| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Medical Imaging, Erasme University Hospital, 808 Route de
Lennik, Brussels 1070, Belgium.
2 Department of Pathology, Erasme University Hospital, Brussels 1070,
Belgium.
3 Department of Pediatrics, University Children's Hospital Queen Fabiola, 15 Av
JJ Crocq, Brussels 1020, Belgium.
4 Department of Pediatric Imaging, University Children's Hospital Queen Fabiola,
Brussels 1020, Belgium.
Received February 21, 2003; accepted after revision September 16, 2003.
Address correspondence to M. Cassart.
OBJECTIVE. The objectives of our study were to evaluate the contribution of adding MRI findings to inconclusive sonographic data when assessing fetal urinary tract anomalies and to determine how this addition may affect the management of pregnancy.
SUBJECTS AND METHODS. We prospectively used MRI to study 16 third-trimester fetuses in whom sonography suggested bilateral urinary tract anomalies but failed to provide a definite diagnosis. These anomalies included enlarged hyperechoic kidneys (n = 6), bilateral pelvicaliceal dilatation (n = 6), renal cystic lesions (n = 2), and renal agenesis associated with severe oligohydramnios (n = 2).
RESULTS. The addition of MRI to sonography modified the diagnosis in five fetuses. In a fetus with suspected bilateral ureteropelvic obstruction, the diagnosis of bilateral ureterohydronephrosis associated with reflux or ureterovesical junction obstruction was made. In a fetus with an enlarged bladder at 32 weeks' gestational age, a possible diagnosis of megacystic microcolon was excluded on the basis of the normal appearance of the colon. In two fetuses with enlarged hyperechoic kidneys, MRI showed localized medullary hyperintense lesions suggesting autosomal recessive polycystic kidney disease in one fetus and medullary cystic dysplasia in another fetus with Jeune's syndrome. In a patient with suspected unilateral renal agenesis, MRI showed bilateral agenesis. In four fetuses, the addition of MRI to sonography led to a diagnosis that modified the decision to continue or terminate the pregnancy.
CONCLUSION. MRI can accurately show many urinary tract anomalies in third-trimester fetuses. It may be a useful complementary tool in the assessment of bilateral urinary tract anomalies of fetuses, particularly in cases with inconclusive sonographic findings.
Urinary tract malformations of the fetus are commonly detected on obstetric sonography. Their incidence varies from 0.1% to 1% of all pregnancies. Uropathies correspond to 30–50% of all structural abnormalities found at birth [1–3]. The spectrum of malformations is wide, and their prognosis is significantly poorer in fetuses with bilateral lesions and decreased volume of amniotic fluid. Antenatal diagnosis is primarily based on sonographic findings, yet the fetal anatomy may be difficult to analyze in some patients because of maternal obesity or the presence of oligohydramnios, which is commonly associated with urinary tract malformations [4, 5].
The recent development of fast sequences has extended the use of MRI to antenatal diagnosis. In this context, MRI has been validated as an efficient technique to evaluate equivocal fetal sonographic findings [6, 7], especially for neurologic anomalies, but its contribution to the diagnosis of fetal uropathies, to our knowledge, has not been clearly shown. This pilot study was undertaken to determine the potential contribution of adding MRI to sonography in the diagnosis of fetal urinary tract anomalies in third-trimester fetuses. More specifically, our aims were to assess the role of MRI as a complementary diagnostic tool in the presence of equivocal or inconclusive sonographic data and to determine how adding it to sonography might affect the decision to continue or terminate the pregnancy.
Subjects and Methods
Study Design
At our institution, all the cases of fetuses with positive sonographic
findings are discussed at a weekly multidisciplinary meeting that involves
obstetricians, radiologists, pediatric surgeons, and neonatologists. In the
context of this study, MRI was proposed whenever sonographic findings
suggested bilateral urinary tract anomalies but failed to provide a definite
diagnosis. Sonography and MRI examinations were always performed by the same
radiologist who used sonographic data to choose optimal MRI sequences and
interpret images. Results of the two examinations were discussed at a second
meeting of the multidisciplinary group; the group made a qualitative
evaluation of whether the addition of MRI to sonography provided a diagnosis
that differed from the one based on sonography alone.
Fetal sonographic and MRI findings were correlated with postnatal findings and were considered correct if the renal size and aspect (parenchyma and cavities) on antenatal imaging and the typical changes caused by the disease in the neonate were concordant. The postnatal workup (n = 12) performed during the first week of life consisted of sonography (n = 12), voiding cystourethrography (n = 8), and urographic MRI (n = 1) (3D T2-weighted images: TR/TE, 3,500/600). The technique selected was based on the abnormality involved. Postmortem findings were available in all four terminated pregnancies.
Patients
The study group consisted of 16 consecutive pregnant women referred between
May 1998 and January 2002 for prenatal MRI because of bilateral urinary tract
anomalies suspected in the fetus as a result of sonographic findings
(n = 14) or equivocal urinary tract findings due to poor sonographic
conditions related to oligohydramnios (n = 2).
Techniques
All sonographic examinations were performed by experienced radiologists
within the week preceding MRI. For MRI, verbal informed consent was obtained
from the parents. Sonographic examinations were performed on an SSD-5500
scanner (Aloka, Wallingford, CT) or a Power Vision 8000 scanner (Toshiba,
Yashimoto, Japan). The entire procedure was performed in accordance with the
recommendations of the local institutional review board.
MRI was performed using a 0.5-T magnet (n = 8) (Gyroscan NT-5, Philips, Eindhoven, The Netherlands) or a 1.5-T magnet (n = 8) (Gyroscan Intera Master, Philips). All acquisitions were obtained with a phased array body coil. The patients were not sedated and were comfortable in the supine position. Acquisitions consisted of 12–20 coronal, sagittal, and axial slices relative to the fetal abdomen using T2-weighted turbo spin-echo sequences (5,324/140) without respiratory triggering of the mother. Half-Fourier single-shot turbo spin-echo heavily T2-weighted acquisitions (duration per slice, 829 msec; TE, 80 msec; 60% partial Fourier acquisition) were also obtained. Finally, balanced fast-field-echo T2- and T1-weighted sequences (4.2/2.1; flip angle, 60°) were acquired at 1.5-T only (20 sections acquired within 19 sec). In all patients, T1-weighted images were acquired with a turbo field-echo sequence with inversion prepulse to enhance the T1 contrast (18/6.9; flip angle, 22°; inversion time, 1,354 msec; shot duration, 2,574 msec; 13 sections with 5-mm thickness; number of excitations, 6; duration, 4 min). The field of view was 340 mm to avoid wraparound artifacts, and slice thickness was 3–4 mm. The acquisition was completed in no more than 30 min.
Results
At the time of the MRI, gestational age ranged from 27 to 37 weeks (mean, 31 weeks). MRI examinations were well tolerated by the patients, and fetal movements did not alter image quality, although no maternal sedation was used.
In 11 of 16 fetuses, MRI and sonography showed concordant findings and provided a diagnosis that was subsequently confirmed by the postnatal workup or the postmortem examination: Four fetuses had bilateral dilatation seen on sonography and MRI; one male fetus had a urethral valve, which was suspected on antenatal sonography performed at 31 weeks' gestational age because of the presence of an enlarged posterior urethra. This diagnosis was confirmed by antenatal MRI and pathologic findings. Another fetus presented with right renal reflux and left ureteropelvic obstruction, which were confirmed on postnatal sonography and voiding cystourethrography. No cause for the hydronephrosis was found in two other patients who had persistent postnatal dilatation of the renal pelvis seen on sonography without reflux on voiding cystourethrography. They were diagnosed as having nonobstructive hydronephrosis [8].
Four other fetuses had enlarged kidneys (i.e., at least 3 SDs greater than normal fetal renal size derived from Cohen et al. [9]) showing a corticomedullary differentiation and a hyperechoic cortex. In the first fetus, sonography performed at 30 weeks' gestational age showed bilateral hyperechoic stripes, and similarly hypointense signal stripes were shown on MRI performed at 31 weeks' gestational age. These findings were consistent with the diagnosis of bilateral renal vein thrombosis, which was confirmed by the presence of vascular calcifications and multiple venous collateral vessels on postnatal sonography. No cause could be found before birth in the three other fetuses. At the age of 2 months, two of these patients developed cortical and medullary cystic lesions. The third patient showed persistent enlarged hyperechoic kidneys on postnatal sonography. All four patients had normal kidney function at birth.
Three fetuses showed a hydronephrotic kidney on sonography performed at 28, 27, and 33 weeks' gestational age. This finding was associated with a contralateral multicystic kidney in one fetus and with multiple adjacent pelvic cystic lesions in the other two. These findings were confirmed on antenatal MRI. Postnatal investigations confirmed renal multicystic dysplasia in the three neonates and associated with a contralateral ureterovesical reflux in two fetuses and with a contralateral obstructive dilatation in one.
MRI provided additional information that modified the sonographic diagnoses in five fetuses: In the first one, bilateral hydronephrosis was diagnosed at 36 weeks' gestational age. The sonographic appearance suggested a bilateral ureteropelvic junction obstruction (Fig. 1A), but MRI showed dilated ureters (Fig. 1B). These findings narrowed the differential diagnosis to a bilateral reflux or a bilateral ureterovesical junction obstruction. The latter diagnosis was eventually confirmed by postnatal voiding cystourethrography and MRI (Fig. 1C).
|
|
|
In the second fetus, findings of sonography performed at 32 weeks' gestational age revealed bilateral ureterohydronephrosis. The diagnoses of prune-belly syndrome and posterior urethral valve were considered unlikely because earlier sonographic findings were normal. Therefore, the most likely diagnosis was bilateral reflux, but megacystic microcolon was a possible diagnosis that could not be excluded on sonography because there was a huge bladder filling the entire fetal abdomen. As a result, the colon was difficult to visualize and to differentiate from the small intestine. MRI findings confirmed the presence of a bilateral hydronephrosis (Fig. 2A) and a dilated bladder (Fig. 2B), but more important, it allowed exclusion of an associated colonic anomaly (Fig. 2C). A bilateral high-grade reflux was confirmed after birth.
|
|
|
In two other fetuses, enlarged hyperechoic kidneys were diagnosed at 31 (Fig. 3A) and 26 weeks' gestational age. In both, MRI showed hyperintense lesions in the medulla, which best suited the features of ectatic collecting ducts (Fig. 3B); these lesions were not detected because of poor quality sonography due to maternal obesity and oligohydramnios. Subsequently, a diagnosis of autosomal recessive polycystic kidney disease seemed most likely for the first fetus and a diagnosis of Jeune's syndrome (based on family history) was proposed for the second one. These two diagnoses were confirmed at autopsy.
|
|
In the last patient with a severe oligohydramnios at 26 weeks' gestational age, sonographic findings suggested the presence of a hypoplastic solitary kidney (Fig. 4A) and a small filled bladder (Fig. 4B). MRI showed the complete absence of renal parenchyma and confirmed the presence of a bladder (Fig. 4C). These findings were confirmed at autopsy.
|
|
|
The additional information provided by MRI modified the management of the pregnancy in four patients. In three fetuses, the added information led the multidisciplinary group to propose termination of the pregnancy. This decision was based on either probable early renal failure associated with autosomal recessive polycystic kidney disease, nonviability of the neonate in the case of bilateral renal agenesis, or the poor prognosis for the neonate affected by Jeune's syndrome. On the other hand, MRI excluded megacystic microcolon in the fetus with bilateral ureterohydronephrosis and enlarged bladder (Fig. 2C). This information reassured the obstetrician and the parents, and it was decided to continue the pregnancy. In our study, extraabdominal anomalies unsuspected on sonography were not detected on MRI because this examination was focused on the abdomen.
Discussion
Most urinary tract malformations can be detected on obstetric sonography. These anomalies vary from minor to severe and potentially lethal. However, oligohydramnios and maternal obesity or both may limit the diagnostic accuracy of sonography, making additional imaging techniques potentially useful. MRI is currently accepted as a valuable technique for imaging the fetus [10, 11], and so far, no deleterious effect has been shown even with high magnetic fields [12–14].
Recent developments in MRI techniques have significantly improved fetal imaging. Fast imaging sequences used in our study (half-Fourier acquisition single-shot turbo spin echo, balanced fast field echo) have decreased the total duration of the acquisitions to about 30 min and have rendered maternal sedation unnecessary [7, 10]. Artifacts due to fetal movements are also minimized because acquisition of a single section with these sequences lasts for approximately 1 sec. The total duration of scanning is from 15 to 20 sec. The sequences used here (half-Fourier acquisition single-shot turbo spin echo and T2-weighted turbo spin echo) were heavily T2-weighted to provide good visualization of the renal pelvis, the bladder, dilated urinary tract, or cystic renal lesions. In addition, a mixed T2- and T1-weighted sequence (balanced fast field echo) was used to assess the integrity of the renal parenchyma. Furthermore, the use of a four-element phased array body coil rendered imaging less dependent on fetal position and provided an increased signal-to-noise ratio that could be reinvested into a better image resolution. Altogether, these technical improvements substantially increased the quality of images obtained in our study as compared with previous studies [15, 16].
In our study, findings of MRI modified the sonographic diagnoses in five fetuses. In one fetus, sonographic findings suggested bilateral ureteropelvic obstruction, but no dilated ureters could be seen because of huge pelvicaliceal dilatation that filled the upper part of the abdomen (Fig. 1A) and because of the iliac bones shadowing the fetal pelvis. On the other hand, the dilated ureters were easily identified on MRI (Fig. 1B), and this finding narrowed the differential diagnosis to bilateral ureterovesical junction obstruction versus high-grade bilateral reflux.
In a second fetus, MRI allowed exclusion of a potential megacystic microcolon (Figs. 2B and 2C), which is one of the differential diagnoses of fetal megacystis. Because this diagnosis carries a very poor prognosis, termination of pregnancy would have been considered an acceptable option [17]. On sonographic examination, the extremely enlarged bladder prevented proper visualization of the fetal bowel. In a healthy fetus in the third trimester, the colon and small bowel are easily identified on MRI: the colon appears hyperintense on T1-weighted sequences and the small bowel appears hyperintense on T2-weighted sequences because the former is filled with meconium and the latter with fluid [18]. In our patient, the colon appeared normal on MRI sequences; therefore, the diagnosis of microcolon was excluded. This diagnosis could hardly be made or excluded on MRI before the third trimester because the colon is not filled with meconium at this early stage of pregnancy.
In two other fetuses with enlarged hyperechoic kidneys, the medullary lesions were not clearly detected on sonography (Fig. 3A) because of poor scanning conditions; in comparison, the better contrast resolution provided by MRI on T2-weighted sequences individualized the highly hyperintense lesions in the medulla, which were considered to be ectatic collecting ducts (Fig. 3B). In the fetus with a familial history of Jeune's syndrome, the presence of medullary hyperintense lesions confirmed transmission of the disease; at 26 weeks' gestational age, the characteristic skeletal anomalies of the chest and long bones were not present.
In a previous study, Nishi [13] found that MRI is an appropriate diagnostic technique in cases of recessive polycystic kidney disease because it shows enlarged well-delineated kidneys with high signal intensity on T2-weighted sequences due to the high water content of the renal parenchyma. On the other hand, Poutamo et al. [16] concluded that in cystic kidneys, MRI does not provide more information than sonography.
In our patients, visualization of hyperintense lesions in the medulla helped to differentiate autosomal recessive polycystic kidney disease from other causes of hyperechoic kidneys, such as dominant polycystic kidney disease, congenital nephrotic and Bardet-Biedl syndromes, or cytomegalovirus infection. These diseases, unlike autosomal recessive polycystic kidney disease [19–21], do not include medullary lesions in utero. Because of the small number of patients studied here, the sensitivity and specificity of MRI in the evaluation of large echogenic kidneys should be confirmed by larger series.
In the fifth fetus, findings of sonography showed a small filled bladder and suggested the presence of a hypoplastic solitary kidney. MRI also showed this small bladder (Fig. 4C), but it showed the absence of renal parenchyma, which was confirmed at autopsy. At autopsy, a cystic cavity was isolated in the pelvis and was covered with urothelium; it did not correspond to urogenital sinus or cloaca. The presence of a bladder has been previously reported in patients with bilateral renal agenesis in whom only the trigone was lacking [22]. The unexpected finding of liquid in this bladder might suggest that a kidney had been present at an earlier stage of the pregnancy, had transiently secreted urine, and eventually atrophied (as may occur in multicystic kidneys) [23–25]. Alternatively, a transient patency of the urachus with reflux of amniotic fluid during early gestation might explain the presence of fluid in the pouch.
Most important, on four occasions, MRI associated with sonography played a pivotal role in the decision to continue or interrupt the pregnancy. In one fetus, MRI excluded a microcolon; as a result, it was decided to continue the pregnancy to term. In the three other fetuses (with either bilateral renal agenesis, autosomal recessive polycystic kidney disease, or Jeune's syndrome), the addition of MRI to sonography yielded a diagnosis associated with poor outcome and hence led to the proposition of terminating the pregnancy.
Legal issues may limit the practical relevance of the findings reported in our study. Medical termination of pregnancy during any trimester is authorized under restricted conditions by Belgian law, but the legislation is more restrictive in many other countries and prohibits termination of pregnancy after the second trimester. In addition, this option may be rejected by the parents, regardless of the severity of the anomaly seen in the fetus. However, even when the decision to continue pregnancy is made, the precise diagnosis provided by combining MRI and sonography may help health care professionals prepare the parents for the neonatal outcome.
In conclusion, this study shows that in third-trimester fetuses, MRI may help to better characterize bilateral urinary tract anomalies when sonography is inconclusive. In selected cases, MRI combined with sonography may establish a more precise diagnosis and hence impact the decision to continue or terminate the pregnancy. Because this series included a small number of fetuses, our findings should be confirmed by a larger study.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Barseghyan, H. A. Jackson, R. Chmait, R. E. De Filippo, and D. A. Miller Complementary Roles of Sonography and Magnetic Resonance Imaging in the Assessment of Fetal Urinary Tract Anomalies J. Ultrasound Med., November 1, 2008; 27(11): 1563 - 1569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Farhataziz, J. E. Engels, R. M. Ramus, M. Zaretsky, and D. M. Twickler Fetal MRI of Urine and Meconium by Gestational Age for the Diagnosis of Genitourinary and Gastrointestinal Abnormalities Am. J. Roentgenol., June 1, 2005; 184(6): 1891 - 1897. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. P. Wood Fetal Urinary Tract Malformations AAP Grand Rounds, November 1, 2004; 12(5): 55 - 56. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
