Nonlinear Dynamics of Red Blood Cell Distribution Width and Visceral Adiposity Index: Insights from the 2005-2018 NHANES Data

Technical Summary: Nonlinear Dynamics of Red Blood Cell Distribution Width and Visceral Adiposity Index

Problem Statement
Obesity, particularly the accumulation of visceral fat, is a critical public health issue linked to metabolic dysfunction, cardiovascular disease, and increased mortality. While the Visceral Adiposity Index (VAI) is a well-validated proxy for visceral fat and metabolic risk, the relationship between VAI and Red Blood Cell Distribution Width (RDW) remains unexplored. RDW, a standard hematological parameter reflecting erythrocyte size variation, is increasingly recognized as a marker of systemic inflammation and oxidative stress. However, existing literature presents conflicting evidence regarding RDW's association with metabolic conditions; some studies suggest elevated RDW correlates with metabolic syndrome, while others indicate an inverse relationship or no association. This study aims to clarify the association between RDW and VAI in a nationally representative adult population and investigate potential nonlinear dynamics and subgroup variations.

Methodology
The study utilized a cross-sectional design analyzing data from the National Health and Nutrition Examination Survey (NHANES) spanning 2005 to 2018.

• Study Population: From an initial pool, 14,568 adults (aged ≥20 years) with complete data for RDW, VAI components (BMI, waist circumference, triglycerides, HDL-C), and covariates were selected. Exclusion criteria included participants under 20, pregnant individuals, those with a history of cancer, and individuals with missing data for key variables.
• Variables:
  • Exposure: RDW was analyzed as both a continuous variable and in quartiles (Q1: 9.7–12.4%, Q2: 12.4–12.9%, Q3: 13.0–13.7%, Q4: 13.8–37.8%).
  • Outcome: VAI was calculated using sex-specific formulas incorporating waist circumference, BMI, triglycerides, and HDL-C.
  • Covariates: Analyses adjusted for age, gender, race/ethnicity, education, poverty-to-income ratio, marital status, smoking, physical activity, hypertension, diabetes, BMI, alcohol use, and various lipid and hematological markers.
• Statistical Analysis: The study employed weighted statistical methods to ensure national representativeness. Techniques included:
  • Multivariate linear regression to assess the linear association between RDW and VAI.
  • Smoothed spline regression and two-piecewise linear regression models to detect nonlinear relationships and identify inflection points.
  • Subgroup analyses and interaction tests to evaluate consistency across demographics (sex, age, race) and clinical conditions (hypertension, diabetes, BMI).
  • Software used included R (v4.2.2) and EmpowerStats (v4.2).

Key Results

• Inverse Association: Multivariate linear regression revealed a significant inverse association between RDW and VAI. In the fully adjusted model, each unit increase in RDW was associated with a decrease in VAI ($\beta = -0.13$; 95% CI: -0.21 to -0.05). When RDW was categorized by quartiles, higher quartiles (Q2, Q3, Q4) showed significantly lower VAI levels compared to Q1, with a statistically significant trend ($p < 0.001$).
• Nonlinear Dynamics: Curve fitting and threshold effect analysis identified a nonlinear relationship. An inflection point was detected at an RDW value of 12.9%. Below this threshold, the negative correlation between RDW and VAI was significant; above 12.9%, the association became nonsignificant.
• Subgroup Variations: The inverse relationship was not uniform across all populations. Significant interactions were observed for sex and hypertension.
  • Sex: The negative correlation was more pronounced in males ($\beta = -0.29$) compared to females ($\beta = -0.09$).
  • Hypertension: A significant inverse correlation was found in hypertensive patients ($\beta = -0.25$), whereas no significant relationship was detected in non-hypertensive groups.
  • No significant interactions were found for age, race, physical activity, smoking, diabetes, or BMI.

Key Contributions

1. Novel Association: This is the first study to examine the relationship between RDW and VAI in a large, nationally representative adult cohort, establishing a significant inverse correlation.
2. Nonlinearity Discovery: The study uniquely identifies a nonlinear pattern in this relationship, pinpointing a critical inflection point at RDW = 12.9%, suggesting that the physiological mechanisms linking RDW to visceral adiposity may differ depending on the RDW level.
3. Stratified Insights: The research highlights that the RDW-VAI association is not universal but is particularly relevant for male patients with hypertension, offering a more nuanced understanding of metabolic risk markers.

Significance and Claims
The authors claim that RDW serves as a valuable, cost-effective indicator for identifying metabolic dysfunction and visceral adiposity, particularly in high-risk populations such as hypertensive males. The findings suggest that lower RDW levels (specifically below 12.9%) may be associated with a more stable erythropoiesis process and lower systemic inflammation, correlating with a healthier metabolic profile in these specific subgroups.

The paper posits that the observed inverse relationship may be driven by complex physiological mechanisms, including systemic inflammation, oxidative stress, and erythropoiesis regulation. In hypertensive males, the authors hypothesize that higher androgen levels and increased blood viscosity may drive physiological adaptations (such as more uniform red blood cell production) that lower RDW, thereby creating a specific inverse correlation with VAI.

Limitations
The authors explicitly acknowledge that the cross-sectional design precludes causal inferences. Additionally, while adjustments were made for numerous covariates, residual confounding cannot be entirely ruled out. The study does not propose specific clinical interventions but suggests that defining optimal RDW thresholds (e.g., the 12.9% inflection point) could aid in future metabolic risk assessments.