OBJECTIVE: The presence of liver fibrosis in patients with beta-thalassemia major has been demonstrated to be an important negative prognostic factor. Identification of liver fibrosis in early stage would be of great value. Hyaluronic acid (HA), type III pre-collagen (PC III), collagen IV (C IV) and laminin (LN) as serum markers were widely used in the diagnosis of liver fibrosis in patients with chronic viral infections or alcoholic liver diseases. However, their values in thalassemic liver fibrosis have not been studied. This work was to determine the serum HA, PC III, C IV and LN levels in children with beta-thalassemia major and evaluate the diagnostic utility. METHOD: Serum HA, PC III, C IV and LN in 49 hospitalized children with beta-thalassemia major (aged 1 - 15 years with the media age of 6.27 years) and 41 healthy children served as controls (aged 1 - 13 years with media age of 6.40 years) were detected by radioimmunoassay (RIA). Forty-five of 49 cases were performed percutaneous liver biopsies, and the histopathological fibrosis was compared with the four serum markers. The correlation and discriminate analysis were used. RESULTS: All the serum levels of HA, PC III, C IV and LN in beta-thalassemia were significantly higher than those in controls (P < 0.01). In 36 of 45 cases, the histopathology showed liver fibrosis including stage I and stage II by biopsies with a positive rate of 80%. The serum levels of four markers increased successively with the aggravation of liver fibrosis from stage 0 to stage II, and significant correlation was observed between the level of HA or PC III and the stage of fibrosis (HA, r = 0.379, P = 0.017; PC III, r = 0.455, P = 0.04). While there was no difference between the level of C IV or LN and fibrosis (C IV, r = 0.312, P = 0.053; LN, r = 0.310, P = 0.055). Using discriminate analysis, the discriminate function of co-detection of the four markers for the diagnosis of fibrosis was 0.002 HA + 0.003 PC III + 0.002 C IV + 0.006 LN-1.859, which had a sensitivity of 93.88%, specificity of 68.29%, predictive value of positive test and negative test of 77.97% and 90.32%, respectively. Moreover, there was a significant correlation between the serum level of HA or PC III and the liver iron concentration (HA, r = 0.318, P = 0.035; PC III, r = 0.305, P = 0.044). CONCLUSION: The results suggest that, in beta-thalassemia major with chronic liver disease, HA and PC III showed more practical value in diagnosing liver fibrosis than the levels of C IV and LN. The combination of the four serum markers could improve the accuracy and reliability of the diagnosis. A validation study is necessary before introducing into the prediction function during the clinical practice.
OBJECTIVE: The presence of liver fibrosis in patients with beta-thalassemia major has been demonstrated to be an important negative prognostic factor. Identification of liver fibrosis in early stage would be of great value. Hyaluronic acid (HA), type III pre-collagen (PC III), collagen IV (C IV) and laminin (LN) as serum markers were widely used in the diagnosis of liver fibrosis in patients with chronic viral infections or alcoholic liver diseases. However, their values in thalassemic liver fibrosis have not been studied. This work was to determine the serum HA, PC III, C IV and LN levels in children with beta-thalassemia major and evaluate the diagnostic utility. METHOD: Serum HA, PC III, C IV and LN in 49 hospitalized children with beta-thalassemia major (aged 1 - 15 years with the media age of 6.27 years) and 41 healthy children served as controls (aged 1 - 13 years with media age of 6.40 years) were detected by radioimmunoassay (RIA). Forty-five of 49 cases were performed percutaneous liver biopsies, and the histopathological fibrosis was compared with the four serum markers. The correlation and discriminate analysis were used. RESULTS: All the serum levels of HA, PC III, C IV and LN in beta-thalassemia were significantly higher than those in controls (P < 0.01). In 36 of 45 cases, the histopathology showed liver fibrosis including stage I and stage II by biopsies with a positive rate of 80%. The serum levels of four markers increased successively with the aggravation of liver fibrosis from stage 0 to stage II, and significant correlation was observed between the level of HA or PC III and the stage of fibrosis (HA, r = 0.379, P = 0.017; PC III, r = 0.455, P = 0.04). While there was no difference between the level of C IV or LN and fibrosis (C IV, r = 0.312, P = 0.053; LN, r = 0.310, P = 0.055). Using discriminate analysis, the discriminate function of co-detection of the four markers for the diagnosis of fibrosis was 0.002 HA + 0.003 PC III + 0.002 C IV + 0.006 LN-1.859, which had a sensitivity of 93.88%, specificity of 68.29%, predictive value of positive test and negative test of 77.97% and 90.32%, respectively. Moreover, there was a significant correlation between the serum level of HA or PC III and the liver iron concentration (HA, r = 0.318, P = 0.035; PC III, r = 0.305, P = 0.044). CONCLUSION: The results suggest that, in beta-thalassemia major with chronic liver disease, HA and PC III showed more practical value in diagnosing liver fibrosis than the levels of C IV and LN. The combination of the four serum markers could improve the accuracy and reliability of the diagnosis. A validation study is necessary before introducing into the prediction function during the clinical practice.