1053
Improving the Differentiation of Mild from Significant Liver Fibrosis in HIV/HCV-co-infected Patients Using Transient Elastometry
Jose del Valle*1, J Macias1, P Barreiro2, A Rivero3, A Gutierrez4, M Rios-Villegas5, D Merino6, M Gonzalez-Serrano7, E Vispo2, and J Pineda8
1Hosp Univ de Valme, Seville, Spain; 2Hosp Carlos III, Madrid, Spain; 3Hosp Univ Reina Sofia, Cordoba, Spain; 4Hosp Univ Virgen del Rocio, Seville, Spain; 5Hosp Univ Virgen de la Macarena, Seville, Spain; 6Hosp Juan Ramon Jimenez, Huelva, Spain; 7Hosp Univ Virgen de la Victoria, Malaga, Spain; and 8Hosp Univ de Valme, Seville, Spain
Background: Transient elastometry is accurate for
detecting cirrhosis (F=4) in HIV/hepatitis C virus (HCV) -co-infected patients.
However, this procedure is less precise to differentiate mild (F ≤1) from
significant (F ≥2) fibrosis. Thus, using the proposed cut-off value of
7.2 kPa, 24% HIV/HCV-co-infected patients are misclassified as having F ≤1
while they show F ≥2 in the liver biopsy, what may leave untreated
patients with clear histological indication. The aim of this study was to
elaborate and validate cut-off values of liver stiffness to better discriminate
F ≤1 from F ≥2 in HIV/HCV-co-infected subjects.
Methods: Liver stiffness measurements were performed
in 197 co-infected patients who had undergone a liver biopsy within 12 months
of transient elastometry. Patients had received no prior therapy against HCV
infection.
Results: The area under the receiver
operating-characteristic curve (95% confidence interval) was 0.86 (0.78 to 0.93)
for F ≥2. To diagnose F ≥2, a cut-off value of 9.0 kPa showed a positive
predictive value of 87%. Applying this cut-off in the validation set, 5 (12.5%)
patients were misclassified. To discard F ≥2, a cut-off value of 6.0 kPa
showed a negative predictive value of 90%. Using this cut-off in the validation
group, 5 (15%) patients were misclassified. Considering all the patients, 61
(31%) yielded liver stiffness values ≤6.0 kPa and 81 (41%) showed liver
stiffness values ≥9.0 kPa; 55 (28%) patients yielded intermediate values;
no patient showing liver stiffness value ≤6.0 kPa presented F ≥3 in
the liver biopsy. Likewise no patient with a liver stiffness value ≥9.0
kPa had F=0.
Conclusions: We found that the diagnostic accuracy
of transient elastometry to discriminate F ≤1 from F ≥2 was
enhanced using a combination of two different cut-off values, liver stiffness
≤6.0 kPa was diagnostic of F ≤1 and LS ≥9.0 kPa detected
reliably patients with F ≥2, compared with previously proposed single
cut-off. Using these new criteria most HIV/HCV-co-infected patients can be
classified and, consequently, spared of liver biopsy. Only patients with
intermediate values would need invasive procedures.
|
