Role of Hepcidin to Identify the Type of Anemia in Chronic Kidney Disease in Children
Chronic kidney disease (CKD) may present with anemia of chronic disease (ACD), iron-deficiency anemia, or both (mixed anemia). Common hematologic parameters may not distinguish type of anemia in CKD. Hepcidin is a new variable considered to guide management of anemia in CKD. This study aimed to determine type of anemia in children with CKD, and determine the level of hepcidin in those patients and its relationship with degree of CKD, hemoglobin, and ferritin. This was a cross sectional study in 2-18 years non-dialyzed children with CKD.
Subjects were divided into group I (CKD stage 1-2) and group II (CKD stage 3-5). Each group consisted of 29 subjects. Anemia occurred in 34 of 58 subjects, 24 were ACD and 10 were mixed anemia. Median of hepcidin levels in group II were significantly higher than group I (33.4 vs 12.5 ng/mL). Hepcidin has positive correlation with ferritin. ROC analysis showed that hepcidin level of >18 ng/mL may predict ACD. Ferritin level of >99.7 ng/dL can predict hepcidin >18ng/mL (sensitivity 74.2% and specificity 70.4%). This study concluded that ACD is the most type of anemia in CKD besides mixed anemia.
Venous blood (6 mL) was collected for laboratory analysis in Prodia Laboratory, Kramat Jakarta Pusat using Quantikine IVD Human sTFR immunoassay kit and DRG hepcidin-25 (bioactive) ELISA kit. Subjects were grouped into two groups, Group I (CKD stage I-II or GFR >60mL/min/1,73m2 ) and Group II (CKD stage 3-5 or GFR 100 ng/dL, sTFR 2,7 mg/L and/or sTFR/log ferritin index >1,5). Data were analyzed using SPSS v20.
The normality test was conducted for all numerical data with Kolmogorov-Smirnov test for sample size more than 50, or Shapiro-Wilk test for sample size less than 50. Normally distributed data is presented as mean ± standard deviation (SD), while skewedly distributed data is presented as median (range). The bivariate analysis for numerical data was analyzed using independent T-test, or Mann-Whitney test for skewed distribution.
Categorical data was analyzed with chi-square test. Correlation analysis was performed using Pearson test, or Spearman test for skewed data. The sensitivity and specificity and cut-off point were determined with ROC curve and AUC. A p value < 0,05 is considered to be statistically significant.