OBJECTIVE: To evaluate the utility of two biochemical tests (APRI and FIB-4) to predict liver fibrosis (LF) in prison inmates with chronic hepatitis C. METHOD: We performed a cross-sectional study in 165 inmates with chronic hepatitis C and liver biopsy from two Spanish prisons. LF was staged according to the Metavir Index and was subsequently reclassified as mild or absent (LF<or=1), significant (LF>or=2) or serious (LF>or=3). APRI and FIB-4 were calculated in all patients. The predictive value was calculated by the area under the curve and the optimal cut-off was obtained based on the best specificity value for each LF stage. Then, we analyzed the sensitivity, positive predictive value (PPV) and negative predictive value (NPV) for each cut-off point. RESULTS: LF<or=1 was found in 116 inmates (80.3%), significant LF in 49 (29.3%) and serious LF in 24 (14.5%). An APRI value >or=0.55 or FIB-4 >or=1.0 showed a PPV of 91% and 92%, respectively, for the presence of any grade of LF (>or=1), but both tests had low diagnostic sensitivity: 61.8% and 61.1%, respectively. An APRI value >or=0.86 or FIB-4 >or=1.3 showed a high NPV for serious LF (92.5% and 88.4% respectively). CONCLUSIONS: Both tests have a high predictive capacity to detect the presence of LF in inmates with chronic hepatitis C, but their predictive value in detecting intermediate stages of LF is low. Moreover, a significant number of inmates with LF are not identified.
OBJECTIVE: To evaluate the utility of two biochemical tests (APRI and FIB-4) to predict liver fibrosis (LF) in prison inmates with chronic hepatitis C. METHOD: We performed a cross-sectional study in 165 inmates with chronic hepatitis C and liver biopsy from two Spanish prisons. LF was staged according to the Metavir Index and was subsequently reclassified as mild or absent (LF<or=1), significant (LF>or=2) or serious (LF>or=3). APRI and FIB-4 were calculated in all patients. The predictive value was calculated by the area under the curve and the optimal cut-off was obtained based on the best specificity value for each LF stage. Then, we analyzed the sensitivity, positive predictive value (PPV) and negative predictive value (NPV) for each cut-off point. RESULTS: LF<or=1 was found in 116 inmates (80.3%), significant LF in 49 (29.3%) and serious LF in 24 (14.5%). An APRI value >or=0.55 or FIB-4 >or=1.0 showed a PPV of 91% and 92%, respectively, for the presence of any grade of LF (>or=1), but both tests had low diagnostic sensitivity: 61.8% and 61.1%, respectively. An APRI value >or=0.86 or FIB-4 >or=1.3 showed a high NPV for serious LF (92.5% and 88.4% respectively). CONCLUSIONS: Both tests have a high predictive capacity to detect the presence of LF in inmates with chronic hepatitis C, but their predictive value in detecting intermediate stages of LF is low. Moreover, a significant number of inmates with LF are not identified.