This presentation is centered on evaluating the current place of ultrasonography in assessing gastrointestinal tumors. Comparison to survey radiographs, and radiographic contrast studies will emphasize the value of this diagnostic tool.

Survey radiographs are very helpful in assessing the presence of mass/thickening of the gastrointestinal wall; however, the detection of such findings is highly dependent on the size and location of the lesion. Abnormal fluid/gas accumulation can also be detected proximal to the (sub)-obstruction site when the tumor creates significant narrowing of the lumen. The "gravel sign" is useful in identifying abnormal accumulation of heavy particles (like bony fragments) at a site of suboptimal transit, which is commonly seen with constrictive lesion such as intestinal carcinoma. However, that sign is not pathognomonic of a tumoral process as it can be seen in any disease associated with a (sub)-obstruction.

Radiographic contrast studies such as gastrogram, UGI are often necessary to better assess the size, location, shape and contours of the affected segment of the GI tract. Intramural lesions frequently are irregular and produce either eccentric or annular luminal narrowing with thickening or rigidity of the bowel wall. Pedunculated, intraluminal masses often appear as filling defects in the barium column. Extraluminal masses usually create gradual and smooth narrowing of the lumen. The contrast studies are less commonly performed since ultrasonography has taken a large part in diagnosing these lesions.

Ultrasonography of gastrointestinal tumors is now the modality of choice. The most common ultrasonographic sign of GI neoplasia is thickening of the wall associated with loss of layering.

Alimentary lymphoma (involving the GI tract and/or the mesenteric lymph nodes) is one of the most common sites of occurrence for feline lymphoma. In cats, the most common ultrasonographic findings were transmural gastric or intestinal thickening associated with diffuse loss of normal wall layering, reduced wall echogenicity, localized decreased motility, and regional lymphadenopathy. These features can be found in both the stomach and bowel. Most lymphosarcomas are associated with regional lymphadenopathy. Jejunal lymphadenopathy is a common finding in intestinal lymphosarcoma, and in some instances, is responsible for most of the mass-effect. Despite extensive wall involvement, mechanical ileus is rarely present. Similar ultrasonographic features of transmural thickening of the GI wall with decreased echogenicity and loss of layering, decreased motility at the affected site and regional lymphadenopathy are present in dogs.

Mesenchymal tumors (smooth muscle and GIST) are typically found incidentally within the GI tract of geriatric patients. These tumors can be discrete, small (2 to 3 cm in diameter) and of uniform echogenicity; but often, the tumors are large and can be cavitated. These tumors can be incidental findings during the ultrasonographic evaluation, and they uncommonly cause obstruction. In geriatric patients, leiomyoma can be an underlying cause for intussusception. Occasional cases of metastasis from tumors initially described as leiomyomas suggest that these typically benign lesions should, in fact, be monitored with sequential abdominal ultrasound examinations. Several studies have reported a detection rate of 50% or less to detect abdominal masses with radiographs. The use of positive-contrast upper GI studies increased the sensitivity of radiography but still failed in a significant number of animals. Transcutaneous abdominal ultrasonography is more sensitive than survey radiography in detecting GI leiomyosarcomas. These tumors originate intramurally and grow out of the serosa as large eccentric, extraluminal masses, or less commonly project into the bowel lumen. Because of their common exophytic distribution and their large size, it is difficult to assess the anatomical origin of the mass and even more so to determine the precise layer of tumor origin. During real-time evaluation, it is important to identify the gas and/or the small amount of fluid associated with the lumen within the mass. Large GI mesenchymal tumors tend to be inhomogeneous, with a mixed echogenic pattern. The presence of anechoic and hypoechoic foci within the mass may correlate with the areas of central degeneration and necrosis frequently found in these large lesions. In dogs, metastases from GI leiomyosarcomas, when they occur, frequently involve the liver and the mesenteric lymph nodes. In our experience, it is uncommon to detect hepatic or lymphatic metastases at the time of the initial examination. However, regular (every 2 to 3 months) ultrasonographic follow-up evaluations after surgical resection might be warranted to assess possible metastatic spread.

Carcinomas are considered the most common GI tumor in the dog. In the author's experience, gastric carcinomas appear to have an ultrasonographic distinction from those with an intestinal location.

The most common ultrasonographic findings are transmural thickening of the wall with altered wall layering. This altered layering appeared as a moderately echogenic zone surrounded by an outer and inner poorly echogenic lines. Because of this particular appearance and to avoid confusion with the term "layering" usually reserved to describe the normal appearance of the GI wall, we called this feature "pseudo-layering." This pseudo-layering seems to correlate with the unevenly layered tumor distribution noted histopathologically. This particular ultrasonographic feature, when present, may be indicative of gastric carcinoma.

Regional lymphadenopathy is present in most of the dogs with gastric carcinoma. The lymph nodes often have a target appearance, with a poorly echogenic rim and a hyperechoic center. The extent of the tumor and the presence of lymphadenopathy suggest that the diagnosis had been frequently made in the late stage of the disease with a consequently poor prognosis.

Intestinal carcinoma is well documented in dogs and cats, although it is much less commonly encountered in cats. A review in our institution of intestinal carcinoma in 14 dogs and 6 cats was performed. The most common ultrasonographic findings were transmural thickening with complete loss of layering, and often with associated regional lymphadenopathy. The lumen contour was moderately irregular and the lesion tends to be short (less than 4–5 cm long). In the majority of the cases, there is evidence of fluid accumulation proximal to the intestinal thickening/mass associated with localized ileus.

Intestinal carcinoma has several similar ultrasonographic features to intestinal lymphosarcoma. Signalment, clinical presentation and ultrasonographic signs may support a tentative diagnosis, but a final histopathological confirmation remains necessary to diagnose the lesion type with certainty.

Percutaneous ultrasound-guided fine-needle aspiration or automated microcore biopsy is a safe and valid alternative to endoscopic or surgical biopsy. The guided techniques of fine-needle aspiration using either a 22-gauge spinal needle or a 20-gauge Westcott needle, and of microcore automated biopsy using a 18-gauge Tru-Cut-like needle assisted by an automated biopsy gun are commonly used.

Other GI masses such as granulomatous lesions can also be encountered in dogs and cats. Underlying fungal or bacterial infections can be the inciting causes for the development of these masses. Feline eosinophilic sclerosing fibroplasia often appears as an inhomogeneous mass, this entity will be discussed in this presentation. Other lesions that can be confused as tumors as they present as focal thickening with loss of layering, are granulomas secondary to previously perforating foreign material, site of dehiscence or previous enterectomy sites.