FECAL CALPROTECTIN AS A NON-INVASIVE MARKER OF ENDOSCOPIC ACTIVITY IN INFLAMMATORY BOWEL DISEASE: A RETROSPECTIVE STUDY

Abstract

Introduction: Inflammatory bowel disease (IBD) is a chronic, relapsing–remitting inflammatory disorder of the gastrointestinal tract. Its monitoring relies mainly on endoscopy and on clinical scores, which are partly subjective. Fecal and blood biomarkers are increasingly used to reduce the need for endoscopy.

Objective: To assess the correlation of fecal calprotectin (FC), C-reactive protein (CRP), hemoglobin and platelet count with clinical and endoscopic activity, and to compare the diagnostic performance of FC and CRP for detecting endoscopic activity.

Methods: We conducted a retrospective study of IBD patients who underwent FC testing between July 2021 and July 2023 at Mohammed VI University Hospital, Oujda, Morocco. A total of 235 assessments from 155 patients were analyzed; the number of assessments varied by analysis according to data availability (137 with an endoscopic score, 132 with a classifiable endoscopic activity). Disease activity was defined using GETAID-recommended scores: Harvey- Bradshaw Index and CDEIS in Crohn’s disease (CD), Mayo score and UCEIS in ulcerative colitis (UC). Correlations were assessed using Spearman’s coefficient (ρ), and diagnostic performance using receiver operating characteristic (ROC) curves. Disease-specific analyses were considered primary and pooled analyses exploratory. Sensitivity analyses accounted for repeated assessments from the same patients and for censored FC values.

Results and discussion: The cohort included 101 patients with CD, 33 with UC and 21 with unclassified or undocumented IBD; 54.8% were women and the mean age was 37.8 years. Median FC was 78 μg/g and median CRP 4.39 mg/L. In CD, FC correlated weakly with the CDEIS (ρ = 0.30, p = 0.003, n = 97) and the HBI (ρ = 0.18, p = 0.022, n = 155), as did CRP (CDEIS: ρ = 0.21, p = 0.039; HBI: ρ = 0.17, p = 0.039); no significant correlation was found in UC. In CD, the AUC for endoscopic activity was 0.723 (95% CI: 0.619–0.827) for FC and 0.659 (95% CI: 0.542–0.776) for CRP. In the exploratory pooled analysis, the AUC was 0.667 (95% CI: 0.567–0.767; p = 0.001) for FC and 0.603 (95% CI: 0.493–0.712; p = 0.07) for CRP; the difference was not significant (p = 0.35). Clinical and endoscopic activity were not correlated (ρ = 0.01, p = 0.89). At ≥ 354 μg/g, FC had a specificity of 90.9% for endoscopic activity. The association of FC with endoscopic activity persisted when restricted to one assessment per patient (CD: ρ = 0.41, AUC 0.782), with patient-level cluster bootstrap and with alternative handling of censored values, whereas its weak correlation with clinical activity did not.

Conclusion: FC was the only marker with a significant AUC for endoscopic activity in the pooled analysis, with a better performance in CD, where CRP also had a significant AUC, and high FC values were highly specific for endoscopic inflammation. However, its correlations with disease activity were weak, its advantage over CRP was not significant, and clinical scores did not reflect endoscopic activity. FC is a useful complementary tool to select patients for endoscopy, but it cannot replace endoscopy.