200
Equine vet. J. (2010) 42 (3) 200-207
Magnetic resonance imaging of the equine
temporomandibular joint anatomy
M. J. RODRÍGUEZ*, A. AGUT, M. SOLER, O. LÓPEZ-ALBORS†, J. ARREDONDO§, M. QUEROL‡ and R. LATORRE†
Department of Medicine and Surgery, Veterinary School, University of Murcia, 30100 Espinardo, Murcia, Spain; †Department of Anatomy and
Comparative Anatomy, Veterinary School, University of Murcia, 30100 Espinardo, Murcia, Spain; ‡Resonancia de Murcia, Medicina y Gestión
Sanitaria, S.L., Murcia, Spain; and §Faculty of Veterinary Medicine & Zootechny, Autonomous University of the State of Mexico, Mexico.
Keywords: horse; temporomandibular joint; magnetic resonance imaging; anatomy; plastination
Summary
Reasons for performing study: In human medicine, magnetic
resonance imaging (MRI) is considered the ‘gold standard’
imaging procedure to assess the temporomandibular joint
(TMJ). However, there is no information regarding MRI
evaluation of equine TMJ.
Objectives: To describe the normal sectional MRI anatomy of
equine TMJ by using frozen and plastinated anatomical
sections as reference; and determine the best imaging planes
and sequences to visualise TMJ components.
Methods: TMJs from 6 Spanish Purebred horse cadavers (4
immature and 2 mature) underwent MRI examination. Spin-
echo T1-weighting (SE T1W), T2*W, fat-suppressed (FS)
proton density-weighting (PDW) and fast spin-echo
T2-weighting (FSE T2W) sequences were obtained in oblique
sagittal, transverse and dorsal planes. Anatomical sections
were procured on the same planes for a thorough
interpretation.
Results: The oblique sagittal and transverse planes were the
most informative anatomical planes. SE T1W images showed
excellent spatial resolution and resulted in superior anatomic
detail when comparing to other sequences. FSE T2W
sequence provided an acceptable anatomical depiction but
T2*W and fat-suppressed PDW demonstrated higher contrast
in visualisation of the disc, synovial fluid, synovial pouches
and articular cartilage.
Conclusions: The SE T1W sequence in oblique sagittal and
transverse plane should be the baseline to identify anatomy.
The T2*W and fat-suppressed PDW sequences enhance the
study of the articular cartilage and synovial pouches better
than FSE T2W.
Potential relevance: The information provided in this paper
should aid clinicians in the interpretation of MRI images of
equine TMJ and assist in the early diagnosis of those problems
that could not be diagnosed by other means.
Introduction
The recent emphasis placed on the equine temporomandibular joint
(TMJ) by clinicians and owners results from an improved
*Authors to whom correspondence should be addressed.
[Paper received for publication 07.04.09; Accepted 02.08.09]
© 2010 EVJ Ltd
EQUINE VETERINARY JOURNAL
doi: 10.1111/j.2042-3306.2010.00030.x
understanding of this region. Much of this is thanks to an increase
in the amount of research work concerning the TMJ (Weller et al.
2002; Bonin et al. 2006; Rodríguez et al. 2006; Ramzan et al.
2008) and equine dentistry (Schumacher 2001; Weller et al. 2001),
in which the relationship of this articulation to different aspects of
the horse’s performance is considered (Cooks 2006; Ramzan
2006). Reports found in the literature are mainly related to clinical
cases involving fractures (Devine et al. 2005), luxations (Hardy
and Shiroma 1991) and septic and nonseptic arthritis (Warmerdam
et al. 1997; Carmalt and Wilson 2005; Nagy and Simhofer 2006).
The main clinical signs associated with these diseases are localised
inflammation, mouth-opening limitation, malocclusion and pain on
palpation. Other clinical signs such as bitting, quidding,
headshaking, masticatory and dental problems or even sour attitude
and gait alterations have been recently associated with TMJ
disorders (May et al. 2001; Ramzan 2006); most of them can
fluctuate over time, with activity or be overlooked because of their
subtlety. Therefore, the information collected from the clinical
history and physical examination alone is not sufficient to permit
the diagnosis of a TMJ disorder (Schumacher 2001; Ramzan 2006).
Several techniques have been described for imaging the equine
TMJ, namely radiography, ultrasonography, scintigraphy,
arthroscopy and computed tomography (CT). Radiography is
traditionally employed to image the osseous structures of the
equine head, but the overlapping of adjacent structures makes the
acquisition of diagnostic images and its interpretation difficult
(Weller et al. 1999a; Townsend and Cotton 2007; Ramzan et al.
2008). Ultrasonography has the capacity of visualising the lateral
aspect of soft tissues, articular cartilage and bone surfaces of the
joint with a reduced cost (Weller et al. 1999b; Rodríguez et al.
2007). Nuclear scintigraphy has been proved as a high sensitivity
method for detection of TMJ disorders although it lacks specificity
in characterising the pathology (Weller et al. 1999a, 2001).
Arthroscopy allows direct diagnosis and treatment of the
articulation with excellent detail of intra-articular structures but is
an invasive technique (May et al. 2001; Weller et al. 2002).
In human medicine, CT, in particular multislice helical CT, and
magnetic resonance imaging (MRI) are standard procedures for the
TMJ examination offering the possibility of multiplanar imaging
acquisitions and 3D analysis (Shimazaki et al. 2007). Both CT and
MRI can provide a high spatial resolution and better image
M. J. Rodríguez et al. 201

TABLE 1: Imaging parameters used for this study

Sequences parameters SE T1W T2* Fat-suppressed PDW FSE T2W

TR (ms) 300–400 350 2800 2800–3140

TE (ms) 12 15 22.9 91.5
Flip angle 30°
Turbo factor 10–12 10–12
Slice thickness (mm) 4.0 4.0 4.0 4.0
Interslice gap (mm) 0.0 0.0 0.0 0.0
Field of view (cm) 16.0 16.0 16.0 16.0
Matrix size 512 ¥ 512 512 ¥ 512 512 ¥ 512 512 ¥ 512
No. excitations 4 4 4 4
Scan time (min) 4–6 5–6 7–12 5–6

SE T1W = spin-echo T1-weighting; T2* = gradient echo T2-weighting; PDW = proton density-weighting; FSE T2W = fast spin-echo T2-weighting; TR = repetition
time; TE = echo time; turbo factor = echo train length.

contrast, giving an early evidence of pathological changes, and selected using the 3 spatial gradients, and obtained in 3 planes.
consequently, enhancing the effectiveness of therapeutic strategies Oblique sagittal images were orientated perpendicular to the
(Stehling et al. 2007). Although CT is useful for studying soft lateromedial axis of the mandibular condyle visualised on the dorsal
tissues, it is particularly effective in assessing bone TMJ structures plane. Oblique transverse images were obtained parallel to the
(humans: Vilanova et al. 2007; horses: Tucker and Farrell 2001; lateromedial axis of the mandibular condyle showed on the dorsal
Rodríguez et al. 2008). However, MRI has lately been considered plane, and dorsal images were orientated parallel to the articular
the best suited imaging procedure to evaluate the human TMJ due surface of the mandibular condyle displayed on the transverse plane.
to its capacity of displaying both bone components and soft tissues Four sequences were used in this study and the imaging parameters
with a superior tissue contrast (Schmitter et al. 2005). In equine are shown in Table 1. MRI analysis software (ADW 4.1)1 was
medicine, MRI has been mainly used for orthopaedics (Kleiter employed for image interpretation and cross-referencing. Signal
et al. 1999; Widmer et al. 2000; Murray et al. 2006) and head intensity and homogeneity, definition of anatomical margins, and
evaluation (Chaffin et al. 1997; Arencibia et al. 2000, 2001; Ferrell
et al. 2002), but there are no references concerning MRI study of
the equine TMJ anatomy.
The objectives of this study were to describe in detail the
normal sectional MRI anatomy of the equine TMJ by using frozen
and plastinated anatomical sections as reference, and to determine
the best imaging planes and sequences to visualise the TMJ
components (both soft tissues and bone).

Materials and methods

Animals

Whole heads were collected from 6 Spanish Purebred horses (4

immature and 2 mature horses) that were subjected to euthanasia
for medical reasons unrelated to mouth disturbances, including
dental and masticatory problems as well as TMJ disorders (Weller
et al. 2001; Ramzan et al. 2008). The age range was 1–15 years and
weight 213–564 kg bwt. Each head was disarticulated at the
atlanto-occipital joint. From ultrasonography of the joints, joint
effusion or bony irregularities of their lateral aspect were not
visualised.

Magnetic resonance imaging examination

MRI examination was carried out using a 1 Tesla magnet (Signa
Horizon LX)1 with a 7.62 cm radiofrequency single surface coil1
placed on the lateral aspect of the TMJ region. Heads were held
vertically on the table as in a standing live horse. Each joint was
scanned separately, one TMJ first and the other one later, within 2 h
of euthanasia to reduce the interference of post mortem changes on
the image quality. The protocol followed by this study was based on
standard protocols described for evaluating human TMJs in a closed-
mouth position (Vilanova et al. 2007). Scan slice orientations were
Fig 1: Oblique sagittal spin-echo T1-weighting image of the left
temporomandibular joint. The edge of the articular tubercle, retroarticular
process and disc has been outlined. The yellow line (a) is the baseline that
joins both edges of the retroarticular process and articular tubercle. The
yellow spot (b) is the midpoint of the mandibular condyle. The red line (c)
is perpendicular to the line a and passes through the midpoint of the
condyle. The blue line (d) is tangential to the caudal margin of the disc and
passes through the midpoint of the condyle. a is the angle between the line
c and d and shows the position of the disc relative to the mandibular
condyle. D = dorsal. M = medial. R = rostral.

© 2010 EVJ Ltd

202 MRI-Equine TMJ

relationship with adjacent structures were described for TMJ
components such as articular surfaces, articular disc, articular
cartilage, capsule and ligaments.
The position of the disc relative to the mandibular condyle
was studied in each TMJ on SE T1W sagittal images following
the guideline reported by Tominaga et al. (2007) in humans. The
reference images to obtain data were those where the edge of the
retroarticular process and articular tubercle of the zygomatic
process of the temporal bone, and the caudal margin of the disc
were well discriminated (Fig. 1). A base line was drawn between
the edge of the retroarticular process and articular tubercle of the
zygomatic process of the temporal bone. The midpoint of the
mandibular condyle was marked. Then, a line perpendicular to
the base line passing through the midpoint of the condyle was
drawn; a line tangentially to the caudal margin of the disc
crossing through the midpoint of the condyle was also drawn. The
angle between these 2 lines was measured to determine the disc
position.
Anatomical study
After imaging the specimens, blood vessels and synovial pouches
of 2 heads were injected with coloured latex (Rodríguez et al.
2006). Blocks containing only the TMJ and adjacent structures
were cut off from all 6 frozen heads (-30°C for 48 h).
These blocks were then frozen at -70°C for 7 days and sliced into
contiguous 4-mm thick sections in the same planes as MRI
images: oblique sagittal, oblique transverse or oblique
dorsal planes (4 articulations per plane). Each section was
photographed and plastinated (Latorre et al. 2003) to facilitate the
interpretation and comparison of the anatomical structures to the
corresponding MRI images. The identified structures were
labelled according to an internationally accepted nomenclature
(Anon 2005).
Results
The quality of the MR images obtained from all specimens was
good, resulting in a detailed visualisation of TMJ components,
which were accurately matched with the corresponding anatomic
sections. The signal intensities displayed by the TMJ components
and adjacent structures in the different MRI sequences used are
shown in Table 2.
On the oblique sagittal images, the majority of the TMJ
components were visualised (Figs 2b–f). These images also
allowed to study the relationship of TMJ components to vital
adjacent structures namely the ear, guttural pouch, nerves and
vessels (Figs 2b,c).
The oblique transverse images allowed better assessment of
the mandibular condyle and the lateromedial axis of the joint
(Figs 3b,c). On these images, the lateral ligament and the lateral
insertion of the disc to the capsule and mandibular condyle were
identified. The articular cartilage was difficult to identify on the
lateral aspect of the joint.
On the oblique dorsal images (Figs 4b,c), the lateromedial and
rostrocaudal axis of the mandibular condyle were assessed
simultaneously. This plane allowed to identify the lateral and
caudal ligaments; however, the rest of the TMJ components, mainly
the articular cartilage and the zygomatic process of the temporal
bone, were more difficult to asses.
On spin-echo T1-weighting (SE T1W) sequence (Fig 2b), bone
marrow showed high signal intensity with a granular appearance
owing to the fatty infiltration. The subchondral bone and the
cortical had a uniform homogeneous surface.
The disc displayed homogeneous low signal intensity in the
rostral and caudal portions and intermediate signal intensity in the
thin central portion. The caudal fibrous expansion of the disc was
identified as a thin band with low signal intensity located between
the retroarticular process and mandibular condyle. The transition
between the caudal portion of the disc and the caudal fibrous
expansion was observed as an intermediate signal intensity
vertical line. This finding was used as a reference to mark the
caudal margin of the disc which allowed measurement of the
angle formed between the disc and the mandibular condyle; the
angle differed from 20–30° in the TMJs studied. The attachment
of the rostral portion of the disc to the capsule displayed a low
signal intensity defining 2 separated dorsal and ventral synovial
pouches in which the synovial fluid had lower signal intensity
than the synovium.

TABLE 2: Signal intensity showed by the TMJ components and adjacent structures in the MRI sequences used for this study

Structures SE T1W T2*W FS PDW FSE T2W

1. Cortical and subchondral Signal void Signal void Signal void Signal void
bone
2. Bone marrow Hyperintense Intermediate signal intensity Hypointense Intermediate signal intensity
3. Articular cartilage Low-intermediate signal intensity Hyperintense Hyperintense Intermediate signal intensity
4. Articular disc Hypointense (in poles) intermediate Low-intermediate signal Hypointense Hypointense
signal intensity (in central portion) intensity
5. Synovial fluid Low-intermediate signal intensity Hyperintense Hyperintense Hyperintense
7. Synovium Intermediate signal intensity Hyperintense Hyperintense Hyperintense
8. Intra-articular fat tissue Hyperintense Intermediate signal intensity Hypointense Intermediate signal intensity
9. Capsule Hypointense Hypointense Hypointense Hypointense
10. TMJ ligaments Intermediate signal intensity Intermediate signal intensity Intermediate signal Intermediate signal intensity
intensity
11. Vessels Signal void Signal void Signal void Signal void
12. Masticatory muscles Intermediate signal intensity High-intermediate signal Hypointense Low-intermediate signal
intensity intensity
13. Nerves Hypointense Intermediate signal intensity – Intermediate signal intensity

SE T1W = spin-echo T1-weighting; T2* = gradient echo T2-weighting; FS PDW = fat-suppressed fast proton density-weighting; FSE T2W = fast spin-echo
T2-weighting; TMJ = temporomandibular joint.

© 2010 EVJ Ltd

M. J. Rodríguez et al. 203

d)
a)

e)

b)

f)

c)

Fig 2: a) Transverse image of the left temporomandibular joint (TMJ) displaying the sagittal-sectional planes use as reference for b–f (plane I shown in
green; plane II shown in red). b) Oblique sagittal spin-echo T1-weighting image of the left TMJ of a mature horse obtained at the level of plane I. c) Oblique
sagittal anatomical section of the left TMJ obtained at the level of plane I. d) Oblique sagittal T2*W image of the left TMJ of an immature horse obtained
at the level of plane II. e) Oblique sagittal fat-suppressed proton density-weighting image of the left TMJ of an immature horse obtained at the level of plane
II. f) Oblique sagittal anatomical section of the left TMJ of an immature horse obtained at the level of plane II. 1 = temporal muscle; 2 = zygomatic process
of the temporal bone (bone marrow); 3 = articular tubercle; 4 = retroarticular process; 5 = mandibular condyle; 6 = cortical bone; 7 = subchondral bone;
8 = articular cartilage; 9 = articular disc; 10 = caudal fibrous expansion of the articular disc; 11 = articular capsule; 12 = caudal ligament of the TMJ;
13 = guttural pouch; 14 = parotid salivary gland; 15 = temporal venous sinus; 16 = maxillary artery; 17 = maxillary vein; 18 = masseteric vein (and its
branches); 19 = masticatory nerve; 20 = lateral pterygoid muscle; 21 = external auditory meatus; 22 = caudal compartment of the dorsal synovial pouch;
23 = rostral compartment of the dorsal synovial pouch; 24 = rostral compartment of the ventral synovial pouch; 25 = rostrolateral attachment of the disc
onto the capsule; 26 = transverse facial vein. B = brain; C = mandibular condyle (TMJ); D = dorsal; M = medial; R = rostral.

© 2010 EVJ Ltd

204
a)
b)
The articular capsule was visualised as a thin hypointense
strip barely discernible from adjacent structures. The caudal
ligament was easily identified as an intermediate signal intensity
thick band delimited by 2 hypointense lines. However, the lateral
ligament was not visualised on this plane. The parotid salivary
gland produced heterogeneous signal intensity as a result of the
low signal intensity of the glandular tissue and the high signal
intensity of the fatty infiltration. The masticatory muscles were
observed with intermediate signal intensity (higher than the disc
and lower than the articular cartilage) showing various shades of
grey. Nerves were difficult to distinguish from adjacent tissues,
but the surrounding fat enhanced their low signal intensity.
Gradient echo T2-weighting (T2*W) and fat-suppressed
(FS) fast proton density-weighting (PDW) images showed poor
signal to noise ratio, exhibiting lower anatomical detail than the
SE T1W images, although their tissue contrast was superior
(Figs 2d,e). The higher signal intensity shown by the articular
cartilage made easier its differentiation from the articular disc and
allowed the measurement of articular cartilage thickness, which
was 1 mm thick in the 2 mature animals (Fig 2b) and 3 mm thick
in the 4 immature horses (Figs 2d,e). The disc morphology and
caudal fibrous expansion were outlined by the high signal
© 2010 EVJ Ltd
MRI-Equine TMJ
c)
Fig 3: a) Sagittal image of the left temporomandibular
joint (TMJ) displaying the transverse-sectional plane
used as reference for b and c. b) Oblique transverse
fast spin-echo T2-weighting image of the left TMJ of a
mature horse obtained at the level of the plane. c)
Oblique transverse anatomical section of the left TMJ
of a mature horse obtained at the level of the plane.
1 = squamous portion of the temporal bone; 2 =
zygomatic process of the temporal bone; 3 =
mandibular condyle (bone marrow); 4 = ramus of the
mandible; 5 = cortical bone; 6 = subchondral bone; 7
= articular cartilage; 8 = articular disc; 9 = caudal
compartment of the dorsal synovial pouch (intra-
articular fat tissue); 10 = articular capsule; 11 =
parotid salivary gland; 12 = guttural pouch;
13 = parotidoauricular muscle; 14 = masseter muscle;
15 = medial pterygoid muscle; 16 = lateral pterygoid
muscle; 17 = temporal muscle; 18 = transverse facial
vessels; 19 = maxillary vein; 20 = pterygoid venous
plexus; 21 = ventral petrous venous sinus; 22 = facial
nerve; 23 = mandibular nerve. D = dorsal; M = medial;
R = rostral; C = mandibular condyle; T = zygomatic
process of the temporal bone.
intensity of the dorsal and ventral synovial pouches where the
fluid and synovium appeared indistinguishable. The parotid
salivary gland and the masticatory muscles appeared with similar
though slightly brighter signal intensity than on the SE T1W
sequence.
Fast spin-echo T2-weighting (FSE T2W) images (Figs 3b, 4b)
showed better anatomical detail but poorer contrast than T2*W and
FS PDW images. The bone marrow was visualised less bright than
in SE T1W images. The articular cartilage showed intermediate
signal intensity. The synovium and the synovial fluid were
observed with a high signal intensity but less bright than on T2*W
and FS PDW images.
Discussion
The quality of images obtained in this study was excellent due to
their high signal-to-noise ratio acquired by the use of adjusted
imaging parameters (Stoller and Jacobson 1997; Stehling et al.
2007). Consequently, the SE T1W sequence was the best one to
assess the articular and peri-articular anatomy as reported in
previous human TMJ studies (Kober et al. 2007). The FSE T2W
sequence provided simultaneously an acceptable anatomical
M. J. Rodríguez et al.
a)
b)
resolution and sensitivity to pathology in a shorter scan time
(Westbrook and Kaut 2000). The reduced quality exhibited by
T2*W and FS PDW images could be due to the lower signal-to-
noise ratio achieved even using these adjusted imaging parameters
(Chaffin et al. 1997; Blaik et al. 2000). However, this deficiency
was accepted because of the high contrast shown by these
sequences, which were considered the best choice to evaluate the
disc, the synovial fluid, the synovial pouches and the articular
cartilage.
The oblique sagittal and transverse planes were the most
informative anatomical planes. The articular cartilage, the articular
disc and the synovial pouches were better studied on the oblique
sagittal plane. Indeed, the articular cartilage was hardly visualised
on the lateral aspect of the joint probably due to the obliquity of the
image. Nevertheless, both planes complemented each other when
investigating other TMJ components (e.g. bone surfaces, ligaments
and capsule), adjacent structures and their relationship. These
findings were similar to descriptions reported in humans (Schmitter
et al. 2005; Vilanova et al. 2007).
Early MRI anatomic studies on equine specimens have
confirmed the valid use of cadavers for representing clinical
205
c)
Fig 4: a) Transverse image of the left temporomandibular joint (TMJ)
displaying the dorsal-sectional plane used as reference for b and c. b)
Oblique dorsal fast spin-echo T2-weighting image of the left TMJ of
a mature horse obtained at the level of the plane. c) Oblique dorsal
anatomical section of the left TMJ of a mature horse obtained at the
level of the plane. 1 = mandibular condyle; 2 = coronoid process of
the mandible; 3 = cortical bone; 4 = articular capsule (outlined in
white); 5 = caudal ligament of the TMJ (outlined in blue); 6 = lateral
ligament of the TMJ; 7 = lateral pterygoid muscle; 8 = medial
pterygoid muscle; 9 = masseter muscle; 10 = temporohyoid joint; 11
= guttural pouch; 12 = parotid salivary gland; 13 = masseteric vein
(and its branches); 14 = pterygoid venous plexus. D = dorsal; R =
rostral; M = medial; B = brain; C = mandibular condyle.
conditions (Blaik et al. 2000; Arencibia et al. 2001; Murray et al.
2006). Signal intensities of some TMJ components observed in our
study differed from those described in human patients and horses.
The principal difference found was the appearance of several
vessels, which were of signal void in all sequences probably caused
by the air content. Nevertheless, some small vessels displayed high
signal intensity due to clotted blood, similarly to findings reported
by Kleiter et al. (1999) and Latorre et al. (2006) for MRI of the
equine extremities. These differences could be mainly attributed to
post mortem changes as it has been mentioned previously in other
imaging studies for equine specimens (CT, Morrow et al. 2000). In
accordance with human TMJ researches, these findings must be
identified since vessels in live horses will be displayed hiper- or
hypointense, depending on the sequences used, due to their blood
flow (Stoller and Jacobson 1997; Larheim et al. 2001).
Other little variations in signal intensity were observed in SE
T1W images. The articular cartilage showed low-intermediate
signal intensity differing from the low signal intensity described
in humans (Stoller and Jacobson 1997) and the high signal
intensity reported in horses for limbs (Kleiter et al. 1999; Widmer
et al. 2000). Synovial fluid displayed low to intermediate signal
© 2010 EVJ Ltd
206
intensity, conversely to the low signal intensity displayed in
human TMJ (Larheim et al. 2001) and equine limbs (Blaik et al.
2000) MRI studies. These little differences could be attributed to
variations in the technical support applied and the use of cadaver
specimens and their manipulation before scanning. In our study,
heads were collected and scanned within 2 h post mortem in spite
of after freezing the specimens, as reported by previous authors
(Kleiter et al. 1999; Widmer et al. 2000). Immature specimens
showed a thicker articular cartilage (3 mm) in the MR images,
which was in accordance with previous results reported by Weller
et al. (1999b) and Rodríguez et al. (2007) for the equine TMJ
ultrasonography. The articular cartilage thickness for mature
horses reported here was similar to the descriptions in human
adult patients (Larheim et al. 2001). In the anatomical
sections, the articular cartilage thickness and the size of the
synovial pouches were smaller than in MRI images due to the
plastination process. The different signal intensities showed by
the portions of the articular disc have been also described in
previous human reports (Vilanova et al. 2007). This heterogeneity
is probably due to the variations in the disc histological
composition (elastic fibres in the central portion since it is the
weight-bearing zone and fibrous tissue of the bilaminar
retrodiscal zone) (humans: Stoller and Jacobson 1997; horses:
Rodríguez et al. 2006).
In the present study, TMJs were imaged individually in an
upright position, unlike in humans where simultaneous bilateral
joint scanning is possible thanks to dual coil designs (Vilanova
et al. 2007). However, these devices are not currently suitable for
horses because of their head size. The upright position used in this
research was selected to facilitate the manipulation of the
specimins. This approach could not be reproduced in live horses
due to the current gantry designs, which involve some physical
difficulties while placing the head and coils rightly within the
magnet; therefore, in clinical conditions, the horse can be
positioned in lateral recumbency. The use of general anaesthesia
restricts the scan time; however, the acquisition time for sequences
in the present study was considered acceptable for application in a
clinical context (less than 40 min per TMJ).
Dynamic explorations are performed routinely in MRI studies
of the human TMJ in order to evaluate the disc position (Drace and
Enzmann 1990). That is not the case for anaesthetised horses
hence, the position of the disc in relation to the mandibular condyle,
measured as an angle, could be used as a good indicator of a normal
condition in this species. In human medicine, this angle is used to
assess the disc position as well as to quantify the degree of disc
displacement which is one of the most common TMJ disorder
(Larheim et al. 2001). Our measurements reproduced an angle
range of 20–30°. This result differs from human studies (Tominaga
et al. 2007) where a normal disc position is considered when the
posterior band of the disc is no larger than 10° from the 12 o’clock
position (Drace and Enzmann 1990); this angle variation between
both species could be due to their different TMJ conformation. To
the authors’ knowledge, there is no published information
concerning the normal disc position in horses. Therefore, further
studies are needed to establish the significance of the disc position
in equine TMJ pathologies.
MRI is considered in human medicine as the prime diagnostic
noninvasive modality of choice for studying patients suspicious for
TMJ pathology (Stoller and Jacobson 1997). In equine medicine,
MRI is now widely used thanks to the overcoming of some
procedural difficulties such as size limitations, availability of MR
© 2010 EVJ Ltd
MRI-Equine TMJ
units, suitable tables and radiofrequency coils (Chaffin et al. 1997).
However, no previous publications have been found where MRI has
been used to study and/or diagnose TMJ pathologies. The authors
believe that MRI could be helpful in this purpose, as it is in human
medicine where MRI has demonstrated a high sensitivity and
specificity in the evaluation of TMJ, being 95% accurate in
assessing disc position and morphology and 93% accurate in
assessing bone changes (Stoller and Jacobson 1997; Shimazaki
et al. 2007). Logically, further studies are required in normal and
abnormal live horses to improve clinical diagnostic information
concerning this method and TMJ conditions. To support this, a
thorough knowledge of the interpretation principles and TMJ
anatomy is essential (Tucker and Farrell 2001; Latorre and
Rodríguez 2007).
In conclusion, MRI is a promising imaging modality for
evaluation of the equine TMJ. The correlation of MRI images with
anatomical sections confirmed that the TMJ components could be
assessed using the protocol described in this study. T2*W and FS
PDW sequences in an oblique sagittal and transverse anatomical
plane are the best choice to evaluate the articular cartilage, the
articular disc and synovial pouches. The information provided in
this paper should aid clinicians in the interpretation of MR images
of equine TMJ, contributing to a better understanding of the joint
and helping in the early diagnosis of those problems that could not
be diagnosed by other means.
Manufacturer’s address
1
General Electric Co., Waukesha, Wisconsin, USA.
References
Anon (2005) Nomenclature World Association of Veterinary Anatomists (W.A.V.A.)
Nomina anatomica veterinaria. 5th Editorial Committee Hannover, Columbia,
Gent, Sapporo: International Committee on Veterinary Gross Anatomical
Nomenclature. pp 1-190.
Arencibia, A., Vazquez, J.M., Jaber, R., Gil, F., Ramírez, J.A., Rivero, M., González,
N. and Wisner, E.R. (2000) Magnetic resonance imaging and cross sectional
anatomy of the normal equine sinuses and nasal passages. Vet. Radiol. Ultrasound
41, 313-319.
Arencibia, A., Vazquez, J.M., Ramírez, J.A., Ramírez, G., Vilar, J.M., Rivero, M.A.,
Alayon, S. and Gil, F. (2001) Magnetic resonance imaging of the normal equine
brain. Vet. Radiol. Ultrasound 42, 405-408.
Blaik, M.A., Hanson, R.R., Kincaid, S.A., Hatchcock, J.T., Hudson, J.A. and Baird,
D.K. (2000) Low-field magnetic resonance imaging of the equine tarsus: normal
anatomy. Vet. Radiol. Ultrasound 41, 131-141.
Bonin, S.J., Clayton, H.M., Lanovaz, J.L. and Johnson, T.J. (2006) Kinematics of the
equine temporomandibular joint. Am. J. vet. Res. 67, 423-428.
Carmalt, J.L. and Wilson, D.G. (2005) Arthroscopic treatment of temporomandibular
joint sepsis in a horse. Vet. Surg. 34, 55-58.
Chaffin, M.K., Walker, M.A., McArthur, N.H., Perris, E.E. and Matthews, N.S. (1997)
Magnetic resonance imaging of the brain of normal neonatal foals. Vet. Radiol.
Ultrasound 38, 102-111.
Cooks, W.R. (2006) Correspondence to the editor. Equine vet. J. 38, 361.
Devine, D.V., Moll, H.D. and Bahr, R.J. (2005) Fracture, luxation, and chronic septic
arthritis of the temporomandibular joint in a juvenile horse. J. vet. Dent. 22, 96-99.
Drace, J. and Enzmann, D. (1990) Defining the normal temporomandibular joint:
closed-, partially open-, and open-mouth MR imaging of asymptomatic subjects.
Radiol. 177, 67-71.
Ferrell, E.A., Gavin, P.R., Tucker, R.L., Sellon, D.C. and Hines, M.T. (2002) Magnetic
resonance for evaluation of neurologic disease in 12 horses. Vet. Radiol.
Ultrasound 43, 510-516.
Hardy, J. and Shiroma, J. (1991) What is your diagnosis? Rostral luxation of the right
temporomandibular joint. J. Am. vet. med. Ass. 198, 1663-1664.
M. J. Rodríguez et al.
Kleiter, M., Kneissl, S., Stanek, Ch., Mayrhofer, E., Baulain, U. and Deegen, E. (1999)
Evaluation of magnetic resonance imaging techniques in the equine digit. Vet.
Radiol. Ultrasound 40, 15-22.
Kober, C., Hayakawa, Y., Kinzinger, G., Gallo, L., Otonari-Yamamoto, M., Sano, T.
and Sader, R.A. (2007) 3D-visualization of the temporomandibular joint with
focus on the articular disc based on clinical T1-, T2-, and proton density weighted
MR images. Int. J. Cars. 2, 203-210.
Larheim, T.A., Westesson, P.-L. and Sano, T. (2001) MR grading of
temporomandibular joint fluid: association with disk displacement categories,
condyle marrow abnormalities and pain. Int. J. Oral Maxillofac. Surg. 30, 104-
112.
Latorre, R., Arencibia, A., Gil, F., Rivero, M., Henry, R.W., Ramírez, G. and Vazquez,
J.M. (2006) Correlation of magnetic resonance images with anatomic features of
the equine tarsus. Am. J. vet. Res. 67, 756-761.
Latorre, R., Arencibia, A., Gil, F., Rivero, M., Ramírez, G., Vázquez-Autón, J.M. and
Henry, R.W. (2003) P-40 and S-10 plastinated slices: an aid to interpreting MR
images. J. Int. Soc. Plast. 18, 14-22.
Latorre, R. and Rodríguez, M.J. (2007) In search of clinical truths: equine and
comparative studies of anatomy. Equine vet. J. 39, 263-268.
May, K.A., Moll, H.D., Howard, R.D., Pleasant, R.S. and Gregg, J.M. (2001)
Arthroscopic anatomy of the equine temporomandibular joint. Vet. Surg. 30,
564-571.
Morrow, K.L., Park, R.D., Spurgeon, T.L., Stashak, T.S. and Arceneaux, B. (2000)
Computed tomographic imaging of the equine head. Vet. Radiol. Ultrasound 41,
491-497.
Murray, R.C., Schramme, M.C., Dyson, S.J., Branch, M.V. and Blunden, T.S. (2006)
Magnetic resonance imaging characteristics of the foot in horses with palmar foot
pain and control horses. Vet. Radiol. Ultrasound 47, 1-16.
Nagy, A.D. and Simhofer, H. (2006) Mandibular condylectomy and meniscectomy for
the treatment of septic temporomandibular joint arthritis in a horse. Vet. Surg. 35,
663-668.
Ramzan, P.H., Marr, C.M., Meehan, J. and Thompson, A. (2008) Novel oblique
radiographic projection of the temporomandibular articulation of horses. Vet. Rec.
162, 714-716.
Ramzan, P.H.L. (2006) The temporomandibular joint: component of clinical
complexity. Equine vet. J. 38, 102-104.
Rodríguez, M.J., Agut, A., Gil, F. and Latorre, R. (2006) Anatomy of the equine
temporomandibular joint: study by gross dissection, vascular injection and
section. Equine vet. J. 38, 143-147.
Rodríguez, M.J., Latorre, R., López-Albors, O., Soler, M., Aguirre, C., Vázquez,
J.M., Querol, M. and Agut, A. (2008) Computed tomographic anatomy
of the temporomandibular joint in the young horse. Equine vet. J. 40, 566-571.
Rodríguez, M.J., Soler, M., Latorre, R., Gil, F. and Agut, A. (2007) Ultrasonographic
anatomy of the temporomandibular joint in healthy Pure Bred Spanish horses. Vet.
Radiol. Ultrasound 48, 149-154.
Schmitter, M., Kress, B., Ludwig, C., Koob, A., Gabbert, O. and Rammelsberg, P.
(2005) Temporomandibular joint disk position assessed at coronal MR imaging in
asymptomatic volunteers. Radiol. 236, 559-564.
207
Schumacher, J. (2001) The present state of equine dentistry. Equine vet. J. 33, 2-1.
Shimazaki, Y., Saito, K., Matsukawa, S., Onizawa, R., Kotake, F., Nishio, R. and Abe,
K. (2007) Image quality using dynamic MR imaging of the temporomandibular
joint with true-FSIP sequence. Magn. Reson. Med. Sci. 6, 15-20.
Stehling, C., Vieth, V., Bachmann, R., Nassenstein, I., Kugel, H., Kooijman, H.,
Heindel, W. and Fischbach, R. (2007) High-resolution magnetic resonance
imaging of the temporomandibular joint: image quality at 1.5 and 3.0 Tesla in
volunteers. Invest. Radiol. 42, 428-434.
Stoller, D.W. and Jacobson, R.L. (1997) The temporomandibular joint. In: Magnetic
Resonance Imaging in Orthopaedics & Sports Medicine, 2nd edn., Ed: D.W.
Stoller, Lippincott-Raven Publishers, Philadelphia. pp 995-1020.
Tominaga, K., Konoo, T., Morimoto, Y., Tanaka, T., Habu, M. and Fukuda, J. (2007)
Changes in temporomandibular disc position during growth in young Japanese.
Dentomaxillofac. Radiol. 36, 397-401.
Townsend, N.B. and Cotton, J.C. (2007) A novel radiographic technique for
investigation of the equine temporomandibular joint. Clinical Research. In:
Proceedings of the 46th Congress of the British Equine Veterinary Association,
Equine Veterinary Journal Ltd, Newmarket. p 424.
Tucker, R.L. and Farrell, E. (2001) Computed tomography and magnetic resonance
imaging of the equine head. Vet. Clin. N. Am: Equine Pract. 17, 131-144.
Vilanova, J.C., Barceló, J., Puig, J., Remollo, S., Nicolau, C. and Bru, C. (2007)
Diagnostic imaging: magnetic resonance imaging, computed tomography, and
ultrasound. Semin. Ultrasound CT MRI 28, 184-191.
Warmerdam, E., Klein, W. and van Herpen, B. (1997) Infectious temporomandibular
joint disease in the horse: computed tomographic diagnosis and treatment of two
cases. Vet. Rec. 141, 172-174.
Weller, R., Cauvin, E.R., Bowen, I.M. and May, S.A. (1999a) Comparison of
radiography, scintigraphy and ultrasonography in the diagnosis of a case of
temporomandibular joint arthropathy in a horse. Vet. Rec. 144, 377-379.
Weller, R., Livesey, L., Maierl, J., Nuss, K., Bowen, I.M., Cauvin, E.R.J., Weaver, M.,
Schumacher, J. and May, S.A. (2001) Comparison of radiography and scintigraphy
in the diagnosis of dental disorders in the horse. Equine vet. J. 33, 49-58.
Weller, R., Maierl, J., Bowen, I.M., May, S.A. and Liebich, H.G. (2002) The
arthroscopic approach and intra-articular anatomy of the equine
temporomandibular joint. Equine vet. J. 34, 421-424.
Weller, R., Taylor, S., Maierl, J., Cauvin, E. and May, S. (1999b) Ultrasonographic
anatomy of the equine temporomandibular joint. Equine Vet. J. 31, 529-532.
Westbrook, C. and Kaut, C. (2000) Pulse sequences. In: MRI in Practice, 2nd edn.,
Blackwell Science Ltd., London. pp 104-140.
Widmer, W.R., Buckwalter, K.A., Fessler, J.F., Hill, M.A., VanSickle, D.C. and
Ivancevich, S. (2000) Use of radiography, computed tomography and magnetic
resonance imaging for evaluation of navicular syndrome in the horse. Vet. Radiol.
Ultrasound 41, 108-116.
Author contributions The initiation, conception, planning and
writing of this study were by M.J.R., A.A. and R.L. All
authorscontributed to the execution.
© 2010 EVJ Ltd