Clinical history
Max had been on long term treatment for a grade VI heart murmur for 1 year, with frusemide 20 mg tablets quarter daily and lanoxin 0.0625 mg tablets half twice daily. He was stable on the medication and doing well. Then he was bitten by a dog and suffered a glenoid fracture of the right scapula. This was repaired and postoperatively the dog was doing well. Two months later the owner presented Max for coughing and wheezing which had been getting progressively worse for the last 3 weeks. Max was booked in for radiography and ultrasonography, the owner having signed a consent form for the procedures and general anaesthesia. The dog was starved from the night before with access to water until 2 hours prior to premedication.

Restraint for radiography
This was achieved with general anaesthesia. 

Choice of film/screen/grid
Agfa CP B Blue 100 NIF film was used with calcium tungstate screens. These screens emit blue light so a blue light sensitive film has to be used with them. No grid was used. The tissue depth was 10 cms which was borderline for use of a grid, although some sources state that when radiographing the thorax up to 15 cms can be radiographed without the use of a grid.

Positioning
Right Lateral
The dog was placed in right lateral recumbency with the thorax on the cassette. A foam pad was placed between the fore limbs and the forelegs were extended cranially and tied. A foam wedge was placed under the sternum to prevent axial rotation of the thorax. The neck was placed in a neutral position.

Dorsoventral
The dog was placed on it's sternum with the forelimbs drawn forwards and tied. The hind limbs were extended and drawn caudally. This is to help achieve a straight positioning. The spine should be straight and axial rotation should be avoided by extending both forelimbs equally and rotating the dog until the sternum is directly under the thoracic spine. A right marker was placed on the appropriate side of the animal.

All safety precautions taken during radiography are described in Appendix 1.

Centering
Lateral
The primary beam was centred just caudal to the caudal border of the scapula and midway between the thoracic spine and ventral thoracic wall.

Dorsoventral
As above.

Collimation
Lateral
The primary beam was collimated from a position just cranial to the manubrium to the last rib. It was also collimated from the spinous process of T12 to the ventral thoracic wall.

Dorsoventral
Laterally the primary beam was collimated just outside each thoracic wall and also from just cranial to the manubrium to the last rib.

Exposures
The following exposures were used:

Both exposures were taken on inspiration.

 

Radiographic appraisal
Choice of film and screen
The tungstate screens and blue light sensitive film are standard at this practice. Possibly a better detailed radiograph could have been obtained using a rare earth screen and green light sensitive film such as Agfa CP-G PLUS. Rare earth screens are more efficient at absorbing x-ray photons and converting them to visible light, resulting in smaller exposures being necessary and better quality films. As the tissue depth was on the border line for use of a grid, a slightly better quality radiograph might have been obtained by using one.

Positioning
Lateral
The forelegs are drawn well forward, out of the cranial thoracic field. There is a little axial rotation which can be seen where the ends of the ribs are not in alignment. Perhaps more time spent assessing the elevation of the sternum could have improved this. However, this dog had a barrel shaped thorax which makes visual assessment difficult.

Dorsoventral
This projection appears to be quite straight. It is difficult to see the sternum which would indicate that it is directly under the spine. The spine itself is reasonably straight and both forelegs appear to be equally extended.

Centering
Lateral
The primary beam is centred over the base of the heart. This would seem to be an acceptable centring point for the general view of the thorax. In some animals it is quite difficult to be accurate both with centring and collimation - correct collimation will leave the centring point slightly out of place and, similarly correct centring can sometimes make correct collimation impossible. This is because they are inter-linked, with the centre of the beam always falling within the centre of the field of collimation.

Dorsoventral
The primary beam is centred over the spine at the level of T8. This is a little caudal to the point mentioned in literature, which would appear to be at T6, if centred just caudal to the scapula. However, this is an example of the collimation/centring problem mentioned above. For the purposes of this radiograph the centring would probably be accurate enough.

Collimation
Good collimation is necessary for both film quality and personnel safety. All 4 film edges are unexposed which demonstrates the extent of the primary beam. Close collimation is required to reduce scatter from tissues which are not under investigation. This scatter will reduce the overall quality of the radiograph's detail. Both of these projections are collimated reasonably well. Both are collimated correctly, just cranial to the manubrium and to the last rib. Laterally the collimation is acceptable on the dorsoventral projection, and dorsally on the lateral projection the collimation is correct, as it is important to see the dorsocaudal lung fields.

Exposure
The exposure of these projections is reasonable. The thorax can be a difficult area to obtain good overall exposure. The lungs having a very low physical density and the heart having a high physical density. Therefore one or other organ will be under or overexposed. For this reason a high kV and low mAs are used - higher kV's produce less contrast between different physical densities of tissues. The other advantage of this technique is that exposure times can be shorter which helps with the problem of movement blur that occurs when radiographing the thorax. There is a large range of densities on both projections, and the definition (sharpness is reasonable). With regard to detail, the pulmonary vessels can be seen and the caudal vena cava is visible on the lateral projection. The aorta is not visible but that may be due to the heart enlargement.

On the dorsoventral projection the heart is well penetrated, allowing site of the bifurcation of the trachea (carina) at the level of T5 and the bronchi. Detail could be better with regard to the internal structures of the lungs. Perhaps manual inflation of the lungs to maintain maximum lung expansion without respiratory movement whilst the exposure is made might have improved detail. This is not possible at this practice, as no one is allowed in the room whilst an exposure is made.

The lateral projection was taken on inspiration. The caudal crura meet the vertebral column at T13. On inspiration there should also be a space between the heart and diaphragm. However, as this radiograph demonstrates, if the heart is enlarged it is not a reliable indicator of the respiratory cycle.

If the same premise can be applied to the dorsoventral projection the crura meet the vertebral column at the level of T12, so it could be assumed that this exposure was not taken at full inspiration.

Processing
This radiograph was put through an automatic processor and appears to have been processed correctly. Over development would result in blackening of the metal marker.

Artefacts
The processor rollers have left some marks over the length of the film. Despite fully dis- mantling and cleaning every week there still artefacts appearing on radiographs. This is very frustrating. The metal wire appearing on both projections is not a true artefact but the wire used for the repair of the fractured shoulder.

There are several fingerprints on the film, caused after processing. This is difficult to avoid as these are 'working' radiographs, and have been handled and examined by veterinary surgeons!

Labelling
The system of labelling radiographs as they emerge from the processor is not ideal and all radiographs should be labelled at the time of exposure to avoid the possibility of mis- labelling.

Diagnosis
The radiography enabled the size of the cardiac shadow to be assessed - in this case it is grossly enlarged. Pulmonary oedema was also seen within the lungs. With the results of both radiography and ultrasonography the treatment regime was changed to frusemide 40mg quarter twice daily, lanoxin pg 0.25 mls per os twice daily and Cardiovet 5mg quarter twice daily for 2 days then half twice daily.

General comments on radiography
In this case a complete diagnosis could not be made with radiography. Ultrasonography was also used prior to general anaesthesia to assess the physical condition of the heart. The results of which showed - a massive volume overload of the left atria and ventricle. The mitral valve is thickened and can actually be seen leaking when fully closed. See figure 1.  

References
Manual Sm An Diagnostic Imaging. BSAVA 1997
Radiographic Techniques - The Dog. Morgan, Duval, Samii. Schlutersche. 1998.