Inflammatory bowel disease (IBD) and alimentary lymphoma (ALA) are two of the more common gastrointestinal (GI) disorders in cats. The clinical signs observed in both are quite similar; with the most common signs being anorexia, weight loss, vomiting, and/or diarrhea. This commonality makes diagnosis between the two disorders complicated, plus there is an increased chance of intestinal lymphoma developing when a previous diagnosis of IBD is made. Currently, the diagnosis of IBD and ALA depends on the histopathologic examination of tissue samples taken by biopsy of the small intestine.
Upon histology, IBD samples are usually characterized by an accumulation of lymphocytes and plasma cells (and to a lesser degree others cells such as neutrophils, eosinophils, and plasma cells) in the lamina propria of the stomach and small intestine. Alimentary lymphoma also involves the infiltration of lymphoid cells in similar locations; however, the infiltrating lymphoid cells in ALA are not limited and extend to involve and efface the epithelial lining, submucosa, tunica muscularis, and serosa. Because of the similarity of clinical signs and overlapping appearance histologically between the two diseases, diagnosis by veterinary pathologists and veterinarians can be difficult in some instances. Obtaining full thickness intestinal wall tissue samples can overcome this difficulty but not in all instances and this approach requires a more invasive procedure for the patient.
The researchers’ objective with this study was to model the influence of IBD and ALA on different complete blood count (CBC) and serum chemistry (SC) variables and determine if using three data mining approaches – naïve Bayes, decision trees, and artificial neural networks classification algorithms – could help distinguish between the two diseases.
This study was retrospective, examining 3 groups of 40 (120 total) client-owned cats which were identified as normal, IBD, and ALA. Twelve CBC variables were evaluated: hematocrit, hemoglobin, platelet count, red, blood cell count, white blood cell count, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, neutrophils, lymphocytes, monocytes, and eosinophils. The nineteen SC variables were: albumin, alkaline phosphatase, alanine transferase, calcium, cholesterol, choride, CO2, creatinine, glucose, phosphorus, potassium, total protein, sodium, urea nitrogen, anion gap, total bilirubin, direct bilirubin, indirect bilirubin and globulin.
The study results showed cats with IBD or ALA were more likely to have mild non-regenerative anemia associated with chronic disease and suppression of hematopoiesis as previously noted in other studies. Cats with IBD and ALA had significantly lower red blood cells counts than normal cats, and cats with ALA had higher white blood cell counts compared to the normal group. Also, cats with IBD and ALA had significantly lower plasma total protein and albumin concentrations compared to the normal group of cats.
The authors state that the study indicates utilizing prediction models using machine learning provides a method for distinguishing between ALA and IBD, ALA and normal, and IBD and normal. The naïve Bayes and artificial neural networks classifiers outperformed the decision tree. They comment that these models can provide another noninvasive diagnostic tool to assist in differentiating between IBD and ALA cases, and between diseased and nondiseased cats.
Related Topic:
