Search

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.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.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).Compared to conventional models, ANNs performance in predicting CHC cirrhosis is slightly better and more reproducible.

Summary Background Biopsy is the gold standard for assessing cirrhosis in patients with chronic hepatitis C virus infection, but it is expensive and at risk of complications. Alternative non-invasive methods have been developed but their usefulness remains uncertain. Aim To compare the accuracy of five non-invasive scores in detecting cirrhosis. Methods We reviewed the charts and liver biopsies of 228 consecutive, treatment-naive, hepatitis C virus-positive patients, 13.2% of whom with histological diagnosis of cirrhosis. The five alternative scores were age-platelet index, cirrhosis discriminant score, aspartate transaminases to platelet ratio index, Pohl's index, and aspartate transaminases/alanine transaminases ratio. Results The specificities of the scores were good (87–100%), but not so their sensitivities (17–67%). Accordingly positive likelihood ratios were generally good but negative likelihood ratios were suboptimal. Combinations of the scores independently related to cirrhosis only slightly change this diagnostic accuracy. Using double cut-offs to exclude/diagnoses cirrhosis, cirrhosis discriminant score classified 21% of patients without misdiagnoses and aspartate transaminases to platelet ratio index classified 85% of case with 9% of misdiagnoses. Conclusions The five scores showed variable sensitivities and specificities in detecting liver cirrhosis, both individually and in combination. The use of double cut-off points may make the cirrhosis discriminant score and aspartate transaminases to platelet ratio index useful to reduce the number of patients submitted to liver biopsy.