Scholarlink

Objectives <p>To develop and validate a dynamic nomogram for early prediction of complicated appendicitis (CA) in children using a multi-stage feature selection strategy.</p> Methods <p>A retrospective cohort of 398 pediatric patients undergoing laparoscopic appendectomy (January 2022–December 2024) was analyzed and split (7:3) into training and validation sets. Candidate predictors were screened by Boruta, LASSO, and Venn intersection, with final variables identified by multivariable logistic regression. Model performance was assessed by discrimination, calibration, and decision curve analysis.</p> Results <p>Eighty-two cases (20.6%) were classified as CA. C-reactive protein (CRP), neutrophil count (NEU), and respiratory rate (RR) were retained as independent predictors. The nomogram achieved excellent discrimination (AUC 0.937, training; 0.909, validation), good calibration, and favorable clinical utility. A dynamic version is available at: <a href="https://c-xxsrjb.shinyapps.io/dynnomapp/">https://c-xxsrjb.shinyapps.io/dynnomapp/</a>.</p> Conclusions <p>CRP, NEU, and RR are effective predictors of pediatric CA. The proposed nomogram offers a simple, clinically practical tool for risk stratification and early decision-making.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development and validation of a nomogram based on C-reactive protein, neutrophil count, and respiratory rate for predicting complicated appendicitis in children

  • 期刊论文

  • 发表日期：

    2025年12月17日

查看全文
  • Jun Chen,
  • Yanli Guo,
  • Baolian Chen,
  • Zhihong Fu,
  • Mingqing Liu,
  • Xin Qu,
  • Tao Zhou

摘要

Objectives

To develop and validate a dynamic nomogram for early prediction of complicated appendicitis (CA) in children using a multi-stage feature selection strategy.

Methods

A retrospective cohort of 398 pediatric patients undergoing laparoscopic appendectomy (January 2022–December 2024) was analyzed and split (7:3) into training and validation sets. Candidate predictors were screened by Boruta, LASSO, and Venn intersection, with final variables identified by multivariable logistic regression. Model performance was assessed by discrimination, calibration, and decision curve analysis.

Results

Eighty-two cases (20.6%) were classified as CA. C-reactive protein (CRP), neutrophil count (NEU), and respiratory rate (RR) were retained as independent predictors. The nomogram achieved excellent discrimination (AUC 0.937, training; 0.909, validation), good calibration, and favorable clinical utility. A dynamic version is available at: https://c-xxsrjb.shinyapps.io/dynnomapp/.

Conclusions

CRP, NEU, and RR are effective predictors of pediatric CA. The proposed nomogram offers a simple, clinically practical tool for risk stratification and early decision-making.