| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Obstetric Imaging |
1 From the Departments of Radiology (D.L., T.M.) and Obstetrics and Gynecology (D.L., G.W.), Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA 02215; Department of Radiology and the Advanced Fetal Care Center, Children’s Hospital, Boston, Mass (C.E.B., J.E.); and Department of Radiology, Hôpital Saint-Luc, Centre hospitalier de l’Universite de Montreal, Rue St-Denis, Montreal, Quebec, Canada (I.T.). Received May 20, 2002; revision requested July 16; final revision received November 25; accepted December 16. Supported in part by National Institutes of Health grant NS37945. Address correspondence to D.L. (e-mail: dlevine@caregroup.harvard.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: US and MR imaging data from 83 MR examinations of 74 fetuses with thoracic abnormalities and confirmatory US performed within 1 week before MR imaging were compared with respect to resulting changes in patient counseling and/or care. Lung parenchyma and lesion signal intensities and vascularity, airway, esophagus, and diaphragm appearances were reviewed retrospectively on MR images. Student t tests and analyses of variance were performed.
RESULTS: MR imaging yielded information additional to that acquired with US in 28 (38%) of 74 fetuses. The additional findings were confirmed in 19 of the 28 fetuses at postnatal follow-up; no follow-up data were available for the other nine fetuses. Thoracic MR information affected care with regard to six (8%) of 74 fetuses. Mean gestational age of 15 fetuses with lung signal intensity (SI) slightly lower than that of amniotic fluid (28.4 weeks ± 6.8 [SD]) at T2-weighted MR imaging was significantly older than that of 18 fetuses with intermediate SI (21.3 weeks ± 4.3) (P < .05). Mean SI of 13 congenital cystic adenomatoid malformations (CCAMs) and/or sequestrations (1.74 ± 1.05) at T2-weighted MR imaging was significantly higher than that of the normal lungs of 33 fetuses (2.63 ± .63) (P < .001). Among nine studies in which vessels were visualized in CCAMs and/or sequestrations, six involved a normal vascular branching pattern. Portions of the esophagus were seen in 31 (36%) of 85 fetuses. Nonvisualization of a major airway was not sufficient for diagnosis of pulmonary atresia. Visualization of a portion of the esophagus did not correlate with esophageal atresia. In all except one fetus, who had anhydramnios and pulmonary hypoplasia, and the fetuses with congenital diaphragmatic hernia, at least a portion of the diaphragm was visualized at MR imaging.
CONCLUSION: MR imaging yields information additional to that yielded with US in fetuses with thoracic abnormalities.
© RSNA, 2003
Index terms: Fetus, abnormalities, 856.8721, 856.875 • Fetus, cardiovascular system, 856.8751 • Fetus, MR, 856.121411, 856.121412, 856.121413, 856.121416 • Fetus, respiratory system, 856.8754, 856.8755, 856.8756, 856.8758 • Fetus, US, 856.129
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The MR and US examinations performed at Boston Children’s Hospital (n = 23) were ordered because of clinical indications in patients who were being referred for counseling regarding fetal chest masses. Verbal consent for review of the images and medical records was obtained, and the internal review board approved the use of the data for publication. Written informed consent for fetal MR imaging was obtained in all cases.
Eighty-five MR examinations were performed in 76 fetuses with thoracic abnormalities; there were a total of 75 pregnancies because thoracoomphalopagus conjoined twins were counted as two fetuses. Sixty-six pregnant women (60 with a single fetus and six with twin fetuses, including one set of thoracoomphalopagus conjoined twins) were examined once, eight patients (six with singleton pregnancy, one carrying twin fetuses, and one carrying triplet fetuses) were examined twice, and one patient with a singleton pregnancy was examined three times. Six co-twins and two co-triplets (without thoracic abnormality) of affected fetuses were examined for comparison of lung signal intensity only. Gestational age ranged from 16.0 to 38.0 weeks (mean, 25.2 weeks ± 6.3 [SD]).
The sequences varied according to individual MR imaging unit used, type of fetal anomaly, and gestational age; 1.5-T magnets (Vision, Symphony, or Quantum unit, Siemens, Erlangen, Germany; Signa LX, version 8.3, GE Medical Systems, Milwaukee, Wis) were used. Phased-array or surface coils were used for all but two patients. The patients were placed in a supine, feet-first position in the magnet to minimize claustrophobia. In the fetuses referred because of thoracic abnormalities, MR images were acquired in the fetal transverse, coronal, and sagittal planes with respect to the fetal chest. In the fetuses referred because of central nervous system abnormalities, MR images were obtained in planes with respect to the fetal brain but included the thorax in at least two imaging planes.
Fast T2-weighted MR images were acquired with the Siemens units (62 examinations) by using half-Fourier single-shot rapid acquisition with relaxation enhancement. A typical sequence performed with a Siemens unit involved the use of 4.2-msec echo spacing, an effective echo time of 60 msec, an echo train length of 72, one signal acquired, a 4-mm section thickness, a 30 x 30-cm field of view, and a 192 x 256 matrix. A 130° refocusing pulse was used to minimize the amount of radio-frequency power deposition. A 1-second delay between image acquisitions minimized the specific absorption rate. Thus, the time to acquire 13 sections in a single sequence was 17 seconds, and the time for each single image acquisition was 430 msec. For fetuses of early gestational age (<20 weeks) or in whom a small structure was being evaluated, the section thickness of at least one set of images was decreased to 3 mm. With use of the Signa LX magnet (23 examinations), the sequences varied during the course of the examination, with the final sequence (used in 17 acquisitions) being that of single-shot fast spin-echo T2-weighted imaging with a 98-msec effective echo time, a 62.5-kHz bandwidth, one signal acquired, a 4-mm section thickness, a 36 x 36-cm field of view, and a 256 x 256 matrix. All fetuses were examined with T2-weighted MR imaging.
T1-weighted MR imaging was performed by using a fast low-angle shot technique with the Siemens magnets. The fast low-angle shot sequence is a gradient-echo sequence that involves the use of a spoiler gradient to disperse residual transverse magnetization. The sequence used with the Siemens magnets was performed during breath holding by using 126/4 (repetition time msec/echo time msec), an 80° flip angle, a 24 x 32-cm field of view, a 116 x 256 matrix, a 5-mm section thickness, and one signal acquired, for an acquisition time of 12 seconds. The sequences used with the Signa LX unit varied during the course of the examination, with the final sequence being that of T1-weighted MR imaging performed with a 38-msec echo time, a 2,000-msec inversion time, a 62.5-kHz bandwidth, and a 4-mm section thickness. T1-weighted MR imaging was performed in cases in which the position of the fetal liver was of concern and in which there was a fetal mass that required characterization additional to that possible with fast T2-weighted MR imaging. Thirty fetuses were examined with T1-weighted MR imaging.
US Imaging
Patients were typically referred for US because of an anomaly seen on a previously obtained sonogram. When a referral US examination was performed outside of the two institutions, confirmatory US was performed at one of the two institutions. The confirmatory US (to confirm diagnosis of anomaly seen on previously obtained sonogram) diagnoses were compared with the MR imaging results. The US examinations conformed to American Institute of Ultrasound in Medicine and American College of Radiology guidelines (10) and included views of biometric features and the fetal anatomy; acquisition of additional images of the extremities, the right and left ventricular outflow tracts, and the face; and assessment of the vasculature of any lung lesions. In cases of suspected congenital diaphragmatic hernia (CDH), additional views were used to assess the position of the left lobe of the liver. Sonograms were obtained by using one of four US units (model 128XP or Sequoia, Acuson, Mountain View, Calif; model 3000 or 5000, Advanced Technology Laboratory, Bothell, Wash).
Data Acquisition and Analysis
The prospective US and MR findings were tabulated and compared with the fetal outcome data from pre- and postnatal echocardiography (43 patients), postnatal imaging (36 patients), surgery with or without pathologic analysis (24 patients), and autopsy (11 patients). One outcome examination was performed in each of 23 fetuses, and two or more outcome examinations were performed in 40 fetuses. In two fetuses, none of these examinations was performed and outcomes were determined by means of physical examination. In 11 fetuses, no outcome examinations were performed because the maternal patient declined permission for autopsy or the fetal specimen was macerated and insufficient for diagnosis.
The anomalies in each fetus were counted only once for the acquisition of summary data. The data acquired with US and MR imaging were compared by one author (D.L.). Changes in the referral diagnoses after US and MR imaging were tabulated. One of the authors (D.L., T.M., or C.E.B.) queried the clinicians about any changes in maternal-fetal care after the MR examination. The MR information was considered to have influenced care only in those cases in which there was a clear change in care due to data gleaned from MR imaging.
Two authors (D.L., C.E.B.) together assessed the appearances of normal and abnormal fetal lungs and fetal lung lesions with respect to signal intensity, vascular pattern, and visualization of the major airways, esophagus, and diaphragm. Interpretation differences were resolved by consensus.
Lung signal intensity on T2-weighted MR images was graded by using a five-point scale as follows: A grade of 1 meant the signal intensity was as high as that of fluid (for comparison, either amniotic fluid or cerebrospinal fluid at a similar distance from the coil); 2, the signal intensity was slightly lower than that of fluid; 3, the signal intensity was intermediate between that of fluid and that of muscle (for comparison, fetal skeletal muscles at a similar distance from the coil); 4, the signal intensity was slightly higher than that of muscle; and 5, the signal intensity was similar to that of muscle.
Lung signal intensity on T1-weighted MR images was graded as follows: A grade of 1 meant the signal intensity was as low as that of fluid; 2, the signal intensity was slightly higher than that of fluid but lower than that of muscle; 3, the signal intensity was similar to that of muscle; 4, the signal intensity was slightly higher than that of muscle; and 5, the signal intensity was very high.
Each lung and each lung lesion was assessed separately. Findings were tabulated with respect to the diagnosis. A fetus was considered to have normal lungs if there was no pulmonary lesion and no mediastinal shift. In the cases of twin fetuses in which one co-twin had a thoracic abnormality and the other did not, the signal intensity of the abnormal lungs was also compared with that of the normal lungs of the other fetus. These were the only data from the healthy co-twins that were used.
The vascular pattern in each lung and each lung lesion was described as follows: normal (with either normal spacing, a stretched appearance, or a collapsed appearance), abnormal (distorted vessels or abnormal branching), few vessels seen (but those seen appear normal), or no vessels seen.
Visualization of the major airways was described with respect to the portion of the airway seen—that is, the trachea, the carina, and/or the mainstem bronchi. Visualization of the esophagus was described as follows: seen in the cervical region, seen at the thoracic inlet, seen in portions of the chest, and/or seen at the gastroesophageal junction. The visualized diaphragm was described as well demarcated (even if not completely visualized) or poorly demarcated. The diaphragm was considered to have been visualized if a linear structure with a signal intensity slightly lower than that of the liver was seen in the expected location of the diaphragm (ie, between the liver and lungs or between the stomach and/or spleen and lungs) on T2-weighted MR imaging. In cases of CDH, the position of the left liver lobe at MR imaging was assessed and compared with the position of the left liver lobe at both US and surgery.
The signal intensities of two normal lungs in individual fetuses were compared with each other. The signal intensities of lung lesions were compared with those of the adjacent parenchyma and the contralateral lungs in individual fetuses. In the group of fetuses with normal lungs, the signal intensities of the lungs were compared with respect to gestational age. The signal intensities of normal lungs, compressed or collapsed lungs, hyperexpanded lungs, and lung lesions were compared. The vascular patterns of normal lungs were compared with respect to gestational age. The appearances of pulmonary vasculature, airways, the esophagus, and the diaphragm were compared with respect to gestational age and final diagnosis.
Statistical Analysis
The two-tailed Student t test (Excel software; Microsoft, Redmond, Wash) was used to compare the gestational ages of the fetuses with a lung signal intensity slightly lower than that of amniotic fluid at T2-weighted MR imaging with the gestational ages of the fetuses with intermediate lung signal intensity and to compare the mean signal intensity between lung lesions and normal lungs. Analysis of variance (Minitab software; Minitab, State College, Pa) was used to compare vascular pattern, airway visualization, and esophageal visualization with respect to lung signal intensity and gestational age. P < .05 was considered to indicate a significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Comparison of US and MR Imaging Findings
Confirmatory US findings led to changed or clarified thoracic diagnoses for 19 (26%) of 74 fetuses. For 28 (38%) of 74 fetuses, MR imaging findings either led to a change in diagnosis or represented information additional to that acquired with confirmatory US (also referred to as additional information or additional findings) (Table 1). In 19 of these 28 fetuses, postnatal follow-up results confirmed the prenatal additional findings. For nine fetuses, either no follow-up was performed or the follow-up data were insufficient for verifying the findings. Included in the 28 fetuses with additional MR imaging findings were four fetuses in whom the additional findings were in the central nervous system and not the thorax. The fetal thorax information yielded from MR imaging was believed by the referring clinician to have affected care for six (8%) of 74 fetuses. In all cases, except those in which there was poor assessment of congenital heart disease, MR imaging enabled detailed assessment of the thorax (Table 2). However, MR imaging yielded additional information in the thorax and abdomen in three cases of heterotaxy (Fig 1). In one case of conjoined twins, US better depicted the details of the shared vasculature, but MR imaging better depicted the overall appearance of the fetuses.
|
|
|
|
|
|
|
Eighteen fetuses had CDH and underwent 20 MR examinations. The mean signal intensity of the lungs in the fetuses with CDH was not significantly different from that of the normal lungs. In the fetuses with CDH, the signal intensity of the lungs in the older-gestational-age fetuses was higher than that in the younger-gestational-age fetuses (P < .05). The signal intensity of the lungs in one fetus with CDH, who was one of a set of triplets, was lower than that in the two co-triplets at two MR examinations performed at 20 and 26 weeks gestation.
Details about the signal intensities of the lungs and lung lesions in the fetuses at T1-weighted MR imaging are summarized in Table 4. The normal lungs and collapsed or compressed lungs were of higher signal intensity than the CCAMs and/or sequestrations (P < .05).
|
|
|
Major airways.—At T2-weighted MR imaging, the trachea, carina, and both mainstem bronchi were seen in their entirety in 29 (34%) (mean gestational age of fetuses, 25.8 weeks ± 5.7), portions of the airways were seen in 28 (33%) (mean gestational age of fetuses, 24.7 weeks ± 6.9), and no portions of the airways were seen in 18 (21%) (mean gestational age of fetuses, 24.9 weeks ± 6.8) of 85 examinations. There was no significant difference in gestational age when the three groups were compared. The extent of airway visualization with respect to the final diagnoses is summarized in Table 7. The airways were best visualized in the cases in which the lungs were not hypoplastic or compressed. With regard to five cases in which there was a mediastinal shift of unclear origin (ie, no CDH or mass), the ipsilateral mainstem bronchus was not seen in three and no pulmonary vessels were seen in two cases; however, lung tissue was visualized at all examinations.
|
|
The position of at least a portion of the left lobe of the liver was above the diaphragm in 13 (72%) of 18 fetuses with CDH. The entire liver was below the diaphragm in five (28%) of 18 fetuses. The position of the liver was equally well visualized with US and MR imaging, and an accuracy of 100% for depiction of the liver position was confirmed at surgery.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The performance values in our study are much lower than those calculated by Hubbard et al (3), who found that of 18 fetuses with chest masses who were evaluated with US and fast MR imaging, nine were given an incorrect US-based diagnosis. Hubbard et al stated that the US examinations were performed "at several institutions" and that the overall image quality was good, but the incorrect diagnoses were due to a "misinterpretation of the sonographic findings" (3). We believe that our finding that MR imaging had a somewhat lower incremental benefit compared with US was the result of performing confirmatory US before MR imaging and using the confirmatory US finding as the diagnosis for comparison, because 19 diagnoses were changed at the time of confirmatory US. If we had not performed confirmatory US, the diagnoses for 13 of these fetuses (excluding those in whom confirmatory US depicted congenital heart disease, since this is not expected to be visualized at MR imaging) would have been changed after MR imaging.
The MR imaging performance values that we calculated are also somewhat lower than those determined in the Coakley et al study (1), in which MR imaging directly influenced the care of four (17%) of 24 fetuses who were evaluated for complex disorders. In an additional eight cases (33%) in that study, MR imaging provided supplementary findings but did not affect fetal care (1). The difference in results between the Coakley et al study and our investigation is attributable to differing patient populations: Our patient population was limited to fetuses with thoracic abnormalities. It is also important to note that the changes in care were subjective in this study. There were cases in which the patient’s understanding of the abnormality improved owing to MR imaging findings and thus aided in decision making. MR imaging findings also had subtle effects on clinician confidence in diagnoses that were difficult to assess.
In fetuses with CDH, the location of the left lobe of the liver is an important factor when counseling patients about the expected outcome, because isolated "liver-up" and "liver-down" CDHs are associated with reported mortality rates of 57% and 7%, respectively (12–14). In studies performed by Hubbard et al (3,15), MR imaging was better at depicting the location of the liver in the chest. However, in our study, there was 100% concordance between US and MR imaging determinations of liver position; this accuracy value was based on postnatal surgical findings. Such a high concordance rate will likely be secondary to the use of confirmatory US before MR imaging, with US images obtained specifically to determine the liver location in any fetus with CDH.
As expected, in our study MR imaging was less sensitive than US in the diagnosis of cardiac abnormalities. Since fetal MR images are not gated for fetal cardiac motion, cardiac chambers are not adequately assessed (16). The small outflow tracts also cannot be adequately evaluated with the MR sequences used in this study. However, we did find MR imaging to be helpful in better characterizing the cases of heterotaxy—for example, those requiring visualization of polysplenia.
We found that lung signal intensity was higher in the older-gestational-age fetuses with normal lungs. This is in concordance with the findings of Duncan et al (17), who reported higher T2-weighted MR imaging measurements in older-gestational-age fetuses. With respect to the lungs of the fetuses with space-occupying lesions, the lungs of the fetuses with CDH (who are expected to have associated lung hypoplasia) had a signal intensity similar to that of the normal lungs of the fetuses of a similar gestational age. This signal intensity was higher in older-gestational-age fetuses. Similarly, the lungs of a fetus with massive pericardial effusion were of normal signal intensity. These findings suggest that lungs may not have low signal intensity in cases of pulmonary hypoplasia.
Other authors have suggested that the signal intensity of the lungs can be used to assess for pulmonary hypoplasia. For example, Ikeda et al (2) reported differences in lung signal intensity between 33-week-gestation twins in whom one was affected with pulmonary hypoplasia. However, as shown in our study, this finding is not uniformly present in the second trimester: In three sets of twins in which one co-twin had pulmonary hypoplasia, the co-twins had similar lung signal intensities. It may be that the relatively low signal intensity of the lungs does not become apparent until the third trimester. Lung volume, which was not assessed in this study, is probably a better indicator of pulmonary hypoplasia (18,19) than subjectively assessed signal intensity alone.
We found that most CCAMs and sequestrations were of high signal intensity and that the signal intensities of these abnormalities were higher than those of the surrounding lung parenchyma and the normal lungs in fetuses of similar gestational age. However, the signal intensities of one CCAM and one sequestration, each in a third-trimester fetus, were low and lower than the signal intensities of normal lungs and typical CCAMs and sequestrations. We know that fetal lung lesions can shrink in utero, and it is likely that low signal intensity is a part of this regression process. This is an important observation since in previous studies of fetal lung lesions, CCAM and sequestration lesions were always described as having high signal intensity (3–5,9). Unlike Quinn et al (4), we were unable to distinguish CCAMs without cysts from sequestrations in any case, except one in which a feeding vessel off the aorta was visualized.
Another interesting finding in our study was the frequency with which portions of the esophagus were visualized. We visualized portions of the esophagus in 36% (n = 31) of the examinations. The only fetus in our study in whom a long segment of the esophagus was visualized was given a postnatal diagnosis of microgastria and gastroesophageal reflux. A portion of the esophagus—the upper part—was visualized in only one of three fetuses with esophageal atresia, and this esophageal region was not dilated. This finding is in contrast to that in a study performed by Langer et al (7), in which a series of 10 fetuses with a small or absent stomach at US were evaluated: A dilated upper region of the esophagus at MR imaging was always seen in association with esophageal atresia, and nonvisualization of the intrathoracic portion of the esophagus at MR imaging was sufficient to exclude the diagnosis of esophageal atresia.
We evaluated the airways and central pulmonary vasculature to determine whether nonvisualization of the mainstem bronchus in a fetus with a mediastinal shift but no mass was sufficient for the diagnosis of unilateral pulmonary agenesis, as suggested by Kalache et al (6). We agree that these findings should be absent in pulmonary agenesis but in our series observed the frequency with which the airways and central pulmonary vasculature can be poorly visualized at MR imaging in cases of an unexplained mediastinal shift.
MR imaging of fetal thoracic abnormalities yields information additional to that acquired with US. The additional information yielded at MR imaging frequently leads to changes in patient counseling but infrequently leads to changes in patient care. We found that performing confirmatory US lessens the incremental benefit of MR imaging. MR imaging yielded information additional to that acquired with US for 28 (38%) of 74 fetuses. However, MR information regarding the thorax affected care with respect to only six (8%) of the 74 fetuses.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, D.L.; study concepts, D.L., C.E.B.; study design, D.L.; literature research, D.L.; clinical studies, T.M., D.L., C.E.B., J.E., I.T.; experimental studies, T.M., D.L., C.E.B., I.T.; data acquisition, T.M., D.L., C.E.B., I.T.; data analysis/interpretation, D.L.; statistical analysis, D.L.; manuscript preparation and definition of intellectual content, D.L.; manuscript editing, D.L., C.E.B., I.T., T.M.; manuscript revision/review, T.M., D.L., C.E.B., J.E., I.T.; manuscript final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. N. Saleem Feasibility of MRI of the Fetal Heart with Balanced Steady-State Free Precession Sequence Along Fetal Body and Cardiac Planes Am. J. Roentgenol., October 1, 2008; 191(4): 1208 - 1215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Barth and E. Rubesova Fetal Magnetic Resonance Imaging: Anomalies of the Neck, Chest, and Abdomen NeoReviews, August 1, 2007; 8(8): e313 - e335. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. L. Ward, M. Nishino, H. Hatabu, J. A. Estroff, C. E. Barnewolt, H. A. Feldman, and D. Levine Fetal Lung Volume Measurements: Determination with MR Imaging--Effect of Various Factors. Radiology, July 1, 2006; 240(1): 187 - 193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zaretsky, R. Ramus, D. McIntire, K. Magee, and D. M. Twickler MRI Calculation of Lung Volumes to Predict Outcome in Fetuses with Genitourinary Abnormalities Am. J. Roentgenol., November 1, 2005; 185(5): 1328 - 1334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. Brewerton, R. S. Chari, Y. Liang, and R. Bhargava Fetal Lung-to-Liver Signal Intensity Ratio at MR Imaging: Development of a Normal Scale and Possible Role in Predicting Pulmonary Hypoplasia in Utero Radiology, June 1, 2005; 235(3): 1005 - 1010. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ankermann, H. C. Oppermann, S. Engler, I. Leuschner, and C. S. von Kaisenberg Congenital Masses of the Lung, Cystic Adenomatoid Malformation Versus Congenital Lobar Emphysema: Prenatal Diagnosis and Implications for Postnatal Treatment J. Ultrasound Med., October 1, 2004; 23(10): 1379 - 1384. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Grassi, V. Esposito, M. Scaglione, M. Cirillo, S. Cappabianca, G. Guglielmi, F. S. Sasso, and A. Rotondo Multi-Detector Row CT for Depicting Anatomic Features of Cephalothoracopagus Varieties: Revised Approach RadioGraphics, September 1, 2004; 24(5): e21 - e21. [Abstract] [Full Text]
|
|
|
|
|
|
M. C. Frates, A. J. Kumar, C. B. Benson, V. L. Ward, and C. M. Tempany Fetal Anomalies: Comparison of MR Imaging and US for Diagnosis Radiology, August 1, 2004; 232(2): 398 - 404. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
