#299 The association between body fat distribution, albuminuria, and mortality in relatively healthy adults

Abstract

Abstract Background and Aims Albuminuria is a risk factor for kidney disease progression and cardiovascular disease (CVD). Metabolic syndrome has been associated with incident albuminuria. We aimed to understand which measures of adiposity are associated with albuminuria and mortality in relatively healthy adults. Method We included 8,215 adults in the National Health and Nutrition Examination Survey (years 2003-6 and 2011-2016) without prior CVD, estimated glomerular filtration rate <60 ml/min/1.73 m2 (eGFR), diabetes, hypertension, or prior dialysis. We used multivariable adjusted linear regression modelling to evaluate sex-stratified relationships between urine albumin excretion and multiple measures of adiposity, controlling for urine creatinine. We then performed sex-stratified multivariable adjusted Cox regression models of the association between adiposity and all-cause mortality. Measures of adiposity included android fat mass index (FMI), gynoid FMI, android-to-gynoid ratio, subcutaneous FMI, abdominal FMI, visceral FMI, body mass index (BMI), and waist circumference. Models were adjusted for age, race or ethnicity, education, food insecurity, health insurance, survey year, physical activity, eGFR, total cholesterol, systolic blood pressure, hemoglobin A1c, and urine albumin:creatinine ratio (UACR), and menopausal status (among women). Models were subsequently adjusted for BMI and waist circumference. Results The sample was 52.3% women, the mean (standard deviation, SD) BMI was 27.1 kg/m2 (SD 6.7), and the median UACR was 5.4 mg/g [3.8, 8.6]. Over a median follow up of 7.8 years [IQR 5.3-14.3], there were 128 deaths among men and 89 deaths among women. Among men, a 1-SD increase in android fat mass index (FMI) was associated with 13.1% (95% CI 4.5, 22.5) higher albuminuria. Similarly, a 1-SD increase in visceral FMI was associated with 8.3% (95% CI 1.3, 15.7) higher albuminuria. BMI was non-linearly associated with albuminuria among men. Among women, a 1-unit increase in android to gynoid ratio was associated with 47.7% lower albuminuria (% change = −47.7 (95% CI −61.4, −29.3). A 5 kg/m2 increase in BMI was associated with 11.0% lower albuminuria (% change = −11.0 (95% CI −15.0, −6.7) and a 3 cm increase in waist circumference was associated with 1.6% (95% CI 0.03, 3.0) higher albuminuria. Among men, a 1-SD increase in abdominal FMI was associated with 2.09-fold higher risk of all-cause mortality (HR 2.09 [95% CI 1.16-3.74]), independent of BMI and waist circumference. Among women, a 1-SD increase in gynoid FMI was associated with 56% decreased risk of all-cause mortality (HR 0.44 [95% CI 0.20-0.88]), independent of BMI and waist circumference. Conclusion Adiposity is differently associated with albuminuria and mortality in relatively healthy men and women. Sex-specific patterns of adipose deposition may inform CVD risk stratification.