Hepatic Granulomas
A multifactorial infiltrative liver disorder with or without additional hepatic
inflammation and fibrosis. The term "granulomatous hepatitis" is often used,
but the condition is not a true hepatitis. Hepatic granulomas are found in about
3 to 10% of liver biopsies. They may be insignificant incidental findings, but
more often they reflect clinically relevant disease --usually a systemic disorder
rather than primary liver disease.
Etiology
The causes of liver granulomas are legion. Infectious disorders are the most
important: bacterial (eg, TB and other mycobacterial infections, brucellosis,
tularemia, actinomycosis); fungal (eg, histoplasmosis, cryptococcosis, blastomycosis);
parasitic (eg, schistosomiasis --the most important worldwide, toxoplasmosis,
visceral larva migrans); viral infections, which are less common (eg, infectious
mononucleosis, cytomegalovirus); and numerous others (eg, Q fever, syphilis,
cat-scratch fever).
Sarcoidosis is the most important noninfectious cause; liver involvement occurs
in about 2/3 of patients and occasionally is the dominant clinical manifestation.
A variety of drugs can be responsible (eg, quinidine, sulfonamides, allopurinol,
phenylbutazone). Hepatic granulomas can also occur in polymyalgia rheumatica
and other collagen-vascular diseases; in Hodgkin's disease, sometimes without
other morphologic evidence of the lymphoma; and in a host of other systemic
conditions.
Granulomas are less common in primary liver disease. Of these, primary biliary
cirrhosis is the only important cause; periportal granulomas are typical in
this disorder, especially in the early stages, usually coupled with other characteristic
histologic features. Small granulomas are occasionally seen in various other
liver diseases, most often associated with fat droplets (lipogranulomas), but
are of no clinical significance.
In many cases no etiology can be established. A few such patients have a syndrome
of recurrent fevers, myalgias, fatigue, and other systemic symptoms, often occurring
intermittently for years. Whether this "idiopathic granulomatous hepatitis"
is a specific syndrome or a variant of sarcoidosis is debated.
Pathophysiology
Granuloma formation is incompletely understood. The lesions are regarded as
a host attempt to protect against poorly soluble exogenous or endogenous irritants.
Immunologic mechanisms convert cells of the mononuclear phagocytic system into
the typical collection of epithelioid cells that constitute a granuloma; multinucleated
giant cells are believed to derive from fusion of macrophages.
In the liver, granulomas often incite little or no hepatocellular reaction
and merely serve as the morphologic clue to some underlying systemic process;
hepatic disease is not apparent clinically and liver function is well preserved.
However, when granulomas are part of a broader inflammatory reaction involving
the liver (eg, drug reactions, infectious mononucleosis), clinical and biochemical
evidence of hepatocellular dysfunction is usually present. Sometimes an aggressive
inflammatory response ensues around the granulomas, resulting in progressive
hepatic fibrosis and portal hypertension. This is typical of schistosomiasis
and occasionally occurs in extensive sarcoidal infiltration.
Symptoms, Signs, and Laboratory Findings
Clinical features reflect the underlying etiology. Granulomas themselves are
typically subclinical; even extensive infiltration usually produces only minor
hepatomegaly and little or no jaundice. Fever, malaise, and other systemic symptoms
are the usual presenting manifestations of an infective etiology; prolonged
FUO is especially common in TB and fungal infections. The history is critical
in establishing a drug etiology. Various systemic features may provide the clue
to sarcoidosis, collagen-vascular disease, lymphoma, and other causes. Signs
of primary liver disease are usually lacking, and hepatosplenomegaly is typically
absent or mild except in schistosomiasis.
In most situations, liver function test results are only mildly deranged, usually
with a disproportionate elevation of alkaline phosphatase. Bilirubin levels
are typically normal or only mildly elevated, unless concomitant hepatocellular
injury coexists. Enzyme values may simulate viral hepatitis if extensive hepatocellular
necrosis is present (eg, in infectious mononucleosis or a drug reaction). A
predominant cholestatic reaction suggests primary biliary cirrhosis, especially
if long-standing. Other laboratory abnormalities depend on the specific cause.
Diagnosis
Liver biopsy is essential for diagnosis and should be considered whenever a
systemic granulomatous disorder is suspected, even in the absence of apparent
hepatic involvement. Biopsy demonstrates granulomas and may provide histologic
evidence of the specific etiology (eg, schistosomal ova, caseation of TB, fungal
organisms, primary biliary cirrhosis). However, the morphologic pattern is often
nonspecific and the diagnosis must be pursued with appropriate studies (eg,
cultures, skin tests, laboratory and x-ray studies, and other tissue specimens).
Infective etiologies are especially important to establish in patients with
FUO; this task often proves challenging. A portion of the fresh biopsy specimen
should be sent for culture; special stains for acid-fast bacilli, fungi, and
other organisms can sometimes prove the cause, though negative results do not
exclude an infective etiology.
Prognosis and Treatment
Hepatic granulomas of infective or drug etiology regress completely after appropriate
therapy. Sarcoid granulomas may disappear spontaneously or persist for years,
usually without clinically important hepatic disease, but progressive fibrosis
and portal hypertension occasionally develop (sarcoidal cirrhosis). In schistosomiasis,
progressive portal scarring is the rule ("pipestem fibrosis"); hepatic function
usually remains well preserved, but increasing portal hypertension leads to
marked splenomegaly and risk of variceal hemorrhage.
Treatment depends on the underlying etiology. Without an etiologic diagnosis,
it is generally best to follow the patient rather than blindly treat with antibiotics
or other therapies. Antituberculous therapy may be justified in a patient with
prolonged fever, a compatible clinical picture, and a downhill systemic course.
Patients with progressive hepatic sarcoidosis may benefit from corticosteroid
therapy, though it remains unclear whether this prevents hepatic fibrosis; corticosteroids
are not indicated for most patients with sarcoidosis and should be given only
if TB and other infective disorders can confidently be excluded. Corticosteroids
usually suppress the recurrent fevers in patients with the syndrome of idiopathic
granulomatous hepatitis.
|