The role of peripheral white blood cell counts in the association between central adiposity and glycemic status

The study addresses how central adiposity, commonly measured by waist-to-hip ratio (WHR), relates to glycemic deterioration and whether inflammatory processes, captured by peripheral white blood cell (WBC) counts, mediate this association. Diabetes prevalence is rising in China and globally, and abdominal obesity is more strongly linked to type 2 diabetes mellitus (T2DM) risk than general obesity. Central adiposity is thought to induce systemic low-grade inflammation, and WBC counts (neutrophils, lymphocytes, monocytes, eosinophils, basophils) are markers of such inflammation. Prior evidence links elevated total WBC to glucose metabolism disorders. The purpose of this cross-sectional analysis was to quantify associations between WHR, WBC counts, and glycemic status (normoglycemia, prediabetes, diabetes), and to estimate the extent to which WBC counts mediate the relationship between central adiposity and diabetes.

Prior large cohort and Mendelian randomization studies show abdominal obesity (WHR) has a stronger association with T2DM than BMI, with each 0.1 unit increase in WHR markedly elevating diabetes risk. Obesity induces low-grade inflammation via adipokines and leukocyte infiltration; WBC counts correlate with insulin resistance and predict incident diabetes. Studies have reported higher total WBC and specific subtypes (neutrophils, lymphocytes, monocytes, etc.) in individuals at elevated diabetes risk. However, few studies have specifically examined circulating WBC subtypes as mediators of the obesity–diabetes relationship, motivating the current mediation analysis.

Design and setting: Cross-sectional analysis using baseline data (July 2020–June 2021) from the Fuqing cohort study in Gaoshan town, Fuqing City, Fujian province, China. The Fuqing cohort is a prospective multi-purpose program investigating chronic diseases. Participants: Inclusion criteria were age 35–75 years and residency in Fuqing City. Exclusions: history of cancer or type 1 diabetes mellitus, and hypoglycemia at baseline. Of 6,790 initially included, 177 with missing waist circumference, blood parameters, or glycemic indicators were excluded, leaving 6,613 participants. Ethics approval was obtained (Fujian Medical University, 2017-07 and 2020-58), with written informed consent. Data collection: Trained interviewers collected sociodemographic data (age, sex, education, occupation), lifestyle (smoking, alcohol, tea drinking), medical history (hypertension, hyperlipidemia), and medication use (lipid-lowering drugs). Physical examinations measured waist and hip circumference via standardized protocols. Central adiposity definition: WHR calculated as waist/hip circumference. Central adiposity defined per WHO cut-offs: WHR ≥0.90 (men) and ≥0.85 (women). WBC measurements: Venous blood analyzed with automated blood cell analyzer (Xs-1000i, Sysmex, Osaka, Japan) for total WBC and differential counts: neutrophils, lymphocytes, monocytes, eosinophils, basophils. Total WBC was the sum of the five subtypes. Glycemic status: Classified per WHO criteria into normoglycemia, prediabetes, and diabetes. Prediabetes: FPG 6.1–6.9 mmol/L (110–125 mg/dL) with 2h-OGTT <7.8 mmol/L (<140 mg/dL), or FPG <7.0 mmol/L (<126 mg/dL) with 2h-OGTT 7.8–11.0 mmol/L (140–199 mg/dL), in those without prior diabetes. Diabetes: prior physician diagnosis or current diabetes medication, or FPG ≥7.0 mmol/L (≥126 mg/dL) and/or 2h-OGTT ≥11.1 mmol/L (≥200 mg/dL). Covariates: age, sex, education, occupation, smoking, alcohol drinking, tea drinking, hypertension, hyperlipidemia, and lipid-lowering medications. Definitions: Hypertension defined by self-report/diagnosis, treatment, or BP ≥140/90 mmHg. Hyperlipidemia defined by TC ≥6.2 mmol/L, TG ≥2.3 mmol/L, HDL-c <1.0 mmol/L, or LDL-c ≥4.1 mmol/L. Statistical analysis: Baseline characteristics summarized by glycemic status; categorical variables compared by chi-squared tests; continuous variables (non-normal) expressed as medians (IQR) and compared via Kruskal–Wallis tests. Associations of WHR or WBC counts with prediabetes and diabetes were evaluated using logistic regression models adjusting for covariates. Spearman rank correlations assessed relationships between WHR and WBC counts. Mediation analysis: Causal mediation decomposed total effect of WHR (central adiposity vs normal) on outcomes (prediabetes vs normoglycemia; diabetes vs normoglycemia) into natural direct and indirect effects through each WBC mediator using CMAverse in R (RStudio 4.2.2), with 10,000 bootstrap samples for bias-corrected 95% CIs. Both simple (single mediator) and parallel multiple mediation (five WBC subtypes simultaneously) models were performed, adjusting for the same covariates. Software: Stata 16.0 for primary analyses; R (CMAverse) for mediation. Sensitivity analysis: Excluded participants who used antibiotics for infection in the prior 6 months and those with extreme total WBC (<2.5th or >97.5th percentiles), resulting in n=4,374. Re-estimated WBC–glycemic associations and mediation effects.

- Study population: 6,613 adults (35–75 years). Prediabetes n=1,625; diabetes n=1,199; normoglycemia n=3,789. Those with prediabetes/diabetes were older, less educated, and had higher BMI, waist circumference, and WHR. - WHR and glycemic status: Higher WHR associated with greater risk. Adjusted ORs per central adiposity: prediabetes OR 1.53 (95% CI 1.35–1.74); diabetes OR 2.89 (95% CI 2.45–3.40). By sex: males—prediabetes OR 1.61 (1.30–1.99), diabetes OR 2.58 (2.00–3.33); females—prediabetes OR 1.47 (1.26–1.72), diabetes OR 3.14 (2.52–3.90). - WHR and WBC: Positive correlations between WHR and total/differential WBC counts (Spearman r=0.12–0.23; all P<0.01). - WBC and glycemic status (adjusted for WHR and covariates): Higher quartiles of total WBC, neutrophils, lymphocytes, and monocytes were associated with increased odds of prediabetes and diabetes. Examples: total WBC Q4 vs Q1—prediabetes OR 1.59 (1.33–1.89), diabetes OR 2.60 (2.08–3.24). Eosinophils showed no significant association with prediabetes or diabetes; basophils showed limited associations. - Mediation (simple models): For prediabetes, significant indirect effects via total WBC (proportion mediated 9.92%), neutrophils (6.98%), lymphocytes (6.07%), monocytes (3.84%); eosinophils and basophils not significant. For diabetes, significant mediation via total WBC (11.79%), neutrophils (11.51%), lymphocytes (6.29%), monocytes (4.78%), and basophils (1.76%). - Mediation (parallel multiple mediators): Combined five WBC subtypes mediated 10.09% of the WHR–prediabetes association (indirect effect OR 1.04, 95% CI 1.02–1.07) and 12.77% of the WHR–diabetes association (indirect effect OR 1.09, 95% CI 1.06–1.14). - Sensitivity analysis (n=4,374): Findings were consistent. WBC subtypes remained associated with glycemic status; total WBC mediated 15.03% of the WHR–diabetes effect; combined WBCs mediated 11.71%.

The findings demonstrate that central adiposity, measured by WHR, is associated with elevated peripheral WBC counts and with higher odds of both prediabetes and diabetes. This supports the hypothesis that low-grade systemic inflammation accompanies central fat accumulation. By applying causal mediation models, the study shows that a measurable portion of the association between central adiposity and dysglycemia is transmitted through WBC counts—particularly total WBC, neutrophils, lymphocytes, and monocytes—suggesting that leukocyte-related inflammatory processes partially explain the pathway from central adiposity to impaired glycemic control. The mediated proportions (approximately 10–13% for diabetes) imply that while inflammation contributes meaningfully, other mechanisms beyond WBC-measured inflammation also link central adiposity to diabetes. These results align with prior evidence that WHR is a stronger predictor of diabetes than BMI and that leukocytosis relates to insulin resistance and incident T2DM. The mediation analysis adds specificity by quantifying WBCs’ contribution and highlighting differential roles across subtypes. Clinically, the data underscore the importance of central adiposity reduction and suggest that inflammatory markers, including WBC counts, may help risk stratify individuals with abdominal obesity for glycemic deterioration.

Central adiposity was positively associated with prediabetes and diabetes in this Chinese adult cohort. Elevated circulating white blood cell counts, especially total WBC, neutrophils, lymphocytes, and monocytes, partially mediated this association, accounting for about 10–13% of the total effect on diabetes. These findings suggest subclinical inflammation reflected by WBC counts is one mechanistic pathway linking abdominal obesity to worsening glycemic status. Future research should employ longitudinal designs to establish temporality, assess additional inflammatory and metabolic mediators, and evaluate whether modifying inflammation can attenuate obesity-related diabetes risk.

- Cross-sectional design limits inference on temporality and causality among central adiposity, WBC counts, and glycemic status. - Incomplete data on medications and comorbidities affecting inflammation (e.g., steroid/NSAID use, antibiotic history and durations, periodontal disease) could confound WBC levels. - Potential residual confounding remains despite adjustment for multiple covariates. - Other unmeasured mediators may contribute to the central adiposity–diabetes association, so mediation proportions may be underestimated or incomplete.