The Diagnostic Value of Inflammation Related Indicators for Inflammatory Bowel Disease in Children and the Construction of Nomogram Prediction Model
Objective To evaluate the diagnostic value of inflammation related indicators for inflammatory bowel disease(IBD)in children and to construct an IBD nomogram prediction model.Methods Children with inflammatory bowel disease and non-inflammatory bowel disease were selected from the Department of Gastroenterology at Wuxi Children's Hospital from January,2020 to August,2023.We compared the general clinical information of these two groups of children and conducted correlation analysis to examine the potential correlation between various indicators and the occurrence of inflammatory bowel disease in children.Receiver operating characteristic curve(ROC)was drawn to evaluate the diagnostic value of individual and combined indicators.We then screened the final variables through Lasso and stepwise regression,and constructed a nomogram prediction model.Results This study included a total of 387 samples,including 163 cases in the inflammatory bowel disease group and 224 cases in the non-inflammatory bowel disease group.The intergroup comparison results showed statistically significant differences in family history,alkaline phosphatase,neutrophil count,C-reactive protein,erythrocyte sedimentation rate,neutrophil lymphocyte count ratio(NLR),platelet to lymphocyte ratio(PLR),platelet to albumin ratio(PAR),systemic immune inflammation index(SII),lymphocyte count,platelet count,and albumin comparison between the two groups of patients(P＜0.05).The correlation analysis showed that family history(r=0.199,P＜0.001),neutrophil count(r=0.915,P＜0.001),erythrocyte sedimentation rate(r=0.608,P＜0.001),SII(r=0.464,P＜0.001),alkaline phosphatase(r=0.491,P＜0.001),C-reactive protein(r=0.423,P＜0.001),NLR(r=0.480,P＜0.001),PLR(r=0.353,P＜0.001),and PAR(r=0.371,P＜0.001)were positively correlated with the occurrence of inflammatory bowel disease;Lymphocyte count(r=-0.711,P＜0.001),platelet count(r=-0.136,P=0.007),and albumin(r=-0.638,P＜0.001)were negatively correlated with the occurrence of inflammatory bowel disease(all P＜0.05).ROC curves showed that intestinal alkaline phosphatase(AUC=0.741,95％CI 0.689-0.793),C-reactive protein(AUC=0.747,95％CI 0.690-0.804),NLR(AUC=0.781,95％CI 0.734-0.828),PLR(AUC=0.706,95％CI 0.653-0.759),PAR(AUC=0.717,95％CI 0.666-0.768),and SII(AUC=0.756,95％CI 0.706-0.806)had certain diagnostic values,and the combination of all indicators significantly improved the diagnostic efficacy of IBD(AUC=0.706,95％CI 0.653-0.759).AUC=0.964,95％CI 0.946-0.982).A nomogram predictive model for IBD was established using alkaline phosphatase,C-reactive protein,NLR,PAR,SII,family history,erythrocyte sedimentation rate,and albumin variables.The results showed that the C-index of the model was 0.988,indicating a good discriminability.Bootstrap sampling validation showed that the model had a high consistency,and further clinical decision curve analysis showed that the model had significant clinical net benefits.Conclusion Inflammatory related indicators based on intestinal alkaline phosphatase,C-reactive protein,NLR,PLR,PAR,and SII have certain diagnostic values for pediatric inflammatory bowel disease,and the combined diagnostic value of these indicators is even higher.The nomogram prediction model for pediatric inflammatory bowel disease constructed in this study has certain clinical application values,which is beneficial for the early screening and diagnosis of the disease and can provides guidance for timely clinical interventions.
万方数据
维普
