HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Coakley, F. V. Articles by Filly, R. A. Search for Related Content PubMed PubMed Citation Articles by Coakley, F. V. Articles by Filly, R. A. Social Bookmarking                      What's this?

Fetal MRI: A Developing Technique for the Developing Patient

¹ All authors: Department of Radiology, University of California San Francisco, 505 Parnassus Ave., San Francisco, CA 94143-0628.

Received May 8, 2003; accepted after revision July 8, 2003.

 
Address correspondence to F. V. Coakley.

Introduction

Top Introduction History of Fetal MRI Safety of Fetal MRI Technique of Fetal MRI Indications Conclusion References

 
Sonography is the primary technique for fetal imaging because of its proven utility, widespread availability, and relatively low cost. However, limitations include a small field of view, limited soft-tissue acoustic contrast, beam attenuation by adipose tissue, poor image quality in oligohydramnios, and limited visualization of the posterior fossa after 33 weeks' gestation because of calvarial calcification [1]. Accordingly, sonographic findings are occasionally inconclusive or insufficient to guide treatment choices [2, 3]. Over the past decade, fetal MRI has emerged as a clinically useful supplement to sonography and is rapidly moving from the realm of select academic medical centers into community practice. Advances in fetal medicine and surgery have also driven the development of fetal MRI [4, 5]. Any radiologist who performs prenatal sonography can expect to see occasional patients who will benefit from the incremental information provided by MRI. This article aims to provide a timely and general review of fetal MRI, including a discussion of history, safety, current techniques, and common indications. Practical aspects are emphasized. Readers should be aware that the field of fetal MRI is still evolving and that the material presented necessarily reflects the authors' institutional experience and bias.

History of Fetal MRI

Top Introduction History of Fetal MRI Safety of Fetal MRI Technique of Fetal MRI Indications Conclusion References

 
MRI of women during pregnancy was first described in 1983 [6]. Initial obstetric applications were primarily related to maternal and placental abnormalities [2, 7]. Fetal applications were largely confined to volumetric measurements using echoplanar imaging because of the image degradation introduced by fetal motion on standard sequences [8–10]. Attempts to eliminate fetal motion artifact included the administration of muscle relaxants directly into the umbilical vein [11].

During the early 1990s, fetal MRI was revolutionized by the development of the single-shot rapid acquisition sequence with refocused echoes [12] (Fig. 1A, 1B, 1C). Single-shot rapid acquisition with refocused echoes is a high-quality T2-weighted sequence that has a slice acquisition time of less than a second, effectively "freezing" fetal motion [13]. Equipment for single-shot rapid acquisition with refocused echoes is commercially available as a single-shot fast spin-echo unit (General Electric Medical Systems, Milwaukee, WI) and a half-Fourier acquisition single-shot turbo spin-echo unit (Siemens Medical Solutions, Iselin, NJ).

View larger version (156K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1A. —Comparison of fetal MRIs obtained during previous 12 years. Oblique sagittal gradient-echo fetal image (TR/TE, 0/6; flip angle, 33.30°) obtained in 1991 shows flow-related enhancement in aneurysm in veins of Galen (asterisk). Anatomic relationships are difficult to see.

View larger version (156K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1B. —Comparison of fetal MRIs obtained during previous 12 years. Axial spoiled gradient-echo T1-weighted fetal image (140/4.2; flip angle, 70°) obtained in 1998 shows flow-related enhancement in aneurysm in veins of Galen (asterisk) and in draining straight sinus (arrow).

View larger version (170K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1C. —Comparison of fetal MRIs obtained during previous 12 years. Sagittal single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes of same fetus as shown in B shows aneurysm (asterisk) as focal signal void.

Safety of Fetal MRI

Top Introduction History of Fetal MRI Safety of Fetal MRI Technique of Fetal MRI Indications Conclusion References

 
United States Food and Drug Administration guidelines [14] require labeling of MRI devices to indicate that the safety of MRI with respect to the fetus "has not been established." Safety concerns arise for both mother and fetus. Maternal safety concerns are the same as for a nonpregnant patient and are addressed by standard MRI screening. Fetal safety concerns are related to teratogenesis and acoustic damage.

Most studies suggest MRI during pregnancy is safe [15–18], but several animal studies have raised the possibility of teratogenetic effects in early pregnancy [19–21]. Although these studies may not be applicable to humans or may represent statistical quirks, they suggest that a cautious approach to adopting the use of MRI in women during the first trimester may be advisable. The guidelines of the National Radiological Protection Board in the United Kingdom [22] state, "It might be prudent to exclude pregnant women during the first three months of pregnancy." An additional concern in the first trimester is the underlying relatively high rate of spontaneous abortion during this period. An MRI study could be coincidentally followed by a spontaneous abortion that might appear iatrogenic to the patient. That said, when a strong clinical indication has been established, MRI is probably still preferable to any study involving ionizing radiation [23].

The loud noises generated by the coils of the MR scanner as they are subjected to rapidly oscillating electromagnetic currents could potentially cause acoustic damage to the fetus. Two reports from the United Kingdom [24, 25] provide reassuring clinical and experimental evidence that the risk of acoustic injury is negligible. In summary, pregnant women in the second and third trimester can be reassured that MRI poses no known risk to the fetus. Although safety has not been positively established, any hazard appears negligible and is outweighed by the potential diagnostic benefit. A more cautious approach should be taken when MRI is required during the first trimester.

Technique of Fetal MRI

Top Introduction History of Fetal MRI Safety of Fetal MRI Technique of Fetal MRI Indications Conclusion References

 
The mother should fast for 4 hr before the examination to reduce bowel peristalsis artifacts and to prevent postprandial fetal motion and should empty her bladder immediately before undergoing MRI. Standard MRI screening procedures should be used. We believe written consent is advisable, although it is not mandatory, and local practice may also influence the consent process. Fetal sedation, by maternal oral administration of 1 mg of flunitrazepam has been recommended in Europe to reduce fetal movement [1], although we have not found routine sedation necessary for acquisition of T2-weighted images.

A surface phased array multicoil should be used to improve spatial resolution [26]. The mother can ordinarily be scanned in a supine position, but a left lateral decubitus position is usually preferable during late pregnancy to prevent compression of the inferior vena cava by the gravid uterus. After a localizer sequence is acquired, images are graphically prescribed in planes anatomic to the fetus. T2-weighted images are useful to assess both anatomy and pathology; for most studies, we acquire single-shot rapid acquisition T2-weighted images with refocused echoes in the axial, coronal, and sagittal planes. Repetition of some sequences may be required because the image is degraded by fetal motion during acquisition (Fig. 2A, 2B) or because fetal motion between sequences results in images that are not in true anatomic planes. Variants of the single-shot rapid acquisition sequence with refocused echoes that allow changes in scanning parameters, including the imaging plane during sequence acquisition, are in development and may prove useful in the future. Fat and hemorrhage may be shown on T1-weighted images [27]. Fetal bowel content may also be of high T1 signal intensity, and this finding can be used to identify the gastrointestinal tract. Satisfactory T1-weighted images can be difficult to obtain without sedation. Multislice spoiled gradient-echo appears to be the most robust sequence [28]. Lateralization of fetal anatomy as right or left should be based on analysis of fetal position relative to the mother because fetal landmarks may be unreliable as a result of transposition.

View larger version (175K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2A. —Comparison of MRIs with fetal motion. Image degradation by fetal motion during acquisition can usually be overcome by repetition of sequence. Axial single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes of fetal brain obtained during fetal motion is markedly degraded.

View larger version (176K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2B. —Comparison of MRIs with fetal motion. Image degradation by fetal motion during acquisition can usually be overcome by repetition of sequence. Subsequent axial single-shot rapid acquisition T2-weighted image (infinite/100) with refocused echoes obtained few minutes after degraded image (A) shows that fetus is not moving. This image is of diagnostic quality.

Maternally administered IV gadolinium crosses the placenta and is not approved for use in pregnant women. To date, no role for IV contrast material has been shown in fetal MRI. Our protocol for fetal MRI is shown in Table 1. High spatial resolution (i.e., small field of view, thin sections, and large matrix) is desirable but should not be overdone because gains are offset by "wrap" artifact and reduced signal-to-noise ratio (Fig. 3A, 3B).

View larger version (127K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3A. —Comparison of fields of view in fetal MRIs. Small field of view is intuitively desirable for fetal MRI but may be detrimental if carried to extremes. Coronal single-shot rapid acquisition T2-weighted images (TR/TEeff, infinite/100) with refocused echoes of fetal brain differ markedly. Image obtained with 14-cm field of view (A) is grainy and degraded by "wrap" artifact (asterisk, A), whereas image obtained with 20-cm field of view (B) is much less grainy and is not degraded by wrap artifact. Absence of corpus callosum can now be seen.

View larger version (165K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3B. —Comparison of fields of view in fetal MRIs. Small field of view is intuitively desirable for fetal MRI but may be detrimental if carried to extremes. Coronal single-shot rapid acquisition T2-weighted images (TR/TEeff, infinite/100) with refocused echoes of fetal brain differ markedly. Image obtained with 14-cm field of view (A) is grainy and degraded by "wrap" artifact (asterisk, A), whereas image obtained with 20-cm field of view (B) is much less grainy and is not degraded by wrap artifact. Absence of corpus callosum can now be seen.

Indications

Top Introduction History of Fetal MRI Safety of Fetal MRI Technique of Fetal MRI Indications Conclusion References

 
Overview
Major indications for fetal MRI include the confirmation of inconclusive sonographic findings and the evaluation of sonographically occult diagnoses. Many of the described applications are based on case reports or small series, and definitive recommendations on the appropriateness of fetal MRI in a given situation are not currently available. Nevertheless, studies requested to answer a specific clinical question are more likely to affect treatment than studies with a less well-defined focus [29]. The impact of fetal MRI on treatment can be particularly difficult to assess because a contemporaneous standard of reference is frequently lacking. Furthermore, studies in which fetal MRI performed at academic centers is compared with sonography performed at community hospitals [30], rather than sonography performed at equivalent academic centers [29], tend to exaggerate the apparent advantages of MRI. Common neurologic and nonneurologic indications for fetal MRI at our institution are described in the next sections. Other less common applications, including the evaluation of abdominal and genitourinary abnormalities, have been described but are not included in this review.

Neurologic Indications
Ventriculomegaly.—Ventriculomegaly is the most common central nervous system abnormality identified on prenatal sonography. Ventriculomegaly is defined as an atrial width greater than 10 mm measured at the posterior margin of the glomus of the choroid plexus on an axial plane through the thalami [31]. Despite the growth of the surrounding brain, the atrial diameter is relatively constant from 15 to 35 weeks' gestation, so the lateral ventricles appear proportionately larger early in gestation [32]. Many disorders can result in fetal ventriculomegaly, and 70–84% of fetuses with ventriculomegaly show associated structural or chromosomal anomalies [33–37]. Associated structural abnormalities include neural tube defects, agenesis of the corpus callosum, Dandy-Walker syndrome, holoprosencephaly, cortical malformations, intracranial hemorrhage, and porencephaly [36, 38].

Children with isolated prenatal ventriculomegaly appear to have a better neurodevelopmental outcome than those in whom additional abnormalities are present [34, 38–44]. In a series of 194 fetuses diagnosed with prenatal ventriculomegaly, the frequency of developmental delay was 37% in children with isolated ventriculomegaly, compared with 84% in children with additional abnormalities [39]. In patients with mild isolated ventriculomegaly, defined as an atrial width of 10–15 mm with no chromosomal or additional sonographic abnormalities, the frequency of neurodevelopmental abnormality ranges from 0% to 36% [38, 45–47]. The risk is lower if the atrial diameter is less than 12 mm and if the fetus is male [38, 46, 47]. Counseling parents after a prenatal diagnosis of isolated mild fetal ventriculomegaly is challenging, and several studies have investigated the potential role of prenatal MRI in this setting.

MRI shows additional central nervous system abnormalities in up to 50% of fetuses with sonographically isolated ventriculomegaly, including agenesis of the corpus callosum, cortical malformations, periventricular heterotopia, periventricular leukomalacia, multicystic encephalomalacia, and intracranial hemorrhage [48–53]. The prognostic implications of these additional findings remain under investigation. Periventricular white matter injury may manifest as focal periventricular T2 hyperintensity (Fig. 4A, 4B), focal defects in the germinal matrix, subtle irregularity of the ventricular margin, or large areas of abnormal signal in the developing white matter and overlying cortex. The latter may or may not be associated with volume loss. Hemorrhage is usually detected as foci of T1 hyperintensity and T2 hypointensity in the germinal matrix, ventricles, or brain parenchyma. Blood in the ventricles may layer or form a discrete clot. Small subependymal hemorrhage may be difficult to identify on rapid acquisition images with refocused echoes because the germinal matrix has signal characteristics similar to those of blood.

View larger version (147K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 4A. —25-gestational-week-old fetus referred for MRI because of mild bilateral ventriculomegaly seen on routine prenatal sonography. Axial single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes of fetal brain shows mild enlargement of lateral ventricles and focal hyperintensity (arrow) adjacent to frontal horn of left lateral ventricle.

View larger version (143K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 4B. —25-gestational-week-old fetus referred for MRI because of mild bilateral ventriculomegaly seen on routine prenatal sonography. Coronal single-shot rapid acquisition T2-weighted image (infinite/100) with refocused echoes obtained through frontal horns confirms periventricular focal hyperintensity (arrow), consistent with parenchymal injury; finding suggests worse postnatal developmental outcome than that expected with isolated mild bilateral ventriculomegaly.

Agenesis of the corpus callosum.—The corpus callosum reaches adult form by 18–20 weeks' gestation [54]. The prevalence of callosal agenesis in the general population is estimated to be 0.2–0.7%, rising to 3% in mentally disabled patients [55]. Most patients with callosal agenesis have neurodevelopmental disorders, including developmental delay, mental disability, and epilepsy [55–58]. At autopsy, 85% of adults with callosal agenesis have additional central nervous system anomalies, and 62% have anomalies outside the central nervous system [59]. Approximately 50% of fetuses with callosal agenesis have detectable additional central nervous system anomalies such as Dandy-Walker syndrome, Chiari's malformation type II, gray matter heterotopia, holoprosencephaly, schizencephaly, and encephalocele [59–62]. These fetuses have a higher incidence of motor or mental disorders compared with those with isolated agenesis [59, 62]. Findings of callosal agenesis—including enlarged atria and occipital horns with a teardrop configuration of the lateral ventricles, absence of the cavum septum pellucidum, a high-riding third ventricle, and radiating medial sulci—can be difficult to identify sonographically. MRI offers improved detection of both callosal agenesis and associated anomalies [1, 3, 48, 63]. MRI can directly depict the corpus callosum on images obtained in the sagittal and coronal planes (Fig. 5A, 5B). In a study of 50 patients with proven callosal agenesis, all cases were identified on prenatal MRI, but only 33 cases were identified on sonography [3].

View larger version (180K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 5A. —22-gestational-week-old fetus referred for MRI because of suspected agenesis of corpus callosum on routine prenatal sonography. Coronal single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes shows absence of corpus callosum and abnormal morphology of medial brain surface and continuity of third ventricle (black arrow) with interhemispheric fissure (white arrow).

View larger version (168K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 5B. —22-gestational-week-old fetus referred for MRI because of suspected agenesis of corpus callosum on routine prenatal sonography. Midline sagittal single-shot rapid acquisition T2-weighted image (infinite/100) with refocused echoes confirms complete absence of corpus callosum.

Posterior fossa abnormalities.—Posterior fossa abnormalities that can be evaluated by prenatal MRI include Dandy-Walker syndrome, Dandy-Walker variant (Fig. 6), mega cisterna magna, arachnoid cyst, and Chiari's malformation type II [1, 3, 53, 64–66]. In patients with Dandy-Walker syndrome, fetal MRI may display additional abnormalities that indicate a worse prognosis, including agenesis of the corpus callosum, polymicrogyria, neuronal heterotopia, and occipital encephalocele [67–69]. Chiari's malformation type II may also be associated with other anomalies, such as callosal agenesis, polymicrogyria, gray matter heterotopia, cerebellar dysplasia, syringohydromyelia, diastematomyelia, and diplomyelia.

View larger version (163K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 6. —22-gestational-week-old fetus referred for MRI because routine prenatal sonography raised suspicion of Dandy-Walker syndrome. Sagittal single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes shows hypogenesis of cerebellar vermis (arrow) with normal corpus callosum and no additional abnormalities. Appearances are consistent with those of Dandy-Walker variant.

Complications of monochorionic twin pregnancies.—Twin pregnancies carry substantially higher morbidity and mortality than singleton pregnancies. Monochorionic twins (monozygotic twins contained in a single chorionic membrane, as established on sonography) are subject to several specific complications that may be indications for fetal intervention, including twin–twin transfusion syndrome, twin-embolization syndrome, acardiac syndrome, and conjoined twinning [70]. Several of these complications are associated with neurologic impairment because of presumed thrombotic end-organ ischemia [71, 72]. Parenchymal destruction can be identified on fetal MRI [1, 3, 51] as focal or diffuse areas of increased T2 signal in the germinal matrix, developing white matter, or cortex (Fig. 7). We have found that these injuries are best visualized as cavitary lesions if imaging is performed at least 2 weeks after a possible ischemic episode such as a co-twin demise or fetal intervention. In the future, diffusion MRI may allow earlier identification of ischemic injuries.

View larger version (175K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 7. —24-gestational-week-old monochorionic twin fetus with periventricular white matter injury. In utero endoscopic laser ablation of placental vascular connections was performed 2 weeks before scanning for treatment of twin–twin transfusion syndrome. Findings on concurrently obtained sonogram of brain (not shown) were unremarkable, but autopsy confirmed coagulative necrosis of periventricular white matter. Coronal single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes shows twin pregnancy. Difference in size of fetuses is consistent with twin–twin transfusion syndrome. Area of increased T2 signal intensity (arrow) is seen adjacent to frontal horn of left lateral ventricle in smaller fetus (i.e., donor twin) with focal ventricular dilatation.

Malformations of cerebral cortical development.—Neuronal precursors originate from the germinal matrix lining the ventricles and migrate to the developing cortex between 7 and 20 weeks' gestation. Abnormalities of neuronal development and migration may be sonographically occult but are detectable on MRI because of its superior tissue contrast [3, 50, 64]. In a study of 20 patients with proven migrational disorders, fetal MRI was superior to sonography for identifying schizencephaly, lissencephaly, polymicrogyria, and gray matter heterotopia [3].

Identification of cortical malformations requires an understanding of normal brain maturation as seen on MRI, and several excellent reviews of this topic are available [1, 50, 73, 74]. Subependymal heterotopia appears as nodules along the ventricular walls that are isointense relative to the germinal matrix (Fig. 8A, 8B). These nodules are radiologically indistinguishable from the subependymal tubers of tuberous sclerosis. Other manifestations of tuberous sclerosis such as transmantle dysplasia, cortical tubers, and cardiac rhabdomyoma may help in differentiation. Schizencephaly appears as a gray matter–lined cleft between the ventricle and subarachnoid space (Fig. 9A, 9B). Shallow Sylvian fissures, absence of normal multilayered brain parenchyma, and a reduction in gestationally appropriate cortical sulcation are the characteristic features of classical lissencephaly.

View larger version (187K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 8A. —23-gestational-week-old fetus with subependymal heterotopia, subsequently confirmed at autopsy, referred for possible inferior vermian agenesis seen on routine prenatal sonography. Other images (not shown) confirmed normal vermis. Axial single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes of fetal brain shows nodule of decreased signal (arrow) along right lateral ventricle.

View larger version (168K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 8B. —23-gestational-week-old fetus with subependymal heterotopia, subsequently confirmed at autopsy, referred for possible inferior vermian agenesis seen on routine prenatal sonography. Other images (not shown) confirmed normal vermis. Coronal single-shot rapid acquisition T2-weighted image (infinite/100) with refocused echoes confirms subependymal nodule (arrow).

View larger version (184K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 9A. —33-gestational-week-old fetus with bilateral open lip schizencephaly referred for MRI after routine sonography suggested possible holoprosencephaly. Coronal single-shot rapid acquisition T2-weighted image (TR/TEeff, infinite/100) with refocused echoes shows bifrontal clefts (arrows) extending from ventricles to subarachnoid space. Clefts are lined with areas of low signal intensity. Septum pellucidum is absent.

View larger version (160K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 9B. —33-gestational-week-old fetus with bilateral open lip schizencephaly referred for MRI after routine sonography suggested possible holoprosencephaly. Axial T2-weighted single-shot rapid acquisition image (infinite/100) with refocused echoes shows abnormal gyral pattern (arrow) adjacent to clefts.

Nonneurologic Indications
Congenital diaphragmatic hernia.—Congenital diaphragmatic hernia is a developmental defect in the posterolateral diaphragm with herniation of abdominal viscera into the thorax. Congenital diaphragmatic hernia has an incidence of 1 in 3,000–4,000 live births, and 90% of cases are left-sided [75]. The cause is unknown, but one third of cases are associated with chromosomal or additional anatomic abnormalities and have a mortality rate of 76% [76]. The position of the liver and the degree of pulmonary hypoplasia are important prognostic factors in isolated congenital diaphragmatic hernia because mortality is predominantly caused by compression of the lungs from the herniated abdominal viscera. From 60% to 86% of left-sided congenital diaphragmatic hernias [75, 77] are "liver-up" and have a mortality of 57% compared with 7% for "liver-down" cases [76, 77].

The sonographic diagnosis of congenital diaphragmatic hernia and the evaluation of liver position can be difficult because lung and liver are of similar echogenicity. On prenatal MRI, lung, liver, stomach, and bowel are easily identified (Fig. 10A, 10B). Because of its excellent soft-tissue contrast, MRI can be used to confirm the diagnosis, evaluate liver position (Fig. 11), and perform lung voluntary [78]. Lung volume measured by planimetry on MRI can be expressed as a percentage of the expected lung volume based on fetal size, a measurement known as the relative lung volume [79]. This measurement appears to be of prognostic importance; in a preliminary study of isolated left congenital diaphragmatic hernia, three of four fetuses with a relative lung volume of less than 40% died postnatally despite intensive treatment, and all seven fetuses with a relative lung volume of greater than 40% survived [80].

View larger version (120K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 10A. —29-gestational-week-old fetus without congenital diaphragmatic hernia. Coronal spoiled gradient-echo T1-weighted image (TR/TE, 140/4.2; flip angle, 70°) shows normal chest and abdomen. Liver (arrow) is of high T1 signal intensity.

View larger version (135K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 10B. —29-gestational-week-old fetus without congenital diaphragmatic hernia. Coronal T2-weighted single-shot rapid acquisition image (TR/TEeff, infinite/100) with refocused echoes shows lungs (white arrows) as high signal intensity. Fluid is visible in stomach (black arrow).

View larger version (123K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 11. —Fetus with left-sided congenital diaphragmatic hernia. Coronal spoiled gradient-echo T1-weighted image (TR/TE, 140/4.2; flip angle, 70°) shows upward herniation of left hepatic lobe (arrow). Prognosis is worse than that for patient with congenital diaphragmatic hernia, but left hepatic lobe remains in abdomen.

Pulmonary sequestration.—Pulmonary sequestration is a developmental mass of nonfunctioning bronchopulmonary tissue that is separate from the tracheobronchial tree and receives arterial blood from the systemic circulation (usually from the aorta). Postnatal pulmonary sequestrations are classified as extralobar (15–25%) or intralobar (75–85%), depending on whether the sequestration has a separate pleural investment or is in the pleura of the lung. Most, if not all, prenatal sequestrations are extralobar and are characterized pathologically by diffuse dilatation of bronchioles, alveoli, and subpleural lymphatic vessels. Cysts are present occasionally. Pulmonary sequestrations account for up to 23% of prenatally detected lung lesions [81], and MRI is increasingly used as a supplement to obstetric sonography in the detection of congenital thoracic anomalies [30]. On MRI, sequestrations typically appear as well-defined masses in the chest that are of higher T2 signal intensity than normal lung [30] but lower than that of free amniotic fluid (Fig. 12). The frequency with which MRI reveals feeding vessels has not been systematically established [30, 82], and the incremental benefit of MRI over sonography remains under investigation. We have found MRI helpful in the prenatal distinction of subdiaphragmatic sequestration from neuroblastoma [82].

View larger version (175K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 12. —Typical fetal extralobar sequestration. Axial T2-weighted single-shot rapid acquisition image (TR/TEeff, infinite/100) with relaxation enhancement shows fetal chest sequestration (black asterisk) as large left-sided triangular mass of increased signal intensity relative to displaced and compressed normal lungs (arrows). Lungs and heart (white asterisk) are displaced to right.

Congenital cystic adenomatoid malformation.—Congenital cystic adenomatoid malformation is an abnormal developmental lung mass composed of a proliferation of terminal bronchioles. The blood supply is usually drawn from the pulmonary arteries. Communication with the bronchial tree or gastrointestinal tract may be present. Congenital cystic adenomatoid malformations may consist of a few large or medium-sized cystic spaces (macrocystic type) or of multiple tiny cysts (microcystic type). The microcystic type may appear solid on prenatal sonography. Small- to moderate-sized congenital cystic adenomatoid malformations usually have a benign course and are treated by postnatal resection. Large congenital cystic adenomatoid malformations are increasingly recognized as a cause of prenatal death because progressive enlargement can lead to compression of the esophagus, vena cava, and lungs, resulting in impaired swallowing, reduced venous return, pulmonary hypoplasia, polyhydramnios, and hydrops fetalis. Prenatally detected congenital cystic adenomatoid malformations, especially if large, should be closely monitored for the development of polyhydramnios or hydrops, which are indications for early delivery in a mature fetus and for prenatal resection in an immature fetus [83, 84].

On prenatal MRI, congenital cystic adenomatoid malformations are seen as intrapulmonary masses of increased T2 signal intensity [30]. Discrete cysts may be identified (Fig. 13). In the absence of a visible feeding artery from the aorta suggesting the diagnosis of sequestration, congenital cystic adenomatoid malformation and sequestration may be indistinguishable.

View larger version (138K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 13. —Fetus with right-sided congenital cystic adenomatoid malformation. Coronal T2-weighted single-shot rapid acquisition image (TR/TEeff, infinite/100) with refocused echoes shows chest with right-sided congenital cystic adenomatoid malformation (black arrow). Heart (H) and left lung (L) are displaced to left. Macrocyst (white arrow) is visible in lesion.

Airway obstruction.—Airway obstruction at birth is life-threatening. Congenital obstruction of the upper airway is usually extrinsic, caused by either a cervical lymphangioma (including cystic hygroma) or a teratoma [85]. Cervical lymphangiomas are composed of dilated lymphatic spaces, possibly stemming from local failure of lymphatic connections during development, and are often complicated by hydrops, probably caused by compression of the neck vessels. Chromosomal anomalies are present in 30–70% of fetuses. Cervical teratomas are usually benign isolated tumors that can be cured by surgery if the airway can be maintained during and after delivery. Both lymphangioma and teratoma may appear solid or cystic on prenatal imaging. The finding of a predominantly solid tumor or a cystic tumor with solid nodules favors the diagnosis of teratoma, and intrathoracic extension favors the diagnosis of lymphangioma [85]. Congenital high airway obstruction syndrome is a rare intrinsic form of obstruction of the larynx or upper trachea [86] that results in retention of bronchial secretions and pulmonary distention by the retained fluid. Overinflation of the lungs with flattening or aversion of the diaphragm is thought to impair venous return to the heart, resulting in fetal hydrops and ascites. This results in a characteristic constellation of sonographic findings including large bilateral echogenic fetal lungs, flattening or aversion of the diaphragm, dilated fluid-filled airways below the level of obstruction, and fetal hydrops or ascites. These findings can also be recognized on MRI [79]. Major airway obstruction below the carina may result in ipsilateral pulmonary hyperinflation (Fig. 14A, 14B).

View larger version (171K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 14A. —Fetus with right-sided bronchial atresia. Axial T2-weighted single-shot rapid acquisition image (TR/TEeff, infinite/100) with refocused echoes shows that right lung (asterisk) is grossly overexpanded and heart and left lung (arrow) are displaced to left.

View larger version (149K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 14B. —Fetus with right-sided bronchial atresia. Sagittal T2-weighted single-shot rapid acquisition image (infinite/100) with refocused echoes shows dilatation of proximal right airways (thick black arrow) below level of obstruction. Diaphragm (white arrow) is inverted because of overexpansion of right lung. Ascites (thin black arrow) visible in abdomen indicates development of hydrops fetalis.

Volumetric measurements.—The acquisition of multiple contiguous slices on MRI allows easy and accurate measurement of fetal volumes both of the entire fetus and of individual fetal organs [8–10]. Assessment of fetal liver volume on prenatal MRI may facilitate recognition of intrauterine growth retardation, which is difficult to diagnose accurately using clinical or sonographic criteria. In a study of 32 high-risk pregnancies, 11 resulted in the birth of a fetus with intrauterine growth retardation [10]. Ten of these 11 fetuses had an abnormally small liver volumes on prenatal MRI, and the remaining 21 fetuses had normal liver volumes.

Conclusion

Top Introduction History of Fetal MRI Safety of Fetal MRI Technique of Fetal MRI Indications Conclusion References

 
Technical and therapeutic advances have driven the development of fetal MRI, which is likely to become an increasingly important modality in the evaluation of sonographically complex or occult anomalies of the fetal brain and body. All radiologists involved in prenatal imaging should be aware of the applications and limitations of this modality.

References

Top Introduction History of Fetal MRI Safety of Fetal MRI Technique of Fetal MRI Indications Conclusion References

 

1. Garel C, Brisse H, Sebag G, Elmaleh M, Oury JF, Hassan M. Magnetic resonance imaging of the fetus. Pediatr Radiol1998; 28:201 –211[Medline]
2. Angtuaco T, Shah H, Mattison D, Quirk J. MRI in high risk obstetric patients: a valuable complement to US. RadioGraphics1992; 12:91 –109[Abstract]
3. Sonigo PC, Rypens FF, Carteret M, Delezoide AL, Brunelle FO. MR imaging of fetal cerebral anomalies. Pediatr Radiol1998; 28:212 –222[Medline]
4. James D. Fetal medicine. BMJ1998; 316:1580 –1583[Free Full Text]
5. Coakley FV. Role of MRI in fetal surgery. Top Magn Reson Imaging 2001;1:39 –51[Medline]
6. Smith FW, Adam AH, Phillips WDP. NMR imaging in pregnancy. Lancet 1983;1:61 –62[Medline]
7. Stark DD, McCarthy SM, Filly RA, Parer JT, Hricak H, Callen PW. Pelvimetry by magnetic resonance imaging. AJR1985; 144:947 –950[Abstract/Free Full Text]
8. Baker PN, Johnson IR, Gowland PA, et al. Fetal weight estimation by echo-planar magnetic resonance imaging. Lancet1994; 343:644 –645[Medline]
9. Baker PN, Johnson IR, Gowland PA, Freeman A, Adams V, Mansfield P. Estimation of fetal lung volume using echo-planar magnetic resonance imaging. Obstet Gynecol1994; 83:951 –954[Medline]
10. Baker PN, Johnson IR, Gowland PA, et al. Measurement of fetal liver, brain, and placental volumes with echo-planar magnetic resonance imaging. Br J Obstet Gynaecol1995; 102:35 –39[Medline]
11. Yuh WTC, Nguyen HD, Fisher DJ, et al. MR of fetal central nervous system abnormalities. AJNR1994; 15:459 –464[Abstract]
12. Kiefer B, Grassner J, Hausman R. Image acquisition in a second with half Fourier acquisition single-shot turbo spin-echo. (abstr) J Magn Reson Imaging 1994;4(P):86 –87
13. Levine D, Barnes PD, Sher S, et al. Fetal fast MR imaging: reproducibility, technical quality, and conspicuity of anatomy. Radiology1998; 206:549 –554[Abstract/Free Full Text]
14. United States Food and Drug Administration. Magnetic resonance diagnostic device: panel recommendation and reports on petitions for MR reclassification. Federal Register1988; 53:7575 –7579
15. Mevissen M, Buntenkotter S, Loscher W. Effects of static and time-varying (50 Hz) magnetic fields on reproduction and fetal development in rats. Teratology1994; 50:229 –237[Medline]
16. Beers GJ. Biological effects of weak electromagnetic fields from 0 Hz to 200 Hz: a survey of the literature with special emphasis on possible magnetic resonance effects. Magn Reson Imaging1989; 7:309 –331[Medline]
17. Schwartz JL, Crooks LE. NMR imaging produces no observable mutations or cytotoxicity in mammalian cells. AJR1982; 139:583 –585[Free Full Text]
18. Wolff S, Crooks LE, Brown P, Howard R, Painter R. Test for DNA and chromosomal damage induced by nuclear magnetic resonance imaging. Radiology1980; 136:707 –710[Abstract/Free Full Text]
19. Heinrichs WL, Fong P, Flannery M, et al. Midgestational exposure of pregnant BALB/c mice to magnetic resonance imaging. Magn Reson Imaging 1988;6:305 –313[Medline]
20. Tyndall DA, Sulik KK. Effects of magnetic resonance imaging on eye development in the C57BL/6J mouse. Teratology1991; 43:263 –275[Medline]
21. Yip YP, Capriotti C, Talagala SL, Yip JW. Effects of MR exposure at 1.5 T on early embryonic development of the chick. J Magn Reson Imaging 1994;4:742 –748[Medline]
22. National Radiological Protection Board. Principles for the protection of patients and volunteers during clinical magnetic resonance diagnostic procedures: documents of the NRPB, vol.2 . London, England: HM Stationery Office,1991
23. Shellock FG, Kanal E. Policies, guidelines, and recommendations for MR imaging safety and patient management. J Magn Reson Imaging 1991; 1:97 –101[Medline]
24. Baker PN, Johnson IR, Harvey PR, Gowland PA, Mansfield P. A three-year follow-up of children imaged in utero with echo-planar magnetic resonance. Am J Obstet Gynecol1994; 170:32 –33[Medline]
25. Gover P, Hykin J, Gowland P, Wright J, Johnson I, Mansfield P. An assessment of the intrauterine sound intensity level during obstetric echo-planar magnetic resonance imaging. Br J Radiol1995; 68:1090 –1094[Abstract]
26. Campeau NG, Johnson CD, Felmlee JP, et al. MR imaging of the abdomen with a phased-array multicoil: prospective clinical evaluation. Radiology1995; 195:769 –776[Abstract/Free Full Text]
27. Barkovich AJ, Atlas SW. Magnetic resonance imaging of intracranial hemorrhage. Radiol Clin North Am1988; 26:801 –820[Medline]
28. Yamashita Y, Namimoto T, Abe Y, et al. MR imaging of the fetus by a HASTE sequence. AJR1997; 168:513 –519[Abstract/Free Full Text]
29. Coakley FV, Hricak H, Filly RA, Barkovich AJ, Harrison MR. Impact on management of breath-hold MR imaging of complex fetal disorders: preliminary clinical experience. (abstr) Radiology1998; 209(P):189
30. Hubbard AM, Adzick NS, Crombleholme TM, et al. Congenital chest lesions: diagnosis and characterization with prenatal MR imaging. Radiology1999; 212:43 –48[Abstract/Free Full Text]
31. Cardoza JD, Goldstein RB, Filly RA. Exclusion of fetal ventriculomegaly with a single measurement: the width of the lateral ventricular atrium. Radiology1988; 169:711 –714[Abstract/Free Full Text]
32. Filly RA, Goldstein RB. The fetal ventricular atrium. Radiology1994; 193:315 –317[Free Full Text]
33. Pretorius DH, Davis K, Manco-Johnson ML, Manchester D, Meier PR, Clewell WH. Clinical course of fetal hydrocephalus: 40 cases. AJR 1985;144:827 –831[Abstract/Free Full Text]
34. Nyberg DA, Mack LA, Hirsch J, Pagon RO, Shepard TH. Fetal hydrocephalus: sonographic detection and clinical significance of associated anomalies. Radiology1987; 163:187 –191[Abstract/Free Full Text]
35. Vintzileos AM, Campbell WA, Weinbaum PJ, Nochimson DJ. Perinatal management and outcome of fetal ventriculomegaly. Obstet Gynecol 1987;69:5 –11[Abstract/Free Full Text]
36. Cochrane DD, Myles ST, Nimrod C, Still DK, Sugarman RG, Wittmann BK. Intrauterine hydrocephalus and ventriculomegaly: associated abnormalities and fetal outcome. Can J Neurol Sci1985; 12:51 –59[Medline]
37. Chervenak FA, Duncan C, Ment LR, et al. Outcome of fetal ventriculomegaly. Lancet1984; 2:179 –181[Medline]
38. Vergani P, Locatelli A, Strobelt N, et al. Clinical outcome of mild fetal ventriculomegaly. Am J Obstet Gynecol1998; 178:218 –222[Medline]
39. Gupta JK, Bryce FC, Lilford RJ. Management of apparently isolated fetal ventriculomegaly. Obstet Gynecol Surv1994; 49:716 –721[Medline]
40. Goldstein RB, La Pidus AS, Filly RA, Cardoza J. Mild lateral cerebral ventricular dilatation in utero: clinical significance and prognosis. Radiology1990; 176:237 –242[Abstract/Free Full Text]
41. Mahony BS, Nyberg DA, Hirsch JH, Petty CN, Hendricks SK, Mack LA. Mild idiopathic lateral cerebral ventricular dilatation in utero: sonographic evaluation. Radiology1988; 169:715 –721[Abstract/Free Full Text]
42. Drugan A, Krause B, Canady A, Zador IE, Sacks AJ, Evans MI. The natural history of prenatally diagnosed cerebral ventriculomegaly. JAMA 1989; 261:1785 –1788[Abstract]
43. Bromley B, Frigoletto FD, Benacerraf BR. Mild fetal lateral cerebral ventriculomegaly: clinical course and outcome. Am J Obstet Gynecol 1991; 164:863 –867[Medline]
44. Nicolaides KH, Berry S, Snijders RJ, Thorpe-Beeston JG, Gosden C. Fetal lateral cerebral ventriculomegaly: associated malformations and chromosomal defects. Fetal Diagn Ther1990; 5:5 –14[Medline]
45. Bloom SL, Bloom DD, DellaNebbia C, Martin LB, Lucas MJ, Twickler DM. The development outcome of children with antenatal mild isolated ventriculomegaly. Obstet Gynecol1997; 90:93 –97[Abstract]
46. Patel MD, Filly AL, Hersh DR, Goldstein RB. Isolated mild fetal cerebral ventriculomegaly: clinical course and outcome. Radiology 1994;192 : 759–764[Abstract/Free Full Text]
47. Pilu G, Falco P, Gabrielli S, Perolo A, Sandri F, Bovicelli L. The clinical significance of fetal isolated cerebral borderline ventriculomegaly: report of 31 cases and review of the literature. Ultrasound Obstet Gynecol 1999;14:320 –326[Medline]
48. Levine D, Barnes PD, Madsen JR, Abbott J, Mehta T, Edelman RR. Central nervous system abnormalities assessed with prenatal magnetic resonance imaging. Obstet Gynecol1999; 94:1011 –1019[Abstract/Free Full Text]
49. Wagenvoort AM, Bekker MN, Go AT, et al. Ultrafast scan magnetic resonance in prenatal diagnosis. Fetal Diagn Ther2000; 15:364 –372[Medline]
50. Simon EM, Goldstein RB, Coakley FV, et al. Fast MR imaging of fetal CNS anomalies in utero. AJNR2000; 21:1688 –1698[Abstract/Free Full Text]
51. de Laveaucoupet J, Audibert F, Guis F, et al. Fetal magnetic resonance imaging (MRI) of ischemic brain injury. Prenat Diagn 2001;21:729 –736[Medline]
52. Coakley FV, Hricak H, Filly RA, Barkovich AJ, Harrison MR. Complex fetal disorders: effect of MR imaging on management—preliminary clinical experience. Radiology1999; 213:691 –696[Abstract/Free Full Text]
53. d'Ercole C, Girard N, Boubli L, et al. Prenatal diagnosis of fetal cerebral abnormalities by ultrasonography and magnetic resonance imaging. Eur J Obstet Gynecol Reprod Biol1993; 50:177 –184[Medline]
54. Barkovich AJ. Pediatric neuroimaging, 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2000:252 –264
55. Aicardi J, Chevrie JJ, Baraton J. Agenesis of the corpus callosum. In: Vinken PJ, Bruyn GW, Klawans HL, eds. Handbook of clinical neurology. Amsterdam, The Netherlands: Elsevier,1987 : 149–173
56. Lacey DJ. Agenesis of the corpus callosum. Am J Dis Child 1985;139:953 –955[Medline]
57. Goodyear PW, Bannister CM, Russell S, Rimmer S. Outcome in prenatally diagnosed fetal agenesis of the corpus callosum. Fetal Diagn Ther 2001; 16:139 –145[Medline]
58. dos Santos AC, Midleton SR, Fonseca RL, dos Santos SR, Llerena JC Jr, Vargas FR. Clinical, neuroimaging and cytogenetic findings in 20 patients with corpus callosum dysgenesis. Arq Neuropsiquiatr2002; 60:382 –385[Medline]
59. Gupta JK, Lilford RJ. Assessment and management of fetal agenesis of the corpus callosum. Prenat Diagn1995; 15:301 –312[Medline]
60. Barkovich AJ, Norman D. Anomalies of the corpus callosum: correlation with further anomalies of the brain. AJNR1988; 9:493 –501
61. Vergani P, Ghidini A, Strobelt N, et al. Prognostic indicators in the prenatal diagnosis of agenesis of corpus callosum. Am J Obstet Gynecol 1994; 170:753 –758[Medline]
62. Blum A, Andre M, Droulle P, Husson S, Leheup B. Prenatal echographic diagnosis of corpus callosum agenesis. Genet Couns 1990;1:115 –126[Medline]
63. d'Ercole C, Girard N, Cravello L, et al. Prenatal diagnosis of fetal corpus callosum agenesis by ultrasonography and magnetic resonance imaging. Prenat Diagn1998; 18:247 –253[Medline]
64. Levine D, Barnes PD, Madsen JR, Li W, Edelman RR. Fetal central nervous system anomalies: MR imaging augments sonographic diagnosis. Radiology1997; 204:635 –642[Abstract/Free Full Text]
65. Kubik-Huch RA, Huisman TA, Wisser J, et al. Ultrafast MR imaging of the fetus. AJR 2000;174 :1599 –1606[Abstract/Free Full Text]
66. Dihn DH, Wright RM, Hanigan WC. The use of magnetic resonance imaging for the diagnosis of fetal intracranial anomalies. Child's Nerv Syst 1990;6:212 –215[Medline]
67. Golden JA, Rorke LB, Bruce DA. Dandy-Walker syndrome and associated anomalies. Pediatr Neurosci1987; 13:38 –44[Medline]
68. Bindal AK, Storrs BB, McLone DG. Management of Dandy-Walker syndrome. Pediatr Neurosurg1990–91; 16:163 –169
69. Maria BL, Zinreich SJ, Carson BC, Rosenbaum AE, Freeman JM. Dandy-Walker syndrome revisited. Pediatr Neurosci1987; 13:45 –51[Medline]
70. Bajoria R, Wee LY, Anwar S, Ward S. Outcome of twin pregnancies complicated by single intrauterine death in relation to vascular anatomy of the monochorionic placenta. Hum Reprod1999; 14:2124 –2130[Abstract/Free Full Text]
71. Haverkamp F, Lex C, Hanisch C, Fahnenstich H, Zerres K. Neurodevelopmental risks in twin-to-twin transfusion syndrome: preliminary findings. Eur J Paediatr Neurol2001; 5:21 –27
72. van Heteren CF, Nijhuis JG, Semmekrot BA, Mulders LG, van den Berg PP. Risk for surviving twin after fetal death of co-twin in twin-twin transfusion syndrome. Obstet Gynecol1998; 921:215 –219
73. Girard N, Raybaud C, Poncet M. In vivo MR study of brain maturation in normal fetuses. AJNR1995; 16:407 –413[Abstract]
74. Brisse H, Fallet C, Sebag G, Nessmann C, Blot P, Hassan M. Supratentorial parenchyma in the developing fetal brain: in vitro MR study with histologic comparison. AJNR1997; 18:1491 –1497[Abstract]
75. Guibaud L, Filiatrault D, Grignon A, Dubois J, Miron MC, Dalliere L. Fetal congenital diaphragmatic hernia: accuracy of sonography in the diagnosis and prediction of outcome after birth. AJR1996; 166:1195 –1202[Abstract/Free Full Text]
76. Adzick SN, Harrison MR, Glick PL, Nakayama DK, Manning FA, deLorimier AA. Diaphragmatic hernia in the fetus: prenatal diagnosis and outcome in 94 cases. J Pediatr Surg 1985;20 : 357–361[Medline]
77. Metkus AP, Filly RA, Stringer MD, Harrison MR, Adzick NS. Sonographic predictors of survival in fetal diaphragmatic hernia. J Pediatr Surg 1996;31:148 –152[Medline]
78. Leung JWT, Coakley FV, Hricak H, et al. Prenatal MRI of congenital diaphragmatic hernia. AJR2000; 174:1607 –1612[Free Full Text]
79. Coakley FV, Lopoo JB, Lu Y, et al. Volumetric assessment of normal and hypoplastic fetal lungs by prenatal single-shot RARE MR imaging. Radiology2000; 216:107 –111[Abstract/Free Full Text]
80. Paek BW, Coakley FV, Lu Y, et al. Prenatal evaluation of congenital diaphragmatic hernia by MR lung volumetry: preliminary experience. Radiology2001; 220:63 –67[Abstract/Free Full Text]
81. Adzick NS, Harrison MR, Crombleholme TM, Flake AW, Howell LJ. Fetal lung lesions: management and outcome. Am J Obstet Gynecol 1998;179:884 –889[Medline]
82. Dhingsa R, Coakley FV, Albanese CT, Filly RA, Goldstein R. Prenatal sonography and MR imaging of pulmonary sequestration. AJR 2003;180:433 –437[Free Full Text]
83. Farmer DL. Fetal surgery: a brief review. Pediatr Radiol 1998;28:409 –413[Medline]
84. Quinn TM, Adzick NS. Fetal surgery. Obstet Gynecol Clin North Am 1997;24:143 –157[Medline]
85. Hubbard AM, Crombleholme TM, Adzick NS. Prenatal MRI evaluation of giant neck masses in preparation for the fetal EXIT procedure. Am J Perinatol 1998;15:253 –257[Medline]
86. Hedrick MH, Ferro MM, Filly RA, Flake AW, Harrison MR, Adzick NS. Congenital high airway obstruction syndrome (CHAOS): a potential for perinatal intervention. J Pediatr Surg1994; 29:271 –274[Medline]

CiteULike

Complore