Diagnostic Approach to a Liver Case
BVSc (Hons) MMedVet (Med) PhD Dipl. ECVIM (Internal Medicine)
Bryanston Veterinary Hospital
PO Box 67092, Bryanston, 2021, South Africa
The liver plays a central role in a diverse array of processes including carbohydrate, lipid, and protein metabolism; detoxification of metabolites; storage of vitamins, trace metals, fat and glycogen; fat digestion; and immune regulation. The clinical signs, clinical findings, and clinicopathologic abnormalities associated with hepatic disease reflect deficiencies in these varied functions. The liver, however, has a tremendous reserve capacity to perform these functions and thus the appearance of relatively specific signs of hepatobiliary disease (icterus, hypoglycaemia, bleeding tendencies, hepatic encephalopathy (HE), or ascites), occur only late in disease progression. Early clinical signs of hepatic disease, such as intermittent anorexia, polyuria/polydipsia (PuPd), vomiting, and lethargy, mimic those seen with disease in other organ systems. In most cases, early detection of hepatobiliary disease relies on clinicopathologic evaluation and histopathological examination of hepatic tissue.
Although most early signs of liver disease are non-specific, several clues should raise the awareness as to the presence of hepatobiliary disease. These include strong breed predispositions and the occurrence of several clinical recognizable syndromes that accompany hepatobiliary disease, such as gastrointestinal haemorrhage, HE, icterus, coagulopathies, and ascites.
Several breed predispositions occur in hepatobiliary disease. Congenital portosystemic shunts (PSS) are most frequent in purebred dogs particularly the Irish wolfhound, Australian cattle dog, Maltese, Cairn terrier, miniature Schnauzer, Yorkshire terrier, dachshund, Labrador retriever, and golden retriever. Feline PSS occur most often in mixed breed cats, but Persians and Himalayans are over-represented. Copper-associated hepatopathy occurs in Bedlington terriers, West Highland white terriers, Dalmatians, Skye terriers, and Siamese cats. Several breeds have a predisposition to develop idiopathic chronic inflammatory hepatobiliary disease, including Doberman pinschers, cocker spaniels, standard poodles, and Labrador retrievers. Hepatic amyloidosis occurs in Chinese Shar Pei dogs and in Abyssinian, Oriental, and Siamese cats.
Historical findings that suggest the presence of hepatobiliary disease include recent ingestion of a known hepatotoxic substance or treatment with a potentially hepatotoxic drug. Stunted growth or anaesthesia or drug intolerance (or a combination of these factors) in a young animal, particularly in a predisposed breed, suggests the presence of PSS. A recently stressed obese cat that becomes anorexic is a classic historical example of idiopathic hepatic lipidosis.
Intermittent GI signs such as diarrhoea, anorexia, and vomiting, are common signs of hepatobiliary disease. Chronic hepatic disease predisposes dogs to GI tract haemorrhage so the presence haematemesis, abdominal pain, and melena, should prompt consideration of an underlying hepatopathy.
Hepatic encephalopathy is a neurological condition associated with failure of the liver to detoxify inhibitory neurotoxins generated in the intestinal tract and is most often seen with either congenital or acquired PSS, but may also accompany acute fulminant hepatic failure. The clinical signs are those of diffuse cerebral disease.
Chronic liver disease, particularly in the dog, can be associated with the development of ascites secondary to the presence of portal hypertension or hypoalbuminaemia. Portal hypertension develops either due to capillarisation of the hepatic sinusoidal endothelium from fibrosis, increased hepatic blood flow such as occurs with arterio-venous fistula, or secondary obstruction of the portal vein or its tributaries.
A highly variable, but often predominant, early clinical sign in many dogs and cats with chronic hepatobiliary disease or PSS is the presence of PuPd, caused by one or more of the following: psychogenic polydipsia, alterations in portal vein osmoreceptors, decreased hepatic urea production resulting in disruption of the renal medullary concentration gradient, potassium depletion, stimulation of thirst centres due to HE, and increased endogenous cortisol levels associated with increased adrenal production or decreased hepatic degradation.
Animals with hepatobiliary disease may also have signs referable to ammonium biurate urolithiasis, which form secondary due to chronic hyperammonaemia and decreased hepatic processing of uric acid.
Although icterus should prompt a consideration of hepatobiliary disease, pre- and post-hepatic aetiologies should also be considered.
Ammonia biurate crystaluria can be associated with liver disease especially with PSS. Repeated examination of fresh urine specimens may be necessary to document the presence of these crystals. Uric acid, a by-product of purine nucleotide catabolism, is normally converted to allantoin by hepatic urate oxidase. In hepatic disease, a deficiency of this enzyme may lead to hyperuricaemia. In the presence of concurrent hyperammonaemia, increased concentrations of both ions appear in the urine, resulting in the precipitation of ammonia biurate crystals.
Bilirubinuria is the presence of conjugated bilirubin in the urine. Bilirubinuria in the dog is not abnormal, as they have a low renal threshold for bilirubin and their renal tubular epithelium is capable of bilirubin production. Cats on the other hand, have a high renal threshold for bilirubin and feline kidneys do not make bilirubin, thus feline bilirubinuria is always abnormal and suggests the presence of a hepatobiliary or haemolytic disorder.
Although spontaneous bleeding associated with coagulopathies is rare in animals with hepatic disease, blood loss with a regenerative anaemia may occur after provocative procedures or secondary to GI ulceration. Non-regenerative anaemia, a more common finding in hepatic disease, is usually normocytic and normochromic and often associated with inefficient use of systemic iron stores. Microcytosis occurs in dogs and cats with congenital PSS, in dogs with acquired shunting secondary to cirrhosis, and in some cats with lipidosis. Target cells and poikilocytes may be seen in dogs and cats with hepatic disease. These morphologic changes may be associated with alterations in the erythrocyte plasma membrane lipoprotein content resulting in altered cell deformability.
Evaluation of serum values of hepatobiliary enzymes such as alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and γ-glutamyl transpeptidase (GGT), are commonly used to screen for the presence of hepatobiliary disease, because consistent increases in the serum concentration of these enzyme activities occur after hepatobiliary injury. Although these serum enzymes have a high sensitivity for the detection of hepatobiliary disease, interpretation of abnormalities is hampered by their lack of specificity for hepatobiliary disease. Although the magnitude of serum enzyme elevation is usually proportional to the severity of active hepatobiliary damage, the degree of elevation is not predictive of hepatobiliary functional capacity. Marked increases in serum enzymes may indicate substantial hepatobiliary injury but, due to tremendous hepatic regenerative capacity, are not necessarily indicative of a poor prognosis. Alternatively, in severe end-stage chronic liver disease, serum enzymes may be normal or only mildly increased, because replacement of hepatocytes by fibrosis, prolonged enzyme leakage, or both has resulted in depletion of total liver enzyme content. A single serum enzyme determination should never be used to establish a prognosis. The prognostic value of serum enzyme is improved by following sequential serum enzyme determinations, especially in conjunction with a hepatic function test or hepatic biopsy.
The liver is the exclusive site of albumin synthesis. Because synthesis occurs at 33% of maximum capacity and the serum half life of albumin is approximately 10 days, serum hypoalbuminaemia is most often seen in chronic hepatic disorders. In chronic liver disease hypoalbuminaemia, portal hypertension, and sodium and water retention lead to the development of ascites. In ascitic animals, hypoalbuminaemia may reflect both third-party sequestration of albumin in the abdominal fluid and hepatic synthetic failure. Hypoalbuminaemia is not specific for hepatic disease and may occur with protein-losing enteropathies, protein-losing nephropathies, exudative cutaneous lesions, vasculitis, or acute blood loss. Albumin is also a negative acute phase protein and will consequently be lower in systemic inflammatory diseases.
Immunoglobulins are not synthesized in the liver but may be increased in chronic inflammatory hepatic disease. A polyclonal increase in γ-globulins has been documented in chronic canine hepatic disease and is seen in 50% of feline cases of chronic cholangiohepatitis. Hypergammaglobulinaemia in chronic liver disease may be associated with enhanced systemic immune reactivity due to abnormal Kupffer cell processing of portal antigens or secondary to autoantibody production.
Coagulation abnormalities are quite common in dogs and cats with hepatobiliary disease as the hepatocyte synthesizes all of the coagulation factors, except factor VIII, as well as critical inhibitors of coagulation and fibrinolysis (antithrombin III, antiplasmin) and fibrinolytic proteins (plasminogen). In addition, the liver is responsible for clearance and catabolism of activated coagulation factors, plasminogen activators, and breakdown products of fibrinolysis such as fibrin degradation products (FDP). The liver is also the site of vitamin K-dependent activation of factors II, VII, IX, and X and protein C. In addition both quantitative and qualitative platelet defects can also accompany hepatobiliary disease.
Assessment of coagulation status is important in animals with liver disease because altered haemostasis can contribute to clinical manifestations of hepatic disease and complicate invasive diagnostic procedures. The complexity and overlap of the liver's synthetic and clearance functions, however, makes the interpretation of haemostatic testing difficult. The commonly used tests to assess coagulation include determination of prothrombin time (PT) and partial thromboplastin time (PTT). Abnormalities in these coagulation tests in hepatobiliary disease may be indicative of hepatic synthetic failure, vitamin K deficiency, or the presence of a consumption coagulopathy such as disseminated intravascular coagulation (DIC).
The liver is responsible for the majority of ammonia detoxification. Ammonia is generated primarily in the GI tract by bacterial degradation of amines, amino acids, and purines; by the action of bacterial urease on urea; and by intestinal catabolism of glutamine. Ammonia readily diffuses through the intestinal mucosa and into the portal circulation where it travels to the liver. After uptake by the hepatocyte, ammonia is detoxified either by enzymatic conversion to urea in the mitochondrial urea cycle or by consumption in the synthesis of glutamine. Urea then undergoes renal excretion. Ammonia, which escapes hepatic metabolism, enters the systemic circulation where other tissues, including the kidney, muscle, brain, and intestines, detoxify it by the formation of glutamine.
Failure of the liver to detoxify ammonia or shunting of portal blood away from the liver results in hyperammonaemia. Because the urea cycle operates at only 60% capacity, hepatic failure must be fairly advanced for blood ammonia concentrations to rise. Because intrahepatic or extrahepatic shunting of portal blood directly deposits ammonia in the systemic circulation, blood ammonia concentrations are more sensitive in the detection of hepatobiliary disorders associated with shunting. Hyperammonaemia also occurs in animals with urea cycle enzyme deficiencies and with pathologic conditions that result in decreased availability of urea cycle substrates. Hyperammonaemia has been described in dogs with a deficiency of the urea cycle enzyme, argininosuccinate synthetase, and in cats fed a diet deficient in arginine, an essential substrate in the urea cycle.
The biggest obstacle in the clinical use of blood ammonia determinations is the difficulty in sample handling. Samples must be drawn directly into cold heparinized tubes, immediately transferred to the laboratory on ice for refrigerated centrifugation, and preferably assayed within 1 hour. Feline, but not canine, plasma samples may be frozen at -20°C for 48 hours without sacrificing accuracy. Because erythrocytes contain two to three times the amount of ammonia as plasma, if they are not separated from the plasma immediately or if haemolysis is found in the sample, spurious increases in blood ammonia will result.
Approximately 65% of dogs and cats with PSS have decreased serum urea. Decreases in serum urea are thought to arise secondary to decreased urea production in the atrophied liver. Serum urea is not specific for liver disease as it can be influenced by hydration status, dietary protein content, GI haemorrhage, glomerular filtration rate, and fluid or solute diuresis.
Bilirubin is a yellow pigment formed in the reticulo-endothelial cell system (RES) by the enzymatic processing of haeme. Bilirubin released from RES cells is not water soluble and is transported in plasma reversibly bound to albumin. This unconjugated bilirubin is extracted by the liver and conjugated by esterification with glucuronic acid. Water-soluble conjugated bilirubin is actively secreted against a concentration gradient into the bile. The bilirubin in bile enters the intestinal tract via the bile duct, where it is either excreted unchanged in the faeces or is converted to urobilinogen by the action of enteric bacteria. Most of the urobilinogen is further degraded into the brown-pigmented stercobilins. The absence of stercobilins in the faeces results in pale-coloured acholic stools. Small amounts of urobilinogen enter the portal venous system and undergo enterohepatic circulation. The majority is re-excreted into the bile, but a small amount undergoes urinary excretion.
Icterus is the clinical manifestation of bilirubin retention within tissues. Although less sensitive than serum liver enzymes for the detection of hepatobiliary disease, hyperbilirubinaemia is more specific. During prolonged cholestasis, excess conjugated bilirubin may become irreversibly bound to albumin. These so-called biliproteins are clinically significant in that they are measured as direct reacting bilirubin, but the half life approximates that of albumin. Their presence can result in persistently elevated serum bilirubin weeks after the resolution of the underlying cholestatic liver disorder.
Bile acids are synthesized exclusively in the liver from cholesterol. After conjugation to either taurine or glycine, bile acids are excreted into bile and collected, stored, and concentrated in the gallbladder. After ingestion of a meal, cholecystokinin release stimulates gallbladder contraction and the transport of bile acids into the intestine. In the intestinal lumen, bile acids aid in the solubilization and absorption of fats. When the bile acids reach the ileum, they are efficiently transported back into the portal circulation from which they are re-extracted by the hepatocyte sinusoidal bile acid transporter. This enterohepatic circulation of bile acids operates at 98% efficacy.
Disruption of the enterohepatic circulation of bile acids results in increases in the concentration of serum bile acids (SBA). In normal animals, SBA concentrations are determined by the spill over of bile acids that escape from the enterohepatic circulation. During fasting, when the enterohepatic circulation of bile acids is low, SBA are low. After a meal, bile acids are released into the intestines and subsequently absorbed into the portal circulation. Increased portal vein bile acid concentrations are reflected in a transient elevation in SBA. This endogenous challenge to the enterohepatic circulation of bile acids is used clinically. In a typical bile acid test, SBA are determined after a 12-hour fast, a test meal is fed and then postprandial SBA determined 2 hours later.
Maltese dogs may have increased postprandial SBA in the absence of hepatobiliary disease; however, the aetiology remains undetermined.
A number of factors that influence the enterohepatic circulation of bile acids in normal animals can affect SBA values. These include the completeness of gallbladder emptying, the rate of gastric emptying, intestinal transit rate, the efficiency of ileal bile acid reabsorption, and the frequency of enterohepatic cycling. Inadequate fat or amino acid content in the test meal or consumption of an insufficient amount of food can result in failure of cholecystokinin release and gallbladder contraction. The presence of concurrent disease that delays gastric emptying may result in failure to stimulate gallbladder contraction. Alterations in intestinal transit time so that the movement of conjugated bile acids to the ileum is delayed can result in less than optimum timing for determination of the postprandial values. Severe ileal disease can result in decreased bile acid reabsorption and inadequate challenge to the enterohepatic circulation. The presence of small bowel overgrowth leads to bacterial deconjugation of bile acids and decreased ileal absorption of bile acids.
Occasionally, fasting SBA concentrations are higher than postprandial values. This happens when interdigestive gallbladder contraction occurs during the course of the fast preceding the test. It may also be associated with individual variations in gastric emptying, response to cholecystokinin release, and intestinal transit time.
The interpretation of abnormal SBA is subject to a number of limitations: SBA are not capable of discriminating one hepatobiliary disease from another; and there is little correlation between the severity of histological changes or the degree of portosystemic shunting and the extent of SBA elevation. A SBA value is either normal or abnormal. When evaluating serial determinations of SBA to monitor disease progression or response to therapy, only a return to normal can be used as a reliable indicator of clinical remission.
The major function of the liver in carbohydrate metabolism is to maintain normoglycaemia during the fasting state. The liver has a large reserve for maintaining glucose homeostasis so that greater than 70% of hepatic function must be lost before hypoglycaemia occurs. Hypoglycaemia occurs most often in acute fulminant hepatic failure and in small breed dogs with PSS. In acute hepatocellular injury, hypoglycaemia may be a relatively early indicator of severe hepatic failure. Hypoglycaemia is a rare complication of end-stage chronic inflammatory liver disease and is a negative predictor of survival.
Plasma fatty acids released from adipose tissue are extracted by the hepatocytes, after which they are either converted to triglycerides or undergo mitochondrial β-oxidation. Triglycerides are either stored or packaged as very low-density lipoproteins and released into the vasculature. The liver also extracts chylomicron remnants and low-density lipoproteins from plasma. This is the major route by which cholesterol enters into the liver, although the liver is also capable of cholesterol synthesis. Hepatic cholesterol can be esterified and then packaged and secreted in lipoproteins or stored in the liver. The majority of unesterified (free) cholesterol in the liver undergoes biliary excretion.
Hypercholesterolaemia is associated with decreased hepatic synthesis, decreased biliary excretion of cholesterol, and PSS. Hypocholesterolaemia can occur with late-stage chronic liver disease in dogs.
Radiography can be used to assess changes in the size and opacity of the liver. Diffuse hepatomegaly is indicated by rounding of the liver edges, extension of the hepatic shadow beyond the costal arch, and caudal displacement of the stomach axis. Hepatomegaly may occur due to congestion, infiltrative disease (neoplasia, lipidosis, glycogen accumulation, amyloidosis), inflammatory disease, RES cell hyperplasia, or extramedullary haematopoiesis. Focal hepatomegaly can be discerned by observing displacement of the structures bordering the liver, and it may occur with cysts, granulomas, neoplasia, regenerative nodules, haematomas, or abscesses. Micro-hepatica is visualized radiographically by a decreased size of the hepatic shadow and a shift of the gastric axis to a more upright orientation with cranial displacement. Micro-hepatica is observed with hepatic atrophy and fibrosis. Cholelithiasis or choledocholithiasis may be visualized as mineralization in the liver. The former is recognized as a discrete round opacity in the cranial right ventral liver shadow, and the latter is seen as diffuse mineralization. Focal mineralization may also be seen with chronic gallbladder infection or neoplasia, granulomatous lesions, abscesses, resolving haematomas, and within regenerative nodules.
Gas opacities in the liver may be associated with hepatic abscesses, emphysematous cholecystitis, or after long-term bile duct obstruction. The finding of gas within the portal vessel indicates the entry of GI gas or infection with gas-producing organisms and is a grave sign seen with gastric torsions and severe necrotizing gastroenteritis.
Radiographic contrast imaging of the portal venous system is used to localize PSS and can be accomplished by cranial mesenteric arteriography, splenoportography, or mesenteric portography. Colorectal scintigraphy can be used for the detection of PSS.
Ultrasonography enables differentiation between focal and diffuse disease, evaluation of alterations in hepatic parenchyma in diffuse disease, evaluation of the biliary system and portal vasculature, and procurement of tissue for hepatic histopathology.
Because patterns of clinicopathologic abnormalities do not reliably differentiate one form of hepatic disease from another, it is often necessary to obtain hepatic tissue for histopathological or cytological evaluation. The indications for examination of hepatic tissue include persistent increases in serum liver enzymes, increased serum bile acids or bilirubin, unexplained ascites, unexplained hepatomegaly, to assess response to therapy or disease progression, and to evaluate for breed-specific hepatopathies. Because the most common complication encountered in the procurement of hepatic tissue is haemorrhage, before undergoing hepatic tissue acquisition, all dogs and cats should have a PT, PTT, and platelet count assessed. Acquisition of hepatic tissue is contraindicated in the presence of a severe coagulopathy, although some animals may be biopsied safely after the administration of fresh frozen plasma to supply deficient coagulation factors.
In general, hepatic biopsy is preferred over fine needle aspirate (FNA) to characterize liver disorders. However, in animals with bleeding disorders or those with large cavitary lesions or suspected abscesses, ultrasound guided fine needle aspiration (22 gauge needle) can safely be performed with minimal danger of haemorrhage or spread of infection. In studies done comparing FNA to hepatic biopsy, complete agreement is typically seen in around 30% of cases. While FNA cytology may be helpful in identifying vacuolar or neoplastic disease such as lymphoma, they frequently miss the presence of inflammatory or vascular disease.
Hepatic biopsy can be obtained by ultrasound-guided needle biopsy, laparoscopic biopsy, or wedge biopsy by laparotomy. The advantages of the latter two methods are the ability to grossly evaluate the liver at the time of biopsy, the acquisition of relatively large tissue samples, and the identification and control of post-biopsy haemorrhage. Because the tissue obtained by wedge or needle biopsy represents only a tiny fraction (maybe 1/10,000) of the entire liver, one of the major limitations of hepatic biopsy is sampling error. Sampling error arises due to uneven distribution of lesions in what appears by diagnostic imaging to be diffuse disease.
To fully appreciate the value of histopathological evaluation of the liver, one must understand what information to reasonably expect from tissue samples. One should expect the sample to determine the category of disease (inflammatory, neoplastic, vacuolar, or vascular), to define the extent of disease (mild, moderate, or severe) and to access the chronicity of the lesion (presence of fibrosis, type of cellular infiltrate). One should not expect every hepatic biopsy to provide an aetiological diagnosis. The biopsy, however, may provide clues as to cause. Special stains can help to identify infectious agents or determine if copper or iron overload exists. Samples for anaerobic and aerobic culture may confirm suspicion of primary or secondary bacterial infection.