Statistics from Altmetric.com
Q1: What does the computed tomogram show and in the light of this, what is the most appropriate investigation?
Computed tomography of the abdomen confirms hepatosplenomegaly, a small amount of ascites and small bilateral pleural effusions (fig 1; see p 242). In view of the ascites, transjugular rather than transabdominal liver biopsy should be performed. All patients should have both a coagulation screen, platelet count, and ultrasound scan of the liver performed before liver biopsy. If the prothrombin time is 4 seconds above the control value (corresponding to international normalised ratio >1.4) then the coagulopathy should be corrected before liver biopsy. Ultrasound is essential to exclude intrahepatic duct dilatation, hydatid cyst, haemangiomata, and the presence of ascites. A transjugular liver biopsy was performed without complications.
Q2: What is the histological diagnosis? List the possible causes for this
Hepatic histology demonstrates granulomatous hepatitis. There are characteristic granulomas, both intralobular and portal tract in distribution (fig 2; see p 242). One of the granulomas is large containing fibrin-like material. Ziehl-Neelsen stains for tuberculosis and periodic acid-Schiff stains for fungi were all negative. The surrounding parenchyma showed regenerative change with focal steatosis. No fibrin rings (doughnuts) or central fat vacuoles are seen. There was fibrosis around the portal tracts of a mild degree but there was no bridging fibrosis or cirrhosis. These findings are in keeping with granulomatous hepatitis.
Possible causes for granulomatous hepatitis include:
Infections: tuberculosis, brucellosis, Q fever, syphilis, histoplasmosis, infectious mononucleosis, and AIDS.
Drugs: sulphonamides, allopurinol, carbamezpine, quinine, and phenylbutazone.
Others: sarcoidosis, beryllosis, primary biliary cirrhosis, and lymphoma.
In 5%–10% no cause is found.
Q3: Can you link the histological diagnosis with the clinical history described and suggest the management of this condition?
This man presented with features of hepatic decompensation due to granulomatous hepatitis. The diagnosis was confirmed after discharge when the atypical pneumonia screen results were available. Q fever titres to the phase II antigen were greatly raised at 1280 indicating acute Q fever infection. Phase I antigen titres were less than 80. The diagnosis of Q fever is based on serology. Antibodies to the phase II antigen occur early in the course of Q fever, followed by a late rise in antibodies to the phase I antigen in the chronic state. Hepatitis is known to be associated with a high titre of antibodies to phase II antigen, as seen in this case. Q fever hepatitis usually manifests with a small increase in liver enzymes, rising to approximately twice or three times the upper limit of normal.1 Liver involvement may occur in the absence of pulmonary involvement.2 The association between the fibrin ring or doughnut granulomas and Q fever has been confirmed.3 They are, however, not pathognomonic for this disease and occur in a variety of conditions including cytomegalovirus and Epstein-Barr virus infections, systemic lupus erythematosis, and allopurinol hypersensitivity. The histological response pattern to coxiella infection is often varied.
In our patient, initial testing for hepatitis C and HIV demonstrated reactive antibodies. It has been found that patients with Q fever hepatitis and endocarditis have circulating immune complexes to a component of Coxiella burnetti.4 These can interfere with complement fixation antibody assays, thus producing false negative results.
Q fever is endemic in Northern Ireland, with incidence levels only second to south western England in the United Kingdom. The incidence of Q fever peaks in April to May, which can probably be in part be explained by the calving and lambing season.2 In Northern Ireland there were six deaths between 1962 and 1989 attributable to Q fever.5 These deaths were due to cardiac complications and one suicide. Q fever can be thought of as an occupational hazard, affecting farmers, abattoir workers, and veterinarians. However the organisms can survive for long periods in the environment, so the population as a whole is at risk. There were no occupational risk factors in this case.
Q fever is a treatable disease, with tetracyclines being the antibiotics of choice.1 Our patient was treated initially with a two week course of oxytetracyline 500 mg four times a day, and subsequently re-treated with a further four week course in view of ongoing night sweats. He remained well at six month follow up with no significant symptoms and mildly cholestatic liver enzymes.
As our case demonstrates Q fever may cause acute hepatitis and acute liver decompensation. The diagnosis should always be considered in patients with an unexplained pyrexia and hepatitis, especially in endemic regions.