The clinical signs of hepatic insufficiency in horses are highly variable, nonspecific and depend on the extent and duration of hepatic disease.
Their onset often is abrupt, because typically at least 75 percent of the liver mass must be affected before the signs become apparent, regardless of the cause of hepatic disease.
The most common clinical signs are weight loss, hepatic encephalopathy, icterus and colic. Less common signs are hepatogenic photosensitization, diarrhea, bilateral laryngeal paralysis, hemorrhage and fever. Rarely reported clinical signs of hepatic insufficiency in horses are ascites, dependent abdominal edema, endotoxic shock and hemolysis.
Significant weight loss and failure to thrive are most consistently present during chronic hepatic insufficiency. However, chronic liver disease may be present without apparent weight loss. Weight loss is due to anorexia and the loss of normal hepatocellular metabolic activities.
Hepatic encephalopathy (HE) is a complex clinical syndrome characterized by abnormal mental status that accompanies severe hepatic insufficiency of any etiology.
The earliest phase of HE manifests as subtle behavioral changes. As it progresses, there may be signs of depression, head-pressing, circling, mild ataxia, aimless walking, persistent yawning and other inappropriate behavior (Photo 1). In advanced cases, somnolence develops and obtunded behavior ensues. At this stage horses often show aggressive or violent behavior interspersed with the periods of stupor. Occasionally seizures occur during the later stages of HE, but in general they are atypical.
The severity of encephalopathy corresponds to the degree of hepatic dysfunction; however, neither of these parameters correlates with type or reversibility of the underlying hepatic disease.
The diagnosis of HE is based on the presence of signs of cerebral dysfunction, with physical examination and laboratory findings compatible with liver disease. Other possible causes for the neurologic signs should be excluded, because there are no specific features of HE that enable this syndrome to be definitively distinguished from other encephalopathies.
Icterus, or jaundice, is caused by hyperbilirubinemia and is most apparent in nonpigmented skin, mucous membranes and the sclerae. When hepatocellular disease or obstruction to bile flow is present, icterus develops as a consequence of impaired conjugation of bilirubin and impeded excretion of conjugated bilirubin into the biliary tract, respectively.
Finally, certain drugs (steroids, heparin) can impede bilirubin uptake and conjugation by hepatocytes, despite an otherwise normally functioning liver.
Abdominal pain or colic may be a result of acute hepatic swelling or pressure from obstruction of bile flow. Signs of colic and liver disease also may be caused by alterations in intestinal motility. Interestingly, many horses with liver disease and signs of colic have clinically significant gastric impactions.
Diarrhea and hemostasis
Because the liver is responsible for the synthesis of numerous factors involved in coagulation, abnormal hemostasis may be a sequela to hepatic insufficiency. Clinical signs may vary from ecchymotic hemorrhages, to hemorrhage after trauma or venipuncture, to spontaneous hemorrhage (Photo 2).
Especially sensitive to hepatic disease is the synthesis of fibrinogen and the vitamin K-dependent factors (II, VII, IX, X), which have relatively short half-lives.
Although not common, a fever may be present in horses with hepatic abscesses, acute hepatitis, chronic active hepatitis, obstructive cholelithiasis, fatty liver failure or neoplasia.
Intravascular hemolysis is a rarely seen, but grave prognostic indicator of fulminate hepatic failure in horses. The exact cause of hemolysis is not known, but is believed to be the result of increased erythrocyte fragility.
Hypoalbuminemia and water retention can occur with chronic liver failure and may result in dependent edema. Because the half-life of albumin is relatively long in the horse, edema is a rare clinical sign. Dependent abdominal edema may form if there is significant portal hypertension and ascites.
The Kupffer cell plays an important role in removing bacterial endotoxin that is normally absorbed from the lumen of the intestinal tract and carried to the liver via the portal circulation. Failure of Kupffer-cell phagocytosis of endotoxin may result in clinical and laboratory evidence of endotoxemia.
Laboratory findings of hepatic insufficiency
Because massive hepatic disease must be present before alterations are seen with some laboratory tests, and because different liver functions are variably altered by disease, the laboratory diagnosis of hepatic disease can be challenging.
The most useful diagnostic tests for evaluation of hepatic disease in horses are quantitation of sorbitol dehydrogenase (SDH) and gamma-glutamyl transpeptidase or -transferase (GGT) activity and serum bile acids concentration (SBA).
In the face of clinically significant liver disease, at least one of the three former serum tests typically is abnormal. Although increases in SDH, GGT and SBA are highly specific for liver disease, they are not specific for the type of disease.
Sorbitol dehydrogenase (SDH) has been widely used in the evaluation of acute liver disease in horses. The short half-life of this liver cytosolic enzyme makes it ideal for the evaluation of acute ongoing disease, as values usually return to baseline within three to five days after a transient hepatic insult. Its short half-life necessitates analysis within hours of collection.
Although mild variations exist between laboratories, the normal blood activity SDH in horses is usually less than 8 units/L. Gamma-glutamyl transpeptidase or transferase (GGT) is primarily associated with microsomal membranes in the biliary epithelium. Production and release of GGT is induced by cholestasis.
Some clinicians consider GGT the test of highest sensitivity in evaluating horses for liver disease. The half-life of GGT is about three days, and it is stable two days in serum at room temperature.
Mild increases may be seen following acute hepatocellular necrosis and may continue to rise for one to two weeks despite improvement in clinical signs. Increases are more persistent in chronic disease, especially with cholestasis. Normal values for GGT in adult horses typically are less than 30 units/L, but may be two to three times greater in healthy donkeys, burros and asses.
The normal liver removes greater than 90 percent of bile acids from the enterohepatic circulation. Thus, the blood concentration of bile acids may be increased with liver disease and quantitation provides an excellent screen of liver function.
The concentration of total serum bile acids is not affected significantly by fasting or eating, so only a single blood sample is needed. Increased serum bile acid concentrations are highly specific for the presence of liver disease (may increase within 24 to 48 hours after the onset of hepatic disease), but are not specific for the type of liver disease.
A value less than 20 micromolel/L appears to be a good predictor in ruling out significant functional liver disease and should be included in the evaluation of horses suspected to have hepatic disease. Bile acid concentrations are highest in biliary obstructive diseases and portosystemic shunts.
Other cytosolic liver enzymes include aspartate aminotransferase (AST), alkaline phosphatase (ALP), lactate dehydrogenase (LD) and alanine aminotransferase (ALT). These enzymes also are found with high activity in other tissues, or are inducible. Thus, increases in these enzymes are not specific for liver disease in horses. Because some of these enzymes are frequently reported in equine biochemical profiles, they may serve as a crude indicator of liver disease; however, the limitations of their usefulness must be recognized.
Other nonspecific tests
Additional nonspecific tests of liver disease in the horse include quantitation of bilirubin, albumin, globulins, ammonia, BUN, coagulation proteins, glucose and esterified triglycerides.
Serum bilirubin concentration is not a sensitive indicator of liver disease in horses, as hemolysis, anorexia and the administration of certain drugs will increase the unconjugated bilirubin concentration.
An increase in the conjugated bilirubin fraction is more reliably indicative of hepatic disease in horses than is the unconjugated bilirubin concentration. When the conjugated bilirubin concentration is greater than 25 percent of the total bilirubin value, hepatocellular disease should be suspected. If the conjugated bilirubin concentration is greater than 30 percent of the total value, cholestasis should be suspected.
In normal horses, the total bilirubin concentration is in the range of 0.2 to 5.0 mg/dL, with conjugated bilirubin in the range of 0 to 0.4 mg/dL. Conjugated bilirubin is water-soluble and detectable in the urine of horses only if blood concentrations become sufficiently increased to surpass the renal threshold (Photo 3); thus when urine tests positive for the presence of bilirubin, cholestatic disease should be suspected.
The half-life of albumin in horses is relatively long (19-20 days), thus a decrease in the albumin concentration is rarely detectable until greater than 80 percent of the liver mass is lost for more than three weeks.
The globulin fraction often is increased in chronic hepatic disease as a result of decreased Kupffer-cell mass and thus wider dissemination of enteric-derived foreign antigens.
Plasma cells respond to the general increased antigen load, resulting in polyclonal gammopathy.
Because the liver primarily is responsible for removing ammonia from circulation and converting it to urea for renal excretion, increases in the blood ammonia concentration or a decrease in the blood urea nitrogen (BUN) concentration (<9 mg/dL) may be indicative of chronic hepatocellular disease. Normal values for ammonia vary among laboratories, but have been reported in the range of 13 to 108 microgram/dL.
Because the liver also is responsible for the synthesis of coagulation factors, evaluation of hemostatic function may be useful, especially if one is considering a liver biopsy.
The vitamin K–dependent factor with the shortest half-life is factor VII. Thus, abnormalities are frequently first observed in the prothrombin time (PT). However, adequate evaluation of hemostatic function necessitates determination of the activated partial thromboplastin time (APTT), the fibrinogen and fibrin degradation products (FDP) concentrations and a platelet count. Typically, a 50 percent to 70 percent decrease in the blood concentration of the coagulation factors is necessary before a change in these clotting time-based assays is detectable.
Changes in the blood-glucose concentration are rarely seen in horses with liver insufficiency. Hyperglycemia may be seen with stress-associated catecholamine and glucocorticoid release. Hypoglycemia (glucose <60 mg/dL) may occur in acute massive hepatic failure, but is more likely in chronic liver disease as anorexia progresses, glycogen stores are depleted and gluconeogenesis and glycolysis are impaired by increased glucagon concentrations.
The concentration of blood triglycerides may become increased during hepatic insufficiency as a result of increased mobilization from adipose tissue to support energy-requiring processes, coupled with decreased clearance by the liver.
The second article on this topic will address differential diagnosis and treatment of liver disease in horses.
Michelle Henry Barton is the Josiah Meigs Distinguished Teaching Professor at the University of Georgia's College of Veterinary Medicine, where she is a large-animal internist in academic practice. She received her DVM from the University of Illinois in 1985, her PhD in physiology at the University of Georgia in 1990 and became an ACVIM diplomate in 1990.