Research Update: Is abdominal ultrasonography a high-yield diagnostic test in dogs with chronic vomiting?

Chronic vomiting (vomiting of more than two weeks' duration) is a common presenting complaint from owners of dogs evaluated at both primary care hospitals and referral institutions. The order in which diagnostic tests are performed in these cases is determined not only by the most likely differential diagnoses but also by the perceived answers that could be provided by each test when balanced with invasiveness, ease, and—of particular importance to owners—cost. For example, ultrasonography is a rapid, minimally invasive, and widely available technique for evaluating the abdominal cavity. However, this imaging modality is often cost-prohibitive, particularly with the likelihood of more testing being required afterward. Unfortunately, no studies have evaluated the utility of abdominal ultrasonography in dogs with chronic vomiting with the goal of identifying a subset of patients that would most or least benefit from this diagnostic test. The authors of this prospective study hypothesized that the diagnostic utility of abdominal ultrasonography would be highest in dogs with gastrointestinal neoplasia and lowest in dogs with chronic inflammatory gastrointestinal disorders, such as inflammatory bowel disease or gastritis.

For this study, a complete blood count, a serum chemistry profile, and abdominal ultrasonography were performed on 89 dogs with chronic vomiting. A survey radiographic evaluation before ultrasonography was not required or evaluated in this study. The ultrasonographic evaluation of the gastrointestinal tract included determination of gastric and intestinal wall thickness, appearance of wall layers, and evaluation of luminal contents, diameter, and motility. The decision of whether histologic evaluation of gastric or intestinal tissues was needed for a clinical diagnosis, as well as whether tissues were collected endoscopically or through celiotomy, was at the discretion of the attending clinician. Owners recorded the frequency of vomiting and diarrhea and any treatments that were administered for a minimum of five months or until patients were euthanized or died. Monthly evaluation of these logs was required to support presumptive clinical diagnoses and track disease progression. Only dogs with a final diagnosis of intra-abdominal gastrointestinal tract disease (including the pancreas) as the cause of their vomiting were included in the final study.

Upon a dog's discharge from the hospital, two clinicians not involved in the evaluation of that particular case were asked to confirm whether the final clinical diagnosis was correct and, if so, were asked to grade the diagnostic utility of the ultrasonography results on a five-point scale, ranging from being critically useful (diagnosis was obtained by ultrasonography) to being associated with a negative result (ultrasonography provided conflicting information or hindered diagnosis).

Both reviewers agreed that in 61 of 89 dogs (68.5%), the diagnostic utility of abdominal ultrasonography was low, meaning that the same diagnosis would have been reached without results of this imaging modality and that no new information that contributed to the final diagnosis was obtained. In 20 of 89 dogs (22.5%), both reviewers agreed that abdominal ultrasonography positively contributed to the final diagnosis. In the final eight dogs (9%), the reviewers disagreed about the usefulness of the results, and the average score (3.5 out of 5) reflected questionable diagnostic utility.

Chronic inflammatory disorders of the gastrointestinal tract (gastritis or inflammatory bowel disease) were diagnosed in 51 of 89 dogs (57.3%); thus, they were the most common diagnoses in dogs enrolled in this study. Gastrointestinal neoplasia (primary gastrointestinal lymphoma, gastric adenocarcinoma, and other primary intestinal neoplasms) was the second most common diagnosis in enrolled dogs, diagnosed in 20 (22.5%) patients. Other diagnoses included pancreatitis, small bowel foreign bodies, duodenal ulceration, and intussusceptions.

The diagnostic utility of abdominal ultrasonography was lowest in patients with chronic inflammatory disorders, with only one of the 51 (2%) dogs thought to have ultrasonographic findings that were important in reaching the final diagnosis. The diagnostic utility of abdominal ultrasonography was judged useful in five of seven (71.4%) dogs with lymphoma and was significantly higher in dogs with this diagnosis than in dogs with chronic inflammatory disorders. The diagnostic utility of ultrasonography in dogs with gastric adenocarcinomas was judged useful in three of 10 (30%) dogs and was significantly higher than in dogs with inflammatory bowel disease but not in those with gastritis or those with concurrent gastritis and inflammatory bowel disease. Other diagnoses in which abdominal ultrasonography was judged to be useful included pancreatitis (three of three [100%] dogs), small bowel foreign bodies (two of three [67%] dogs), duodenal ulceration (two of three [67%] dogs), and intussusceptions (one dog); however, the number of dogs with these diagnoses was too small to allow statistical analysis.

When select patient signalment and historical and physical examination findings were evaluated, the only factors that were significantly associated with increased diagnostic utility of abdominal ultrasonography were increased patient age (the older the patient, the more likely that ultrasonography would be judged to be diagnostically useful) and a clinical diagnosis of gastrointestinal neoplasia. Although the frequency of vomiting and percent of body weight loss were also initially found to be associated with increased diagnostic utility, these associations were not independent of age or diagnosis. However, despite the large number of cases in which abdominal ultrasonography was found to not contribute to the final diagnosis, in no case did results hinder or mislead clinicians.

COMMENTARY

This study provides preliminary support for more selective use of abdominal ultrasonography in dogs with chronic vomiting. Based on the population evaluated at the authors' referral institution, ultrasonography would have the highest diagnostic utility in an older patient with suspected neoplasia. In contrast, the yield of abdominal ultrasonography in a younger dog with minimal clinical suspicion of neoplasia will likely be lower, and the authors concluded that gastroduodenoscopy with biopsy would allow a definitive diagnosis without the added expense of ultrasonography.

Several caveats must be made before these conclusions are extrapolated to other dog populations or veterinary practices. First, the diagnostic utility of ultrasonography was only assessed in those patients with a final diagnosis of gastrointestinal disease. In many cases of chronic vomiting, it is unclear before ultrasonography whether the cause of this clinical sign is solely due to gastric or intestinal disease, and, thus, this imaging modality is used to evaluate multiple intra-abdominal organs. If the authors had included those cases in which the final clinical diagnosis was renal or hepatic disease, for example, then the reported diagnostic utility may have been unpredictably increased or decreased.

Second, all ultrasound examinations of dogs in this study were performed by a single board-certified radiologist. Thus, the likelihood of identifying subtle pathologic changes within the abdomen was likely higher than what would be anticipated if ultrasonography was performed by other specialists or general practitioners, and the diagnostic utility may have been higher than most veterinarians should expect.

Finally, because this study was performed at a referral institution, some causes of chronic vomiting, such as pancreatitis, foreign bodies, and intussusceptions, which may be easier to identify by nonspecialty trained ultrasonographers before referral, would have been lower in the final population. If this study had been performed at a primary care facility, these diagnoses may have made up a larger percentage of the study population and, thereby, increased the overall diagnostic utility of abdominal ultrasonography calculated by the study authors.

One of the most important points highlighted by this study is that veterinarians should continue to evaluate the utility of specific diagnostic tests in specific populations. Performing diagnostic tests on animals without fully appreciating their overall usefulness may lead to increased patient mortality secondary to unnecessary client expense. This report concentrates on a small group of patients with a particular subset of diseases, but it makes important points that enforce the problem-oriented approach to diagnosis and, thus, may affect the decision-making process when evaluating patients with other presenting complaints.

Leib MS, Larson MM, Panciera DL, et al. Diagnostic utility of abdominal ultrasonography in dogs with chronic vomiting. J Vet Intern Med 2010;24:803-808.

The information in "Research Updates" was provided by Scott Owens, DVM, and Barrak Pressler, DVM, PhD, DACVIM, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Purdue University, West Lafayette, IN 47907.