Debra L. Zoran,     College Station, Texas

Cause of Protein-Losing Enteropathies

Classification of PLE in dogs is not based on the same system used in humans, in which the diseases are placed into one of three groups: (1) erosive GI diseases resulting in protein leakage across an abnormal barrier (e.g., Crohn’s disease), (2) nonerosive GI diseases associated with protein leakage across tight junctions or changes in protein uptake (e.g., intestinal parasites, cobalamin deficiency), and (3) diseases associated with increased lymphatic pressure and leakage of high protein lymph (e.g., lymphangiectasia, neoplasia) (Dossin and Lavoue, 2011). However, classification can be useful in consideration of therapeutic options and prognosis, and as such canine PLE generally are grouped according to their primary causes: (1) lymphangiectasia or crypt disease (e.g., Yorkshire terriers, rottweilers), (2) IBD (e.g., lymphoplasmacytic, eosinophilic, or granulomatous forms in many breeds), (3) breed-associated forms of PLE (e.g., enteropathies of basenjis, soft-coated wheaten terriers, and Norwegian lundehunds), and (4) miscellaneous diseases of the GI tract, many of which are associated with GI hemorrhage (e.g., infectious diseases such as histoplasmosis or parvovirus, severe parasitism with hookworms or giardiasis, and neoplastic diseases such as lymphoma). Lymphangiectasia is a relatively well described, if poorly understood, cause of PLE associated with dilation of lacteals, crypts, and lymphatic ducts that appears to be due to an inflammatory process but also may be idiopathic. A newly described but specific disease associated with PLE in dogs is crypt disease, a dilation of the intestinal crypts with mucus, sloughed epithelial cells, and sometimes inflammatory cells, which is not associated with histologic signs of IBD or lymphangiectasia. Further, the lesions may be isolated or patchy in distribution. They often are separated by large areas of normal mucosa and appear to be more common in Yorkshire terriers and rottweilers (Willard et al, 2000). PLE secondary to infiltration of the bowel wall or other diseases, such as IBD, lymphoma, or infectious agents, are the most common forms of the disease and can be variable in severity and presentation. However, German shepherd dogs appear predisposed to development of PLE secondary to these conditions. Primary (idiopathic or breed associated) PLE occurs in just a few canine breeds (i.e., Norwegian lundehunds, Maltese, and shar-peis) and causes the most severe changes in the lymphatic vessels, including development of lymphogranulomas around lymphatic vessels (Berghoff et al, 2007). Breed-associated severe IBD with concurrent PLE occurs in soft-coated wheaten terriers, and a rare immunoproliferative enteropathy affects basenjis; however, both represent severe forms of disease that can present a significant therapeutic challenge. The key feature of PLE is its wide variety of causes; the resultant clinical variability leads to an unpredictable clinical course and outcome.

Clinical Presentation and Diagnosis

The most consistent clinical feature of dogs with PLE is weight loss, especially loss of muscle mass, and may or may not be associated with diarrhea at the time of diagnosis. Diarrhea, either persistent small bowel diarrhea or chronic, relapsing, and intermittent diarrhea, is the next most prevalent clinical sign. Vomiting and inappetence, the other signs indicative of GI disturbance, are much less common and are unreliable indicators of PLE. Other signs of PLE occur as a result of the severe hypoproteinemia and include edema of limbs or ventrum, ascites, chylothorax, or pleural effusion; each of which may be present to a variable degree. Gut wall edema likely occurs in most dogs with PLE. It contributes to additional GI protein loss and increases the difficulty of getting high-quality intestinal biopsies. Other complications of severe PLE or IBD in dogs is malabsorption of fat-soluble vitamins (e.g., A, D, E, K) and decreased absorption of both divalent cations, calcium and magnesium, which in severe cases can lead to clinical hypocalcemia and resultant hypocalcemic tetany or seizures. Severe bleeding resulting from lack of vitamin K clotting factors has not been reported but is reasonable to consider. In addition to loss of albumin, other proteins of similar size and significance are lost from the GI tract. One protein in particular known to become deficient in PLE is antithrombin. Low antithrombin concentrations in dogs with PLE increase the risk of thromboembolism, a clinical complication reported in several separate case reports and studies of clotting profiles in dogs with PLE (Goodwin et al, 2011). However, the loss of antithrombin appears to explain only part of the process. Even dogs with PLE that have improved clinical values after treatment continue to have abnormal thromboelastograms. Furthermore, in a group of Yorkshire terriers with crypt disease and PLE, several had pulmonary thromboembolic events (Goodwin et al, 2011).

The diagnosis of PLE is a multistep process that requires a careful review of the patient’s history and signalment, including characterization of the diarrhea, review of any and all diet changes and therapy, and a thorough physical examination, followed by laboratory confirmation of panhypoproteinemia, hypocholesterolemia, hypocalcemia, and lymphopenia, the laboratory hallmarks of PLE. Once hypoproteinemia is confirmed, the first step is to rule out other causes of protein loss as previously noted. This evaluation should include a urinalysis and urine protein-to-creatinine ratio to rule out protein-losing nephropathy; assessment of liver function, either via bile acid assays, ammonia testing, or possibly liver biopsy; and imaging studies. Imaging may be used to assess cardiovascular function, if indicated, and to collect a sample of fluid accumulated to determine the type of effusion (particularly pleural, but also abdominal) and to rule out a septic or inflammatory fluid. Abdominal ultrasound also is important to complete the diagnostic picture and to rule out other causes of protein loss (e.g., lymphoma or histoplasmosis can be found on FNA of liver or spleen in many cases). Furthermore, ultrasound can be the ideal means of determining whether endoscopy is indicated. For example, if the lesions appear to be focal or located in a region of the intestinal tract not reachable by the endoscope, planning for possible surgical biopsies can be initiated. Imaging also documents the degree of effusion and is useful in guiding the operator when performing pleurocentesis. In addition to the above, measurement of serum cobalamin and folate levels is indicated to assess the need for cobalamin replacement therapy, which is a common complicating factor. Finally, all dogs should have coagulation parameters measured (especially antithrombin, D-dimers and, if possible, a thromboelastogram) to assess clotting status and serum vitamin D status assessed, particularly if iCa²⁺ levels are low.

Ultimately, collection of biopsies from the GI tract, either via endoscopy or full-thickness biopsies obtained in surgery, is necessary to make a definitive diagnosis, determine the cause, evaluate the extent and severity of the disease, and provide prognostic information. There is an ongoing unresolved discussion about the best approach for obtaining biopsies. Although surgical biopsies are necessary in a small number of cases, the author prefers endoscopic examination and procurement of biopsies because they allow visualization of mucosa, are less invasive, and require shorter recovery time. In addition, in the author’s practice, endoscopy of the upper and lower GI tract is the standard method to obtain biopsies from the stomach, upper small intestine, and ileum. As has been reported recently in cats with lymphoma, lymphangiectasia in dogs can be found more commonly in the ileum. Therefore, if biopsies are not taken from the duodenum and ileum, the diagnosis may be missed or in error (e.g., IBD found in the duodenum with lymphangiectasia only present in the ileum) (Casamian-Sorrosal et al, 2010). The key in using endoscopy is collection of appropriate samples for biopsy (e.g., samples that are of sufficient depth to include the submucosa) so that crypt lesions and other structural changes can be identified adequately. This, of course, requires a skill set that must be developed and also requires appropriate patient preparation. For example, before the endoscopic procedure, the degree of gut edema must be reduced by use of colloid therapy so that the operator can grasp mucosal and submucosal tissue with the biopsy forceps and not just edematous tips of villi. In addition, to maximize the lacteal dilation for diagnostic purposes, administration of 1 tsp (5 cc) corn oil (or another liquid fat source) per 10 kg body weight 3 to 4 hours before the procedure is recommended. In cases in which a surgical approach is deemed most appropriate or endoscopy has failed to make the definitive diagnosis, one or two biopsies should be taken from each of the same sites. Despite concerns for hypoalbuminemia being a risk factor for suture site dehiscence, two separate studies did not document this complication (Shales et al, 2005). Thus, in PLE cases in which an exploratory is indicated, the clinician should proceed using the same preanesthetic and procedure preparation and careful attention to surgical detail.