No abstract

The prevalence of lumbosacral intervertebral disk bulge and herniation on sagittal magnetic resonance (MR) images was determined in 45 pregnant subjects and 41 asymptomatic nonpregnant women of childbearing age. MR technique differed for the pregnant and nonpregnant groups. Fifty-three percent of pregnant and 54% of nonpregnant women had an abnormal disk (bulge or herniation) at one or more levels (L3-4, L4-5, or L5-S1). The difference was not statistically significant. There was also no significant difference in the distribution of outcomes among the 45 pregnant subjects, 17 parous nonpregnant women, and 24 nulliparous women. The results suggest that lumbosacral disk bulges or herniations are common in women of childbearing age, whether pregnant or not, and that, contrary to an earlier report, pregnant women do not have an increased prevalence of disk abnormalities.

Twenty-four pregnant women carrying 26 fetuses (two sets of twins) were imaged with magnetic resonance (MR) imaging at 0.35 T following sonographic evaluation. Each study was retrospectively evaluated to determine which of 33 normal fetal structures were visible on the images and which imaging parameters were most useful for depicting fetal anatomy. Fetal motion degraded fetal images in all but two cases, both with oligohydramnios and in the third trimester of gestation. Nevertheless, many fetal structures were identifiable, particularly in the third trimester. Visualization of fetal anatomy improved with intravenous maternal sedation in five cases. Relatively T1-weighted images occasionally offered the advantage of less image degradation owing to fetal motion and improved contrast between different fetal structures. More T2 weighting was believed to be advantageous in one case for outlining the fetal head and in one case for delineation of the brain. In many cases, structures were similarly identifiable (though with different signal intensities) regardless of the parameters selected. The authors conclude that MR imaging of many fetal structures is currently unsatisfactory and is probably of limited value, particularly in the first and second trimesters. However, the relative frequency and detail with which the fetal head and liver can be depicted indicate that these may be areas for further investigation, and the potential utility of imaging fetal fat warrants further investigation.

Sixteen pregnant patients with pelvic masses detected with ultrasound (US) were studied with magnetic resonance (MR) imaging at 0.35 T. Two patients were in their first trimester, eight in the second, and six in the last. On MR images, 34 masses were seen, including 19 uterine leiomyomas. With US, 20 masses were detected. All masses not seen with US were leiomyomas. Only 20 masses (in 11 patients) were removed for histologic study. In nine cases, MR imaging and US provided similar information on the origin, extent, and type of mass. In seven patients, MR imaging contributed additional information. MR imaging depicted more leiomyomas than US in three patients. In another case, MR images showed that a mass depicted with US was actually a loop of bowel. MR images allowed differentiation between a solid soft-tissue mass and a hemorrhagic fluid-containing mass, correctly depicted the presence of an abdominal pregnancy, and allowed evaluation of the parametrium for spread of cervical carcinoma.

Eight patients with angiographically proved arteriovenous malformations (AVMs) of the extremities (seven congenital, one posttraumatic) were evaluated with magnetic resonance (MR) imaging using a 0.35-T superconducting system and spin-echo pulse sequences. Congenital AVMs appeared as accumulations of dilated tortuous blood vessels infiltrating the involved muscles. A posttraumatic acquired AVM of the shoulder consisted of a large feeding artery associated with a pseudoaneurysm and a soft-tissue mass. MR imaging allowed precise anatomic localization and provided details concerning the size and extent of the AVMs. The relationship of AVMs to specific muscle groups, bones, and vascular structures could be accurately determined. Although major feeding and draining vessels were identified, the exact arteries and veins supplying and draining the AVM could not be ascertained. Images obtained in the transverse plane consistently yielded the most useful information. MR imaging and angiography may be complementary techniques in the initial evaluation, follow-up, and treatment planning of AVMs of the extremities.