Introduction

Transthoracic ultrasonography for the assessment of non-cardiac related thoracic disease in small animals has become a valuable diagnostic procedure. One advantage of transthoracic ultrasound is that the thoracic wall and the surface of the lung and to some extent the pleura can be examined with transthoracic ultrasound in animals. Cranial mediastinal masses can usually be well visualized transthoracically and tissue sampling can in many instances also be performed. However, transthoracic ultrasound can be often unrewarding for intrathoracic anatomy, such as tracheobronchial lymph nodes, the esophagus, and lesions in the caudal mediastinum and thorax that are hidden from view by the aerated lungs, bony thorax or pneumothorax. Obesity poses an additional barrier. In these instances, either a computed tomography or endoscopic ultrasound examination of the thorax is warranted. Below is a description of how transthoracic ultrasound can complement the thoracic radiographic findings to increase diagnostic accuracy. Ultrasound-guided tissue sampling will be included in the discussion.

Indications

Transthoracic ultrasonography can be used to detect small amounts of free fluid, examine the internal structures of the thorax in the presence of larger amounts of pleural fluid, detect mediastinal masses, characterize pulmonary masses and examine the pleural surface. Furthermore, thoracic wall lesions, rib masses and the diaphragm can be assessed. Main limitations to examination of the thorax include pneumothorax and the aerated lung.

Determining the source of large amounts of pleural effusion can be aided with the use of transthoracic ultrasound. In conjunction with radiographs, neoplastic diseases of the lung, mediastinum and pleura may be detected. Furthermore, ultrasound-guided tissue aspiration or thoracocentesis may be performed for either cytological or histological analysis of specific lesions.

Technique

It is not necessary to clip the hair from the entire thorax in order to examine it. The cranial mediastinum can be examined in most cats and dogs simply by pulling the front limbs forward and applying alcohol to the lateral aspect of the cranial thoracic wall behind the upper limb. The same technique can be used to quickly screen for pleural effusion.

It may be necessary to examine animals with respiratory difficulties in sternal recumbency as they may not be able to tolerate lateral recumbency. If large amounts of effusion are present and the patient is dyspneic, evacuation of the thorax and stabilization of the patient should take precedence over performing the ultrasound examination. Both right and left lateral recumbency can be used in relaxed patients, the side to be determined based on the radiographic findings.

Multiple scan fields are often necessary for complete examination of the thorax. In practice, the main region of interest is generally determined from thoracic radiographs and if the fur is too thick to examine only by applying alcohol, then the hair of that region can be clipped. When pleural fluid is absent, it may be difficult to find a window for examination of the structures of interest.

The transducer is generally positioned intercostally and pivoted from cranial to caudal and dorsal to ventral. Therefore, microconvex probes with a small footprint are ideally suited for examination of intrathoracic structures. The transducer can also be placed in the thoracic inlet and directed caudally. The windows used for routine cardiac imaging can also be used for examination of non-cardiac structures.

Normal Structures

The subcutaneous tissues as well as the intercostal muscles and ribs can be examined. Ideally, high frequency linear array transducers should be used for this purpose. They provide the highest resolution images of the superficial structures of the thoracic wall. Intrathoracic structures can be examined with high frequency curved array probes. The pleural surface of the body wall and lung should be observed during respiration. Solid lesions originating from either the body wall or lung can be differentiated in this way. The lung and body wall appear to move in opposite directions and glide over one another during respiration. The parietal pleura will appear as a smooth, hyperechoic surface. The visceral pleura appears as a highly reflective smooth surface that produces reverberation artifacts. The mediastinum is poorly visible in normal patients. It is surrounded by the air-filled lung which prevents most of it from being examined. By examining the mediastinum through the thoracic inlet, the cranial vena cava, jugular veins and brachiocephalic veins can be identified. The aortic arch, brachiocephalic trunk and left subclavian artery can also be examined. Doppler ultrasound can be used to confirm which vessels are being examined. Normal mediastinal lymph nodes are generally not visible. The diaphragm is generally a structure that is examined during abdominal sonography. When pleural or peritoneal fluid is present, it can be better examined for defects, infiltrations or masses. Radiography is the imaging modality of choice for examination of the diaphragm.

Abnormalities

Thoracic wall: When a soft tissue swelling of the thoracic wall or destructive rib lesion is identified radiographically, ultrasound can be used to further differentiate the lesions. Invasion of thoracic wall lesions into the pleural cavity can be identified. Disruption of the contour of the parietal pleural is a major indicator of thoracic invasion.

Pleura: an irregular pleural surface can be detected with pleuritis, chronic effusions and carcinomatosis. Chronic granulomatous or fibrotic changes of the lungs may produce similar findings. Free pleural fluid can be readily recognized and can be either anechoic or more echogenic in nature. Anechoic fluid is generally associated with transudates and modified transudates. More echogenic fluid can be seen with hemothorax and pyothorax as well as neoplastic disease. Hyperechoic strands may be visible and indicate chronicity of the effusion. The mediastinum may be well visible when fluid is present in both the right and left thoracic halves. The cranial and caudal vena cava and mediastinal reflections are often visible. Focal accumulations of fluid may be due to trauma, hemorrhage or abscesses following puncture or foreign body migration.

Mediastinum: mediastinal widening can be recognized most readily with thoracic radiography. However, if they are too small, or hidden by the presence of fluid, they may go undetected. Masses are the most common cause of mediastinal widening and are often easily detected with ultrasound. Most masses in the cranial mediastinum appear similar regardless of their type. Lymphoma, histiocytic sarcoma, carcinomas and reactive lymph nodes and hematomas and thymomas may all appear similar. Ultrasound-guided tissue aspiration is required for a definitive diagnosis.

Lungs: the lung parenchyma can be visualized when they are atelectic, consolidated, or when a mass is present. The non-aerated lung appears similar to a small liver lobe in both instances. In lobar pneumonia, the lung lobe will appear hypoechoic and usually homogenous. It may or may not be reduced in size. Lung lobe torsions will appear similar, however, the lobe is generally increased in size when torsed. In consolidated lobes, fluid tends to accumulate within the bronchi and fluid bronchograms may be recognized. They appear as anechoic branching structures, similar to that of vessels in the liver. During therapy, when air returns to the alveoli and bronchi, reverberation artifacts can be identified within the small airways. Consolidated lung lobes should not be mistaken for herniated liver lobes in the thorax.

Disruptions in the smooth surface of the visceral pleura are the first indication that a pulmonary lesion exists. Nodules and masses usually appear as hypoechoic structures that seem to displace the air in the lung away from them. Large masses are often difficult to examine completely due to the overlying ribs and inability to see the lateral and deep margins of the mass. Neoplastic lesions are often solid, hypoechoic and homogeneous. Heterogeneity may also occur due to necrosis as can mineralization.

Fluid-filled cavities, such as abscesses, may also be detected in the pulmonary parenchyma. Radiographically, it is not always possible to differentiate solid from fluid-filled structures. They may occur secondarily to aspiration pneumonia or foreign bodies, either inhaled or migrating. Migrating parasites can also create fluid-filled cavities within the pulmonary parenchyma. Ultrasound-guided puncture of fluid-filled cavities will provide a more definitive diagnosis. If in endemic regions, fecal analysis for parasites should be performed when pulmonary cysts are detected.

Diaphragm: the presence of pleural and/or peritoneal fluid makes the assessment of the diaphragm in cases of rupture possible. When thoracic radiographs are suggestive but not definitive the diagnosis of diaphragmatic rupture, ultrasound can be used to assess the integrity of the diaphragm. However, the large structure, large surface area and domed shape of the diaphragm, as well as its numerous attachments, make it difficult to examine completely with ultrasound. In cases of suspected diaphragmatic rupture, the diaphragm is generally examined from a substernal approach. Congenital peritoneopericardial diaphragmatic hernias can be identified by the presence of abdominal contents in the pericardial sac. Hiatal hernias are better diagnosed radiographically with the help of positive contrast medium and fluoroscopy.

Ultrasound-Guided Tissue Sampling in the Thorax

Ultrasound-guided interventions of the thorax can be performed to remove pleural fluid and perform fine-needle aspiration or tru-cut biopsies of solid lesions or the lung parenchyma. Ultrasound-guided fluid aspiration is most helpful when small amounts of fluid are present and blind techniques are unrewarding. Twenty-two gauge needles are generally adequate for fluid removal, but 20 and 18 may be used when extremely large quantities must be removed or the fluid is hyperechoic or adhesive strands are evident floating within the fluid. A needle connected to an extension set with a 3-way stopcock attached to a 20-60 ml syringe is optimal when large amounts of fluid are present. Patients are ideally placed in sternal recumbency and the needle positioned in the ventral aspect of the thorax and cranial to the heart. Sedation is generally not required.

Fine-needle aspiration of mediastinal and thoracic masses can be performed using a 22-gauge needle by finding an acoustic window where there is no air-filled lung obstructing the path between the needle and the tissue to be sampled. Pulmonary masses are often solid and devoid of air, making the risk of pneumothorax lower than when the air-filled lung is sampled. Hemorrhage is a small risk and can be reduced by sedating the patient to reduce lung movement due to respiration. The thoracic radiograph is a valuable tool for localizing the mass by counting the ribs to target the ultrasound examination appropriately. If the pulmonary mass cannot be identified due to overlying air, the patient should be positioned in lateral recumbency with the affected side down to induce atelectasis. This will allow the lung lesion to be more easily identified and sampled. For mediastinal lesions, Doppler ultrasound can be useful for identifying large vessels in that region prior to aspiration. This is usually not a problem for larger masses that contact the thoracic wall. Tru-cut biopsies usually require general anesthesia so that respiratory motion can be completely controlled to avoid lung laceration. A coagulation profile should be performed. 16-18 gauge automatic or semi-automatic devices can be used. Masses should be at least 2cm in diameter if they are to be biopsied percutaneously. A small stab incision with a No.11 blade over the sampling site aids passage of the biopsy needle.