Thoracic radiography of respiratory cases
Radiography is the most used tool in the approach, screening and diagnosis of thoracic pathology. The main advantage over other more recent techniques is the inherent revelation of the morphology of thoracic structures. Its sensitivity and specificity can be increased with the use of special techniques, employment of contrast medium and follow-up exposures. Its limitations are the inability to differentiate between structures of the same radiographic density and the fact that it is a representation of a 3-dimensional structure in a 2-dimensional medium. As with all imaging techniques it does not allow for the evaluation of function although it may be possible to estimate this from radiographic appearance or with the use of special procedures such as contrast media techniques.
Thoracic radiography, as with all diagnostic procedures, should not be performed “to see what can be found”. It is not intended to replace physical examination and neither is it the first step in the management of the patient or the disease. It is an auxiliary procedure that is intended to confirm or exclude prior suspicions (Table 1) based on the history, clinical examination and other diagnostic procedures (1). This basic tenet is sometimes lost, leading to ambiguous interpretation, overdiagnosis and failure to detect primary or secondary etiologies.
Since radiography is a 2-dimensional representation of a 3-dimensional structure the quality of the image is very dependent on the positioning of the subject. Some patients will permit a complete radiographic study with minimal restraint but others will require sedation or anesthesia (2). The decision on whether or not to use sedation/anesthesia, and which protocol to use, depends on both the patient and on the radiographic study. Some patients may only require sedation to alleviate anxiety or fear while with others (e.g. trauma patients) pain control may be the primary objective (3) (Table 2). In some cases the risk of stressing critically ill patients during positioning for radiography is higher than the risk associated with sedating them, and the veterinarian should make a careful assessment of a patient’s state before proceeding (2).
The duration of the procedure and the type of manipulation required to position the patient must also be taken into consideration since some radiographic studies may be brief whilst others may require a longer time frame. In such cases anesthesia may be a safer option than strong sedation (2).
The positioning of the patient is extremely important since any errors that occur cannot be subsequently compensated. Generally four standard projections may be used (Table 3).
When the patient is in lateral recumbence, the lung parenchyma closer to the table cannot inflate completely, since all other structures above compress it and relative collapse ensues. This causes the contrast to be suboptimal since the more inflated the lung tissue, the more air there is to provide contrast. A similar situation occurs in the case of thoracic effusion provided the fluid is free to move within the thorax (1). This is summarized in Table 4.
It is also important that all thoracic radiographs be performed at maximum inspiration. Exceptions exist as in the case of radiographic studies to diagnose tracheal collapse where inspiratory and expiratory radiographs are used to distinguish between extrathoracic and intrathoracic collapse (1). Although these techniques can be useful they do not possess the specificity and sensibility of definitive dynamic studies such as fluoroscopy (4).
Esophagic contrast studies
In order to better delineate the soft tissue densities associated with the esophagus the use of contrast media is advised. The indications and contraindications for esophageal contrast studies are outlined in Table 5.
• Positioning of a patient in lateral recumbence. The front and hind limbs should be pulled cranially and caudally respectively; care should be taken in order to maintain them parallel and slightly elevated from the table in order to avoid rotation of the thorax or spine (Figure 1). The neck should be straight and pulled forward. Some patients will respond to manipulation with arching of the spine and pelvis which will result in thoracic misalignment even if the limbs are located correctly.
• Positioning of a patient in ventro-dorsal recumbence. The front and hind limbs should be pulled cranially and caudally respectively, care should be taken in order to maintain them parallel (Figure 2). The neck should be straight and stretched forward. If the limbs are excessively stretched some patients will arch their spine which will result in thoracic misalignment even if the limbs are located correctly.
Radiographic interpretation of lung patterns
The classical radiographic patterns described are the alveolar, interstitial, bronchial and vascular (1). The limitations with this approach are that these radiographic descriptions may overlap in some patients and present a poor correlation with microscopic findings. One example is the fact that the interstitium is composed of bronchioles, alveoli, blood and lymphatic vessels so the classification of interstitial pattern is insufficiently specific (5).
The normal pulmonary parenchyma is radio-transparent and only the blood vessels should be visualized. Alterations in radiographic density can vary and may be presented as slight increases that only obscure the visualization of blood vessels or lower airways. In more severe presentations the increase in density may lead to the inability to differentiate the normal thoracic morphology such as the heart or diaphragm. This increase in radiodensity is associated with an increase in severity of the clinical presentation (5). In extreme cases the pulmonary parenchyma can create a contrast with the airways that contain air and create air bronchograms.
Apart from soft tissue structures the pulmonary radiodensity also depends on the presence of air in the lung. So increased or decreased radiodensity may be encountered if the lung is respectively over-inflated or under-inflated. As a consequence the inflation of the lung lobe has to be taken into consideration since an increase in radiographic density may be a consequence of increased parenchyma density or decreased inflation. Signs of under-inflated lung lobes are cardiac/diaphragmatic displacement in the direction of the affected lung and crowding of the ribs since the lung occupies less space in the thorax (5). Increases in lung density due to decreased air content at the alveolar level are a consequence of underinflation. Increased lung density as a consequence of parenchyma alteration is associated with normal lung size.
An approach that expands on the classical pattern interpretation has been proposed and is based on including consideration of the location, radiodensity and pulmonary expansion of tissues (5). As mentioned earlier the pulmonary parenchyma also comprises blood vessels and lymphatics and their involvement can be evaluated by the observation of lines or rings, so that the presence of these findings is not exclusive to the lower airways. This new classification is demonstrated in Figures 3, 4, 5 and 6.
Abnormal organ position
The abnormal presence of abdominal organs in the thorax, which indicates the loss of diaphragmatic integrity, poses no difficulties in terms of diagnosis.
Several radiographic abnormalities may be identified in pathological alterations of the esophagus whether functional or structural in origin. The fact that the esophagus presents as a soft tissue density means that in normal conditions it is not easily recognized in radiographs. The exception to this is in the case of foreign bodies or the presence of gas.
This implies that for the correct evaluation of morphology, integrity and function contrast studies are necessary. Contrast studies may reveal the presence of radiolucent foreign bodies, dilation and perforation. Contrast media can also be used in studies to evaluate function as in fluoroscopy and sequential radiographs.
The use of contrast studies may even delineate abnormalities in the esophago-gastric continuity (Figure 7).
Thoracic involvement by metastatic or primary neoplasia is frequently encountered. Radiographic techniques that involve anesthesia, multiple views and lung inflation may permit identification of small masses over 3 mm in diameter.
Pleural space anomalies
Two distinct radiographic patterns are commonly found in patients with pleural cavity pathology, namely increased radiolucency and increased radiodensity. Both abnormalities are space-occupying anomalies and will restrict tidal ventilation (Figure 8).
Pulmonary parenchyma abnormalities
The radiodensity of the pulmonary parenchyma depends not only on the structure of the parenchyma but also on the amount of air present in the alveoli.
The easy availability, non-invasive nature and simplicity required for the execution of radiographic studies make this technique one of the most useful in day-to-day clinical practice. It also serves for screening and as a first step for the consideration of further diagnostic protocols. However the clinician should take care with patient preparation and positioning to optimize results.
This article was kindly provided by Royal Canin, makers of a range of veterinary diets for dogs and cats. For the full range please visit www.RoyalCanin.co.uk or speak to your Veterinary Business Manager:
1. Suter PF, Lord PF. Thoracic radiography: a text atlas of thoracic diseases of the dog and cat. Switzerland, Wettswil 1984; 683-734.
2. Scrivani PV, Bednarski RM, Dykes NL, et al. Restraint methods for radiography in dogs and cats. Compend Cont Ed 1996; 18: 899-917.
3. Sigrist NE, Doherr MG, Spreng DE. Clinical findings and diagnostic value of post-traumatic thoracic radiographs in dogs and cats with blunt trauma. J Vet Emerg Crit Care 2004; 14: 259-268.
4. Macready DM, Johnson LR, Pollard RE. Fluoroscopic and radiographic evaluation of tracheal collapse in dogs: 62 cases (2001–2006). JAVMA 2007; 230: 1870-1876.
5. Nykamp SG, Scrivani PV. Radiographic signs of pulmonary disease: an alternative approach. Compend Cont Educ 2002; 24: 25-36.
This article was first displayed on VetGrad.co.uk in 2012.