Work

7714 Two different nomogram models for the risk of insulin resistance in overweight and obese children: A representative survey in western China

Year: 2024

Type: article

Abstract: Disclosure: L. Ouyang: None. L. Feng: None. G. Jie: None. F. Yang: None. P. Li: None. Background: Insulin resistance (IR) is the co-pathogenic basis of metabolic abnormalities and greatly influences the treatment and prognosis in children and adolescents with overweight and obesity. Therefore, early detection of the groups with high risk of IR is essential. At present, there is no effective IR warning model that can be used in the primary visit of overweight and obese children. Methods: 181 children and adolescents with overweight and obese were enrolled in this study. The least absolute shrinkage and selection operator and multiple logistic regression were used to screen the anthropometric and hematologic indicators, and the nomogram models for the risk of IR were developed. The model discrimination was verified in terms of the area under the curve (AUC) of the receiver operating characteristic (ROC) and the Akaike information criterion (AIC). The calibration of the nomogram model was validated in terms of the calibration curve. The internal validation of the nomogram model was performed by the Bootstrap method with 500 replicate samples, and the consistency index (C-Index) was calculated using Harrell’s C-Index to verify the model prediction performance. Results: Two nomogram models were successfully built, model 1 enrolled four anthropometric indicators including waist-to-height ratio (WHtR), body mass index (BMI), gender, and height, and model 2 enrolled the above four anthropometric indicators and three hematologic indicators including the levels of serum 25-hydroxyvitamin D (25OHD), white blood cells (WBC) and red blood cell distribution width (RDW). The females, higher values of the WHtR, BMI and height were the IR-related risk factors for overweight and obesity, and higher levels of WBC and RDW were also the IR-related risk factors. However, 25OHD is a protective factor, when the level of 25OHD is greater than 20ng/ml, the less the risk of IR. The AUC of nomogram model 1 was 0.870 and 0.861 of model 2. Internal validation results show a C-Index of 0.86125 in model 1 and 0.8438 in model 2, the mean absolute error was 0.013 and 0.028 separately. The results showed the Akaike information criterion of model 1 was 170.79, and model 2 was 155.51. Conclusions: In summary, both the two models can early warning IR in children with overweight and obese at the first visit. Model 1 can be used as the rapid primary screening tool at the first visit, in particular, it can provide an important primary screening tool for areas with backward medical conditions or scarce medical resources. Model 2 is more accurate, it provides an important reference for further specific examination of glucose metabolism and diagnosis of abnormal glucose metabolism. The two nomogram models are of great significance in avoiding the occurrence of chronic diseases such as diabetes in overweight and obese children due to delayed diagnosis of IR. Presentation: 6/1/2024

Source: Journal of the Endocrine Society

DOI: 10.1210/jendso/bvae163.1527

Authors: L Ouyang, L Feng, G Jie, F Yang, P Li

Institution: West China Second University Hospital of Sichuan University

Language: English

Topic: Diabetes, Cardiovascular Risks, and Lipoproteins

Subfield: Endocrinology, Diabetes and Metabolism

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $3,418