Article Text

PDF

Dissecting out crytogenic liver disease
  1. J G C KINGHAM
  1. Department of Gastroenterology
  2. Singleton Hospital, Swansea, SA2 8QA, UK Email:jerry.kingham{at}swansea-tr.wales.nhs.uk

    Statistics from Altmetric.com

    See article on page 429

    The report from Pamplona in this issue ofGut 1 gives insight into the spectrum of liver disease in Southern Europe and emphasises the importance of careful laboratory evaluation (see page 429). A total of 1075 consecutive patients who had undergone liver biopsy because of elevated transaminases were assessed retrospectively. In 90% of cases the cause was clear on conventional clinical and serological criteria: most commonly chronic viral hepatitis was responsible with hepatitis C predominating. The study focuses on the remaining 10% in whom the cause was not obvious, in particular using polymerase chain reaction (PCR) on stored serum to detect hepatitis viral sequences, and careful review of the biopsy specimens. The histological review showed cirrhosis in 13, chronic hepatitis in 39, non-alcoholic steatohepatitis (NASH) in 16, and non-specific changes in 33 patients.

    The most important finding was the high prevalence of hepatitis B and C sequences, albeit at low levels: 19 patients had hepatitis B virus (HBV) DNA and nine had hepatitis C virus (HCV) RNA in their serum (one had both). None had evidence of HBV or HCV infection by conventional antigen/antibody testing although 30 had evidence of previous exposure to HBV. Sera from 59 seronegative blood donors of similar age were tested as controls and revealed one case each of HBV DNA and HCV RNA positivity. There was a strong correlation between cirrhosis/chronic hepatitis and viraemia but none with serological evidence of past hepatitis B exposure. Thus of the starting cohort of 1075 patients with elevated transaminases, almost two thirds were eventually shown to have chronic viral hepatitis (C being five times commoner than B) of which a small proportion were surreptitious.

    The problem of occult hepatitis infection and coinfection is increasingly recognised. Last year Cacciola and colleagues2 in Sicily reported HBV sequences in the sera of one third of 200 patients with hepatitis C related liver disease all of whom were hepatitis B virus surface antigen (HBsAg) negative. Most had antibodies to hepatitis B core antigen but a sizeable minority lacked any serological HBV markers. The occult nature of chronic hepatitis infection was taken a step further in a study from Edinburgh.3 Of 98 consecutive patients with antibodies to HCV, there were 12 repeatedly negative for HCV RNA in serum by sensitive reverse transcription-PCR. However, 10 of these 12 were shown to harbour low levels of HCV RNA in liver tissue.

    Surreptitious viral hepatitis infection has also been shown to play a crucial role in hepatocellular carcinoma (HCC). The European cooperative study reported by Brechot and colleagues4in 1998 investigated hepatitis B and C in more than 500 patients with HCC from six European countries comparing serology with molecular testing. Conventional serology was positive for active hepatitis B and C infection in only 19% and 40% of cases whereas PCR detected genomic sequences of B and C viruses in 48% and 43%, while 20% were infected by both.

    The message from these observations is that conventional serology detects most cases of hepatitis C infection in chronic hepatitis, cirrhosis, and HCC but may miss 20–50% of cases of hepatitis B infection. PCR of serum will pick up most missed cases of viral infection depending on the sensitivity of the assay but PCR of liver tissue may be more sensitive still.

    These data highlight the dominance of chronic viral hepatitis over other forms of liver disease in many parts of the world.5Those working in areas where viral hepatitis is uncommon must take these reports in context. Occult hepatitis B and C has been shown to be numerically important only where overt infection is highly prevalent. The studies cited show that PCR analysis of serum or liver tissue can improve the diagnostic yield of conventional serology by 10–30% depending on the local balance of hepatitis B to hepatitis C. This would not have much impact for instance in South Wales and doubtless other provincial areas of the UK where chronic viral hepatitis is still uncommon. PHLS data for a population of 977 000 in South West Wales showed that in 1999 there were only 44 positive tests for hepatitis B and 134 for hepatitis C out of a total of 16 593 sera tested (Dr M Isaac, PHLS, Swansea NHS Trust). In this part of the world alcohol is a much more important cause of abnormal liver function, cirrhosis, and HCC although sometimes alcohol and hepatitis infection combine to give more advanced disease.6 The Dionysos study7recording the prevalence of chronic liver disease in two towns in Northern Italy showed that there too, alcohol was by far the most important cause (23%). Chronic viral hepatitis, although often associated with more serious disease, accounted for only 4.5% of cases.

    Coeliac disease is another cause of chronic transaminase elevation that has come to light. It was found to be responsible for approximately 10% of apparently idiopathic cases from Italy in two recent reports.8 9 This should be an important consideration among populations of Celtic origin.

    NASH has risen from obscurity and scepticism to great prominence over the past 20 years.10 This diagnosis was made in one in six of the cryptogenic Pamplona cases1 but there may be some dispute about definition. The diagnostic criteria put forward in 1990 by Powell and colleagues11 included alcohol consumption below 40 g per week. In Pamplona, NASH was diagnosed in patients drinking up to 80 g perday (that is, a 14-fold difference). This may reflect the more generous and tolerant attitude to alcohol in Spain. I suspect one man's NASH is another man's ASH.

    Within a generation we have moved from labelling much or even most liver disease idiopathic to identifying a cause in over 90%. The challenge is to find causes for the remainder.

    See article on page 429

    References

    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles