Page 24 - EASL POSTGRADUATE COURSE
P. 24
Challenges in the diagnosis of NASH
NON-INVASIVE DIAGNOSIS OF FIBROSIS IN NAFLD:
HOW RELIABLE IS IT?
Leon A. Adams
School of Medicine and Pharmacology QEII Medical Centre Unit,
The University of Western Australia (M503),
WA, Australia
Email: leon.adams@uwa.edu.au
Take home messages
• Simple and complex serum based tests have >90% predictive value for excluding cirrhosis, though
are poorly predictive of cirrhosis.
• The NAFLD Fibrosis Score (NFS) and FIB-4 algorithm are clinical/serum based tests which can
confirm or exclude bridging fibrosis/cirrhosis but have indeterminate values in one quarter of patients.
• Serum-based tests, including the NFS, FIB-4 and Fibrotest, predict mortality outcomes in patients
with NAFLD or those with metabolic risk factors for NAFLD.
• Transient elastography is excellent at excluding advanced fibrosis and cirrhosis and has modest
predictive values, although acquisition is unsuccessful in one quarter of patients.
• All tests have been developed in the hospital clinic setting without validation in the general population
and have poor correlation when used as screening tools in this setting.
Introduction
Significant liver fibrosis is present in 5-10% of patients with NAFLD in the primary medical care setting.
Reliable identification of these patients is important as fibrosis is the dominant histological feature
that predicts outcomes and thus the need for treatment and surveillance. Furthermore, determination
of fibrosis is useful to monitor disease progression or treatment response over time. The limitations
of liver biopsy have led to the refinement of non-invasive techniques to predict liver fibrosis. These
include simple easy-to-use clinic-based tools, serum tests or imaging-based techniques that predict
fibrosis based on the physical elasticity of the liver. In general, these investigations provide information
on a continuous scale as opposed to the semi-quantitative histological staging systems utilized in liver
biopsies. Thus, non-invasive tests may be more sensitive to subtle alterations in fibrosis than biopsy and
provide greater prognostic information.
The accuracy of non-invasive fibrosis tests is often described using the area under the receiver operator
characteristic curve (AUROC), where a value of 1.0 reflects a ‘perfect’ test with 100% sensitivity and
specificity, and a value of 0.5 reflects a test as good as chance. Although two tests may have equivalent
AUROC values, they may have different sensitivity and specificity values depending upon the characteristics
of the test and where the individual test cut-offs have been drawn. Typically, values at one end of the test
result spectrum will have a high sensitivity and low specificity, whereas values at the opposite end of the test
result spectrum will have a low sensitivity and high specificity. Test results that fall in-between often have
moderate sensitivity and specificity and are not clinically meaningful, and thus comprise an ‘indeterminate
range’.Therefore, the majority of fibrosis tests will produce inconclusive results for a proportion of patients
falling within the indeterminate range for a specific fibrosis end-point. It is also important to note that the
24 Postgraduate Course Syllabus • Metabolic Liver Disease
NON-INVASIVE DIAGNOSIS OF FIBROSIS IN NAFLD:
HOW RELIABLE IS IT?
Leon A. Adams
School of Medicine and Pharmacology QEII Medical Centre Unit,
The University of Western Australia (M503),
WA, Australia
Email: leon.adams@uwa.edu.au
Take home messages
• Simple and complex serum based tests have >90% predictive value for excluding cirrhosis, though
are poorly predictive of cirrhosis.
• The NAFLD Fibrosis Score (NFS) and FIB-4 algorithm are clinical/serum based tests which can
confirm or exclude bridging fibrosis/cirrhosis but have indeterminate values in one quarter of patients.
• Serum-based tests, including the NFS, FIB-4 and Fibrotest, predict mortality outcomes in patients
with NAFLD or those with metabolic risk factors for NAFLD.
• Transient elastography is excellent at excluding advanced fibrosis and cirrhosis and has modest
predictive values, although acquisition is unsuccessful in one quarter of patients.
• All tests have been developed in the hospital clinic setting without validation in the general population
and have poor correlation when used as screening tools in this setting.
Introduction
Significant liver fibrosis is present in 5-10% of patients with NAFLD in the primary medical care setting.
Reliable identification of these patients is important as fibrosis is the dominant histological feature
that predicts outcomes and thus the need for treatment and surveillance. Furthermore, determination
of fibrosis is useful to monitor disease progression or treatment response over time. The limitations
of liver biopsy have led to the refinement of non-invasive techniques to predict liver fibrosis. These
include simple easy-to-use clinic-based tools, serum tests or imaging-based techniques that predict
fibrosis based on the physical elasticity of the liver. In general, these investigations provide information
on a continuous scale as opposed to the semi-quantitative histological staging systems utilized in liver
biopsies. Thus, non-invasive tests may be more sensitive to subtle alterations in fibrosis than biopsy and
provide greater prognostic information.
The accuracy of non-invasive fibrosis tests is often described using the area under the receiver operator
characteristic curve (AUROC), where a value of 1.0 reflects a ‘perfect’ test with 100% sensitivity and
specificity, and a value of 0.5 reflects a test as good as chance. Although two tests may have equivalent
AUROC values, they may have different sensitivity and specificity values depending upon the characteristics
of the test and where the individual test cut-offs have been drawn. Typically, values at one end of the test
result spectrum will have a high sensitivity and low specificity, whereas values at the opposite end of the test
result spectrum will have a low sensitivity and high specificity. Test results that fall in-between often have
moderate sensitivity and specificity and are not clinically meaningful, and thus comprise an ‘indeterminate
range’.Therefore, the majority of fibrosis tests will produce inconclusive results for a proportion of patients
falling within the indeterminate range for a specific fibrosis end-point. It is also important to note that the
24 Postgraduate Course Syllabus • Metabolic Liver Disease
