Harmonisation of LFT profiles facilitated by Tony Badrick

6 November 2013


Routine laboratory tests such as liver function tests (LFTs) are frequently ordered for patients with non-specific symptoms or as part of a health check. In Australia, LFTs comprise between eight and 11 analytes. However, the composition of test profiles varies between laboratories, states and countries, causing confusion and debate about the cost-effectiveness of these tests. In addition, the existing literature is predominantly retrospective and derived from hospital practice. There is no direct evidence that validates LFTs for the prediction of liver disease in primary care.

Paper(s) presented

The Birmingham and Lambeth Liver Evaluation Testing Strategies (BALLETS) study1 in the UK was the first prospective study that aimed to assess the prognostic value of LFTs for significant liver disease in primary care. 1290 patients with no obvious or pre-existing liver disease, and one or more abnormal analytes in an LFT panel of eight, were fully investigated for liver disease.

The main conclusions were: 

  • Less than 5% of people with incidental abnormal LFTs had a specific liver disease. This is only slightly higher than the prevalence of liver disease in the general population.
  • Abnormal alanine aminotransferase (ALT) was associated with hepatocellular disease while alkaline phosphatase (ALP) was associated with biliary disease and tumours of the hepatobiliary system.
  • When the sole purpose of testing is to exclude liver disease in primary care, a restricted panel of just two analytes (ALT and ALP) is sufficient to exclude liver disease of viral, genetic, autoimmune or neoplastic origin.
  • Requesting a standard LFT panel with a view to repeating it if abnormal is inefficient.

The results from this study call for a major rethink in the use of LFTs for liver disease diagnosis in primary care. It also provides evidence to limit and harmonise the LFT test profile. This issue is also addressed in a paper by WSA Smellie2, summarizing a draft proposal from the UK Association for Clinical Biochemistry. The ACB suggests a 4-analyte profile comprising Bilirubin, ALP, ALT and albumin.


  • Everyone agreed that the current practice of LFT use for the diagnosis or prognosis of liver disease in primary care is inefficient. The reason for this inefficiency likely stems from the fact that LFTs were originally developed in the hospital setting to monitor critically ill patients and not to diagnose or predict disease.
  • A large part of the discussion was focused on the costs of laboratory tests. You can currently get a rebate for up to five LFT analytes, but additional tests only cost the diagnostic lab AUD 0.05. Infrastructure as well as sample collection, transport and storage are the significant hidden costs in a laboratory test; once the sample is in the lab, adding more tests costs very little. Hence reducing the number of tests will only provide moderate cash savings to both the laboratory and the patient.  Medical laboratories offer larger test panels usually to provide a competitive advantage or simply because commercial test kits include multiple analytes. This is also why GPs sometimes receive Gonorrhoea test results in addition to the Chlamydia screening that they ordered.
  • A much more important reason to reduce the number of test analytes is that more tests can lead to more false-positives.
  • GPs order LFTs for screening, monitoring and diagnosis. One cost-effective solution to reduce the number of tests could be to introduce a cascade model of funding. Run two tests initially, then if they come back abnormal, repeat the test with more analytes.
  • Reference intervals are a real issue: there was a significant difference in ALP reference values between the three labs that participated in the BALLETS study. This should be taken into account when attempting to harmonise test profiles.
  • Measuring liver enzymes does not assess the degree of liver injury or the clinical outcome. The AST/ALT ratio (De Ritis ratio) does and has indeed been suggested as a marker of liver fibrosis and disease prognosis.


  1. Lilford RJ, Bentham LM, Armstrong MJ, et al. What is the best strategy for investigating abnormal liver function tests in primary care? Implications from a prospective study. BMJ Open (2013);3:e003099. doi:10.1136/bmjopen-2013-003099  
  2. Smellie WSA. Time to harmonise common laboratory profiles. BMJ (2012) Mar 20;344:e1169. doi: 10.1136/bmj.e1169.

(Summary provided by Michele Weber)

Click on the links to read the articles