Original Research

Differences in the Association of Hip Cartilage Lesionsand Cam-Type Femoroacetabular Impingement WithMovement Patterns: A Preliminary StudyDeepak Kumar, PT, PhD, Alexander Dillon, Lorenzo Nardo, MD, Thomas M. Link, MD,Sharmila Majumdar, PhD, Richard B. Souza, PT, PhD

D.K. Musculoskeletal Quantitative ImagingResearch Group, Radiology and BiomedicalImaging, University of CaliforniaeSan Fran-cisco, 1700 4th St, Suite 203, Byers Hall,UCSF Mission Bay, San Francisco, CA 94158.Address correspondence to: D.K.; e-mail:deepak.kumar@ucsf.edu or krdeepak2pro@gmail.comDisclosures related to this publication: grant(money to institution), NIH R01 AG017762,P50 AR060752

A.D. Musculoskeletal Quantitative ImagingResearch Group, Radiology and BiomedicalImaging, University of CaliforniaeSan Fran-cisco, San Francisco, CADisclosures: nothing to disclose

L.N. Musculoskeletal Quantitative ImagingResearch Group, Radiology and BiomedicalImaging, University of CaliforniaeSan Fran-cisco, San Francisco, CADisclosures related to this publication: grant(money to institution), NIH

T.M.L. Musculoskeletal Quantitative ImagingResearch Group, Radiology and BiomedicalImaging, University of CaliforniaeSan Fran-cisco, San Francisco, CADisclosures related to this publication: grant(money to institution), NIH. Disclosuresoutside this publication: grants/grants pending(money to institution), GE; Royalties (moneyto author), Springer

S.M. Musculoskeletal Quantitative ImagingResearch Group, Radiology and BiomedicalImaging, University of CaliforniaeSan Fran-

Objective: To investigate the differences in hip movement patterns during different dailyand athletic activities in persons with cam-type femoroacetabular impingement (FAI) withand without cartilage lesions compared with control subjects in a preliminary study.Design: Controlled laboratory study using a cross-sectional design.Setting: Research institution with a tertiary care medical center.Participants: Fifteen subjects [M:F, 13:2; age, 31.6 � 9.7 years (range, 22-52 years);body mass index, 24.9 � 4.6 (range, 18.8-38.4); FAI:control, 7:8].Methods: All subjects had 3-Tesla magnetic resonance imaging of the hip and also un-derwent 3-dimensional motion capture during walking, deep-squat, and drop-landing tasks.Experienced radiologists graded cartilage lesions on clinical magnetic resonance images.Outcomes: Peak kinematic and kinetic variables were compared between subjects whodid and did not have FAI, and subjects who had FAI and cartilage lesions were comparedwith subjects who did not have cartilage lesions.Results: Subjects who had FAI demonstrated no significant differences for walking ordrop landing compared with control subjects. However, during the deep-squat task,subjects with FAI adducted more and had a greater internal rotation moment. Subjects whohad cartilage lesions in the presence of a cam lesion demonstrated (1) no difference forwalking; (2) greater adduction, greater internal rotation moment, and lower transverseplane range of motion during the deep-squat task; and (3) greater adduction and lowerinternal rotation during the drop-landing task compared with subjects who did not havecartilage lesions.Conclusions: We observed differences in movement patterns between subjects who hadFAI compared with control subjects. However, the differences were more pronouncedbetween subjects with FAI who had cartilage lesions compared with subjects who did nothave cartilage lesions. These findings highlight the importance of understanding thecomplex interplay between bony morphologic features, cartilage lesions, and movementpatterns in persons with cam-type FAI.

PM R 2014;6:681-689

cisco, San Francisco, CADisclosures related to this publication: grant (money to institution), NIH

R.B.S. Musculoskeletal Quantitative ImagingResearch Group, Radiology and BiomedicalImaging, University of CaliforniaeSan Fran-cisco, San Francisco, CA, and Physical Ther-apy, University of CaliforniaeSan Francisco,San Francisco, CADisclosures related to this publication: grant(money to institution), NIH

Funding provided by NIH grants R01AG017762 and P50 AR060752.

Peer reviewers and all others who controlcontent have no relevant financial relation-ships to disclose.

Submitted for publication February 12, 2013;accepted February 12, 2014.

INTRODUCTION

Hip osteoarthritis (OA) is a major cause of pain and disability, with 4.4% of adults olderthan 55 years reporting symptomatic hip OA [1]. Femoroacetabular impingement (FAI) hasbeen identified as a risk factor for the development of hip OA [2-4]. FAI is primarilyclassified into cam and pincer types based on the morphologic abnormalities of the femoralhead-neck junction or the acetabulum, respectively [5-7]. However, FAI with features ofboth cam- and pincer-type FAI has also been reported [2]. The medical literature suggeststhat 66%-75% of persons with FAI present with the cam-type deformity [8]. Cam-typeimpingement is identified as a decreased offset at the anterolateral femoral head-neckjunction or a nonspherical extension of the femoral head [5,9,10]. It is thought that thecam-type lesion could be associated with abrasive damage as the femoral head translatesover the acetabulum during the repetitive activities of flexion and internal rotation [2].

PM&R1934-1482/14/$36.00

Printed in U.S.A.

ª 2014 by the American Academy of Physical Medicine and RehabilitationVol. 6, 681-689, August 2014

http://dx.doi.org/10.1016/j.pmrj.2014.02.002681

682 Kumar et al HIP CARTILAGE LESIONS, FAI, AND MOVEMENT PATTERNS

Symptomatic cam-type FAI clinically presents as groin painand reduced internal rotation at the hip [5,11-14]. It has beenreported that up to 14%of asymptomatic adultmen and 6%ofasymptomatic adult women have the cam-type lesion as well[6,7]. Furthermore, a recent report showed that 24% ofasymptomatic adults have cartilage defects [15]. Adding to theconfusion, not all persons with the cam-type FAI progress tohip OA [16-19]. The degree of deformity (ie, the size of thelesion), soft-tissue features (eg, the presence, size, and locationof cartilage and labral tears), function (amount and types ofphysical activity), and other unknown factors are likely relatedto the risk that persons with a cam-type lesion will be symp-tomatic or progress to OA [17-19]. Hence it is critical to un-derstand how morphologic features and function interact inpatients with FAI to be able to selectively identify persons atrisk of further cartilage changes.

Abnormal hip movement patterns during daily activitiessuch as walking and squatting have been reported in patientswith cam-type FAI [20,21]. Repetitive excessive hip jointloading due to abnormal movement patterns could be associ-ated with cartilage damage over a period, resulting in hip OA.Hence it is critical to identify these cartilage defects early in thedisease process so that surgical or nonsurgical interventions canbe implemented before further damage occurs. Furthermore, itis important to characterizemovement patterns in the setting ofcartilage lesions so that movement retraining interventions canbe developed and implemented. Earlier reports of kinematicand kinetic differences in persons with FAI have focused onwalking and squatting [20,21]. Because FAI is commonly seenin athletes, it is important to characterize hip movement pat-terns during high-demand tasks such as drop landing; char-acterization of these patterns has not been described thus far.

Movement patterns are commonly characterized with useof kinematic and kinetic descriptors from a motion analysisevaluation [22]. During a motion analysis evaluation, datacollected include segment motion and ground reaction force.These data are used to quantify the kinematic patterns (eg,measures of peak joint motion and excursions in the sagittal,coronal, and transverse planes) and kinetic patterns (eg,peak joint moments and powers in the 3 planes) at the hipjoint during different activities [23,24]. Joint moments are anestimate of the loading across the hip joint and are reportedas net moments (ie, no information on muscle cocon-traction). For instance, a net external flexion moment at thehip during early stance indicates that there is a momenttending to flex the hip that is countered by an equal andopposite net internal extension moment generated by mus-cles and other soft tissues. Joint powers are calculated as theproduct of the joint moment and joint angular velocity [23].A positive joint power indicates concentric work at the jointbecause the joint moment and joint motion are in the samedirection. A negative joint power, on the other hand, in-dicates eccentric work at the joint because the joint momentand joint motion are in opposite directions. These tech-niques have not been used to explore the differences in the

movement patterns of persons with FAI who may or may nothave cartilage lesions compared with control subjects.

Hence the aim of this preliminary exploratory study wasto investigate the differences in movement patterns duringwalking, squatting, and drop landing between persons withFAI and control subjects and also between persons with FAIwho have cartilage lesions compared with persons withoutcartilage lesions.

MATERIALS AND METHODS

Subjects

Seven subjects with unilateral symptomatic FAI (age, 36.6 �9.7 years; bodymass index [BMI], 26.2� 6.9 kg/m2) [25] and8 asymptomatic volunteers (age, 27.3 � 7.7 years; BMI, 23.9� 2.3 kg/m2) participated in the study. All procedures wereapproved by an Institutional Committee on Human Research,and all subjects signed informed consent prior to datacollection. The subjects with FAI were recruited from the or-thopedic faculty practice at our institution after having apositive anterior impingement test (ie, pain accompanying hipflexion, adduction, and internal rotation), visual cam diseaseon anteroposterior and frog-leg lateral radiographs, and analpha angle greater than 55� on oblique axial magnetic reso-nance imaging (MRI; a common diagnostic criteria for cam-type FAI) [10]. The control participants were recruited fromthe student population at our institution and related com-munity networks and were included only if they had no lowerlimb pain, a clinical history negative for serious lower ex-tremity injury or surgery, and no radiologic features consistentwith FAI. Subjects from either groupwere excluded if they hadany neurologic disorders that would affect their ability toperform the motion analysis tasks or any implanted metaldevices that could interact with the magnetic field.

MRI Acquisition

All imaging was performed with a 3-Tesla MR scanner (GEMR750, GE Healthcare, Waukesha, WI) and an 8-channelcardiac coil (GE Healthcare). Following our establishedstandard operations manual, patient-positioning aids wereused to immobilize and support patients, resulting in aconsistent, reproducible, and comfortable hip position dur-ing scanning. The lower extremities were fully extended witha foam lower leg immobilizer to maintain neutral rotation.The feet were comfortably taped together to minimizemovement. Sagittal and coronal T2-weighted fat-saturatedfast spineecho scans (repetition time/echo time [TR/TE],3678/60 msec; field of view [FOV], 14 cm; matrix, 288 �224; slice thickness, 4 mm; no gap; echo train length, 16;bandwidth, 50 kHz; number of excitations, 4) were acquiredfor clinical grading of femoral and acetabular cartilage.Oblique T2-weighted fat-saturated fast spineecho images(TR/TE, 3678/60 msec; slice thickness, 3 mm; no gap;

PM&R Vol. 6, Iss. 8, 2014 683

matrix, 288 � 224; FOV, 18 cm) were acquired for themeasurement of the alpha angle (Figure 1).

Alpha Angle

The alpha angle was measured on the oblique MR studies asshown in Figure 1 [10]. Themeasurements were performed byan experienced radiologist (LN). The interclass coefficient forinterrater reproducibility of the alpha angle has been reportedto be 0.83 [26]. The interclass coefficient for intrarater repro-ducibility has been reported to be between 0.96-0.98 [26].

Semiquantitative Clinical CartilageGrading

Experienced board-certified musculoskeletal radiologists(TML and LN) performed the clinical grading for cartilagelesions on the coronal and sagittal MR studies [27]. The ra-diologists were blinded to the group assignment of the sub-jects. The femoral and acetabular segmentswere divided into 6subregions (4 femoral and 2 acetabular) on the coronal studiesand 4 subregions (2 femoral and 2 acetabular) on the sagittalstudies for a total of 10 subregions (Figure 2). Themid portionof the femoral head was defined on the sagittal images andsubdivided into 4 subregions on the coronal images, fromlateral tomedial (Figure 2A, B). The landmark for division wasthe lateral acetabular rim for the lateral and superolateralsubregions, a vertical line from the center of the femoral headfor the superolateral and supermedial subregions, and theligamentum teres for the supermedial and inferior subregions.

Figure 1. The alpha angle measurement shown on an obliqueaxial T2-weighted fast spineecho image. A ¼ anterior; R ¼ right;L ¼ left; P ¼ posterior.

On the sagittal MR study, the anterior subregion representedthe anterior 1 cm of the femoral head and the posterior sub-region represented the posterior 1 cm of the femoral head(Figure 2C, D). The division was based on a simplified versionof the geographic zonemethod described by Ilizaliturri Jr. et al[28] for hip arthroscopy, which showed superior interob-server reproducibility compared with the clock-face method.Cartilage defects were graded as 0 (no defect), 1 (partialthickness) and 2 (full thickness). Consensus readings wereperformed in case of a disagreement.

Intrarater and interrater reliability for the cartilage lesiongrading was established by 2 radiologists on a cohort of 30subjects not included in this study. Intrareader percent agree-ment was 85%, and the interrater percent agreement was 78%.

Motion Capture

Motion capture was always performed after the MRI onthe same day or on a different day within 2 weeks of the MRI.All subjects performed 3 tasks. During these tasks, 3-dimen-sional kinematic data (at 250 Hz) were collected from bilaterallower extremities using a passive 10-camera system (Vicon,Oxford Metrics, Oxford, UK). Kinetic data (at 1000 Hz) wereconcurrently collected from 2 embedded force platforms(AMTI, Watertown, MA). Tasks recorded included (1)walking at 1.3 m/sec; (2) a deep squat to 25% of the total bodyheight at a speed of 2 seconds for descent and 2 seconds forascent; and (3) drop landing from a 12-inch high platform,where the subjects were asked to step off and land synchro-nously with one foot on each force plate, and then jump ashigh as possible. During the deep squat, the subjects wereinstructed to stand with their feet parallel, facing anteriorlyand with their arms outstretched. However, the distance be-tween the feet was not controlled, and theywere not instructedtomaintain heel contact [21]. Five trials were acquired for eachtask. For the walking task, a trial was considered acceptablewhen there was a clean foot strike on either of the force plat-forms and the speed was within �5% of the first good trial.Fourteen-millimeter spherical retroreflective markers wereplaced on the bony landmarks of the bilateral lower extrem-ities for identification of joint centers. These anatomicmarkerswere placed on the sacrum and bilaterally on the iliac crest,anterior superior iliac spine, greater trochanter, medial andlateral femoral condyles, medial and lateral malleoli, and firstand fifth metatarsal head. Rigid marker clusters placed bilat-erally on the lateral surface of the subject’s thighs, legs, andheel shoe counters were used to track segment motions.

Motion Analysis Data Processing

Kinematic and kinetics were calculated using Visual3D (C-Motion, Georgetown, MD). All net joint moments are ex-pressed as internal moments and normalized to body weight(BW) and height (Ht) (Nm/BW*Ht). In the convention used,flexion, abduction, and internal rotation were positive. Peak

Figure 2. Subregions of articular cartilage for acetabulum on coronal (A) and sagittal (C) images and subregions for the femur oncoronal (B) and sagittal (D) images. (A) Acoronalmagnetic resonance (MR) imagedemonstrating acetabular superolateral (ASL) andsuperomedial (ASM) subregions divided by a vertical line extending from the femoral head center. (B) A coronal MR image demon-strating femoral lateral (FL), superolateral (FSL), superomedial (FSM), and inferior (FIM) subregions divided by a line extending from thefemoral head center to the lateral acetabular rim to straight vertical direction and to the ligamentum teres attachment. (C) A sagittalMR image demonstrates acetabular anterior (AA) and acetabular posterior (AP) subregions, demarcated by a vertical line 1 cm fromthe most anterior and posterior aspect of the femoral head. (D) A sagittal MR image demonstrates femoral anterior (FA) and posterior(FP) subregions, demarcated by a vertical line 1 cm from the most anterior and posterior aspect of the femoral head.

684 Kumar et al HIP CARTILAGE LESIONS, FAI, AND MOVEMENT PATTERNS

kinematic and kinetic variables, as well as joint excursions,were calculated for the hip joint during the stance phase of eachtask in sagittal, coronal, and transverse planes. The averageof 5 trials during each task was calculated for each subject.

Statistical Analysis

Analyses were performed between FAI subjects and controlsubjects and were stratified based on anatomic morphologicfeatures as described earlier. In addition, the subjects withFAI who had cartilage lesions (a grade >0 in any subregion)were compared with the remainder of the cohort withoutcartilage lesions (a grade of 0 in all subregions). One-way

Table 1. Subject age, body mass index, and gender distribution in pand subjects with and without cartilage lesions

Control subjects (n [ 8) FAI (n [ 7) P

Age (y)* 27.3 (7.7) 36.6 (9.7)BMI (kg/m2)* 23.9 (2.3) 26.2 (6.9)Male:female 8:0 5:2FAI:control subjects e e

FAI ¼ femoroacetabular impingement; BMI ¼ body mass index.*Mean (standard deviation).

analysis of variance tests were performed to evaluate thedifferences in peak hip joint moments, excursions, andpowers during the 3 tasks for both comparisons.

RESULTS

Subject Demographics

Group means for age, BMI, and gender distribution for sub-jects stratified by bony morphologic features (subjects withFAI and control subjects), as well as subjects with a cartilagelesion in the presence of FAI and without cartilage lesions, areshown in Table 1. Six of the FAI subjects had cartilage lesions

atients with femoroacetabular impingement, control subjects,

value No lesion (n [ 9) FAI D lesion (n [ 6) P value

.059 27.8 (7.4) 37.3 (10.4) .057

.400 24.2 (2.3) 26.1 (7.6) .473

.104 0:9 2:4 .063e 1:8 6:0 .001

PM&R Vol. 6, Iss. 8, 2014 685

and 9 subjects did not have any cartilage lesions (8 controlsubjects and 1 subject with FAI). None of these cases requiredconsensus readings. The differences in age, BMI, and genderdistribution between subjects with FAI and control subjectsand subjects with cartilage lesions in presence of FAI andwithout cartilage lesions were not statistically significant. Themean alpha angle for the FAI group was 74� (range, 58�-90�),and for the control group it was 42� (range, 33�-52�).

Walking

Average kinematic and kinetic patterns during walking forall groups are shown in Figure 3. Hip powers are shown inTable 2. No significant differences in kinematics or kineticswere seen when comparing subjects with and without FAI, orwhen subjects with cartilage lesions and FAI were comparedwith subjects who did not have cartilage lesions (Figure 3).

Deep Squat

Average kinematic andkinetic patterns duringdeep squats for allgroups are shown in Figure 4. Hip powers are shown in Table 2.

Figure 3. Average (standarddeviation) symptomatic hip kinematic (le(B,E), and transverse (C,F) planes during walking gait. Femoroacetabufirst column. FAIþCL (CL: black) and no cartilage lesion (No-CL: grey) a

Subjects with FAI performed the deep squat with greater hipadduction (P ¼ .005; Figure 4B) and higher peak internalrotationmoment (P¼ .008; Figure 4F)when comparedwith thecontrol group. When evaluating subjects stratified by cartilagelesions, the group with FAI and cartilage lesions also demon-strated greater peak hip adduction (P ¼ .038; Figure 4B) and ahigher peak internal rotationmoment (P¼ .015) comparedwithsubjects without cartilage lesions (Figure 4F). However, thisgroup also demonstrated less transverse plane range of motion(ROM; cartilage lesion group¼ 9.9� � 4.0�; no cartilage lesiongroup ¼ 17.8� � 4.2�; P ¼ .006).

Drop Landing

Average kinematic and kinetic patterns during drop landingfor all groups are shown in Figure 5. Hip powers are shownin Table 2. During the drop-landing task, subjects with FAIlanded with their feet closer together (FAI ¼ 0.30 � 0.05 m;control ¼ 0.37 � 0.04 m; P ¼ .029). No other kinematic orkinetic differences were noted when comparing the FAI andcontrol groups (Figure 5). In contrast, the group with FAIand a cartilage lesion landed with a smaller base of support

ft panel) and kinetic (right panel) patterns for sagittal (A,D), frontallar impingement (FAI: black) and control (grey) are shown in there in the second column. BW ¼ body weight; Ht ¼ height.

Table 2. Peak positive and negative hip powers at the symptomatic hip in watts for walking, deep-squat, and drop-landing tasks

Task Control subjects, M (SD) FAI, M (SD) P value No lesion, M (SD) FAI D lesion, M (SD) P value

WalkingPeak positive power 88.5 (21.2) 72. 9 (23.5) .199 88.9 (19.9) 69.6 (23.9) .112Peak negative power �33.1 (7.5) �34.4 (14.5) .824 �34.2 (7.8) �32.9 (15.3) .827

Deep squatPeak positive power 35.2 (11.7) 44.9 (27.3) .379 41.9 (22.9) 34.8 (12.9) .542Peak negative power �33.1 (12.7) �38.6 (22.5) .574 �38.1(19.0) �30.8 (13.2) .462

Drop jumpPeak positive power 554.1 (72.6) 546.0 (204.6) .919 582.1 (108.1) 493.9 (178.9) .267Peak negative power �879.3 (417.8) �665.1 (261.4) .293 �911.8 (402.8) �563.6 (91.0) .085

FAI ¼ femoroacetabular impingement; SD ¼ standard deviation.

686 Kumar et al HIP CARTILAGE LESIONS, FAI, AND MOVEMENT PATTERNS

(cartilage lesion group ¼ 0.29 � 0.03 m; control group ¼0.37 � 0.04 m; P ¼ .001) and exhibited greater peak hipadduction (P ¼ .031; Figure 5B) and less peak hip internalrotation (P ¼ .033) when compared with subjects withoutcartilage lesions (Figure 5C).

DISCUSSION

This preliminary exploratory study was aimed at investi-gating the influence of FAI and cartilage lesions on

Figure 4. Average (standard deviation) symptomatic hip kinematic (lfrontal (B,E), and transverse (C,F) planes during a deep squat. Femoroain the first column. FAIþCL (CL: black) and no cartilage lesion (No-CL:

kinematics and kinetics of the hip during various tasks. Weobserved differences in movement patterns between sub-jects who had FAI compared with control subjects. How-ever, the differences were more pronounced betweensubjects with FAI who had cartilage lesions compared withsubjects who did not have cartilage lesions. These findingshighlight the importance of understanding the complexinterplay between bony morphologic features, cartilage le-sions, and movement patterns in persons with cam-typeFAI.

eft panel) and kinetic (right panel) patterns for the sagittal (A,D),cetabular impingement (FAI: black) and control (grey) are showngrey) are in the second column. BW ¼ body weight; Ht ¼ height.

Figure 5. Average (standarddeviation) symptomatichipkinematic (leftpanel) and kinetic (rightpanel) patterns for sagittal (A,D), frontal(B,E), and transverse (C,F) planes during drop jump. Femoroacetabular impingement (FAI: black) andcontrol (grey) are shown in the firstcolumn. FAIþCL (CL: black) and no cartilage lesion (No-CL: grey) are in the second column. BW ¼ body weight; Ht ¼ height.

PM&R Vol. 6, Iss. 8, 2014 687

When comparing the movement patterns at the hip dur-ing walking between (1) subjects with and without cam-typeFAI and (2) between subjects with FAI and cartilage lesionsand subjects without cartilage lesions, we did not observe anystatistically significant differences. In an earlier study, Ken-nedy et al [20] showed lower sagittal ROM and hip abductionduring walking in persons with cam-type FAI comparedwith control subjects. Our sample size (n ¼ 7) was smallercompared with that of Kennedy et al [20] (n ¼ 17), and theyexclusively studied male subjects, with an alpha angle cutoffof 50.5� for diagnosis of FAI. Also, their finding of lower hipabduction occurred during the swing phase, which we didnot analyze because our aim was to understand the rela-tionship of hip loading during stance with cartilage damage.These differences could explain the disparity in the findings.Walking is the most common daily activity, and patientswith FAI often report pain after walking [29]. Future studieswould be needed to assess the long-term implications ofthese kinematic asymmetries.

Lamontagne et al [21] performed a study to quantify theeffect of cam-type FAI on movement patterns during a deep-

squat task. They did not find any significant differences in thehip motion between subjects with and without FAI but didreport lower squat depth for persons with FAI and differencesin pelvic motion. In our study, we did not control the widthbetween the feet at the start of the squat or require that the heelbe in contact with the ground during the squat. However, oursubjects were instructed to stand with the feet parallel andfacing anteriorly. Lamontagne et al [21] instructed their sub-jects to “stand with feet shoulder width apart, parallel to eachother, and facing anteriorly.” They also instructed the subjectstomaintain heel contact at all times.Wewanted the subjects toperform the squats as naturally as possible, and hence we usedthe base of support as one of the variables in our analyses. It ispossible that the differences in the starting hip position acrossthe subjectsmay have influenced our results. Lamontagne et al[21] did not control for squat depth, but in our study, wecontrolled for the squat depth and compared the differences inmovement patterns between groups. Using this strategy, wefound that persons with FAI and persons with FAI and carti-lage lesions maintained their hips in more adduction andgenerated greater peak hip internal rotation moments.

688 Kumar et al HIP CARTILAGE LESIONS, FAI, AND MOVEMENT PATTERNS

Reduced hip abduction during these activities may be amechanism to reduce the demands on the musculature. Anearlier study has shown that patients with FAI present withweakness of all hipmuscle groups [30]. Maintaining the hip inmore adduction is one strategy to reduce the demands on thehip musculature, as well as to increase the internal momentarm for the hip abductor muscles, potentially providing morestability. These results suggest that the presence of cartilagelesions with a cam lesion could be associated with furtherabnormalities in functional movement patterns. Longitudinalstudies would be needed to investigate the cause and effectrelationship between movement patterns and cartilage lesionsin persons with cam-type FAI.

No previous studies have compared the hip kinematicsand kinetics in persons with FAI and in control subjectsduring a drop-landing task. We chose to include a drop-landing task because it is a common athletic activity for thisage group (eg, in basketball and gymnastics) and entailsrapid hip loading with significant hip motion, which is verydifferent from the slow loading in the deep squat and thesmaller ROM during walking. Surprisingly, we did not findany statistically significant differences in hip kinematics andkinetics between subjects with and without FAI during thisactivity, except that patients with FAI landed with their feetcloser together, again presumably to lower the demands onthe hip musculature. In contrast, when we compared thesubjects who had FAI and cartilage lesions with subjects whodid not have cartilage lesions, we observed patterns similarto the other tasks, in that the subjects with FAI and lesionsmaintained a smaller base of support with greater hipadduction. We also found less hip internal rotation duringthis task in subjects who had cartilage lesions, suggestingthat subjects may have used a compensatory strategy toavoid anterior impingement. These findings also suggest thatthe presence of cartilage lesions in persons with cam-typeFAI may further alter the loading patterns at the hip, whichneeds to be confirmed in longitudinal studies.

It is important to note that only one subject with FAI did nothave any cartilage lesions. Hence the differences between thefindings when the FAI group was compared with controlsubjects versus when the FAI and cartilage lesion group wascompared with the no-cartilage lesion group are driven by thisone particular subject. This subject was a 32-year-oldmanwithan alpha angle of 58�. Hence this subject, although symp-tomatic, indeed had the lowest alpha angle among the FAIsubjects. It may be possible that larger magnitudes of camlesion are associated with cartilage damage. A cutoff of 55� isusually used to diagnose cam-type FAI, along with presence ofFAI-related symptoms [10].However, recent large-scale cohortstudies suggest that a radiographic alpha angle of 60� suggeststhe presence of a cam lesion and that a radiographic alpha angleof 78� indicates the presence of a pathologic cam lesion [31].

One of the main difficulties in evaluating hip cartilageon MR is the thinness of the articular cartilage. Our cartilagegrading is a simplified version of the Outerbridge

classification, which is a widely accepted cartilage lesionclassification originally based on knee cartilage arthroscopy,with elimination of grades that would not be reliablydiscernible on hip MRI [28]. The MR arthrogram has been themodality of choice for identification of cartilage lesions at thehip [32]. However, optimized noncontrast hip MRI has beenshown to have an excellent ability to identify cartilage lesionsat the hip [33]. The ability to visualize the cartilage in our studywas enhanced by the use of an optimized noncontrast hipMRIprotocol, using a small FOV on a 3-Tesla scanner.

The limitations of this work relate to the facts that thesample size was small, we did not adjust for age, BMI, andgender distribution, and the analyses were cross-sectional. Thesubjects in the control group and the no-cartilage lesion groupswere approximately 10 years younger than the subjects in theFAI and cartilage lesion group, respectively. Although thisdifference was not found to be statistically significant in ourstudy, the P values were close to significance. Because theincidence of cartilage lesions increases with age and risk of OA,our findings would need to be confirmed in a larger cohort andeither group matched for potential covariates or adjusted sta-tistically in the data analysis. Given the small sample of thecurrent investigation, we were unable to adjust for these vari-ables in our analyses, which may or may not confound theresults. Authors of future studies should also consider a lon-gitudinal design because our data suggest a complex interplaybetween bony and cartilage morphologic features in this pop-ulation, and understanding this relationship is not possiblegiven the cross-sectional nature of our study. Other hip ab-normalities, including labral tears, bone marrow edemaelikelesions, herniation pits, and tendinopathies, which all maybe related to hip movement patterns, were not evaluated inour study and should be considered in future investigations.Finally, our findings are limited to cam-type impingement.Subjectswith pincer or combination impingementmaypresentvery differently than our cohort, and caution is required whencomparing our findings with other impingement cohorts.

CONCLUSION

In conclusion, our data show that patients with FAI haveabnormal movement patterns during functional tasks whencompared with control subjects. The differences are amplifiedwhen comparing subjects with andwithout cartilage lesions inthe presence of a cam lesion. These findings highlight theimportance of understanding the complex interplay betweenbony morphologic features, cartilage lesions, and movementpatterns in persons with cam-type FAI.

ACKNOWLEDGMENTS

We thank Dr Anthony Luke for providing space andequipment for motion analysis and subject recruitment,Dr Thomas P. Vail for subject recruitment, and Ms ThelmaMunoz for assistance with subject scheduling.

PM&R Vol. 6, Iss. 8, 2014 689

REFERENCES1. Lawrence RC, Felson DT, Helmick CG, et al. Estimates of the preva-

lence of arthritis and other rheumatic conditions in the United States.Part II. Arthritis Rheum 2008;58(1):26-35.

2. Beck M, Kalhor M, Leunig M, Ganz R. Hip morphology influences thepattern of damage to the acetabular cartilage: Femoroacetabularimpingement as a cause of early osteoarthritis of the hip. J Bone JointSurg Br 2005;87(7):1012-1018.

3. Ganz R, Parvizi J, Beck M, Leunig M, Notzli H, Siebenrock KA. Fem-oroacetabular impingement: A cause for osteoarthritis of the hip. ClinOrthop Relat Res 2003;(417):112-120.

4. TanzerM,NoiseuxN.Osseous abnormalities and early osteoarthritis: Therole of hip impingement. Clin Orthop Relat Res 2004;(429):170-177.

5. Ito K, Minka MA 2nd, Leunig M, Werlen S, Ganz R. Femoroacetabularimpingement and the cam-effect. A MRI-based quantitative anatomicalstudy of the femoral head-neck offset. J Bone Joint Surg Br 2001;83(2):171-176.

6. Jung KA, Restrepo C, Hellman M, AbdelSalam H, Morrison W,Parvizi J. The prevalence of cam-type femoroacetabular deformity inasymptomatic adults. J Bone Joint Surg Br 2011;93(10):1303-1307.

7. Pulido L, Parvizi J. Femoroacetabular impingement. Semin Muscu-loskelet Radiol 2007;11(1):66-72.

8. Byrd JW, Jones KS. Arthroscopic femoroplasty in the management ofcam-type femoroacetabular impingement. Clin Orthop Relat Res 2009;467(3):739-746.

9. Myers SR, Eijer H, Ganz R. Anterior femoroacetabular impingementafter periacetabular osteotomy. Clin Orthop Relat Res 1999 Jun;(363):93-99.

10. Notzli HP, Wyss TF, Stoecklin CH, Schmid MR, Treiber K, Hodler J.The contour of the femoral head-neck junction as a predictor for therisk of anterior impingement. J Bone Joint Surg Br 2002;84(4):556-560.

11. Wyss TF, Clark JM, Weishaupt D, Notzli HP. Correlation betweeninternal rotation and bony anatomy in the hip. Clin Orthop Relat Res2007;460:152-158.

12. Clohisy JC, Baca G, Beaule PE, et al. Descriptive epidemiology offemoroacetabular impingement: A North American cohort of patientsundergoing surgery. Am J Sports Med 2013;41(6):1348-1356.

13. Clohisy JC, Knaus ER, Hunt DM, Lesher JM, Harris-Hayes M,Prather H. Clinical presentation of patients with symptomatic anteriorhip impingement. Clin Orthop Relat Res 2009;467(3):638-644.

14. Lodhia P, Slobogean GP, Noonan VK, Gilbart MK. Patient-reportedoutcome instruments for femoroacetabular impingement and hip labralpathology: A systematic review of the clinimetric evidence. Arthroscopy2011;27(2):279-286.

15. Register B, Pennock AT, Ho CP, Strickland CD, Lawand A, Philippon MJ.Prevalence of abnormal hip findings in asymptomatic participants: Aprospective, blinded study. Am J Sports Med 2012;40(12):2720-2724.

16. Hartofilakidis G, Bardakos NV, Babis GC, Georgiades G. An examinationof the association between different morphotypes of femoroacetabularimpingement in asymptomatic subjects and the development of osteo-arthritis of the hip. J Bone Joint Surg Br 2011;93(5):580-586.

17. Ganz R, Leunig M, Leunig-Ganz K, Harris WH. The etiology of oste-oarthritis of the hip: An integrated mechanical concept. Clin OrthopRelat Res 2008;466(2):264-272.

CME QuestionDuring a drop-landing maneuver, what kinematic/kinetic difference is seen

a. base of support distanceb. thigh adduction momentc. internal rotation momentd. transverse plane velocity

Answer online at me.aapmr.org

18. Allen D, Beaule PE, Ramadan O, Doucette S. Prevalence of associateddeformities and hip pain in patients with cam-type femoroacetabularimpingement. J Bone Joint Surg Br 2009;91(5):589-594.

19. Bardakos NV, Villar RN. Predictors of progression of osteoarthritis infemoroacetabular impingement: A radiological study with a minimumof ten years follow-up. J Bone Joint Surg Br 2009;91(2):162-169.

20. Kennedy MJ, Lamontagne M, Beaule PE. Femoroacetabular impinge-ment alters hip and pelvic biomechanics during gait: Walking biome-chanics of FAI. Gait Posture 2009;30(1):41-44.

21. Lamontagne M, Kennedy MJ, Beaule PE. The effect of cam FAI on hipand pelvic motion during maximum squat. Clin Orthop Relat Res2009;467(3):645-650.

22. Winter DA, Wells RP. Kinetic assessments of human gait. J Bone JointSurg Am 1981;63(8):1350.

23. Eng JJ, Winter DA. Kinetic analysis of the lower limbs during walking:What information can be gained from a three-dimensional model?J Biomech 1995;28(6):753-758.

24. Winter DA, Sidwall HG, Hobson DA. Measurement and reduction ofnoise in kinematics of locomotion. J Biomech 1974;7(2):157-159.

25. Tannast M, Siebenrock KA, Anderson SE. Femoroacetabular impinge-ment: Radiographic diagnosis—what the radiologist should know. AJRAm J Roentgenol 2007;188(6):1540-1552.

26. Mast NH, Impellizzeri F, Keller S, Leunig M. Reliability and agreementof measures used in radiographic evaluation of the adult hip. ClinOrthop Relat Res 2010;469(1):188-199.

27. Kumar D, Wyatt CR, Lee S, et al. Association of cartilage defects, andother MR findings with pain and function in individuals with mild-moderate radiographic hip osteoarthritis and controls. OsteoarthritisCartilage 2013;21(11):1685-1692.

28. Ilizaliturri VM Jr, Byrd JW, SampsonTG, et al. A geographic zonemethodto describe intra-articular pathology in hip arthroscopy: Cadaveric studyand preliminary report. Arthroscopy 2008;24(5):534-539.

29. Beaule PE, Le Duff MJ, Zaragoza E. Quality of life following femoralhead-neck osteochondroplasty for femoroacetabular impingement.J Bone Joint Surg Am 2007;89(4):773-779.

30. Casartelli NC, Maffiuletti NA, Item-Glatthorn JF, et al. Hip muscleweakness in patients with symptomatic femoroacetabular impinge-ment. Osteoarthritis Cartilage 2011;19(7):816-821.

31. Agricola R,Waarsing JH, ThomasGE, et al. Cam impingement: Defining thepresence of a cam deformity by the alpha angle: Data from the CHECKcohort and Chingford cohort. Osteoarthritis Cartilage 2014;22(2):218-225.

32. Schmid MR, Notzli HP, Zanetti M, Wyss TF, Hodler J. Cartilage lesionsin the hip: Diagnostic effectiveness of MR arthrography. Radiology2003;226(2):382-386.

33. Mintz DN,Hooper T, Connell D, Buly R, Padgett DE, PotterHG.Magneticresonance imaging of the hip: Detection of labral and chondral abnor-malities using noncontrast imaging. Arthroscopy 2005;21(4):385-393.

This journal-based CME activity is designated for 1.0 AMA PRA Category1 Credit� and can be completed online at www.me.aapmr.org. Thisactivity is FREE to AAPM&R members and available to non-members fora nominal fee. For assistance with claiming CME for this activity, pleasecontact (847) 737-6000.

between the control and femoroacetabular impingement groups?