Albuminuria and echocardiographic measures of cardiac structure in type 2 diabetes: effect modification by body mass index
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background Albuminuria is an established marker of cardiovascular risk in type 2 diabetes mellitus (T2DM), but whether its association with cardiac structure differs according to body mass index (BMI) remains unclear. We treated confounding and effect modification as distinct questions and asked whether BMI confounds this association or modifies it. Methods In this single-center cross-sectional study, 227 hospitalized patients with T2DM had a valid albumin-to-creatinine ratio, and the nine with macroalbuminuria were excluded from the primary cohort of 218. Cardiac structure was characterized by two prespecified echocardiographic parameters, left atrial diameter (LAD) and left ventricular mass index (LVMI). Albuminuria was modeled as ln(ACR), and BMI was entered as a continuous variable in the prespecified primary interaction test. Multivariable linear regression, interaction testing, BMI-stratified subgroup analyses, and sensitivity analyses were performed. Results Higher ln(ACR) was associated with LAD (β = 0.066, P = 0.014) and LVMI (β = 2.842, P = 0.033) before adjustment, although both effect sizes were small. Both estimates attenuated after adjustment and were no longer significant (LAD P = 0.178; LVMI P = 0.244). The prespecified continuous BMI by ln(ACR) interaction was significant for both outcomes (LAD P = 0.010; LVMI P = 0.026). In BMI-stratified analyses, ln(ACR) was associated with LAD at normal BMI (β = 0.108, 95% CI: 0.031–0.185; Q = 0.024). Associations in the overweight and obesity strata were non-significant and imprecise. The corresponding LVMI estimate (β = 3.902, 95% CI: −0.117 to 7.921) did not remain significant after FDR correction ( Q = 0.184). Across sensitivity analyses the LAD interaction persisted, but the LVMI interaction did not in the complete cohort of 227 patients ( P = 0.064). Conclusions In this cross-sectional cohort, the association between albuminuria and cardiac structure varied with BMI, but the adjusted main associations were not significant. The interaction indicates heterogeneity across BMI strata rather than a causal role for BMI. The subgroup evidence was strongest for LAD at normal BMI and did not extend to LVMI. These findings are hypothesis-generating and do not support BMI-stratified screening or individualized risk stratification. Prospective studies with adjudicated outcomes are needed to confirm these observations.
Source: Frontiers in Nutrition
Authors: Shuan Wang(Writing – original draft, Writing – review & editing), Zimeng Zhu(Writing – original draft, Writing – review & editing), Hui Chen(Writing – original draft), Junxuan Chen(Writing – original draft), Xiaotian Chen(Writing – review & editing), Xiaosu Chen(Writing – original draft), Qing Zhao(Writing – original draft), Yue Xu(Writing – original draft, Writing – review & editing), Qin Wang(Writing – review & editing), Yong Wang(Writing – review & editing)
Language: English
Subfield: Cardiology and Cardiovascular Medicine
Field: Medicine
Domain: Health Sciences
Keywords: albuminuria, type 2 diabetes, body mass index, left atrial diameter, left ventricular mass index, cardiac structure, albumin-creatinine ratio, effect modification, obesity, overweight, confounding, cardiovascular risk, hospitalized patients, echocardiography, interaction testing, stratified analysis, FDR correction, sensitivity analysis
Open Access status: gold
Estimated APC paid: $3,801
Funder: Nanjing Drum Tower Hospital
OpenAlex ID: W7236437055
