elastography, fibroscan, liver stiffness measurement.

1. Liver stiffness measurement (Fibroscan®)
Principles - indications - results - limitations
Samir Haffar M.D.
Assistant Professor of Gastroenterology

2. Fibrose
Blood markers
FibroScan®
Imaging
(US, MRI, endoscopy)
Hepatic biopsy
Biological work-up
Clinical Examination

3. Ideal non-invasive test for diagnosis
of liver fibrosis
• Simple
• Reproducible
• Readily available
• Less expensive than biopsy
• Predicts full spectrum of fibrosis
• Reflects changes occurring with therapy

4. Evaluation of chronic liver injury according
to health care level
Physical examination
Liver function tests
Serum Hyaluronate
APRI or other simple tests
Primary health care
Ultrasound
Fibroscan®
Serum markers &
algorithms
Secondary health care
Fibroscan®
ARFI*
MR elastography*
Tertiary health care
Liver biopsy
HVPG
* Promising but currently under investigation
ARFI: Acoustic Radiation Force Impulse Imaging
HVPG: Hepatic Venous Pressure Gradient
Castéra L et al. Gut 2010 ; 59 : 861 – 866.

5. Liver stiffness
• Assessed by US (FibroScan®) & more recently by MRI
• Evaluates velocity of propagation of a shock wave
within liver tissue (examines a physical parameter of
liver tissue which is related to its elasticity)
• Rationale Normal liver is viscous
Not favorable to wave propagation
Fibrosis increases hardness of tissue
Favors more rapid propagation
Bedossa P. Liver Int 2009 ; 29 (s1): 19 – 22.

6. Fibroscan® device
• Electronic platform
– Ultrasonic signals acquisition
– Numerical signal processing
• Integrated computer
– Stiffness measurement
– Examinations database
• Dedicated probes with unique
technology
Vibrator (50 Hz)
US Transducer
(3,5 MHz)
Fibroscan® (Echosens, Paris, France)

7. Position of probe & explored volume
Cylinder of 1 cm wide & 4 cm long
From 25 mm to 65 mm below skin surface
This volume is at least 100 times bigger than a biopsy sample

8. Results
Stiffness (kPa)
Median value of 10 shots
3.9 Kilo Pascals
 At least 10 shots
 Success Rate: ≥ 60%
 IQR * (kPa)
Interval around median
Contains 50% of valid shots
≤ 25% of median value

9. Manufacturer’s criteria for LSM interpretation
• First step Number of shots ≥ 10
• Second step Success rate ≥ 60 %
• Third step Interquantile range(IQR) ≤ 25%
Failure
Zero valid shot
Unreliable results
< 10 valid shots
Success rate ≤ 60%
IQR ≥ 25%

10. Liver stiffness values in healthy subjects
429 subjects
Roulot D et al. J Hepatol 2008 ; 48 : 606 – 613.
5.2 1.5 kPa 5.8 1.5 kPa
p = 0.0002

11. Liver stiffness values in healthy subjects
with & without metabolic syndrome
Roulot D et al. J Hepatol 2008; 48 : 606 – 613.
5.3 1.5 kPa 6.5 1.6 kPa
p < 0.0001

12. Liver stiffness cut-offs in chronic liver diseases
F2
Sign
F3
Severe
F4
Cirrhosis
Matavir F0-F1
MildFibrosis
Castéra L et al. J Hepatol 2008 ; 48 : 835 – 847.
LSM 2.5 – 7 kPa → Mild or absent fibrosis is likely
LSM > 12.5 kPa → Cirrhosis is likely

13. Progression of fibrosis in viral hepatitis
Photomicrographs (magnification 40; trichrome stains)
Faria SC et al. RadioGraphics 2009 ; 29 : 1615 – 1635.
F 0 F 1 F 2
F 3
F 4

14. Perform LSM
≤ 6 kPa
No significant fibrosis
No biopsy
F0 F1F
Intermediate values
Grey area
Biopsy if results
influence management
F2
Implementation of
other NI tests
≥ 12 kPa
Advanced fibrosis
No biopsy
F4
Treatment or
Follow-up
F3
Vizzutti et al. Gut 2009;58:156-60.

15. Shear wave propagation velocity according to
severity of hepatic fibrosis (Metavir score)
Sandrin L. Ultrasound Med Biol 2003 ; 29 : 1705 – 1713.
E = 3.0 kPa
F 0
E = 7.7 kPa
F 2
E = 27 kPa
F 4

16. Liver stiffness for each Metavir stage in CHC
Box-and-whiskers plot
Vertical axis is in logarithmic scale (wide range of F4 values)
Gastroenterology 2005; 28:343 – 350.Hepatology 2005;41:48 – 54.

17. Correlation between LSM & fibrosis stage
* Gastroentérol Clin Biol 2008;32,58-67.
** J Hepatol 2009;49:1062-68, Aliment Pharmacol Ther 2008;28:1188-98.
*** Hepatology 2010;51:454-62. Gastroentérol Clin Biol 2008;32:58-67.

18. Accuracy of a diagnostic test
• Dichotomous test (only 2 results)
Sensibility (Sn)
Specificity (Sp)
Positive Predictive Value (PPV)
Negative Predictive Value (NPV)
Likelihood Ratios + & – (LRs)
Diagnostic Odds Ratio (OR)
• Multilevel test (> 2 results)
Receiver Operating Characteristic (ROC)
Newman TB & Kohn MA. Evidence-based diagnosis.
Cambridge University Press, Cambridge, UK, 1st edition, 2009.
CIs

19. Hypothetical ROC curve
Pines JM & Everett WW. Evidence-Based emergency care: diagnostic testing & clinical
decision rules. Blackwell’s publishing, West Sussex, UK, 2008.

20. Accuracy of diagnostic test using AUC of ROC
Value Accuracy
0.90 - 1.00 Excellent
Pines JM & Everett WW. Evidence-Based emergency care: diagnostic testing & clinical
decision rules. Blackwell’s publishing – West Sussex – UK – 2008.
AUROC of a ‘‘good” test should be ≥ 0.80
0.80 - 0.90 Good
0.70 - 0.80 Fair
0.60 - 0.70 Poor

21. Meta-analysis of TE for staging liver fibrosis
Severe fibrosis (F3): 0.89
(95% CI: 0.88 – 0.91)
Friedrich R et al. Gastroenterology 2008 ; 134 : 960 – 974.
Cirrhosis (F4): 0.94
(95% CI: 0.93 – 0.95)
Cut-off value: 13.0 kPa
Significant fibrosis (F2): 0.84
(95% CI: 0.82 – 0.86)
Cut-off value: 7.7 kPa
50 studies – random effect – all type of CLD

22. Significance of wide range of LSM in cirrhosis
13 - 75 kPa
Ascites
HCC ?
Variceal bleeding
Foucher J et al. Gut 2006 ; 55 : 403 – 408.
EV stage 2 or 3
26
Child-Pugh B or C
36 49 53 622.5 7513

23. Cumulative incidence of HCC based on LSM
866 CHC – Mean follow-up 3 years
LSM: Liver Stiffness Measurement – HR: Hazard Ratio
Masuzaki R et al. Hepatology 2009 ; 49 : 1954 – 1961.
LSM > 25 kPa HR 45.5 (p< 0.001)
LSM ≤ 10 kPa HR 0
10 < LSM ≤ 15 kPa HR 16.7 (p< 0.001)
15 < LSM ≤ 20 kPa HR 20.9 (p< 0.001)
20 < LSM ≤ 25 HR 25.6 (p< 0.001)

24. Reproducibility of TE in assessing hepatic fibrosis.
Bland Altman Plot
Fraquelli M et al. Gut 2007 ; 56 : 968 – 973.
200 patients with chronic liver disease
2 different operators within 3 days (800 exams)
8 patients scored outside limits of agreement
Upper limit
Lower limit
Mean
95% limit of
agreement

25. Cost of FibroScan® versus liver biopsy
• Liver biopsy*
Cost of liver biopsy 703 – 1 566 € in a French hospital
with a one day observation period
• Fibroscan® **
FibroScan equipment 70 000 €
Low running cost except probe calibration twice/year
Cost per FibroScan exam 100 € with 150 exams annually
* Blanc J et al. Hepatol Res 2005 ; 32 : 1 – 8.
** Canadian Agency for Drugs and Technologies in Health (CADTH).
Transient Elastography (FibroScan) for Non-invasive Assessment of Liver Fibrosis; 2006.

26. Liver biopsy size
• Because grading, & staging of nonneoplastic diffuse
parenchymal liver disease is dependent on adequate
sized biopsy, a biopsy of at least 2-3 cm in length
& 16-gauge in caliber is recommended
• Presence of fewer than 11 complete portal tracts in
pathology report may be incorrect in recognition of
grading, & staging due to insufficient sample size
AASLD guidelines. Hepatology, 2009 ; 49 : 1017 – 1044.

27. Limitations of liver biopsy
• Sampling errors
Extremely small portion of liver (1/50 000)
• Intraobserver & interobserver variation
Even when widely validated systems used for score
• Invasive procedure
Morbidity: pain in 20% of patients
Major complications: bleeding or hemobilia in 0.5%
Mortality:

28. Grading & staging systems for chronic hepatitis
IASL1 Batts–Ludwig2 Metavir3
1 Desmet VJ et all. Hepatology 1994;19:1513-1520.
2 Batts KP et all. Am J Surg Pathol 1995;19:1409-1417.
3 Bedossa P et all. Hepatology 1996;24:289-293.
Grading system
Minimal activity
Mild activity
Moderate activity
Marked activity
Marked activity & bridging
Grade 1
Grade 2
Grade 3
Grade 4
Grade 4
A1
A2
A3
A3
A3
Staging system
No fibrosis
Fibrous portal expansion
Few bridges or septa
Numerous bridges
Cirrhosis
Stage 0
Stage 1
Stage 2
Stage 3
Stage 4
F0
F1
F2
F3
F4
(kappa 0.2 – 0.6)
(kappa 0.5 – 0.9)

29. kappa score ≥ 0.6 indicates good agreement
Interpretation of different values of kappa
Kappa from Greek letter κ
Value of kappa Strength of agreement
0 – 0.20 Poor
0.21– 0.40 Fair
0.41– 0.60 Moderate
0.61– 0.80 Good
0.81–1.00 Very good
Perera R, Heneghan C & Badenoch D. Statistics toolkit.
Blackwell Publishing & BMJ Books, Oxford, 1st edition, 2008.

30. Liver biopsy is not the “gold standard”
but is the “best available gold standard”

31. Contraindications of liver biopsy
• Uncooperated patients
• Disorders in coagulation profile
• Severe ascites
• Cystic lesions
• Vascular tumors (hemangiomas)
• Amiloidosis
• Congestive liver disease

32. R0C curves for FibroScan, FibroTest, & APRI
for cirrhosis (F0 – F3 vs F4)
Castera L et al. Gastroenterology 2005 ;128 : 343 – 50.
Castera L et al. Lancet 2010 ; 375 : 1419 – 20.

33. Pitfalls of liver stiffness measurement
 Obesity
 Operator experience
 Acute liver injury
 Extrahepatic cholestasis
 Increased CVP
 Ascites
 Narrow intercostal spaces

34. Obesity & operator experience

35. Limitations of liver stiffness measurement
13 369 examinations – 5 year prospective study – 5 operators
BMI > 30 kg/m2 (OR 7.5)
Operator experience (OR 2.5)
Age > 52 years (OR 2.3)
Type 2 diabetes (OR 1.6)
Failure (3%)
BMI > 30 kg/m2 (OR 3.3)
Operator experience (OR 3.1)
Age > 52 years (OR 1.8)
Female sex (OR 1.4)
Hypertension (OR 1.3)
Type 2 diabetes (OR 1.1)
Unreliable results (16%)
Castéra L et al. Hepatology 2010 ; 51 : 828 – 835.
LSM uninterpretable in one of five cases
Main raisons: obesity ( WC) – operator experience

36. Failure rates according to BMI
7261 patients at the time of first examination
Castéra L et al. Hepatology 2010 ; 51 : 828 – 835.

37. Unreliable results according to BMI
6968 patients at the time of first examination
Castéra L et al. Hepatology 2010 ; 51 : 828 – 835.

38. Feasibility of LSM with FibroScan® using XL probe
New probe for obese patients
de Lédinghen V et al. Liver International 2010 ; : 1043 – 1048.
60% not measured by M probe successfully measured by XL probe

39.  Acute liver injury

40. Acute viral hepatitis increases liver stiffness
18 patients with acute viral hepatitis
I Peak increase in aminotransferase
II Aminotransferase ≤ 50% of the peak
III aminotransferase levels ≤ 2 ULN
Arena U et al. Hepatology 2008 ; 47 : 380 – 384.

41. Acute viral hepatitis increases liver stiffness
18 patients with acute viral hepatitis
I Peak increase in aminotransferase
II Aminotransferase ≤ 50% of the peak
III aminotransferase levels ≤ 2 ULN
Arena U et al. Hepatology 2008 ; 47 : 380 – 384.

42.  Extrahepatic cholestasis

43. Obstructive jaundice due to GIST occluding CBD
Millonig G et al. Hepatology 2008 ; 48 : 1718 – 1723.
Stent occlusion
Bilirubin 8 mg/dL
Stiffness 10 kPa
Bilirubin 3.5 mg/dL
Stiffness 5.7 kPa
Stent placement
Bilirubin 2 mg/dL
Stiffness 5 kPa

44. Liver stiffness as a function of bile duct ligation
10 German landrace pigs: 5 controls – 5 BD ligation
Millonig G et al. Hepatology 2008 ;48 : 1718 – 1723.
4.6 kPa
8.8 kPa
6.1 kPa

45.  Increased CVP

46. Representation of clamping site of the IVC
5 German landrace pigs
Millonig G et al. J Hepatol 2010 ; 52 : 206 – 210.
Experiment approved by local committee for Animal Welfare
University of Heidelberg – Germany

47. LSM after clamping & reopening of IVC
5 anesthesized landrace pigs
Millonig G et al. J Hepatol 2010 ; 52 : 206 – 210.
P < 0.001 P < 0.001
Before
clamping
3.1 kPa
5 min after
reopening
5.1 kPa
27.8 kPa
5 min after
clamping

48. Liver stiffness directly influenced by CVP
10 patients with CHF before & after recompensation
Millonig G et al. J Hepatol 2010 ; 52 : 206 – 210.
Median 40.7 Median 17.8
p = 0.004

49.  Ascites

50. Ascites in liver cirrhosis
Ascites grade 1: detectable only by ultrasound

51.  Narrow intercostal spaces

52. Position of FibreScan® probe
Dorsal decubitus position
Right arm in maximal abduction

53. Interpretation of the results of LSM should
always be done by expert clinicians according
to clinical context

54. Transient elastography in clinical practice
Examination quality 10 shots at least
Success rate ≥ 60%
IQR ≤ 25% of median value
Liver disease Not used in acute hepatitis
Not used in acute exacerbation
Not used in ascites & EH cholestasis
Choice of cutoff point Cutoffs different for each CLD
Range of value rather than cutoff
De Lédinghen V et al. Gastroentérol Clin Biol 2008 ; 32 : 58 – 67.

55. Questions

56. Thank You