Abstract
Abstract 1413
Hepatitis virus C (HCV) infection is common in patients with inherited bleeding disorders due to the past use of plasma-derived clotting factor concentrates not treated with virucidal methods. The prognosis of the infection and the outcome of antiviral therapy are related to the stage of liver fibrosis. Since liver biopsy, the gold standard to grade fibrosis, is rarely performed in these patients for cost-benefit reasons, it is important to consider non invasive methods to assess fibrosis such as liver stiffness measurement with transient elastography (TE, Fibroscan®), a technique already validated in non hemophilic patients. We measured TE in 170 patients with inherited bleeding disorders and HCV infection (positive serum HCV-RNA). The main characteristics of these patients are reported in the Table. Steatosis was detected by abdominal ultrasound. Cirrhosis was defined by the presence of irregular liver edge, splenomegaly, dilated portal vein and/or esophageal varices combined either with low platelet count and/or reduced albumin/cholinesterase levels. TE was successfully performed in all but 3 patients, 2 of whom for Body Mass Index (BMI) > 30 kg/m2. Overall, the median value of liver stiffness was 7.2 kPa (interquartile range, IQR: 5.3–11.1) with a median success rate of 100% (IQR: 91–100) and a median IQR value of 1.0 (IQR: 0.7–1.9). HCV genotype or the presence of steatosis did not influence the TE values, whereas higher values were observed in patients with cirrhosis than in those without (median 19.8 kPa, IQR: 14.3–28.1 vs 6.8 kPa, IQR: 5.1–9.1, respectively; p< 0.01). In particular, 18/22 (82%) cirrhotic patients had a liver stiffness value ≥ 12.0 kPa, a cut-off previously identified as associated with severe fibrosis in HCV infected patients. Overall, splenomegaly was present in 51 patients (30%), 16 with cirrhosis and 35 without. In 31/35 (89%) of the latter, TE values were < 12 kPa. Moreover, among patients without cirrhosis, 12 (8%) had TE values ≥ 12 kPa: those patients had ALT and GGT levels significantly higher than patients with TE values < 12 kPa (p<0.05 for both variables). In our cohort TE had a 83% sensitivity, a 95% specificity and a 94% negative predictive value for the detection of severe fibrosis. In the same patients we measured the aspartate aminotransferase-to-platelet ratio index (APRI), a simple non-invasive biochemical marker of liver fibrosis. Median APRI values were significantly higher in patients with cirrhosis than in those without (1.6 vs 0.5, respectively; p<0.01), and a value > 1.5 was observed in 12/22 (55%) patients with cirrhosis. An APRI >1.5 had a 96% specificity and a 93% negative predictive value for the detection of severe fibrosis. Univariate and multivariate linear regression analyses were performed to investigate the relationship between log transformed TE and demographic (age, BMI) or laboratory (ALT, GGT, APRI) variables potentially influencing the TE values. By univariate analysis a linear association was found with age, ALT, GGT, APRI and BMI values (p<0.01 for each). In multivariate analysis APRI, ALT and GGT showed the strongest association with TE, while the statistical significance for BMI and age was marginal. The entire model explained about 50% of the variance of TE (R2= 0.49). Our results confirm that TE is a good tool to assess liver fibrosis also in patients with inherited bleeding disorders and chronic hepatitis C and shows that it can be performed safely in a great proportion of patients with a high success rate. The value of biochemical markers of necroinflammation (such as ALT and GGT) at the time of TE performance may influence the result and should be taken into account in the interpretation of the test.
Disclosures:
No relevant conflicts of interest to declare.
Diagnosis of Significant Liver Fibrosis in Patients With Chronic Hepatitis B By Using a Deep Learning-Based Data Integration Network
