Least significant change with repeat bioelectrical impedance analysis measurements in children with metabolic dysfunction-associated steatotic liver disease: A descriptive cohort study.

Abstract

Bioelectrical impedance is used clinically to assess body composition. To determine true (rather than chance/measurement error) change in bioelectrical impedance analysis measurements over time, it is necessary to know their least significant change. Least significant change represents values exceeding the 95% prediction interval of the precision error of repeat measurements. The least significant change of repeat bioelectrical impedance analysis measurements in children with obesity and metabolic dysfunction-associated steatotic liver disease is currently unknown. This is a prospective, single-center, descriptive cohort study in youth 6-20 years of age with obesity and metabolic dysfunction-associated steatotic liver disease. Two same-day bioelectrical impedance analysis measurements were performed on a multifrequency, octopolar device (InBody 370). Fat mass and fat-free mass were adjusted for age using respective indices (dividing by height squared). Fasting status was determined by patient report. Descriptive statistics (medians with interquartile ranges and means with standard deviations and proportions) were used; the least significant change between repeat measurements with precision interval was calculated. We recruited 43 patients (81% male; 33% Hispanic; median age: 14 years [interquartile range: 11-16]; median body mass index z score: 2.31 [interquartile range: 2.84-2.65]). A total of 60% were fasting, for a median of 12 h. The least significant change of fat mass index was 0.5 (precision interval: -0.4 to 0.5) kg/m2, fat-free mass index was 0.3 (precision interval: -0.3 to 0.3) kg/m2, and body fat percent was 1.5% (precision interval: -1.3 to 1.50). Fasting status affected least significant change measurements. In youth with obesity and metabolic dysfunction-associated steatotic liver disease, repeat bioelectrical impedance analysis measurements beyond least significant change and precision interval determined in this study likely represent true changes in body composition over time, vs measurement error.