A simple noninvasive index to predict significant liver fibrosis in patients with advanced schistosomiasis japonica

• HA gave the best AUROC for the diagnosis of significant fibrosis.
• HA and INR were independent predictors for significant fibrosis.
• New index showed the highest AUROC for the diagnosis of significant fibrosis.
• New index had better diagnostic values than APRI and FIB-4.

BackgroundSchistosoma japonicum causes marked liver fibrosis, while lethal syndromes present in advanced schistosomiasis patients. Its management depends on the degree of fibrosis present.Patients and methodsFifty-two patients were recruited to assess the diagnostic value of bio-markers in patients with advanced schistosomiasis japonica. Fibrosis was assessed in liver biopsies using METAVIR system. The correlation between conventional parameters and significant fibrosis (F2-F4) was assessed using univariate analysis and logistic regression. The method of area under receiver operating characteristic curves (AUROCs) was used as a measurement of diagnostic efficacy.ResultsWhite blood cell counts, platelet counts and albumin (all P < 0.05) were significantly lower, while prothrombin time, international normalized ratio (INR), hyaluronic acid (HA), IV collagen and ultrasound fibrosis scores (all P < 0.01) were significantly elevated in F2-F4 patients compared with F0-F1 patients. HA and INR were identified as independent predictors by multivariate analysis (P = 0.023 and P = 0.013, respectively). Of the routine laboratory tests for the diagnosis of significant fibrosis, HA gave the best AUROC of 0.875 (95% confidence interval (CI): 0.701–0.997). We constructed a new simple index (INR × HA/100) to discriminate between F2-F4 patients and F0-F1 patients. It showed the highest AUROC of 0.921 (95% CI: 0.828-1.000), and had better diagnostic values than APRI and FIB-4.ConclusionHA and INR were reliable markers for differentiating significant liver fibrosis in patients with advanced schistosomiasis japonica. And the new simple index can easily predict significant liver fibrosis with a high degree of accuracy.

Figure optionsDownload as PowerPoint slide