David C. Twedt, DVM, DACVIM
The identification of abnormal liver enzymes usually indicates liver damage but rarely provides a diagnosis or etiology. Abnormal liver enzymes are common and in a study of 1,022 blood samples taken from both healthy and sick dogs and cats, one diagnostic laboratory found 39% had ALP increases and 17% had ALT increases. Some of those animals may have had primary liver disease but likely most had secondary liver involvement from a primary non-hepatic disorder. As a clinician it is also difficult to know how to proceed when presented with a patient having abnormal liver enzymes but no clinical signs. It is important to remember that the liver has a great reserve capacity and often signs of liver disease do not occur until the liver damage is advanced.
Tests of Hepatocellular Injury
Increases in either alanine aminotransferase (ALT) or aspartate aminotransferase activity (AST) indicate hepatocellular membrane damage with leakage of these enzymes. This could be due to death of the hepatocyte or from hepatocyte degeneration where the membrane becomes permeable. Conceptually ALT and AST should be thought of as hepatocellular "leakage" enzymes. Subsequent to an acute, diffuse injury, the magnitude of increase crudely reflects the number of affected hepatocytes. The plasma half-life of ALT activity is about 2.5 days (60 hours) in dogs; however, concentrations may take days to weeks to decrease following an acute insult to the liver. Persistent elevations or increase over weeks is characteristic of chronic hepatitis in the dog or other chronic long standing damage. As a general rule, ALT increases should be investigated when they are greater than twice normal or persistently abnormal over weeks to months. Hepatic AST is located predominately in hepatocyte mitochondria (80%) but is also soluble in the cytoplasm. Because of the mitochondrial location, AST elevations are more sensitive for liver disease than ALT and reflect more significant cell damage. On the other hand, AST is less specific than ALT because of the presence in other tissues.
Following an acute injury resulting in a moderate-to-marked increase in the serum ALT and AST concentrations, due to their difference in plasma half- life, the serum AST will return to normal more rapidly (hours to days) than the serum ALT (days). Drugs such as phenobarbital and corticosteroids both can cause increases in ALT and AST.
Tests of Cholestasis and Drug Induction
Alkaline phosphatase (ALP) and gamma-glutamyltransferase (GGT) show minimal activity in normal hepatic tissue but can become increased in the serum subsequent to increased enzyme production stimulated by either impaired bile flow or drug-induction. These enzymes have a membrane bound location at the canalicular surface; ALP associated more with the canalicular membrane and GGT associated more with epithelial cells comprising the bile ductular system. With cholestasis and increased surface tension in the canaliculi and bile ductules increases and production of these surface enzymes is then upregulated. An increase in the serum ALP and GGT activity can be induced by endogenous, topical or systemic glucocorticoids, anticonvulsant medications (ALP only) and possibly other drugs or herbs. The plasma half-life for hepatic ALP in the dog is 66 hours in contrast to 6 hours for the cat and the magnitude of enzyme increase (presumably a reflection of the synthetic capacity) is greater for the dog than the cat. Bone source arises from osteoblastic activity and is elevated in young growing dogs before their epiphysial plates close or in some dogs with bone tumors or lytic lesions. One study identified that increased ALP concentrations in some dogs with osteogenic bone tumors tended to indicate a poorer prognosis, probably from diffuse bone metastasis. Osteoarthritis does not causes increases in ALP. In the adult without bone disease, an increased serum ALP activity is usually of hepatobiliary origin. Hepatic GGT is located predominately on the canalicular membrane and bile ducts. Chronic elevations in GGT tend to reflect hepatobiliary tract disease, with he most marked elevations resulting from diseases of the biliary epithelium such as bile duct obstruction, cholangiohepatitis, cholecystitis or neoplasia. In dogs GGT has a lower sensitivity (50%) but higher specificity (87%) for hepatobiliary disease than total ALP. If ALP is elevated with a concurrent increase in serum GGT, specificity for liver disease increases to 94%. Bone does not contain GGT and the administration of anticonvulsant medications to dogs does not cause an increase in the serum GGT activity.
The diagram below depicts a general algorithm for the workup of dogs that have abnormal liver enzymes. The identification of abnormal liver enzymes may occur when the sick patient is presented for evaluation or during a routine health screen in the healthy patient. Abnormal liver enzymes in the sick patient could either be the result of primary liver disease/damage or secondary, due to a multitude of other non-hepatic disorders. The most common cause of abnormal liver enzymes is in fact, not primary liver disease at all but rather the result of reactive hepatic changes occurring secondary to other non-hepatic causes. Generally, secondary hepatic changes are reversible once the primarily disease is treated. Successful resolution of the non-hepatic disease and continued abnormal liver enzymes would be a strong indication for further investigation of the liver for a primary disease process.
Evaluation of Liver Function
On a routine biochemical profile it is important to note the liver function tests including bilirubin, albumin, glucose, BUN, and cholesterol. Albumin is exclusively made in the liver and if not lost from the body (urine or GI), sequestered or diluted, a low albumin concentration would suggest significant hepatic dysfunction. It may take greater than 60% hepatic dysfunction for albumin concentrations to decline. Major clotting factors are also made in the liver (except 8) therefore prolonged clotting time suggests hepatic dysfunction. Decrease in hepatic production of coagulation factors suggests significant liver dysfunction and a poor prognosis. Decreases in PT, APTT. BUN and glucose occur with significant abnormal liver function and a guarded prognosis. Cholesterol concentrations are quite varied with liver disease and changes are rather nonspecific. In general cholestatic disorders have increases in cholesterol while portosystemic shunt cases have decreases in cholesterol.
It appears blood ammonia is infrequently evaluated with liver disease. Elevations in serum ammonia generally are associated with hepatic encephalopathy. Ammonia elevations best reflect portosystemic shunting rather than direct parenchymal damage.
The most sensitive function tests available are serum bile acids. The fasting serum total bile acid concentration (FSBA) and a 2-hour postprandial serum total bile acid (PPSBA) are performed and if abnormal suggests hepatic disease or portal vascular anomalies. Values greater than 25 µmol/L are considered to be abnormal. Elevations in bile acids suggest portosystemic shunting, cholestasis or hepatic parenchymal disease. The highest sensitivity of the test involves taking both a fasted and postprandial sample.
In the asymptomatic patient with an increased liver biochemical test(s) the increased value should be confirmed. If no likely explanation for the laboratory abnormalities can be found there are two courses of action that one can take; either begin a diagnostic evaluation of the patient starting with bile acid determinations, or re-evaluate the patient's liver enzymes at a later date. With the patient having no clinical signs a reasonable waiting time to repeat the liver enzymes is 6–8 weeks. If enzymes are abnormal at that time I suggest performing serum bile acids. With continued abnormal liver enzymes with or without abnormal bile acids I believe that the liver should be further investigated. If the bile acids are also normal significant liver function is also involved and the liver evaluation should be more aggressive.
Routine abdominal radiographs are helpful in determining liver size and shape and for detection of other intra-abdominal disorders. Ultrasonography is noninvasive, readily available and is the most informative initial imaging modality for liver disease. Frequently, fine needle aspiration (FNA) for cytological evaluation is performed. One should be cautious in over interpretation of those results however.
A biopsy is required for a definitive determination of the nature and extent of hepatic damage and to appropriately direct the course of treatment. The method for liver biopsy procurement may be surgery, needle biopsy or laparoscopy.
Abnormal liver enzymes should not be ignored and should be investigated in a systematic manner as previously discussed. Asymptomatic animals with no evidence of significant or treatable disease or in situations where financial constraints limit further work up the patient should be fed a quality maintenance diet for the patient's stage of life and the possibility of instituting specific liver support therapy should be explored. Finally, the most common cause for abnormal live enzymes are the secondary reactive hepatopathies resulting from a primary non-hepatic condition.
1. Center SA. Interpretation of liver enzymes. Vet Clin North Am Small Anim Pract. 2007;37:297–333.
2. Center SA. Diseases of the gallbladder and biliary tree. Vet Clin North Am Small Anim Pract. 2009;39(3):543–598.
3. Cole TL, Center SA, Flood SN, et al. Diagnostic comparison of needle and wedge biopsy specimens of the liver in dogs and cats. J Am Vet Med Assoc. 2002;220:1483–1490.
4. Comazzi SC, Pieralisi C, Bertazzolo W. Haematological and biochemical abnormalities in canine blood: frequency and associations in 1022 samples. J Small Anim Pract. 2004;45(7):343–349.
5. Dial SM. Clinicopathologic evaluation of the liver. Vet Clin North Am Small Anim Pract. 1995;25(2):257–273.
6. Ruland K, Fischer A, Hartmann K. Sensitivity and specificity of fasting ammonia and serum bile acids in the diagnosis of portosystemic shunts in dogs and cats. Vet Clin Pathol. 2010;39:57–64.
7. Wiedmeyer CE, Solter PE, Hoffmann WE. Kinetics of mRNA expression of alkaline phosphatase isoenzymes in hepatic tissues from glucocorticoid-treated dogs. Am J Vet Res. 2002;63(8):1089–1095.