Evaluation of the Agreement between Fibrotest , Fibroscan and by johnrr2

VIEWS: 186 PAGES: 18

									Evaluation of the Agreement between Fibrotest®, Fibroscan® and APRI for the Assessment of Significant or Severe Liver Fibrosis in HIV/HCV Coinfected Patients Participating in the Prospective ANRS C013 - HEPAVIH French Cohort Study
Y. Benhamou, M.A. Loko, M. Winnock, P. Sogni, G. Pialoux, C. Katlama, M.A. Valantin, D. Neau, F. Dabis, D. Salmon

4th IAS Conference , Sydney, Australia, 22-25 July 2007

BACKGROUND
• Several non invasive methods have been developed as an alternative to liver biopsy for the assessment of liver fibrosis. • Among these, FibroTest®, APRI and transient elastography (FibroScan®), although not yet validated, are commonly used to assess liver fibrosis in HIV-HCV co-infected patients.
• Whether these different methods agree for the diagnosis of significant or severe liver fibrosis remains unknown.

AIM
• To assess the agreement between FibroTest ®, APRI and FibroScan ®, taken two-by-two, for the determination of significant (F ≥ 2) or severe (F ≥ 3) liver fibrosis in HIV-HCV coinfected patients.

METHODS
• Cross sectional study (ANRS CO13 HEPAVIH prospective French Cohort ) • Patients : • Positive HCV RNA • No current anti-HCV therapy • Biological samples available within 6 months of Fibroscan performance.

METHODS (2)
• Non-invasive methods : • Transient elastography (FibroScan®, FS)
- measures ‘liver stiffness’ - expressed in Kilopascals (KPa)

• FibroTest® (FT)
- biochemical marker-based algorithm :
a2 macroglobulin, haptoglobin, gGT, bilirubin, apolipoprotein A1

• APRI score*
- based on aspartate transaminases and platelet counts :
[AST(IU/ULN) x 100 / platelets (10^9/L)]
* Wai et al. Hepatology, 2003;38:518-526

METHODS (3)
• Thresholds for scoring liver fibrosis : Significant
( F ≥ 2)

Severe
(F ≥ 3)

FibroScan (KPa) FibroTest Index APRI

> 7.1 > 0.48 > 1.5

> 9.5 > 0.58 -----

METHODS (4)
• Agreement, measured by the kappa coefficient, was considered as :
poor : fair : moderate : good : very good : 0 – 0.20 0.21 – 0.40 0.41 – 0.60 0.61 – 0.80 0.81 – 1.00

RESULTS
• Characteristics of patients (N=154) *
Age (yrs) Gender (% male) CD4 counts (/mm3)
Current antiretroviral treatment (%)

44 (26-64) 80 % 440 (47-1877)
91.6%

HCV genotype 1 2 or 3 other
* Median (range), N (%)

55.3% 23.4% 21.3%

Prevalence of significant (F ≥ 2) vs minimal liver fibrosis depending on the test used
(%) Fibrotest (FT) Fibroscan (FS)

APRI

Agreement for the diagnosis of significant liver fibrosis ( F ≥ 2)
FibroScan (N,%) APRI FibroTest Kappa CI
(95 %)

99 (64.3)
106 (68.8)
FibroTest

0.23
0.39
Kappa

0.10 – 0.36
0.26 – 0.53
CI

APRI

78 (51.3)

0.16

0.07 – 0.25

Prevalence of severe (F ≥ 3) liver fibrosis depending on the test used
100 80 60 40
25

(%) (%)

FT
75

FS

Fibrotest (FT)
50 50

Fibroscan (FS)

20 0 F0-F2 ≥F3

Agreement for the diagnosis of severe liver fibrosis (F ≥ 3)
FibroScan (N,%) Kappa CI
(95 %)

FibroTest

105 (68.2)

0.36

0.24 – 0.49

Disagreement between Fibrotest and Fibroscan
(> 1 point of fibrosis)
• Significant fibrosis • Severe fibrosis 39 patients 25.3% 47 patients 31%

Causes of disagreement between Fibrotest and Fibroscan for severe fibrosis
47 patients
Fibrotest > FibroScan
N = 42/47 (89%)
- ATAZANAVIR :  bilirubin (N=19) - inflammatory syndrome (N=2) - hemolysis (N=3) - acute hepatitis (N=1) - incomplete information (N=15)

FibroScan > Fibrotest
N = 5/47 (11%)
- incomplete information (N=5)

Agreement for diagnosis of significant liver fibrosis (F >2)
Patients treated by Atazanavir NOT included (N=95)

FibroScan
(N,%)

Kappa

CI
(95 %)

APRI FibroTest

62 (65) 70 (74)
FibroTest
(N,%)

0.27 0.48
Kappa

0.11 – 0.43 0.30 – 0.65
CI
(95 %)

APRI

57 (60)

0.24

0.10 – 0.37

Agreement for diagnosis of severe liver fibrosis (F > 3)
Patients treated by Atazanavir NOT included (N=95)

FibroScan
(N,%)

Kappa

CI
(95 %)

FibroTest

76 (80)

0.57

0.41 – 0.73

CONCLUSION (1)
• In HIV-HCV co-infected patients, poor to fair agreement between APRI, FibroTest and FibroScan for grading significant or severe fibrosis.
APRI seems to underestimate significant fibrosis as compared to other tests

•

•

Between FibroTest and FibroScan, disagreement of more than one point:  25% for significant fibrosis (F > 2)  31% for severe fibrosis (F > 3)  fibrosis stage higher by FibroTest than by FibroScan in 89%

CONCLUSION (2)
• Agreement improved when patients receiving atazanavir were not considered.
• At least, in such patients, liver fibrosis is better evaluated by Fibroscan.

• A limitation of our study is that these tests have not yet been compared with the gold standard assessment obtained by a liver biopsy.
• Such a study is on going in order to optimize the use of non-invasive methods of fibrosis evaluation that are becoming important treatment-decision making tools.


								
To top