314 - Novel body composition assessments in very preterm infants: Skin fold thickness as a marker of regional fat distribution

Poster Number: 314
Publication Number: 314.129

Catherine O. Buck, Yale School of Medicine, Guilford, CT, United States; Kristin Leone. Santoro, Harvard Medical School/ Beth Israel Deaconess Medical Center, Boston, MA, United States; Veronika Shabanova, Yale University, New Haven, CT, United States; Tessa Kehoe, Yale School of Medicine, Wethersfield, CT, United States; Rachel L. Hansen, Beth Isreal Deaconess Medical Center, Boston, MA, United States; Marianna B. Castellanos, Beth Israel Deaconess Medical Center, Boston, MA, United States; Denisse V. Moncada, Beth Israel Deaconess Medical Center, Jamaica Plain, MA, United States; Kelly Walter, Beth Israel Deaconess Medical Center, Boston, MA, United States; Camilia R. Martin, Weill Cornell Medicine, New York City, NY, United States; Sarah N. Taylor, Yale University, MADISON, CT, United States

Background: Although minimally studied, regional fat distribution may be a marker of metabolic health and brain growth in preterm infants. Point of care ultrasound (US) has been used to assess regional fat in term infants but has not been used widely in preterm infants.

Objective: To describe changes in bedside body composition assessments in very preterm infants from birth through 36-weeks post-menstrual age (PMA).

Design/Methods: Infants born at two level IV NICUs between 24 0/67 and 31 6/7 weeks gestational age (GA) were enrolled after birth. Body composition assessments were completed from birth through 36-weeks PMA, including body circumferences, skin fold thickness (SFT) by caliper and SFT and skin through muscle thickness by US. SFT measures were obtained in duplicate, and thickness measures obtained from US images were measured by two independent raters. Mean change in measures were plotted over time, and linear and mixed effects regression was used to model change in body composition assessments from birth through 36-weeks PMA using constant slope (linear PMA effect) or time-varying slope (polynomial PMA effect).

Results: Among 48 infants in the analysis, 35% were born extremely preterm, 37% were female sex, and average birth weight z-score was 0.06 (±1.04) with 10% of infants born small for GA (Table 1). Overall, there were average increases across PMA for each body composition outcome (Table 2), with some observed GA at birth cohort effects in the ultrasound measures (Figure 1). Triceps, biceps, and subscapular SFT by US increased by 0.19 mm (95% CI: 0.15, 0.22), 0.12 mm (95% CI: 0.09, 0.14), and 0.16 mm (95% CI: 0.13, 0.20) each post-menstrual week, respectively. While gains in mid-arm and abdominal circumference remained constant over time, mid-thigh circumference remained constant between birth and 26 weeks PMA, after which thigh circumference increased between 0.32 cm (95% CI: 0.15, 0.49) and 0.88 cm (95% CI: 0.75, 1.01) per week from 26 through 37 weeks PMA (Table 2). Similarly, triceps and subscapular caliper SFT and triceps and abdominal SFT by US remained constant from birth through 30 weeks PMA, after which these measures increased through 37 weeks PMA (Table 2).

Conclusion(s): In this cohort, body composition assessments by ultrasound demonstrated PMA-dependent trajectories in skin fold and skin through muscle thickness increases from birth through 36-weeks PMA. Use of ultrasound as a novel method to assess regional fat development in very preterm infants is feasible. Additional investigation is warranted to determine gestational age cohort and sex specific effects of these measures over time.