Body Composition in 995 Acutely ill or Chronically ill Patients at Hospital Admission: A controlled population Study

Abstract

Objective To determine if fat-free mass and fat mass in acutely ill and chronically ill patients differed from healthy controls at hospital admission and if prevalence of malnutrition differed by body mass index (BMI) or fat-free mass percentile.Subjects/Setting 995 consecutive patients 15 to 100 years of age admitted to the hospital were measured in the hospital admission center and compared with 995 healthy age- and height-matched subjectsDesign Cross-sectional study. Fat-free mass, fat mass, and percentage fat mass were determined by 50 kHz bioelectrical impedance analysis. Prevalence of malnutrition was determined by BMI <20kg/m2 or fat-free mass in the 10th percentile.Statistical Analysis Analysis of variance was used to examine differences between acutely ill and chronically ill patients and controls and between age groups.Results Fat-free mass was significantly lower in patients than controls (P<.05), and the difference with age in fat-free mass in patients was greater than the age-related difference in the controls. A higher percentage fat mass was found in spite of lower BMI in chronically ill patients older than 55 years. Among participants, 25% of acutely ill and 37.3% of chronically ill patients fell below fat-free mass in the 10th percentile, compared with 15.6% of acutely ill and 18.9% of chronically ill patients falling below BMI ≤20kg/m2.Applications/Conclusion Weight and BMI do not evaluate body compartments and therefore do not reveal if weight changes result in loss of fat-free mass or gain in fat mass. In spite of minimal differences in BMI between patients and controls, we found that fat-free mass was lower and fat mass was higher in acutely ill and chronically ill patients than controls. The objective measurement of body composition, as part of a comprehensive nutritional assessment, helps to identify subjects who have low fat-free mass or high fat mass. J Am Diet Assoc. 2002;102:944–948,953–955.