Objective To explore the predictive indexes for the efficacy of infliximab induction therapy in children with Crohn's disease. Methods The clinical data of children diagnosed with Crohn's disease and initially treated with infliximab in the Gastroenterology Department from January 1, 2018 to July 31, 2023 were retrospectively analyzed. The clinical features and laboratory tests before treatment were compared between children with mucosal healing and those without mucosal healing after 14 weeks of infliximab induction therapy. Results Fifty-three children with Crohn's disease (39 boys and 14 girls) were included, and the median age of diagnosis was 12.0 (11.0-13.0) years. The median duration from disease onset to the first infliximab treatment was 5.0(2.0-12.0) months. All the 53 patients had moderate to severe Crohn's disease, including 27 cases with anal fistula, 9 cases with anal fistula combined with intestinal stenosis, 8 cases with intestinal stenosis, 7 cases with extensive intestinal lesions, and 2 cases with growth retardation. Multivariate logistic regression analysis showed that simple endoscopic score for Crohn's disease (SES-CD), C-reactive protein (CRP) and procalcitonin (PCT) might be related to the outcome of mucosal non-healing under endoscopy after infliximab induction therapy (P<0.05). The area under the receiver operating characteristic (ROC) curve (AUC) of baseline SES-CD, CRP and PCT for predicting mucosal non-healing under endoscopy after infliximab induction therapy was 0.73, 0.77 and 0.77, respectively. The sensitivity, specificity and AUC of combined prediction of the three indicators were 0.96, 0.68 and 0.88, respectively. Conclusions The combination of SES-CD, CRP and PCT before infliximab treatment is more accurate in predicting the efficacy of infliximab induction therapy in children with Crohn's disease.