Development and validation of a nomogram based on clinical factors and standard laboratory tests for prediction of clinically significant liver fibrosis in chronic hepatitis C virus infection.

Objectives Staging liver fibrosis in chronic viral hepatitisC (HCV) patients is essential for prompting surveillanceand treatment. The aim of this study was to develop anomogram, on the basis of simple clinical and laboratoryvariables, to predict three clinically significant stages offibrosis (nil–mild, moderate, advanced/cirrhosis), usinghistology as reference, and to compare its performancewith that of FibroTest, a widely used noninvasive fibrosisscore.Materials and methods Nomograms are graphicalrepresentations of a mathematical formula, used aspredictive tools. The study retrospectively recruited 406HCV patients undergoing liver biopsy. Nomogram wasdeveloped in a training set of 252 patients and testedin a validation set of 154 patients. Histology was stagedaccording to the Metavir system. Fibrosis stages weresubgrouped as follows: advanced fibrosis/cirrhosis(F3/F4, 24%), nil–mild (F0/F1, 36%), and moderate (F2,40%). Age at biopsy, aspartate aminotransferase,c-glutamyl transpeptidase, albumin, platelet count, andprothrombin activity formed the basis for the so-calledFibro-Nomogram, which, in one graphical representation,estimates probability for different stages of fibrosis.Results Areas under the receiver-operating characteristiccurves for advanced fibrosis/cirrhosis were similar fortraining (0.86) and validation sets (0.87). For nil–mildfibrosis, area under the receiver-operating characteristicswere 0.81 and 0.79. Compared with FibroTest,Fibro-Nomogram performed slightly better at predictingsevere fibrosis (F3/F4) with positive likelihood ratio (LR+)5.07 (95% confidence interval 3.08–8.37) versus LR+ 3.82(95% confidence interval 2.56–5.71) for FibroTest. Fornil–mild fibrosis, the two tests showed limited butcomparable performances.Conclusion In HCV patients, Fibro-Nomogram, aninexpensive and readily available predictive tool, couldenable clinicians to interpret patients’ profile, concurrentlystratifying patients into three clinically relevant probabilitycategories with good overall performance. Eur JGastroenterol Hepatol 00:000–000