Idiopathic Hepatitis

a nd Cirrhosis in Do gs
Robert P. Favier, DVM

KEYWORDS
 Hepatitis  Cirrhosis  Dog  Treatment  Prognosis

INCIDENCE AND PATHOGENESIS

Primary hepatitis (PH) is the most frequently occurring liver disease in dogs and should
be distinguished from nonspecific reactive hepatitis (NSRH). A previous study found
that 1% of all referred patients that presented to the author’s university clinic had
a form of canine PH. In contrast to human hepatology, the diagnosis of canine hepa-
titis is based mainly on histologic morphology, and the term, hepatitis, often is used
regardless of cause. Regularly encountered forms of PH in dogs include acute hepa-
titis (AH) and chronic hepatitis (CH) (with or without cirrhosis); less frequently encoun-
tered forms are lobular dissecting hepatitis (LDH), granulomatous hepatitis (GH), and
eosinophilic hepatitis (EH). For each of these forms, the World Small Animal Veterinary
Association (WSAVA) Liver Standardization Group has published standards for diag-
nosis.1 Of 101 cases of PH referred to the author’s clinic between 2002 and 2006
(Department of Clinical Sciences of Companion Animals, Utrecht University),
21 (21%) had AH (of which CH developed in at least five at a later date), 67 (66%)
CH, seven (7%) LDH, one (1%) GH, and one (1%) EH. Of the CH cases, after re-eval-
uation of the biopsies with copper staining, 36% seemed copper associated (CH[ca]),
higher than expected, and 64% of the CH cases had an unknown cause and were
considered idiopathic (CH[i]). Among dogs with CH(i) and CH(ca), approximately
50% had cirrhosis at initial diagnosis, and both groups contained several female
Labrador retriever dogs (seven and five, respectively).
In different publications and case reports, a wide variety of causes for hepatopathy
in general has been documented, including microorganisms, toxins and drugs,
immune-mediated reactions, and breed-associated metabolic errors.2 The inherited
disorders of copper metabolism, in particular, have received much attention in the
last few decades.3–9 In spite of a large effort, however, the majority of PH cases remain
of idiopathic origin. Although hepatitis in dogs has been characterized extensively,
there are no data published on the occurrence of the various WSAVA-classified forms
of hepatitis in a clinical population, progression between those forms, or occurrence of
idiopathic and copper-associated forms of hepatitis.10–12

Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, P.O.

Box 80154, 3508 TD Utrecht, The Netherlands
E-mail address: [email protected]

Vet Clin Small Anim 39 (2009) 481–488

doi:10.1016/j.cvsm.2009.01.002 vetsmall.theclinics.com
0195-5616/09/$ – see front matter ª 2009 Elsevier Inc. All rights reserved.
482 Favier

ACUTE HEPATITIS

In the field, most cases of AH probably are missed. Dogs with AH are ill for a couple
days, after which they recover spontaneously with or without supportive care without
knowing what has happened. The most aggressive form of AH, fulminant hepatitis, is
rapidly progressive within hours or days. In the author’s referral clinic, these patients
are not often seen (alive), probably because the time to get to the clinic via the referring
veterinarian is too long. Canine adenovirus-1 (CAV1) is a known cause for the devel-
opment of AH (sometimes fulminant). Because of vaccination, AH caused by CAV1
has been controlled effectively and practically eliminated from the domestic dog
population.
There is an impression that, although initially diagnosed as acute, some cases
remain acute for months on histopathology and might be considered more or less
CH, although fibrosis is lacking. This form of hepatitis, although not a WSAVA-classi-
fied form, might be considered subacute hepatitis. When starting with prednisone
treatment of subacute hepatitis in the clinically chronic stage, this hepatitis does not
seem to respond clinically or histopathologically.
Twenty-five percent of the dogs with AH had AH(ca). Hepatitis resulting from
primary copper accumulation starts somewhere in the process of developing hepatitis
and, depending on the stage at which the diagnosis is made, can be acute or chronic.
This finding suggests that even when dealing with a dog that has hepatitis with
a history of sudden onset, suggesting an acute inflammation, copper may be the
cause, and it is advisable to ask for a routine copper staining in this type of patient.
From 21 dogs with AH initially diagnosed, at least five had CH when a second liver
biopsy was taken 6 weeks later.

CHRONIC HEPATITIS AND CIRRHOSIS

CH is a frequently occurring disease in dogs. Approximately two thirds of the patients

with PH referred to the author’s university clinic had CH at initial diagnosis, and several
more patients with AH progressed histologically to CH. In humans, the diagnosis of CH
is based on patient history and results of histopathologic examination of liver biopsies.
Because the diagnosis is etiology-based, a specific etiology-focused treatment
approach is possible. In contrast, diagnosis in dogs is mainly morphology based
with severity based on the type and distribution of inflammatory cells, hepatocellular
apoptosis and necrosis, and the abundance and localization of fibrosis.1 In most
cases, there is no evidence of a cause, resulting in a large proportion of CH(i) cases.2
This poor understanding of the cause of CH(i) results in limited options for adequate
treatment and in variable results. Although probably not the initial cause, oxidative
stress plays an important role in the maintenance and progression of disease, and,
for this reason, antioxidants might play a beneficial role in the treatment of CH, at least.
For decades, canine CH(i) patients have been treated mainly with orally administered
immunosuppressive medication, of which prednisone is most commonly used. The
efficacy of prednisone has been described in one publication.10 In this retrospective
study, a prolonged survival time was demonstrated for prednisone-treated canine
CH patients in comparison with untreated patients. These results indicate a positive
long-term effect of prednisone in the treatment of CH(i). In addition to anti-inflamma-
tory effects, corticosteroids have (weak) antifibrotic properties. A retrospective histo-
pathologic evaluation of inflammatory activity and fibrosis formation in dogs with CH(i)
before and after a 6-week treatment with prednisone, although not double-blind or
placebo controlled, found a reduced inflammatory activity and a stable fibrotic
 Idiopathic Hepatitis and Cirrhosis in Dogs 483

situation, the latter suggesting a fibrosis inhibitory effect of prednisone (Poldervaart, in

preparation).

CIRRHOSIS

Fibrosis, a hallmark of CH, is defined as a detectable deposit of extracellular matrix

(ECM). Cirrhosis is the end stage of CH and is defined as a diffuse process character-
ized by fibrosis of the liver and the conversion of normal liver architecture into
structurally abnormal nodules, micro- or macronodular.1 Cirrhosis is the result of an
accumulation of ECM materials, which is the resultant of increased synthesis or
decreased breakdown. The bulk of ECM in the fibrotic liver is produced by myofibro-
blast (MF)-like cells. Three different MF-like cells are described based on location and
immunohistochemical profile.13 These comprise portal or septal MF; interface MF; and
perisinusoidally located hepatic stellate cells, which are nonparenchymal, quiescent
cells that are activated by hepatic injury and produce most of the factors that lead
to hepatic fibrosis. One of the most important of these factors is transforming growth
factor-beta (TGF-b), which acts via its two receptors, TGF-b receptor type I (TGF-b RI)
and TGF-b receptor type II (TGF-b RII), at the cell surface and the intracellular
substrates, the Smad proteins. TGF-b stimulates fibrosis by inducing the up-regula-
tion and the release of many of the ECM components (collagens and glycoaminogly-
cans) and inhibitors of the metalloproteinases, preventing the breakdown of the ECM.
Important regulators of ECM turnover and breakdown are plasminogen activator–
plasmin system components. Urokinase-type plasminogen activator (uPA) generates
plasmin from circulating plasminogen by proteolytic cleavage. This plasmin is capable
of degrading ECM components directly by proteolysis and indirectly by inhibiting
deposition of ECM by activation of matrix metalloproteinases. In this way, an up-regu-
lation of uPA in the liver might inhibit the deposition of ECM and reverse hepatic
fibrosis. Overall, matrix remodelling is an important component of liver regeneration.
Cirrhosis is considered irreversible, although the point at which this happens is not
well defined. Until now, no antifibrotic therapy has been clinically available, and in
humans, liver transplantation remains the only treatment option in cases of hepatic
dysfunction resulting from cirrhosis. Recent evidence suggests that the process of
fibrogenesis might be reversible,14,15 opening possibilities for evaluation of newly
designed antifibrotic therapies.

LOBULAR DISSECTING HEPATITIS

The clinical symptoms in dogs with LDH are more or less acute, but on histopathology,
they are diagnosed as having a CH with cirrhosis based on a massive deposition of
fibrous tissue around individual or small groups of hepatocytes. The cause of this
form of hepatitis is unknown. In most cases, patients with LDH are young animals
(average age 2.3 years in the author’s clinic) and die shortly after diagnosis, with an
estimated median survival time of 0.7  01 months (n 5 7, 2002–2006) despite treat-
ment consisting mostly of prednisone and sometimes diuretics because of the devel-
opment of severe ascites resulting from portal hypertension.

NONSPECIFIC REACTIVE HEPATITIS

Nonspecific reactive hepatitis (NSRH) is an aspecific response to extrahepatic disease

processes, especially inflammation somewhere in the splanchnic bed (gastrointestinal
tract or pancreas) or a systemic illness with fever. It also can be found as a residual
lesion of a previous inflammatory primary intrahepatic disease. NSRH, a secondary
484 Favier

problem, does not have to be treated. It is essential to look for the primary cause
(gastrointestinal tract or systemic illnesses) and to treat that underlying cause.

CLINICAL PRESENTATION

Patients with PH seen in the author’s clinic present, in most cases, with nonspecific
clinical signs. The most noticed symptoms mentioned by owners are decreased appe-
tite (50%), vomiting (48%), polyuria-polydipsia (47%), reduced activity (39%), weight
loss (28%), jaundice (24%), diarrhea (23%), and abdominal distension (21%). Episodic
neurologic symptoms resulting from hepatic encephalopathy rarely are mentioned.
Acholic feces, a specific indicator for extrahepatic biliary obstruction, are not seen
in cases of PH. Depending on the form of hepatitis, signs can be obvious from hours
(acute fulminant hepatitis) to months (CH).

DIAGNOSTIC EVALUATION
Physical Examination
At general appearance, dogs with CH, but not AH, have a slight to moderate muscle
atrophy. Dogs with ascites resulting from portal hypertension show abdominal
distension.
Physical examination related to liver diseases concentrates on the mucous
membranes and abdominal palpation. In cases of hepatomegaly, examination of the
circulation is indicated to detect or exclude cardiac disease. The mucous membranes
are normal in most patients with liver disease. Abnormalities may include icterus,
pallor, and indications for coagulopathy. Very pale mucous membranes in the pres-
ence of icterus indicate that the liver dysfunction is secondary to hemolytic anemia,
and a further workup should focus on this problem. Petechiae, an indicator for throm-
bocytopenia or thrombocytopathy, although rarely seen in patients with PH, can be
found as a result of disseminated intravascular coagulopathy.
Hepatomegaly, as a finding at abdominal palpation, is rare in dogs with PH. Ascites,
leading to abdominal distension, resulting from portal hypertension or hypoproteine-
mia, also may be an indication for liver disease, but there are many diseases of origins
other than hepatic that may induce ascites formation. Abdominal palpation also may
reveal splenomegaly in cases of portal hypertension.
In most dogs with hepatitis, physical examination reveals no specific information.
Therefore, in the majority of cases that involve symptoms (described previously),
laboratory investigation is required to detect or exclude a liver disease.

Blood
Blood work can be used for two purposes: first, to detect or to exclude a (primary) liver
problem, and second, to screen for the overall status of a patient. To address the first,
blood work should consist of, at minimum, serum bile acids, alkaline phosphatase
(AP), and alanine aminotransferase (ALT). In almost all cases of PH, one of these
parameters is out of the upper reference range. Liver enzymes are indicators for
cellular damage, whereas bile acids are a functional parameter. Of the liver enzymes,
ALT is the first to increase when PH is present. Adding more (liver) enzymes to the
diagnostic panel does not give more information. When patients with jaundice present
to a clinic, this part of the blood workup is not necessary. For the second purpose,
a complete blood cell count and a biochemistry and coagulation profile are necessary.
The biochemistry profile should include at minimum—in addition to serum bile acids,
AP, and ALT—urea, creatinine, total protein, albumin, sodium, and potassium. If
further workup liver biopsies are needed, a coagulation profile has to be determined.
 Idiopathic Hepatitis and Cirrhosis in Dogs 485

This test should be performed shortly before the biopsy procedure because coagula-
tion parameters may change quickly in patients with hepatitis (inadequate vitamin K
absorption, reduced production of coagulation factors, or increased consumption
[disseminated intravascular coagulopathy]). In the author’s clinic, prothrombin time,
activated thromboplastin time, fibrinogen concentration, and platelet count are
measured. Fibrinogen concentration, in particular, is a critical indicator, and a concen-
tration less than 1 g/L is a contraindication for taking a liver biopsy, which happens in
approximately 8% of cases of PH. This lowered fibrinogen concentration increases in
more than 90% of these cases above the critical level after a 1-week treatment with
prednisone/prednisolone so that a liver biopsy can be taken safely at that time.
There is some debate on what type of liver function tests to use for a functional
evaluation. Worldwide, the serum bile acid tolerance test (comparison of pre- and
postprandial serum bile acids) commonly is used. A major reason for this is that it is
easily accessible for private clinics because samples can be sent to laboratories for
measurement. The serum bile acid tolerance test does not give much additional
information regarding the liver function above only preprandial serum bile acid
concentration; for screening for portosystemic shunting, determination of the basal
plasma NH3 concentration is a better test.16 When the basal plasma NH3 measure-
ment is not informative, the rectally applied NH3 tolerance test confirms or excludes
the presence of portosystemic shunting. Measurement of plasma NH3, which should
be done immediately after sampling, can be more problematic in private clinics,
although in recent years equipment for ammonia measurement has become more
accessible for private clinics.

Ultrasound scan
For further diagnostic workup, ultrasound scan is needed. With ultrasonography, the
liver parenchyma, gallbladder/biliary tree, portal vein, acquired shunting (when
present), and ascites can be evaluated. In a recently performed evaluation of patients
with PH, in 20% of the cases no abnormalities were found at abdominal ultrasonog-
raphy. In approximately 25% of the cases, the liver was enlarged, irrespective of
type of hepatitis. Ascites, in cases of liver disease, resulting from portal hypertension,
often combined with a slightly to moderately decreased plasma albumin concentra-
tion, was found mainly in dogs with CH and LDH. In cases of ascites resulting from liver
failure, there also is a high chance of finding acquired portosystemic collaterals. The
best place to look is the region caudal to the left kidney. The finding of enlarged portal
lymph nodes, ascites, or a decreased liver size has a negative prognostic value.
Finally, ultrasonography is necessary for guiding liver biopsies with Tru-Cut needles
(Manan Medical Products, PBN Medicals, Stenlose, Denmark).

CT/MRI
For a further workup of patients with hepatitis, CT or MRI normally is not necessary. In
cases in which a primary liver tumor is suspected, based on ultrasonography, and
surgical intervention is needed, preoperative screening with CT or MRI with contrast
is helpful in estimating size and localization of the tumor and visualizing the presence
of tumor metastasis.

Liver Biopsy: Pathology

A liver biopsy is considered the gold standard for establishing a diagnosis of PH and to
differentiate, when necessary, PH from NRSH. Fine-needle aspiration is not sufficient
for diagnosing any form of hepatitis (primary or secondary). Liver biopsies can be ultra-
sound guided with a True-Cut needle or blind by aspiration using a syringe attached to
486 Favier

the needle (Menghini technique). At least two or more samples are advisable to mini-
mize sampling errors. Because approximately one third of the cases of PH referred to
the author’s clinic are copper associated, the author advocates routine staining for
copper (eg, with rubeanic acid) in addition to hematoxylin-eosin staining as standard
procedure for liver histology in dogs.

TREATMENT AND PROGNOSIS

Most cases of idiopathic AH do not need treatment, but, depending on the severity of
vomiting and presence of dehydration, antiemetic treatment and fluid therapy are indi-
cated. Most dogs with AH recover after several days without medical interference.
Progression from (initial) AH to its chronic counterpart may occur, resulting in recur-
rence of clinical signs.2 It is advisable to repeat the liver biopsy 6 to 8 weeks after
the initial diagnosis to control if the hepatitis has been solved or has progressed to
CH. CH(i) is treated as an immune-mediated disease with oral submission of predni-
sone or prednisolone combined with supportive therapy (eg, antiemetic, antidiuretic,
fluid therapy, and dietary adjustments). As discussed previously, only one publication
(a retrospective evaluation) is available on the efficacy of prednisone in the treatment
of CH in dogs, which showed a prolonged survival time for dogs with CH when treated
with prednisone (0.6–1.1 mg/kg per day).10 The response to prednisone therapy is
controlled on a regular basis by liver biopsy, usually at a 6-week interval, and therapy
is continued until histologically no hepatocellular death and inflammation are
observed. In humans, the application of glucocorticoid treatment is indicated in
alcohol-induced cirrhosis and autoimmune hepatitis in contrast to virally induced
hepatitis, where it is contraindicated. Histologic similarities between human virally
induced hepatitis and canine CH could indicate a reverse effect of prednisone
efficacy. The majority of dogs with CH(i) referred to the author’s clinic (2002–2006,
n 5 36) treated with prednisone (1 mg/kg per day), initially aimed at a 6-week treat-
ment period, showed an estimated median survival time of 9.9 months. When only
the CH(i) with cirrhosis cases treated with prednisone (n 5 19) were included, the
median survival time was 1.3 months, stressing that the presence of cirrhosis is
a strong negative prognostic indicator. In the past, when unacceptable side effects
(extreme polyuria-polydipsia, severely increased appetite, and reduced exercise toler-
ance) resulting from prednisone/prednisolone medication occurred, the author and
colleagues tapered the dosage of prednisone/prednisolone and started a combined
therapy with azathioprine (1 mg/kg per day) for 6 weeks. Because of increased aware-
ness of toxic side effects of cytostatic drugs in households (young children and preg-
nant women), however, and because treatment of CH(i) has no proved benefit, the
author does not advocate this combination therapy any longer.
Other than immunosuppressive medication, proposed medicinal options for treat-
ment of CH(i), based mainly on extrapolated human data and personal experiences,
are ursodeoxycholic acid (UDCA) (7.5 mg/kg twice a day); antioxidants, such as S-ad-
enosyl-L-methionine (SAMe) (10 mg/kg twice a day)17; silymarin (100–200 mg/dog
single oral administration); vitamin E (100–400 IU/day); and the antifibrotic drug,
colchicine (0.025 mg/kg/day). UDCA is a synthetic nontoxic hydrophilic bile acid
that provides a few positive actions. First, UDCA enhances the bile flow, and in this
way it stimulates the excretion of inflammatory products. Second, UDCA decreases
by dilution the concentration of the endogenous, more toxic bile acids. Third, it modu-
lates the immune system, resulting in a reduction of the immune response, and fourth,
there is proof that UDCA has antioxidative properties. The author’s clinic recently
started a trial with UDCA to evaluate if this drug might be a fair alternative for
 Idiopathic Hepatitis and Cirrhosis in Dogs 487

prednisone as a treatment of CH(i). SAMe is a natural metabolite in hepatocytes and is

a precursor of glutathione. It is important in the defense against oxidative stress, and
exhaustion might occur as a result of exposure to toxic substances in patients with
CH. Silymarin seems to act as a strong free-radical scavenger by increasing cellular
levels of superoxidedismutase, important in enzymatic defenses against oxidative
stress—it regulates cell membrane permeability and has been shown to inhibit leuko-
triene synthesis and the effects of tumor necrosis factor alpha. Evidence from many
human and veterinary reports underlines the protective effects of silymarin in patients
with mushroom or acetaminophen intoxications.18,19 Vitamin E is a nutritional antiox-
idant that protects against different routes of membrane peroxidation.
Colchicine has been proposed for treating CH fibrosis presumably by decreasing
the formation and increasing the breakdown of collagen, but benefit is unproved,
and there is little experience with colchicine in dogs. It can be a toxic drug after small
overdoses. It is not advisable to use this drug in dogs until it is proved effective.
Many of the medications discussed previously generally are accepted and clinically
used for the treatment of liver diseases, mostly as part of a multidrug therapy.
Unfortunately, until now, for most of these drugs, critical scientific evaluation of their
effectiveness is lacking.
If there is clinical evidence of portal hypertension (ascites or hepatic encephalop-
athy), treatment with spironolactone, a potassium-sparing diuretic (1–2 mg/kg twice
a day), lactulose (0.5 mL/kg 2 to 3 times a day), and dietary adjustments (high-quality
protein diets, moderately restricted) can be started. Spironolactone is preferred to
furosemide because of the underlying pathophysiology of portal hypertension in which
the rennin-angiotensin-aldosterone system is activated. In cases of severe ascites,
a combination of spironolactone and furosemide may be effective. Lactulose,
a synthetic disaccharide fermented by colonic bacteria into short-chain fatty acids,
helps acidify the colonic environment to trap NH3 (NH41) so that it remains mainly in
the feces and does not enter the portal circulation, reducing clinical signs of HE.
When the portosystemic collaterals are optimally activated, and plasma albumin
concentration is above the edema border (>15 g/L), ascites can disappear, and
diuretic treatment might be stopped. This activation of collaterals normally takes
2 to 3 weeks. Symptomatic treatment of gastric erosions and ulceration consists of
sucralfate (1 g by mouth 3 times a day) and a H2-blocker (ranitidine [1 mg/kg twice
a day], famotidine, or omeprazole [1 mg/kg once daily] [avoid cimetidine]).
In cases of CH(ca) or AH(ca), an etiology-based specific therapeutic approach is
applied by feeding a low-copper diet, submission of a copper chelator (eg, D-penicil-
lamine, 10–15 mg/kg twice a day), or submission of exogenous zinc (10 mg elemental
zinc/kg twice a day). Penicillamine has, in addition to metal chelating properties, an
immunomodulatory effect and possesses antifibrotic activity via inhibition of collagen
crosslinking, causing collagen to be more susceptible to degradation.

SUMMARY

Poor understanding of the causes of PH, especially CH(i), results in limited options for
adequate treatment and variable results. Elucidating the causes, aside from the
copper-associated form of hepatitis, is of utmost importance to find etiology-based
treatments for canine (chronic) hepatitis, when possible, most likely resulting in a better
prognosis. The prognosis for patients with CH(i), with developed cirrhosis, is poor.
Because many AH and CH cases are concluded to be copper associated
(25%–30%), it is advisable to ask for a copper staining (eg, rubeanic acid) in addition
to routine hematoxylin-eosin staining when sending liver materials to a pathology
488 Favier

department; otherwise, this diagnosis can be missed and patients do not get appro-
priate treatment with copper-binding agents.

REFERENCES

1. Van den Ingh TSGAM, Van Winkle TJ, Cullen JM, et al. Morphological classification
of parenchymal disorders of the canine and feline liver. In: Rothuizen J, Bunch SE,
Charles JA, et al, editors. Standards for clinical and histological diagnosis of canine
and feline liver diseases. Philadelphia: Saunders Elsevier; 2006. p. 85–101.
2. Watson PJ. Chronic hepatitis in dogs: a review of current understanding of the
aetiology, progression, and treatment. Vet J 2004;167:228–41.
3. Twedt DC, Sternlieb I, Gilbertson SR. Clinical, morphologic, and chemical studies
on copper toxicosis of Bedlington Terriers. J Am Vet Med Assoc 1979;175:269–75.
4. Thornburg LP, Shaw D, Dolan M, et al. Hereditary copper toxicosis in West
Highland white terriers. Vet Pathol 1986;23:148–54.
5. Haywood S, Rutgers HC, Christian MK. Hepatitis and copper accumulation in
Skye terriers. Vet Pathol 1988;25:408–14.
6. Mandigers PJ, van den Ingh TS, Bode P, et al. Association between liver copper
concentration and subclinical hepatitis in Doberman Pinschers. J Vet Intern Med
2004;18:647–50.
7. Webb CB, Twedt DC, Meyer DJ. Copper-associated liver disease in Dalmatians:
a review of 10 dogs (1998–2001). J Vet Intern Med 2002;16:665–8.
8. van de Sluis B, Rothuizen J, Pearson PL, et al. Identification of a new copper
metabolism gene by positional cloning in a purebred dog population. Hum Mol
Genet 2002;11:165–73.
9. Hoffmann G, van den Ingh TS, Bode P, et al. Copper-associated chronic hepatitis
in Labrador Retrievers. J Vet Intern Med 2006;20:856–61.
10. Strombeck DR, Miller LM, Harrold D. Effects of corticosteroid treatment on
survival time in dogs with chronic hepatitis: 151 cases (1977–1985). J Am Vet
Med Assoc 1988;193:1109–13.
11. Sevelius E. Diagnosis and prognosis of chronic hepatitis and cirrhosis in dogs.
J Small Anim Pract 1995;36:521–8.
12. Shih JL, Keating JH, Freeman LM, et al. Chronic hepatitis in Labrador Retrievers:
clinical presentation and prognostic factors. J Vet Intern Med 2007;21:33–9.
13. Yzer J, Roskams T, Molenbeek RF, et al. Morphological characterisation of portal
myofibroblasts and hepatic stellate cells in the normal dog liver. Comp Hepatol
2006;5:7.
14. Bonis PA, Friedman SL, Kaplan MM. Is liver fibrosis reversible? N Engl J Med
2001;344:452–4.
15. Friedman SL. Reversibility of hepatic fibrosis and cirrhosis—is it all hype? Nat Clin
Pract Gastroenterol Hepatol 2007;4(5):236–7.
16. Gerritzen-Bruning MJ, van den Ingh TS, Rothuizen J. Diagnostic value of fasting
plasma ammonia and bile acid concentrations in the identification of portosyste-
mic shunting in dogs. J Vet Intern Med 2006;20:13–9.
17. Center SA, Karen LW, McCabe J, et al. Evaluation of the influence of S-adenosyl-
methionnine on systemic and hepatic effects of prednisolone in dogs. Am J Vet
Res 2005;66:330–41.
18. Vogel G, Tuchweber B, Trost W, et al. Protection by silibinin against Amanita
phalloides intoxication in beagles. Toxicol Appl Pharmacol 1984;73(3):355–62.
19. Saller R, Brignoli R, Melzer J, et al. An updated systematic review with meta-analysis
for the clinical evidence of silymarin. Forsch Komplementmed 2006;15(1):9–20.