Measurement of fetal fat in utero in normal and diabetic pregnancies using magnetic resonance imaging

OBJECTIVES

To assess the reliability of magnetic resonance imaging (MRI) to measure fetal fat volume in utero, and to study fetal growth in women with and without diabetes in view of the increased prevalence of macrosomia in the former.

METHODS

We studied 26 pregnant women, 14 with pre-gestational diabetes and 12 non-diabetic controls. Fetal assessment took place at 24 weeks' gestation and again at 34 weeks by standard ultrasound biometry followed by MRI at 1.5 T. Fetal fat volume was determined from T1-weighted water-suppressed images using a semi-automated approach based on pixel intensity and taking into account partial volume effects. Fetal volume was also determined from the MRI images. Fetal weight was calculated using published fat and lean tissue densities.

RESULTS

There was little fetal fat at 24 weeks' gestation, but at 34 weeks the fetal fat content was considerably higher in the women with diabetes, with a mean fat content of 1090 ± 417 cm(3) compared with 541 ± 348 cm(3) in the controls (P = 0.006). Measurements of fetal fat volume showed low intra- and interobserver variability at 34 weeks, with intraclass correlation coefficients consistently above 0.99. Birth-weight centile correlated with fetal fat volume (R(2)  = 0.496, P < 0.001), percentage of fetal fat (R(2)  = 0.362, P = 0.008) and calculated fetal weight (R(2)  = 0.492, P < 0.001) at 34 weeks.

CONCLUSIONS

MRI appears to be a promising tool for the determination of fetal fat, body composition and weight in utero during the third trimester of pregnancy.

Changes in myometrial 'perfusion' during normal labor as visualized by three-dimensional power Doppler angiography

OBJECTIVES

Myometrial contractions are one of the most important aspects of effective labor. For cells within the myometrium to work efficiently they need to be well oxygenated and this requires an adequate blood supply. This study used quantitative three-dimensional (3D) power Doppler angiography to calculate the percentage change in myometrial blood flow during a relaxation-contraction-relaxation cycle of active labor.

METHODS

Transabdominal 3D power Doppler ultrasound imaging was used to acquire volumetric data during the first stage of spontaneous labor in 20 term, nulliparous women. 3D datasets were acquired during a single cycle of uterine relaxation, contraction and subsequent relaxation for each subject. The resultant datasets were analyzed independently by two investigators on two occasions using Virtual Organ Computer-aided AnaLysis to define a volume of interest within the myometrium; the power Doppler signal within this volume was quantified to provide 3D indices of vascularity: vascularization index (VI), flow index (FI) and vascularization flow index (VFI). The percentage change in these indices, during a uterine contraction, was calculated from the baseline value during the initial uterine relaxation phase (taken as a maximum of 100%).

RESULTS

Myometrial blood flow fell significantly during the uterine contraction and returned during the subsequent relaxation phase of the cycle (P < 0.001 for VI and VFI, P = 0.002 for FI). From the initial baseline relaxation value, VI dropped to 43.9%, FI to 85.5% and VFI to 40.8% during uterine contraction, and returned to 86.7%, 98.1% and 89.1%, respectively, during the subsequent relaxation. The intraclass correlation coefficients in blood flow measurements of 0.982-0.999 between the two investigators were indicative of good interobserver reliability.

CONCLUSIONS

This study confirms that myometrial perfusion, as measured by quantitative 3D power Doppler angiography, significantly falls during uterine contractions, returns during the subsequent relaxation phase, and can be quantified reliably from stored datasets. Further work is now required to establish clinical applicability for this non-invasive investigation.