Introduction In-utero substance exposure is associated with increased risk of infant neurodevelopment impairment. We measured fetal brain volume and surface area in the 3^rd trimester using 3T MRI to investigate: (1) performance of a fetal motion correction algorithm; (2) effect of maternal substance misuse (methodone, diazepam, trazadone).

Methods Three sets of HASTE acquisitions were acquired in each of axial, sagittal and coronal planes in 46 pregnant women at 36–38 weeks’ gestation. Images were motion-corrected and reconstructed to form a single 3D high-resolution brain image for each fetus.¹ Motion-corrected data was rated based on image quality: best (A), good (B), moderate (C), poor (D), reconstruction failed (E). Resulting 3D-datasets with A–C were used to estimate fetal brain intracranial (whole brain including cerebrospinal fluid), cerebellum and cerebrum volumes, and cerebral hemisphere and cerebellum surface areas using stereology methods (Isotropic Cavalieri).² All measures were adjusted for gestational age at scan, birthweight and gender.

Results After motion correction, 36 (78%) datasets met the criteria for subsequent morphological measurements. Intracranial volume (p = 0.011) and cerebral volume (p = 0.017) were significantly lower in fetuses exposed to maternal substance misuse.

Discussion Quality of motion-corrected data depends greatly on degree of fetal motion. In particular gyri and sulci may still not be clearly depicted on the reconstructed 3D image. Imaging earlier in pregnancy may be informative at determining the gestation of fetal brain vulnerability to maternal substance misuse.

References

1. Kim K, et al. IEEE Trans. Med. Imaging. 2010; 29( 1): 146–158

2. Cruz-Orive LM, et al. J Microsc. 2010;240(2):94–110