SENSITIVITY OF RADIOGRAPHIC DETECTION OF LUNG METASTASES IN THE DOG |
| |
| Authors: | JOHANN LANG DR MED VET JEFFREY A. WORTMAN VMD PHD LARRY T. GLICKMAN VMD DR PH DARRYL N. BIERY DVM W. HARKER RHODES VMD M MED SCI |
| |
| Affiliation: | Department of Clinical Studies, Philadelphia, School of Veterinary Medicine, University of Pennsylvania, where Dr. Lang was a resident in Radiology, Dr. Wortman is an Assistant Professor, Dr. Glickman is an Associate Professor, and Drs. Biery and Rhodes are Professors. |
| |
| Abstract: | Thoracic radiographs were taken in right lateral recumbent (RLR), left lateral recumbent (LLR), and ventrodorsal (VD) projections in 80 dogs with known or suspected malignant tumors. Radiographs in each projection were interpreted, independently of those in the other two projections, by four veterinary radiologists, and classified as positive or negative for one or more lung metastases. The four radiologists then met, to agree on a diagnosis for each patient by simultaneously reviewing all three views. This panel diagnosis became the standard for subsequently classifying individual views as either true positive or true negative. As determined by a consensus of the panel, 25 dogs had at least one positive thoracic radiograph.
The three radiographic projections differed in sensitivity for the detection of lung metastases. In this study, the RLR view was the most sensitive, and the VD view the least sensitive. The sensitivity of detection improved significantly with multiple readers. From these data we recommend that, for detection of lung metastases in the dog, the RLR view be included in any diagnostic protocol; a minimum of two readers be used; and a three-view protocol be used if only one reader is available. |
| |
| Keywords: | radiographic sensitivity canine lung metastases |
|
|
