Introduction

There are many indications to perform spinal and pelvic radiographs in dogs and cats, the most common ones including hindlimb lameness, mono- or paraparesis or -plegia, spinal and pelvic pain, trauma to the back or pelvic region. Survey radiographs are almost always warranted even if further imaging such as Computed Tomography or Magnetic Resonance Imaging is pursued. The goal of spinal radiography is to rule out traumatic lesions (fractures, luxations), aggressive lesions such as vertebral tumors and discospondylitis, and to determine if there are degenerative changes or changes to a disc space that could indicate disc disease. Pelvic radiographs are often performed to evaluate for fractures or changes to the coxofemoral joints. Good radiographic technique is of utmost importance in order to detect these lesions with confidence and on the flip side poor radiographic technique can obscure lesions, or just as detrimental, create the appearance of lesions where there are none. Obtaining the right projections is also very important depending on the indication. In this presentation examples of common diseases and pitfalls associated with spinal and pelvic radiography will be shown.

Spinal Radiography

As a general rule, heavy sedation or general anesthesia is needed for radiography of the spine. It is rare that a patient is relaxed enough without sedation that the spine can be positioned properly. The exception to this rule is the acute trauma patient where muscle contraction around a fracture should not be compromised with sedation to avoid destabilization. Orthogonal (lateral and ventrodorsal) radiographs are the standard, however, there are instances where lateral radiographs are sufficient for example to screen for larger trauma or discospondylitis. Trauma patient should be moved as little as possible and ventrodorsal projections should only be obtained if they can be performed with a horizontal beam. Main problems/artifacts that can hinder interpretation include patient rotation, too large of a field of view and in lateral projections insufficient padding of portions of the spine that naturally deviate towards the table such as the cervical spine and the caudal lumbar spine. Superimposition of structures such as the ear cartilages, the shoulder joints, and collars and harnesses can obscure lesions. Centering the beam to an area of suspected abnormality is very important as for example disc spaces are markedly distorted in the periphery of a large field of view due to the angled, divergent nature of the x-ray beam.

For good quality lateral cervical radiographs, the patient should be positioned with the head extended and supported so that it is not rotated (for example by placing a foam wedge underneath the chin). The ears should be pulled forward so that they are not superimposed over the atlantoaxial junction. The thoracic limbs are pulled caudally to avoid superimposition of the shoulder joint over the caudal cervical spine. Padding is added under the caudal cervical spine to have the spine parallel to the x-ray table. In a perfect lateral view there is complete superimposition of the wings of the atlas. This, unfortunately, leads to poor visibility of the odontoid process of C2 and, therefore, a slightly oblique projection (head slightly rotated) is a helpful view when atlantoaxial subluxation is suspected as it projects the dens dorsally over the vertebral canal and fragmentation or blunting of the dens would be more easily visible. Extended and flexed views are not recommended when evaluating dogs for atlantoaxial luxation as it could result in spinal cord damage. For a ventrodorsal view the patient is ideally placed in a trough and straightened with the thoracic limbs also pulled caudally. Flexion and extension is occasionally used to evaluate presence of instability in the caudal cervical spine in dogs with suspected cervical spondylomyelopathy. Subtle lesions that involve the pedicles and transverse processes can be highlighted with oblique ventrodorsal projections, this projection also allows evaluation of the intervertebral foramina. Rotation and motion artifact are the main problems when radiographing the thoracic spine. Similar to the caudal cervical spine, padding should be added underneath the lumbar spine directly cranial to the pelvis to ensure that the spine is positioned parallel to the table. The last thoracic vertebrae or lumbosacral junction should be included in all views of the lumbar spine so that lesions can be localized to the correct vertebra. Lateral flexed and extended radiographs of the lumbosacral junction are taken to determine presence of instability at the lumbosacral junction in dogs with suspected cauda equina syndrome. When evaluating the lumbosacral endplates for defects such as discospondylitis or osteochondrosis, a ventrodorsal view with the pelvic limbs in flexed (frog-leg) position is helpful as it highlights a slightly different portion of the endplate and endplate defects are often better visible than in extended hip views in general.

Pelvic Radiography

A routine pelvic radiographic series should include a lateral and ventrodorsal view. The lateral view is helpful to evaluate concomitant lesions of the caudal lumbar and lumbosacral spine, the pelvic canal and the caudal abdominal soft tissues. Periosteal reactions along the pelvic margins may be seen with prostatic cancer and are often best seen in the lateral view. The coxofemoral joints are more difficult to evaluate as there is superimposition with the pelvis. Ventrodorsal projections can be obtained with the pelvic limbs extended which is the standard view for evaluation of the coxofemoral joints. To evaluate for joint alignment and possible laxity it is very important to have proper positioning without axial rotation, otherwise the coxofemoral joints can artificially look better or worse than they are. Good positioning is recognized by symmetry of the ilial wings, similar size and shape of the obturator foramina and centered position of the spinous processes of the caudal lumbar spine. The femora should be parallel with the patella centered over the femoral trochlear groove. This is only possible in the heavily sedated patient as muscle tone otherwise prevents extension and internal rotation of the proximal pelvic limbs.

Flexed ventrodorsal projections are particularly useful when evaluating the lumbosacral junction for endplate defects, and in dogs and cats with suspected femoral neck fractures. Femoral neck fractures (traumatic or developmental as with slipped capital physes in cats) can be minimally displaced and difficult to detect in extended hip views as the fracture is reduced through positioning.

Symmetric positioning of the pelvis is also very important when evaluating patients for aggressive or traumatic lesions of the pelvis. Slight rotation can occasionally produce an unusual projection of a portion of the pelvis resulting in over- or under interpreting changes in bone opacity. Other artifacts that can be problematic is superimposition of fecal material in the rectum. Gas lucency within the fecal material my result in the false impression of a radiolucent lytic lesion. It also hinders interpretation of the complex bone structure of the sacrum and cranial coccygeal vertebrae. Presence of gas in the anal sacs results in rounded radiolucent “lesions” superimposed over the ischium and should not be confused with a pathologic aggressive lesion.

Additional, specialized pelvic views to evaluate dogs for hip dysplasia include distraction methods (PennHIP for example) and dorsal acetabular rim views that are mostly used for surgical planning.

References

1.  Vet Clin North Am Small Anim Pract. 2016;46(2):265–275.

2.  Vet Clin North Am Small Anim Pract.2018:48(1):85–94.

3.  Vet Clin North Am Small Anim Pract. 2017:47(4):777–793.