Comparative study on determination of fecal calprotectin by enzyme-linked immunosorbent assay and fluorescence immunochromatography assay
10.3760/cma.j.cn101480-20250702-00100
- VernacularTitle:酶联免疫吸附分析与荧光免疫层析法测定粪钙卫蛋白的比较研究
- Author:
Sinan XIAO
1
;
Haitao SHI
1
;
Kairuo WANG
1
;
Kairong SU
1
;
Xin LIU
1
Author Information
1. 西安交通大学第二附属医院消化内科，西安 710004
- Publication Type:Journal Article
- Keywords:
Inflammatory bowel disease;
Fecal calprotectin;
Enzyme-linked immunosorbent assay;
Fluorescence immunochromatography assay;
Comparative study;
Cut-off valu
- From:
Chinese Journal of Inflammatory Bowel Diseases
2025;09(5):404-411
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To compare the diagnostic efficacy and consistency of fecal calprotectin (FC) detected by enzyme-linked immunosorbent assay (ELISA) and fluorescence immunochromatography assay (FICA) in assessing the disease activity of inflammatory bowel disease (IBD) .Methods:The paired-design diagnostic test comparison study was conducted. A total of 61 IBD patients from the Second Affiliated Hospital of Xi'an Jiaotong University who underwent simultaneous ELISA and FICA testing from May to June 2025 were prospectively enrolled. Using Best Crohn's disease activity index and modified Mayo score as gold standards, optimal FC cut-offs for assessing the disease activity were determined by receiver operating characteristic (ROC) curve analysis. Numerical consistency was evaluated via Spearman correlation, Passing-Bablok regression, and Bland-Altman analysis. Classification consistency was assessed by Cohen's Kappa coefficient based on both manufacturer-recommended cut-offs (ELISA: 200 μg/g, FICA: 100 μg/g) and ROC-optimized cut-offs.Results:Of the 61 patients, 28 were male and 33 were female, with a median age of 48 (34, 61) years and a disease duration of 48 (12, 109) months; 43 had ulcerative colitis (UC) and 18 had Crohn's disease (CD) ; 35 were in remission and 26 were in the active stage. Median FC concentrations were 178.0 (30.0, 1 342.0) μg/g by ELISA and 67.2 (15.0, 275.6) μg/g by FICA. The area under the curve (AUC) for ELISA in diagnosing activity of IBD was 0.930, with a sensitivity of 80.0% and specificity of 96.2% at the optimal cut-off of 154.0 μg/g. The AUC for FICA was 0.784, with a sensitivity of 80.0% and specificity of 80.8% at the optimal cut-off of 81.2 μg/g. DeLong test showed that the overall diagnostic efficacy of ELISA was significantly superior to that of FICA ( Z = 2.550, P = 0.011). Spearman correlation analysis showed a correlation coefficient of 0.62 (95% CI: 0.41-0.73, P < 0.001) between ELISA and FICA results. The Cusum linearity test indicated a linear relationship between the two methods ( P = 0.291). Passing-Bablok regression yielded the equation y = -53.38 + 5.56x, indicating both significant constant and proportional systematic errors between ELISA and FICA, and the errors increased with the concentrations. Bland-Altman analysis demonstrated ELISA assay values were systematically higher than those of FICA (overall mean bias: 74.5%, 95% limits of agreement: -101.0% to 250.0%), with larger differences in active disease than remission (the mean bias: 100.4% vs. 48.0%). Classification consistency improved markedly when using ROC-optimized cut-offs compared with manufacturer-recommended cut-offs (Kappa: 0.608 vs. 0.474) . Conclusions:Both ELISA and FICA can effectively identify active IBD but exhibit concentration-dependent systematic bias (ELISA > FICA). The consistent use of a single assay is recommended for disease monitoring and ROC-optimized cut-offs are adopted to improve the accuracy of disease activity stratification.
