266EQUINE VETERINARY JOURNAL
Equine vet. J. (2006) 38 (3) 266-270
Reasons for performing study: An alternative technique of
radiographing the pelvis in the standing horse is required, to
avoid the risks associated with general anaesthesia.
Hypothesis: That lateral oblique radiography in the standing
horse would be a useful technique in the investigation of
Objectives: To describe the technique of lateral oblique pelvic
radiography in the standing horse and demonstrate the
feasibility and usefulness of this technique.
Methods: A technique for lateral oblique radiography in the
standing horse was devised and retrospective review made of
radiographic findings in 18 clinical cases.
Results: The caudal iliac shaft, greater trochanter of the
femur, femoral head, acetabulum and coxofemoral
articulation on the side under investigation were visualised
consistently using this technique. Of the 18 cases, 3 iliac
shaft fractures, 1 acetabular fracture, 2 coxofemoral
luxations and 4 horses with new bone formation
around the coxofemoral joint and/or proximal femur
Conclusions: Lateral oblique radiography in the standing,
conscious horse can be used to investigate conditions
affecting the caudal iliac shaft, coxofemoral articulation and
proximal femur in the horse.
Potential relevance: The technique is straightforward,
noninvasive and useful in the investigation of horses with
suspected pelvic injury. However, not all pelvic injuries
would be identified, and normal radiographic findings do
not rule out injury or fractures elsewhere in the pelvis.
Indications for equine pelvic radiography include investigation of
hindlimb lameness in the absence of detectable abnormalities or
response to local analgesia in the lower limb, asymmetry of the
pelvic region and the presence of crepitus in the pelvic region.
The most common indication is in the definitive diagnosis of
suspected pelvic fractures, which often have a nonspecific
clinical presentation. Although affected horses typically
demonstrate significant hindlimb lameness, the presence of
crepitus and/or abnormal findings on rectal examination have
been found to be inconsistent (Jeffcott 1982; Hendrickson 1987;
Little and Hilbert 1987).
Radiography is reported as a reliable diagnostic aid for equine
pelvic fractures (Little and Hilbert 1987). Radiographic
techniques include ventrodorsal radiographs obtained with the
horse in dorsal recumbency under general anaesthesia (Lewis and
Heinze 1971; Kangstrom 1972), and ventrodorsal (May et al.
1991) and lateral (Jeffcott 1979; Little and Hilbert 1987)
radiographs in the standing horse.
Additional risks associated with general anaesthesia of horses
with suspected pelvic fracture are the potential for fracture
displacement and/or haemorrhage due to laceration of internal iliac
arteries during recovery. May et al. (1991) described a technique of
obtaining ventrodorsal radiographs of the standing horse with the
x-ray machine positioned ventral to the abdomen and the cassette
positioned dorsal to the sacrum. This technique avoids the risks
associated with recovery from general anaesthesia, but there are
some disadvantages. Centring and collimation of the x-ray beam
and safe handling of the cassette can be difficult. The technique
also poses some risk to the x-ray machine in fractious horses.
Lateral radiographs of the pelvis of the standing horse are
considered to be of little value (Little and Hilbert 1987) except in
the identification of some sacral fractures (Hendrickson 1987).
Scintigraphy is highly sensitive but nonspecific for bone
pathology. With its ability to image the whole skeleton in the
standing sedated horse, it has had a major impact in the diagnosis of
pelvic fractures (Driver 2003). However, there are several factors
that may influence its sensitivity, including attenuation of the
gamma-photons by the pelvic muscle mass and the position of the
urinary bladder (Hendrickson 1987; Pilsworth et al.1993; Hornof et
al. 1996; Attenburrow 1997; Erichsen and Berger 2001). Bone
scanning the horse too early following the originating incident also
leads to false negatives, presumably because the ‘up-regulation’ of
bone turnover has not yet occurred. It is commonly recommended
to wait at least 7 days before performing a bone scan when
investigating a suspected fracture. It is also possible to produce false
negative images of the pelvis when a fracture is present without a
definitive underlying cause, up to 14 days after injury (A.J. Driver,
unpublished data). Scintigraphy is the technique of choice for the
diagnosis of suspected stress fractures (Hornof et al. 1996).
Ultrasonography is also useful in the assessment of pelvic
damage, but is limited to the detection of discontinuity of bony
surfaces and identification of soft tissue changes (Shepherd and
Pilsworth 1994; Busoni 2001; Tomlinson et al. 2000, 2001;
Almanza and Whitcombe 2003).
This article describes a technique for obtaining lateral
oblique views of the pelvis in a standing, conscious horse.
A technique for pelvic radiography in the standing horse
E. L. BARRETT*, A. M. TALBOT, A. J. DRIVER, F. J. BARR and A. R. S. BARR
Department of Clinical Veterinary Science, University of Bristol, Langford House, Langford, Bristol BS40 5DU; and Greenwood Ellis and
Partners, 166 High Street, Newmarket CB8 9WS
Keywords: horse; pelvis; radiography; pelvic fracture; acetabulum
*Author to whom correspondence should be addressed.
[Paper received for publication 25.08.05; Accepted 28.11.05]
E. L. Barrett et al.
Materials and methods
The results of lateral oblique radiographic views obtained in
18 horses at the Equine Diagnostic Centre, University of Bristol,
UK, between March 2000 and June 2005 were evaluated.
Lateral oblique radiography (lateral 30° dorsal-lateroventral
oblique) was carried out with the horse standing. In fractious
animals, sedation was helpful. The rectum was evacuated where
possible; an air-filled rectum provides good contrast for the
visualisation of the bony structures and avoids the potential for
artefacts created by overlying faecal material. The horse was, as
far as possible, standing squarely on all 4 limbs. A vertically
positioned x-ray cassette (X-Omatic)1was positioned against the
side of the pelvis under examination and the x-ray tube angled
approximately 30° ventrally from horizontal. It was centred
between the level of the greater trochanter and the base of the tail
on the tube side, approximately two-thirds along the craniocaudal
distance between the palpable landmarks of the tuber sacrale and
tuber ischiae (Fig 1). The height and craniocaudal position of the
cassette were adjusted to coincide with the path of the x-ray
beam. Rare earth screens (Lanex Regular)1and a stationary
parallel grid (parallel 12:1 grid)2were used, with a film-focal
distance of approximately 100 cm. For the x-ray machine
(UD150B-10 generator, maximum exposure factors 150 kV and
1000 mA, with Comet tube, Argostat tube crane and Bucky
system)2and film (MG SR Medical Film)3-screen-grid
combination in our clinic, the settings used varied from 90 to
130 kV and 125 to 400 mAs, depending on the size of the horse
and the pelvic muscle mass.
The lead strips within the parallel grid were orientated
vertically, so that the dorsoventral angulation of the x-ray beam
did not result in the phenomenon of grid cut-off. Due to the
relatively large exposure factors used, cassettes were not held by
hand but by a gantry-mounted holder. However, a cassette
mounted or suspended from a vertical stand would be equally
suitable. To minimise the radiation hazard to personnel, only the
person holding the horse’s head was present at the time of
exposure. Lead protective clothing is essential to protect
personnel from the scattered radiation associated with the high
Areas of the pelvis visualised during standing lateral oblique pelvic
radiography included the iliac shaft, greater trochanter of the femur,
femoral head, acetabulum and coxofemoral articulation on the side
under examination (Fig 2). Depending on the craniocaudal centring
of the x-ray beam, the ischium was also visualised. The structures on
the side being imaged were projected dorsally, ideally superimposed
on a gas-filled rectum (Fig 3). Areas of the pelvis that may not be
Fig 2: Areas of the pelvis visualised during standing lateral oblique
Fig 1: Positioning the x-ray tube and cassette for standing lateral oblique
radiography of the equine pelvis.
Fig 3: a) Normal standing lateral oblique view of the equine pelvis.
b) The same image with diagram of visible structures overlaid.
268A technique for pelvic radiography in the standing horse
seen using this technique include the iliac wings, tubera coxae,
tubera sacrale, sacroiliac joints, sacrum, pubis and entire ischium.
Eighteen horses underwent standing lateral oblique pelvic
radiography at the Equine Diagnostic Centre, University of Bristol
between March 2000 and June 2005. A variety of breeds were
examined, with bodyweight 125–550 kg (Table 1). Six horses
(Cases 1, 5, 7, 10, 11 and 16) were unable fully to bear weight on
the affected limb. Diagnostic quality radiographs were achieved in
Iliac shaft fractures (Fig 4) were diagnosed conclusively in
3 horses and an acetabular fracture in one horse. In 3 of these horses,
the findings were confirmed with ultrasonography, scintigraphy
and/or further radiographs. In one horse (Case 1), ventrodorsal
radiographs demonstrated additional fractures of the ischium and
pubis. In 2 horses (Cases 6 and 16), dorsal luxation of the
coxofemoral joint (Fig 5) was diagnosed and confirmed either at
surgery or with ventrodorsal radiographs. In 4 horses (Cases 12, 13,
15 and 17), irregular new bone formation was identified. In one
horse (Case 12) the new bone was located on the dorsal aspect of the
acetabulum, and in another (Case 17) on the femoral neck, with
associated joint incongruity. In both horses, this was considered
consistent with osteoarthritic change. In 2 horses (Cases 13 and 15)
the new bone was located on the cranial aspect of the greater
trochanter of the femur. In all 4 horses, these locations corresponded
to areas of increased radiopharmaceutical uptake identified using
scintigraphy. In one horse (Case 10), the lateral oblique radiograph
TABLE 1: Age, breed, sex, history, radiographic findings and additional diagnostic information for the 18 clinical cases included in this study
Radiographic findings from
lateral oblique view of pelvisAge, breed, sexHistory Additional diagnostic investigations
1 7-year-old TBx gelding Lame LH (grade 9/10) following a fall,
rectal findings unremarkable
L iliac shaft fracture cranial to
Confirmed on ventrodorsal radiographs,
L ischial and pubic fractures also seen;
Scintigraphy demonstrated mild IRU in
area of L acetabulum
2 6-year-old TB gelding Acute onset lameness LH, grade 9/10 Left iliac shaft and coxofemoral
joint appeared normal
No abnormalities detected3 37-year-old TB mare Traumatic injury to perineal region,
pelvis appeared tilted on visual
Cauda equina syndrome4 22-year-old Arab gelding No abnormalities detectedLateral radiographs of sacrococcygeal
spine and pelvic ultrasonography
Ultrasonography confirmed R iliac shaft
fracture and haematoma
58-month-old IDx filly Lame RH (grade10/10) following a fall,
pelvis appeared tilted on visual
examination, no crepitus detected
Consistent upward fixation of L patella
5-week history RH lameness
(grade 10/10), pelvis appeared tilted
on visual examination, muscle wastage
3-week history of severe RH
lameness and upward fixation of the
R patella after a fall
Found in stable with haematoma
over R coxofemoral joint
Acute onset RH lameness (grade 10/10) Apparent displacement of proximal
after jumping out of field
Acute onset RH lameness
R iliac shaft fracture
7-year-old Shetland mare
2-year-old TB mare
Dorsal luxation of L coxofemoral joint
Fracture through right acetabulum
Confirmed on ventrodorsal radiograph
Confirmed, but less well visualised, on
8 9-year-old TB gelding No abnormalities detected Scintigraphy demonstrated fractures of
R tuber coxae and R iliac wing, confirmed
-9 9-year-old TB mare No abnormalities detected
10 5-year-old TB gelding Post mortem demonstrated proximal
Ultrasonography confirmed R iliac shaft
fracture and haematoma
Scintigraphy demonstrated corresponding
but diffuse area of IRU
Over-riding fracture R iliac shaft11 3-year-old TB gelding
1217-year-old cob mare 3-week history of RH lameness
following a fall
New bone evident on dorsal aspect
of acetabulum; differential diagnoses
include old nondisplaced fracture or
Irregular new bone on cranial aspect Scintigraphy demonstrated corresponding
of greater trochanter RH area of IRU
No abnormalities detectedVentrodorsal radiographs appeared normal
134-year-old Arab mare4-week history of RH lameness
after becoming cast
Intermittent and severe hindlimb
RH lameness of 4 months’ duration
144-year-old Shire mare
1512-year-old TB mare Irregular new bone on cranial aspect
of greater trochanter RH
Apparent mild diffuse IRU over R hemipelvis
(because of reduced attenuation due to
muscle wastage?); stifle ultrasonography
demonstrated damage to lateral collateral
ligament of stifle
Confirmed at surgery 1610-year-old cob stallionTrapped foot in gate 10 days
previously, severe LH lameness
Chronic LH lameness, severe muscle
wastage L gluteal muscles
6 months’ RH lameness (diagnostic
analgesia localised pain to the foot;
right coxofemoral joint radiographed
due to scintigraphic findings)
Dorsal luxation of L coxofemoral joint
1712-year-old TB mare Osteoarthritis around L femoral neck,
with joint incongruity
No abnormalities detected
Corresponding area of IRU
Scintigraphy demonstrated IRU
over R coxofemoral joint
18 6-year-old TB gelding
TB = Throughbred; ID = Irish Draught; RH = right hindlimb; LH = left hindlimb; IRU = increased radiopharmacetical uptake.
E. L. Barrett et al.
demonstrated an abnormal relationship between the proximal aspect
of the femur and acetabulum. Post mortem examination
demonstrated a proximal femoral fracture, but this was not apparent
on the radiographs. In 7 horses (Cases 2, 3, 4, 8, 9, 14 and 18) the
lateral oblique pelvic radiographs were considered unremarkable. In
6 horses this conclusion was supported by ultrasonography
(Case 4), additional radiographic views (Cases 4 and 14) and/or
clinical outcome (Cases 2, 3, 9 and 18). In one horse (Case 8),
scintigraphy demonstrated fractures of the ipsilateral tubera coxae
and iliac wing, which were confirmed with ultrasonography.
The lateral oblique view of the pelvis in the standing horse
consistently facilitates visualisation of the iliac shaft, greater
trochanter of the femur, femoral head, acetabulum and coxofemoral
articulation of the side under examination. This technique is
straightforward, noninvasive to perform and provides an invaluable
screening technique in horses with suspected pelvic injury, without
the need for general anaesthesia.
In comparison with the more established technique of
ventrodorsal pelvic radiography, the most important advantage of
this lateral oblique technique is the fact that it can be carried out in
the conscious horse. This increases both patient safety and the
convenience of the technique, facilitating both initial and follow-up
radiographic examinations of the pelvis in horses in which general
anaesthesia may not be appropriate. In the 4 horses which
underwent both ventrodorsal and lateral oblique pelvic radiography,
distortion of the anatomy due to the obliquity of the projection was
minimal. The coxofemoral joint and associated pathology were
easier to view on the oblique view, where they were superimposed
on the air-filled rectum, than on the ventrodorsal view, where they
were superimposed on the pelvic musculature.
Disadvantages of the lateral oblique radiographs include the
lack of visualisation of the pubis, sacrum and iliac wings and the
inability to assess right-left symmetry within the same image.
There are some situations in which further modification of the
technique may help to improve the quality of the radiographs.
Manual evacuation of the rectum did not always result in an air-
filled rectum. In such cases, it may be possible to improve contrast
by inflating the rectum with additional air. However, this
technique was not evaluated in this study, as diagnostic
radiographs were achieved even when the rectum was not air
filled. In the 6 horses which were unable to take weight on the
affected limb, the pelvis was tilted such that the affected side was
positioned further ventrally. In order to compensate for this, it was
necessary to increase the angle between the x-ray beam and the
horizontal by 5–10°, and to centre several centimetres further
ventrally than described for a horse standing squarely.
All radiographic techniques for imaging the pelvic region of
the horse require the use of large x-ray exposure values due to the
large muscle mass of this area. All techniques therefore pose a risk
of scattered radiation reaching personnel involved in the x-ray
procedure. It is essential that minimal personnel are present during
the exposure and that no person is within 2 m of the collimated area
of the primary x-ray beam. Those present should wear protective
lead clothing. Close collimation of the x-ray beam to the area of
interest helps minimise scattered radiation. The cassette should be
mounted on a gantry or stand, and should never be hand-held.
Careful patient preparation and positioning should minimise the
need for repeated radiographs due to poor radiographic technique.
The dosimeters of the personnel involved in the radiography of the
18 horses described in this report recorded no radiation exposure.
The need for high exposure factors also results in the use of
relatively long exposure times, with a risk of movement blur. Use
of the highest possible mA allows a reduction in the length of
exposure for the same mAs value. Observation of the respiratory
cycle allows the exposure to be made during the expiratory pause,
thereby limiting the effects of respiratory movement. Judicious
use of sedation and comfortable positioning of the horse also help
reduce movement artefact. A high-speed film-screen combination
maximises the sensitivity of the film to the penetrating x-rays,
TABLE 2: Pelvic fracture locations in 100 horses (Rutkowski and Richardson 1989)
coxae Fracture locationIliumIschium Pubisand ischium Total
Acetabulum not involved
Fig 4: Iliac shaft fracture, as seen on standing lateral oblique radiography.
Fig 5: Coxofemoral luxation, as seen on standing lateral oblique
270A technique for pelvic radiography in the standing horse Download full-text
helping to keep exposure factors as low as possible. The use of a
digital radiography system allows sophisticated post processing of
the image, which should reduce the need for repeated radiography
due to poor radiographic technique. It has been suggested that
digital radiography may allow reduction of the exposure factors
used, but this remains unresolved (Marshall et al. 1994; Dendy
and Heaton 2003).
In this series of 18 horses, 6 were finally diagnosed with pelvic
fractures. All 3 iliac shaft fractures (Cases 1, 5 and 11) and the
acetabular fracture (Case 7) were identified conclusively using
lateral oblique pelvic radiography. However, in one horse (Case 1),
the full extent of concurrent ischial and pubic involvement could
not be appreciated. Fractures of the iliac wing and tuber coxae in
one horse (Case 8) were not identified. In the final horse
(Case 10), a proximal femoral fracture was not diagnosed until
post mortem, although it was noted from the lateral oblique
radiographs that the proximal femur was abnormally positioned.
Previous analyses of equine pelvic fractures demonstrated a
relatively high incidence of iliac and acetabular involvement. In a
study of 19 horses (Little and Hilbert 1987) 9 had iliac fractures, with
5 of these involving the acetabulum (determined from ventrodorsal
radiographs). Six further fractures were described as involving the
acetabulum together with the ischium and/or pubis. A case series of
pelvic fractures in 100 horses (Rutkowski and Richardson 1989)
recorded fracture location in 76 horses (Table 2) as determined by
ventrodorsal radiographs (n = 57) or post mortem examination
(n = 19). Fractures of the ilium were reported in 42 horses, involving
the acetabulum in 21 of these. Overall, acetabular involvement was
reported in 49 of the 76 horses. It was possible to visualise the
acetabulum in all 18 horses in our case series; theoretically,
therefore, acetabular fractures should be visualised consistently with
this technique. However, as in Case 1 reported here, it would not be
possible to visualise fully the extent of ischial and pubic involvement.
In addition, 6 fractures of the tuber coxae were reported in the case
series of 100 horses (Rutkowski and Richardson 1989). These would
not have been visualised with standing lateral oblique pelvic
radiographs, as demonstrated in the present study (Case 8). Neither
of the 2 previous case series distinguished between fractures of the
iliac wing and iliac shaft. While most iliac shaft fractures should be
identified with lateral oblique radiography, iliac wing fractures will
not be seen, as demonstrated in our study (Case 8).
Coxofemoral luxation in the horse is rare (Jeffcott 1982):
however, the technique described here may be used reliably to
diagnose this condition (Cases 6 and 16). Lateral oblique
radiography in the standing horse also has the potential to investigate
such conditions as coxofemoral osteoarthritis (Cases 12 and 17),
fracture of the greater trochanter, fracture of the femoral neck or
slipped proximal femoral epiphysis. These conditions are important
differential diagnoses for pelvic fractures and can be difficult to
diagnose definitively. The clinical significance of the new bone seen
on the cranial aspect of the greater trochanter of the femur in Cases
13 and 15 is unclear, and would be interesting to investigate further.
This study demonstrates the potential use of lateral oblique
pelvic radiography in the standing horse for investigation of pelvic
lameness. However, it is important to appreciate that not all pelvic
injuries can be identified and that normal radiographic findings do
not rule out injury or fractures elsewhere in the pelvis. It should
also be noted that the largest horse radiographed in this series
weighed 550 kg. It may not be possible to obtain images of
adequate diagnostic quality in very large or very fat horses. Lateral
oblique pelvic radiographs should therefore be considered a useful
and noninvasive technique to be used alongside other diagnostic
techniques in the investigation of suspected pelvic injury.
The authors gratefully acknowledge the assistance of Mr John
Conibear and Mr Nick Crabb, Department of Clinical Veterinary
Science, University of Bristol, in the preparation of the
photographs and line drawings. We would also like to thank the
staff within the Department of Clinical Veterinary Science who
contributed to the investigation and care of these horses, in
particular those involved in the radiographic examinations, and
our colleagues in the Department of Pathology for the results of
the necropsy examination in Case 10.
1Eastman Kodak Company, New York, New York, USA,
2Shimadzu Medical Imaging Systems, London, UK.
3Konica Minolta Medical and Graphical Inc., Tokyo, Japan.
Almanza, A. and Whitcombe, M.B. (2003) Ultrasonographic diagnosis of pelvic
fractures in 28 horses. Proc. Am. Ass. equine Practnrs. 49, 50-54.
Attenburrow, D.P. (1997) The application of scintigraphy by using a gamma camera in
clinical equine practice 2: Interpretation of results. Equine vet. Educ. 9, 103-109.
Busoni, V., Gabriel, A., Lacroix, S. and Snaps, F. (2001) Ultrasonography of the back
and pelvis in the horse. Vet. Radiol. Ultrasound 42, 178. [Abstr.]
Dendy, P.P. and Heaton, B. (2003) Special radiographic techniques. In: Physics for
Diagnostic Radiology, 2nd edn., Eds: R.F. Mould, C.G. Orton, J.A.E. Spaan and
J.G. Webster. Institute of Physics Publishing, Bristol and Philadelphia. p 230.
Driver, A.J. (2003) Basic principles of equine scintigraphy. In: Equine Scintigraphy,
Eds: S.J. Dyson, R.C. Pilsworth, A.R. Twardock and M.J. Martinelli, Equine
Veterinary Journal Ltd, Newmarket. pp 17-24.
Erichsen, C. and Berger, M. (2001) Pitfalls in the interpretation of pelvic scintigrams
caused by soft tissue attenuation. Vet. Radiol. Ultrasound 42, 178. [Abstr.]
Hendrickson, D.A. (1987) The pelvis. In: Adams’Lameness in Horses, 4th edn., Ed:
T.S. Stashak, Lea & Febiger, Philadelphia. pp 1044-1053.
Hornof, W.J., Stover, S.M., Koblik, P.D. and Arthur, R.M. (1996) Oblique views of
the ilium and the scintigraphic appearance of stress fractures in the ilium. Equine
vet. J. 28, 355-358.
Jeffcott, L.B. (1979) Radiographic examination of the equine vertebral column. Vet.
Radiol. 20, 135-138.
Jeffcott, L.B. (1982) Pelvic lameness in the horse. Equine Pract. 4, 21-49.
Kangstrom, L.E. (1972) A simple technique for radiographic examination of the
equine pelvis, hip joint and proximal femur. Acta Radiol., Suppl. 319, 79-82.
Lewis, R.E. and Heinze, C.D. (1971) Radiographic examination of the equine pelvis:
technique. J. Am. vet. med. Ass. 159, 1387-1390.
Little, C. and Hilbert, B. (1987) Pelvic fractures in horses: 19 cases (1974-1984). J.
Am. vet. med. Ass. 190, 1203-1206.
Marshall, N.W., Faulker, K., Busch, H.P., Marsh, D.M. and Pfenning, H. (1994) An
investigation into the radiation dose associated with different imaging systems
for chest radiology. Br. J. Radiol. 67, 353-359.
May, S.A., Patterson, L.J., Peacock, P.J. and Edwards, G.B. (1991) Radiographic
technique for the pelvis in the standing horse. Equine vet. J. 23, 312-314.
Pilsworth, R.C., Holmes, M.A. and Shepherd, M. (1993) An improved method for the
scintigraphic detection of acute bone damage to the equine pelvis by probe point
counting. Vet. Rec. 133, 490-495.
Rutkowski, J.A. and Richardson, D.W. (1989) A retrospective study of 100 pelvic
fractures in horses. Equine vet. J. 21, 256-259.
Shepherd, M.C. and Pilsworth, R.C. (1994) The use of ultrasound in the diagnosis of
pelvic fractures. Equine vet. Educ. 6, 223-227.
Tomlinson, J., Sage, A. and Turner, T.A. (2000) Ultrasonographic examination of the
normal and diseased equine pelvis. Proc. Am. Ass. equine Practnrs. 46, 375-377.
Tomlinson, J.E., Sage, A.M. and Turner, T.A. (2001) Detailed ultrasonographic mapping
of the pelvis in clinically normal horses and ponies. Am. J. vet. Res. 62, 1768-1775.