Scholarlink

Background <p>The commonly used approaches for GDM screening and diagnosis mainly include fasting plasma glucose testing in early pregnancy and the oral glucose tolerance test at 24-28 weeks of pregnancy [4]. However, these methods still have limitations.</p> Objective <p>We aimed to evaluate whether neck circumference measured in early pregnancy is associated with subsequent gestational diabetes mellitus (GDM) and to assess its discriminative value for early risk stratification.</p> Methods <p>This prospective case-control study enrolled 400 pregnant women, including 200 in the GDM group and 200 in the control group. Physical parameters, including neck circumference, waist circumference, and blood pressure, were measured between 11 and 13 weeks of gestation. Pre-pregnancy body mass index (BMI) and fasting plasma glucose (FPG) were also recorded. Participants underwent an oral glucose tolerance test (OGTT) between 24 and 28 weeks. Baseline comparisons and univariate and multivariate logistic regression analyses were performed to examine the independent association of neck circumference with GDM. Random forest (RF), eXtreme gradient boosting (XGBoost), and support vector machine (SVM) models were then constructed based on significant variables. Area under the curve (AUC), accuracy, precision, recall, and F1 score were calculated to summarize discrimination and threshold-based classification within this balanced dataset.</p> Results <p>Baseline comparisons showed that the GDM group had significantly higher age, pre-pregnancy BMI, waist circumference, neck circumference, SBP, DBP, and early-pregnancy FPG than the control group (all <i>p</i> &lt; 0.05). Univariate analysis indicated that neck circumference was significantly associated with GDM. Multivariate logistic regression analysis showed that neck circumference remained an independent predictor after adjustment for age, pre-pregnancy BMI, waist circumference, and early-pregnancy FPG. All three machine learning models demonstrated good discriminative performance (based on AUC), with the SVM model performing best. The RF and XGBoost models also maintained stable performance.</p> Conclusions <p>Early-pregnancy neck circumference was independently associated with GDM and showed stable discriminative performance across logistic regression and multiple machine learning models, supporting its potential use as an adjunct marker for early risk stratification rather than as a stand-alone screening or diagnostic tool.</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

Association and discriminative value of early-pregnancy neck circumference for gestational diabetes mellitus: A prospective study based on logistic regression and machine learning models

  • 期刊论文

  • 发表日期：

    2026年4月28日

查看全文
  • Feng Wang,
  • Xiaohua Li,
  • Huanliu He,
  • Weidi Guo,
  • Jiali Huang,
  • Dan Li

摘要

Background

The commonly used approaches for GDM screening and diagnosis mainly include fasting plasma glucose testing in early pregnancy and the oral glucose tolerance test at 24-28 weeks of pregnancy [4]. However, these methods still have limitations.

Objective

We aimed to evaluate whether neck circumference measured in early pregnancy is associated with subsequent gestational diabetes mellitus (GDM) and to assess its discriminative value for early risk stratification.

Methods

This prospective case-control study enrolled 400 pregnant women, including 200 in the GDM group and 200 in the control group. Physical parameters, including neck circumference, waist circumference, and blood pressure, were measured between 11 and 13 weeks of gestation. Pre-pregnancy body mass index (BMI) and fasting plasma glucose (FPG) were also recorded. Participants underwent an oral glucose tolerance test (OGTT) between 24 and 28 weeks. Baseline comparisons and univariate and multivariate logistic regression analyses were performed to examine the independent association of neck circumference with GDM. Random forest (RF), eXtreme gradient boosting (XGBoost), and support vector machine (SVM) models were then constructed based on significant variables. Area under the curve (AUC), accuracy, precision, recall, and F1 score were calculated to summarize discrimination and threshold-based classification within this balanced dataset.

Results

Baseline comparisons showed that the GDM group had significantly higher age, pre-pregnancy BMI, waist circumference, neck circumference, SBP, DBP, and early-pregnancy FPG than the control group (all p < 0.05). Univariate analysis indicated that neck circumference was significantly associated with GDM. Multivariate logistic regression analysis showed that neck circumference remained an independent predictor after adjustment for age, pre-pregnancy BMI, waist circumference, and early-pregnancy FPG. All three machine learning models demonstrated good discriminative performance (based on AUC), with the SVM model performing best. The RF and XGBoost models also maintained stable performance.

Conclusions

Early-pregnancy neck circumference was independently associated with GDM and showed stable discriminative performance across logistic regression and multiple machine learning models, supporting its potential use as an adjunct marker for early risk stratification rather than as a stand-alone screening or diagnostic tool.