Equations for Prediction of Body Adiposity in Patients With Chronic Kidney Disease Undergoing Hemodialysis

Abstract

Objectives: The prevalence of obesity in the population has increased and excess body adiposity is one of the main nutritional disorders in patients with chronic kidney disease (CKD) on hemodialysis (HD). The objective of this study was to develop equations using anthropometric measurements to predict the total and abdominal body adiposity of patients with CKD on HD. Methods: This is a cross-sectional study evaluating 323 patients with CKD on HD in city in northeastern Brazil. Measurements and anthropometric indicators were correlated with percentage of body fat (%BF) and visceral fat (VF, in kg) measured by dual energy X-ray absorptiometry. Multiple linear regression models based on different combinations of anthropometric measurements were adjusted to develop the equations, with subsequent cross-validation. Results: Of the 323 patients, 62.2% were male and 46.5% were aged between 40 and 59 years. The equation selected to estimate %BF included weight, height, waist and hip circumferences, and triceps and suprailiac skin folds, presenting high predictive capacity (R2 = 0.771). The equation selected to estimate VF included weight, height, waist circumference, hip circumference, and sum of skin folds (R22 = 0.796). Conclusions: The proposed equations efficiently predicted the %BF and VF (kg) of patients with CKD on HD, thereby serving as viable indicators in clinical practice.