Skip to main content
. Author manuscript; available in PMC: 2018 Nov 1.
Published in final edited form as: Metabolism. 2017 Jul 17;76:11–22. doi: 10.1016/j.metabol.2017.07.001

Table 2.

Associations of birthweight-for-gestational age z-score (BW/GA) and birth size categories with cord blood metabolomics factor scores

β (95% CI)a in BW/GA z-score per 1 unit factor score OR (95% CI)b of LGA vs. SGA+AGA
BW/GA z-score P LGA
n = 23		 P
Factor 1 0.06 (−0.12, 0.23) 0.53 1.21 (0.71, 2.07) 0.49
Factor 2 0.00 (−0.17, 0.17) 0.98 0.87 (0.53, 1.42) 0.57
Factor 3 0.11 (−0.06, 0.27) 0.21 1.39 (0.86, 2.23) 0.18
Factor 4 0.06 (−0.11, 0.23) 0.46 0.96 (0.60, 1.54) 0.87
Factor 5 0.28 (0.12, 0.43) 0.0004 1.81 (1.09, 3.01) 0.02
Factor 6 −0.09 (−0.27, 0.09) 0.33 0.93 (0.55, 1.57) 0.79
BCAAc 0.20 (0.03, 0.36) 0.02 1.51 (0.92, 2.47) 0.10
a

Estimates are from a linear regression model with the factor score as the exposure, BW/GA as the outcome, and maternal race/ethnicity, infant sex, and gestational age at delivery as covariates. Bolded estimates indicate statistical significance after accounting for multiple comparisons using Bonferroni’s correction (P < 0.05/6).

b

Estimates are from a logistic regression model with the factor score as the exposure, LGA (n = 34) vs. SGA (n = 2) + AGA (n = 101) as the outcome, and includes the same covariates as footnote (a).

c

A weighted metabolite pattern score based on a previously-characterized metabolite pattern implicated in excess adiposity and metabolic risk among Project Viva participants during mid-childhood (10).