[Predictive value of neck circumference for cardiometabolic risk in children]
Y N Shi,W Yan,M Y Cao,L T Peng,M Zhao,L Wang,Q Q Liu,X N Li
DOI: https://doi.org/10.3760/cma.j.cn112140-20240108-00029
2024-08-02
Abstract:Objective: To investigate the predictive value of neck circumference on cardiometabolic risk in children. Methods: This was a cross-sectional study of natural sources. As the prediction cohort, clinical data were collected from 3 443 children aged 5-14 years who underwent physical examination in the Department of Child Healthcare, Children's Hospital of Nanjing Medical University from July 2021 to September 2022. As the validation cohort for external validation, clinical data were collected from 604 children aged 5-14 years who underwent physical examination in the Department of Child Healthcare, Children's Hospital of Nanjing Medical University from October 2022 to March 2023. Height, weight, neck circumference, waist circumference and body composition were measured in both groups, and body mass index, neck circumference to height ratio (NHtR), waist circumference to height ratio, body fat percentage and skeletal muscle percentage were calculated. Systolic blood pressure, diastolic blood pressure, fasting blood glucose, blood lipid and uric acid and other cardiovascular and metabolic risk indicators were collected in both groups. The prediction cohort was further stratified into clustered and non-clustered groups based on the clustering of cardiometabolic risk factors (CCRF). Various variables between these 2 groups were compared using the Mann-Whitney U test. Pearson correlation and binary Logistic regression were conducted to investigate the correlations between neck circumference and cardiovascular metabolic risk factors. The accuracy of NHtR in predicting the CCRF was evaluated using the area under the curve (AUC) of receiver operating characteristic (ROC). The cutoff value was determined using the Youden index. The validation cohort was then divided into groups above and below the cutoff value, and the detection rate of CCRF between the 2 groups was compared using the χ2 test for validation. Results: In the prediction cohort of 3 443 children (2 316 boys and 1 127 girls), 1 395 (40.5%) children were overweight or obese, and 1 157 (33.6%) children had CCRF. Pearson correlation analysis revealed all significant positive correlations (all P<0.01) between neck circumference and systolic blood pressure (r=0.47, 0.39), diastolic blood pressure (r=0.27, 0.21), uric acid (r=0.36, 0.30), and triglycerides (r=0.20, 0.20) after adjusting for age in both males and females. Among both males and females, neck circumference both showed significant negative correlation (both P<0.01) with high-density lipoprotein cholesterol (r=-0.27, -0.28), and no correlation with fasting glucose levels (r=0.03, -0.03, both P>0.05). After adjusting for gender, age, and body fat percentage, increased body mass index, neck circumference, or waist circumference increased the risks of hypertension (OR=1.23, 1.39, 1.07, all P<0.001), hyperuricemia (OR=1.16, 1.23, 1.05, all P<0.001), hypertriglyceridemia (OR=1.08, 1.16, 1.02, all P<0.01), low high-density lipoprotein cholesterol (OR=1.10, 1.27, 1.03, all P<0.01), and the CCRF (OR=1.51, 1.73, 1.15, all P<0.01). The areas under the ROC curves of NHtR in predicting CCRF was 0.73, with sensitivity and specificity at 0.66 and 0.71, respectively. The corresponding optimal cut-off value was 0.21. Validation with 604 children confirmed that the detection of CCRF in the NHtR≥0.21 group was 3.29 times (60.5% (112/185) vs. 18.7% (79/422),χ2=107.82, P<0.01) higher compared to the NHtR <0.21 group. Conclusions: Neck circumference is associated with cardiovascular metabolic risks such as hypertension, hyperlipidemia, hyperglycemia, and hyperuricemia in children. When the NHtR is ≥0.21, there is an increased likelihood of CCRF.