Hypertension was common in this group: 68.5 percent of participants had high blood pressure, a figure that underscores how frequently the two conditions travel together. To probe the relationship, the team used multivariable logistic regression, a statistical method that estimates the odds of hypertension while accounting for other influencing factors. They divided participants into four groups, or quartiles, based on their CVAI and LAP scores.

Compared with people in the lowest quartile, those in the highest quartile had an odds ratio of 3.27 for hypertension according to CVAI (95 percent confidence interval: 2.66 to 4.02) and 3.01 according to LAP (95 percent confidence interval: 2.43 to 3.72). The trend across quartiles was statistically significant for both indices, with a P value for trend below 0.001. The analysis went beyond simple linear assumptions.

Using restricted cubic spline modeling, a technique that allows the shape of a relationship to bend rather than forcing it into a straight line, the researchers uncovered a striking asymmetry between the two indices. LAP showed a significant nonlinear association with hypertension, with a P value for non-linearity below 0.001, and a threshold effect emerged at a LAP value of approximately 44.2 (95 percent confidence interval: 42.141 to 46.233). CVAI, by contrast, maintained a linear relationship with hypertension in both sexes.

The P values for non-linearity were 0.177 in men and 0.055 in women, neither of which reached the threshold for a statistically significant departure from linearity. In practical terms, this means that for the visceral adiposity index, each incremental increase in the score was associated with a steady rise in hypertension odds across the observed range, without an obvious inflection point. To test whether the findings held up across different types of patients, the investigators performed stratified subgroup analyses, examining whether the association varied by characteristics such as sex.

The association between the obesity indices and hypertension remained consistent across all strata, with P values for interaction all exceeding 0.05, indicating no statistically meaningful modification of the relationship by the subgroup variables examined. The team also took steps to address a common weakness of observational datasets: missing data. Covariates with absent values were handled through multiple imputation, a statistical procedure that fills in gaps by generating several plausible complete datasets and pooling the results.

As a sensitivity evaluation, the researchers repeated their analysis using only complete cases, checking that the conclusions did not hinge on how the missing values were treated. Several caveats frame the results. As a cross-sectional study, the design captures a single snapshot in time, so it can establish association but not causation; it cannot determine whether higher visceral fat precedes the development of hypertension or whether the two conditions influence each other.

The participants were drawn from a single hospital in one Chinese city, which may limit how broadly the findings generalize to other populations. The odds ratios describe odds of hypertension within this diabetic cohort rather than absolute risk in the general population.

Subject of Research: Association of visceral adiposity indices with hypertension in type 2 diabetes

Article Title: The association between obesity index and hypertension in type 2 diabetes: a cross-sectional study

Article References: Xing, Y., yin, Y., zhao, Q., zhang, L., & Xiu, X. (2026). The association between obesity index and hypertension in type 2 diabetes: a cross-sectional study. Frontiers in Endocrinology, 17, 1998153. https://doi.org/10.3389/fendo.2026.1998153

Image Credits: AI Generated

DOI: 10.3389/fendo.2026.1998153

Keywords: type 2 diabetes, hypertension, visceral adiposity, CVAI, lipid accumulation product, cross-sectional study, logistic regression, restricted cubic spline, obesity, blood pressure, China, association