0198-021 1/91/1105-0264$03.00/0 FOOT& ANKLE Copyright 0 1991 by the American Orthopaedic Foot and Ankle Society, InC
Magnetic Resonance Imaging of the Calcaneus: Normal Anatomy and Application in Calcaneal Fractures Jacob Zeiss, M.D.,* Nabil Ebraheim, M.D.,t Joe Rusin, M.D.,t and Robert J. CoombsS Toledo, Ohio
tissues and its inability to do direct multiplanar imaging. Although CT has proven capable of determining the location of tendon bundles" which are frequently displaced with calcaneal f r a c t ~ r e s ~ ,or ' ~ other , ' ~ types of injury,14 it has been limited in its ability to discriminate more subtle soft tissue structures. Multiplanar images can, of course, be obtained by CT with computer reconstruction, but these images lack the detail of those achieved by direct scanning. High resolution magnetic resonance imaging (MRI) of the musculoskeletal system, on the other hand, offers a dramatic advantage with respect to imaging soft tissues and imaging multiple planes directly. Its ability to discretely show tendons, ligaments, cartilage, nerves, and a variety of other soft tissues and joints in multiple projections has been well d e ~ c r i b e d as , ~ has its usefulness in the evaluation of bone marrow.'2 Some specific MRI applications to the foot and ankle have been addressed.',2,20-22 The most obvious shortcoming of this modality in the musculoskeletal system is its limited capability in imaging structures containing calcium. Because of the high incidence of injury to soft tissue structures with calcaneal fractures, particularly the peroneal tendons, one might wonder whether MRI can play a significant role in the evaluation of these injuries. The purpose of this paper, therefore, is to define the normal anatomy near the calcaneus and subtalar joint and to examine the capability of MRI in the evaluation of calcaneal fractures with respect to both bony and soft tissue structures as seen acutely and in follow-up.
ABSTRACT Images obtained from normal volunteers demonstrate highly detailed anatomy of the soft tissue and bony structures near the calcaneus and subtalar joint. Cortical bone, marrow, articular cartilage, ligaments, tendons, muscles, fibrous tissue, vascular bundles, and nerves can be identified. However, images obtained of acute calcaneal fractures were found to offer unsatisfactory depiction of bony anatomy. The presence of small fragments was obscured by a change in normal marrow signal by contusion, hemorrhage, and edema, and by the inability of magnetic resonance (MR) to image small pieces of cortical bone. Only in rare instances might MR be helpful in the acute setting when the location or displacement of tendons cannot be clearly ascertained with computed tomography (CT). MR may prove more useful in the long-term followup of healed fractures with persistent pain. In this setting it might be used in the diagnosis of complications such as residual or recurrent tendon displacement, tenosynovitis, heel fat integrity, and tarsal tunnel evaluation. However, this paper did not directly compare the efficacy of MR with that of CT in the long term. Therefore, the degree to which MR may eventually supplement or supplant CT is unclear and further study is required before the use of MR can be recommended in the routine clinical follow-up of calcaneal fractures.
Radiographic evaluation of the calcaneus consists primarily of plain films, plain film tomography, and computed t~mography.','~,'~ The latter has become almost indispensable in the evaluation of complex calcaneal f r a c t u r e ~ . ~ o,~l ~ Bony ~ * - ' anatomy can be exquisitely portrayed allowing accurate evaluation of fracture architecture. There are, however, two shortcomings to computed tomography (CT) capabilities in the musculoskeletal system in general: its inability to resolve fine soft
MATERIAL AND METHODS
Ten normal volunteers underwent calcaneal imaging with multiplanar MRI scanning. Five of the volunteer calcanei were imaged in sagittal, transverse, and coronal planes, while the other five were scanned in sagittal and oblique coronal projections. The latter was done at approximately 45' off the straight coronal plane. Three cadaver calcanei were imaged in sagittal, transverse,
From the Medical College of Ohio, Toledo, Ohio. * To whom all correspondence and reprint requests should be
addressed at: Medical College of Ohio, Department of Radiology, P.O. Box 10008, Toledo, Ohio 43699. t Department of Orthopedic Surgery. Department of Radiology.
*
264
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Foot & Ankle/Vol. 1 1, No. 5/April 1991
MRI OF THE CALCANEOUS
265
Fig. 1. Five sagittal T1 weighted MRI sections of a normal volunteer from lateral to medial hindfoot are presented. The most lateral section ( A ) shows the normal course and appearance to the peroneal tendons just behind the fibula. Image B demonstrates the lateral aspect of the posterior subtalar joint and the calcaneocuboid joint to good advantage and also shows the origin of the adductor digiti minimi muscle. Just lateral to midline (C) one sees the midportion of the posterior facet, the anterior facet (variable), the sinus tarsi and interosseous ligament, calcaneocuboid ligament, and talonavicular joint. Near midline and just medial to it (D), one can identify the medial aspect of posterior facet, the middle facet, the calcaneonavicular or spring ligament, as well as the quadratus plantae and flexor digitorum brevis muscles and Achille's tendon insertion. The most medial section ( E ) demonstrates the sustentaculum tali and middle facet as well as the flexor hallucis longus tendon. ADM = adductor digiti minimi muscle, AF = anterior facet, AT = Achilles tendon, C = calcaneous, CCJ = calcaneo-cuboidjoint, CCL = calcaneocuboid ligament, Cu = cuboid, F = fibula, FDB = flexor digitorum brevis muscle, FHL = flexor hallucis longus tendon, IL = interosseous ligament, MF = middle facet, N = navicular. PTB = peroneal tendon bundle, PF = posterior facet, QP = quadratus plantae muscle, SL = Spring ligament, ST = sustentaculum tali, SiT = sinus tarsi, Ta = talus, Ti = tibia.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Fig. 2. Three oblique-coronal T1 weighted MRI sections through a normal calcaneous are presented from posterior to anterior. A, is taken throiJgh the posterior facet, 8, through the middle facet, and C, through the anterior facet. All of the sections demonstrate medial and lateral
266
Downloaded from fai.sagepub.com at University College London on June 5, 2016
MRI OF THE CALCANEOUS
Foot & Ankle/Vol. 11, No. S/April 1991
267
Fig. 3. A, Normal axial T1 weighted MRI section through the peroneal tunnel at the level of talar dome and lateral malleolus shows intact superior peroneal retinaculum confining the peroneal tendons AT = Achilles tendon, C = calcaneous, F = fibula, PB = peroneal brevis tendon, PL = peroneal longus tendon, SPR = superior peroneal retinaculum, Ta = talus B, Normal oblique coronal T1 weighted MRI cut through the middle of tarsal tunnel shows this structure as defined by the medial malleolus, calcaneous, and flexor retinaculum The posterior tibial nerve within the tunnel can be discretely identified C = calcaneous, FHL = flexor hallucis longus tendon, FR = flexor retinaculum. PTN = posterior tibial nerve, Ta = talus, VB = vascular bundle
and coronal planes, then sectioned with a bandsaw and correlated to the MRI images. The three cadaver and 10 volunteer calcanei in conjunction with standard anatomical references were used to establish normal baselines. Sixteen intra-articular calcaneal fractures from 12 individuals were divided into three groups for MRI evaluation. Eight fractures were imaged within 48 hr of injury, four were imaged at approximately 3 to 4 months, and four were scanned at 2 years postinjury. The acute group all presented with a history of falling onto their heels with consequent pain and swelling. Fractures were first diagnosed by plain film and intraarticular fractures were selected for subsequent CT
and MRI evaluation. The subacute group was defined as healing intra-articular fractures at 3 to 4 months with varying degrees of residual discomfort, but no specific complaints. The healed calcaneal fractures at 2 years presented with a variety of complaints. One was asymptomatic, one had persistent lateral ankle pain with focal swelling, one complained of ankle instability with repeated inversion injuries, and one had an unsatisfactory response to tarsal tunnel release. The eight acute cases were evaluated with plain films, coronal CT with reconstruction, and MRI. The two follow-up groups had only plain films and MRI. MRI scans were done on a 1.5 Tesla General Electric Signa Scanner with a GE (General Electric, Milwaukee,
tendons, appropriate muscle groups and the medial and lateral plantar neurovascular bundles. ADM = adductor digiti minimi muscle, AH = adductor hallucis muscle, AF = anterior facet, C = calcaneous, F = fibula, FDB = flexor digitorum brevis muscle, FDL = flexor digitorum longus tendon, FHL = flexor hallucis longus tendon, LPNB = lateral plantar neurovascular bundle, MPNB = medial plantar neurovascular bundle, MF = middle facet, PA = plantar aponeurosis, PB = peroneal brevis tendon, PF = posterior facet, PL = peroneal longus tendon, QP = quadratus plantae muscle, SL = Spring ligament, ST = sustentaculumtali, Ta = talus, Ti = tibia, TP = tibialis posterior tendon.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
268
Foot & Ankle/Vol. 11, No. 5/April 1991
ZEISS ET AL.
Fig. 4. Corresponding coronal CT ( A ) and T1 weighted MRI (13)images in same individual with a comminuted calcaneal fracture are shown Contrast the sharply defined bony anatomy of the CT image to the poor bony detail of the MRI image The injured cancellous bone is all but obscured by low signal infiltration of hemorrhage and edema The displaced peroneal tendons can be identified in either image C = calcaneous, F = fibula, PB = peroneal brevis tendon, PL = peroneal longus tendon, PF = posterior facet, Ta = talus, Ti = tibia
WI) extremity coil. Each calcaneus was done with T1 and T2 weighting. The T1 scans were done in sagittal, axial, and coronal planes in half of the calcanei. Only sagittal and oblique coronal T1 images were done in the other half. The oblique coronals were done at approximately 45' to the straight coronal or straight axial plane. All of the calcanei were done with T2 weighting in one plane only, usually the oblique coronal. Minimum scan boundaries consisted of the anatomical limits of the calcaneous posteriorly and inferiorly, the midcuboid anteriorly, and the lateral malleolus superiorly. T1 technique was done with repetition time (TR) of 600 msec and echo time (TE) of 20 msec. T2 technique was done with TR of 2000 msec and TE of 80 msec. Slice thickness was 5 mm and no interslice gap was used. A 12-cm or 16-cm field of view was employed with a 256 x 128 matrix and two excitations. Scanning time for four T1 scans totalled approximately 30 min and for the T2 scan approximately 20 min. RESULTS
The MRI images of both the cadaver and volunteer calcanei showed excellent delineation of soft tissue and bony anatomy in all planes imaged. Each view demon-
strated particular strengths depending on the area of interest. For instance, sagittal views can nicely show all three facets of the subtalar joint, calcaneal height, and the calcaneocuboid joint. Transverse cuts are particularly useful in showing tendon position, tarsal and peroneal tunnel anatomy, and calcaneal widening. Coronal sections can demonstrate tendon position, calcaneal height and width, and subtalar joint, especially anterior facet. Oblique coronal views offer cross sections of the subtalar joint, especially posterior and middle facets, and combine many of the advantages of coronal and transverse cuts in a single scan. Figures 1 and 2 present images of the sagittal and oblique coronal planes at different levels through the volunteer calcanei. Figure 3A shows a transverse section through the lateral malleolus which demonstrates the peroneal tunnel, while Figure 38 is an oblique coronal image which demonstrates the tarsal tunnel. CT evalution of the eight acute calcaneal fractures showed them all to be comminuted fractures extending into the posterior subtalar joint, usually laterally. The number of fragments varied, but usually two large fragments, medial and lateral, were produced with varying degrees of additional comminution. The typical frac-
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Foot & Ankle/Vol. 11, No. 5/April 199 1
MRI OF THE CALCANEOUS
Fig. 5. T1 weighted axial MRI cut obtained through the level of the peroneal tunnel in a patient with acute calcaneal fracture. Note the disruption of the superior peroneal retinaculum and dislocation of the peroneal tendons from their usual position in the fibular groove. Compare this image with Figure 3A showing the normal appearance of retinaculum and peroneal tendons. F = fibula, FG = fibular groove, PTB = peroneal tendon bundle, SPR = superior peroneal retinaculum, Ta = talus.
ture consisted of a nondisplaced anteromedial sustentacular fragment, a larger posterolateral tuberosity fragment, axial widening secondary to lateral shift or spread of the tuberosity fragment, and craniocaudal flattening. CT uniformly presented a clear-cut demonstration of fragment type, number and position, and was of significant value in the proper classification of these fractures.” Peroneal tendon dislocation with displacement from behind the lateral malleolus was definite by CT in two of the eight fractures, very probable in one, but indefinite in another. The remaining four showed no dislocation, but varying degrees of lateral bowing of the peroneal tendons were seen in virtually every case because of the lateral spread of the tuberosity fragment. None of the fractures in this study showed significant distortion of the medial tendons. Upon viewing the MRI images of the acute calcaneal fractures it became immediately apparent that there
269
was a striking difference between the images of the normal volunteers and the injured patients. The normal high signal bone marrow within the general vicinity of the injured bone was replaced with a diffuse infiltrating signal of low intensity on the T I scans which blended almost imperceptibly with the injured adjacent soft tissues. On the T2 scans these abnormal marrow areas showed a heterogeneous mixture of predominantly low signal with some focal areas of high signal. Presumably this represented areas of microfracture with bone contusion consisting of hemorrhage, inflammation, and edema. Its effect was to negate the normal high marrow signal by which medullary bone is characterized on MRI, thereby making the margins of cancellous bone and small fragments of bone indistinguishable from adjacent hematoma or soft tissues. Cortical bone, because of its essentially unchanging low signal on both T1 and T2 scans, remained identifiable.Often, however, the cortical fragments were too small or too thin, or too close to a structure of similar intensity (e.g., tendon) to be recognizable. MRI was unable to even approach the sharp bone resolution in the face of acute trauma that one has come to expect from CT. The more extensively comminuted fractures demonstrated the greatest degree of this low T1 and mixed T2 signal infiltration and thereby the greatest lack of bony discrimination. Therefore, while larger fragments as well as the general fracture architecture and facet anatomy could be determined, MRI was of no significant help in fracture classification. Figure 4 demonstrates corresponding CT and MRI cuts from an acutely fractured calcaneous, and clearly points out the superiority of CT in this setting. MRI soft tissue resolution in the acute phase also suffered from the infiltrative effect of hemorrhage and edema when compared to the normal volunteer scans. The effect, however, was not nearly as detrimental to soft tissue discrimination as it was to bone. Soft tissue detail remained well above that of CT and the MRI images easily confirmed the tendon abnormalities found by CT. In the one patient where the peroneal tendons could not be definitively localized by CT because of diffuse soft tissue density around the region of the tendon bundle, MRI was able to demonstrate the tendon group with slight lateral subluxation and with accompanying rupture of the superior peroneal retinaculum. This finding was confirmed at surgery and repaired. Figure 5 demonstrates the typical MRI appearance of peroneal tendon dislocation. MRI evaluation of the subacute calcaneal fractures was compromised to a lesser extent than in the acute setting. Bone detail was significantly improved and one could see very low signal bands forming between the major fragments to indicate healing. The improved vis-
Downloaded from fai.sagepub.com at University College London on June 5, 2016
270
Foot & Ankle/Vol. 1 1, No. S/April 1991
ZEISS ET AL.
Fig. 6. Sagittal T1 weighted MRI images are shown at time of injury ( A ) and at approximately 3 months after injury ( B ) . Note the increasing bony detail as the fracture heals and the hemorrhage and edema resorb from the injured cancellous bone. AT = Achilles tendon, C = calcaneous, Cu = cuboid, Fx = fracture, Ta = talus.
ualization can probably be attributed to the diminishing marrow hemorrhage and edema secondary to healing. Figure 6 presents an example of a calcaneous 4 months after fracture and compares it to the initial MRI obtained at the time of injury. MRI soft tissue resolution in the subacute phase improved slightly from the acute phase as one might expect, and almost rivaled that in normal volunteers. The MRI calcaneal images obtained at 2 years postinjury demonstrated an appearance of bone and soft tissue structures essentially the same as that of the normal volunteers except for the varying degrees of residual deformity in the injured group. Marrow signal intensity was normal, cortical margins were well-defined, and soft tissue structures were sharply delineated. Bony alignment and facet anatomy could be easily evaluated. In this healed fracture group MRI evaluation did present some rather unique advantages. One patient was imaged after ineffectual release of his flexor retinaculum for tarsal tunnel syndrome secondary to old calcaneal fracture. The T1 examination revealed the caudal portion of the retinaculum to be intact. Reoperation resulted in successful release of the retinaculum and relief of the patient’s symptoms. Figure 7 shows two levels through this individual’s calcaneus revealing the retinaculum prior to reoperation to be released only proximally and intact distally. A second patient was imaged with persistent ankle instability and a tendency for inversion. MRI with T1
weighting demonstrated a disrupted superior peroneal retinaculum and a minimally subluxed peroneal tendon bundle, missed at the time of original treatment. This was confirmed at surgery and repaired (Fig. 8). A third patient had complained of persistent lateral ankle pain and focal swelling about the lateral malleolus. T2 weighted MRI images demonstrated findings consistent with tenosynovitis of a dislocated peroneal tendon bundle. The tendon bundle was not positioned in its usual location behind the fibula, but was lateral to it. The signal intensity of the tendon, normally homogeneously low or dark, was heterogeneous and abnormally high to indicate edema within it. It was also surrounded by a homogeneously high intensity periphery to indicate surrounding synovial fluid. DISCUSSION
This study indicates that MRI of the acutely fractured calcaneous can give a general impression of fracture architecture and a reasonable assessment of subtalar joint involvement. It is incapable, however, of adequately defining the existence and location of small bone fragments. The reason for this is the infiltration of hemorrhage and inflammation into the injured cancellous bone due to microfractures and contusion which alters the typical high marrow signal and makes it indistinguishable from surrounding soft tissue changes, both on T1 and T2 images. Small segments of cortical bone and the trabeculae of cancellous bone are insufficiently distinct to be discriminated from adjacent he-
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Foot & Ankle/Vol. 11, No. 5/April 1991
MRI OF THE CALCANEOUS
271
Fig. 7. Oblique-coronal T1 weighted MRI images are presented through posterocephalad( A ) and anterocaudal(8) levels of the tarsal tunnel in a patient who failed to respond to flexor retinaculum release for tarsal tunnel syndrome following healing of calcaneal fracture. The cephalic part of the flexor retinaculum was released as indicated by the arrow in A , but the caudal portion of the retinaculum was inadvertently left intact as seen in 8.Note the healed though broadened calcaneous. Symptoms were relieved following reoperation and complete release. C = calcaneous, FR = flexor retinaculum, MF = middle facet, PF = posterior facet, TA = talus.
matoma or other structures of similar intensity such as tendon bundles. Those fractures with the greatest degree of comminution and hemorrhage are most compromised in MRI evaluation, but all are affected to a significant degree. CT, therefore, remains the obvious choice in bony evaluation of calcaneal fractures. While MRI enhances ability to discriminate soft tissue structures, does it have a role in the evaluation of acute calcaneal injuries? In most cases, CT has sufficient soft tissue resolution to determine impingement on or displacement of peroneal or flexor hallucis tendons. Occasionally, however, the desired soft tissues structures cannot be adequately discriminated by CT.6One might also encounter an instance of superior peroneal retinaculum tear with very little or no displacement of the peroneal tendons, similar to that seen in Figure 8.22In such rare instances MRI can be of help, but this would certainly not warrant routine use of MRI in a fracture setting.
In the subacute phase, with resolution of the adjacent hematoma, MRI visualization improves, but is still suboptimal from a bone evaluation standpoint. The use of MRI in this timeframe would depend on the major clinical concern. CT indications would include determination of healing and residual bony deformity, while MRI might be used for residual tendon displacement, avascular necrosis, persistent localized edema, and osteomyelitis. Unlike the limitations encountered in MRI imaging of acute and subacute calcaneal fractures, healed fractures can be imaged with essentially the same sharpness of bony and soft tissue detail as healthy volunteers. At this stage soft tissue and cancellous bone hemorrhage and inflammation have resolved and the normal high marrow signal is re-established. Cortical edges and facets are sharply outlined. Therefore, the most helpful contribution of MRI evaluation of calcaneal injury may be found in long-term evaluation. T1 weighted scans are ideal for anatomical depiction, while
Downloaded from fai.sagepub.com at University College London on June 5, 2016
272
Foot & Ankle/Vol. 11, No. S/April 1991
ZEISS ET AL.
of calcaneal fractures is undefined. Since MRI is presently more costly, more time-consuming, and less available than CT, further experimental study is required before its routine clinical use can be recommended. CONCLUSIONS
The usefulness of MRI evaluation of calcaneal fractures in acute and subacute evaluation will most likely be very limited to occasional instances where CT does not clearly show soft tissue anatomy such as tendon displacement, or where complicating avascular necrosis or osteomyelitis is of concern. It could be more helpful in evaluation of persistent pain complicating healed fractures. As it completes directly with CT, additional investigationwill be necessary to more clearly define those instances where it may be of particular value.
REFERENCES
Fig. 8. Transverse Tl weighted MRI through the level of the peroneal tunnel in a patient with persistent pain along the lateral ankle and inversion instability 2 years following calcaneal fracture. The superior peroneal retinaculum is avulsed from its fibular attachment but the peroneal tendons are only slightly subluxed from their normal location. F = fibula, PTB = peroneal tendon bundle, SPR = superior peroneal retinaculum, Ta = talus.
T2 weighting is useful with infectious or inflammatory complications. Common causes of persistent discomfort in healed calcaneal fractures include heel pad pain, fibulocalcaneal impingement, scarring, and posttraumatic arthritis of the subtalar joint, peroneal tendonitis due to compression and displacement, plantar calcaneal spurs due to malunion and associated with disruption of the heel fat, arthritis of the calcaneocuboid, talonavicular and ankle joints, and nerve entrapment of medial and lateral plantar branche~.~.” These entities have been evaluated with a variety of diagnostic modalities including physical examination, plain radiography, CT, and nerve conduction studies. While each of these has a place in any given situation, the abilities of MRI should be particularly useful in evaluating articular cartilage and joint anatomy, residual tendon displacement, tenosynovitis, and nerve entrapment. A specific comparison between MRI and CT in the evaluation of chronic calcaneal injuries was not made in this study, and the exact role of MRI in the follow-up
1 . Beltran, J., Noto, A.M., Mosure, J.C., Shamam, O.M., Weiss, K.L., and Zuelzer, W.A.: Ankle: surface coil MR imaging at 1.5T. Radiology, 161:203-209, 1986. 2. Crim, J.R., Cracchiolo, A., Bassett, L.W., Seeger, L.L., Soma, C.A., and Chatelaine, A.: Magnetic resonance imaging of the hindfoot. Foot Ankle, lO(1):l-7, 1989. 3. Deyerle, W.M.: Long term follow-up of fractures of the 0s calcis. Orthop. Clin. North Am., 4(1):213-227. 1973. 4. Ehman, R.L., Berquist, T.N., and McLeod, H.A.: MR imaging of the musculoskeletal system: a 5-year appraisal. Radiology, 166~313-320,1988, 5. Gilmer, P.W., Herzenberg, J., Frank, J.L., Silverman, P., Martinez, s., and Goldner, J.L.: Computerized tomographic analysis of acute calcaneal fractures. Foot Ankle, 6(4):184-193, 1986. 6. Guyer, B.H., Levinsohn, M., Fredrickson, B.E., Bailey, G.L., and Formikell, M.: Computed tomography of calcaneal fractures: anatomy, pathology, dosimetry and clinical relevance. Am. J. Roentgen., 145911-919, 1985. 7. Heger, L., and Wulff, K.: Computed tomography of the calcaneus: normal anatomy. Am. J. Roentgen., 145:123-129, 1985. 8. Heger, L., Wulff, K., and Seddiqi, M.S.A.: Computed tomography of calcaneal fractures. Am. J. Roentgen., 145:131-137, 1985. 9. Hindman, B.W., Ross, S.D.K., and Sowerby, M.R.R.: Fractures of the talus and calcaneus: evaluation by computed tomography. J. Comput. Tomogr., 10(2):191-196, 1986. 10 Pablot, S.M., Daneman, A., Stringer, D.A., and Carroll, N.: The value of computed tomography in the early assessment of comminuted fractures of the calcaneus: a review of three patients. J. Pediatr. Orthop., (5)4:435-438, 1985. 1 1 Paley, D., and Hall, H.: Calcaneal fracture controversies. Orthop. Clin. North Am., 20(4):665-677, 1989. 12 Porter, B.A.: MR may become routine for imaging bone marrow. Diagn. Imaging, 104-1 08, 1987. 13 Romash, M.M.: Calcaneal fractures: three-dimensional treatment. Foot Ankle, 8(4):180-213, 1988. 14 Rosenberg, Z.S., Feldman, F., and Singson, R.D.: Peroneal tendon injuries: CT analysis. Radiology, 161:743-748, 1986.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
MRI OF THE CALCANEOUS
Foot 8, Ankle/Vol. 1 1, No. S/April 1991 15. Rosenberg, Z.S., Feldman, F.,and Singson, R.D.:Intra-articular calcaneal fractures: computed tomographic analysis. Skeletal Radiol., 16:105-113, 1987. berg, z,s., Feldman, F., and singson, R.D.: peroneal 16, R tendon injury associated with calcaneal fractures: CT findings. Am. J. Roentgen., 149:125-129, 1987. 17. Rosenberg, Z.S., Feldman, F., Singson, R.D., and Kane, R.: Ankle tendons: evaluation with CT. Radiology, 166:221-226, 1988. 18. Sartoris, D.J., and Resnick, D.: Diagnostic imaging approach to calcaneal fractures. J. Foot Surg., 26(6):524-529, 1987.
273
19. Solomon, M.A., Gilula, L.A., Oloff, L.M., and Oloff, J.: CT scanning of the foot and ankle: 2. Clinical applications and review of the literature. Am. J. Roentgen., 146:1204-1214, 1986. 20. Solomon, M.A., and Oloff-Solomon, J.: Magnetic resonance imaging in the foot and ankle. Clin. Podiatr. Med. Surg., 5(4):945965, 1988. 21. Zeiss, J., Fenton, P., Ebraheim, N., and Coombs, R.: Normal magnetic resonance anatomy of the tarsal tunnel. Foot Ankle, 10(4):214-218, 1990. 22. Zeiss, J., Saddemi, S.R., and Ebraheim, N.A.: MR imaging of the peroneal tunnel. J. Comput. Assist. Tomogr., 13(5):840844, 1989.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Magnetic Resonance Imaging of the Calcaneus: Normal Anatomy and Application in Calcaneal Fractures Jacob Zeiss, M.D.,* Nabil Ebraheim, M.D.,t Joe Rusin, M.D.,t and Robert J. CoombsS Toledo, Ohio
tissues and its inability to do direct multiplanar imaging. Although CT has proven capable of determining the location of tendon bundles" which are frequently displaced with calcaneal f r a c t ~ r e s ~ ,or ' ~ other , ' ~ types of injury,14 it has been limited in its ability to discriminate more subtle soft tissue structures. Multiplanar images can, of course, be obtained by CT with computer reconstruction, but these images lack the detail of those achieved by direct scanning. High resolution magnetic resonance imaging (MRI) of the musculoskeletal system, on the other hand, offers a dramatic advantage with respect to imaging soft tissues and imaging multiple planes directly. Its ability to discretely show tendons, ligaments, cartilage, nerves, and a variety of other soft tissues and joints in multiple projections has been well d e ~ c r i b e d as , ~ has its usefulness in the evaluation of bone marrow.'2 Some specific MRI applications to the foot and ankle have been addressed.',2,20-22 The most obvious shortcoming of this modality in the musculoskeletal system is its limited capability in imaging structures containing calcium. Because of the high incidence of injury to soft tissue structures with calcaneal fractures, particularly the peroneal tendons, one might wonder whether MRI can play a significant role in the evaluation of these injuries. The purpose of this paper, therefore, is to define the normal anatomy near the calcaneus and subtalar joint and to examine the capability of MRI in the evaluation of calcaneal fractures with respect to both bony and soft tissue structures as seen acutely and in follow-up.
ABSTRACT Images obtained from normal volunteers demonstrate highly detailed anatomy of the soft tissue and bony structures near the calcaneus and subtalar joint. Cortical bone, marrow, articular cartilage, ligaments, tendons, muscles, fibrous tissue, vascular bundles, and nerves can be identified. However, images obtained of acute calcaneal fractures were found to offer unsatisfactory depiction of bony anatomy. The presence of small fragments was obscured by a change in normal marrow signal by contusion, hemorrhage, and edema, and by the inability of magnetic resonance (MR) to image small pieces of cortical bone. Only in rare instances might MR be helpful in the acute setting when the location or displacement of tendons cannot be clearly ascertained with computed tomography (CT). MR may prove more useful in the long-term followup of healed fractures with persistent pain. In this setting it might be used in the diagnosis of complications such as residual or recurrent tendon displacement, tenosynovitis, heel fat integrity, and tarsal tunnel evaluation. However, this paper did not directly compare the efficacy of MR with that of CT in the long term. Therefore, the degree to which MR may eventually supplement or supplant CT is unclear and further study is required before the use of MR can be recommended in the routine clinical follow-up of calcaneal fractures.
Radiographic evaluation of the calcaneus consists primarily of plain films, plain film tomography, and computed t~mography.','~,'~ The latter has become almost indispensable in the evaluation of complex calcaneal f r a c t u r e ~ . ~ o,~l ~ Bony ~ * - ' anatomy can be exquisitely portrayed allowing accurate evaluation of fracture architecture. There are, however, two shortcomings to computed tomography (CT) capabilities in the musculoskeletal system in general: its inability to resolve fine soft
MATERIAL AND METHODS
Ten normal volunteers underwent calcaneal imaging with multiplanar MRI scanning. Five of the volunteer calcanei were imaged in sagittal, transverse, and coronal planes, while the other five were scanned in sagittal and oblique coronal projections. The latter was done at approximately 45' off the straight coronal plane. Three cadaver calcanei were imaged in sagittal, transverse,
From the Medical College of Ohio, Toledo, Ohio. * To whom all correspondence and reprint requests should be
addressed at: Medical College of Ohio, Department of Radiology, P.O. Box 10008, Toledo, Ohio 43699. t Department of Orthopedic Surgery. Department of Radiology.
*
264
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Foot & Ankle/Vol. 1 1, No. 5/April 1991
MRI OF THE CALCANEOUS
265
Fig. 1. Five sagittal T1 weighted MRI sections of a normal volunteer from lateral to medial hindfoot are presented. The most lateral section ( A ) shows the normal course and appearance to the peroneal tendons just behind the fibula. Image B demonstrates the lateral aspect of the posterior subtalar joint and the calcaneocuboid joint to good advantage and also shows the origin of the adductor digiti minimi muscle. Just lateral to midline (C) one sees the midportion of the posterior facet, the anterior facet (variable), the sinus tarsi and interosseous ligament, calcaneocuboid ligament, and talonavicular joint. Near midline and just medial to it (D), one can identify the medial aspect of posterior facet, the middle facet, the calcaneonavicular or spring ligament, as well as the quadratus plantae and flexor digitorum brevis muscles and Achille's tendon insertion. The most medial section ( E ) demonstrates the sustentaculum tali and middle facet as well as the flexor hallucis longus tendon. ADM = adductor digiti minimi muscle, AF = anterior facet, AT = Achilles tendon, C = calcaneous, CCJ = calcaneo-cuboidjoint, CCL = calcaneocuboid ligament, Cu = cuboid, F = fibula, FDB = flexor digitorum brevis muscle, FHL = flexor hallucis longus tendon, IL = interosseous ligament, MF = middle facet, N = navicular. PTB = peroneal tendon bundle, PF = posterior facet, QP = quadratus plantae muscle, SL = Spring ligament, ST = sustentaculum tali, SiT = sinus tarsi, Ta = talus, Ti = tibia.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Fig. 2. Three oblique-coronal T1 weighted MRI sections through a normal calcaneous are presented from posterior to anterior. A, is taken throiJgh the posterior facet, 8, through the middle facet, and C, through the anterior facet. All of the sections demonstrate medial and lateral
266
Downloaded from fai.sagepub.com at University College London on June 5, 2016
MRI OF THE CALCANEOUS
Foot & Ankle/Vol. 11, No. S/April 1991
267
Fig. 3. A, Normal axial T1 weighted MRI section through the peroneal tunnel at the level of talar dome and lateral malleolus shows intact superior peroneal retinaculum confining the peroneal tendons AT = Achilles tendon, C = calcaneous, F = fibula, PB = peroneal brevis tendon, PL = peroneal longus tendon, SPR = superior peroneal retinaculum, Ta = talus B, Normal oblique coronal T1 weighted MRI cut through the middle of tarsal tunnel shows this structure as defined by the medial malleolus, calcaneous, and flexor retinaculum The posterior tibial nerve within the tunnel can be discretely identified C = calcaneous, FHL = flexor hallucis longus tendon, FR = flexor retinaculum. PTN = posterior tibial nerve, Ta = talus, VB = vascular bundle
and coronal planes, then sectioned with a bandsaw and correlated to the MRI images. The three cadaver and 10 volunteer calcanei in conjunction with standard anatomical references were used to establish normal baselines. Sixteen intra-articular calcaneal fractures from 12 individuals were divided into three groups for MRI evaluation. Eight fractures were imaged within 48 hr of injury, four were imaged at approximately 3 to 4 months, and four were scanned at 2 years postinjury. The acute group all presented with a history of falling onto their heels with consequent pain and swelling. Fractures were first diagnosed by plain film and intraarticular fractures were selected for subsequent CT
and MRI evaluation. The subacute group was defined as healing intra-articular fractures at 3 to 4 months with varying degrees of residual discomfort, but no specific complaints. The healed calcaneal fractures at 2 years presented with a variety of complaints. One was asymptomatic, one had persistent lateral ankle pain with focal swelling, one complained of ankle instability with repeated inversion injuries, and one had an unsatisfactory response to tarsal tunnel release. The eight acute cases were evaluated with plain films, coronal CT with reconstruction, and MRI. The two follow-up groups had only plain films and MRI. MRI scans were done on a 1.5 Tesla General Electric Signa Scanner with a GE (General Electric, Milwaukee,
tendons, appropriate muscle groups and the medial and lateral plantar neurovascular bundles. ADM = adductor digiti minimi muscle, AH = adductor hallucis muscle, AF = anterior facet, C = calcaneous, F = fibula, FDB = flexor digitorum brevis muscle, FDL = flexor digitorum longus tendon, FHL = flexor hallucis longus tendon, LPNB = lateral plantar neurovascular bundle, MPNB = medial plantar neurovascular bundle, MF = middle facet, PA = plantar aponeurosis, PB = peroneal brevis tendon, PF = posterior facet, PL = peroneal longus tendon, QP = quadratus plantae muscle, SL = Spring ligament, ST = sustentaculumtali, Ta = talus, Ti = tibia, TP = tibialis posterior tendon.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
268
Foot & Ankle/Vol. 11, No. 5/April 1991
ZEISS ET AL.
Fig. 4. Corresponding coronal CT ( A ) and T1 weighted MRI (13)images in same individual with a comminuted calcaneal fracture are shown Contrast the sharply defined bony anatomy of the CT image to the poor bony detail of the MRI image The injured cancellous bone is all but obscured by low signal infiltration of hemorrhage and edema The displaced peroneal tendons can be identified in either image C = calcaneous, F = fibula, PB = peroneal brevis tendon, PL = peroneal longus tendon, PF = posterior facet, Ta = talus, Ti = tibia
WI) extremity coil. Each calcaneus was done with T1 and T2 weighting. The T1 scans were done in sagittal, axial, and coronal planes in half of the calcanei. Only sagittal and oblique coronal T1 images were done in the other half. The oblique coronals were done at approximately 45' to the straight coronal or straight axial plane. All of the calcanei were done with T2 weighting in one plane only, usually the oblique coronal. Minimum scan boundaries consisted of the anatomical limits of the calcaneous posteriorly and inferiorly, the midcuboid anteriorly, and the lateral malleolus superiorly. T1 technique was done with repetition time (TR) of 600 msec and echo time (TE) of 20 msec. T2 technique was done with TR of 2000 msec and TE of 80 msec. Slice thickness was 5 mm and no interslice gap was used. A 12-cm or 16-cm field of view was employed with a 256 x 128 matrix and two excitations. Scanning time for four T1 scans totalled approximately 30 min and for the T2 scan approximately 20 min. RESULTS
The MRI images of both the cadaver and volunteer calcanei showed excellent delineation of soft tissue and bony anatomy in all planes imaged. Each view demon-
strated particular strengths depending on the area of interest. For instance, sagittal views can nicely show all three facets of the subtalar joint, calcaneal height, and the calcaneocuboid joint. Transverse cuts are particularly useful in showing tendon position, tarsal and peroneal tunnel anatomy, and calcaneal widening. Coronal sections can demonstrate tendon position, calcaneal height and width, and subtalar joint, especially anterior facet. Oblique coronal views offer cross sections of the subtalar joint, especially posterior and middle facets, and combine many of the advantages of coronal and transverse cuts in a single scan. Figures 1 and 2 present images of the sagittal and oblique coronal planes at different levels through the volunteer calcanei. Figure 3A shows a transverse section through the lateral malleolus which demonstrates the peroneal tunnel, while Figure 38 is an oblique coronal image which demonstrates the tarsal tunnel. CT evalution of the eight acute calcaneal fractures showed them all to be comminuted fractures extending into the posterior subtalar joint, usually laterally. The number of fragments varied, but usually two large fragments, medial and lateral, were produced with varying degrees of additional comminution. The typical frac-
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Foot & Ankle/Vol. 11, No. 5/April 199 1
MRI OF THE CALCANEOUS
Fig. 5. T1 weighted axial MRI cut obtained through the level of the peroneal tunnel in a patient with acute calcaneal fracture. Note the disruption of the superior peroneal retinaculum and dislocation of the peroneal tendons from their usual position in the fibular groove. Compare this image with Figure 3A showing the normal appearance of retinaculum and peroneal tendons. F = fibula, FG = fibular groove, PTB = peroneal tendon bundle, SPR = superior peroneal retinaculum, Ta = talus.
ture consisted of a nondisplaced anteromedial sustentacular fragment, a larger posterolateral tuberosity fragment, axial widening secondary to lateral shift or spread of the tuberosity fragment, and craniocaudal flattening. CT uniformly presented a clear-cut demonstration of fragment type, number and position, and was of significant value in the proper classification of these fractures.” Peroneal tendon dislocation with displacement from behind the lateral malleolus was definite by CT in two of the eight fractures, very probable in one, but indefinite in another. The remaining four showed no dislocation, but varying degrees of lateral bowing of the peroneal tendons were seen in virtually every case because of the lateral spread of the tuberosity fragment. None of the fractures in this study showed significant distortion of the medial tendons. Upon viewing the MRI images of the acute calcaneal fractures it became immediately apparent that there
269
was a striking difference between the images of the normal volunteers and the injured patients. The normal high signal bone marrow within the general vicinity of the injured bone was replaced with a diffuse infiltrating signal of low intensity on the T I scans which blended almost imperceptibly with the injured adjacent soft tissues. On the T2 scans these abnormal marrow areas showed a heterogeneous mixture of predominantly low signal with some focal areas of high signal. Presumably this represented areas of microfracture with bone contusion consisting of hemorrhage, inflammation, and edema. Its effect was to negate the normal high marrow signal by which medullary bone is characterized on MRI, thereby making the margins of cancellous bone and small fragments of bone indistinguishable from adjacent hematoma or soft tissues. Cortical bone, because of its essentially unchanging low signal on both T1 and T2 scans, remained identifiable.Often, however, the cortical fragments were too small or too thin, or too close to a structure of similar intensity (e.g., tendon) to be recognizable. MRI was unable to even approach the sharp bone resolution in the face of acute trauma that one has come to expect from CT. The more extensively comminuted fractures demonstrated the greatest degree of this low T1 and mixed T2 signal infiltration and thereby the greatest lack of bony discrimination. Therefore, while larger fragments as well as the general fracture architecture and facet anatomy could be determined, MRI was of no significant help in fracture classification. Figure 4 demonstrates corresponding CT and MRI cuts from an acutely fractured calcaneous, and clearly points out the superiority of CT in this setting. MRI soft tissue resolution in the acute phase also suffered from the infiltrative effect of hemorrhage and edema when compared to the normal volunteer scans. The effect, however, was not nearly as detrimental to soft tissue discrimination as it was to bone. Soft tissue detail remained well above that of CT and the MRI images easily confirmed the tendon abnormalities found by CT. In the one patient where the peroneal tendons could not be definitively localized by CT because of diffuse soft tissue density around the region of the tendon bundle, MRI was able to demonstrate the tendon group with slight lateral subluxation and with accompanying rupture of the superior peroneal retinaculum. This finding was confirmed at surgery and repaired. Figure 5 demonstrates the typical MRI appearance of peroneal tendon dislocation. MRI evaluation of the subacute calcaneal fractures was compromised to a lesser extent than in the acute setting. Bone detail was significantly improved and one could see very low signal bands forming between the major fragments to indicate healing. The improved vis-
Downloaded from fai.sagepub.com at University College London on June 5, 2016
270
Foot & Ankle/Vol. 1 1, No. S/April 1991
ZEISS ET AL.
Fig. 6. Sagittal T1 weighted MRI images are shown at time of injury ( A ) and at approximately 3 months after injury ( B ) . Note the increasing bony detail as the fracture heals and the hemorrhage and edema resorb from the injured cancellous bone. AT = Achilles tendon, C = calcaneous, Cu = cuboid, Fx = fracture, Ta = talus.
ualization can probably be attributed to the diminishing marrow hemorrhage and edema secondary to healing. Figure 6 presents an example of a calcaneous 4 months after fracture and compares it to the initial MRI obtained at the time of injury. MRI soft tissue resolution in the subacute phase improved slightly from the acute phase as one might expect, and almost rivaled that in normal volunteers. The MRI calcaneal images obtained at 2 years postinjury demonstrated an appearance of bone and soft tissue structures essentially the same as that of the normal volunteers except for the varying degrees of residual deformity in the injured group. Marrow signal intensity was normal, cortical margins were well-defined, and soft tissue structures were sharply delineated. Bony alignment and facet anatomy could be easily evaluated. In this healed fracture group MRI evaluation did present some rather unique advantages. One patient was imaged after ineffectual release of his flexor retinaculum for tarsal tunnel syndrome secondary to old calcaneal fracture. The T1 examination revealed the caudal portion of the retinaculum to be intact. Reoperation resulted in successful release of the retinaculum and relief of the patient’s symptoms. Figure 7 shows two levels through this individual’s calcaneus revealing the retinaculum prior to reoperation to be released only proximally and intact distally. A second patient was imaged with persistent ankle instability and a tendency for inversion. MRI with T1
weighting demonstrated a disrupted superior peroneal retinaculum and a minimally subluxed peroneal tendon bundle, missed at the time of original treatment. This was confirmed at surgery and repaired (Fig. 8). A third patient had complained of persistent lateral ankle pain and focal swelling about the lateral malleolus. T2 weighted MRI images demonstrated findings consistent with tenosynovitis of a dislocated peroneal tendon bundle. The tendon bundle was not positioned in its usual location behind the fibula, but was lateral to it. The signal intensity of the tendon, normally homogeneously low or dark, was heterogeneous and abnormally high to indicate edema within it. It was also surrounded by a homogeneously high intensity periphery to indicate surrounding synovial fluid. DISCUSSION
This study indicates that MRI of the acutely fractured calcaneous can give a general impression of fracture architecture and a reasonable assessment of subtalar joint involvement. It is incapable, however, of adequately defining the existence and location of small bone fragments. The reason for this is the infiltration of hemorrhage and inflammation into the injured cancellous bone due to microfractures and contusion which alters the typical high marrow signal and makes it indistinguishable from surrounding soft tissue changes, both on T1 and T2 images. Small segments of cortical bone and the trabeculae of cancellous bone are insufficiently distinct to be discriminated from adjacent he-
Downloaded from fai.sagepub.com at University College London on June 5, 2016
Foot & Ankle/Vol. 11, No. 5/April 1991
MRI OF THE CALCANEOUS
271
Fig. 7. Oblique-coronal T1 weighted MRI images are presented through posterocephalad( A ) and anterocaudal(8) levels of the tarsal tunnel in a patient who failed to respond to flexor retinaculum release for tarsal tunnel syndrome following healing of calcaneal fracture. The cephalic part of the flexor retinaculum was released as indicated by the arrow in A , but the caudal portion of the retinaculum was inadvertently left intact as seen in 8.Note the healed though broadened calcaneous. Symptoms were relieved following reoperation and complete release. C = calcaneous, FR = flexor retinaculum, MF = middle facet, PF = posterior facet, TA = talus.
matoma or other structures of similar intensity such as tendon bundles. Those fractures with the greatest degree of comminution and hemorrhage are most compromised in MRI evaluation, but all are affected to a significant degree. CT, therefore, remains the obvious choice in bony evaluation of calcaneal fractures. While MRI enhances ability to discriminate soft tissue structures, does it have a role in the evaluation of acute calcaneal injuries? In most cases, CT has sufficient soft tissue resolution to determine impingement on or displacement of peroneal or flexor hallucis tendons. Occasionally, however, the desired soft tissues structures cannot be adequately discriminated by CT.6One might also encounter an instance of superior peroneal retinaculum tear with very little or no displacement of the peroneal tendons, similar to that seen in Figure 8.22In such rare instances MRI can be of help, but this would certainly not warrant routine use of MRI in a fracture setting.
In the subacute phase, with resolution of the adjacent hematoma, MRI visualization improves, but is still suboptimal from a bone evaluation standpoint. The use of MRI in this timeframe would depend on the major clinical concern. CT indications would include determination of healing and residual bony deformity, while MRI might be used for residual tendon displacement, avascular necrosis, persistent localized edema, and osteomyelitis. Unlike the limitations encountered in MRI imaging of acute and subacute calcaneal fractures, healed fractures can be imaged with essentially the same sharpness of bony and soft tissue detail as healthy volunteers. At this stage soft tissue and cancellous bone hemorrhage and inflammation have resolved and the normal high marrow signal is re-established. Cortical edges and facets are sharply outlined. Therefore, the most helpful contribution of MRI evaluation of calcaneal injury may be found in long-term evaluation. T1 weighted scans are ideal for anatomical depiction, while
Downloaded from fai.sagepub.com at University College London on June 5, 2016
272
Foot & Ankle/Vol. 11, No. S/April 1991
ZEISS ET AL.
of calcaneal fractures is undefined. Since MRI is presently more costly, more time-consuming, and less available than CT, further experimental study is required before its routine clinical use can be recommended. CONCLUSIONS
The usefulness of MRI evaluation of calcaneal fractures in acute and subacute evaluation will most likely be very limited to occasional instances where CT does not clearly show soft tissue anatomy such as tendon displacement, or where complicating avascular necrosis or osteomyelitis is of concern. It could be more helpful in evaluation of persistent pain complicating healed fractures. As it completes directly with CT, additional investigationwill be necessary to more clearly define those instances where it may be of particular value.
REFERENCES
Fig. 8. Transverse Tl weighted MRI through the level of the peroneal tunnel in a patient with persistent pain along the lateral ankle and inversion instability 2 years following calcaneal fracture. The superior peroneal retinaculum is avulsed from its fibular attachment but the peroneal tendons are only slightly subluxed from their normal location. F = fibula, PTB = peroneal tendon bundle, SPR = superior peroneal retinaculum, Ta = talus.
T2 weighting is useful with infectious or inflammatory complications. Common causes of persistent discomfort in healed calcaneal fractures include heel pad pain, fibulocalcaneal impingement, scarring, and posttraumatic arthritis of the subtalar joint, peroneal tendonitis due to compression and displacement, plantar calcaneal spurs due to malunion and associated with disruption of the heel fat, arthritis of the calcaneocuboid, talonavicular and ankle joints, and nerve entrapment of medial and lateral plantar branche~.~.” These entities have been evaluated with a variety of diagnostic modalities including physical examination, plain radiography, CT, and nerve conduction studies. While each of these has a place in any given situation, the abilities of MRI should be particularly useful in evaluating articular cartilage and joint anatomy, residual tendon displacement, tenosynovitis, and nerve entrapment. A specific comparison between MRI and CT in the evaluation of chronic calcaneal injuries was not made in this study, and the exact role of MRI in the follow-up
1 . Beltran, J., Noto, A.M., Mosure, J.C., Shamam, O.M., Weiss, K.L., and Zuelzer, W.A.: Ankle: surface coil MR imaging at 1.5T. Radiology, 161:203-209, 1986. 2. Crim, J.R., Cracchiolo, A., Bassett, L.W., Seeger, L.L., Soma, C.A., and Chatelaine, A.: Magnetic resonance imaging of the hindfoot. Foot Ankle, lO(1):l-7, 1989. 3. Deyerle, W.M.: Long term follow-up of fractures of the 0s calcis. Orthop. Clin. North Am., 4(1):213-227. 1973. 4. Ehman, R.L., Berquist, T.N., and McLeod, H.A.: MR imaging of the musculoskeletal system: a 5-year appraisal. Radiology, 166~313-320,1988, 5. Gilmer, P.W., Herzenberg, J., Frank, J.L., Silverman, P., Martinez, s., and Goldner, J.L.: Computerized tomographic analysis of acute calcaneal fractures. Foot Ankle, 6(4):184-193, 1986. 6. Guyer, B.H., Levinsohn, M., Fredrickson, B.E., Bailey, G.L., and Formikell, M.: Computed tomography of calcaneal fractures: anatomy, pathology, dosimetry and clinical relevance. Am. J. Roentgen., 145911-919, 1985. 7. Heger, L., and Wulff, K.: Computed tomography of the calcaneus: normal anatomy. Am. J. Roentgen., 145:123-129, 1985. 8. Heger, L., Wulff, K., and Seddiqi, M.S.A.: Computed tomography of calcaneal fractures. Am. J. Roentgen., 145:131-137, 1985. 9. Hindman, B.W., Ross, S.D.K., and Sowerby, M.R.R.: Fractures of the talus and calcaneus: evaluation by computed tomography. J. Comput. Tomogr., 10(2):191-196, 1986. 10 Pablot, S.M., Daneman, A., Stringer, D.A., and Carroll, N.: The value of computed tomography in the early assessment of comminuted fractures of the calcaneus: a review of three patients. J. Pediatr. Orthop., (5)4:435-438, 1985. 1 1 Paley, D., and Hall, H.: Calcaneal fracture controversies. Orthop. Clin. North Am., 20(4):665-677, 1989. 12 Porter, B.A.: MR may become routine for imaging bone marrow. Diagn. Imaging, 104-1 08, 1987. 13 Romash, M.M.: Calcaneal fractures: three-dimensional treatment. Foot Ankle, 8(4):180-213, 1988. 14 Rosenberg, Z.S., Feldman, F., and Singson, R.D.: Peroneal tendon injuries: CT analysis. Radiology, 161:743-748, 1986.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
MRI OF THE CALCANEOUS
Foot 8, Ankle/Vol. 1 1, No. S/April 1991 15. Rosenberg, Z.S., Feldman, F.,and Singson, R.D.:Intra-articular calcaneal fractures: computed tomographic analysis. Skeletal Radiol., 16:105-113, 1987. berg, z,s., Feldman, F., and singson, R.D.: peroneal 16, R tendon injury associated with calcaneal fractures: CT findings. Am. J. Roentgen., 149:125-129, 1987. 17. Rosenberg, Z.S., Feldman, F., Singson, R.D., and Kane, R.: Ankle tendons: evaluation with CT. Radiology, 166:221-226, 1988. 18. Sartoris, D.J., and Resnick, D.: Diagnostic imaging approach to calcaneal fractures. J. Foot Surg., 26(6):524-529, 1987.
273
19. Solomon, M.A., Gilula, L.A., Oloff, L.M., and Oloff, J.: CT scanning of the foot and ankle: 2. Clinical applications and review of the literature. Am. J. Roentgen., 146:1204-1214, 1986. 20. Solomon, M.A., and Oloff-Solomon, J.: Magnetic resonance imaging in the foot and ankle. Clin. Podiatr. Med. Surg., 5(4):945965, 1988. 21. Zeiss, J., Fenton, P., Ebraheim, N., and Coombs, R.: Normal magnetic resonance anatomy of the tarsal tunnel. Foot Ankle, 10(4):214-218, 1990. 22. Zeiss, J., Saddemi, S.R., and Ebraheim, N.A.: MR imaging of the peroneal tunnel. J. Comput. Assist. Tomogr., 13(5):840844, 1989.
Downloaded from fai.sagepub.com at University College London on June 5, 2016
