imaging of talar dome chondral and osteochondral lesions of talar dome chondral... · abstract:...

11
Imaging of Talar Dome Chondral and Osteochondral Lesions James M. Linklater, MB, BS Abstract: Talar dome chondral and osteochondral lesions are a com- mon cause of ankle pain and subjective instability. The goal of imaging these lesions is primarily their detection, demonstration of their position and extent, including status of the chondral surface, demonstration of any associated chondral delamination, assessment of the integrity of the subchondral plate, and assessment of the cancellous subchondral bone for bone marrow edema like signal, sclerosis, cystic change, and for the presence of an unstable osteochondral fragment. Although plain radiography, computerized tomography, and bone scan may be helpful in the detection and characterization of these lesions, magnetic resonance imaging is the only imaging modality that will provide a comprehensive assessment of all these issues. Technical aspects of plain radiography, computerized tomography, and magnetic resonance imaging are discussed, and imaging findings are presented. Key Words: ankle, talar dome, chondral, osteochondral, MRI, CT, radiography (Top Magn Reson Imaging 2010;21: 3Y13) HISTORICAL PERSPECTIVE Chondral and osteochondral lesions in the ankle are a well- recognized cause of persistent ankle pain and subjective insta- bility after ankle injury. 1Y4 Clinically, there may be an overlap between a posttraumatic ankle joint synovitis, posttraumatic chondral and osteochondral injury, and extraarticular patholo- gies. Berndt and Harty 5 introduced a plain radiographic classi- fication of talar dome osteochondral lesions. In the 1970s and 1980s, bone scan became widely used in the workup of sus- pected talar dome lesions. In the late 1980s computerized tomographic assessment of talar dome lesion became wide- spread 6 and magnetic resonance imaging (MRI) began to be applied to the imaging of the ankle. The goal of imaging chondral and osteochondral lesions in the ankle is primarily their detection, demonstration of their position and extent, including status of the chondral surface, demonstration of any associated chondral delamination, assess- ment of the integrity of the subchondral plate, and assessment of the cancellous subchondral bone for bone marrow edema like signal, sclerosis, cystic change and for the presence of an unstable osteochondral fragment. Although plain radiography, computerized tomography (CT), and bone scan may be helpful in the detection and characterization of these lesions, MRI is the only imaging modality that will provide a comprehensive assessment of all these issues. PLAIN RADIOGRAPHY Technical Aspects Radiographs should be performed weight bearing in order to provide a more relevant assessment of alignment. In addition, joint space loss may be more conspicuous or even only be present on weight bearing. A routine ankle series should include antero-posterior (AP), lateral, and mortise views. Computerized radiography (CR) and digital radiography (DR) are gradually replacing film-screen radiography units. Computerized radiog- raphy and DR allow wider latitude in radiographic exposure, resulting in less repeat exposures and providing a better as- sessment of bone and soft tissues. Whereas previously it was common to see overexposed radiographs in which the soft tis- sues are ‘‘blacked out,’’ with CR and DR, this should no longer be the case. Computerized radiography and DR also offer improvements in patient throughput, not achievable with film- screen systems. The digital image data from CR and DR is also readily transferred to picture archive and communication system. Diagnostic Features Plain radiography is of limited sensitivity in the demon- stration of chondral and osteochondral pathology in the ankle. 7 In the setting of acute trauma, plain radiographs may demon- strate an acute osteochondral fracture (Fig. 1). In the chronic setting, central osteophyte formation may occur at the site of an overlying chondral defect (Fig. 2). Plain radiographs may also demonstrate the development of subchondral cystic change in a chronic osteochondral lesion (Fig. 3). In situ, unstable osteo- chondral fragments and displaced osteochondral loose bodies may also be demonstrable on plain radiography. Ankle joint osteoarthritis is an uncommon complication of an osteochondral lesion in the ankle. It is indicated at plain radiography by the presence of joint space loss and osteophyte formation. COMPUTERIZED TOMOGRAPHY Technical Aspects Modern multislice helical CT involves acquisition of a thin slice volume data set in the transverse plane that can be recon- structed in any plane and also reconstructed into 3-D images. Computerized tomography arthrography will provide detailed assessment of the chondral surfaces and may be used as an alternative means of assessment when MRI is contraindicated. 8 The disadvantage of CT arthrography compared with MRI lies in its invasive nature and its use of ionizing radiation. Com- puterized tomography single proton emission computerised tomography bone scans will demonstrate areas of increased osteoblastic activity associated with an osteochondral lesion. The greater anatomical detail provided by fusing the bone scan data with CT scan data allows more accurate localization of the ‘‘hot spot.’’ Diagnostic Features In the acute setting, CT can provide an assessment for fractures in cases where plain radiographs are equivocal or demonstrate a complex fracture. In the setting of chronic osteochondral lesions, CT will demonstrate the presence of ORIGINAL ARTICLE Top Magn Reson Imaging & Volume 21, Number 1, February 2010 www.topicsinmri.com 3 From the Castlereagh Sports Imaging, North Sydney Orthopaedic and Sports Medicine Centre, Crows Nest, New South Wales, Australia. Reprints: James M. Linklater, MB, BS, Castlereagh Sports Imaging, North Sydney Orthopaedic and Sports Medicine Centre, 286 Pacific Hwy, Crows Nest, New South Wales 2065, Australia (e-mail: linklj@ bigpond.com). This article was previously published: Linklater, JM. Imaging of talar dome chondral and osteochondral lesions. Tech Foot Ankle Surg. 2008;7:140Y151. Copyright * 2011 by Lippincott Williams & Wilkins Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Upload: dinhcong

Post on 08-Jun-2018

233 views

Category:

Documents


1 download

TRANSCRIPT

Imaging of Talar Dome Chondral and Osteochondral LesionsJames M. Linklater, MB, BS

Abstract: Talar dome chondral and osteochondral lesions are a com-mon cause of ankle pain and subjective instability. The goal of imagingthese lesions is primarily their detection, demonstration of their positionand extent, including status of the chondral surface, demonstration of anyassociated chondral delamination, assessment of the integrity of thesubchondral plate, and assessment of the cancellous subchondral bonefor bone marrow edema like signal, sclerosis, cystic change, and forthe presence of an unstable osteochondral fragment. Although plainradiography, computerized tomography, and bone scan may be helpful inthe detection and characterization of these lesions, magnetic resonanceimaging is the only imaging modality that will provide a comprehensiveassessment of all these issues. Technical aspects of plain radiography,computerized tomography, and magnetic resonance imaging arediscussed, and imaging findings are presented.

Key Words: ankle, talar dome, chondral, osteochondral, MRI,CT, radiography

(Top Magn Reson Imaging 2010;21: 3Y13)

HISTORICAL PERSPECTIVEChondral and osteochondral lesions in the ankle are a well-

recognized cause of persistent ankle pain and subjective insta-bility after ankle injury.1Y4 Clinically, there may be an overlapbetween a posttraumatic ankle joint synovitis, posttraumaticchondral and osteochondral injury, and extraarticular patholo-gies. Berndt and Harty5 introduced a plain radiographic classi-fication of talar dome osteochondral lesions. In the 1970s and1980s, bone scan became widely used in the workup of sus-pected talar dome lesions. In the late 1980s computerizedtomographic assessment of talar dome lesion became wide-spread6 and magnetic resonance imaging (MRI) began to beapplied to the imaging of the ankle.

The goal of imaging chondral and osteochondral lesions inthe ankle is primarily their detection, demonstration of theirposition and extent, including status of the chondral surface,demonstration of any associated chondral delamination, assess-ment of the integrity of the subchondral plate, and assessmentof the cancellous subchondral bone for bone marrow edemalike signal, sclerosis, cystic change and for the presence of anunstable osteochondral fragment. Although plain radiography,computerized tomography (CT), and bone scan may be helpfulin the detection and characterization of these lesions, MRI isthe only imaging modality that will provide a comprehensiveassessment of all these issues.

PLAIN RADIOGRAPHY

Technical AspectsRadiographs should be performed weight bearing in order

to provide a more relevant assessment of alignment. In addition,joint space loss may be more conspicuous or even only bepresent on weight bearing. A routine ankle series should includeantero-posterior (AP), lateral, and mortise views. Computerizedradiography (CR) and digital radiography (DR) are graduallyreplacing film-screen radiography units. Computerized radiog-raphy and DR allow wider latitude in radiographic exposure,resulting in less repeat exposures and providing a better as-sessment of bone and soft tissues. Whereas previously it wascommon to see overexposed radiographs in which the soft tis-sues are ‘‘blacked out,’’ with CR and DR, this should no longerbe the case. Computerized radiography and DR also offerimprovements in patient throughput, not achievable with film-screen systems. The digital image data from CR and DR is alsoreadily transferred to picture archive and communication system.

Diagnostic FeaturesPlain radiography is of limited sensitivity in the demon-

stration of chondral and osteochondral pathology in the ankle.7

In the setting of acute trauma, plain radiographs may demon-strate an acute osteochondral fracture (Fig. 1). In the chronicsetting, central osteophyte formation may occur at the site of anoverlying chondral defect (Fig. 2). Plain radiographs may alsodemonstrate the development of subchondral cystic change in achronic osteochondral lesion (Fig. 3). In situ, unstable osteo-chondral fragments and displaced osteochondral loose bodiesmay also be demonstrable on plain radiography. Ankle jointosteoarthritis is an uncommon complication of an osteochondrallesion in the ankle. It is indicated at plain radiography by thepresence of joint space loss and osteophyte formation.

COMPUTERIZED TOMOGRAPHY

Technical AspectsModern multislice helical CT involves acquisition of a thin

slice volume data set in the transverse plane that can be recon-structed in any plane and also reconstructed into 3-D images.Computerized tomography arthrography will provide detailedassessment of the chondral surfaces and may be used as analternative means of assessment when MRI is contraindicated.8

The disadvantage of CT arthrography compared with MRI liesin its invasive nature and its use of ionizing radiation. Com-puterized tomography single proton emission computerisedtomography bone scans will demonstrate areas of increasedosteoblastic activity associated with an osteochondral lesion.The greater anatomical detail provided by fusing the bone scandata with CT scan data allows more accurate localization ofthe ‘‘hot spot.’’

Diagnostic FeaturesIn the acute setting, CT can provide an assessment for

fractures in cases where plain radiographs are equivocal ordemonstrate a complex fracture. In the setting of chronicosteochondral lesions, CT will demonstrate the presence of

ORIGINAL ARTICLE

Top Magn Reson Imaging & Volume 21, Number 1, February 2010 www.topicsinmri.com 3

From the Castlereagh Sports Imaging, North Sydney Orthopaedic and SportsMedicine Centre, Crows Nest, New South Wales, Australia.

Reprints: James M. Linklater, MB, BS, Castlereagh Sports Imaging, NorthSydney Orthopaedic and Sports Medicine Centre, 286 Pacific Hwy,Crows Nest, New South Wales 2065, Australia (e-mail: [email protected]).

This article was previously published: Linklater, JM. Imaging of talardome chondral and osteochondral lesions. Tech Foot Ankle Surg.2008;7:140Y151.

Copyright * 2011 by Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

FIGURE 1. Undisplaced (A) and displaced (B) flake type lateral talar dome osteochondral fractures on AP radiographs. (Reprinted withpermission from Atlas of Imaging in Sports Medicine, McGraw Hill, Sydney, 2008).

FIGURE 2. A, Acute undisplaced lateral talar dome osteochondral fracture on coronal CT image. B, Subsequent development ofsubchondral cystic change and central osteophyte on x-ray 6 months later.

FIGURE 3. Normal articular cartilage. Cadaveric sagittal (A) and coronal (B) proton density fast spin echo magnetic resonancearthrographic images. Note low signal line at cartilage-cartilage interface (arrow).

Linklater Top Magn Reson Imaging & Volume 21, Number 1, February 2010

4 www.topicsinmri.com * 2011 Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

subchondral sclerosis and cystic change, central osteophytesand in situ and displaced ossific loose bodies. Computerizedtomography is superior to MRI in demonstrating small cracks inthe subchondral plate associated with cystic change which mayallow propulsion of synovial fluid into the cyst. Computerizedtomography will not demonstrate the chondral surfaces and willnot demonstrate bone marrow edema. Therefore, in general,MRI is preferred over CT in the assessment of talar domechondral and osteochondral lesions.

MAGNETIC RESONANCE IMAGINGMagnetic resonance imaging provides a noninvasive

assessment of the chondral surfaces, subchondral plate, sub-chondral bone, joint fluid, capsule, and ligaments. Magneticresonance imaging is at present the best single imagingmodality for assessment of ankle joint chondral and osteo-chondral lesions.

Technical Aspects

Field StrengthMagnetic resonance imaging units in clinical practice range

in field strength from 0.3 to 3.0 Tesla. In general, low fieldstrength MRI units (G0.5 Tesla) do not provide detailed articularcartilage assessment.

Pulse SequencesThe pulse sequences typically used in ankle MRI consist

of proton density (PD) and fat-suppressed proton densitysequences (PD-FS).9,10 The pulse sequences should have suffi-cient spatial and contrast resolution to provide detailed assess-ment of articular cartilage. Spatial resolution is a function ofthe slice thickness (usually 2.5Y4 mm), field of view (usually10Y16 cm) image matrix (the number of data points in the imagegrid, expressed numerically in 2 axes) typically between 512 �384 and 256 � 192.

The nonfat-suppressed PD sequence renders joint fluidbright and provides high-resolution images of articular cartilage,enabling assessment of the chondral surface, demonstratingchondral fissure/fracture and basal chondral delamination.9,11

Chondral fibrillation, commonly seen in patellar articular carti-lage, is uncommon in the ankle.12 Proton density sequencingalso provides some assessment of the subchondral bone, inthat low signal intensity change in the subchondral bone isindicative of subchondral sclerosis or fibrosis. Bone marrowedema is not demonstrable on a nonfat-suppressed PD sequence.

Fat suppression is achieved by the use of a radio frequencypulse that reduces the amount of signal derived from fat,rescaling the contrast characteristics of the pulse sequence towhich it is applied. If applied to a PD sequence, areas ofincreased water content within bone marrow will be of higher

FIGURE 4. Minimally displaced chondral flap midlateral talar dome, with a small crack in the chondral surface at the medial andposterior margins of the delaminated chondral fragment and mild subchondral bone marrow edema. Note loss of the normal lowsignal intensity in the deep layer of the delaminated chondral fragment on the coronal PD sequence. (A) Sagittal PD, (B) sagittalfat-suppressed PD, and (C) coronal PD MR images.

FIGURE 5. Undisplaced chondral flap postero-lateral talar dome, with surface crack at the lateral gutter articular facet (arrow) andmild subchondral bone marrow edema. Sagittal fat-suppressed PD (A) and coronal PD MR images (B).

Top Magn Reson Imaging & Volume 21, Number 1, February 2010 Imaging of Talar Dome Chondral and Osteochondral Lesions

* 2011 Lippincott Williams & Wilkins www.topicsinmri.com 5

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

signal (brighter) compared with reduced signal derived from fatwithin marrow. Fat-suppressed PD sequencing will demonstratethe chondral surfaces, the interface between articular cartilageand the subchondral plate and will also demonstrate subchondralbone marrow edema and cystic change. In the ankle, with someMRI systems, fat suppression can, on occasion, be somewhatinhomogeneous.When this occurs, it can be difficult to distin-guish between bone marrow edema and an area of artefactualhigh signal related to inhomogeneous fat suppression. This iscommonly seen in the medial malleolus.

Short tau inversion recovery (STIR) sequencing representsan alternative method of suppressing fat signal and demon-strating bone marrow edema. STIR sequencing does not sufferfrom the inhomogeneity issues that are potentially seen withfrequency selective fat suppression. The disadvantage of STIRsequencing is that it is intrinsically a lower signal to noisesequence compared with a fat suppression sequence, limitingthe spatial resolution that can be achieved and thus limitingassessment of articular cartilage.

Occasionally, immature subchondral cystic change maynot be frankly of fluid signal intensity and may be more con-spicuous on a T1-weighted sequence. The disadvantage of T1sequencing is that joint fluid is of similar signal intensity to

articular cartilage and thus the chondral surfaces are not welldemonstrated.12

Fat-suppressed T1 gradient echo sequencing renders artic-ular cartilage bright, joint fluid intermediate to low signal andmarrow dark. It is not commonly used in clinical imaging but itcan be used to calculate articular cartilage volumes and providesome assessment of the chondral surfaces.13Y15

Recent developments in MRI assessment of articular car-tilage include quantitative T2 mapping, T1Q mapping, diffusionimaging, and delayed gadolinium enhanced magnetic resonanceimaging of cartilage. These techniques can provide someassessment of water and proteoglycan content and integrity ofthe collagen matrix in articular cartilage before the onset ofmacrostructural changes.16Y19 These techniques remain largely aresearch tool not widely used in clinical practice.

Diagnostic Features

Normal Articular CartilageNormal articular cartilage in the ankle is of intermediate

signal intensity on PD sequencing, becoming mildly hyper-intense (bright) on fat-suppressed PD sequencing. The basallayer of articular cartilage is normally of low signal intensity on

FIGURE 6. Minimally displaced chondral flap postero-lateral talar dome (arrows), with crack in the chondral surface at the superiormargin of the lateral gutter articular facet (arrowhead). No subchondral bone marrow edema. Sagittal PD (A) and sagittalfat-suppressed PD MR images (B).

FIGURE 7. Displaced chondral fracture with resultant full thickness chondral defect superior margin of the medial gutterarticular facet of the talar dome (arrow), the displaced chondral fragment in the posterior recess (arrowhead). Coronal PD (A)and sagittal PD MR images (B).

Linklater Top Magn Reson Imaging & Volume 21, Number 1, February 2010

6 www.topicsinmri.com * 2011 Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

FIGURE 8. Chronic midlateral talar dome osteochondral lesion, with extensive, poorly demarcated area of softened cancelloussubchondral bone, small foci of cystic changewithmixed pattern of surrounding fibrotic, and bonemarrow edema like signal. Note overlyingfull thickness chondral flap and crack in the chondral surface (arrow). Sagittal PD (A) and sagittal fat-suppressed PD MR images (B).

FIGURE 9. Medial talar dome osteochondral lesion, with 2.7 cm fluid signal intensity subchondral cyst extending into the talar neckand overlying basal chondral delamination (arrow), and communication with small crack in the subchondral plate (arrowhead).Sagittal PD (A) and coronal PD MR images (B).

FIGURE 10. Chronic medial talar dome osteochondral lesion, with 1.3 cm subchondral cyst containing fibrotic debris. Mixed patternof surrounding fibrotic and bone marrow edema like signal. Small overlying crack in the chondral surface (arrowhead), leading into asubtle area of basal chondral delamination (arrows). Sagittal PD (A), sagittal fat-suppressed PD (B), and coronal PD MR images (C).

Top Magn Reson Imaging & Volume 21, Number 1, February 2010 Imaging of Talar Dome Chondral and Osteochondral Lesions

* 2011 Lippincott Williams & Wilkins www.topicsinmri.com 7

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

FIGURE 11. Chronic lateral talar dome osteochondral lesion, with subchondral cyst containing fibrotic debris and displacedosteochondral fragment (arrow). Note sclerosis deep to the cystic change, manifest as low signal intensity on PD sequencing, withmild surrounding bone marrow edema. Sagittal PD (A), sagittal fat-suppressed PD (B), and coronal PD MR images (C).

FIGURE 12. Lateral talar dome osteochondral lesion, with small subchondral cyst (black arrow), extensive surrounding bone marrowedema, small overlying full thickness chondral defect (white arrow) and undermining of the chondral margins, with resultant unstablechondral fragment. AP mortise radiograph (A), coronal PD (B), and sagittal fat-suppressed PD MR images (C).

FIGURE 13. In situ osteochondral lesion with confluent fluid signal intensity at the interface between the osteochondral fragment andthe base of the potential osteochondral crater. Mild depression in contour of the chondral surface at the central aspect of the lesionbut no discernible chondral surface defect at the lesion margins. The lesion is likely to be ballotable. Sagittal (A) and PD MR images (B).

Linklater Top Magn Reson Imaging & Volume 21, Number 1, February 2010

8 www.topicsinmri.com * 2011 Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

PD and fat-suppressed PD sequencing, related to the sub-chondral plate and zone of calcification. It is normal to see a lowsignal intensity line at the interface between 2 chondral surfacesthat are in contact with each other (Fig. 3). This should not bemistaken for a fibrous band or loose body.

Chondral Contusion, Fracture and ChondralDelamination/Flap

Experimental studies suggest that blunt trauma to articularcartilage produces profound changes in histologic, biochemical,and ultrastructural changes in the absence of surface disrup-tion.20,21 Contusional chondral injury is rarely demonstrablewith routine clinical MRI techniques. Newer MRI techniquesincluding delayed gadolinium enhanced magnetic resonanceimaging of cartilage, T2 mapping, T1Q, and diffusion imagingoffer some promise in demonstrating these changes.16Y19

Acute traumatic chondral fractures are relatively com-mon4 and can have a subtle appearance on MRI, often onlyappreciable with high-resolution imaging.11 They are typicallyseen as a linear focus of high signal on PD or fat-suppressedPD sequencing at the medial or lateral margin of the talardome on coronal sequencing, sagittal in orientation, breachingthe chondral surface.11 These may be associated basal chon-dral delamination, with resultant chondral flap.4 The basal de-lamination manifests on MRI as a loss of the normal low signalat the basal layer of articular cartilage, related to delamination atthe interface between the subchondral plate and the zone ofcalcification (Fig. 4). The degree of signal hyperintensity canvary from mildly hyperintense to frankly that of fluid. If there isno displacement, chondral delamination lesions can be quitesubtle. These lesions are rendered more conspicuous if there isan adjacent subchondral bone marrow edema (Fig. 5). Occa-sionally, a chondral lesion may be seen without an adjacentsubchondral bone marrow edema (Fig. 6). Displaced chon-dral fragments are usually seen in the anterior or posteriorrecesses (Fig. 7).

Acute Osteochondral InjuryAcute osteochondral injury in the ankle may range from

contusional injury to the articular cartilage and subchondralbone, to a transchondral, transosseous fracture. Berndt andHarty5 theorised that there are 4 stages to an acute impactionalinjury that produces a talar dome fracture.

& Stage 1: compression fracture of subchondral bone, withintact chondral surface and subchondral plate.

& Stage 2: partially detached osteochondral fragment, breechingthe articular cartilage and subchondral plate.

& Stage 3: completely detached in situ osteochondral fragment,with breech of the chondral surface and subchondral plate.

& Stage 4: detached migrated osteochondral fragment.

This staging system for the acute injury has been widelyadopted as a radiographic and arthroscopic classification. How-ever, its clinical use is limited as there is no demonstrated cor-relation with prognosis and it does not guide management. Inaddition, it does not take into account the status of the articularsurface or the presence of subchondral cystic change or bonemarrow edema. A number of different CT, MRI, and arthro-scopic staging systems have been advocated, however, theirclinical use and prognostic significance are not yet established,limiting their clinical application.6,7,11,22

FIGURE 14. Chronic medial talar dome osteochondral lesion, with defined osteochondral fragment demonstrating extensive bonemarrow edema. Note low signal intensity fibrotic signal at the interface between the fragment and the base of the potential osteochondralcrater (arrow) and intact overlying chondral surface. Sagittal fat-suppressed PD (A) and coronal PD MR images (B).

FIGURE 15. Large osteochondral lesion with unstableosteochondral fragment on a sagittal PD MR image. Note fluidsignal intensity cleft at the surface at the anterior margin of thelesion (white arrow) and fluid signal intensity at the interfacebetween the fragment and the base of the lesion (black arrow).

Top Magn Reson Imaging & Volume 21, Number 1, February 2010 Imaging of Talar Dome Chondral and Osteochondral Lesions

* 2011 Lippincott Williams & Wilkins www.topicsinmri.com 9

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Contusional injury to the cancellous subchondral bone ismanifested on MRI as a bone marrow edemalike signal on fat-suppressed PD or T2 sequencing and on STIR sequencing andwas seen in 17% to 27% of patients after ankle sprains in twoseries.23Y25 Bone marrow edema is usually evident within hoursof injury and will generally persist for several months. Micro-fracture, hemorrhage, and edema have been seen as histologic

correlates in areas of posttraumatic bone marrow edema in ani-mal studies.26,27 This differs from the histologic findings seenin areas of subchondral bone marrow edema in osteoarthritis,where variable findings of trabecular remodeling, marrownecrosis, edema, and fibrosis and rarely, hemorrhage may beseen.28 There are conflicting reports as to the prognosticsignificance of subchondral bone marrow edema after ankle

FIGURE 16. Chronic medial talar dome osteochondral lesion with in situ, unstable osteochondral fragment. Note crack in thechondral surface at the posterior margin of the lesion (arrow), fluid signal intensity at the interface between the fragment and theadjacent base of the potential crater, sclerotic margins at the interface and adjacent bone marrow edema. Sagittal (A) and coronalfat-suppressed PD MR images (B).

FIGURE 17. Large chronic medial talar dome osteochondral lesion, with in situ, unstable osteochondral lesion. Sagittal T1 (A),fat-suppressed PD (B), T2 (C) and coronal PD (D) images. Note low signal intensity sclerotic osseous component best appreciated onthe sagittal T1 and coronal PD images, cystic change and small crack in the chondral surface at the lateral margin on the coronal PDimage (arrow) (D).

Linklater Top Magn Reson Imaging & Volume 21, Number 1, February 2010

10 www.topicsinmri.com * 2011 Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

injury.23,29,30 Anecdotally, bone marrow edema after anklesprains is associated with a slower recovery.30 In the chronicsetting, bone marrow edema is often interpreted as being amarker for activity of the osteochondral lesion. Anecdotally,there would appear to be a correlation between pain and bonemarrow edema in talar dome lesions, however, there is littleevidence to substantiate this clinical impression.

Transchondral, transosseous fractures are usually appre-ciable on MRI, however, the osseous component of the injurymay be relatively inconspicuous on MRI, when compared witha CT scan. On PD sequencing, the fracture line is evident asfocal disruption of trabecular architecture, with linear signalhyperintensity. Bone marrow edema is typically present at thefracture margins.

Subchondral Cystic ChangeThe outcome after an acute talar dome osteochondral

impaction injury is variable. The chondral surface may remainintact and there may be resolution of the subchondral bonemarrow edema, without complication.31 However, the relativelypoor, retrograde blood supply of the talar dome predisposes toincomplete healing and the development of focal areas ofischaemic/devascularized subchondral bone in which theremay be resorption of subchondral trabeculae, replacement withfibrotic tissue, and subsequent formation of cystic change

(Fig. 8). In addition, if there is a crack in the chondral surface andsubchondral plate, synovial fluid may be forced into the sub-chondral bone by virtue of a ball valve type effect (Fig. 9). Cysticchange almost invariably develops at sites of pre-existingbone marrow edema signal.32 The cystic change associated withtalar dome osteochondral lesions may vary from a predominatelyfibrotic to predominantly fluid content12 (Figs. 9Y11). Cysticchange is usually readily appreciated on routine MRI sequenc-ing. Sometimes cystic change can be a little more conspicuouson T1 sequencing. Magnetic resonance imaging is of limited usein demonstrating small cracks in the subchondral plate asso-ciated with cystic change which may allow propulsion ofsynovial fluid into the cyst. If this is suspected and clinicallyimportant, then CT may be helpful. Subchondral cystic changemay become extremely large, presenting a challenge for surgicalmanagement (Fig. 9). In one series, the presence of small areasof cystic change was not associated with a worse outcomeafter arthroscopic debridement, when compared with lesionswithout cystic change.33 Progressive increase in cyst size wasassociated with poor clinical outcome in one series.34

Unstable Chronic Osteochondral Lesions andChronic Osteochondral Defects

The destabilization of a talar dome osteochondral lesionmay commence at the time of an acute osteochondral fracture or

FIGURE 18. Chronic lateral talar dome osteochondral lesion with mildly displaced osteochondral fragment (arrow) and unconstrainedlateral border of osteochondral crater. Coronal (A) and sagittal PD MR images (B).

FIGURE 19. Chronic unconstrained postero-medial osteochondral defect. Note deficient subchondral plate of the medial gutterarticular facet of the talar dome (arrow). Sagittal fat-suppressed PD (A) and coronal PD MR images (B).

Top Magn Reson Imaging & Volume 21, Number 1, February 2010 Imaging of Talar Dome Chondral and Osteochondral Lesions

* 2011 Lippincott Williams & Wilkins www.topicsinmri.com 11

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

develop as a complication of a chronic osteochondral lesion(Figs. 12, 15 and 18). The MRI criteria for assessment ofstability of an osteochondral lesion depend to some extent onthe criteria used for arthroscopic assessment. There may be de-stabilization of the osseous interface of an osteochondral lesionin the setting of an intact overlying chondral surface (Fig. 13).In this situation, the lesion may be ballotable at arthroscopy.Although this lesion may be interpreted at arthroscopy as beingunstable, this picture differs from that of a destabilized osteo-chondral fragment with breach of the subchondral plate and theoverlying articular cartilage. Magnetic resonance imaging hasbeen demonstrated to accurately predict the stability of osteo-chondral lesions.11,22 The principal criterion for stability fordiagnosis of instability of an osteochondral lesion is the presenceof a line of high signal on PD or fat-suppressed PD sequencingbreaching the chondral surface, subchondral plate and extendingalong the interface between the osteochondral fragment and thebase of the osteochondral lesion.11,22,35 It has been suggestedthat a high signal line may be present in the setting of reparativegranulation tissue at the interface between an osteochondrallesion and the base of the potential osteochondral crater andthat this is not necessarily a marker of instability if the overlyingarticular cartilage is intact.35 A low signal intensity line at theinterface of an osteochondral fragment is consistent with fibrotictissue (Fig. 14).

Displaced osteochondral fragments are usually readilyidentified on MRI. Fragments may be seen adjacent to theosteochondral defect, in the anterior or posterior recess, orwithin the syndesmotic recess (Fig. 16).

Large osteochondral defects of the talar dome are asso-ciated with significant changes in the contact characteristics ofthe ankle joint that may result in increased loading of articularcartilage at the margins of the defect (Fig. 17).36 Talardome osteochondral defects can be classified as constrained orunconstrained, depending on whether the defect extends tobreach the subchondral plate at either the medial or lateral gutterarticular facet of the talar dome (Figs. 18 and 19).

REFERENCES

1. Taga I, Shino K, Inoue M, et al. Articular cartilage lesions in ankleswith lateral ligament injury. An arthroscopic study. Am J Sports Med.1993;21:120Y126. [discussion 126Y127].

2. Hintermann B, Boss A, Schafer D. Arthroscopic findings in patientswith chronic ankle instability. Am J Sports Med. 2002;30:402Y409.

3. Lantz BA, McAndrew M, Scioli M, et al. The effect of concomitantchondral injuries accompanying operatively reduced malleolar fractures.J Orthop Trauma. 1991;5:125Y128.

4. Loren GJ, Ferkel RD. Arthroscopic assessment of occult intra-articularinjury in acute ankle fractures. Arthroscopy. 2002;18:412Y421.

5. Berndt AL, Harty M. Transchondral fractures (osteochondritisdissecans) of the talus. J Bone Joint Surg. 1959;41:988Y1020.

6. Anderson IF, Crichton KJ, Grattan-Smith T, et al. Osteochondralfractures of the dome of the talus. J Bone Joint Surg Am.1989;71:1143Y1152.

7. Loomer R, Fisher C, Lloyd-Smith R, et al. Osteochondral lesions ofthe talus. Am J Sports Med. 1993;21:13Y19.

8. Schmid MR, Pfirrmann CW, Hodler J, et al. Cartilage lesions in theankle joint: comparison of MR arthrography and CT arthrography.Skeletal Radiol. 2003;32:259Y265.

9. Potter HG, Linklater JM, Allen AA, et al. Magnetic resonanceimaging of articular cartilage in the knee. An evaluation with use offast-spin-echo imaging. J Bone Joint Surg Am. 1998;80-A:1276Y1284.

10. Bredella MA, Tirman PF, Peterfy CG, et al. Accuracy of T2-weightedfast spin-echo MR imaging with fat saturation in detecting cartilagedefects in the knee: comparison with arthroscopy in 130 patients.Am J Roentgenol. 1999;172:1073Y1080.

11. Mintz DN, Tashjian GS, Connell DA, et al. Osteochondral lesions ofthe talus: a new magnetic resonance grading system with arthroscopiccorrelation. Arthroscopy. 2003;19:353Y359.

12. Linklater J. Ligamentous, chondral, and osteochondral ankle injuriesin athletes. Semin Musculoskelet Radiol. 2004;8:81Y98.

13. Tan TC, Wilcox DM, Frank L. MR imaging of articular cartilage inthe ankle: comparison of available imaging sequences and methods ofmeasurement in cadavers. Skeletal Radiol. 1996;25:749Y755.

14. Disler DG, McCauley TR, Kelman CG, et al. Fatsuppressedthree-dimensional spoiled gradient-echo MR imaging of hyalinecartilage defects in the knee: comparison with standard MR imagingand arthroscopy. Am J Roentgenol. 1996;167:127Y132.

15. Eckstein F, Cicuttini F, Raynauld JP, et al. Magnetic resonance imaging(MRI) of articular cartilage in knee osteoarthritis (OA): morphologicalassessment. Osteoarthritis Cartilage. 2006;14(Suppl A):A46YA75.

16. Young AA, Stanwell P, Williams A, et al. Glycosaminoglycan contentof knee cartilage following posterior cruciate ligament rupturedemonstrated by delayed gadolinium-enhanced magnetic resonanceimaging of cartilage (dGEMRIC). A case report. J Bone Joint Surg Am.2005;87:2763Y2767.

17. Hannila I, Nieminen MT, Rauvala E, et al. Patellar cartilage lesions:comparison of magnetic resonance imaging and T2 relaxation-timemapping. Acta Radiol. 2007;48:444Y448.

18. Li X, Benjamin Ma C, Link TM, et al. In vivo T(1rho) and T(2) mappingof articular cartilage in osteoarthritis of the knee using 3 T MRI.Osteoarthritis Cartilage. 2007;15:789Y797.

19. Mamisch TC, Menzel MI, Welsch GH, et al. Steady-state diffusionimaging for MR in-vivo evaluation of reparative cartilage aftermatrix-associated autologous chondrocyte transplantation at 3tesla-Preliminary results. Eur J Radiol. 2007.

20. Thompson RC Jr, Oegema TR Jr, Lewis JL, et al. Osteoarthriticchanges after acute transarticular load. J Bone Joint Surg.1991;73A:990Y1001.

21. Donohoe JM, Buss D, Oegema TR Jr, et al. The effects of indirectblunt trauma on adult canine articular cartilage. J Bone Joint Surg.1983;65A:951Y957.

22. De Smet AA, Fisher DR, Burnstein MI, et al. Value of MR imaging instaging osteochondral lesions of the talus (osteochondritis dissecans):results in 14 patients. Am J Roentgenol. 1990;154:555Y558.

23. Alanen V, Taimela S, Kinnunen J, et al. Incidence and clinicalsignificance of bone bruises after supination injury of the ankle.A double-blind, prospective study. J Bone Joint Surg Br.1998;80:513Y515.

24. Pinar H, Akseki D, Kovanlikaya I, et al. Bone bruises detected bymagnetic resonance imaging following lateral ankle sprains. Knee SurgSports Traumatol Arthrosc. 1997;5:113Y117.

25. Sijbrandij ES, van Gils AP, Louwerens JW, et al. Posttraumaticsubchondral bone contusions and fractures of the talotibial joint:occurrence of ‘‘kissing’’ lesions. Am J Roentgenol. 2000;175:1707Y1710.

26. Rangger C, Kathrein A, Freund MC, et al. Bone bruise of the knee:histology and cryosections in 5 cases. Acta Orthop Scand.1998;69:291Y294.

27. Ryu KN, Jin W, Ko YT, et al. Bone bruises: MR characteristics andhistological correlation in the young pig.Clin Imaging. 2000;24:371Y380.

28. Zanetti M, Bruder E, Romero J, et al. Bone marrow edema pattern inosteoarthritic knees: correlation between MR imaging and histologicfindings. Radiology. 2000;215:835Y840.

Linklater Top Magn Reson Imaging & Volume 21, Number 1, February 2010

12 www.topicsinmri.com * 2011 Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

29. Zanetti M, De Simoni C, Wetz HH, et al. Magnetic resonanceimaging of injuries to the ankle joint: can it predict clinical outcome?Skeletal Radiol. 1997;26:82Y88.

30. Zanetti M, Steiner CL, Seifert B, et al. Clinical outcome of edema-likebone marrow abnormalities of the foot. Radiology. 2002;222:184Y188.

31. Elias I, Jung JW, Raikin SM, et al. Osteochondral lesions of the talus:change in MRI findings over time in talar lesions without operativeintervention and implications for staging systems. Foot Ankle Int.2006;27:157Y166.

32. Carrino JA, Blum J, Parellada JA, et al. MRI of bone marrowedema-like signal in the pathogenesis of subchondral cysts.Osteoarthritis Cartilage. 2006;14:1081Y1085.

33. Han SH, Lee JW, Lee DY, et al. Radiographic changes and clinicalresults of osteochondral defects of the talus with and withoutsubchondral cysts. Foot Ankle Int. 2006;27:1109Y1114.

34. Shearer C, Loomer R, Clement D. Nonoperatively managed stage5 osteochondral talar lesions. Foot Ankle Int. 2002;23:651Y654.

35. O’Connor MA, Palaniappan M, Khan N, et al. Osteochondritisdissecans of the knee in children. A comparison of MRI andarthroscopic findings. J Bone Joint Surg Br. 2002;84:258Y262.

36. Christensen JC, Driscoll HL, Tencer AF. 1994 William J. StickelGold Award. Contact characteristics of the ankle joint. Part 2. The effectsof talar dome cartilage defects. J Am Podiatr Med Assoc.1994;84:537Y547.

Top Magn Reson Imaging & Volume 21, Number 1, February 2010 Imaging of Talar Dome Chondral and Osteochondral Lesions

* 2011 Lippincott Williams & Wilkins www.topicsinmri.com 13

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.