Abstract

Clinical hepatobiliary disease investigation in reptiles appears deficient given the lack of publications reporting antemortem diagnoses. However, the clinical approach is not dissimilar to that recommended in domesticated species. Detailed history, physical examination findings, clinicopathology, diagnostic imaging, endoscopy, biopsy histopathology, and microbiology are typically able to determine a definitive diagnosis. Definitive diagnosis provides an accurate prognosis and facilitates more appropriate, specific therapy.

Introduction

The reptilian liver is similar in structure and function to that of other vertebrates. It is the largest single visceral organ and may be elongated in snakes and some lizards, or transverse in testudines and other lizards. Liver disease is not uncommon in reptiles, and yet a review of published hepatopathies indicates that the vast majority of cases were diagnosed postmortem. This implies deficiency in the current clinical approach to hepatobiliary disease diagnosis. Therefore, the aim of this presentation is to provide clear guidance on antemortem diagnosis of hepatobiliary diseases in reptiles.¹

History and Physical Examination

The approach to hepatobiliary disease investigation in reptiles is similar to that in mammals. Recently, the World Small Animal Veterinary Association (WSAVA) published guidelines for the diagnosis of known hepatobiliary diseases in dogs and cats. Although these guidelines relate specifically to dogs and cats, they serve as a useful guide for the appropriate investigation of hepatobiliary disease in reptiles. They also serve as a useful reminder that just because a specific disease process has not been identified in reptiles, it does not mean that it cannot occur. As clinicians involved in the evolution of herpetological medicine, we must continually strive to watch for the first case of a portal systemic shunt or hepatocutaneous syndrome.

Detailed information on performing a physical examination is covered elsewhere. Clinical signs, physical examination findings, and clinicopathologic abnormalities associated with hepatobiliary disease often reflect disruptions in normal structure and function. The liver possesses a large functional reserve, and therefore the appearance of potentially hepatic-associated signs such as icterus and ascites typically represent exhaustion of these reserves that occur late in the course of disease. Therefore, any reptile exhibiting such signs should be investigated as a matter of urgency. Nonspecific signs including intermittent anorexia, vomiting, and lethargy can occur with diseases associated with the liver or a variety of other organ systems. Therefore, diagnostic investigation is also required to differentiate from hepatic and non-hepatic diseases.

Historical events such as the ingestion of known hepatotoxic substances (e.g., aflatoxins, pyrrolizidine alkaloids) or treatment with potentially hepatotoxic drugs (e.g., NSAIDs, glucocorticoids) may suggest the presence of hepatobiliary disease. An obese reptile that becomes anorectic may be predisposed to hepatic lipidosis, while chronic hepatic disease often predisposes to gastrointestinal ulceration, hematemesis, coelomic pain, and melena.

The presence of icterus is the most significant and specific sign associated with hepatobiliary disease. In those squamates that produce bilirubin, a yellow discoloration to the mucous membranes may be obvious, while in other species, a greener pigment (especially in urine/urates) is often appreciated. However, in many cases, mucous membrane pigmentation often complicates evaluation. Commonly reported in mammals, polyuria/polydipsia is seldom appreciated in reptiles; however, coelomic enlargement due to ascites or hepatomegaly, anorexia, poor body condition, neurologic depression, and lethargy might also be indicators of an underlying hepatobiliary disorder.

Clinical Enzymology

Although evaluations of hepatobiliary enzymes (e.g., ALT, AST, ALP, GGT, and SDH) are routine in human and domesticated animal medicine, it would be wrong to assume that they are as reliable in reptiles. All of the commercial enzyme measurements are designed for mammals and are run at 37°C. Very few studies have attempted to objectively measure hepatocellular enzyme activities in healthy reptiles, and even fewer have attempted to demonstrate changes following a hepatic insult. Worse still, very few (if any) of the published studies that this author is aware of have managed to satisfy the requirements of the American Society for Veterinary Clinical Pathologists (ASVCP) for reference range determination. Therefore, clinicians should be cautious of including or excluding hepatobiliary disease on the basis of plasma biochemistry results. Taking the green iguana (Iguana iguana) as an example, moderate activities of LDH and AST were found in many tissues, making differentiation between hepatic and non-hepatic (especially muscle) sources impossible, whereas ALT and ALP activities were also generally low but again found in a variety of tissues. Recent research at the University of Georgia demonstrated significant elevations in a number of parameters, compared to controls, following an acute hepatotoxic insult. AST, SDH and LDH levels increased more than 15-fold, and ALT increased fourfold, whereas ALP and GGT did not change significantly.

Non-hepatic tissue trauma (especially involving muscle) can also cause increases in AST and LDH, and these parameters are considered less specific. Enzyme elevations remained high for at least 48 hours after insult but decreased by day 6 in most cases. Therefore, elevation of AST, ALT, LDH and SDH are consistent with the initial phase of acute liver injury, but there is little information on the plasma half-lives of reptile hepatic enzymes. Furthermore, although the degree of enzyme elevations is considered proportional to the degree of hepatobiliary damage, it is not predictive of hepatic function and therefore does not indicate prognosis.

Plasma Proteins

Albumin is exclusively produced by the liver, but because of functional capacity and protein half-life, hypoalbuminemia is typically only seen during chronic hepatic disease. Liver cirrhosis may result in hypoalbuminemia, sodium and water retention, and lead to ascites. Hypoalbuminemia is not specific and can occur secondary to protein-losing nephropathies/enteropathies, exudative cutaneous lesions (e.g., severe burns), vasculitis, acute blood loss, or excessive blood collection. Inadequate nutrition and systemic inflammatory conditions can curtail albumin synthesis. Unfortunately, the measurement of albumin by the bromocresol green dye-binding method has been shown to be inaccurate in turtles when compared to the gold standard of protein electrophoresis. Recent work at the University of Georgia has also demonstrated inaccuracies between albumin measurements by bromocresol green when compared to electrophoresis in the bearded dragon (Pogona vitticeps). Based upon current knowledge, the measurement of albumin in chelonians and bearded dragons by bromocresol green is inaccurate. Indeed, until proven otherwise, albumin measurements in all reptiles by anything other than protein electrophoresis should be considered inaccurate and avoided.

The plasma globulin fraction is composed of immunoglobulins and non-immunoglobulins, and hepatic impairment can result in decreases in α- and β-globulins. However, as many immunoglobulins are acute-phase proteins and hepatic production is increased during systemic response to inflammatory disease, hepatitis may actually lead to hypergammaglobulinemia. Some commercial laboratories are able to offer acute-phase protein assays, including C-reactive protein, haptoglobin, and amyloid A, although their accuracy and relevance to reptile diseases have not been proven.

Coagulation abnormalities are common in mammals with severe hepatopathy, but they appear to be rare in reptiles. In reptiles, the extrinsic and common pathways are thought to play major roles in coagulation. The extrinsic pathway can be assessed using prothrombin times (PT); however, the use of commercially available tests using mammalian thromboplastin results in prolonged coagulation times or lack of coagulation. Recent investigations into PT in green iguanas (Iguana iguana) and red-eared sliders (Trachemys scripta elegans) using reptile-derived thromboplastin have demonstrated clotting times in the range of 22–37 seconds, which are not dissimilar to those reported in endotherms. The value of activated partial thromboplastin time (APTT) in reptiles is likely negligible due to the relative lack of intrinsic pathway clotting factors. Although not completely understood, the impacts of temperature and calcium need to be considered when assessing coagulation in reptiles.

Ammonia

The liver is responsible for detoxifying ammonia of intestinal bacterial origin through the urea cycle. Ammonia is an important, but not exclusive cause of hepatic encephalopathy; however, it is currently the only measurable toxin. Hepatic failure or shunting of portal blood away from the liver can result in hyperammonemia, although in mammals, test sensitivity is greater for shunting disorders. The greatest issue facing testing is sample handling, as blood must be collected into cold heparinized tubes and transported on ice for immediate laboratory evaluation. Currently, there is very little information on the use of ammonia in reptiles.

Uric Acid and Blood Urea Nitrogen

Decreased urea levels have been documented in mammals with severe hepatic fibrosis due to decreased hepatic production. It would also seem likely, therefore, that urea and uric acid levels in ureotelic and uricotelic reptiles would also be decreased due to decreased hepatic production. A recent study investigating hepatotoxin-induced hepatopathy in iguanas reported resting levels of urea and uric acid to be 3.5 mg/dL and 4.1 mg/dL, respectively. Peak levels of 11.5 mg/dL and 4.4 mg/dL were recorded at 96 hours post-insult. These elevations (rather than decreases) emphasize the many confounding variables that can influence urea and uric acid levels, which include hydration status, dietary protein content, recent feeding, gastrointestinal hemorrhage, glomerular filtration rate, drug therapy (e.g., allopurinol), and diuresis.

Bilirubin and Biliverdin

In those snakes that can reduce biliverdin to bilirubin, hepatobiliary disease would be expected to cause an elevation in unconjugated bilirubin. Although elevated bile pigments are more specific for hepatobiliary disease, they are generally less sensitive than hepatocellular enzymes. Hyperbilirubinemia and hyperbiliverdinemia can be prehepatic, hepatic, or posthepatic. Prehepatic causes include severe hemolysis and can be confirmed by demonstrating a decrease in packed cell volume. Hepatic causes can be due to impaired uptake, conjugation, or excretion that cause severe intrahepatic cholestasis. Differentiation between hepatic and posthepatic causes is difficult but important, as posthepatic conditions typically require surgical decompression of the gallbladder, and intrahepatic cases are treated medically. A recent experimental study in iguanas documented resting bilirubin levels of 0.2±0.1 mg/dL, rising to 0.6±0.1 mg/dL 96 hours following the administration of hepatotoxin. However, for most non-serpentine reptiles, measurement of biliverdin is required and, despite the existence of validated assays, few are commercially offered.

Bile Acids

In normal animals during fasting, when the enterohepatic recirculation of bile acids is low, total serum or plasma bile acids are also low. In dogs and cats, bile acids are highly specific for hepatobiliary disease and are an ideal hepatic function test. However, with the exception of portosystemic shunts, the sensitivity of bile acids is insufficient to warrant use as a screening test, and the same may be true for reptiles. In clinically healthy green iguanas and bearded dragons, mean fasting levels of 7.5 (range 2.6–30.3) µmol/L and 4.6 (range 0.8–33.7) µmol/L, respectively, have been documented. Significant postprandial effects have also been demonstrated, and in iguanas, mean elevations to 33.3±22.0 µmol/L have been recorded 3 hours post-feeding. In bearded dragons, mean elevations to 6.9 (range 1.4–37.6) µmol/L and 11.9 (4.3–31.0) µmol/L have been documented at 4 and 24 hours post-feeding. The duration of this postprandial effect has not been determined, but it suggests that reptiles should be fasted for several days prior to sampling, probably longer in large carnivorous reptiles. A number of factors can affect total plasma bile acids, including degree of gallbladder emptying, rate of gastric emptying, intestinal transit rate, efficiency of ileal bile acid reabsorption, frequency of enterohepatic cycling, fat and amino acid content of the test meal, and amount of test meal consumed. Therefore, a consistent approach to performing a bile acid stimulation test is important.

Glucose

Hypoglycemia occurs in mammals with hepatic failure but is rare in cases of chronic disease. Given the already low resting glucose levels of most reptiles and the effects of nutrition and handling stress, the detection of hypoglycemia may be difficult unless accurate baseline values for a specific patient have been repeatedly documented over time. Following hepatotoxin insult to iguanid livers, no significant changes in glucose levels were appreciated over a 6-day period.

Cholesterol, Triglycerides, and Lipids

The liver converts fatty acids into triglycerides, which are either stored or released from the liver as very low-density lipoproteins (VLDL). Cholesterol may be synthesized within the liver or acquired through chylomicron remnants and low-density lipoproteins (LDL). Hepatic cholesterol can be esterified and secreted in lipoproteins or stored in the liver. Hypercholesterolemia can be associated with excessive production or decreased excretion (e.g., extrahepatic biliary obstruction). Normal female reptiles undergoing vitellogenesis often exhibit increases in cholesterol and triglycerides; however, sustained elevations or elevations in males are often seen in cases of hepatic lipidosis. There is a variable postprandial effect on triglyceride levels (less so for cholesterol) in bearded dragons, and fasting prior to sampling is recommended. In addition to cholesterol and triglycerides, some commercial laboratories offer lipoprotein profiles.

Urinalysis

Urinalysis in mammals is useful for the demonstration of ammonia biurate crystalluria and uric acid; however, given the variable expectation for such nitrogenous products in many reptiles, their diagnostic usefulness is probably limited as an adjunct to hepatobiliary disease diagnosis. Bilirubinuria is considered abnormal in cats but not in dogs. Only snakes appear capable of producing bilirubin in significant quantities, but the renal threshold for urinary bilirubin is unknown.

Hematology

Coagulopathies are rare, but bleeding could result in regenerative anemia. Nonregenerative anemia may be present in cases of chronic hepatobiliary disease, and normocytic, normochromic anemia would be expected due to inefficient utilization of systemic iron stores (anemia of chronic disease).

Diagnostic Imaging

Radiography

The size, shape, and position of the liver can be difficult to determine from plain radiographs of the cranial coelom. Unfortunately, due to lack of diffuse fat tissue, serosal detail is typically poor in reptiles, and evaluation of liver margins is often not possible. In addition, although hepatomegaly can cause displacement of the stomach, there is little appreciation of changes in gastric axis in reptiles. Hepatomegaly commonly occurs due to infiltrative disease (e.g., neoplasia, lipidosis, amyloidosis) and inflammatory disease (e.g., bacterial hepatitis). Focal enlargement due to a neoplasm, cyst, granuloma, or abscess may be appreciated by displacement of adjacent tissues and, in squamates, compression on the lungs. Microhepatica, often associated with fibrosis and atrophy (including chronic cachexia), is seen as a decrease in the size of the liver silhouette. Normally the liver is homogeneous in appearance. Mineralization can be diffuse (choledocholithiasis) or as a focal, discrete opacity associated with the right liver lobe in lizards and chelonians, or just caudal to the liver in snakes. Gas opacities may be associated with abscesses or as a grave sign associated with severe necrotizing gastroenteritis.

Computed Tomography (CT)

The size and structure of the liver can be assessed using CT, in much the same way as in mammals. Pre- and post-contrast images in both soft tissue and bone algorithms should be collected. Computed tomography can be used to detect alterations in the x-ray attenuation of hepatic parenchyma. In healthy juvenile sea turtles (Chelonia mydas), a mean attenuation value of 60.09±5.3 Hounsfield units (HU) has been reported; in general, values between 50–70 HU are considered normal for chelonians. However, in male captive red-footed tortoises (Chelonoidis carbonaria), hepatic values were much lower at 11.2±3.0 HU and were attributed to multiple cases of hepatic lipidosis in the group, although this was not confirmed histologically.

Ultrasonography

Ultrasonography is useful for the evaluation of coelomic viscera, including the liver. However, ultrasonographic anatomy has only been described for a few species, but is generally considered useful for the differentiation between focal and diffuse disease, as well as assessment of displacement/compression, parenchymal changes, gallbladder disease, and vascular abnormalities.

In the green iguana, hepatic parenchyma is of uniform echotexture with medium echogenicity, similar to that of the spleen, testes, and fat bodies. Hepatic lobation is generally indistinct. The gallbladder is easily identified adjacent to and partially embedded in the caudoventral border of the liver, right of midline. Gallbladder contents are normally anechoic, but small amounts of dependent material of mineral echogenicity can sometimes be identified. The subadult iguanid gallbladder is an elongated oval in the sagittal plane, with mean±SD long-axis dimensions of 1.67±0.40 cm (1.00–2.72 cm) and mean short-axis dimensions in the sagittal plane of 0.67±0.16 cm (range, 0.34–1.00 cm). However, significant linear correlations between body weight and gallbladder dimensions appear to be lacking. Preliminary investigations into contrast-enhanced hepatic ultrasonography have indicated that peak enhancement of 19.9±7.5% (11.7–34.6%) is achieved after 134.0±125.1 seconds (59.6–364.5). However, as the distribution of contrast medium in iguanas differs from that of mammals, specific reference ranges of hepatic perfusion for diagnostic evaluation of the reptilian liver are still outstanding.

Ultrasonographic liver anatomy has also been briefly described for the desert tortoise (Gopherus agassizii) via left and right mediastinal or cervicobrachial acoustic windows. No descriptions of hepatic parenchymal structure or measurements were given, although visualization of the gallbladder was limited by the amount of gastrointestinal gas present. The liver of the Boa constrictor (Boa constrictor) has been described as uniformly heteroechoic, fusiform, and primarily located on the right side of the coelom, with the vena cava ventral and hepatic vein dorsal. The gallbladder, anechoic with hyperechoic wall, is located just caudal to the caudal margin of the liver, in close association with the pancreas, spleen, and fat body.

Magnetic Resonance Imaging (MRI)

The hepatobiliary system can be evaluated using MRI, with most current descriptions based upon chelonians. Image quality is largely dependent upon the size of the animal, strength of the magnet, and scanner software. In general, units capable of generating magnetic fields of 3+ tesla provide better-quality images for most reptiles. The liver is more signal intensive in T1-weighted images compared to those that are T2 weighted. In addition to hepatic size, parenchymal homogeneity can be assessed, and its fat content compared to that of fat bodies. The internal parenchymal structure is interrupted by the vessels, which (along with the biliary system) are better evaluated using T2-weighted images.

Endoscopy

Endoscopic techniques for the evaluation of the chelonian and saurian coelom have been well described. Endoscopic evaluation of the liver provides magnified, color visualization of the hepatic surface and gallbladder, although deeper lesions can often be appreciated, especially if large. In addition, significant lesions can often be identified endoscopically, despite the lack of obvious clinicopathologic, radiographic, ultrasonographic, or cross-sectional imaging abnormalities. In laterally recumbent lizards and chelonians, a lateral paralumbar approach provides visualization of the ventral and dorsal surfaces of the ipsilateral liver lobe. In dorsal recumbency, a paramedian approach provides access to the ventral surface of the entire liver. In green iguanas and red-eared sliders, entry of a 2.7-mm telescope within a 4.8-mm operating sheath and visualization of the liver from both left and right paralumbar approaches were easily performed without complications; however, the gallbladder could only be reliably seen from a right approach. In snakes, a targeted coelioscopic approach is often required, as, in most cases, no single entry can permit examination along the length of the entire liver. However, examination from within the air sac does permit evaluation over a much greater hepatic length.

Fine-Needle Aspirate Cytology

Fine-needle aspiration from the liver is possible under ultrasound or CT guidance. However, due to the loss of tissue architecture, ultrasound-guided cytological aspirates are generally considered inferior to endoscopic or surgical biopsy. A recent study investigated hepatotoxin-induced disease in green iguanas, and invariably ultrasound-guided fine-needle aspirates were of poor diagnostic value. Fine-needle aspirates may be more useful in disease processes that are highly cellular (e.g., neoplasia), for microbiologic evaluation, and where longer periods of general anesthesia carry unacceptable risks.

Biopsy

Clinicopathologic changes may be attributable to hepatobiliary disease; however, they do not reliably differentiate between different disease processes and do not provide an accurate prognosis. To reach a definitive diagnosis, it is essential to demonstrate both (i) an accurate host pathological response and (ii) the etiologic or causative agent. While paired rising titers are definitive, there are few commercial assays available for reptiles, and those that are available typically require samples 6–9 weeks apart. Given the poorer diagnostic value of cytology compared to histopathology, tissue biopsy is generally preferred. Tissue biopsies can be used for histopathologic, microbiologic, parasitologic, and toxicologic evaluations and can typically provide a definitive diagnosis within days of collection. Indications for hepatic biopsy include elevations of hepatic enzymes (especially AST and SDH, bile acids, biliverdin/bilirubin, unexplained coelomic effusion/ascites, and hepatomegaly). The most common complication concerning hepatic biopsy (especially following needle biopsy) is hemorrhage, and therefore consideration should be given to platelet count, PT, PTT, and buccal mucosal bleeding time. Biopsy samples may be collected using cutting biopsy needles (e.g., Tru-Cut), endoscopically, or by standard wedge or excisional surgical techniques.

Handling of biopsies should, whenever possible, be undertaken with atraumatic instruments to minimize damage. Small endoscopic and needle biopsies should be gently shaken free into a sterile saline and not removed using cotton-tipped applicators or, worse still, a hypodermic needle. Small samples should be placed into labeled histology filters before being submitted in 10% neutral buffered formalin for histopathology. For transmission electron microscopy (TEM), glutaraldehyde is often used, while ethanol or methanol may be preferred for urate tophi or parasite evaluations. Tissue samples for microbiology should be sent in appropriate transport media depending upon whether fungal, bacterial, or viral investigations are most important. Biopsies can be frozen at -20°C while histopathology results are pending. Then, if histopathologically indicated, such frozen tissue can be submitted for specific microbiologic (often virologic) investigations.

Ultrasound-guided, percutaneous hepatic biopsies have been reported from 1–25-kg boid snakes using an 18-gauge x 20-cm biopsy needle. All collected biopsies were considered diagnostic, and all snakes survived; however, multiple biopsy attempts were required for some snakes, and iatrogenic gastric perforation was documented in at least one of 15 animals (7%). Ultrasound-guided liver biopsy in green iguanas has also been described, but one of eight animals (12.5%) suffered from fatal hemorrhage; the technique is probably unsafe, especially for smaller lizards.

Although general anesthesia is seldom employed for such biopsies in humans and dogs, it is still recommended for reptiles to ensure adequate restraint. Automated needles are preferred and operated by the ultrasonographer to ensure that vessels and other organs are not in the path of the needle. Holding an anesthetized reptile at maximum inspiration tends to improve localization and visualization of the liver. The biopsy site should be evaluated for post-sample hemorrhage. Comparison between needle biopsies and wedge biopsy (gold standard) in dogs and cats has indicated an overall agreement on only 48%, and similar results would likely be expected for reptiles.

Endoscopic liver biopsies have been experimentally evaluated in green iguanas and freshwater chelonians. Biopsies were collected using 1.7-mm biopsy forceps, and each contained a mean of 1.9 and 3.1 portal tracts in iguanas and sliders, respectively. Although peripheral crush artifact was common, it was minimal, and all samples were considered diagnostic. No adverse effects of endoscopic hepatic biopsy were reported. Multiple samples can be readily collected and should be routinely submitted for histopathology, bacterial (aerobic/anerobic) and fungal cultures. Additional parasitic or toxicologic analysis can also be considered on an individual case basis. This technique enables gross evaluation of most (if not all) of the liver, extrahepatic biliary system, and surrounding structures. The ability to collect multiple samples from various hepatic locations decreases the risks of sampling artifact in cases of regional diversity within the liver. The standard oval/round cup biopsy forceps also produces less hemorrhage than biopsy needles; but if hemorrhage is encountered, direct pressure can be applied by the closed biopsy forceps or a blunt probe, or radiosurgical coagulation can be used. Direct visualization of the hepatic parenchyma enables the clinician to correlate clinical data with liver appearance and histopathology to adjudicate the most accurate diagnosis.

Coeliotomy provides access to the liver for harvesting surgical biopsies. The advantages of open surgical access are the ability to collect larger liver samples, as well as an improved opportunity to more completely evaluate and sample other visceral structures (e.g., intestines). Nevertheless, increased invasion and duration of anesthesia must be considered. Surgical biopsies can be collected using a standard wedge technique. Hemorrhage can be controlled using hemostatic material (e.g., Gelfoam) and digital pressure. Advantages and disadvantages are similar to those of endoscopy.

References

1.  Divers SJ. Hepatology. In: Divers SJ, Stahl SJ, eds. Mader’s Reptile and Amphibian Medicine and Surgery. 3rd ed. St. Louis, MO: Elsevier; 2019:649–668.

An extensive list of references can be found in the above review chapter.