Protein-losing enteropathies (Proceedings)
The protein-losing enteropathies (PLE) comprise a collection of intestinal, usually small intestinal, diseases typically associated with weight loss, hypoproteinemia caused by hypoalbuminemia or panhypoproteinemia, and variable signs of weight loss, vomiting and diarrhea. In dogs, PLE is most often a consequence of lymphocytic/plasmacytic enteritis (LPE) or other inflammatory infiltrative intestinal disease, intestinal lymphangiectasia (IL), or intestinal lymphoma; these diseases are the focus of this discussion. Because the presenting clinical findings and diagnostic features are similar for all three, they will be considered as a group with key differences pointed out where relevant. Readers should be aware that other diseases can contribute to the clinical manifestation of a PLE including hypoadrenocorticism, gastrointestinal ulcerative disease, other intestinal neoplasms, intestinal histoplasmosis and other chronic infections, gastrointestinal parasites, and chronic intestinal intussusception.
PLE can develop in animals of any age. Dog breeds that seem predisposed to the development of PLE include basenjis, Lundehunds, soft-coated wheaten terriers, Chinese Shar Peis, Rottweilers, German shepherds and Yorkshire terriers. Dogs are affected with PLE more often than cats despite the overlap in histological diagnoses of chronic intestinal diseases that exist between the species.
The clinical signs of PLE can be quite variable. Many dogs with PLE exhibit nonspecific signs, such as anorexia or lethargy and weight loss; some dogs may exhibit polyphagia. Diarrhea is one of the most consistent clinical signs referable to the GI tract, but is not seen in all patients with PLE. Those patients with diarrhea typically lack large bowel features such as mucus, fresh blood and tenesmus thus localizing to the small bowel. When diarrhea is not a presenting complaint, PLE should not necessarily be ruled out if otherwise compatible signs, such as weight loss, edema, or ascites, are present. Vomiting is also a frequent clinical sign in dogs with PLE. Dilation of intestinal lymphatics alone in dogs with IL does not seem to cause vomiting, but because many dogs with IL have intestinal inflammation, vomiting likely results from inflammatory stimuli.
Ascites, subcutaneous dependent edema, and pleural effusion are common in PLE patients. In most cases, fluid accumulates secondary to decreased colloid oncotic pressure from hypoproteinemia, particularly hypoalbuminemia. Effusions or edema may develop when serum albumin concentrations fall below 1.5 g/dl; effusions in these situations are usually pure transudates (low protein content, often less than 1 g/dl, and low nucleated cell counts). However, pure transudates in abdominal effusion can sometimes be seen in the context of serum albumin concentrations higher than 1.5 g/dl, suggesting that other pathophysiological factors that govern the formation of effusions, such as increased hydrostatic pressure or lymphatic obstruction, may contribute to fluid accumulation. It is important to remember that pure transudates in the abdomen can also be a consequence of portal hypertension. Chylothorax, and possibly chyloabdomen, can be seen in some patients with PLE secondary to IL.
In some dogs, clinical signs of thromboembolic disease, such as tachypnea and hyperpnea due to pulmonary thromboemboli, can develop; the author has also seen dogs with PLE die acutely from thromboembolic complications. The coagulation abnormalities that lead to thromboembolism in dogs with PLE are poorly characterized and understood but are attributed to imbalances between pro- and anti-coagulant factors, increased platelet aggregability or vascular endothelial injury, which have been documented in humans with inflammatory bowel diseases.
Laboratory abnormalities in dogs with PLE can be quite variable. Anemia of chronic inflammation and neutrophilic leukocytosis secondary to stress or chronic inflammation may be observed in some dogs. Some dogs with IL have lymphopenia, but lymphopenia could also be a component of stress leukograms in such patients. Platelet counts may increase from chronic inflammation. Thrombocytopenia in a dog with PLE is unusual and suggests a complication of the disease, such as thromboembolism or disseminated intravascular coagulation.
Hypoalbuminemia is the most consistent laboratory abnormality in canine PLE. Some dogs, however, have normal albumin levels. Commonly, serum globulin concentration is also low. Globulins can be increased, or remain normal in the face of enteric protein loss, secondary to increased production associated with the inflammatory process. Total serum calcium level is often decreased as an artifact of low serum albumin concentration. Occasionally, the total corrected calcium concentration remains abnormally low, and some dogs will have persistently low ionized calcium concentrations.
Other biochemical abnormalities commonly seen in dogs with PLE include hypocholesterolemia (especially with IL) and increased alanine aminotransferase (ALT) and alkaline phosphatase (AP) activity. Hypocholesterolemia is attributed to GI loss and lipid malabsorption. Increases in liver enzyme activities could be a reflection of some degree of concurrent hepatobiliary disease. Vacuolar changes in hepatocytes have been described in dogs with IL and increased liver enzyme activities. Trypsin-like immunoreactivity (TLI) is expected to be normal in dogs with PLE; TLI testing may be necessary to exclude exocrine pancreatic insufficiency in dogs with signs of diarrhea and weight loss that do not yet exhibit hypoalbuminemia/hypoproteinemia.
Urinalysis results are typically unremarkable in dogs with PLE, and there are no urinalysis abnormalities that would point to the possibility of PLE. However, urinalysis results are important to rule out hypoalbuminemia from renal losses, especially in dogs that are hypoalbuminemic with normal serum globulin concentrations. A urine protein:creatinine ratio less than 1 generally excludes the kidneys as a major site of protein loss, but higher ratios do not eliminate GI albumin loss. It has been speculated that in some dogs with PLE, increased permeability of the GI tract to luminal antigens and immune responses to them may lead to glomerulonephritis and proteinuria.
Because of the overlap in clinical signs and laboratory features between PLE and hepatic disease, bile acid testing may be indicated in some patients to exclude hepatic insufficiency/failure as the cause of clinical signs or laboratory abnormalities (hypoalbuminemia, liver enzyme abnormalities). If the ability to differentiate PLE from hepatic disease is still unclear after bile acids testing, measurements of fecal alpha-1 proteinase inhibitor (see below), a protein approximately the same size as albumin that appears in the feces in states of enteric protein loss, can confirm the existence of enteric protein loss. Lastly, similarities in presentation between hypoadrenocorticism, especially the atypical forms of the disease caused by glucorticoid deficiency only, may mandate basal cortisol measurements, and ACTH stimulation to follow if basal cortisol is <2 µg/dl.
Complete physical examinations and laboratory assessments are important in patients with chronic vomiting, diarrhea, or weight loss to exclude non-GI causes of these signs. Panhypoproteinemia can suggest the possibility of a PLE, but as noted, some dogs will only be hypoalbuminemic. In such cases, exclusion of abnormal hepatic function, or urinary or third-space losses may be needed. Normal fasting and fed bile acids would make abnormal liver function an unlikely cause of hypoalbuminemia, and analysis of effusions should eliminate albumin loss into body cavities as a factor contributing to hypoalbuminemia in most patients. Anorexic patients may be candidates for adrenal function testing.
A number of methods can demonstrate GI protein loss in cases of suspected PLE. While most are not readily accessible to practitioners, measurement of fecal α-protease inhibitor (API) is a simple, noninvasive, and objective method of documenting abnormal GI protein loss in dogs. In normal dogs, API is not excreted to any appreciable extent into the intestinal lumen and, when present, cannot be digested by luminal bacteria. Increased excretion of fecal API would be indicative of abnormal intestinal permeability consistent with PLE. Fecal API measurement does not differentiate the causes of PLE.
Thoracic radiography would be appropriate for patients with signs of respiratory disease that could reflect pleural effusion or thromboembolism, assessing lymph node enlargement or other lesions consistent with neoplasia or systemic inflammatory disease. Plain abdominal radiography is usually not helpful in patients with PLE, but could be considered to exclude partial GI obstructions that can cause similar clinical signs.
Abdominal ultrasonography can be useful in evaluating dogs with GI disease. Thickened small intestinal walls, hyperechoic mesentery, hyperechoic mucosal layer, indistinct wall layering, and small bowel hypermotility can be features of PLE of various causes. Peritoneal effusion, when present, is also readily apparent during abdominal ultrasonography. Abdominal ultrasonography can be normal in dogs with PLE.
Ultimately, definitive diagnosis of PLE is made following histologic assessment of intestinal biopsies obtained endoscopically or surgically. There are advantages and disadvantages to both techniques. Endoscopy typically is less costly and less time-consuming or invasive, important issues in unstable animals. Endoscopy permits direct visualization of the mucosal lesion, and dilation of lacteals in villous tips in patients with lymphangiectasia is readily appreciated in many patients. Lesions of some causes of PLE, such as lymphangiectasia, can occur regionally in the intestine and could be missed due to short endoscope length, poor tissue sampling technique, or both. When endoscopy is used, samples from the duodenum and ileum should be acquired whenever possible to increase the likelihood of obtaining a diagnosis.
Surgical biopsies offer the advantage of being able to see all abdominal organs, and the entire length of the intestine can be evaluated. Acquisition of biopsy samples of adequate size and depth to permit detection of submucosal lesions is possible. Biopsies should still be obtained if no lesions are apparent.
Regardless of the method used to obtain intestinal biopsies, patients with PLE can pose anesthetic difficulties due to body cavity effusions and hypoalbuminemia, conditions that may be aggravated with crystalloid fluid therapy during the anesthetic period. Plasma transfusions or administration of human albumin (10 ml/kg of a 5% solution) before anesthesia may increase the albumin level short-term and reduce the chance of developing complications related to fluid support. Caution should be exercised when administering additional doses of plasma or human albumin in the days following the procedure because sensitized patients may be more susceptible to developing allergic transfusion reactions. Administration of colloidal fluids (e.g., hydroxyethyl starch) can also be useful during the anesthetic period for maintaining plasma oncotic pressure and adequate effective circulating volume. Use of colloids has been associated with an increased risk of bleeding due to interaction with von Willebrand factor, but the risk of bleeding in dogs appears to be minimal if smaller dosages (<6 ml/kg) are used.
If an exploratory laparotomy must be performed, it should also be considered that wound healing may be altered and the risk of suture dehiscence increased. Although the topic is controversial, hypoalbuminemia does not seem to impair wound healing after intestinal biopsies. However, intestinal pathology and malnourishment due to a generalized catabolic state may increase the potential for suture dehiscence. Poorly absorbable or nonabsorbable sutures used in conjunction with serosal patch grafting can help prevent dehiscence.
Treatment of PLE in dogs has three main thrusts: resolution of the underlying disease, when identified, dietary modification, and symptomatic therapy. It is not unusual to implement treatment regimens that address all three aspects simultaneously.
Prednisone at immunosuppressive dosages (2 to 4 mg/kg/day) is commonly used in the initial management of many dogs with lymphocytic/plasmacytic enteritis and intestinal lymphangiectasia. In severe cases or patients that respond poorly to glucocorticoids alone, azathioprine (1 to 2 mg/kg or 50 mg/m2 PO q24–48h) may be added. Cyclosporine has also been of documented benefit in dogs with inflammatory bowel disease.
Dietary strategies often play an important role in the management of dogs with PLE. For dogs with LPE, a strategy of feeding a novel protein/novel carbohydrate diet can be considered on the possibility that immune responses to food antigens drive intestinal inflammation. Although not studied in a clinically scientific manner, dogs with IL should probably be fed fat-restricted diets. In humans with IL, dietary modification is considered the most important aspect of treatment. In people, dietary fat restriction, especially of long-chain triglycerides, is associated with increased concentration of serum proteins, including albumin, less severe histologic evidence of inflammation, and resolution of clinical signs. Long-chain fatty acids have proinflammatory properties, and have been proposed as a cause of lipogranulomas. Medium-chain triglycerides (MCTs) are added routinely to the diet of affected humans to increase the caloric intake without increasing lymph flow.
Studies in dogs have shown that MCTs are incorporated into chylomicrons and contribute to the composition of intestinally derived lymph. Whether the volume of lymph produced is reduced with MCT supplementation has not been established in dogs with IL. Thus, feeding diets with a greater proportion of MCTs may diminish inflammation while preserving the favorable aspects (calories, vitamin absorption) of dietary lipids and potentially allow some dogs to derive clinical benefits from MCT supplementation. The decision to use MCTs should probably be made on a case-by-case basis and responses judged by improvement in clinical signs, weight gain, and laboratory values, especially serum albumin.
Although not studied in detail, patients with PLE may be prone to cobalamin deficiency. It is believed that some animals may experience diarrhea solely due to cobalamin deficiency. Therefore, measuring serum cobalamin concentration may be warranted in patients with PLE. If serum cobalamin concentration is below the reference range, supplementation may be considered at 50 µg/kg (up to 1000 µg) SC once a week for 6 weeks, then once every other week for 6 weeks, and then monthly. Supplementation may need to be continued long term because GI lesions may persist; administration frequency may be best determined by regularly measuring serum cobalamin concentration.
Symptomatic treatment of animals with PLE consists of transfusions of plasma or other colloids, typically to stabilize the patient before surgery. However, colloidal transfusion should be considered as only a short-term treatment because oncotically active particles are quickly lost into the intestine. Diuretics may help relieve clinical signs from severe ascites and thoracic effusion. Thoracocentesis or abdominocentesis may be needed in some dogs to acutely alleviate the clinical signs or discomfort associated with the presence of effusions.
Despite the recognition that dogs with PLE can be predisposed to thromboembolic disease, guidelines for prophylaxis of thromboembolic disease have not been developed for dogs with PLE. Thromboembolism prophylaxis is probably warranted in some cases, but neither criteria for selecting cases warranting prophylaxis nor reasonable protocols have been developed for such patients. Intuitively, dogs with severe hypoalbuminemia, thrombocytosis, or evidence of active inflammation may have a more urgent need for prophylaxis, but no controlled studies have supported such contentions. A possible anticoagulant with few side effects is low-dose aspirin (0.5 to 1 mg/kg bid), given for anti-platelet effects. Other drugs (e.g. clopidogrel [Plavix®]) that impair platelet function are of theoretical benefit, but have not been specifically studied for this purpose in animals with PLE. For dogs in acute stages of known or suspected thromboembolic disease, heparin, low-molecular-weight heparin, or warfarin is indicated. Therapy with warfarin, which is highly protein bound, is demanding because of inconsistencies of absorption from dog to dog and variations in plasma protein concentrations. Therefore, safe warfarin therapy requires close monitoring of coagulation parameters (prothrombin time, partial thromboplastin time). Low-molecular-weight heparin is attractive in terms of efficacy and safety but could be cost prohibitive for many cases. Thrombolytic drugs, such as urokinase, streptokinase, or tissue-plasminogen activator, can be considered in dogs with demonstrated thromboemboli and have been shown to be effective in managing acute thromboembolism in dogs with PLE, but efficacy of these drugs has not been fully evaluated.
The prognosis for dogs with PLE partly depends on the underlying disease. If the underlying disease can be resolved, the prognosis is good with the general exception of intestinal lymphoma, which in dogs tends to carry a poor prognosis. In cases in which the underlying disease cannot be effectively managed, the prognosis is guarded to poor. Some animals live for only a short time after the diagnosis is made and die as a result of disease complications. Owners should be aware that although resolution of the disease may be achieved in some cases, a more attainable goal may simply be control of clinical signs.
Barton CL, Smith C, Troy G, et al: The diagnosis and clinicopathological features of canine protein-losing enteropathy. J Am An Hosp Assoc 14:85–91, 1978.
Kimmel SE, Waddell LS, Michel KE: Hypomagnesemia and hypocalcemia associated with protein-losing enteropathy in Yorkshire terriers: Five cases (1992–1998). J Am Vet Med Assoc 217(5):703–706, 2000.
Kull PA, Hess RS, Linden CE, et al: Clinical, clinicopathological, radiographic, and ultrasonographic characteristics of intestinal lymphangiectasia in dogs: 17 cases (1996–1998). J Am Vet Med Assoc 219(2):197–202, 2001.
Littman MP, Dambach DM, Vaden SL, Giger U: Familial protein-losing enteropathy and/or protein-losing nephropathy in soft-coated wheaten terriers: 222 cases (1983–1997). J Vet Intern Med 14:68–80, 2000.
Melzer KJ, Sellon RK. Canine intestinal lymphangiectasia. Compend Contin Educ Pract Vet 2004; 24:953.
Peterson PB, Willard MD. Protein-losing enteropathies. Vet Clin Small Anim 2003;33:1061.