The Clinical Usefulness of the Ventrodorsal Versus Dorsoventral Thoracic Radiograph in Dogs
Differences exist in the ventrodorsal (VD) and dorsoventral (DV) radiographic views of the canine thorax. One view may be preferred over another because of how it portrays different areas of interest or different disease conditions. The VD view is indicated for evaluation of the cranial and caudal mediastinum, the caudal vena cava, and the accessory lung lobe, and in cases of pleural effusion. Indications for the DV view include assessment of a consistent cardiac silhouette, evaluation of the pulmonary lobar vessels, and evaluation of the structures of the dorsal thorax, such as hilar lymph nodes, the caudal dorsal lungs, trachea, mainstem bronchi, and left atrium.
Introduction
Radiographic projections are named for the pathway the X-ray beam takes through the body or for the point of entrance. A ventrodorsal view (VD) is made with the dog in dorsal recumbency, with the film or detector under the dog and the X-ray tube above the dog. The X-ray beam enters the ventral aspect of the thorax first, passes through the thorax, and exits the dorsal aspect of the chest. With the dorsoventral view (DV), the dog lies in sternal recumbency. The X-ray beam enters the dorsal aspect of the chest, passes through the thorax, and exits ventrally.1
The appearance of thoracic structures varies depending upon whether a dog is in dorsal or ventral recumbency.2–14 While some authors recommend the DV view for routine studies of the canine thorax, there are certain times when a VD view is preferred.2,7–11 The DV view is recommended for evaluation of the cardiac silhouette and pulmonary vessels, while the VD view is often preferred for evaluation of the lungs, especially in cases of pleural effusion.2,7–9
The purpose of this report is to identify differences and the clinical applications of the VD and DV thoracic radiographic views. Six clinical cases are described that illustrate the utility of these views.
Differences in Technique
Radiography is an essential step in the diagnosis of most thoracic diseases. Routine thoracic studies consist of a lateral view and either the VD or DV views. A high kilovoltage peak (kVp) (90 to 110) and low milliampere-second (mAs) (1 to 6) should be chosen for a long scale of contrast.1,15 It is best to obtain films during peak inspiration for full inflation of the lungs to provide maximum contrast.1,15 The shortest exposure time possible is chosen to minimize motion artifacts from respiration.15
To obtain a VD view, the dog is placed in dorsal recumbency, with the sternum superimposed over the spine. The front feet are held lateral to the head, and the head is straight, with the nose positioned exactly on midline. The pelvic limbs may be flexed. Positioning aids, such as sandbags, ropes, and a trough may be used to decrease exposure of personnel to ionizing radiation. The X-ray beam is directed vertically and is centered on the midthoracic spine.1
To obtain a DV view, the dog is placed in sternal recumbency, and the spine is superimposed over the sternum. The front feet are pulled cranially, with the elbows under the neck, so the scapulae are not superimposed over the thorax. If the dog does not tolerate this position, the elbows can be positioned off to the side. The head and neck are in line with the spine, with the nose positioned on midline. The pelvic limbs can be flexed. Just as for the VD view, positioning aids may be used. The X-ray beam is directed vertically and is centered on the midthoracic spine.1
Before interpretation, the technique and positioning of the film are evaluated. The field of view should include the entire thorax from the thoracic inlet to the most caudal aspect of the lungs. On both views, the lungs extend over the diaphragm, and it is important to include the entire lung field.15 The X-ray exposure is made so that the spinous processes of the vertebrae are visible and the peripheral lung fields can be evaluated without a hot light. When evaluating thoracic radiographs, it is important to take into account the divergence of the primary X-ray beam, which leads to magnification of body parts closest to the X-ray tube. Body parts closest to the film are less magnified. With less magnification, these parts have sharper margins and may be more easily identified. Magnification is often helpful, however, even though margins are not as sharp. Because of magnification, the ventral thorax is imaged more clearly on the VD view, and the dorsal thorax is clearer on the DV view.7 The VD view is often preferred for symmetry of thoracic structures, and it allows for maximum animal inspiration and for easier positioning. The DV view is less compromising to dogs with hypovolemic shock or severe dyspnea.9
The appearance of the diaphragm varies between dorsal and ventral recumbency. On a VD view, the X-ray beam is nearly parallel to the diaphragm, which leads to projection of the shadows of the cupula and crura into the thorax. This gives the diaphragm the appearance of having three convex projections [Figure 1A], with the cupula in the center and the two crura superimposed. On a DV view, with the X-ray beam centered on the thorax, the divergent beam hits the diaphragm at approximately 45° and causes superimposition of the crura with the liver. The diaphragm appears as one smooth dome with a continuous, curved contour [Figure 1B]. The peak of the dome lies slightly to the right of midline and represents the cupula.8 Beam centering also affects the appearance of the diaphragm. Two or three separate domes are seen when the animal is in dorsal recumbency and the beam is centered on the midthorax. These multiple domes are seen on the DV view if the beam is centered on the midabdomen.12
The VD view is preferred over the DV view for evaluation of the cranial and caudal mediastinum. More of the caudal vena cava is seen on the VD view [Figures 1A, 1B].2 If a mediastinal mass is suspected, a VD view is usually obtained.2
The shape of the cardiac silhouette is more variable on the VD view than on the DV view.9 The cardiac apex is not fixed to the sternum, so changes in posture cause displacement of the apex. On the VD view, the apex may move laterally (usually to the left) and dorsally. This movement causes elongation of the cardiac silhouette as compared to the DV view [Figures 1A, 1B].8,10 On the VD view, the heart is more parallel with the spine than on the DV view.2 The heart also changes position because of gravity. There is more contact of the right heart with the sternum on the DV view, and the apex of the heart tends to shift to the left.2,8 Ventral recumbency is probably safer for dogs with heart base or mainstem bronchial diseases, because there is less physical compression of these structures than in dorsal recumbency.
Subjectively the long axis of the cardiac silhouette is larger on the VD view, and the magnitude of difference increases as body weight increases.2 This increase in length likely results from magnification produced by the increased distance between the heart and the X-ray film with VD positioning.2 The width of the cardiac silhouette does not vary with dorsal or ventral recumbency, probably because the base of the heart does not vary in distance from the table top as compared to the apex.2
Both radiographic views are used to evaluate the caudal lobar pulmonary vessels. The pulmonary arteries are lateral to their corresponding veins on the VD and DV views [Figures 1A, 1B]. The artery and vein are usually similar in size and are no wider than the ninth rib where they cross it.6 Caudal lobar pulmonary vessels are more easily evaluated on the DV view.2,8,9 This difference may arise because the diverging X-ray beam is more perpendicular to the dorso-caudally oriented vessels. Also, these vessels are more magnified when the dog is in sternal recumbency. When the dog is in sternal recumbency, the dorsal lung fields are more inflated, and the air in these lung fields provides greater contrast for the caudal lobar pulmonary vessels.2 In contrast, the lung fields appear larger on the VD view.8 The accessory lobe is imaged better on the VD view. It is difficult to evaluate the accessory lobe on the DV view, because there is increased cardiac-diaphragmatic contact when the dog is in ventral recumbency.2
Thoracic computed tomography of three beagles demonstrated differences in X-ray attenuation in dorsal versus ventral recumbency.13 When the dogs were placed in dorsal recumbency, X-ray attenuation was increased in the dorsal regions of the thorax and decreased ventrally. When the dogs were placed in ventral recumbency, there were no differences in X-ray attenuation in the dorsal, middle, or ventral lung fields.13 Because of this increased attenuation in dorsal recumbency, VD radiographs may be less valuable in detecting lesions of the dorsal lungs. This attenuation does not occur when dogs are in ventral recumbency, so DV radiographs should be obtained for detection of lesions in the dorsal lung fields.
The most notable differences between the VD and DV views often occur when a moderate to large amount of pleural fluid is present. With pleural effusion, the lungs and cardiac silhouette are better evaluated on the VD view, because in this position free pleural fluid moves to the dependent portion of the thorax (owing to gravity) and accumulates away from the cardiac apex, and the dorsal thorax has a larger volume than the ventral thorax.10,11 The pleural fluid does not obscure thoracic structures as much, because it is distributed over a larger area, allowing for shallower levels. The fluid does not contact the heart or cupula of the diaphragm. There is no border effacement of the cardiac silhouette, because the fluid is dorsal to the heart.4 This fluid lies in the paraspinal gutters and between the lung and thoracic wall.8 The fluid moves caudally as well as dorsally because of the long spinous processes of the cranial thoracic vertebrae, which makes the cranial thorax farther from the table than the caudal thorax.11 Another cause for increased visibility of the cardiac silhouette on the VD view is the dorsally attached lungs floating up and surrounding the heart with air, providing contrast. On the DV view, the lungs pivot dorsally, allowing the fluid to accumulate between the lungs, heart, and diaphragm, causing border effacement.8 When the animal is in ventral recumbency, the fluid accumulates to a greater depth because of the smaller ventral chest cavity, thereby increasing opacity in the thorax [Figure 2].4,11 This ventral accumulation of fluid causes border effacement of the heart, diaphragm, and mediastinum.4,10
When there is only a small amount of pleural effusion, interlobar fissures may be the only radiographic abnormality detected. These fissures are more easily seen on the VD view. When the dog is in ventral recumbency, the fluid accumulates along the sternum and does not enter the interlobar fissures. If the amount of fluid is small enough, it does not cause increased opacity in the thorax, even on the DV view.14 If a small amount of effusion is suspected, the two views should be compared. If there are differences in the appearance of interlobar fissures, it is likely that free pleural fluid is present. If there is little difference in the two views, any pleural fluid is probably loculated.8 When pleural effusion is present, the appearance of the costophrenic sulci also differs when the animal is in dorsal versus ventral recumbency. When the animal is in dorsal recumbency, the pleural fluid flows dorsally and causes rounding of the sulci. In sternal recumbency, the fluid flows ventrally and the sulci have a normal appearance.4,8
Based on a principle similar to pleural fluid accumulation, the appearance of mild to moderate pneumothorax varies between the two views. When the dog is in dorsal recumbency, pleural air accumulates at the midline, so the air is superimposed over the spine and sternum. When the dog is in ventral recumbency, the air accumulates in the costophrenic sulci and is easier to detect.8 The Table summarizes the differences between the VD and DV views.
Clinical Applications and Discussion
Case No. 1
A 2-month-old, female bichon frise was evaluated for a heart murmur, possibly from a patent ductus arteriosus. Physical examination revealed tacky, pale mucous membranes; thin body condition; abnormal femoral pulses; and a continuous grade V/VI heart murmur that was loudest at the heart base. Thoracic radiographs [Figures 3A, 3B] revealed an elongated and widened cardiac silhouette. Pulmonary arteries and veins were enlarged, consistent with pulmonary overcirculation. The proximal descending aorta and main pulmonary artery were prominent. Findings were consistent with patent ductus arteriosus. The diagnosis was confirmed with echocardiography.
In this case, the DV view was helpful in that it allowed better visualization of the caudal pulmonary lobar vasculature. The aneurysmal dilatation at the proximal descending aorta was much more distinct on the DV view, because it was in the dorsal aspect of the thorax. When the dog was in sternal recumbency, the lungs in the dorsal thorax were better aerated, providing more contrast. Magnification of the structures of the dorsal thorax on the DV view also allowed for better visualization.
Case No. 2
A 5-year-old, castrated male German shorthaired pointer-cross was presented for treatment of heartworm disease. The dog had displayed some wheezing after exercise, but no abnormalities were detected on physical examination. Biochemical and complete blood count abnormalities included hyperproteinemia (8.8 g/dL, reference range 5.6 to 7.9 g/dL), with hyperglobulinemia (5.5 g/dL, reference range 1.8 to 4.2 g/dL), elevated creatine kinase (410 U/L, reference range 58 to 241 U/L), monocytosis (900/μL, reference range 100 to 800/μL), eosinophilia (2600/μL, reference range <750/μL), and basophilia (260/μL, reference range <100/μL). Microfilaria were observed on microscopic examination of whole blood.
On thoracic radiographs [Figures 4A, 4B], a diffuse interstitial pattern was present and most prominent in the caudal lung lobes. The caudal lobar arteries were increased in size when compared to their corresponding veins. The right caudal lobar artery was mildly tortuous, and the left caudal lobar artery was blunted. Findings were consistent with previously diagnosed heartworm disease.
In this case, the DV view was essential for evaluating the tortuous, enlarged pulmonary arteries that are a hallmark of heartworm disease in dogs. The caudal lobar pulmonary vessels are in the dorsal thorax; therefore, when the dog was in ventral recumbency, the dorsal lung fields were more aerated, providing better contrast for visualization of these vessels. These structures were also magnified. In addition, these vessels were projected over the liver, creating summation, increasing their radiopacity, and allowing better visualization. The cranial lobar arteries are often normal in heartworm disease, so evaluation of the caudal lobar vessels is critical.17
Case No. 3
A 13-year-old, castrated male Doberman pinscher was evaluated for a 1-week history of inappetence, lethargy, diarrhea, and enlarged lymph nodes. Small, fluctuant subcutaneous masses, peripheral lymphadenopathy, and a grade V/VI systolic murmur were detected on physical examination. Complete blood count abnormalities included monocytosis (970/μL, reference range 100 to 800/μL) and eosinophilia (1400/μL, reference range <750/μL).
Thoracic radiographs [Figures 5A, 5B] revealed a 4-cm, cavitated, soft-tissue mass within the right caudodorsal lung field, and slight dorsal elevation of the caudal trachea. An increased soft-tissue opacity within the perihilar region was causing separation of the caudal mainstem bronchi. Two soft-tissue nodules measuring approximately 3.5 cm and 2 cm, respectively, were located dorsal to the second and third sternebrae. The left auricle was prominent. Abdominal radiographs revealed enlarged sublumbar lymph nodes. The findings of a cavitary pulmonary mass together with perihilar, sternal, and sublumbar lymphadenopathy were most consistent with neoplasia or granulomatous disease. A fine-needle aspirate of a prescapular lymph node was consistent with lymphoma.
When a pulmonary mass is present, the location of the mass within the thorax determines whether the VD or DV view is more helpful. If the lesion is within the ventral aspect of the thorax, a VD view is better. A DV view is advantageous if the lesion is in the dorsal thorax. Although the cavitary pulmonary mass in this case was seen on the lateral views, it was not easily identified on the VD radiograph. The DV view allowed better visualization of the mass, because it was located in the dorsal lung fields. On the DV view, the dorsal lung fields were more aerated, allowing for better contrast of soft-tissue structures in these regions. Because of their dorsal location within the thorax, increased perihilar opacity and separation of mainstem bronchi were more apparent on the DV view. This case also illustrated the differences in appearance of the cardiac silhouette between the VD and DV views. The left auricle appeared larger on the DV view.
Case No. 4
A 12-year-old, spayed female poodle-mix was presented with a history of vomiting, lethargy, loss of appetite, polydipsia, polyuria, and mild pelvic-limb weakness. The only abnormal findings on physical examination were a small, subcutaneous mass on the dog’s back and dental disease. Biochemical abnormalities included a mild elevation in blood urea nitrogen (34 mg/dL, reference range 8 to 30 mg/dL) and elevated alanine aminotransferase (94 U/L, reference range 13 to 79 U/L). Complete blood count abnormalities included decreased hematocrit (33.2%, reference range 37% to 55%), low red blood cell count (4.78 × 106/μL, reference range 5.5 to 8.5 × 106/μL), decreased hemoglobin (11.1 g/dL, reference range 12 to 18 g/dL), mature neutrophilia (23,000/μL, reference range 3000 to 11,500/μL), lymphopenia (250/μL, reference range 1500 to 5000/μL), and monocytosis (1800/μL, reference range 100 to 800/μL). Serum thyroxine (T4) was mildly decreased (9.0 nmol/L, reference range 10.0 to 45.5 nmol/L).
Thoracic radiographs [Figures 6A, 6B] revealed generalized enlargement of the cardiac silhouette, with mild retraction of lung lobe margins and fluid within the pleural space. Partial border effacement of the cardiac silhouette and the caudal vena cava were seen on the lateral radiograph. The pulmonary vasculature was within normal limits. Because of the globoid shape of the cardiac silhouette, pericardial effusion was suspected. Echocardiography revealed pericardial and pleural effusion and a homogeneous, hyperechoic mass at the junction of the right ventricular and right atrial free walls. The owners elected not to pursue treatment, and the dog was lost to follow-up.
This case highlights the differences in visualization of the cardiac silhouette on VD and DV views when a moderate to large amount of pleural effusion is present. The VD view allowed for more complete assessment of the cardiac silhouette.
Case No. 5
A 10-year-old, spayed female rottweiler was presented with a 1-month history of a hacking cough, anorexia, lethargy, weight loss, and ptyalism. Abnormal physical examination findings included enlarged submandibular and prescapular lymph nodes, harsh lung sounds, and a dry cough. Laboratory abnormalities included decreased bicarbonate (14 mmol/L, reference range 16 to 26 mmol/L), elevated alkaline phosphatase (131 U/L, reference range 12 to 122 U/L), and increased anion gap (29, reference range 13 to 26).
Thoracic radiographs [Figures 7A–7C] revealed multiple, well-marginated, soft-tissue opacities within the pulmonary parenchyma. The largest of these opacities measured 8 × 9 × 9 cm and was located caudal to the carina. An area of increased soft-tissue opacity was detected ventral to the intrathoracic portion of the trachea between the second and fourth ribs, as well as border effacement of the cranial portion of the cardiac silhouette. The cranial mediastinum was wider than normal, and the margins were smooth and irregular. Radiographic findings were consistent with perihilar and sternal lymphadenopathy.
Fine-needle aspirates of the pulmonary masses were consistent with chronic, active inflammation. There was no evidence of sepsis or neoplasia. Serum ferritin levels were submitted and were 1500 ng/mL, nearly twice that of normal values (reference range 80 to 800 ng/mL). Elevated serum ferritin has been demonstrated in a dog with canine malignant histiocytosis.16 The dog was euthanized, and postmortem findings were consistent with poorly differentiated histiocytic sarcoma.
Because of their dorsal position in the thorax and better aeration of surrounding lung, the enlarged perihilar lymph node and the abaxial deviation of the caudal mainstem bronchi were much better visualized on the DV radiograph. However, the widening of the cranial mediastinum due to sternal lymphadenopathy was easier to evaluate on the VD view, because the sternal lymph nodes are in the ventral aspect of the cranial mediastinum. The VD view may be more useful in cases where the sternal lymph nodes are greatly enlarged.
Case No. 6
A 15-year-old, spayed female, mixed-breed dog was presented for evaluation of mitral and tricuspid valvular insufficiency. On physical examination, a grade IV/VI systolic murmur was heard over the left hemithorax, and the abdomen was severely distended with fluid.
Thoracic radiographs [Figures 8A–8C] revealed severe cardiomegaly with prominence of the left atrium and auricle, and distention of the caudal vena cava. The pulmonary veins were increased in size compared to their corresponding arteries. No abnormalities within the pulmonary parenchyma were noted. There was decreased detail in the visible portion of the abdomen. Radiographic impressions were that of severe cardiomegaly and pulmonary venous congestion. Echocardiographic findings were consistent with mitral and tricuspid valvular insufficiency. Abdominal ultrasonography revealed abdominal effusion and hepatic venous congestion, consistent with right-sided heart disease.
Because of their dorsal position in the thorax, the left atrial enlargement and pulmonary lobar venous distention were imaged better on the DV view. The pulmonary vessels were much more difficult to evaluate on the VD view because of their dorsal position within the thorax. The abaxial deviation of the caudal mainstem bronchi seen with left atrial enlargement was also easier to see on the DV view. Because the position of the heart changed within the thorax when the dog was in dorsal and ventral recumbency, the left auricular enlargement was more evident on the VD view.
Conclusion
As demonstrated in these six clinical cases, differences exist between the VD and DV views of the thorax. One view may be preferred over another for visualizing different areas of interest and evaluating different disease conditions. The VD view is indicated for evaluation of the cranial and caudal mediastinum, the caudal vena cava, the accessory lung lobe, and cases with pleural effusion. Indications for the DV view include evaluation of the caudal pulmonary lobar vessels, consistent assessment of the cardiac silhouette, and evaluation of structures of the dorsal thorax.
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Citation: Journal of the American Animal Hospital Association 42, 6; 10.5326/0420440
Ventrodorsal (A) and dorsoventral (B) views of a normal, 3-year-old, spayed female golden retriever. Structures identified include the lungs, cardiac silhouette, diaphragm (open arrows), caudal lobar pulmonary arteries (white arrows) and veins (black arrows), caudal vena cava (CVC), and region of the accessory lung lobe (ALL). The three open arrows are used in Figure 1A to identify the three convex projections of the diaphragm sometimes seen on the ventrodorsal view. Notice the region of the ALL is difficult to evaluate on the dorsoventral view. R=right, L=left.
Schematic representation of pleural effusion in ventral (left image) and dorsal (right image) recumbency. When the dog is in ventral recumbency, fluid surrounds the heart, causing border effacement. When the dog is in dorsal recumbency, the fluid does not surround the heart and is shallower, allowing visualization of the cardiac silhouette. V=vertebra, S=sternum, R=right, L=left. (Modified from Thrall DE. The pleural space. In: Thrall DE, ed. Textbook of Veterinary Diagnostic Radiology. 3rd ed. Philadelphia: WB Saunders, 1998:329.)
Figure 3A is a ventrodorsal radiograph of a 2-month-old bichon frise (case no. 1) with a patent ductus arteriosus. The arrowhead points to the aneurysmal bulge of the descending aorta. The arrow identifies the left caudal pulmonary lobar artery. Figure 3B is a dorsoventral radiograph of the same dog. The bulging of the proximal descending aorta (arrowhead) is more distinct in this view, and the left caudal pulmonary lobar artery (arrow) is also easier to evaluate. R=right, L=left.
Figure 4A is a ventrodorsal radiograph of a 5-year-old, castrated male German shorthaired pointer-cross with heartworm disease (case no. 2). The medial and lateral margins of the right and left caudal pulmonary lobar arteries are identified (arrows). Figure 4B is a dorsoventral radiograph of the same dog. Enlargement and tortuosity of the caudal lobar pulmonary arteries (arrows) are more easily seen on the dorsoventral view. R=right, L=left.
Figure 5A is a ventrodorsal radiograph of a 13-year-old, castrated male Doberman pinscher (case no. 3) with a cavitary pulmonary mass (white arrow), sternal and perihilar lymphadenopathy (black arrow), and left auricular enlargement (open arrow). Figure 5B is a dorsoventral radiograph of the same dog. The cavitary mass (white arrow), perihilar lymphadenopathy (black arrow), and left auricular enlargement (open arrow) are more obvious on the dorsoventral view. R=right, L=left.
Figure 6A is a ventrodorsal radiograph of a 12-year-old, spayed female poodle-mix (case no. 4) with pericardial and pleural effusion. The cardiac silhouette is clearly visible in this view. Figure 6B is a dorsoventral radiograph of the same dog. Note the increased opacity in the thorax on the dorsoventral view, with reduced visualization of the cardiac silhouette compared to the ventrodorsal view. R=right, L=left.
Figure 7A is a right lateral radiograph of a 10-year-old, spayed female rottweiler (case no. 5) with perihilar lymphadenopathy (black arrow), cranial mediastinal (white arrow) and sternal lymphadenopathy (open arrow), and pulmonary nodules and masses. Figure 7B is a ventrodorsal view of the same dog. The perihilar lymph node is identified by the black arrow. Figure 7C is a dorsoventral view of the same dog. The perihilar lymph node (black arrow) is delineated more clearly on this view. R=right, L=left.
Figure 8A is a right lateral radiograph of a 15-year-old, spayed female poodle-mix (case no. 6) with mitral and tricuspid valvular insufficiency. Left atrial enlargement (black arrow) is present. Figure 8B is a ventrodorsal radiograph of the same dog. The caudal lobar pulmonary vessels are enlarged (white arrow points to right one). The left atrium is identified by the black arrow. Figure 8C is a dorsoventral view of the same dog. Notice the increased visibility of the right caudal lobar pulmonary artery (white arrow) and the left atrial enlargement (black arrow) on this view. R=right, L=left.
