Cross-sectional associations between central and general adiposity with albuminuria: observations from 400,000 people in UK Biobank

The global increase in chronic kidney disease (CKD) mortality has been linked to rising adiposity levels. While general adiposity, measured by BMI, is associated with advanced CKD, the role of central adiposity (e.g., waist-to-hip ratio) and the influence of factors like diabetes (HbA1c) remain unclear. Albuminuria, an early marker of CKD, offers a valuable outcome for comparing the independent effects of general and central adiposity. Previous studies have yielded inconsistent results regarding the association between central adiposity and albuminuria, particularly after adjusting for BMI. This study aimed to clarify these associations using data from the large UK Biobank cohort, examining the independent effects of both central and general adiposity on albuminuria, considering potential mediating factors like diabetes, vascular disease, and blood pressure, and exploring the modifying effect of HbA1c levels. The importance of this research stems from the potential to improve our understanding of CKD risk factors and guide preventive strategies.

A systematic review preceding this study identified 46 observational studies examining adiposity and albuminuria. However, only five studies incorporated central adiposity measures after adjusting for BMI, and their findings were inconsistent. Some smaller studies found no association after adjusting for BMI, while others found a positive association. A large Japanese study showed an association in men but not women. The lack of consensus highlighted the need for a large-scale study with robust methodology to clarify the relationship between central and general adiposity and albuminuria.

This cross-sectional study utilized data from 408,527 participants in the UK Biobank. Waist-to-hip ratio (WHR) and BMI were selected as primary adiposity measures due to their widespread use and low correlation. Urinary albumin-to-creatinine ratio (uACR) was categorized into undetectable, low normal, high normal, and albuminuria groups. Ordinal logistic regression was employed to analyze the associations between adiposity measures and uACR categories, adjusting for confounders (age, ethnicity, education, deprivation, smoking, physical activity, urinary sodium-to-creatinine ratio) and measurement error using repeated measures. Models were further adjusted for potential mediators (diabetes status, diabetes duration, vascular disease history, blood pressure) to determine the extent to which adiposity's effects on uACR were mediated by these factors. Associations were analyzed separately for men and women and combined using inverse variance weighting. Subgroup analyses were conducted based on diabetes status to explore potential modifying effects of HbA1c. Sensitivity analyses were also performed with and without measurement error correction and using albuminuria as a binary outcome.

After adjusting for confounders and measurement error, each 0.06 increase in WHR was associated with a 55% increase in the odds of a higher uACR category. This association attenuated to 32% after adjusting for BMI. Similarly, a 5 kg/m² increase in BMI was associated with a 47% increase in the odds, reduced to 35% after adjusting for WHR. Approximately 40% of the association between central adiposity and albuminuria was mediated by diabetes, vascular disease, and blood pressure. The associations between both WHR and BMI with albuminuria were apparent across different HbA1c levels, indicating that these associations are not solely explained by diabetes status. Diabetes status itself was strongly associated with increased odds of albuminuria (2-3 fold increase). There was some evidence to suggest that the adiposity-albuminuria associations were somewhat stronger in participants with diabetes compared with those without diabetes or with prediabetes. Analyses of other adiposity measures showed similar trends, except for waist and hip circumferences.

The findings indicate that both central and general adiposity are independently and positively associated with albuminuria, even after accounting for potential mediators and confounders. The persistence of these associations across various HbA1c levels suggests that mechanisms beyond diabetes contribute to the link between adiposity and kidney function. The significant association between adiposity and albuminuria, even in individuals with normal HbA1c levels, emphasizes the importance of managing adiposity for maintaining kidney health across the diabetic spectrum. This study’s large sample size and comprehensive adjustment for confounders and mediators strengthens the conclusions, providing strong evidence for the independent contribution of both central and general adiposity to albuminuria risk.

This large-scale study confirms the independent association of both central and general adiposity with albuminuria risk. These associations appear robust and exist even after accounting for established mediators like diabetes and vascular disease, emphasizing the importance of adiposity management for kidney health. Future research could explore the underlying biological mechanisms linking adiposity to kidney damage and evaluate the effectiveness of interventions targeting adiposity reduction in preventing or slowing CKD progression.

The cross-sectional study design limits the ability to establish causality. Residual confounding remains possible despite the extensive adjustments. The reliance on self-reported data for certain variables introduces potential bias. Generalizability to populations with different ethnic compositions or healthcare access needs further investigation.