The identification of reliable non-invasive biomarkers of Crohn disease (CD) activity is crucial. Fecal calprotectin (FC) is the most promising one. We aimed to define whether its performance for isolated ileal CD could be as useful as for colonic and ileocolonic disease.
Retrospective observational study of all pediatric patients affected with CD with FC and inflammatory blood markers performed within one week from an ileocolonoscopy or magnetic resonance enterography. The sensitivity, specificity, negative predictive value, positive predictive value and accuracy of FC >50 mcg/g were evaluated for isolated ileal CD (L1), colonic CD (L2) and ilecolonic CD (L3) using ilecolonoscopy and magnetic resonance enterography as the reference standard. The best FC cut-off for each disease location was evaluated using receiver operating characteristic curves.
One-hundred seventy-six FC measurements from a total of 98 patients were collected (14.3% L1, 10.2% L2, 75.5% L3). The sensitivity and specificity of FC for L1 CD were 36% and 91%, respectively, compared to 93% and 75% for L2 and 70% and 95% for L3. An FC of 95 mg/kg was identified as the best cut off for identification of active isolated ileal disease, with a sensitivity of 77% and a specificity of 56% [area under the curve (confidence interval [CI] 0.56– 0.97) 0.77, P = 0.02].
With a sensitivity of 36%, FC alone does not accurately reflect isolated L1 disease activity. A significantly lower threshold should be applied for increasing its performance for isolated small bowel CD monitoring.