Systematic Interpretation of Thoracic Radiographs in Dogs and Cats

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Key Takeaways

• A minimum three-view thoracic radiographic study (right lateral, left lateral, and ventrodorsal or dorsoventral) is essential for comprehensive evaluation, with technical adequacy (exposure, rotation, respiration phase, positioning) assessed prior to interpretation to prevent artifact misdiagnosis.
• Systematic interpretation involves an extracardiac survey (pleural space, mediastinum, diaphragm), followed by cardiac and vascular assessment (e.g., Vertebral Heart Score in dogs, left atrial enlargement indicators), and finally, lung pattern recognition (alveolar, bronchial, interstitial, vascular, or mixed).
• Lung pattern classification is critical for differential diagnosis: alveolar patterns suggest edema, pneumonia, or hemorrhage; bronchial patterns indicate chronic bronchitis or asthma; and interstitial patterns can be structured (nodular/reticular) or unstructured, pointing towards neoplasia, granulomas, or fibrosis.
• Common pitfalls include hypovolemia mimicking interstitial disease, expiratory films simulating alveolar disease or cardiomegaly, and misinterpreting normal breed variations in cardiac silhouette or anatomical variants like the azygos vein.
• Pattern recognition guides differential prioritization, with subsequent diagnostic steps like echocardiography, CT, bronchoalveolar lavage, or fine needle aspiration employed to refine diagnoses based on clinical context and signalment.

Thoracic radiography remains the most frequently performed advanced diagnostic imaging study in small animal practice. It is a challenging and error-prone task for veterinarians, with interpretive accuracy depending on technical quality, systematic viewing habits, and familiarity with normal radiographic anatomy across species and body conformations. This article provides a structured framework for reading canine and feline thoracic radiographs, covering a checklist for evaluating the lungs, heart, great vessels, and thoracic cavity, with emphasis on pattern recognition and differential prioritization. It is written for practicing veterinarians who seek a reproducible method for interpreting both screening and problem-oriented thoracic studies.

The framework presented here integrates three complementary approaches: a technical quality assessment, a regional survey of all thoracic structures, and a pattern-based analysis of pulmonary parenchymal disease. The order of these steps matters. Evaluating technical quality before any diagnostic interpretation prevents the common error of attributing a positional or exposure artifact to disease. Surveying the entire thorax before focusing on the lungs prevents the equally common error of missing a lesion outside the primary region of interest. Pattern recognition then provides the differential list, and clinical context, signalment, and adjunctive imaging refine that list into a prioritized plan.

Interpretive confidence grows with experience, but even experienced radiologists benefit from a consistent checklist. The method described below is designed to be applied in the same sequence on every study, regardless of the clinical question. This consistency reduces omission errors and makes the interpretation reproducible across observers and over time.

At a Glance

Foundations of Thoracic Radiographic Interpretation

Physics and Image Formation Relevant to the Thorax

Thoracic radiography relies on differential attenuation of the x-ray beam by air, soft tissue, fluid, fat, and bone. The lung provides natural contrast because air-filled parenchyma attenuates far less than the soft tissue of the heart, vessels, and thoracic wall. This inherent contrast is both an advantage and a trap. The same high contrast that makes pulmonary lesions visible also makes subtle changes in lung opacity difficult to judge without a consistent exposure technique. Overexposure can obscure a mild interstitial pattern, while underexposure can create the false impression of increased pulmonary opacity.

The thoracic cavity presents unique technical challenges. Respiratory motion degrades image sharpness, so exposure should be timed to peak inspiration. In dogs, this typically requires manual exposure triggering or a short exposure time with high tube current. In cats, faster respiratory rates and the tendency to breath-hold during handling make patient cooperation more variable. The phase of respiration changes also lung volume but also the apparent size and position of the heart and vessels, so a film taken at expiration can mimic cardiomegaly or pulmonary disease.

Radiographic Anatomy as the Basis for Interpretation

Normal thoracic radiographic anatomy must be understood before any pattern can be recognized. The cardiac silhouette occupies the middle mediastinum, with its long axis oriented differently on the lateral versus the dorsoventral projection. The cranial mediastinum contains the cranial vena cava, brachiocephalic trunk, and left subclavian artery, which are variably visible depending on patient size and body condition. The caudal mediastinum contains the esophagus, descending aorta, and caudal vena cava. The pulmonary vasculature is evaluated by comparing the diameters of the cranial lobar arteries and veins on the lateral view and the caudal lobar vessels on the dorsoventral view. The trachea courses from the thoracic inlet to the carina, with its bifurcation located at the level of the fourth to sixth intercostal space in most dogs and cats.

Breed and species differences modify the normal appearance. Deep-chested dogs have a narrow, upright cardiac silhouette on the dorsoventral view, while barrel-chested breeds show a wider, more horizontal heart. Cats have a more horizontally oriented heart on the lateral view, and the normal feline cardiac silhouette occupies a larger proportion of the thoracic width than in most dogs. The radiographic anatomy of the thorax is best learned by correlating images with known anatomic structures, an approach that has been formalized in equine radiology and applies equally to small animals.

The Role of Systematic Viewing

The human visual system is biased toward salient features. A large cardiac silhouette or an obvious mass will capture attention, and the rest of the image may receive only cursory inspection. Systematic viewing counters this bias by forcing evaluation of every region in a fixed order. This is particularly important in thoracic radiography because clinically significant findings, such as a small pneumothorax or a subtle bronchial pattern, are easily overlooked when attention is directed elsewhere. Studies of computer-aided classification in canine thoracic radiographs have shown that automated systems can identify multiple concurrent findings, including cardiomegaly, alveolar patterns, and pleural effusion, but the clinical lesson is that multiple abnormalities are common and must each be sought deliberately.

The Systematic Viewing Sequence

A disciplined viewing sequence prevents the common failure mode of premature closure, where the first obvious abnormality halts further search. The sequence below applies to both dogs and cats and works for any thoracic projection.

Begin with the extracardiac thoracic cavity. Assess the pleural space for evidence of gas or fluid. In the dependent hemithorax, pleural fluid appears as a soft tissue opacity that blunts the costophrenic angle on the lateral view and obscures the cardiac silhouette and diaphragm on the ventrodorsal or dorsoventral view. Pneumothorax is recognized by retraction of the lung lobes from the thoracic wall, with the visceral pleura visible as a thin opaque line paralleling the body wall. In the cat, pneumothorax often produces a characteriztic dorsal displacement of the cardiac silhouette on the lateral view because the lungs collapse ventrally.

Next evaluate the mediastinum. The cranial mediastinum in the cat normally contains a visible thymic remnant in animals under one year of age, which can mimic a cranial mediastinal mass. The trachea should be traced from the thoracic inlet to the carina. Dorsal displacement of the trachea at the thoracic inlet in the dog suggests a cranial mediastinal mass or esophageal foreign body. In the cat, the trachea normally courses more dorsally, and subtle changes in its course are less reliable indicators of pathology.

The diaphragm is assessed for continuity and position. Loss of the normal cupula shape, focal elevation, or the presence of soft tissue opacity within the thorax that connects to the abdominal cavity through a diaphragmatic defect indicates rupture. The liver, stomach, or small intestine may be identified within the thoracic cavity. In the cat, herniated liver lobes appear as a rounded soft tissue opacity that may be mistaken for a pulmonary mass if the diaphragm is not carefully evaluated.

Cardiac and Vascular Assessment

The cardiac silhouette is evaluated for size, shape, and position. Vertebral heart score (VHS) remains the most widely used objective measure in dogs, with a normal range of 8.5 to 10.5 vertebrae in most breeds. The method measures the long axis from the ventral border of the carina to the cardiac apex and the short axis at the widest point perpendicular to the long axis, then compares both to the length of the thoracic vertebrae starting at T4. Breed variation is significant. Deep-chested breeds such as the Irish Wolfhound and Greyhound have lower normal values, while barrel-chested breeds such as the Bulldog and Pug have higher values. The VHS has not been validated as a reliable tool in cats, where subjective assessment of cardiac size combined with evaluation of the sternal contact and the vertebral body width at the cardiac base is preferred.

Left atrial enlargement is a key radiographic finding because it distinguishes clinically significant cardiac disease from incidental changes. On the lateral view, the left atrium is normally located caudal to the tracheal bifurcation and ventral to the caudal vena cava. When enlarged, it produces a focal bulge dorsal to the caudal vena cava and may elevate the tracheal bifurcation dorsally. A standardized method has been described in which a line is drawn from the dorsal border of the tracheal bifurcation to the crossing point of the dorsal border of the caudal vena cava and the most cranial crus of the diaphragm, if the left atrium extends dorsally beyond this line, it is considered enlarged. This method shows moderate correlation with echocardiographic left atrial to aortic ratio and may facilitate learning for less experienced observers, though it does not outperform subjective assessment in diagnostic accuracy.

The pulmonary vasculature is assessed on both views. On the lateral view, the cranial lobar arteries and veins are visible dorsal to the trachea. On the ventrodorsal view, the caudal lobar vessels are evaluated where they cross the ninth to eleventh ribs. The normal artery to vein ratio is approximately 1:1 in the dog and 0.8:1 in the cat. Arterial enlargement suggests pulmonary hypertension or increased pulmonary blood flow. Venous enlargement suggests left-sided congestive heart failure. Both arterial and venous enlargement occur with volume overload. In the cat, the pulmonary veins are best evaluated on the lateral view where they enter the left atrium, and venous distension is an early sign of left-sided heart failure.

Lung Pattern Recognition

The interstitial pattern is the most common and the least specific. A mild, unstructured interstitial pattern appears as a diffuse increase in lung opacity with preserved vascular visibility. This pattern is seen with normal aging, obesity, and mild pneumonitis. A structured interstitial pattern, where the opacity has a nodular or reticular character, raises concern for metastatic disease, granulomatous disease, or mycotic infection. The distinction between structured and unstructured interstitial patterns is clinically important because the differential diagnoses differ substantially.

The alveolar pattern is characterized by air bronchograms, obliteration of vascular margins, and lobar sign. Air bronchograms are the most reliable radiographic sign of alveolar disease and confirm that the opacity is within the airspaces instead of in the interstitium or pleura. In the dog, the most common causes are cardiogenic pulmonary edema, pneumonia, and pulmonary hemorrhage. Cardiogenic pulmonary edema typically has a perihilar distribution in the dog, while pneumonia is often cranioventral or lobar. In the cat, cardiogenic pulmonary edema is more variable in distribution and may be patchy or diffuse, making it difficult to distinguish from pneumonia on radiographic grounds alone.

The bronchial pattern appears as thickened, parallel lines or ring shadows representing the bronchial walls. This pattern is seen with chronic bronchitis, asthma, and, in older dogs, bronchiectasis. In the cat, a prominent bronchial pattern is characteriztic of feline asthma and chronic bronchitis. The presence of a bronchial pattern in a young dog should prompt consideration of parasitic bronchitis.

The mixed pattern is the most common presentation in clinical practice. Most pulmonary diseases produce more than one pattern, and the dominant pattern should guide the differential diagnosis. A perihilar alveolar pattern with an interstitial component in a dog with a history of cough and exercise intolerance strongly suggests cardiogenic pulmonary edema. A cranioventral alveolar pattern with air bronchograms in a febrile dog suggests bacterial pneumonia. A diffuse miliary interstitial pattern in a dog with lameness raises concern for systemic mycosis.

Differential Prioritization by Pattern

The Decision Tree for Common Abnormalities

When a thoracic radiograph shows an abnormality, the first decision is whether the abnormality is intrathoracic or extrathoracic. Skin folds, nipples, and artifacts from patient positioning can mimic pulmonary nodules or pleural effusion. Repeat the view with the patient repositioned if an artifact is suspected.

If the abnormality is intrathoracic, determine whether it is primarily pulmonary, pleural, mediastinal, or cardiac. Pulmonary patterns are evaluated as described above. Pleural disease is identified by the presence of fluid or gas in the pleural space. Mediastinal disease is suggested by tracheal displacement, a widened mediastinum, or a soft tissue mass. Cardiac disease is identified by cardiomegaly, vascular changes, or both.

For a pulmonary mass, the next decision is whether it is solitary or multiple. A solitary mass in an older dog is most likely a primary lung tumor, while multiple masses are more consistent with metastatic disease. A solitary mass in a young dog raises concern for granuloma or abscess. CT is the preferred next step for characterization and surgical planning.

For cardiomegaly, the next decision is whether left atrial enlargement is present. If it is, the dog is likely to have stage B2 or C mitral valve disease, and echocardiography is indicated to confirm the diagnosis and guide therapy. If left atrial enlargement is absent, the cardiomegaly may be due to physiologic adaptation, mild valvular disease, or cardiomyopathy, and echocardiography is still recommended for definitive assessment.

For pleural effusion, the next decision is whether the effusion is transudative, modified transudative, exudative, or hemorrhagic. Thoracocentesis with fluid analysis is the definitive diagnostic step. The radiographic appearance does not reliably predict the fluid type.

Documentation and Reporting

The radiographic report should follow a standard structure: patient identification, views obtained, radiographic technique, findings organized by system, and a conclusion that lists the most likely differential diagnoses in order of probability. The report should state the radiographic diagnosis explicitly and recommend the next diagnostic step. If the radiographs are non-diagnostic due to technique, this must be stated clearly and repeat views recommended.

The report should distinguish between findings that are definitive and those that are suggestive. An air bronchogram is definitive evidence of alveolar disease. A mild interstitial pattern is suggestive but not diagnostic of any specific condition. The report should also note the absence of significant findings, as this information is clinically useful for excluding differential diagnoses.

Serial radiographs are often more informative than a single study. Comparison with previous radiographs allows assessment of progression or resolution and is particularly valuable in monitoring response to treatment for congestive heart failure, pneumonia, and neoplasia. When previous studies are available, the report should comment on interval change.

The interpretation of thoracic radiographs remains a challenging task, and computer-aided diagnostic systems are under active development. Deep learning models have shown moderate to good performance in classifying common radiographic findings in dogs, including cardiomegaly, alveolar patterns, and pleural effusion, but they are not yet a substitute for systematic human interpretation. The radiologist should use these tools cautiously and always correlate radiographic findings with the clinical presentation.

Recognized Complications and Failure Modes

Thoracic radiography fails clinically when the image is technically inadequate, when a lesion is missed, or when a finding is misinterpreted. Each failure mode has a characteriztic signature that can be recognized early if the interpreter actively looks for it.

Technical inadequacy. Underexposed radiographs simulate interstitial disease because quantum mottle and reduced contrast obscure normal vascular margins. Overexposure washes out pulmonary vessels and makes the lung appear hyperlucent, mimicking air trapping or pneumothorax. Inspiratory films are essential: an expiratory film increases pulmonary opacity, enlarges the cardiac silhouette, and displaces the diaphragm cranially, producing a false alveolar or interstitial pattern. Motion blur from panting or tremor degrades vascular sharpness and creates a spurious bronchial pattern. The corrective action is to re-evaluate patient positioning and exposure settings before interpreting any abnormality. A radiograph that cannot be interpreted should be repeated, not reported.

Missed lesions. Peripheral pulmonary masses, small pneumothoraces, and mild pleural effusion are the most commonly overlooked findings. The dorsal recumbent view is superior for detecting small volumes of pleural fluid because it pools ventrally and outlines the lung margins. A subtle pneumothorax is best identified on the lateral view by retraction of the cardiac apex from the sternum or by a hyperlucent dorsal lung field with a visible visceral pleural line. Masses hidden within the cardiac silhouette or superimposed over the diaphragm require scrutiny of the full lung field on both projections.

Misinterpretation of normal variants. The azygos vein can be mistaken for a mass on the dorsoventral view. The caudal vena cava may appear tortuous in older dogs. The pulmonary veins are normally larger than the corresponding arteries in the caudal lung lobes, and this asymmetry should not be called cardiomegaly unless the left atrium is also enlarged. A standardized method for detecting left atrial enlargement, drawing a line from the dorsal border of the tracheal bifurcation to the crossing of the caudal vena cava and the diaphragmatic crus, correlates moderately with echocardiographic left atrial to aortic ratio and may help less experienced observers learn subjective assessment (A standardized method for left atrial enlargement on lateral radiographs).

Common Errors and Corrective Actions

Less experienced interpreters tend to overcall interstitial patterns, undercall cardiomegaly, and describe findings without prioritizing them. The most frequent errors are listed below with the discriminating check that resolves each one.

The single most effective corrective action is to develop a fixed viewing sequence and apply it to every study. Interpreting the lungs before the heart, or the heart before the extrathoracic structures, invites systematic omission. The deep learning literature confirms that even automated systems struggle with bronchial and interstitial pattern classification, which underscores the difficulty of these distinctions for human observers as well (Automatic classification of canine thoracic radiographs using deep learning).

Limitations of Current Evidence

The evidence base for thoracic radiograph interpretation in dogs and cats rests largely on expert opinion, retrospective case series, and extrapolation from human medicine. Prospective studies with blinded outcome validation are scarce. The radiographic lung pattern classification itself has not been validated against histopathology or computed tomography in a large prospective cohort, so the inter-observer agreement for interstitial versus bronchial patterns remains modest.

Expert opinion still differs on several points. Whether a mild bronchial pattern in an asymptomatic older dog warrants treatment is contested. The threshold for calling cardiomegaly on radiographs varies with breed and body condition score. The left atrial enlargement line method performs no better than subjective assessment in experienced observers, though it may aid teaching (A novel standardized method for left atrial enlargement). Deep learning classifiers show promise for automating pattern recognition, but their generalization across acquisition systems and patient populations is not yet established, and they do not replace clinical reasoning (Automatic classification of canine thoracic radiographs using deep learning).

Referral, Consultation, and Reporting Thresholds

Referral for specialist imaging or echocardiography is warranted when radiographic findings are equivocal but clinical signs are progressive, when a mass lesion requires characterization beyond radiography, or when cardiac disease is suspected and echocardiography would change management. Computed tomography is indicated for suspected pulmonary metastasis, mediastinal masses, and traumatic diaphragmatic injury where radiography is inconclusive. The American College of Veterinary Radiology maintains resources on imaging standards and specialty consultation pathways (ACVR resources on diagnostic imaging practice).

Laboratory involvement is indicated when radiographic findings suggest infectious, inflammatory, or neoplastic disease requiring cytology, culture, or histopathology. Bronchoalveolar lavage, fine needle aspiration, or biopsy should follow radiographic localization of disease.

Regulatory reporting applies when radiographic findings suggest a notifiable disease. The World Organization for Animal Health terrestrial code defines reportable respiratory diseases that may present with thoracic radiographic abnormalities, and practitioners should consult current national requirements (WOAH terrestrial animal health standards). Reporting obligations vary by jurisdiction, and the responsible approach is to confirm local requirements before acting.

Frequently Asked Questions

How Should I Approach Thoracic Radiographs When Only a Single Lateral View Is Available?

A single lateral view limits detection of unilateral disease, particularly small masses, focal pneumonia, and mild pleural effusion. The right lateral view is preferred in dogs because it reduces magnification of the heart, but in cats the left lateral view often provides better cardiac evaluation. If only one view exists, interpret cautiously and prioritize the most conspicuous findings. Compare the cranial and caudal lung lobes, assess the cardiac silhouette against the sternum and diaphragm, and look for signs of pneumothorax such as cardiac elevation. Recommend a second orthogonal view whenever the clinical picture does not match the radiographic findings. The ACVR professional resources outline minimum standards for diagnostic image acquisition that support this recommendation.

What Are the Practical Limits of Radiographic Pattern Recognition in Obese or Thin Patients?

Body condition alters radiographic interpretation substantially. In obese patients, intrathoracic fat increases apparent soft tissue opacity, which can mimic or mask an interstitial pattern. The cardiac silhouette may appear enlarged due to epicardial fat, and the cranial mediastinum looks wider. In thin or emaciated patients, the lungs appear hyperlucent and vascular markings stand out prominently, which can be mistaken for a bronchointerstitial pattern. In both cases, compare the current study with prior radiographs when available. The MSD Veterinary Manual provides species-specific guidance on normal radiographic variation that helps distinguish true pathology from body condition artefacts.

How Do I Decide Between Radiography and Advanced Imaging for a Suspected Thoracic Mass?

Radiography confirms the presence of a mass but cannot reliably determine its origin or invasiveness. If the mass is solitary, well circumscribed, and confined to one lung lobe, surgical resection may proceed based on radiographs alone. If the mass is ill defined, located near the hilus, or associated with pleural effusion, computed tomography is indicated for staging and surgical planning. Ultrasound can characterize masses that contact the thoracic wall or heart base, and it guides fine needle aspiration. Radiographic signs of mediastinal involvement include tracheal elevation, esophageal displacement, and loss of the cranial cardiac waist. The deep learning classification study demonstrates that automated tools may eventually support mass detection, but they do not replace tissue diagnosis.

How Should I Document Radiographic Findings in the Medical Record?

Record the patient signalment, the views obtained, and the technical quality of each image. Describe each structure systematically: extrathoracic tissues, thoracic wall, pleural space, mediastinum, heart, great vessels, trachea, and lung parenchyma. Use standard terminology such as alveolar, bronchial, interstitial, and vascular patterns, and state the distribution of any abnormality. Include a radiographic diagnosis that prioritizes differentials by likelihood. Note any previous studies and whether changes are static or progressive. The AVMA practice resources provide guidance on medical record standards that support defensible documentation.

Does the Interpretation Framework Differ Between Dogs and Cats?

The framework is similar, but species-specific anatomy changes the emphasis. Cats have a more vertically oriented heart in lateral view, and the cardiac silhouette occupies a smaller proportion of the thorax. Feline bronchial patterns are more commonly associated with asthma or chronic bronchitis, whereas in dogs they suggest chronic airway disease or aging changes. Cats develop pleural effusion with cardiac disease more readily than dogs, and a mild effusion can obscure the cardiac silhouette entirely. The left atrial enlargement method described in a standardized method for left atrial assessment was validated in dogs, so extrapolation to cats requires caution. Always interpret feline radiographs against feline norms, not canine references.

How Do I Explain Radiographic Uncertainty to a Client or Referring Veterinarian?

Be direct about what the radiographs show and what they cannot show. State that a lung pattern indicates a category of disease, not a specific diagnosis, and list the most likely differentials. Explain that additional tests such as echocardiography, bronchoscopy, or computed tomography may be needed to confirm the cause. Avoid giving a false sense of certainty, particularly when the images are technically limited or the pattern is ambiguous. If a second opinion is warranted, say so explicitly and offer to arrange referral. The WOAH terrestrial animal health standards emphasize transparent communication in veterinary practice, which applies equally to diagnostic imaging discussions.

Related Clinical & Scientific Guides

• MRI Monitoring of Brain Tumor Response to Therapy in Dogs
• Ultrasound-Guided Drainage of Abscesses in Small Animals
• Radiographic Monitoring of Total Hip Replacement in Dogs

References and Further Reading

• Automatic classification of canine thoracic radiographs using deep learning.. 2021.
• A Novel Standardized Method for Aiding to Determine Left Atrial Enlargement on Lateral Thoracic Radiographs in Dogs.. 2023.
• Radiographic anatomy of the equine thorax as a basis for radiological interpretation.. 1983.
• Transdiaphragmatic intercostal herniation in the setting of trauma.. 2024.
• Intelligent Interpretation of Veterinary Musculoskeletal X-rays Trained with Human Thoracic Limb X-rays. 2021.
• American College of Veterinary Radiology Resources. American College of Veterinary Radiology.
• MSD Veterinary Manual, Professional Edition. MSD Veterinary Manual.
• American Veterinary Medical Association Practice Resources. American Veterinary Medical Association.
• WOAH Terrestrial Animal Health Code. WOAH.

Related Articles

• Dental Radiography Positioning and Interpretation in Dogs and Cats
• Echocardiography in Dogs and Cats: Indications and Basic Interpretation
• CT-Guided Biopsy of Thoracic and Abdominal Lesions in Dogs and Cats
• Thoracic Radiograph Patterns in Dogs: Differential Diagnosis and Clinical Approach
• Radiographic Assessment of Fracture Healing in Dogs and Cats

This article is educational professional reference material for veterinary audiences. It is not a substitute for veterinary diagnosis, individual clinical judgment, current product labeling, or applicable regulatory requirements.