See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/221768587 Construct Validity of Clinical Tests for Alar Ligament Integrity: An Evaluation Using Magnetic Resonance Imaging Article in Physical Therapy · January 2012 DOI: 10.2522/ptj.20110261 · Source: PubMed CITATIONS 8 READS 47 3 authors, including: Peter Osmotherly University of Newcastle 135 PUBLICATIONS 346 CITATIONS SEE PROFILE Darren A Rivett University of Newcastle 221 PUBLICATIONS 1,460 CITATIONS SEE PROFILE All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately. Available from: Peter Osmotherly Retrieved on: 20 August 2016
doi: 10.2522/ptj.20110261Originally published online January 19, 2012
2012; 92:718-725.PHYS THER. RowePeter G. Osmotherly, Darren A. Rivett and Lindsay J.ImagingIntegrity: An Evaluation Using Magnetic Resonance Construct Validity of Clinical Tests for Alar Ligament
http://ptjournal.apta.org/content/92/5/718found online at: The online version of this article, along with updated information and services, can be
Collections
Tests and Measurements Manual Therapy
Injuries and Conditions: Neck in the following collection(s): This article, along with others on similar topics, appears
e-Letters
"Responses" in the online version of this article. "Submit a response" in the right-hand menu under
or click onhere To submit an e-Letter on this article, click
E-mail alerts to receive free e-mail alerts hereSign up
by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from
Construct Validity of Clinical Tests forAlar Ligament Integrity: An EvaluationUsing Magnetic Resonance ImagingPeter G. Osmotherly, Darren A. Rivett, Lindsay J. Rowe
Background. The alar ligaments are integral to limiting occipito-atlanto-axialrotation and lateral flexion and enhancing craniocervical stability. Clinical testing ofthese ligaments is advocated prior to the application of some cervical spine manualtherapy procedures. Given the absence of validation of these tests and the potentialconsequences if manipulation is applied to an unstable upper cervical spine segment,exploration of these tests is necessary.
Objective. The purpose of this study was to examine the direct effect of theside-bending and rotation stress tests on alar ligaments using magnetic resonanceimaging (MRI).
Design. This was a within-participant experimental study.
Methods. Sixteen participants underwent MRI in neutral and end-range stress testpositions using proton density-weighted sequences in a 3-Tesla system. Measure-ments followed a standardized protocol relative to the position of the axis. Distanceswere measured from dens tip to the inferior margin of the foramen magnum and frommidsubstance of the dental attachment of the ligament to its occipital insertion.Between-side differences were calculated for each measurement to account forinherent asymmetries in morphology. Differences were compared between the testand neutral positions using a Wilcoxon signed rank test.
Results. Side-bending stress tests produced a median between-side difference inligament length of �1.15 mm. Rotation stress tests produced a median between-sidedifference in ligament length of �2.08 mm. Both results indicate increased measure-ment of the contralateral alar ligament.
Limitations. Assessment could be made only in the neutral position due toimaging limitations. Clinical texts state that tests should be performed in 3 positions:neutral, flexion, and extension.
Conclusions. Both side-bending and rotation stress testing result in a measurableincrease in length of the contralateral alar ligament. This finding is consistent withmechanisms that have been described to support their use in clinical practice.
P.G. Osmotherly, BSc, GradDip-Phty, MMedSci, School of HealthSciences, The University of New-castle, University Drive, New-castle, New South Wales, 2308Australia. Address all correspon-dence to Mr Osmotherly at: [email protected].
D.A. Rivett, BAppSc(Phty), Grad-DipManTher(ManipPhty), PhD,School of Health Sciences, TheUniversity of Newcastle.
L.J. Rowe, MAppSc, BMed,FRANZCR, School of Medicineand Public Health, The Universityof Newcastle.
[Osmotherly PG, Rivett DA,Rowe LJ. Construct validity of clin-ical tests for alar ligament integ-rity: an evaluation using magneticresonance imaging. Phys Ther.2012;92:718–725.]
The lack of established validityof the tests of ligamentous sta-bility of the upper cervical
spine brings into question their abil-ity to detect instabilities in thisregion. There is potential for anadverse outcome for a patient withan upper cervical spine instabilityundergoing treatment with manipu-lative techniques.1,2 Given this pos-sibility, this area warrants furtherresearch to improve both the safetyand treatment outcomes for patientsundergoing physical therapy man-agement of upper cervical spine dis-orders. An early step in the validationprocess for these clinical tests is theestablishment of construct validity toassess whether the tests are capableof influencing the alar ligaments.
The alar ligaments have beendescribed primarily as limitingoccipito-atlanto-axial rotation andlateral flexion.3–5 They pass from thesuperolateral aspect of the dens tothe medial surface of the occipitalcondyles.6–8 Loss of integrity of thealar ligaments removes a primarypassive restraint to rotation in theupper cervical spine. A loss of con-trol of rotation is associated withincreased likelihood of adverse neu-rovascular events that have beenassociated with high-velocity thrustand end-range techniques used inthe upper cervical spine.9,10
Both the side-bending and rotationstress tests for the alar ligaments arebased on preventing the inherentcoupling of rotation and lateral flex-ion in the occipito-atlanto-axial com-plex. That is, lateral flexion of theocciput on the atlas is accompaniedby immediate ipsilateral rotation ofthe axis beneath the atlas. This rota-tion was proposed by Dvorak andPanjabi3 to result from tension gen-erated in the alar ligaments.
The side-bending stress test, firstproposed by Aspinall,11 has beendescribed for both sitting12,13 and
supine14 positions. In performingthis test, the spinous process andlamina of the axis are stabilizedby the therapist to prevent bothside bending and rotation of thesegment.11 Slight compression isapplied through the crown of thehead to facilitate atlanto-occipitalside bending. Passive side bendingthen is applied using pressurethrough the patient’s head; in effect,directing the patient’s ear toward theopposite side of the neck.11,12,15,16 Iffixation of the axis is adequate, thenormal coupled movement will notbe permitted to occur. Hence, nolateral flexion should occur. Testingis recommended to be performed in3 planes (neutral, flexion, and exten-sion) to account for variation in alarligament orientation.11,16 For a side-bending stress test to be consideredpositive for an alar ligament lesion,excessive movement in all 3 planesof testing should be evident.15,16
The rotation stress test14–16 isregarded as primarily stressing thecontralateral alar ligament in accor-dance with the biomechanicaldescription of Dvorak et al.6 Again,the test is described for both sit-ting15,16 and supine14 positions. Theaxis is stabilized around its laminaeand spinous process using a lumbri-cal grip. The cranium is grasped witha wide hand span and then rotated,the occiput taking the atlas segmentwith it, to the end of available range.No lateral flexion is permitted. Somerotation will occur during the test,but the extent of rotation within thebounds of normal is subject to somevariation. Estimates of the range ofnormal rotation vary between 20 and40 degrees.12–16 As with the side-bending test, the test is repeated in 3positions in the sagittal plane, withlaxity in all 3 positions necessary toestablish a positive test finding.14–16
The aim of the study was to examinethrough magnetic resonance imag-ing (MRI) the direct effect of clinical
stress tests described for the alarligaments to assess whether thesetests are capable of demonstratingabnormalities of these structures.Using individuals without instability-related pathologies of the craniover-tebral region, we proposed to exam-ine whether a measurable change inligament length occurred when aspecific stress test was applied tothe ligament structure comparedwith measurements taken with thecervical spine in a neutral position.
MethodParticipantsSixteen skeletally mature partici-pants were recruited sequentiallyvia advertisement from the popula-tion of The University of Newcastle,Newcastle, New South Wales, Aus-tralia. To be eligible for inclusion,participants had to be between theages of 18 and 35 years. The upperage limit was imposed to mitigatethe effect of degenerative changeon cervical spine movement duringtesting. Potential participants wereexcluded if they had a history ofcervical spine trauma or recurrentpharyngeal infection, had been diag-nosed with an inflammatory diseaseor an instability of the craniover-tebral region, had any congenitaldisorder recognized to have thepotential for instability of the cranio-vertebral region, or experiencedclaustrophobia or discomfort in con-fined spaces.
Eight male and eight female indi-viduals satisfying all criteria forinclusion volunteered to participate.The average age of the female par-ticipants was 23 years 10.6 months(SD�72.3 months). The average ageof the male participants was 25 years4.6 months (SD�57.6 months).
Clinical Stress Tests ExaminedThe stress tests examined in thisstudy were the side-bending stresstest11 and the rotation stress test.15
Each test was administered by a
Construct Validity of Clinical Tests for Alar Ligament Integrity
May 2012 Volume 92 Number 5 Physical Therapy f 719 by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from
single investigator with the partici-pants in a supine position. The side-bend or rotation movement imposedduring each test was directed to theright in each case. Testing was notperformed away from neutral in thesagittal plane because the requireddegree of sagittal-plane positioningwould move the alar ligaments awayfrom the posterior imaging coil, ren-dering them unclear on the resultantimages.
Imaging of ParticipantsImages were acquired in the coronalplane using a Siemens MagnetomVerio Syngo MR B17 MRI systemwith a 3-Tesla magnet (Siemens AG,Erlangen, Germany). All ligamenttests were performed consecutivelyin a supine position within the MRIbore, with participants enclosed in aphased array neck coil (Fig. 1). Aneutral image was acquired as a ref-erence study at the commencementof each participant’s examination.The neutral head and neck position
for each participant was definedusing criteria published previously,whereby the participant was posi-tioned such that a line between theforehead and chin was horizontaland parallel to the examination tableand an imaginary line running paral-lel to the table extended from thetragus of the ear would pass alongthe axis of the neck longitudinally.17
A proton density-weighted turbospin echo sequence was used withthe following parameters: repetitiontime�1,000 milliseconds, echo time�38, field-of-view 150 � 150 mm,image matrix 320 � 320, image res-olution 0.5 � 0.5 � 1.5 mm (phaseencoding direction right to left).Sixty slices were generated with aslice thickness of 1.5 mm. The totalacquisition time for each sequencewas 3 minutes 20 seconds.
Measurement of MRI ImagesViewing and analysis of all imageswere performed using OsiriX 3.5
Method of standardizing refer-ence position. Each test exam-ined has been described as move-ment of the atlas or occiput withrespect to a stationary, manuallystabilized axis. To ensure consis-tency with the test description andpermit accurate and reproduciblemeasurement, each image was mea-sured with reference to a standard-ized position of the axis.
To create anatomically based scanplanes to standardize axis position,each data set was displayed usinga multiplanar reconstruction. In thesagittal reconstruction, a sectionpassing parallel to the longitudinalaxis of the odontoid process andorthogonal to the plane of the inter-body joint of C2–3 was selected. Inthe coronal reconstruction, a sectionpassing longitudinally down the mid-line of the odontoid process andbisecting the body of the second cer-vical vertebra was selected. In hori-zontal reconstruction, a section cen-tered on the center point of thecross-section of the odontoid pro-cess was selected. To account forany rotation of the axis present, theimage was rotated when necessarysuch that the plane of the sectioncontained the transverse foraminaeof the axis in alignment.
Methods of measurement foreach ligament test position.Each image was measured on 2 sep-arate occasions to establish reliabilityof measurement according to theprotocol that follows.
The effectiveness of the tests in ten-sioning the alar ligaments was mea-sured in the coronal plane usingboth direct and indirect techniques.In the absence of validated publishedmethods to assess alar ligamentlength, indirect estimation was usedto provide concurrent measurement
Figure 1.Patient position in standard neck coil for side-bending testing during imaging.
Construct Validity of Clinical Tests for Alar Ligament Integrity
720 f Physical Therapy Volume 92 Number 5 May 2012 by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from
of change in the relationshipbetween points of bony prominenceadjacent to the attachment sites ofthe ligaments. To estimate displace-ment of the occiput from the sta-bilized axis, the distance from thetip of the odontoid process to theinferior aspect of the foramen mag-num was measured bilaterally. Directestimation was performed by firstselecting the midpoint of the dentalattachment of the alar ligament. Aline corresponding to the axis of theligament between origin midpointand insertion into the occiput wascreated and measured. These mea-surements are depicted in Figure 2(A and B).
Data AnalysisAnalysis of all data was performedusing Stata 11.0 statistical software(Stata Corporation, College Station,Texas). Due to the inherent asym-metries of the morphology of theregion, including variation in orien-tation of the odontoid process andthe individual ligaments in all 3 ana-
tomical planes, analysis of the alarligament tests was undertaken usingthe difference between left-sidedand right-sided measurements as thebase variable of analysis. A variablerepresenting the difference in mea-sured distance between bony land-marks and between alar ligamentlengths was generated for each mea-sure in all test positions as the mea-sure of the right side subtracted fromthe measure of the left side.
Exploratory data analysis was used todescribe the difference and spread ofdata representing the generated vari-ables of the left to right differencefor each measure. The distribution ofeach variable was assessed both visu-ally using histograms of the data andnormal probability plots and statisti-cally using the Shapiro-Wilk test fornormality. Hypothesis testing com-paring the left to right differences inboth measured distance betweenbony landmarks and actual ligamentlength was done by analyzing thedifference estimates in the test posi-
tions compared with the differenceestimates in the neutral position.Each hypothesis test was performedusing the nonparametric Wilcoxonsigned rank test for paired variables.Reliability of measurements for eachimage was assessed by estimation ofintraclass correlation coefficients forthe recorded measurements of theimage taken on 2 separate occasions.
Role of the Funding SourceThis research was supported bya Physiotherapy Research Founda-tion grant.
ResultsThe measured lengths of the dis-tance between the tip of the odon-toid process and the foramen mag-num and the direct measurementsof the alar ligaments for each sideand for each position are given inTable 1. After application of eachstress test for the alar ligament, anincrease in left-sided length was evi-dent in each participant (Fig. 3).
Figure 2.(A) Measurement from the tip of the odontoid process to the inferior margin of the foramen magnum indicated by the arrow. (B)Direct estimation of alar ligament length. A line corresponding to the axis of the ligament (indicated by the arrow) is generated andmeasured between origin midpoint and insertion into the occiput.
Construct Validity of Clinical Tests for Alar Ligament Integrity
May 2012 Volume 92 Number 5 Physical Therapy f 721 by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from
Bony Estimation:Tip of the OdontoidProcess to the Foramen MagnumIn the neutral position, the medianleft-right difference in this measurewas 0.02 mm (interquartile range[IQR]��0.05 to 0.23). After imposi-tion of the side-bending stress test,the median left-right difference wascalculated as 0.85 mm (IQR�0.10 to2.52), indicating a lengthening of theinterval on the left side comparedwith the neutral position. On rota-tion stress testing, the median left-right difference was calculated as1.44 mm (IQR�0.76 to 1.90), againindicating an increased distancebetween landmarks on the left side.
Between-position comparisons usingthe neutral and test position mea-surements were statistically signifi-cant for each alar ligament stress test(Tab. 2).
Direct Measurement ofAlar Ligament LengthThe median left-right differencein alar ligament length in the neutralposition was �0.05 mm (IQR��0.17to 0.36). With the imposition of theside-bending stress test, the medianleft-right difference increased to1.15 mm (IQR�0.58 to 1.67), indi-cating a greater length of the left-sided alar ligament. Upon rotationstress testing, the median left-right
difference increased to 2.08 mm(IQR�1.09 to 2.60), again indicatingan increase in measurable length ofthe left-sided alar ligament. Compar-isons between the stress test posi-tion and the neutral position werestatistically significant for each stresstest examined (Tab. 2).
Intraclass correlation coefficients forthe estimation of left-right differencefor each measurement in each posi-tion are given in Table 3. Reliabilityof measurement ranged from mod-erate to substantial according toaccepted criteria.18
DiscussionThis is the first study to demonstratea direct effect of these clinical testson the alar ligaments, providingsupport for the construct validityof these screening tests. The useof a detailed, standardized protocolto orient and measure the imagesof the testing procedures providesconsiderable rigor and consistencyto the findings of this study. Thisconsistency is underscored by themagnitude of the intraclass corre-lation coefficients assessed for eachmeasurement in each positionimaged. This standardized protocol,using 3-dimensional reconstructionsto create a reference position forthe axis, accounted for movementsoccurring across all 3 planes duringthe testing procedures.
The observed direct effect of testingon the alar ligaments has been cor-roborated by changes measured con-
Table 1.Measurements of Indirect and Direct Methods of Ligament Length for Left and Right Alar Ligaments
currently in the distance betweenthe tip of the odontoid process andthe foramen magnum. This findingindicates that the method of alar lig-ament measurement used is a validrepresentation of its length.
The substantial reliability of imagemeasurement demonstrated that theinherent inaccuracies in measure-ment due to vibration while sus-taining end-range positions wereminimized, despite the subsequentreduction in image quality. Themain undesirable effect of sustainingeach test position in excess of 3minutes would be to lose the end-range position and hence reducethe measured differences betweenthe neutral and test positions. Thus,the changes shown in this study maypossibly be considered an underesti-mate of the potential displacementoccurring during these tests. Theconsistent findings of a measurabledisplacement thus should be consid-ered to be a conservative estimate ofthe true displacement that may beachieved during the application ofthese stress tests.
In the neutral position, no significantdifference was noted in the ligamentlengths measured or the bony esti-mations of ligament attachment.Thus, any left-right difference foundon testing may be attributed tothe application of the test proce-dure. Both the side-bending and therotation stress tests resulted in ameasurable change in the distancesassessed. In each case, the left-sidemeasurements increased relative tothe right side, indicating a directlengthening of the left, that is, con-tralateral alar ligament. This findingindicates that the 2 stress testsapplied in this study both demon-strated a direct effect on the alarligaments.
Aspinall11 proposed the side-bendingstress test as a mechanism for testingthe contralateral alar ligament. These
findings are consistent with thetesting mechanism as described byAspinall. However, based on thedescriptions of Dvorak and Panjabi,3
it also has been suggested that test-ing in both directions is necessaryto infer instability due to both alarligaments tensioning bilaterally dur-ing side bending.16 In the currentstudy, a clear difference betweensides was evident during side-bending testing. This finding indi-cates that within the ranges in whichthese ligaments were tested, a bilat-eral effect on the alar ligaments is notevident and the need for a finding oflaxity in both directions is not nec-essary to infer instability.
The mechanism attributed to therotation stress test is the preventionof coupled movement within theoccipito-atlanto-axial complex. Rota-tion of the occiput over a stationaryaxis results in the contralateralalar ligament being wound around
the odontoid process due to its pos-terior attachment on the odontoid.6
Under normal biomechanical cir-cumstances, the odontoid would bepermitted to shift laterally, thus ten-sioning the ipsilateral ligament.However, if the maintenance of theaxis position is effective and cranio-cervical side bending is effectivelyminimized through manual stabil-ization, the limiting feature of therotation movement should be thetension developed in the contra-lateral alar ligament. The findings ofthe current study are consistentwith this mechanism. A clear differ-ence in length developed betweenthe alar ligaments in each partici-pant, with the contralateral ligamentplaced in a comparatively length-ened position under test positions inall participants.
The side-bending stress test resultedin a mean increase in indirect mea-
Table 2.Summary of Findings Following the Examination of Alar Ligament Stress Testing
Position
Left-Right Difference
Distance Tip ofOdontoid Process to
Foramen Magnum (mm)
DirectMeasurement of
Alar Ligament Length (mm)
Median IQRaP Value forDifference Median IQR
P Value forDifference
Neutral 0.02 �0.05 to 0.23 �0.05 �0.17 to 0.36
Side bending 0.85 0.10 to 2.52 .002 1.15 0.58 to 1.67 �.001
Rotation 1.44 0.76 to 1.90 �.001 2.08 1.09 to 2.60 �.001
a IQR�interquartile range.
Table 3.Intraclass Correlation Coefficients (ICCs) for the Left-Right Difference inAlar Ligament Length Measurements
Position Left-Right Difference Assessed ICC 95% CIa
Neutral Odontoid process to foramen magnum .85 .63 to .95
Direct measurement of ligament length .81 .54 to .93
Side bending Odontoid process to foramen magnum .63 .24 to .86
Direct measurement of ligament length .83 .58 to .94
Rotation Odontoid process to foramen magnum .68 .29 to .88
Direct measurement of ligament length .62 .22 to .85
a CI�confidence interval.
Construct Validity of Clinical Tests for Alar Ligament Integrity
May 2012 Volume 92 Number 5 Physical Therapy f 723 by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from
surements of 1.24 mm and anincrease in direct ligament measure-ment of 1.22 mm. Hence, both indi-cations of ligament length are consis-tent and strongly correlated (r�.76).Rotation stress testing resulted in amean increase in ligament approxi-mation by skeletal measurement of1.60 mm and a direct ligamentmeasurement of 1.88 mm. Again,the measured effect on ligamentlength was consistent in direction,with moderate correlation betweenthese measures (r�.65). From thesefindings, it may be considered thatthe rotation stress test produces agreater measurable effect on the con-tralateral alar ligament than the side-bending stress test.
Although assessment of the alar liga-ments was undertaken only in theneutral plane for imaging reasons asdescribed previously, testing intoboth flexion and extension shouldexhibit the same findings, consider-ing the mechanism of the clinicaltests applied. Moreover, the majorityof alar ligament specimens examinedin previous dissection19 and radiolog-ical20 studies were oriented in thehorizontal plane. Where caudal orcranial orientation was noted, theangles were smaller than illustratedin standard texts.21 Hence, findingsin the neutral position will be trans-ferable to the majority of alar liga-ments in the adult population.
Previous studies also have indicatedthat a proportion of alar ligamentseither have anterior portions thatdo not attach to the odontoid pro-cess or may even bypass the odon-toid process entirely.22,23 It is notpossible to identify people whoseligament arrangement might reflectthis morphology under clinicalexamination. Hence, some radiolog-ically demonstrable ligament injuriesmay not be perceptible in some indi-viduals using these standard clinicalstress tests.
Although clinical texts suggest thatthe alar ligament tests should be per-formed in 3 positions (neutral, flex-ion, and extension), we assessedthese procedures only in the neutralposition. Pretrial piloting of the tech-niques showed that retesting in fur-ther positions in the sagittal planeresulted in extensive loss of MRI sig-nal due to the separation of thepatient from the anterior portion ofthe head coil, rendering the imagesacquired unreadable.
Although the findings of this studyprovide support for the constructvalidity of these 2 clinical tests bydemonstrating a direct effect onalar ligaments in an asymptomaticpopulation, it should be noted thatneither the validity nor the reliabilityof these tests has yet been estab-lished in a clinical population. Inthe only article published previouslyexamining alar ligament testing,Kaale and colleagues24 demonstratedhigh specificity and moderate sensi-tivity for detecting alar ligamentlesions in a mixed population domi-nated by people with a history ofchronic whiplash-associated disor-der based on an assessment of thequality of occipito-atlanto-axial rota-tion performed by one examiner.Assessment of tests under conditionsof pain and muscle spasm and in thepresence of other induced symp-toms is necessary to evaluate theirclinical utility.
ConclusionThis study has established success-fully a reproducible method for theassessment of the clinical stress testsof the upper cervical spine liga-ments. By using rigorously definedmethods of standardization of theaxis as a reference position and aclearly defined measurement proto-col, the measurements producedhave been demonstrated to be highlyreliable. This study has addressedan important limitation of previousstudies using MRI and will permit
more accurate examination of thealar ligaments in future research.
Both the rotation and the side-bending stress tests for the alar liga-ments have been demonstrated toincrease the length of the contra-lateral alar ligament during testing.In contrast to the opinions of someauthors, no bilateral effect wasobserved.
Mr Osmotherly and Dr Rivett providedconcept/idea/research design and writing.All authors provided data collection andfund procurement. Mr Osmotherly and MrRowe provided data analysis. Dr Rivett pro-vided project management. Mr Osmotherlyprovided participants. Dr Rowe providedfacilities/equipment, institutional liaisons,and consultation (including review of man-uscript before submission).
Ethical approval for this study was grantedby the Hunter New England HumanResearch Ethics Committee.
An oral presentation of this research wasgiven at the Australian Physiotherapy Asso-ciation Conference; October 27–30, 2011;Brisbane, Queensland, Australia.
This research was supported by a Physiother-apy Research Foundation grant.
DOI: 10.2522/ptj.20110261
References1 Di Fabio R. Manipulation of the cervical
spine: risks and benefits. Phys Ther.1999;79:50–65.
2 Rivett DA, Milburn P. Complications aris-ing from spinal manipulative therapy inNew Zealand. Physiotherapy. 1997;83:626–632.
3 Dvorak J, Panjabi MM. Functional anatomyof the alar ligaments. Spine. 1987;12:183–189.
4 Harris MB, Duval MJ, Davis JAJ, BerniniPM. Anatomical and roentgenographic fea-tures of atlantooccipital instability. J Spi-nal Disord. 1993;6:5–10.
5 White AAI, Panjabi MM. Clinical Biome-chanics of the Spine. 2nd ed. Philadelphia,PA: JB Lippincott Co; 1990.
6 Dvorak J, Schneider E, Saldinger P, Rahn B.Biomechanics of the craniocervicalregion: the alar and transverse ligaments.J Orthop Res. 1988;6:452–461.
7 Gardner E, Gray DJ, O’Rahilly R. Anatomy:A Regional Study of Human Structure.4th ed. Philadelphia, PA: WB Saunders Co;1975.
Construct Validity of Clinical Tests for Alar Ligament Integrity
724 f Physical Therapy Volume 92 Number 5 May 2012 by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from
8 Panjabi MM, Oxland TR, Parks EH. Quan-titative anatomy of the cervical spine liga-ments, part 1: upper cervical spine. J Spi-nal Disord. 1991;4:270–276.
9 Paciaroni M, Bogousslavsky J. Cerebrovas-cular complications of neck manipulation.Eur Neurol. 2009;61:112–118.
10 Ernst E. Manipulation of the cervical spine:a systematic review of case reports of seri-ous adverse events, 1995–2001. Med JAust. 2002;176:376–380.
11 Aspinall W. Clinical testing for the cranio-vertebral hypermobility syndrome. J OrthopSports Phys Ther. 1990;12:47–54.
12 Gibbons P, Tehan P. Manipulation ofthe Spine, Thorax and Pelvis: An Osteo-pathic Perspective. Edinburgh, Scotland:Churchill Livingstone; 2004.
14 Hing W, Reid D. Cervical Spine Manage-ment: Pre-screening Requirement forNew Zealand. Auckland, New Zealand:New Zealand Manipulative Physiothera-pists Association; 2004.
15 Pettman E. Stress tests of the cranioverte-bral joints. In: Boyling JD, Palastanga N,eds. Grieve’s Modern Manual Therapy:The Vertebral Column. 2nd ed. Edin-burgh, Scotland: Churchill Livingstone;1994:529–537.
16 Beeton K. Instability in the upper cervicalregion: clinical presentation, radiologicaland clinical testing. Man Ther. 1995;27:19–32.
17 Falla DL, Jull GA, Hodges PW. Patientswith neck pain demonstrate reduced elec-tromyographic activity of the deep cer-vical flexor muscles during performanceof the craniocervical flexion test. Spine.2004;29:2108–2114.
18 Shrout P. Measurement reliability andagreement in psychiatry. Stat Meth MedRes. 1998;7:301–317.
19 Osmotherly PG, Rivett DA, Mercer SR. Cra-niovertebral anatomy and implications forclinical testing of ligamentous stability inthe upper cervical spine. Paper presentedat: 9th Congress of the International Fed-eration of Orthopaedic Manipulative Ther-apists; June 8–13, 2008; Rotterdam, theNetherlands.
20 Krakenes J, Kaale BR, Rorvik J, Gilhus NE.MRI assessment of normal ligamentousstructures in the craniovertebral junction.Neuroradiology. 2001;43:1089–1097.
21 Osmotherly P, Rawson O, Rowe L. Therelationship between dens height and alarligament orientation: a radiological study.J Manipulative Physiol Ther. 2011;34:181–187.
22 Osmotherly P, Mercer S, Rivett D. Trans-verse band of the alar ligament: a commonanatomical variant. Clin Anat. 2008;21:274.
23 Poirier PJ, Nicolas A, Charpy A. Traited’Anatotomie Humaine. 3rd ed. Paris,France: Massan; 1911.
24 Kaale BR, Krakenes J, Albrektsen G,Wester K. Clinical assessment techniquesfor detecting ligament and membrane inju-ries in the upper cervical spine region: acomparison with MRI results. Man Ther.2008;13:397–403.
Construct Validity of Clinical Tests for Alar Ligament Integrity
May 2012 Volume 92 Number 5 Physical Therapy f 725 by guest on December 30, 2015http://ptjournal.apta.org/Downloaded from
doi: 10.2522/ptj.20110261Originally published online January 19, 2012
2012; 92:718-725.PHYS THER. RowePeter G. Osmotherly, Darren A. Rivett and Lindsay J.ImagingIntegrity: An Evaluation Using Magnetic Resonance Construct Validity of Clinical Tests for Alar Ligament
References
http://ptjournal.apta.org/content/92/5/718#BIBLfor free at: This article cites 16 articles, 2 of which you can access
