Cynthia R.L. Webster,     Grafton, Massachusetts

Johanna C. Cooper,     Walpole, Massachusetts

Laboratory Evaluation of Hepatobiliary Disease

Clinicopathologic evaluation helps to verify the presence of liver disease and determine the degree to which other organ systems are affected. The initial database should include a complete blood count (CBC), biochemical profile, and urinalysis.

The most consistent CBC abnormalities include changes in erythrocyte size and morphology, including microcytosis, target cells, poikilocytes, and Heinz body formation (cats). Microcytosis without anemia, most likely associated with impaired iron transport, occurs in dogs and cats with congenital PSVA and in dogs with acquired shunting secondary to portal hypertension. Target cells and poikilocytes result from alteration in the erythrocyte plasma membrane lipoprotein content, causing altered cell deformability. Anemia may accompany hepatic disease from a bleeding gastric ulcer, a coagulopathy, or the anemia of chronic disease. Some dogs with hepatobiliary disease have a mild thrombocytopenia, reflecting a systemic infectious disorder, synthetic failure from decreased hepatic thrombopoietin production, or secondary to a thrombotic episode.

Alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and γ-glutamyl transferase (GGT) are serum enzymes used as screening tests for hepatobiliary disease (Center, 2007). Increases occur as a result of (1) leakage from damaged hepatobiliary cells (ALT, AST); (2) elution from damaged membranes (ALP, GGT); or (3) increased synthesis (ALP). Although elevations in these serum enzymes have a high sensitivity for the detection of hepatobiliary damage, increases also occur in the absence of clinically important primary hepatobiliary disease.

This discordance occurs for several reasons. First, increases in serum hepatobiliary enzymes can originate from nonhepatic tissues. Second, in the dog endogenous or exogenous corticosteroids and phenobarbital can induce the production of excess hepatobiliary enzymes in the absence of liver damage. Finally, the liver is uniquely susceptible to secondary injury from primary disease in other organs, particularly the pancreas and gastrointestinal tract (see Box 139-1).

Elevations in ALT and AST result secondary to leakage from damaged hepatocytes. ALT is a liver-specific cytosolic enzyme; however, increases also may occur with severe muscle necrosis in the dog. The half-life () of ALT is days in dogs. No published values are available for cats; however, the is presumed to be much shorter (around 6 hours). The largest elevations in ALT occur with acute hepatocellular necrosis and inflammation. Mild-to-moderate elevations occur with primary hepatic neoplasia. AST, which is present in the cytosol and mitochondria, is more sensitive but somewhat less specific for liver disease than ALT. Increases in AST typically parallel those of ALT but are of a smaller magnitude. AST elevations in excess of ALT indicate either a muscle source or the release of mitochondrial AST caused by severe irreversible hepatocellular injury. The of AST is 22 hours in the dog and 77 minutes in the cat.

ALP is a membrane-bound enzyme (see Web Chapter 50). In dogs ALP has a high sensitivity (80%) but low specificity (51%) for hepatobiliary disease. Its low specificity is caused by the presence of several isoenzymes and its sensitivity to drug induction. In the dog three isoenzymes make up the total serum ALP (T-ALP), including a bone (B-ALP), liver (L-ALP), and corticosteroid (C-ALP) isoenzyme. B-ALP makes up one third of the T-ALP in dogs and is elevated in conditions with increased osteoblastic activity such as bone growth, osteomyelitis, osteosarcoma, and secondary renal hyperparathyroidism. L-ALP is present on the luminal surface of biliary epithelial cells and the hepatocyte canalicular membrane. L-ALP is 70 hours in dogs and 6 hours in cats. Because of the cats’ short and the fact that feline hepatocytes contain less ALP, increases typically do not approach those seen in canine patients. These characteristics and the fact that cats lack C-ALP make T-ALP less sensitive (50%) but more specific (93%) for liver disease than in the dog. The largest increases in L-ALP are associated with focal or diffuse cholestatic disorders and primary hepatic neoplasms. Less dramatic increases are found in chronic hepatitis, hepatic necrosis, and canine nodular hyperplasia. C-ALP is located on the hepatocyte canalicular membrane. C-ALP increases in dogs exposed to exogenous corticosteroids or in cases of spontaneous hyperadrenocorticism; however, increases also are associated with chronic illness, possibly secondary to increases in endogenous glucocorticoid secretion.

Hepatic GGT is located on the hepatocyte canalicular membrane. In dogs GGT has a lower sensitivity (50%) but higher specificity (87%) for hepatobiliary disease than T-ALP. If an elevated ALP is noted with a concurrent increase in serum GGT, specificity for liver disease increases to 94%. The most marked elevations in GGT result from diseases of the biliary epithelium such as bile duct obstruction, cholangiohepatitis, and cholecystitis. Moderate elevations accompany primary hepatic neoplasia, whereas mild elevations result from hepatic necrosis. In cats GGT has a higher sensitivity (86%) but lower specificity (67%) for hepatobiliary disease than T-ALP. Serum GGT may be considerably greater than ALP in some cats with cirrhosis, extrahepatic bile duct obstruction (EHBDO), or cholangitis. In feline idiopathic hepatic lipidosis GGT is typically only mildly elevated.

In dogs substantial increases in hepatobiliary enzymes can result secondary to corticosteroid and phenobarbital therapy. These increases may result from hepatobiliary enzyme induction and/or hepatocyte damage. Exogenous or endogenous excess of corticosteroids increases serum enzyme activity (L-ALP, C-ALP, ALT, and GGT) secondary to induction. In general, the most marked increases occur in ALP and GGT, with lesser elevations in ALT. Typically induction of AST is minimal. Cessation of corticosteroid therapy in the absence of liver damage results in a gradual normalization of enzyme values over a period of 2 to 3 months. Corticosteroids induce morphologic vacuolar change in hepatocytes and have resulted in focal areas of hepatic necrosis in experimental studies; therefore hepatocyte damage may be the cause of some of the increased enzyme activity.

Idiosyncratic hepatotoxic reactions to phenobarbital may occur in dogs, leading to chronic inflammatory disease or the hepatocutaneous syndrome. In both of these disorders moderate-to-marked increases in ALP, moderate increases in ALT, and mild increases in AST typically are seen. Phenobarbital therapy in the dog has been reported to induce the production of hepatobiliary enzymes (primarily ALP). Prospective studies to assess the relative role of induction versus damage in dogs on phenobarbital therapy have resulted in conflicting results. One study demonstrated no increase in ALT and ALP enzyme activity in whole-liver homogenates from phenobarbital-treated dogs, suggesting that induction was not occurring (Gaskill et al, 2005). However, another study found increased T-ALP activity in the liver of phenobarbital-treated dogs, supportive of induction (Unakami et al, 1987). Overall the available literature suggests that mild-to-moderate increases in ALP (up to five times the upper limit of normal) and ALT (usually less than two times the upper limit of normal) may reflect enzyme induction. However, increases in GGT and AST seldom are caused by induction and may be suggestive of primary liver disease.

The magnitude of serum hepatobiliary enzyme elevation is usually proportional to the severity of active hepatobiliary damage; however, the degree of elevation is not predictive of hepatobiliary functional capacity. Marked increases in these enzymes may indicate substantial hepatobiliary injury but are not necessarily indicative of a poor prognosis because of the tremendous regenerative capacity of the liver. Alternatively, normal or only mildly increased serum hepatobiliary enzymes may be seen in end-stage chronic liver disease because of replacement of hepatocytes by fibrosis and/or prolonged enzyme leakage, resulting in depletion of hepatic stores. Thus a single determination of serum hepatobiliary enzyme values has little prognostic significance. Prognostic value increases with sequential evaluation in conjunction with liver function testing and biopsy.

Assessment of liver function requires evaluation of parameters that reflect the synthetic and excretory capacity of the liver. Several hepatic function tests are included on routine biochemical testing, including bilirubin, glucose, cholesterol, blood urea nitrogen (BUN), and albumin. Although these tests are not particularly sensitive or specific, they are obtained easily and may help build a case for primary hepatobiliary disease. Hyperbilirubinemia, the most sensitive and specific of these parameters, in the face of a normal hematocrit is caused by hepatic disease resulting in inadequate uptake, conjugation and/or excretion of bilirubin, posthepatic disease interfering with biliary excretion of bilirubin, or the cholestasis of sepsis. In cholestasis of sepsis, cytokines released during sepsis inhibit the expression of hepatocyte transporters necessary for bilirubin transport. Cholestasis of sepsis may occur in the presence or absence of hepatobiliary damage and appears to be common in the septic cat.

Hypoglycemia occurs only when 75% of hepatic mass is nonfunctional as a result of decreased gluconeogenesis and clearance of insulin. Hypoglycemia also occurs periodically in dogs with congenital PSVA, possibly secondary to impaired glucose production, reduced glycogen stores, decreased responsiveness to glucagon, or a combination of these factors. Serum cholesterol levels in hepatobiliary disease are variable; cholestatic disease frequently is associated with hypercholesterolemia, whereas hypocholesterolemia is seen most often in end-stage liver disease. The BUN may be low in dogs with chronic liver disease or PSVA because the hepatic conversion of ammonia to urea decreases with decreasing hepatic mass or shunting of blood past the liver. The liver is responsible for the synthesis of albumin; therefore hypoalbuminemia may accompany chronic hepatic disease. Synthetic failure occurs only when 70% of hepatic functional mass has been lost. However, serum hypoalbuminemia also may occur with protein-losing nephropathies or enteropathies, vasculitis, blood loss, or from third spacing in ascitic patients.

Urinalysis in cases of hepatobiliary disease may reveal bilirubinuria or ammonium biurate crystals. The canine renal threshold for bilirubin is low and dogs are capable of tubular secretion of bilirubin; therefore bilirubinuria may be present in the absence of bilirubinemia. However, cats have a high threshold for bilirubin and are not bilirubinuric unless also bilirubinemic. Freshly collected urine samples may reveal ammonia biurate crystalluria in dogs and cats with PSVA.