Table 3.
Adjusted changes in cord plasma leptin, adiponectin, IGF-I and IGF-II in relation to per log unit increase in retinol, β-carotene, α- and γ-tocopherols in newborns (n = 248).
| % change (95% CI)a | Retinol | β-Carotene | α-Tocopherol | γ-Tocopherol | γ:α-Tocopherol ratio |
|---|---|---|---|---|---|
| Leptin | 12.1 (−13.5, 45.2) | −1.1 (−16.1, 16.6) | −0.5 (−19.1, 22.4) | −21.2 (−34.8, −4.7) * | −15.6 (−28.0, −1.0) * |
| Adiponectin | 0.9 (−6.9, 9.5) | 3.8 (−1.4, 9.2) | −6.6 (−12.3, −0.4) * | −3.7 (−9.3, 2.2) | 1.4 (−3.5, 6.7) |
| IGF-I | 9.3 (0.6, 18.7) *,b | −1.5 (−6.6, 4.0) | 0.4 (−6.0, 7.3) | −2.5 (−8.5, 3.8) | −2.3 (−7.3, 3.0) |
| IGF-II | 1.6 (−3.5, 7.0) | −2.5 (−5.7, 0.7) | 0.4 (−3.6, 4.6) | 1.4 (−2.5, 5.4) | 0.8 (−2.4, 4.2) |
| Birth weight (z score) | 0.28 (0.07, 0.48) * | −0.09 (−0.22, 0.04) | −0.03 (−0.20, 0.13) | −0.002 (−0.16, 0.15) | 0.018 (−0.11, 0.14) |
aData presented are the percentage change (95% CI) in each biomarker outcome (leptin, adiponectin, IGF-I, IGF-II) per log unit increase in each vitamin measurement, adjusted for gestational diabetes, pre-pregnancy BMI, maternal ethnicity, education, parity, smoking, alcohol use, mode of delivery, infant sex, gestational age, and birth weight z score. For birth weight as the outcome, the effect was presented as z score change. The adjusted % change was calculated from the regression coefficient for the dependent variable (y) in log scale per log unit increase in each vitamin measurement (x): because the regression coefficient (β) represents the proportion of change in y in the original scale: log y1 − log y0 = β, then log (y1/y0) = β, thus y1/y0 = eβ, and the percentage change is (eβ −1) * 100%.
bThere was a significant interaction between infant sex and cord blood retinol in relation to IGF-I (P = 0.003): the association was present in girls [28.0 (11.1–47.5), P < 0.001)], but not in boys (P = 0.75); otherwise, no significant interactions affecting these effect estimates of interest.
*P < 0.05.