P250. Correlation of faecal calprotectin levels with Crohn's disease activity scores using two different immunoassays
M. Oyaert1, D. De Smet1, F. Baert2, F. D'Heygere3, N. Callewaert4, M. Langlois5, H. Vanpoucke1, 1AZ Delta Roeselare Menen, Clinical Laboratory, Roeselare, Belgium, 2AZ Delta Roeselare Menen, Gastroenterology, Roeselare, Belgium, 3AZ Groeninge, Gastroenterology, Kortrijk, Belgium, 4AZ Groeninge, Clinical Laboratory, Kortijk, Belgium, 5AZ St. Jan Brugge Oostende, Laboratory Medicine, Brugge, Belgium
Faecal calprotectin (FC) is a useful marker for detection and follow-up of Inflammatory Bowel Disease. We evaluated two methods for FC measurement and investigated whether FC values correlated with Crohn's disease Activity Index (CDAI) and Harvey–Bradshaw Index (HBI) scores.
Follow-up samples (n = 150) for FC measurement were obtained from consecutive patients (n = 60, median age 39 years [range 18–65 years], 23 males, 37 females) diagnosed with Crohn's disease and included in two prospective clinical trials. CDAI (n = 80) or HBI (n = 70) scores were calculated. FC measurements were performed using an established immunochromatographic Bühlmann POCT method and with a recently introduced automated method (EliA ELISA on ImmunoCAP 250). Spearman's rank correlation was used to investigate the relationship between FC levels and disease activity scores. To assess diagnostic performance, patients were dichotomized according to disease activity: moderate disease or remission (CDAI < 220, HBI < 7) and active disease (CDAI ≥ 220, HBI ≥ 7). ROC analysis was performed and likelihood ratios (LRs) for rationally selected result intervals were calculated.
POCT FC levels correlated significantly with CDAI (rs = 0.81 [95% CI: 0.71 to 0.87]; p < 0.0001) and HBI (rs = 0.74 [95% CI: 0.71 to 0.87]; p < 0.0001). With the automated EliA method similar and significant correlations between FC levels and CDAI (rs = 0.74 [95% CI: 0.62 to 0.83]; p < 0.0001) and HBI (rs = 0.77 [95% CI: 0.65 to 0.85]; p < 0.0001) were obtained.
Prevalence of active Crohn's disease was 30.7% (46/150). ROC analysis yielded excellent and comparable AUCs of 0.98 for POCT versus 0.98 for EliA (p = 0.581). Selection of result intervals and calculation of associated LRs showed that POCT (<168 µg/g) and EliA (<84 µg/g) FC concentrations ruled out active disease in respectively 82.7% (86/104; LR = 0.00 [95% CI 0.00 to 0.20]) and 77.9% (81/104; LR = 0.00 [95% CI 0.00 to 0.22]) of patients with moderate disease or remission (CDAI < 220, HBI < 7). On the contrary, a strong suggestion of active disease was made in 32 (69.6%) and 31 (67.4%) of 46 patients for POCT (>627 µg/g; LR = 72.4 [95% CI 10.19 to 513.57]) and EliA (>383 µg/g; LR = +∞ [95% CI 8.76 to +∞]), respectively.
In conclusion, we found good correlations between FC measurements and disease activity scores CDAI and HBI for both the Bühlmann POCT method and the automated EliA ELISA (Immunocap250). The automated EliA immunoassay is a good alternative to the Bühlmann POCT as a screening assay in patients to detect active Crohn's disease. Use of likelihood ratios for different FC result intervals improves clinical interpretation and enables differentiation between patients with active disease and patients with moderate disease or remission.