Prediction of asymptomatic cirrhosis in chronic hepatitis C patients: accuracy of artificial neural networks compared with logistic regression models.

OBJECTIVE: Models based on logistic regression analysis are proposed as noninvasive tools to predict cirrhosis in chronic hepatitis C (CHC) patients. However, none showed to be sufficiently accurate to replace liver biopsy. Artificial neural networks (ANNs), providing a prediction based on nonlinear algorithms, can improve the diagnosis of cirrhosis, a syndrome characterized by complex, nonlinear biological alterations. We compared ANNs with two logistic regression analysis-based models in predicting CHC histologically proven cirrhosis. METHODS: Liver biopsy was obtained in CHC patients of two different cohorts (an internal cohort including 244 patients and an external cohort including 220 patients). One hundred and forty-four patients from the internal cohort served as a training set to construct ANNs and a logistic regression model (LOGIT). These two models and the aspartate aminotransferase-to-platelet ratio index (APRI) were tested in the remaining 100 patients (internal validation set) and in the external cohort (external validation set). Diagnostic performances were evaluated by standard indices of accuracy. RESULTS: In the internal validation set, ANNs, LOGIT, and APRI showed similar discrimination powers (0.88, 0.87, and 0.87 respectively). However, ANNs showed the best positive predictive value (0.86 vs. 0.67 and 0.56) and positive likelihood ratio (40.2 vs. 13.4 and 8.4). In the external validation set, the discrimination power of ANNs (0.76) was significantly higher than those of LOGIT (0.67) and APRI (0.67). CONCLUSION: Compared to conventional models, ANNs performance in predicting CHC cirrhosis is slightly better and more reproducible.

Alpha 1-antitrypsin mutations in NAFLD: high prevalence and association with altered iron metabolism but not with liver damage.

Hyperferritinemia, a common feature of nonalcoholic fatty liver disease (NAFLD), has been associated with steatohepatitis and fibrosis. Heterozygosity for alpha 1-antitrypsin (AAT) mutations is a cofactor of liver damage, and AAT influences inflammation and iron metabolism. This study evaluated the prevalence of the common AAT PiS/PiZ mutants in 353 patients with NAFLD, 195 of whom had hyperferritinemia, versus 114 matched controls and their influence on iron metabolism and the severity of liver damage in the 212 patients submitted to biopsy. PiS and PiZ alleles were searched for by restriction analysis. Thirty-eight patients (10.8%) carried non-MM genotypes versus 4/114 (3.5%) controls (P = .02). Patients carrying AAT mutations had higher ferritin (573 [454-966] vs. 348 [201-648]; P = .001) with similar transferrin saturation. The difference was more evident in males (P < .0001) and significant in patients not carrying HFE genotypes associated with iron overload (P = .015). The prevalence of non-MM genotypes was higher in patients with hyperferritinemia than in those without (28/195, 14% vs. 10/158, 6%, P = .016), and AAT mutations were associated with higher prevalence of sinusoidal siderosis (17/27, 63% vs. 70/180, 39%; P = .02), and sinusoidal/total iron score (46.3 +/- 38% vs. 25.1 +/- 35%, P = .01). Although ferritin was independently associated with fibrosis (P = .047), AAT mutations favoring sinusoidal iron deposition did not affect liver damage. In conclusion, AAT mutations are associated with hyperferritinemia and sinusoidal iron accumulation, but not with more severe liver damage in NAFLD.

Long-term outcome (35 years) of hepatitis C after acquisition of infection through mini transfusions of blood given at birth.

Long-term follow up studies of hepatitis C virus (HCV) infection rarely exceed 20-25 yr. We studied the outcome of HCV infection in 35-yr-old adults infected at birth (1968) through mini transfusions of blood. A retrospective-prospective study was carried out. The cohort included 31 individuals who were given mini blood transfusions (21-30 ml) collected from a donor subsequently revealed to be HCV infected. At enrollment (1998), 18 of 31 (58.1%) recipients had anti-HCV antibody and 16 (88.9%) of them were HCV-RNA positive. All viremic recipients and the infectious donor had the same genotype 1b. Sequence analysis of E1/E2 and NS5b regions, coupled with phylogenetic analysis, indicated that HCV isolates from donor/recipients were linked. Eleven of the 16 viremic recipients gave consent to liver biopsy. Nine had no fibrosis or mild portal fibrosis and 2 had either discrete (Ishak's staging 3) or marked (Ishak's staging 4) fibrosis. During the prospective follow-up period (1998-2003), 2 patients were given therapy, one of whom achieved sustained clinical and virologic response. A second biopsy, performed in 5 patients at a 5 yr interval, revealed no substantial modifications in 4 cases and progression from absence of fibrosis to mild portal fibrosis in the fifth. In conclusion, taking into account the limited study sample, these findings suggest that HCV infection acquired early in life shows a slow progression and mild outcome during the first 35 yr of infection.