered median nerve stresstest. Arch Phys Med Rehabil 1989;70:44–46.

Martinoli C, Bianchi S, Derchi L. Ultrasonography of peripheralnerves. Semin Ultrasound, CT MR 2000;21:205-–213.

Peer S, Kovacs P, Harpf C, Bodner G. High-resolution sonography oflower extremity peripheral nerves. J Ultrasound Med 2002;21:315–322.

Scrimshaw S, Maher C. Randomized controlled trial of neural mobili-zation after spinal surgery. Spine 2001;26:2647–2652.

Shacklock MO. Neurodynamics. Physiotherapy 1995;81:9–16.Shacklock MO. Clinical neurodynamics: A new system of neuromus-

culoskeletal treatment. Oxford: Butterworth Heinemann, 2005.Szabo RM, Bay BK, Sharkey NA, Gaut C. Median nerve displacement

through the carpal tunnel. J Hand Surg 1994;19:901–906.Totten PA, Hunter JM. Therapeutic techniques to enhance nerve glid-

ing in thoracic outlet syndrome and carpal tunnel syndrome. HandClin 1991;7:505–520.

Wright TW, Glowczewskie F, Wheeler D, Miller G, Cowin D. Excur-sion and strain of the median nerve. J Bone Joint Surg 1996;78A:

1897–1903.