Impact of weight change on albuminuria in the general population

Background. Increased levels of albuminuria have been recognized as a feature of obesity and the metabolic syndrome, and to be associated with an increased risk for cardiovascular and renal disease. The impact of weight change on albuminuria and its possible mechanism has not been studied yet in the general population. We investigated this issue in a cohort of the Northern European population. Methods. A total of 6894 participants of the Prevention of Renal and Vascular Endstage Disease (PREVEND) study were evaluated from baseline to a mean period of follow-up of 4.2 years for weight change (gain/loss), and its impact on albuminuria, renal function and cardiovascular risk factors. Participants were categorized into three groups based on absolute change in weight from baseline to follow-up: significant weight loss (> 10kg reduction in weight), stable weight, or significant weight gain (> 10 kg increase). Multivariate regression analysis was used to evaluate the effect of baseline characteristics and time-dependent changes in these characteristics on the relationship of weight change with urine albumin excretion (UAE). Results. At follow-up 101 subjects experienced significant weight loss (mean change = -14.2 kg), 348 had significant weight gain (mean change = +13.4 kg) and the remaining were defined stable in weight (mean change = +1.4kg). Weight loss was associated with significant improvement in systolic blood pressure (-11 +/- 15 mmHg), diastolic blood pressure (-5 +/- 8 mmHg), and cholesterol (-0.7 +/- 1.3 mmol/l), even after adjustment for the use of medications (P < 0.001). These parameters worsened significantly in those who substantially gained weight (P < 0.001). Similarly, weight loss was significantly associated with a reduction in UAE (mean -2.2 +/- 1.1 mg/24 h), whereas weight gain was associated with a rise in UAE (mean +0.42 +/- 2.0 mg/24 h). Notably, no changes were observed in GFR (assessed as 24 h urinary creatinine clearance) in subjects with weight loss or weight gain. Multivariate regression modelling with changes in UAE as dependent variable-correcting for factors that might explain the association-showed that only part of the relationship between weight changes and changes in UAE was explained by effect of weight change on blood pressure and cholesterol, whereas the association disappeared with changes in CRP in the model (P = 0.50). Conclusion. This is the first population-based longitudinal study to show that changes in weight are associated with parallel changes in albuminuria. This relationship cannot be fully explained by the association between weight and classical cardiovascular risk factors and renal function. Based on our data we hypothesize that weight-induced changes in vascular inflammation may cause changes in albuminuria.