The association between white blood cells (WBCs) and metabolic syndrome (MS) has been consistently reported in previous studies using regional samples. However, it remains unclear whether this relationship has urban–rural differences and is independent of insulin resistance using a large-scale representative sample. Additionally, accurate risk prediction in patients with MS is crucial for developing targeted interventions to enhance the quality of life and prognosis of patients.
The aims of this study were (1) to examine the cross-sectional association between WBCs and MS among the national population and analyze the urban–rural difference and whether insulin resistance plays a moderator effect in this association and (2) to describe the performance to predict MS using machine learning (ML) models.
A cross-sectional study was performed using 7,014 data from the China Health and Nutrition Survey (CHNS).
WBCs were analyzed using an automatic hematology analyzer and MS was defined according to the criteria of the American Heart Association scientific statements of 2009. Variables on sociodemographic characteristics (sex, age, and residence), clinical laboratory (BMI and HOMA-IR), and lifestyle characteristics (smoking and drinking status) were used to construct ML models to predict MS [logistic regression (LR) and multilayer perceptron (MLP) neural network].
We found that 21.1% of participants (1,479/7,014) were classified as having MS. In multivariate logistic regression (including insulin resistance), the result revealed a significant positive association between WBCs and MS. The odds ratios (95% CI) for MS with increasing WBC level were 1.00 (reference), 1.65 (1.18, 2.31), and 2.18 (1.36, 3.50) (
With the aim of confirming the association between WBCs and MS, this cross-sectional study is the first to show that maintaining normal WBC count levels is helpful to prevent the development of MS, and this association is independent of insulin resistance. The results also showed that the MPL algorithm represented a more prominent predictive performance to predict MS.