Application of the "Humanoid" Ventrodorsal Thoracic Radiographic View for Improved Cranial Thoracic Assessment

Summary

Objectives: To describe and provide examples of the clinical use of a non-standard thoracic radiographic projection that optimizes visualization of the cranial thorax.

Materials and Methods: Retrospective analysis of imaging reports and medical records of 44 dogs, with 5 selected as examples of the view’s advantages.

Results: The ventrodorsal thoracic view with caudal limb positioning, also termed “the humanoid view,” provides subjective utility in the detection, characterization, and localization of cranial lung lesions, mediastinal abnormalities, and body wall lesions, as well as the exclusion of pseudo-lesions.

Clinical Significance: Radiography remains the most widely available and accessible imaging modality for evaluation of the thorax in small animal practice.  The humanoid view illustrates that the application of optimized alternative views can provide clinically valuable information in cases that may otherwise have resulted in uncertainty, recommendation of advanced imaging, and additional cost to clients.
 

Introduction

Thoracic radiography is an important and frequently used diagnostic tool for the evaluation of numerous conditions in small animals. Routine thoracic radiographic examination is comprised of imaging in a minimum of two orthogonal planes, commonly evaluated in three views: left lateral to right lateral (right lateral: RL), right lateral to left lateral (left lateral: LL), and either ventrodorsal, or dorsoventral projections. Modification of the standard thoracic examination with removal or addition of views has been recommended according to the needs of individual cases [1-6]. Differences in the radiographic appearance of normal and abnormal thoracic structures depending on positioning have been described, attributed to the effects of gravity, superimposition of anatomical structures, magnification, and differing anatomical contrast provided by greater expansion of the non-dependent lung [7-14].

In the standard VD thoracic radiograph, superimposition of appendicular structures overlying the cranial lung lobes, mediastinum, and thoracic body wall is minimized by cranially extending the thoracic limbs at the scupulohumeral joints. Due to the mobility of the scapulothoracic muscular attachment (also termed the scapulothoracic synsarcosis or syssarcosis), this positioning of the limbs also results in rotation and mild angulation of the scapula, such that the dorsal margin of the scapula is directed caudally and the body of the scapula and associated musculature is superimposed on the cranial thorax [15]. This increases difficulty of interpretation in the region, due to the expected increase in relative opacity of the cranial lung fields when compared to caudal, and the opportunity for illusions of visual sensation and perception, such as foreground/background illusions or production of subjective contours [16,17]. Further, patient specific factors can result in difficulty achieving standard VD limb positioning – these include orthopedic, neurologic, and conformational abnormalities (e.g. chondrodystrophy) that limit range of motion or result in pain on manipulation, as well as behavioral variation such as paw touch aversion or resistance to forelimb extension.  In these cases, an alternative method of achieving a VD view is desirable.

The “human view,” or “humanoid view” (as herein termed) is an alternative positioning technique for VD thoracic radiographs that addresses these downsides of standard VD positioning by positioning the forelimbs in a caudal direction, rather than in cranial extension [3].  The purpose of this article is to draw attention to the humanoid view’s utility and provide clinical examples of its use, having been applied clinically at the authors’ institution for two decades.

Materials and Methods

Thoracic radiographic studies of canids consisting of at least RL, LL, VD and/or DV views, and the humanoid VD view performed at the Veterinary Health Center at Kansas State University from 2005 to 2020 were retrospectively reviewed in Carestream Vue PACS (Carestream Health, Rochester, NY).  All indications, ages, and reproductive statuses were included; incomplete series were excluded.  Forty-four cases were located that met requirements, and the search was discontinued; this does not represent the totality of cases during this time period which would have met inclusion criteria.   Of these, five cases were selected based on author consensus regarding value in illustrating indications for and advantages of the humanoid view.  Medical records of these cases were reviewed.

Performing the Humanoid VD

Animals are placed on the table in dorsal recumbency, with or without the use of a trough. Shielded personnel or positioning devices are utilized cranially to stabilize the head and cervical region, and caudally to position the thoracic limbs.  Caudal traction is placed on the thoracic limbs such that they are oriented longitudinally alongside the animal’s thorax with the manus pointing towards the tail, or slightly abducted, forming an acute angle with the body wall (Fig. 1).

This position can be maintained either by continuing to apply traction, or by placing restraint transversely across the cranial/dorsal aspect of the distal limbs and the ventrum of the cranial abdomen (Fig. 2).

The exposure is made using an overhead source with routine mAs and kVp for standard VD thorax.  The resulting projection resembles the human, anterior-posterior chest view, due to similarity in positioning of the thoracic limbs (Fig. 3).

Results

Case 1

A 14-year-old, male castrated Italian Greyhound presented for recheck thoracic radiography following incomplete excision and recurrence of histologically-diagnosed soft tissue sarcoma associated with the right abdominal body wall.  The initial examination consisted of RL, LL, VD, and DV views.  A rounded focus of homogeneous soft tissue opacity with sharp margins measuring ~7 mm in diameter was identified in the VD view only, centered within the right 3rd intercostal space and superimposed on the right cranial lung lobe and right scapula (Fig. 4). 

Initial differentials for this lesion included neoplasia (with metastasis of top consideration), superimposition/confluence of normal structures, or less likely pulmonary granuloma.  A humanoid VD was requested, in which the lesion was no longer detected, and pulmonary lesions were excluded (Fig. 5).  The final radiographic diagnosis provided was pseudonodule comprised of either adjacency of the right cranial lobar pulmonary vasculature, focal sclerosis of the scapula, or superimposition of an extra-thoracic soft tissue structure.

Case 2

A 7-year-old male castrated West Highland White Terrier was referred for restaging and potential surgical excision of a previously diagnosed cranioventral thoracic mass.  Prior radiographic examinations had been inconclusive for organ of origin of the mass, with both cranial mediastinum and right cranial lung lobe under consideration.  The initial examination consisted of RL, LL, and VD thoracic views.  An ovoid to triangular, homogeneously soft tissue opaque mass with sharp, rounded margins was identified in the right cranioventral thorax, extending from the level of the 2nd to 4th ribs.  Margins of the mass were sharper in LL projection when compared to RL, and a mass effect was appreciated displacing bronchi and pulmonary vascular structures dorsally along the margin of the mass (Fig. 6).

In VD, there was increased opacity of the right cranial lung field from the 1st to 4th intercostal spaces, though the cranial mediastinum was widened at the same level with laterally bowed margins that measured ~2.5x the width of the thoracic spine in VD (Fig. 7).

A humanoid VD projection was requested for further evaluation.  In the additional projection, the cranial mediastinum was assessed as normal, and the caudal margin of the mass was now distinctly seen, confirming the suspected pulmonary location (Fig. 8).

A top differential of primary pulmonary neoplasia was provided (e.g. pulmonary carcinoma, histiocytic sarcoma) with granulomatous pulmonary disease not entirely excluded.  Computed tomography was performed for surgical planning, and findings were consistent with a right cranial lung lobe neoplasm.  The patient was treated surgically via right cranial lung lobectomy, and the lesion was histopathologically diagnosed as a minimally invasive, papillary type, grade 1 pulmonary adenocarcinoma.

Case 3

A 7-year-old, female spayed Beagle presented with a 2-day history of tachypnea, anorexia, adipsia, and hiding behavior.  Per owners’ report, five year prior, the dog had been evaluated at another veterinary hospital with similar signs and had been treated for bacterial pneumonia with complete resolution of signs.  Thoracic radiographs were ordered, and RL, LL, VD, and DV views were acquired.  In RL and LL views, a focal region of peripherally distributed, cranioventral alveolar pulmonary pattern was seen, characterized by air bronchogram formation and complete effacement of the pulmonary vasculature (Fig. 9).

In both orthogonal views, the expected location of the pulmonary changes was superimposed with the scapulae and external soft tissue structures, and bilaterality could not be confirmed (Fig. 10).

In order to accurately gauge progression/resolution of radiographic signs, more precise localization was desired, and a humanoid VD was requested.  With abduction of the scapulae, the alveolar pattern was confirmed to be present within both the left cranial and right cranial lung lobes (Fig. 11).

Top differentials of aspiration or community acquired pneumonia, versus fibrosis secondary to reported prior pulmonary disease were provided. The patient was treated medically for presumed bacterial pneumonia, was discharged, and was doing well, eupneic, and of normal respiratory rate by owner report as of 3 days post-discharge.  No radiographic recheck was performed to the authors’ knowledge.

Case 4

An 8-year-old female spayed mixed-breed small dog presented for restaging of cytologically diagnosed, mediastinal large-cell lymphoma. The dog was currently under treatment with CHOP-based chemotherapy.  Recheck thoracic radiography was compared to the examination performed 6 weeks earlier at the time of diagnosis. A notable reduction in the dorsoventral dimension of the cranial mediastinal mass was noted when comparing lateral views. However, in standard VD, a definitive lateral dimension of the mass was unable to be confidently measured due to greater left sided scapular superimposition, as the subject resented forelimb extension during positioning.  Contribution to the left cranial hemithoracic soft tissue opacity by atelectasis could not be excluded (Fig. 12).

The requested humanoid VD view better revealed margins of the mass and improved the radiographic estimation of mass volume (Fig. 13).

The patient was determined to be in partial remission, indicated by a >50% reduction in volume of the mass, which was calculated using multiple dimensions from all available imaging planes.

Case 5

A 7-year-old, female spayed, mixed-breed terrier presented with a 3-day history of progressive, ascending tetraparesis, generalized hyporeflexia, and increased respiratory rate consisting of short, shallow breaths.  The patient had previously been diagnosed and recovered from idiopathic polyradiculoneuritis/Coonhound paralysis two years prior, and a second occurrence or recurrence of the same condition was expected.  Hypoxia was documented while breathing room air, with an arterial PaO2 of 78.8, and respiratory pattern was non-responsive to oxygen supplementation. Thoracic radiographs were ordered to exclude other causes for the respiratory signs.  In the standard VD projection, a focal region of relative increased opacity of the left cranial lung lobe was seen at the level of the left third rib space, with a faint, curvilinear gas lucency centered within (Fig. 14).

The LL and RL projections were unremarkable. To further assess if this opacity represented a focal alveolar pattern with air bronchogram formation, a humanoid view was requested.  By removing superimposition of the scapula, the axial region of the soft tissue opacity was seen to consist of the normal aortic arch, the lateral aspect was formed by a normal pulmonary vessel, and the curvilinear lucency related to aerated pulmonary parenchyma or bronchus juxtaposed between these structures (Fig. 15).

The thorax was ultimately assessed to be normal, and the dog was discharged with a presumptive diagnosis of and treatment for idiopathic polyradiculoneuritis.  The case was lost to follow-up.

Discussion

As illustrated by the included cases, the utility of the humanoid VD view in radiographic assessment of the cranial thorax is twofold: first, the view can provide added visual information that could confirm or exclude the presence of lesions, and second, the view can assist in anatomical localization of these lesions.  Outside the context of disease, knowledge of this alternative method of positioning might also serve helpful when positioning patients that resist cranial traction of the forelimbs. Though no included cases illustrate this point, a potential disadvantage of the humanoid view is the possibility to obscure thoracic body wall lesions due to summation of the appendicular structures with the subcutis and skin margins.

The value of the information provided by the humanoid view will vary based on case selection and interpreter skill. As such, it is recommended that the view be deployed on a case-by-case basis, and not unilaterally as part of a standard thoracic series. Efficient use of the view that balances diagnostic yield with patient/personnel exposure, equipment degradation, and technician time is recommended.

The major explanation for the improved visualization of the cranial thorax in the humanoid VD view is straightforward. With reorientation of the thoracic limb, the scapula and associated soft tissues are removed from the region of interest, reducing anatomical superimposition. However, it is worth considering that the humanoid position could also alter the biomechanics of respiration such that improved inflation of the cranial lung lobes could provide minor benefit to anatomical contrast within the lungs. In human patients with chronic obstructive pulmonary disease, changes in arm position have been shown to significantly alter measurements of functional residual capacity, inspiratory capacity, and activation of the accessory muscles of inspiration [18,19]. As demonstrated in high school athletes, maximal voluntary inspiration is significantly increased by leaning forward and placing hands on the knees, when compared to clasping hands over the head or resting them neutrally at the side [20]. Comparable spirometric or imaging-based thoracic volume measurements based on body position have not been investigated in canids, and it is possible that humanoid VD positioning could affect lung aeration.

Supplemental views of the thorax have seen varying adoption by practitioners, and are utilized according to case need, personal preference, comfort with interpretation, and hospital protocol. Among these are the thoracic oblique views, advantageous in altering superimposition of the spine and for achieving tangential assessment of peripheral lung and body wall masses; inspiratory/expiratory views, of value in assessing tracheal collapse, small volume pneumothorax, conditions resulting in air trapping/pulmonary emphysema, diaphragmatic paresis/paralysis, and cavitated lung lesions; the tangential ventral cervical view, used in radiographic evaluation of tracheal collapse and luminal tracheal lesions; and various horizontal beam views (standing lateral, decubitus ventrodorsal/dorsoventral), recommended for detection of small volume pneumothorax and pleural effusion [3,5,6]. In the authors' experiences, indications for utilization of the humanoid view occur at similar frequency as those warranting other supplemental thoracic views, and the view is rapidly and easily accomplished with minimal training of technical staff. 

Access to tomographic imaging in veterinary medicine has rapidly increased in the past decades, such that emergency and primary care veterinarians are often able to directly order or perform advanced imaging and are even more commonly able to refer to secondary or tertiary care facilities with these capabilities.  Nevertheless, radiography remains the most widely available, rapid, and familiar diagnostic equipment, accessible to a large majority of practitioners, and taught as an essential component of non-specialty education. Contemporaneously within the human medical space, use of computed tomography and magnetic resonance imaging has ballooned, igniting controversy regarding overuse of diagnostics as it pertains to patient outcomes, finances, liability protection, and utilization of hospital resources [21-24].  Veterinary professionals are acutely aware of these constraints, as they remain common barriers to care for animal patients [25-27]. The humanoid view serves as a reminder that application of alternative and creative radiographic techniques should not be discarded due to the potential to minimize cost to clients and risk to patients, even where advanced imaging might also provide answers to clinical questions. For complex cases that may ultimately require additional imaging, the up-front information provided by radiography for prognostication and cost estimates still argues for its use, and the humanoid view has the potential to amplify the yield of this common diagnostic, when applied appropriately.   

Limitations of this study include lack of definitive diagnosis or follow-up in some cases, and the subjective nature of authors’ assessment of the humanoid view. Future study might attempt to address this limitation by quantifying the diagnostic ability of practitioners presented with standard views alone vs. when given additional humanoid views in the presence of cranial thoracic disease. Additionally, study of the view’s utility in species other than canids could be considered.
 

References

1. Burk R. L. & Feeney D. A. (2003) Chapter 2: The thorax. In: Small Animal Radiology and Ultrasound: A Diagnostic Atlas and Text. 3rd edn. Elsevier-Saunders, St. Louis, MO, USA. pp 25-28.
2. Kealy J. K., McAllister H., & Graham J. P. (2005) Chapter 3: The thorax. In: Diagnostic Radiology and Ultrasonography of the Dog and Cat. 5th edn. Elsevier-Saunders, St. Louis, MO, USA. pp 208-214.
3. Muhlbauer M.C. & Kneller S. K. (2013) Chapter 2: Making quality radiographs. In: Radiography of the Dog and Cat: Guide to Making and Interpreting Radiographs. 1st edn. Wiley-Blackwell, Ames, IA, USA. pp 59-67.
4. Ober C.P. & Barber D. (2006) Comparison of two- vs. three-view thoracic radiographic studies on conspicuity of structured interstitial patterns in dogs. Veterinary Radiology and Ultrasound 47, 542–545.
5. Rudorf H., Taeymans O., & Johnson V. (2008) Chapter 1: Basics of thoracic radiography and radiology. In: BSAVA Manual of Canine and Feline Thoracic Imaging. 1st edn. Ed T. Schwarz and V. Johnson. BSAVA, Gloucester, UK. pp 5-9.
6. Thrall, D.E. (2018) Chapter 28: Principles of Radiographic Interpretation of the Thorax. In: Textbook of Veterinary Diagnostic Radiology. 7th edn. Elsevier, St. Louis, MO, USA. pp 568-582.
7. Avner, A. & Kirberger, R. M. (2005) Effect of various thoracic radiographic projections on the appearance of selected thoracic viscera. Journal of Small Animal Practice 46, 491–498.
8. Brinkman, E. L., Biller, D. & Armbrust, L. (2006) The clinical usefulness of the ventrodorsal versus dorsoventral thoracic radiograph in dogs. Journal of the American Animal Hospital Association 42, 440–449.
9. Carlisle, C. H. & Thrall, D. E. (1982) A comparison of normal feline thoracic radiographs made in dorsal versus ventral recumbency. Veterinary Radiology and Ultrasound 23, 3–9.
10. Groves, T. F. & Ticer, J. W. (1983) Pleural fluid movement. Veterinary Radiology and Ultrasound 24, 99–105.
11. Holmes, R. A., Smith, F. G., Lewis, R. E., et al. (1985) The effects of rotation on the radiographic appearance of the canine cardiac silhouette in dorsal recumbency. Veterinary Radiology and Ultrasound 26, 98–101.
12. Kirberger, R.M., Dvir, E. & Van der Merwe, L. L. (2009) The effect of positioning on the radiographic appearance of caudodorsal mediastinal masses in the dog. Veterinary Radiology and Ultrasound 50, 630–634.
13. Ruehl W. W. & Thrall D. E. (1981) The effect of dorsal versus ventral recumbency on the radiographic appearance of the canine thorax. Veterinary Radiology and Ultrasound 22, 10–16.
14. Spencer, C., Ackerman, N. & Burt, J. (1981) The canine lateral thoracic radiograph. Veterinary Radiology and Ultrasound 22, 262–266.
15. Rochat M.C. (2012) Chapter 51: The Shoulder. In: Veterinary Surgery: Small Animal. 1st edn. Ed K.M. Tobias and S.A. Johnston. Elsevier, St. Louis, MO, USA. p 692.
16. Buckle, C. E., Udawatta, V. & Straus, C. M. (2013) Now you see it, now you don’t: visual illusions in radiology. Radiographics 33, 2087–2102.
17. Farrow, C. S. (2003) Thoracic radiographic disease indicators.  In: Veterinary Diagnostic Imaging: The Dog and Cat. 1st edn. Ed. C. S. Farrow. Elsevier-Mosby, St. Louis, MO, USA. pp 368-373.
18. McKeough Z.J., Alison J.A. & Bye P.T.P. (2003) Arm positioning alters lung volumes in subjects with COPD and healthy subjects. Australian Journal of Physiotherapy 49, 133–137.
19. Kim K.S., Byun M.K., Lee W.H., Cynn H.S., Kwon O.Y. & Yi C.H. (2012) Effects of breathing maneuver and sitting posture on muscle activity in inspiratory accessory muscles in patients with chronic obstructive pulmonary disease. Multidisciplinary Respiratory Medicine 7, 9-14.
20. Skaggs J.R., Joiner E.R.A., Laguardia E., Sini M., Wren T.A.L., Woon R.P. & Skaggs, D.L. (2016). The effect of arm and body position on respiratory ventilation in high school athletes: A pilot study. International Journal of Athletic Therapy and Training 21, 30–34.
21. Greenberg, J. & Green, J.B. (2014) Over-testing: why more is not better. American Journal of Medicine 127, 362–363.
22. Owlia M., Yu L., Deible C., at al. (2014) Head CT scan overuse in frequently admitted medical patients. American Journal of Medicine 127, 406–410.
23. Schwartz, D.T. (2008) Counter-point: are we really ordering too many CT scans? The Western Journal of Emergency Medicine 9, 120–2.
24. Smith-Bindman, R., Kwan, M.L., Marlow, E.C., et al. (2019) Trends in use of medical imaging in US health care systems and in Ontario, Canada, 2000-2016. Journal of the American Medical Association 322, 843–856.
25. Access to Veterinary Care Coalition (2018) Access to Veterinary Care:  Barriers, Current Practices, and Public Policy. Available at: https://avcc.utk.edu/avcc-report.pdf (Accessed: 25 May 2020).
26. Kipperman, B.S., Kass, P.H. & Rishniw, M. (2017) Factors that influence small animal veterinarians’ opinions and actions regarding cost of care and effects of economic limitations on patient care and outcome and professional career satisfaction and burnout. Journal of the American Veterinary Medical Association 250, 785–794.
27. Stull, J.W., Shelby, J.A., Bonnett, B.N., et al. (2018) Barriers and next steps to providing a spectrum of effective health care to companion animals. Journal of the American Veterinary Medical Association 253, 1386–1389.