Page 27 - EASL POSTGRADUATE COURSE
P. 27
Values based on the prediction of advanced fibrosis. Studies with >100 subjects presented. Only ELF
validated in a separate cohort.
Elastography based tests
Hepatic elasticity and distention reduces as hepatic fibrosis worsens. This may be quantified by
measuring the velocity of a transmitted physical or sonographic impulse through the liver (shear wave
elastography), or by measuring the magnitude of liver tissue distention to external pressure or internal
movements such as the cardiac cycle (strain elastography).
Transient elastography (Fibroscan®)
A recent meta-analysis demonstrated modest accuracy for detecting moderate (F2+) fibrosis with
sensitivity and specificity values of <80% and AUROC of 0.79-0.87. Fibroscan® has better accuracy
for F3+ fibrosis (sensitivity 85%, specificity 82%, AUROC 0.76-0.98) and cirrhosis (92% sensitivity
and specificity, AUROC 0.91-0.99) [11]. Correspondingly, predictive values for F2+ fibrosis are modest
(PPVs 55-79%, NPVs 72-95%) but better for cirrhosis (PPVs 41-86%, NPVs 91-100%).
A limitation of Fibroscan® is acquisition failure or unreliable readings related to obesity, which occurs
in one quarter of patients when using the XL probe. Studies are conflicting as to whether BMI impacts
on the accuracy of Fibroscan®; however discordance between biopsy and Fibroscan® results increases
at a BMI threshold of 35 kg/m2.There is also a range of cut-offs in the literature that overlap for different
levels of fibrosis (Fig. 1), making it difficult to determine the significance of a mid-range reading. Cut-
offs should be adapted to the probe used, with readings 1.2-1.3 kPa lower with the XL probe compared
with the standard M probe. Severe hepatic steatosis increases liver stiffness measurements in chronic
hepatitis C infection and thus may also affect liver stiffness measurements in NAFLD.
Figure 1. Recommended Fibroscan® cut-offs for different stages of fibrosis in NAFLD.
Comparative studies have shown higher accuracy, although lower specificity and PPV, with Fibroscan®
when compared to non-proprietary serum-based tests such as FIB-4 and NFS [12, 13]. Specificity and
PPV are increased in Fibroscan® when higher cut-offs are utilized.
Acoustic radiation force impulse (ARFI)
ARFI sonographically measures the velocity of a shear wave generated by an acoustic impulse at a single
point in the liver. Only a limited number of studies have examined the accuracy in NAFLD and have
shown variable accuracy for the detection of advanced fibrosis (AUROC 0.6-0.97) and cirrhosis (0.74-
0.98). Similar to Fibroscan®, successful acquisition may be impacted by BMI and was unsuccessful
in 20% of patients with a BMI between 30-40 kg/m2 [14]. Further clarification of the impact of BMI,
hepatic steatosis, and inflammation on accuracy as well as validation of appropriate cut-offs is required.
The International Liver Congress™ 2015 • Vienna, Austria • April 22–23, 2015 27
validated in a separate cohort.
Elastography based tests
Hepatic elasticity and distention reduces as hepatic fibrosis worsens. This may be quantified by
measuring the velocity of a transmitted physical or sonographic impulse through the liver (shear wave
elastography), or by measuring the magnitude of liver tissue distention to external pressure or internal
movements such as the cardiac cycle (strain elastography).
Transient elastography (Fibroscan®)
A recent meta-analysis demonstrated modest accuracy for detecting moderate (F2+) fibrosis with
sensitivity and specificity values of <80% and AUROC of 0.79-0.87. Fibroscan® has better accuracy
for F3+ fibrosis (sensitivity 85%, specificity 82%, AUROC 0.76-0.98) and cirrhosis (92% sensitivity
and specificity, AUROC 0.91-0.99) [11]. Correspondingly, predictive values for F2+ fibrosis are modest
(PPVs 55-79%, NPVs 72-95%) but better for cirrhosis (PPVs 41-86%, NPVs 91-100%).
A limitation of Fibroscan® is acquisition failure or unreliable readings related to obesity, which occurs
in one quarter of patients when using the XL probe. Studies are conflicting as to whether BMI impacts
on the accuracy of Fibroscan®; however discordance between biopsy and Fibroscan® results increases
at a BMI threshold of 35 kg/m2.There is also a range of cut-offs in the literature that overlap for different
levels of fibrosis (Fig. 1), making it difficult to determine the significance of a mid-range reading. Cut-
offs should be adapted to the probe used, with readings 1.2-1.3 kPa lower with the XL probe compared
with the standard M probe. Severe hepatic steatosis increases liver stiffness measurements in chronic
hepatitis C infection and thus may also affect liver stiffness measurements in NAFLD.
Figure 1. Recommended Fibroscan® cut-offs for different stages of fibrosis in NAFLD.
Comparative studies have shown higher accuracy, although lower specificity and PPV, with Fibroscan®
when compared to non-proprietary serum-based tests such as FIB-4 and NFS [12, 13]. Specificity and
PPV are increased in Fibroscan® when higher cut-offs are utilized.
Acoustic radiation force impulse (ARFI)
ARFI sonographically measures the velocity of a shear wave generated by an acoustic impulse at a single
point in the liver. Only a limited number of studies have examined the accuracy in NAFLD and have
shown variable accuracy for the detection of advanced fibrosis (AUROC 0.6-0.97) and cirrhosis (0.74-
0.98). Similar to Fibroscan®, successful acquisition may be impacted by BMI and was unsuccessful
in 20% of patients with a BMI between 30-40 kg/m2 [14]. Further clarification of the impact of BMI,
hepatic steatosis, and inflammation on accuracy as well as validation of appropriate cut-offs is required.
The International Liver Congress™ 2015 • Vienna, Austria • April 22–23, 2015 27
