[Is APRI score a suitable tool for prediction of fibrosis in Tunisian patients with genotype 1 chronic viral hepatitis C?]. / Le score April est il fiable pour la prediction de la fibrose chez les patients Tunisiens atteints d'hepatite chronique virale C de genotype 1?

ABSTRACT

BACKGROUND: Therapeutic indications in chronic genotype 1 hepatitis C are based on severity of fibrosis. APRI score is a simple, cheap and reproducible biochemical test. Performances of APRI score in Tunisian population with chronic hepatitis C were not previously prospectively studied. AIM: To evaluate the performances of APRI score in prediction of severity of fibrosis in chronic genotype 1 hepatitis C based on a prospective study. METHODS: We prospectively include patients with chronic genotype 1 hepatitis C and positive viral load. Hepatic biopsy was performed in all included patients and abnormalities were classified according to METAVIR classification. In all patients, APRI score was calculated based on biochemical data collected within the 15 days before hepatic biopsy. RESULTS: We studied 140 patients (46 men, mean age 48.4 years (20 - 65 years)). Mean APRI score was 0.89 (0.18 - 3.72). Statistically significant correlation was observed between APRI score and fibrosis severity (r = 0.31 p < 0.0001). APRI score was higher in patients with severe fibrosis (F2, F3 or F4) compared to patients with moderate fibrosis (F0 or F1) (0.97 + 0.68 vs 0.62 + 0.44; p 0.009). Threshold value of APRI score of 0.72 was associated with area under the curve of 0.65 + 0.05 (0.57 - 0.73), sensitivity of 56.3% and specificity of 75.8% in prediction of severe fibrosis. APRI score was also higher in patients with cirrhosis (1.24 + 0.79 vs 0.85 + 0.61; p = 0.01). Threshold value of 0.86 was associated with area under the curve of 0.69 + 0.07 (0.61 - 0.77), sensitivity of 76.4% and specificity of 65.8% in prediction of cirrhosis. CONCLUSION: APRI score is not a good alternative to hepatic biopsy although a strong correlation with fibrosis severity, because of relatively low area under the curve, sensitivity and specificity in prediction of severe fibrosis and cirrhosis.