Radiography is a poor method to use to assess cardiac dynamics but is valuable for assessing the efficacy of treatment and radiographs can be used to evaluate for signs of cardiac decompensation. In this presentation the key signs used to differentiate heart disease from heart failure will be described. This is important as it can influence what treatment is administered. Abnormalities will be discussed with reference to the normal variation that can be misinterpreted as disease, and also with reference to radiographic positioning.

The wide range of normal cardiac appearance is one reason that radiographs are a poor method for cardiac assessment. The appearance of the cardiac silhouette is affected by: 1) body condition, especially the volume of fat in the mediastinum, 2) the body stature of the patient, 3) ventilation, as might be altered by sedation, and 4) the position of the patient during radiography.

In obese patients, fat in the mediastinum will cause the cardiac silhouette to appear enlarged. Occasionally, a demarcation between the mediastinal fat and heart can be seen due to the difference in radiographic opacity of fat vs. soft tissue, but often mediastinal fat just contributes to what appears to be generalized cardiomegaly.

The shape of the cardiac silhouette parallels the overall shape of the body, with lean dogs such as Doberman pinschers or collies having a tall thin heart, and stout dogs such as Boston terriers and Bulldogs having a more round heart. It is easy to misinterpret the normal cardiac silhouette in some breeds as cardiomegaly.

Ventilation has a profound effect on the appearance of the cardiac silhouette. With poor aeration the overall size of the thoracic cavity is diminished and the heart will appear enlarged. Incompletely aerated lungs also have an increased opacity, which is usually overinterpreted as pulmonary disease. Lungs are almost always more poorly inflated in lateral views than in VD or DV views. Lung disease is commonly misdiagnosed by relying on lateral views.

Whether the dog is in dorsal or ventral recumbency also affects the appearance of the cardiac silhouette, especially in large dogs. In DV radiographs, the diaphragm is displaced cranially and this pushes the heart to the left, and the heart appears rounder. These changes are often misinterpreted as cardiomegaly in DV radiographs.

Another consideration with regard to patient positioning is the much improved conspicuity of the cranial lobe pulmonary vessels in left lateral radiographs compared to right lateral radiographs. Pulmonary vessels can provide important information about the physiologic status of the cardiovascular system, so it is important to be able to evaluate them accurately.

Another problem that complicates radiographic interpretation is bias introduced by the history or physical findings. Radiographs of a patient with a heart murmur will be interpreted more critically than when a heart murmur is not present. One's threshold for detection of disease should not be altered by the history. It is also important to think in terms of physiology rather than morphology with assessing thoracic radiographs. The significance of radiographic changes can be assessed more accurately when the pathophysiology of the suspected disease is considered in addition to only morphologic abnormalities.

In general, the ability to define the following abnormalities accurately from thoracic radiographs is poor: 1) left ventricular hypertrophy, 2) right ventricular hypertrophy, and 3) right atrial dilation. Ventricular hypertrophy is not readily detected radiographically because the hypertrophy occurs as the expense of the lumen volume rather than an overall increase in cardiac volume. Right atrial dilation is rare and rarely seen in isolation. Left ventricular dilation and right ventricular dilation can be seen, depending on the degree, but being able to identify those abnormalities specifically from radiographs is poor and the usual conclusion is "generalized cardiomegaly". Echocardiography is needed for an accurate assessment of these abnormalities.

Cardiac measurement techniques have been proposed as a way to make assessment of heart size more standardized. A problem with this technique is that the range of normal values is still wide enough that prediction of cardiac size in an individual patient using this approach is not accurate. Cardiac measurement has its greatest value in following changes in heart size in an individual patient.

The abnormality that can be detected with a moderate degree of accuracy is left atrial dilation. This is fortunate, as left atrial dilation is an important component of the most common acquired cardiac abnormalities, such as mitral valve disease and cardiomyopathy. Echocardiography will still be important to appreciate the deranged function leading to the left atrial dilation.

Left atrial dilation is best assessed in the lateral view. Enlargement causes: 1) increased soft tissue mass and a concave shape at the dorsocaudal aspect of the cardiac silhouette, and 2) elevation, and occasional compression, of the tracheal bifurcation, especially the left main bronchus. In extensive left atrial dilation, there will be a "double wall" appearance of the cardiac silhouette in the VD or DV view due to the enlarged left atrium being superimposed on the ventricular mass.

Moderate to pronounced left atrial dilation also causes splaying of the main stem bronchi as seen in the VD or DV view. This splaying causes the bronchi to be curved rather than straight at the bifurcation.

Thus, the real value of thoracic radiographs in patients with suspected or known heart disease are: 1) a global view of the heart can be obtained, this is not possible with echocardiography, 2) pulmonary vessels can be assessed, and 3) signs of heart failure can be detected.

As noted previously, assessment of pulmonary vessels is critical. Pulmonary vessels in the cranial lobes are better seen in the left lateral view than the right lateral view because they are not superimposed in the left lateral view. Pulmonary vessels in the caudal lobes are better seen in DV views than in VD views because they are less affected by adjacent recumbent atelectasis, and they are more parallel to the primary x ray beam. In lateral views, veins are ventral to the corresponding pulmonary artery and in the VD or DV view, veins are medial to the corresponding pulmonary artery.

There are measurement data defining the normal size of pulmonary vessels but comparing arteries to veins subjectively in lateral and VD/DV views is more valuable; they should be the same size. In VD radiographs, caudal lobe vessels can be compared to the size of the 9th rib where they intersect. The summation shadow created by overlapping should be a rectangle, i.e., the vessel and rib should normally be the same size at the point of intersection.

Finding pulmonary veins larger than pulmonary arteries suggests pulmonary venous hypertension, as might result from mitral valve dysfunction. Finding pulmonary arteries larger than pulmonary veins suggests pulmonary artery hypertension, as might result from heartworm disease, pulmonary vasculopathy or pulmonary fibrosis. Finding both pulmonary arteries and veins enlarged suggests circulatory hypervolemia as might occur in a left-to-right shunt, fluid overload, or fluid retention.

Importantly, pulmonary vessels can be normal even if circulatory dynamics are not (false negative), as with dehydration due to diuretic therapy. Pulmonary vessels can also be abnormal if the inciting cause has been corrected (false positive), as with heartworm disease that has undergone successful treatment.

Radiographs are an excellent method to assess cardiac decompensation. Left heart failure is more common than right heart failure, occurring secondary to mitral valve degeneration, hypertrophic and restrictive cardiomyopathy in cats, and hypertrophic and dilatory cardiomyopathy in dogs. Right heart failure, as might occur from dilatory cardiomyopathy in dogs or heartworm disease, is not discussed herein.

The hallmark radiographic sign of left heart failure is pulmonary edema. The radiographic appearance of the edema depends on the stage of the disease and the treatment that has been administered. Pulmonary edema can be interstitial early-on or following treatment. Pulmonary edema can occasionally be peribronchial, in which situation the appearance is more suggestive of inflammatory disease than heart failure. The exact pathophysiology behind pulmonary edema having a peribronchial distribution is not fully understood. And, as expected most of the time, pulmonary edema usually has an alveolar pattern.

Alveolar pulmonary edema can have a variety of distributions. The most common is a patchy generalized distribution that is not usually as opaque (intensity per unit area of lung) as that of inflammatory lung disease. Pulmonary edema can be often devoid of air bronchograms, the hallmark sign of an alveolar pattern. This is especially true in cats and is likely due to the relatively low intensity per unit area of cardiogenic pulmonary edema. Having uniformly opaque lungs as a result of cardiogenic edema is uncommon.

Cardiogenic pulmonary edema is often described as having a perihilar distribution. This is true in people but I believe this is overemphasized in the dog. Importantly, therapy directed against heart failure can cause a rapid change in the intensity and distribution of cardiogenic pulmonary edema.

One would typically expect pulmonary veins to be distended in patients with left heart failure, and this is often seen, but not always. Pulmonary veins can be normal as a result of prior diuretic therapy. Also, sometimes pulmonary arteries and veins are both enlarged in patients with left heart failure; this results from decreased cardiac output that leads to activation of the renin-angiotensin-aldosterone system, leading to water and salt retention and increased circulating volume.

Though typically thought of as a sign of right heart failure, pleural effusion can develop in cats with left heart failure. The pathophysiology of this change is not completely understood.

In summary, radiography is limited with regard to cardiac evaluation. Findings are not sensitive or specific, except for left atrial dilation. Clinical signs and physical findings can bias the interpretation of thoracic radiographs in patients suspected of having heart disease. Assessing the pulmonary vessels is probably of more value radiographically than trying to assess the heart. Finally, heart disease does not equal heart failure. Cardiac and/or pulmonary vessel enlargement without signs of decompensation, i.e., pulmonary edema, is a reflection of cardiac disease but not failure.