| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Emory University School of Medicine, Atlanta,
GA.
2 Department of Radiology, Children's Healthcare of Atlanta, 1001 Johnson Ferry
Rd., Atlanta, GA 30342.
3 Department of Pediatric Urology, Emory University School of Medicine, Atlanta,
GA.
4 Department of Pediatric Urology, Children's Healthcare of Atlanta, Atlanta,
GA.
Received November 4, 2004;
accepted after revision February 24, 2005.
Address correspondence to J. D. Grattan-Smith.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Sixty-one studies were performed in 50 children with hydronephrosis but without hydroureter. Anatomic criteria assessed included degree of hydronephrosis, morphology of the renal pelvis, atrophy of medulla, swirling contrast material, fluid levels, and the presence of fetal folds and crossing vessels. Functional criteria included renal transit time, differential renal function, and time-intensity curves when available.
RESULTS. Thirty-one kidneys were classified as obstructed, 15 as equivocal, and 15 as nonobstructed. Obstructed systems had more marked hydronephrosis, more extensive medullary atrophy, more fluid levels, and more swirling contrast material. Fetal folds were seen in only the equivocal and nonobstructed groups. Crossing vessels were seen in all groups. Obstructed systems also showed greater functional derangement, decreased split renal function, and abnormal time-intensity curves.
CONCLUSION. MR urography provides both excellent anatomic and functional information in children with UPJ obstruction in a single test that does not use ionizing radiation. MR urography may lead to greater understanding of the pathophysiology of UPJ obstruction.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As a result of this variable outcome, management of UPJ obstruction in children is controversial, with some authors recommending early surgery and others advocating simple observation [8-10]. Most surgeons monitor hydronephrotic kidneys with sonography and use decreasing function or worsening hydronephrosis as an indicator that surgery is required. The problem with this approach is that some obstructed kidneys will deteriorate while under observation. Ideally, it would be better to identify and surgically correct the condition in those patients before nephron loss occurs. At least one study has suggested that pyeloplasty performed early in the natural course of the disease leads to much better outcomes [9].
Previous studies have shown that dynamic contrast-enhanced MR urography has several advantages in the evaluation of hydronephrosis in children because it combines both anatomic and functional information in a single test that does not use ionizing radiation [11-17]. The purpose of this article is to review our experience with MR urography in children with UPJ obstruction and to identify anatomic or functional parameters that may predict which children will benefit from surgery.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
Image Interpretation
Two observers jointly evaluated all studies using anatomic and functional
parameters. Data on which patients went to surgery were collected from the
referring urologists. Anatomic assessment included renal size and the degree
of dilatation of the collecting system and ureter according to the Society for
Fetal Urology (SFU) grading system
[15]. Renal pelvis morphology
was classified as intrarenal or extrarenal; extrarenal pelves that displaced
the lower pole of the kidney laterally were further sub-classified as
"large." Medullary pyramid atrophy or absence was noted (Figs.
1A,
1B, and
1C). Contrast material in the
collecting system was observed for swirling and fluid level formation (Figs.
2A,
2B, and
2C). Fetal folds (Figs.
3A, and
3B) and crossing vessels (Figs.
4A,
4B, and
4C) in relation to the
proximal ureter were also noted.
Functional evaluation began with a measurement of renal transit time, defined as the time between renal cortical enhancement and contrast excretion into the ureter at the level of the lower pole of the kidney. Renal transit times were considered to be normal if less than 245 sec, equivocal if greater than 245 but less than 590 sec, and obstructed if greater than 590 sec [16]. Split renal function was determined after contrast enhancement by calculating the relative volume of the parenchyma of each kidney. Signal intensity curves over time for the cortex and medulla were generated for the latter part of the study and were available in 29 kidneys. Six discrete traits of the time-intensity curves were assessed (Figs. 3B and 5B): peak cortical enhancement amplitude, peak medullary enhancement amplitude, the corticomedullary crossover point (where there is isointensity between the cortex and medulla), the presence of the distal tubular peak (DTP), and both slopes of the final excretory phases for the cortex and the medulla.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Thirty-one kidneys were classified as obstructed on the basis of renal transit time calculation [16]. All the hydronephrotic kidneys had morphologic findings of UPJ obstruction, most frequently atrophy of the medullary pyramids (n = 25, 81%) (Figs. 1A, 1B, and 1C), and development of a large, often pendulous extrarenal pelvis (n = 16, 52%). Dilatation in the obstructed group was greater than in the nonobstructed group, with an average SFU grade of 3.25. Twenty-five (81%) of the obstructed kidneys had either swirling contrast patterns or fluid levels or both in the collecting systems (Figs. 2A, 2B, and 2C). Obstructed kidneys were almost twice as likely to have an extrarenal (n = 20) as an intrarenal (n = 11) pelvis; these 20 kidneys made up more than three quarters of the extrarenal group (n = 26). Almost all (n = 16) of the 20 obstructed extrarenal pelves were classified as large, and all but one of the large extrarenal pelves (n = 17) were obstructed. Seven (23%) of the obstructed kidneys had crossing vessels. Fetal folds were not identified in the obstructed group.
Early in the study, functional assessment was limited to visual inspection of cortical and medullary enhancement for symmetry with the contralateral kidney. Renal cortical enhancement was symmetric in all cases. In the obstructed group, 13 kidneys had medullary atrophy so severe that medullary signal changes could not be evaluated, and in two cases motion during scanning precluded evaluation of medullary signal changes. Enhancement of the renal medulla could be evaluated in 16 kidneys in the obstructed group; it was delayed in eight of those and symmetric in the other eight.
Time-intensity curves were generated for 12 obstructed systems (Table 2), all of which were cases of unilateral hydronephrosis. In four kidneys, the medullary parenchymal loss was so severe that no discernable signal intensity changes could be detected and only the cortical changes could be evaluated. In only two cases did an obstructed kidney have normal time-intensity curves. The most frequent abnormal findings were a decrease in the medullary peak signal intensity (n = 8) and loss of the distal tubular peak (n = 9) (Figs. 5A, and 5B). Delay of the corticomedullary crossover point (Figs. 5A, and 5B) and decrease of the peak cortical signal intensity occurred with intermediate frequency (n = 4 and 8, respectively), and abnormal washout slopes in the cortex and medulla were least common (n = 4 and 5, respectively). A mild decrease of the peak medullary signal intensity correlated with preserved function (average differential renal function, 43%) and swirling contrast patterns (75%), but less so with fluid levels (25%) and pyramid loss (50%).
|
|
Surgery was performed in 16 children (14 pyeloplasties and 2 nephrectomies). Fifteen of these children had UPJs that were obstructed using the renal transit time calculation. One child had surgery on a nonobstructed system for recurrent severe abdominal pain due to a crossing vessel (Fig. 3B). This patient underwent left pyeloplasty, and the surgical finding was a constriction at the UPJ. Retrograde pyelograms were available for review in eight cases, all of which showed narrowing of the UPJ. Children who went to surgery had a slightly higher SFU grade (average, 3.6) than their counterparts with nonoperative but obstructed UPJs (average, 2.7). Fluid levels were much more common (n = 14) in surgical cases. Swirling contrast material was seen equally in both surgical and nonsurgical obstructed systems. Five (71%) of the seven obstructed kidneys with crossing vessels were treated operatively. Patients going to surgery tended to be older (3.3 vs 1.3 years) and have a lower split function (33.5% vs 45.1%). Time-intensity data were collected for nine of the surgical patients. On average, 4.3 of the six curve characteristics were abnormal (compared with 2.3 in the obstructed, nonsurgical group and 0 in the nonobstructed group [as discussed in the next paragraphs]). One child from the obstructed group who had surgical repair of his UPJ obstruction had normal time-intensity curves.
In three patients with obstructed kidneys who underwent surgery, postoperative scans were obtained that showed improved drainage, with renal transit times in the normal range. In all three, the degree of hydronephrosis improved; and in two, medullary pyramids could be identified. In these two patients, the corticomedullary crossover points became symmetric, the medullary peak signal intensity improved, and the distal convoluted tubular peak that was absent on the preoperative scans could be identified. Improvement was also noted in the excretory phase of the curves. In the other kidney, although morphologic improvement was seen, the medulla could not be identified.
Fifteen kidneys were categorized as nonobstructed. These kidneys had lower SFU grades (average, 1.4), preserved medullary pyramids (n = 15, 100%), and symmetric corticomedullary crossover points (n = 15, 100%) (Figs. 4A, 4B, and 4C). These kidneys tended to be larger than their contralateral controls (average differential renal function, 52.5%). Ten had an intrarenal pelvis and five had an extrarenal pelvis; none had a large extrarenal pelvis. One third had crossing vessels (n = 5) and another third (n = 5) had fetal folds in the proximal ureter. Except for one patient with a symptomatic crossing vessel, none of these patients with nonobstructed kidneys was operated on. Time-intensity curves were generated in eight of these patients and were uniformly normal except in two patients (both with crossing vessels) in whom the peak cortical signal intensity was elevated.
The 15 hydronephrotic kidneys in the equivocal category were intermediate in SFU grade (average, 1.9), had predominantly normal medullary pyramids (n = 11, 73%), had symmetric corticomedullary crossover points (n = 9, 60%), and almost exclusively had intrarenal pelves (n = 14, 93%). Crossing vessels were rare (n = 2) in this category. Time-intensity curves were generated for nine of these kidneys, with abnormal results in one that showed loss of the distal tubular peak and delay of the corticomedullary crossover point.
Crossing vessels were identified in 14 kidneys across all categories. These
kidneys were more likely to go to surgery (43% vs 26%) and to occur in older
patients (average age, 5.7 vs 2.2 years). These kidneys did not differ
significantly from other hydronephrotic kidneys in our population in SFU grade
(average, 2.3 vs 2.5), frequency of obstruction (n = 7, 50%), fluid
levels (n = 3, 21%), swirling contrast material (n = 5, 36%)
in the collecting system, large extrarenal pelvis (n = 4, 29%), or
male predominance (
3:1).
Fetal folds in the proximal ureters were diagnosed on the basis of a beaded or corkscrew appearance to the proximal ureter (Figs. 4A, 4B, and 4C). Fetal folds were seen in both nonobstructed and equivocal kidneys. Patients with fetal folds and hydronephrosis tended to be younger (average age, 3 months vs 3.6 years) and have a lower SFU grade (1.4 vs 2.7) than the others in this study. No kidney with associated fetal folds had an extrarenal pelvis, swirling contrast material, fluid levels in the collecting system, or transit time in the obstructed range. No child with fetal folds went to surgery. Time-intensity curves were generated in 10 patients with fetal folds and were uniformly normal (Fig. 3B).
Extrarenal and large extrarenal pelves occurred with greater frequency in the most poorly functioning kidneys. Only one third of kidneys with preserved function had extrarenal pelves, and none had large extrarenal pelves. Two thirds of impaired kidneys (30-40% split function) had extrarenal pelves (half of these large), and almost three fourths of poorly functioning kidneys (< 30% split function) had large extrarenal pelves. Time-intensity curves were generated in nine of the patients with large extrarenal pelves and were never normal, with medullary peak intensity decreased; corticomedullary crossover points and distal tubular peaks were abnormal in all but one and two patients, respectively.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR urography is similar to sonography in its ability to categorize the degree of hydronephrosis [16]. The volume of functioning renal parenchyma is easily identified, even in children with marked hydronephrosis. MR urography easily distinguishes cortical scarring from medullary atrophy. We showed that the renal pyramids and medulla atrophied primarily in UPJ obstruction. Our observations support the contention of those authors who say that poor split function preoperatively leads to poorer pyeloplasty results. Although some have suggested that the presence of pelvic dilatation may act as a capacitor and protect the medulla from the effects of increased pressure [17], our study has not borne out this hypothesis. Extensive pelvic dilatation occurred mostly in patients with the worst differential renal function.
MR urography can show the UPJ with transition in caliber to a normal ureter in most cases. UPJ obstruction related to persistence of fetal folds and crossing vessels can be convincingly shown. Previously, fetal folds were identified using retrograde urography or excretory urography. Fetal folds cause nonobstructive hydronephrosis, which improves with the growth of the child [17]. Our study shows that although fetal folds may cause hydronephrosis and be associated with a slightly delayed renal transit time, they are easily identified on MR urography and are not associated with impairment of renal function. If fetal folds are identified, the hydronephrosis is likely to resolve spontaneously, and simple follow-up with sonography is all that is required.
Because of the high temporal and spatial resolution with MR urography, images obtained immediately after the administration of a contrast agent reliably delineate the renal arteries. Preoperative identification of crossing vessels is particularly important to those surgeons who perform endopyelotomies. The pathophysiology and role of crossing vessels as a source of UPJ obstruction in children has been widely debated [7, 19, 20]. One author has suggested that crossing vessels tend to present later with pain, whereas other causes of UPJ obstruction tend to present earlier with asymptomatic hydronephrosis. Crossing vessels were identified in all categories in our study. We also showed that older patients tend to experience UPJ obstruction with crossing vessels (average age, 5.7 vs 2.2 years for other obstructed patients). Whether these crossing vessels are the direct primary cause or simply have a secondary, exacerbating role remains unclear; however, their presence in the setting of an obstructive pattern is a better predictor of surgery than any other finding, perhaps because they tend to produce symptoms. Furthermore, intermittent UPJ obstruction related to crossing vessels is well described [20-22]; this phenomenon is thought to explain the only nonobstructed system in our study that went to surgery (Figs. 4A, 4B, and 4C).
MR urography also showed secondary signs of UPJ obstruction, such as fluid levels and swirling contrast material in the renal collecting system, as previously identified by Teh et al. [14]. This feature is most easily displayed on maximum-intensity-projection images of the dynamic contrast-enhanced sequences. Although its significance is uncertain at this time, in our population swirling contrast material tended to occur in obstructed kidneys with lower SFU grades and higher differential renal function.
Individual kidney functional assessment is the most exciting aspect of MR urography. Previous studies have suggested that the time-intensity curves described in this article can be used to infer intrarenal dynamics [15-17]. Delay in the corticomedullary crossover point has been suggested to indicate increased intratubular pressure [13]. In our study, all children in the nonobstructed category had symmetric crossover points. Only one crossover point in the equivocal group was delayed. However, in the obstructed group with preserved renal function, 50% showed delayed corticomedullary crossover when compared with the normal contralateral kidney. The corticomedullary crossover was symmetric in the other half. In many children in the obstructed group, the corticomedullary crossover point could not be assessed because of marked medullary atrophy. Interestingly, in the two patients studied both pre- and postoperatively in whom corticomedullary crossover could be identified, the preoperative study showed delay. After surgery, with return of renal transit times to normal, the corticomedullary crossover points became symmetric. It seems logical that the presence of increased tubular pressure in the setting of obstruction would correlate with ongoing renal damage and suggest a need for more urgent pyeloplasty. We are currently evaluating the corticomedullary crossover point in children with obstruction and preserved function to determine whether this criterion has prognostic value.
In patients in the obstructed category, changes were also noted in the ability of the cortex and medulla to concentrate the contrast agent. These changes may reflect early manifestations of renal damage and provide more subtle indicators of progressive renal disease than are currently available. Mild decrease in the peak concentration of the medulla and loss of the distal convoluted tubular peak were the most common abnormalities detected.
Another observation of interest was the abnormally flat slope of the excretory portions of the cortical and medullary curves in kidneys with medullary loss and poor function and secondary signs of obstruction such as fluid levels. This may be the MR correlate of the increasingly dense nephrogram seen with acute obstruction on excretory urography.
MR urography has several limitations. First, because of its sensitivity to patient motion, all our patients younger than 7 years were sedated. The costs associated with MRI are greater than those associated with renal scintigraphy. The postprocessing requires dedicated software and technical expertise. Segmentation of the cortex and medulla can be time-consuming, especially in children with poorly functioning kidneys.
Despite these limitations, MR urography is a single-technique approach to pediatric UPJ obstruction that offers both anatomic and functional insights. Anatomic evaluation combined with renal transit time classification provides a reliable parameter for the identification of obstruction. Although other morphologic findings suggest varying stages of renal dysfunction, their significance in the absence of a gold standard is uncertain. The ability of MR urography to identify fetal folds and crossing vessels offers distinct advantages over other techniques. Individual renal functional assessment with attention to the peak medullary signal intensity, distal tubular peak, and corticomedullary crossover point seems to identify the earliest signs of functional derangement in obstructed systems. Whether these signs simply represent surrogates for obstruction in general or signal impending nephron damage is uncertain at this time and will require continued follow-up.
Because MR urography can combine superior anatomic and functional information in a single test that does not use ionizing radiation, it seems likely that it will become the primary technique in the evaluation of UPJ obstruction in children.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. G. Silverman, J. R. Leyendecker, and E. S. Amis Jr What Is the Current Role of CT Urography and MR Urography in the Evaluation of the Urinary Tract? Radiology, February 1, 2009; 250(2): 309 - 323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.-H. Vivier, M. Dolores, I. Gardin, P. Zhang, C. Petitjean, and J.-N. Dacher In Vitro Assessment of a 3D Segmentation Algorithm Based on the Belief Functions Theory in Calculating Renal Volumes by MRI Am. J. Roentgenol., September 1, 2008; 191(3): W127 - W134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Leyendecker, C. E. Barnes, and R. J. Zagoria MR Urography: Techniques and Clinical Applications RadioGraphics, January 1, 2008; 28(1): 23 - 46. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
