Jaehwan Kim

An improved understanding of the CT characteristics for histologically confirmed primary intestinal tumors would be helpful for guiding prognosis and treatment plans in affected dogs. This retrospective, multi‐center, analytical study aimed to evaluate the CT characteristics for the differentiation of adenocarcinoma, lymphoma, and spindle cell sarcoma (SCS) in dogs. Thirty‐seven dogs who underwent contrast CT and histopathological examinations were included (adenocarcinomas, n = 11; lymphomas, n = 12; SCS, n = 14). Quantitative and qualitative CT parameters, including tumor morphology, contrast enhancement pattern, Hounsfield unit (HU) value, and presence or absence of intraabdominal lymphadenopathy, were evaluated for each included small intestine tumor CT case. Adenocarcinomas tended to show endophytic growth, intestinal obstruction, and a heterogeneous enhancement pattern. Lymphomas tended to show exophytic growth, contrast enhancement of the intestinal tumor mucosal layer, a homogeneous enhancement pattern, and the presence of lymphadenopathies in the abdominal cavity. SCSs tended to show lobulated growth, a large cystic portion within the tumor, a heterogeneous enhancement pattern, a large size with fat stranding sign, and lower HU values in postcontrast images. Cut‐off values of the minimum diameter/fifth lumbar vertebral mid‐body height (≥5.80; area under the curve [AUC] = 0.97, P < 0.001) and minimum HU value/HU value of the aorta (≤0.26; AUC = 0.96, P < 0.001) were derived to discriminate SCS from the two other tumor types. In conclusion, contrast CT characteristics may be useful in differentiating small intestinal adenocarcinomas, lymphomas, and SCSs in dogs.

Magnetic resonance imaging (MRI) is a standard test for diagnosis and treatment planning in dogs with degenerative thoracolumbar intervertebral disc disease (IVDD). However, published studies evaluating three‐dimensional fast‐spin echo (3D‐FSE) pulse sequences for dogs with IVDD are currently limited. Aims of this retrospective, observational study were to compare findings from T2‐weighted two‐ and three‐dimensional fast spin‐echo sequences (2D‐ and 3D‐FSE, respectively) for a group of small breed dogs with thoracolumbar IVDD. Inclusion criteria were dogs with IVDD that underwent 1.5‐Tesla MRI using both 2D‐FSE and 3D‐FSE sequences. For each dog and sequence, five pathologic indices were recorded: epidural fat discontinuation, vertebral canal compromise, spinal cord signal change, disc degeneration, and nerve root compression. Two independent investigators also scored visibility of the facet joint, intervertebral foramen, nerve roots, spinal cord grey‐white matter differentiation, intervertebral discs, and epidural fat. The Wilcoxon signed‐rank test was used to evaluate the between‐sequence differences in pathologic indices and visibility scores. Interobserver agreement was measured using Cohen's weighted kappa along with 95% confidence intervals. A total of 21 dogs were sampled. The 3D‐FSE sequences had higher pathologic indices of vertebral canal compromise ( P = 0.020) and spinal cord signal change ( P = 0.046) than 2D‐FSE sequences. Furthermore, 3D‐FSE sequences had higher visibility scores for the facet joint, intervertebral foramen, and nerve root structures ( P < 0.001). Findings from the current supported the use of 3D‐FSE sequences over 2D‐FSE sequences for the evaluation of IVDD and visualization of spinal structures in small breed dogs.