Abstract
BACKGROUND AND PURPOSE: Offspring exposed to maternal diabetes are at increased risk of neurocognitive impairment, but its origins are unknown. With MR imaging, we investigated the feasibility of comprehensive assessment of brain metabolism ((1)H-MRS), microstructure (DWI), and macrostructure (structural MRI) in third-trimester fetuses in women with diabetes and determined normal ranges for the MR imaging parameters measured. MATERIALS AND METHODS: Women with singleton pregnancies with diabetes (n = 26) and healthy controls (n = 26) were recruited prospectively for MR imaging studies between 34 and 38 weeks' gestation. RESULTS: Data suitable for postprocessing were obtained from 79%, 71%, and 46% of women for (1)H-MRS, DWI, and structural MRI, respectively. There was no difference in the NAA/Cho and NAA/Cr ratios (mean [SD]) in the fetal brain in women with diabetes compared with controls (1.74 [0.79] versus 1.79 [0.64], P = .81; and 0.78 [0.28] versus 0.94 [0.36], P = .12, respectively), but the Cho/Cr ratio was marginally lower (0.46 [0.11] versus 0.53 [0.10], P = .04). There was no difference in mean [SD] anterior white, posterior white, and deep gray matter ADC between patients and controls (1.16 [0.12] versus 1.16 [0.08], P = .96; 1.54 [0.16] versus 1.59 [0.20], P = .56; and 1.49 [0.23] versus 1.52 [0.23], P = .89, respectively) or volume of the cerebrum (243.0 mL [22.7 mL] versus 253.8 mL [31.6 mL], P = .38). CONCLUSIONS: Acquiring multimodal MR imaging of the fetal brain at 3T from pregnant women with diabetes is feasible. Further study of fetal brain metabolism in maternal diabetes is warranted.