RESUMO
The quality-initiative analysis of weekly duplicate PEAPOD® body composition measurements was conducted from clinical practice (January to September 2021) on preterm and term infants without respiratory support. Statistical analysis, including regression analysis, Bland-Altman plots and cv-root-mean-square tests, was performed. A total of 188 duplicate (376 individual) measurements were collected from 119 infants (88 preterm, 31 term). The median absolute difference between duplicates was 31.5 g for fat-free mass (FFM). Linear correlation analysis showed R2 = 0.97 for FFM. The absolute differences in FFM and fat mass did not significantly correlate with increasing age. The %FFM differed (p = 0.02) across body weight groups of 1 kg < BW ≤ 2 kg (1.8%; IQR: 0.8, 3.6) and BW > 3 kg (0.9%; IQR: 0.3, 2.1). The median absolute differences were 1 g (IQR: 0.4, 3.1) for body weight and 5.6 mL (IQR: 2.1, 11.8) for body volume. Body volume estimation is charged with a constant absolute error, which is the main factor for differences between repeated body composition assessments. This error becomes more prominent in infants with lower body weights. Nevertheless, reproducibility of weekly PEAPOD testing is sufficient to monitor body compartment changes, offering a foundation for nutritional decisions in both preterm and term infants.
