Early predictors of induction of remission with exclusive enteral nutrition in pediatric Crohn's disease

Background Exclusive enteral nutrition (EEN) is recommended as the first line therapy for mild to moderate pediatric Crohn's disease (CD) with inducing clinical remission and promoting mucosal healing. However, early factors that could predict the efficacy of EEN in inducing remission of CD is lack. Methods A total of 56 patients in the derivation cohort and 19 patients in the validation cohort were included in this study. Baseline clinical covariates were compared between the clinical remission group and the no clinical remission group and between the mucosal healing group and the no mucosal healing group. More than 100 laboratory covariates before EEN were analyzed with the Pediatric Crohn’s Disease Activity Index(PCDAI) and the Crohn’s Disease Endoscopic Index of Severity (CDEIS)at the end of the 8-week EEN. Predictive models were developed by forward stepwise (binary likelihood) logistic regression, and the probabilities of clinical remission and mucosal healing were calculated for selected baseline levels of covariates. Results The clinical remission diagnostic model performed well inboth the training cohort (AUC 0.93; p<0.05) and the validation cohort (AUC 0.72; p=0.19). Indirect bilirubin (IBIL)>4.95 μmol/L, CD3>76.78% and iron>9.025 mmol/L at baseline were associated with a reduced clinical remission rate. The mucosal healing diagnostic model also presented the optimal discrimination in both the derivation cohort (AUC 0.87; p<0.05) and the validation cohort (AUC 0.83; p=0.07). Factors determining a lower mucosal healing rate were human IL-10>4.35 μmol/L and RDW>14.55%. Conclusion IBIL, CD3 and iron are reliable predictors of the induction of clinical remission with EEN. Human IL-10 and RDW are early predictors of the induction of mucosal healing with EEN.