Thorax: Are the Changes I Am Seeing Clinically Significant?
World Small Animal Veterinary Association World Congress Proceedings, 2003
Robert M. Kirberger, BVSc, MMedVet (Rad), DECVDI
Department of Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria
Onderstepoort, Republic of South Africa

Thoracic radiographic interpretation is the most complex of all body systems. Accurate interpretation of abnormalities involves the following:

1.  excellent radiographic technique, preferably using high output equipment to minimise motion blur

2.  understanding of the effect of positioning on the image

3.  knowledge of normal anatomy and possible variants

4.  interpretative ability to distinguish insignificant findings from pathology

In evaluating a radiograph the following systematic approach should be used:

1.  evaluate technical factors (exposure technique, respiratory phase and positioning)

2.  evaluate extrathoracic structures

3.  evaluate intrathoracic structures

4.  correlate radiological findings with clinical presentation.

Effect of positioning

On lateral radiographs the most dependant lung is compressed by the weight of the mediastinum with its content (including the heart) and the most dependant crus which is displaced cranially by the abdominal content. The uppermost lungs will undergo a degree of compensatory overinflation. The visibility of pulmonary pathology is dependant on contrast between air and soft tissue. The relative lack of air in the dependent lung diminishes this contrast resulting in loss of visibility of pathology. In the upper lung the relative increased contrast enhances pathology visibility.

Right lateral vs. left lateral views

In right lateral positioning the caudal vena cava enters the most cranially positioned right crus. The heart is most accurately positioned for interpretation with the apex and part of the right ventricle in contact with the sternum. In left lateral positioning the caudal vena cava traverses the most cranial crus and enters the more caudally positioned right crus. The cardiac apex tends to lift from the sternum with less right ventricle sternal contact mimicking concentric right ventricular hypertrophy as seen in pulmonic stenosis.

Dorsoventral(DV) vs. ventrodorsal(VD)

The DV view has the most constant position of the cardiac silhouette and is characterised by well defined caudal lung lobe vessels and a prominent diaphragmatic cupula due to pressure of the ventrally located abdominal organs. In the VD position the two diaphragmatic crura are often seen as prominent bulges with a more retracted cupula. The cardiac silhouette tends to displace cranially allowing greater visibility of the accessory lung lobe region. The VD view may result in greater divergence of the main stem bronchi mimicking left atrial enlargement.

Rotated views

Lack of superimposition of the rib origins is indicative of a rotated lateral view. This results in tracheal elevation and in splitting of the main stem bronchi simulating left atrial or hilar lymph node enlargement. Rotated DV/VD views may mimic cardiac chamber enlargement.

Breed differences

Conformation is the single most important cause for apparent cardiomegaly in barrel-chested dogs such as the Bulldog, Yorkshire Terrier and Dachshund which have a relatively large heart and elevated trachea on lateral views. The Dachshund characteristically also has a very wide trachea. Deep chested dogs have a more erect cardiac silhouette and minimal sternal contact.


Fat deposits may accumulate in the sternal region and can mimic pleural effusion but is distinguished from effusion by its decreased opacity relative to the soft tissues of the cardiac silhouette. Subpericardial fat simulates cardiomegaly. Fat in the mediastinum, particularly cranially on DV/VD views, looks like a cranial mediastinal mass. The trachea is however in its normal position, slightly to the right, and on lateral views there is no sign of a mediastinal mass.

Artifactual increased pulmonary pattern

Diffuse increased pulmonary opacities similar to that seen with an interstitial lung pattern may be caused by underexposure, underdevelopment, poorly inflated lungs due to expiratory radiographs, abdominal distention or laryngeal paralysis. Obesity and motion blur may contribute as well as thoracic limbs not pulled fully forward, bandages and wet or dirty hair coat. Hypostasis due to prolonged lateral recumbency also mimics parenchymal pathology.

Is it a pulmonary nodule or MASS?

Numerous structures may simulate pulmonary nodules. Skin nodules, particularly teats, are contrasted by the surrounding air and are thus often mistaken for pulmonary nodules. Bizarre mineralization of the costal cartilages and costochondral junctions and healed rib fractures may also be confusing. Pulmonary blood vessels seen in cross section are the most common cause of mis-diagnosing pulmonary nodules. In the hilar region the main artery supplying the right middle lung lobe seen end on mimics a nodule of up to 1 cm in diameter. Other pulmonary vessels are distinguished from tumour nodules by their greater opacity; being perfectly circular with crisp margins; diminish in size peripherally; and are associated with longitudinally coursing vessels. In older dogs pulmonary osteomata (heterotopic ossification) or calcified pleural plaques also simulate nodules but because of their mineral content are more opaque than similar sized pulmonary soft tissue opacity nodules. Pleural masses as well as small diaphragmatic hernias or eventrations may mimic pulmonary masses.

Artifactual focal areas of pulmonary hyperlucency

Pulmonary cavitatory lesions may be mimicked by intrapulmonary ring shadows caused by curved bronchial walls and vessels as well as extrapulmonary ring shadows caused by superimposed subcutaneous gas, gas filled herniated gastrointestinal content, focal pneumomediastinum, expansile rib lysis and small amounts of oesophageal air including focal gas accumulation in a redundant oesophagus at the thoracic inlet.

Tracheal narrowing and malposition

Tracheal narrowing due to a collapsing trachea or hypoplasia may at the cervicothoracic junction be confused with a superimposing longus colli muscle or oesophagus. Intrathoracic dorsal deviation of the trachea may be caused by ventral bending of the neck and expiratory or rotated views. On the DV/VD radiograph the trachea normally deviates slightly towards the right due to the presence of the normal left aortic arch. This deviation may be accentuated by the same factors causing dorsal deviation of the trachea.


A small amount of oesophageal air is often seen just cranial to the carina region. The caudal oesophagus is sometimes seen as a poorly defined soft tissue opacity dorsal to the caudal vena cava on left lateral recumbent views.


Pseudopneumothorax may be caused by overexposure, relative radiolucency lateral to axillary skin folds and overinflation of the lungs. This increased radiolucency of the pleural cavity can be determined to be artifactual by seeing pulmonary vessels in the suspect region with a hot light.

Cardiac changes

A small heart may be misrepresented by deep inspiration, deep chested dogs, pulmonary overinflation and a heart displaced from the sternum by a mediastinal shift or pneumothorax. In geriatric cats on lateral views the cardiac axis may be markedly cranially tilted (so called lazy heart) and the aorta may be redundant. The latter may also be seen on DV views as well as in brachycephalic and older dogs. Tortuous mineralisation of the coronary arteries is seen as an incidental finding in <1% of dogs. A pulmonary segment bulge at 1-2 o' clock may be seen when the exposure is made during systole simulating a post stenotic pulmonary artery dilatation. This may also be seen in very deep chested dogs or slightly rotated views. Pneumopericardium with its associated delineation of the auricles, is a rare finding following trauma but resolves on its own.

Osseous changes

Narrowed cranial thoracic intervertebral disc spaces, often with associated spondylosis, are a common insignificant finding. The spondylosis must however be differentiated from spondylitis which in the caudal thoracic region is pathognomonic for Spirocerca lupi infection.


1.  Dennis R, Kirberger R M, Wrigley R H and Barr F J. Small Animal Radiological Differential Diagnosis. W B Saunders London. 2001.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Robert M. Kirberger, BVSc, MMedVet (Rad), DECVDI
Department of Companion Animal Clinical Studies
Faculty of Veterinary Science
University of Pretoria
Onderstepoort, Republic of South Africa