Having breakfast has no clinically relevant effect on bioelectrical impedance measurements in healthy adults

• Published in: Nutrition journal Vol. 22; no. 1; pp. 1 - 55
• Main Authors: Korzilius, Julia W, Oppenheimer, Sosha E, de Roos, Nicole M, Wanten, Geert J. A, Zweers, Heidi
• Format: Journal Article
• Language: English
• Published: London BioMed Central Ltd 31.10.2023; BioMed Central; BMC
• Subjects: Anthropometry; Bioelectrical impedance analysis; Body composition; Body mass index; Fasting; Fat-free body mass; Fat-free mass; Impedance; Malnutrition; Nutrition assessment; Pilot projects; Random variables; Sarcopenia; Statistical analysis; Netherlands
• DOI: 10.1186/s12937-023-00882-5
• ISSN: 1475-2891

Bioelectrical impedance analysis (BIA) is commonly used to evaluate body composition as part of nutritional assessment. Current guidelines recommend performing BIA measurements in a fasting state of at least 2 h in a clinical setting and 8 h in a research setting. However, since asking patients with malnutrition or sarcopenia to fast is not desirable and literature to support the strategy in the guidelines is lacking, this study aimed to assess the impact of breakfast on BIA measurements. We performed an explorative, prospective study in healthy volunteers aged between 18 and 70 years, with a normal fluid balance and a body mass index between 18.5 and 30 kg/m.sup.2. BIA measurements were performed according to the standard operating procedure in the fasting state, and 1, 2, 3, and 4 h after ingesting a standardized breakfast meal of about 400 kcal with a 150 mL drink, using the hand-to-food single-frequency BIA (Bodystat500 [R]). The Kyle formula was used to calculate the primary outcome, i.e. fat-free mass (FFM, kg). A linear mixed model was used to compare baseline values with other time points. A difference of 1 kg in FFM was considered clinically relevant. Thirty-nine (85% female) volunteers were included, with a median age of 28 years (IQR 24-38). In 90% of the participants, having breakfast had no clinically relevant impact on the estimated FFM. For the group, the most pronounced mean difference, a statistically but not clinically significant higher value of 0.2 kg (0.4%), was observed after 3 h of fasting compared to baseline. No statistically significant differences were found at the other time points. Eating affects single-frequency BIA measurements, but differences in FFM remain below clinical relevance for most participants when using a standardized breakfast. Thus, the current study suggests performing a BIA measurement in a fasting state is not required.