Send comments/questions via email
or fax 800-556-3288.
Visit CompendiumVet.com for full-text articles, CE testing, and CE test answers.
Nathaniel A. White, DVM, MS, DACVS
Virginia Polytechnic Institute and State UniversityNatasha M. Werpy, DVM
Colorado State University
What a Difference MRI Makes
Consultant’s Corner provides expert answers to questions frequently asked by equine practitioners.
A relatively short time after the usefulness of magnetic resonance imaging (MRI) was discovered, it became commonly used in human medicine. MRI, which relies on the reaction of molecules in a magnetic field, was first used in human medicine in 1971, and the first image of a human head was published in 1978.1Since then, use of super-cooled magnets for MRI has resulted in increases in the magnetic field strength, improved image resolution, and increased speed of image acquisition. MRI became the imaging modality of choice for neurologic scanning during the 1980s and is now con-sidered by many to be the ultimate imaging modality for the musculoskeletal system because of its ability to simultaneously detect changes in bone and soft tissue.2
MRI is a new modality in equine medicine. Some of the larger MRI systems made for human medicine and called high-field magnets have been adapted for horses. Although the larger magnets are expensive to purchase and maintain, they can create high-resolu-tion images. The magnetic field strength, which is measured in tesla units, is usually 1 to 1.5 tesla for most MRI systems used for humans. These high-field magnets have recently been used to image small animals and anesthetized horses3–5(Figure 1).
A new low-field MRI system (Hallmarq Veterinary Imaging Ltd., Guildford, Surrey, UK) designed for use in standing horses uses an open magnet with a field strength of 0.3 tesla. The open magnet can be positioned around the horse’s lower limb or foot to image the area of interest (Figure 2). The resulting images of the weight-bearing limb allow detection of changes in soft tissue and bone similar to those detected using larger conventional magnets. Images can be obtained from the foot to the carpal or tarsocrural joints, and because movement creates image artifacts, antimotion software is part of the image capture system (Figure 3). In some cases, movement artifacts cannot be corrected and general anesthesia is required to obtain diagnostic images. Because of the lower cost of low-field magnets, MRI is becoming available for all animals, including horses. With all MRI systems, the opening into the magnet restricts MRI use in adult horses from the foot to the distal radius and tibia and on the head.
HOW DOES MRI WORK?
Unlike radiography or ultrasonography, which detect structure based on tissue or fluid density, MRI detects changes in proton resonance. Protons make up the nucleus of atoms, and when placed in the magnet, protons align with the magnetic field and become magnetized (Figure 4). When stimulated by a radiofrequency pulse, protons in different types of tissue have a different resonance. For example, protons in fat resonate differently than do protons in fluid.
To detect different types of tissue or abnormalities in tissue, electromagnetic energy or a radiofrequency pulse is emitted from a coil, which is placed around the part of the limb to be scanned. The radiofrequency pulse alters the magnetization alignment of
protons within the magnetic field. When the radiofre-quency pulse is stopped, the protons release energy as they realign within the magnetic field. The released energy or signal is detected by the MRI coil, which
transmits it to the computer, where the impulse is trans-formed into an image (Figure 5).
To detect different types of tissue, the angle of the radiofrequency pulse and its length (known as repetition time), as well as the time during proton realignment when the radiofrequency signal release is transmitted to the MRI computer (known as echo time), are adjusted to image contrast and highlight different types of tissue or fluid. The sequences most commonly used for muscu-loskeletal imaging are called fast spin echo and gradient echo, which alter the way the radiofrequency pulse is delivered and subsequently detected. Each of these sequences can be weighted based on the length of the radiofrequency pulse and the time when the release of radiofrequency from the tissue is detected. The weight-ing is referred to as T1 and T2 weightweight-ing.
The MRI scan is collected in slices. The benefit of MRI is that slices can be made in any plane of the limb. In some sequences, the slice is made with a specific length, depth, and width, whereas in others, the slice is made as an average from the information collected from a specific volume of tissue. The slices are normally taken every 3 to 5 mm, making it possible to miss very small lesions. In most cases, the changes in horses are suffi-Figure 1. A horse under general anesthesia for MRI of a
front limb in a 1-tesla magnet (Ortho One, ONI Medical Systems,Wilmington, MA) at the Orthopedic Research Laboratory at Colorado State University.
Figure 2. The MRI machine (Hallmarq Veterinary Imaging Ltd., Guildford, Surrey, UK) at the Marion duPont Scott Equine Medical Center (Virginia
Polytechnic Institute and State University). An open
magnet can be moved around a horse’s foot or limb and raised or lowered to the necessary height for the region to be scanned. Horses must be sedated and stand still for each scan sequence.
Figure 3. The open magnet is centered around the horse’s foot. The coil is positioned around the part of the limb
ciently large to be detected by MRI.
Although the physics and terminology of MRI are not easily understood without some knowledge of the science behind it, MRI has been refined so that the images resemble structures. Nevertheless, when an MRI scan is examined, the evaluator must remember that both the anatomy and physiology of the tissue are being displayed, rather than just density or structure. The chemical makeup is related to tissue density and accu-rately identifies types of tissue.
WHAT DO MRI SCANS LOOK LIKE?
The three main substances detected by MRI include fluids (e.g., joint fluid, edema), water-rich tissue (e.g., muscle, cartilage), and fat (which is found in bone mar-row). MRI can detect changes in these components due to inflammation, bruising, fracture, or scarring. The sig-nal is the amount of information on the image. The images are made from pixels, each representing the amount of signal from a specific site in the tissue. Tissue that appears white has high signal intensity, whereas tis-sue that is dark has low signal intensity. In most T1 and T2 sequences, fat appears bright with a high signal intensity (Figure 6).
In T1-weighted images, fluid appears gray; in T2-weighted images, fluid is white, and cortical bone, ten-dons, and ligaments appear black. Cartilage, which has a high water content, is white in both sequences. A short T1-inversion recovery (STIR) sequence uses pulse
radiofrequency to invert the pro-ton alignment, thereby suppress-ing the signal from fat. This allows examination of all structures, par-ticularly the bone marrow cavity, so abnormal fluid or blood can be identified (Figure 7).
DOES MRI MAKE A DIFFERENCE IN DIAGNOSIS?
The value of MRI in diagnos-ing musculoskeletal disorders is the ability to detect active inflam-mation at a specific anatomic site. Although changes in bone min-eralization can be detected with radiography, these processes take time, and alterations in bone den-sity may not be evident at the onset of acute inflammation. Scintigraphy can help localize lesions but is not sensitive enough to detect spe-cific morphologic changes. Ultrasonography is a very effective technique for detecting changes in soft tissue structure but cannot always detect early evidence of inflammation or discriminate between fibrous tissue and active inflammation. Detecting inflammation or struc-tural change is even more difficult in a horse’s foot because of the limited ability to evaluate structures within the hoof capsule.
MRI can detect changes in bone and soft tissue that are not evident with the other imaging modalities. Since use of MRI in horses was first described in 1996,6 MRI has been used to detect injuries to tendons, ligaments of the distal interphalangeal joints, the metacarpal pha-langeal joint, the suspensory ligament, the hock, the car-pus, subchondral bone, cartilage, fractures, hoof lamina, the navicular bursa, navicular bone, and the impar ligament.4,7–12
MRI has helped detect changes and inflammation as-sociated with navicular disease. Because pain can come from several structures surrounding and supporting the navicular bone, MRI is helpful in determining which structure(s) is inflamed or scarred. When radiographs and ultrasonograms are normal, MRI can often detect bone edema, navicular suspensory ligament inflamma-tion, deep digital flexor tendinitis, navicular bursa adhe-sions, and changes in the impar ligament.7,13In a study13 of 199 horses with caudal heel pain, 59% of lesions Magnetic field
Magnetic field Figure 4. Protons align their axes when exposed to a magnetic field. The greater
detected by MRI were injuries to the deep digital flexor tendon (Figure 8). The prognosis for full return to soundness for horses with this lesion was only 28%, sug-gesting that these lesions are potentially chronic and not always amenable to currently available treatments.
Lesions identified by MRI in the caudal heel region may make the diagnosis of navicular disease a dilemma in some horses. Although navicular disease is characteristically thought to be due to pain originat-ing from the navicular bone and bursa, MRI is able to distinguish between inflammation of the navicular bone, deep digital flexor tendon, bursa, or suspensory ligament. As a result, MRI has shown that lameness associated with navicular or caudal heel pain can be due to inflammation in different structures.
Injury to the collateral ligaments of the distal inter-phalangeal joint has been detected with MRI (Figure 9). In fact, in one clinical study,9 distal interphalangeal collateral ligament desmitis was the second most fre-quent diagnosis related to foot pain based on MRI find-ings. However, lesions identified by MRI do not always Figure 6. A T1 gradient echo image. The fat in the marrow
cavity has a high signal (i.e., white), whereas cortical bone and tendon have a low signal (i.e., black). Fluid has a medium signal (i.e., gray).
Start
Radiofrequency
pulse
Radiofrequency pulse
Emitted radiofrequency
Stop
Radiofrequency
pulse
Magnetic field
Magnetic field
Magnetic field
Figure 5. The pulse of the radiofrequency changes the proton axis alignment. Once the radiofrequency pulse stops, the
proton axis is realigned in the magnetic field, releasing a radiofrequency that is acquired by the computer at specific times (i.e., echo time) to make images with specific types of tissue contrast.
correlate with ultrasonographic findings. As clinicians become better able to distinguish specific diseases with MRI, its value in prognostication will undoubtedly increase.
Inflammation within the proximal and distal suspen-sory ligaments can be detected with MRI and seen without ultrasonographic evidence of injury. MRI can detect changes in size as well as evidence of fluid accu-mulation within ligaments before there is ultrasono-Figure 7. A STIR sequence suppresses the signal from fat.
Suppression of the signal from fat makes the marrow cavity of the middle phalanx and distal phalanx appear gray, whereas fluid in the coffin joint has a high signal and appears white (arrow).
A STIR sequence from a horse with a bone bruise in the proximal aspect of the middle phalanx (arrow).The abnormal fluid in the bone has a high signal in contrast to the low signal from fat.
Figure 8. T2-weighted image of a deep digital flexor tendon injury adjacent to the navicular bone (N). The focal
increased signal in the tendon (arrow), which is normally black, is indicative of fluid within the tendon. (Courtesy of Wisconsin Equine Clinic and Hospital, Oconomowoc, WI)
Figure 9. Increased signal in a transverse view of a collateral ligament of the distal interphalangeal joint
(arrow). The ligament should have a low signal (i.e., black) as seen
graphic evidence of fluid accumulation or fiber damage. This capability of MRI is extremely valuable in horses with acute injuries (Figure 10).
One of the most valuable features of MRI is its ability to detect inflammation in bone. Both fluid (e.g., bone edema) and sclerosis (e.g., increased bone density) are detectable with MRI; these findings are difficult to impossible to detect by radiography. Bone bruises, infec-tion (Figure 11), subchondral damage due to joint dis-ease, and bone remodeling can all be visualized with MRI before they are evident radiographically. Bone sclerosis with remodeling in young athletic horses is evi-dence of stress remodeling and, with clinical experience, allows early detection of bone injury before appearance of stress fractures on radiographs. Subchondral sclerosis can indicate chronic injury in joints (Figure 12).
Cartilage changes are often difficult to detect, but full-thickness defects, which are large enough to be intersected at one of the MRI slices, can be identified.
Changes to subchondral bone are often concurrent with cartilage damage and can be an early indicator of joint disease or degenerative joint disease6,14,15(Figure 13).
WHAT’S NEXT FOR MRI?
With the advent of the MRI system for standing horses, the number of horses being examined with MRI is increasing rapidly. However, it is important to recog-nize that interpretation of the resulting images is in its infancy. Interpreting MRI scans requires new knowl-edge, with an understanding of what the images repre-sent. Veterinarians trained and experienced in evaluating radiographs and ultrasonograms require additional training to read MRI scans. Otherwise, it could be extremely easy to misinterpret variations within normal tissue as lesions. Nevertheless, MRI can be used to find tissue changes and pathology that cannot be identified using other imaging modalities.
As with ultrasonography and radiography, artifacts exist with MRI and must be recognized.16 Because imperfections can exist in the magnetic field and MRI is new in horses, interpretations of abnormalities must be corroborated with findings from other modalities, including gross and histologic evaluation of affected tis-sue. To minimize misinterpretation of artifacts, a com-Figure 10. Transverse MRI scan of a chronic (i.e.,
2-month duration) rear proximal suspensory ligament between the lateral and medial splint bones (S). There is
disruption of the plantar margin, which is replaced with a mix of fluid and fibrous tissue (i.e., a high signal surrounded by a normal low signal; arrow) in its core, indicating inflammation.
Figure 11. Osteomyelitis of the proximal phalanx and distal cannon bone with fluid (i.e., high-intensity signal) within the bone (arrow).
plete MRI examination protocol with an adequate num-ber of sequences should be used on each region being examined to ensure that all pertinent structures are prop-erly imaged. Interpretation of findings must be made with enough skepticism and objectivity to ensure the diagnosis is based on appropriate criteria rather than conveniently fitting the clinical picture. Although MRI studies4,17–23of normal anatomy have been published, it is clear that many more horses must be scanned to estab-lish normal variations and the limits of this modality.
MRI does not replace other imaging modalities. Lesions identified with MRI can often be detected ret-rospectively by reexamining ultrasonograms or radi-ographs, albeit with the knowledge that a lesion exists. Although it does not make sense to use MRI to facili-tate a radiographic diagnosis, it does make sense to con-duct MRI evaluations on straightforward cases to increase the available data for corroboration of diseases and create a resource for future reference.5,8,10,15,24–27
If the future of equine MRI is anything like what has occurred in human medicine and small animal medi-cine, the only limitation to its usefulness in horses will be the size of the magnet. Motion correction software for use with the standing MRI system is improving.28 Current limitations are based on machines adapted from
human medicine. As MRI technology advances, there are likely to be larger open magnets, which can be used to image more of a horse. It is already clear that MRI is taking equine imaging to a new and exciting level. Although evaluating MRI scans is challenging, the greater challenge will likely be treating what can now be detected with MRI.
REFERENCES
1. McRobbie DW, Moore EA, Graves MJ, Prince MR: MRI From Picture to
Proton. Cambridge, UK, Cambridge University Press, 2003.
2. Rodkey WG, Steadman JR, Ho CP: Role of MR imaging in sports medicine research: Basic science and clinical research studies. Magn Reson Imaging Clin
N Am 7(1):191–203, 1999.
3. Tucker RL, Sande RD: Computed tomography and magnetic resonance imaging of the equine musculoskeletal conditions. Vet Clin North Am Equine
Pract 17(1):145–157, vii, 2001.
4. Denoix JM, Audigie F: Imaging of the musculoskeletal system in horses, in Hinchcliff KW, Kaneps AJ, Geor RJ (eds): Equine Sports Medicine and
Surgery, ed 1. Philadelphia, WB Saunders, 2004, pp 161–187.
5. Ferrell EA, Gavin PR, Tucker RL, et al: Magnetic resonance for evaluation of neurologic disease in 12 horses. Vet Radiol Ultrasound 43(6):510–516, 2002. 6. Martinelli MJ, Baker GJ, Clarkson RB, et al: Magnetic resonance imaging of
degenerative joint disease in a horse: A comparison to other diagnostic tech-niques. Equine Vet J 28(5):410–415, 1996.
7. Dyson S, Murray R, Schramme M, Branch M: Lameness in 46 horses associ-ated with deep digital flexor tendonitis in the digit: Diagnosis confirmed with magnetic resonance imaging. Equine Vet J 35(7):681–690, 2003.
8. Dyson S, Murray R, Schramme M, Branch M: Magnetic resonance imaging
Figure 12. T1-weighted sagittal MRI scan of a foot with sclerosis of the navicular bone (i.e., black; low signal), which normally appears white (i.e., high signal) as in Figure 6. (Courtesy of Equest Imaging, San Marcos, CA)
Figure 13. Sagittal fat–suppressed MRI scan of the fetlock. Interruption of the confluent white cartilage and focal
fluid accumulation in the subchondral bone are the result of a cartilage defect (arrow).
of the equine foot: 15 horses. Equine Vet J 35(1):18–26, 2003.
9. Dyson SJ, Murray R, Schramme M, Branch M: Collateral desmitis of the distal interphalangeal joint in 18 horses (2001–2002). Equine Vet J 36(2):160–166, 2004.
10. Zubrod CJ, Schneider RK, Tucker RL: Use of magnetic resonance imaging to identify suspensory desmitis and adhesions between exostoses of the sec-ond metacarpal bone and the suspensory ligament in four horses. JAVMA 224(11):1789, 1815–1820, 2004.
11. Tapprest J, Audigie F, Radier C, et al: Magnetic resonance imaging for the diagnosis of stress fractures in a horse. Vet Radiol Ultrasound 44(4):438–442, 2003.
12. Kasashima Y, Kuwano A, Katayama Y, et al: Magnetic resonance imaging application to live horse for diagnosis of tendinitis. J Vet Med Sci 64(7):577– 582, 2002.
13. Dyson SJ, Murray R, Schramme MC: Lameness associated with foot pain: Results of magnetic resonance imaging in 199 horses ( January 2001–Decem-ber 2003) and response to treatment. Equine Vet J 37(2):113–121, 2005. 14. Kaplan PA, Helms CA, Dussault R, et al: Musculoskeletal MRI. Philadelphia,
WB Saunders, 2001.
15. Zubrod CJ, Schneider RK, Tucker RL, et al: Use of magnetic resonance imaging for identifying subchondral bone damage in horses: 11 cases (1999–2003). JAVMA 224(3):411–418, 2004.
16. Busoni V, Snaps F: Effect of deep digital flexor tendon orientation on mag-netic resonance imaging signal intensity in isolated equine limbs: The magic angle effect. Vet Radiol Ultrasound 43(5):428–430, 2002.
17. Arencibia A, Vazquez JM, Ramirez JA, et al: Magnetic resonance imaging of the normal equine brain. Vet Radiol Ultrasound 42(5):405–409, 2001. 18. Anastasiou A, Skioldebrand E, Ekman S, Hall LD: Ex vivo magnetic
reso-nance imaging of the distal row of equine carpal bones: Assessment of bone sclerosis and cartilage damage. Vet Radiol Ultrasound 44(5):501–512, 2003. 19. Blaik MA, Hanson RR, Kincaid SA, et al: Low-field magnetic resonance
imaging of the equine tarsus: Normal anatomy. Vet Radiol Ultrasound 41(2):131–141, 2000.
20. Busoni V, Snaps F, Trenteseaux J, Dondelinger RF: Magnetic resonance imaging of the palmar aspect of the equine podotrochlear apparatus: Normal appearance. Vet Radiol Ultrasound 45(3):198–204, 2004.
21. Kleiter M, Kneissl S, Stanek C, et al: Evaluation of magnetic resonance imaging techniques in the equine digit. Vet Radiol Ultrasound 40(1):15–22, 1999.
22. Murray RC, Roberts BL, Schramme MC, et al: Quantitative evaluation of equine deep digital flexor tendon morphology using magnetic resonance imaging. Vet Radiol Ultrasound 45(2):103–111, 2004.
23. Wright IM, Kidd L, Thorp BH: Gross, histological and histomorphometric features of the navicular bone and related structures in the horse. Equine Vet J 30(3):220–234, 1998.
24. Murray RC, Dyson SJ, Schramme MC, et al: Magnetic resonance imaging of the equine digit with chronic laminitis. Vet Radiol Ultrasound 44(6):609–617, 2003.
25. Rand T, Bindeus T, Alton K, et al: Low-field magnetic resonance imaging (0.2 T) of tendons with sonographic and histologic correlation: Cadaveric study. Invest Radiol 33(8):433–438, 1998.
26. Sanders SG, Tucker RL, Bagley RS, Gavin PR: Magnetic resonance imaging features of equine nigropallidal encephalomalacia. Vet Radiol Ultrasound 42(4):291–296, 2001.
27. Zubrod CJ, Farnsworth KD, Tucker RL, Ragle CA: Injury of the collateral ligaments of the distal interphalangeal joint diagnosed by magnetic reso-nance. Vet Radiol Ultrasound 46(1):11–16, 2005.
28. McKnight AL, Manduca A, Felmlee JP, et al: Motion-correction techniques for standing equine MRI. Vet Radiol Ultrasound 45(6):513–519, 2004.
