ABSTRACT
The use of magnetic resonance imaging (MRI) has been shown to be safe during pregnancy
and can accurately assess fetal fat and rate of growth. The purpose of this study
was to investigate with MRI the relationship between maternal weight, fat distribution,
and glucose levels and neonatal birthweight ratio, percent fat, and infant outcome
in pregnancies complicated by gestational diabetes. Twenty gestational diabetic women
were studied at 36 to 38 weeks' gestation with hemoglobin A1c, a glucose tolerance test and maternal body fat by electrical impedance. MRI scans
of the maternal upper arm and from her zyphoid process to her pubic symphysis, including
all fetal components, were analyzed for percent fat of the mother's arm and trunk
and her fetus' trunk. All variables were then related to infant outcome (birthweight,
birthweight ratio, and neonatal morbidity). Maternal body composition was related
to maternal weight (p = 0.012, r = 0.54), as was maternal arm fat (p <0.005, r = 0.60).
Arm fat did correlate with trunk fat, but trunk fat did not correlate was well with
weight (p = NS). Although maternal Hb A1c correlated with maternal weight (p = 0.05, r = 0.43), it did not correlate with infant
birthweight. In contrast, maternal mean glucose levels correlated with both maternal
and infant weight and infant birthweight ratio (p = 0.05, r = 0.60; p = 0.05, r =
0.42; and p = 0.01, r = 0.64, respectively). The infant birthweight ratio was predicted
by the MRI, both by a subjective assessment of the images of fetal fat and by quantitating
the fetal fat by determining the mean of two points of maximal fat thickness in the
abdominal wall (p <0.001, r = 0.88). In addition, when the mother's arm was more than
50 mm2 or the fetus was more than 4.0 mm mean thickness of maximal abdominal fat, the risk
of macrosomia, and neonatal hypoglycemia increased. The MRI at 36 to 38 weeks of gestation
is able to estimate birthweight ratio and neonatal morbidity in women with gestational
diabetes.
