RadioGraphics
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/rg.251045510
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow CME Test (opens in a new window)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lawler, L. P.
Right arrow Articles by Fishman, E. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lawler, L. P.
Right arrow Articles by Fishman, E. K.
Related Collections
Right arrow Computed Tomography
Right arrow Genitourinary Radiology
RadioGraphics 2005;25:121-134
© RSNA, 2005

EDUCATION EXHIBIT

Adult Ureteropelvic Junction Obstruction: Insights with Three-dimensional Multi–Detector Row CT1

Leo P. Lawler, MD, FRCR, Thomas W. Jarret, MD, Frank M. Corl, MS and Elliot K. Fishman, MD

1 From the Russell H. Morgan Department of Radiology and Radiological Science (L.P.L., F.M.C., E.K.F.) and the James Buchanan Urological Institute (T.W.J.), the Johns Hopkins Medical Institutions, 601 N Caroline St, Rm 3254, Baltimore, MD 21287-0801. Presented as an education exhibit at the 2003 RSNA Scientific Assembly. Received February 23, 2004; revision requested April 14 and received June 23; accepted July 28. All authors have no financial relationships to disclose. Address correspondence to L.P.L. (e-mail: llawler@jhmi.edu).


   Abstract
Top
 Abstract
LEARNING OBJECTIVES
 Introduction
 Acquisition and Postprocessing...
 Background of UPJO
 Interpretation of CT Images...
 Conclusions
 References
 
Ureteropelvic junction obstruction (UPJO) is a benign, congenital condition that remains an enigma in terms of both diagnosis and therapy. On the basis of a series of cases that were referred to the authors in their clinical practice, they found that the unprecedented quality and novel perspectives of multi–detector row computed tomography (CT) with two- and three-dimensional postprocessing allow a comprehensive, single-study assessment of the ureterovascular relationships in UPJO. This topic is important because the causative role of crossing vessels may be questioned on the basis of such studies, and the therapeutic approach may be altered by using precise anatomic images customized to the pathologic features of the individual patient. Although CT images can provide only circumstantial or supportive evidence of the pathophysiology of UPJO, they greatly facilitate therapeutic intervention when it is clinically indicated and may eliminate the need for it in select cases. Careful attention to the postprocessing of CT images may show that the crossing vasculature has no direct relationship to the transition point of the UPJO in many cases.

© RSNA, 2005


   LEARNING OBJECTIVES
Top
 Abstract
 LEARNING OBJECTIVES
Introduction
 Acquisition and Postprocessing...
 Background of UPJO
 Interpretation of CT Images...
 Conclusions
 References
 
After reading this article and taking the test, the reader will be able to:


   Introduction
Top
 Abstract
 LEARNING OBJECTIVES
 Introduction
Acquisition and Postprocessing...
 Background of UPJO
 Interpretation of CT Images...
 Conclusions
 References
 
Despite advances in morphologic and functional imaging of ureteropelvic junction obstruction (UPJO), controversies remain surrounding the etiologic and surgical significance of the anatomic relationship of the renal pelvis, ureter, and adjacent vessels. Whether UPJO is symptomatic or an asymptomatic finding, the predicted natural history is unclear. What is clear is that a precise anatomic assessment of the renal parenchyma, collecting system, and vascular pedicle is fundamental to surgical management, providing insight into the condition for each individual patient.

In this article, the topics of discussion include multi–detector row computed tomography (CT) acquisition and image postprocessing for UPJO assessment based on our experience evaluating a series of referred patients. The currently applied design and application of data acquisition and postprocessing are discussed with examples of all current multi–detector row CT three-dimensional (3D) tools for renal tract depiction. The background of UPJO pathophysiology and ureterovascular relationships is reviewed. An approach to the interpretation of the two-dimensional (2D) and 3D CT reformatted images is suggested. The unprecedented quality and novel perspectives of multi–detector row CT allow precise anatomic description of the complex features often present in the setting of suspected UPJO. The approach is illustrated by a series of case examples to show how postprocessed multi–detector row CT may suggest association without causation of crossing vasculature in some cases as well as demonstrate the presence of vessels that are causative. Scanning can help ensure the successful outcome of surgical repair when undertaken.


   Acquisition and Postprocessing of Multi–Detector Row CT Images
Top
 Abstract
 LEARNING OBJECTIVES
 Introduction
 Acquisition and Postprocessing...
Background of UPJO
 Interpretation of CT Images...
 Conclusions
 References
 
Data Acquisition
Small section widths and near-isotropic or isotropic (voxels equal in all dimensions) multi–detector row CT data sets are required for optimal imaging of the renal hilar anatomy (1). The radiation beam is collimated to 0.75–1.5-mm detector systems to produce 0.75–1.5-mm section widths in breath-hold studies with 50% overlap. Smaller section widths do have higher noise and may incur a higher radiation dose to compensate, and 1.5-mm section widths will suffice for most vessel detection and characterization. A larger detector array design facilitates smaller section widths. From the topogram, the z-axis field of view is set to cover the kidneys with superior and inferior extension to allow for any slight respiratory variation and ectopic origin of the vascular supply or drainage. Pelvic imaging is added to the z-axis coverage if CT urography is requested (ie, a 3D CT scan performed with delayed contrast material opacification of the urothelial tract to simulate the information gained from conventional excretory urography). No positive oral contrast material is administered, as its high attenuation may hinder subsequent 3D segmentation.

Noncontrast images are not routinely obtained for specific UPJO evaluation, and indeed they may be misleading when the ureter, pelvis, and vasculature cannot be discriminated. If management of UPJO also requires evaluation for renal stones, a nonenhanced study will likely have higher sensitivity for calculi less than 1 cm, although we find that most calcifications may be detected with good arterial phase imaging providing unopacified renal medullae and collecting systems of the tracts. Our usual UPJO imaging protocol involves arterial (cortical) and venous (nephrographic) phase contrast-enhanced imaging with approximately 30- and 60-second delays, respectively, with use of a contrast material injection rate of 3–3.5 mL/sec. The former phase is for arterial anatomy and is sensitive to asymmetry in perfusion. The latter phase is for venous anatomy (Fig 1) and may show asymmetry in excretion into the tubules. Opacification of the collecting system may require a 2–3-minute delay.



View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a.  (a) Coronal volume-rendered multi-detector row CT angiogram shows the renal arteries. Note the two right renal arteries (arrowheads), with the lower-pole segmental artery (a common variant) arising directly from the inferior aorta. Long arrow = left renal artery, short arrow = left renal vein. (b) Coronal volume-rendered image shows the renal arteries (arrowheads), left renal vein (short arrow), and left gonadal vein (long arrow). (c) Coronal volume-rendered multi-detector row CT angiogram, obtained with slab editing to remove foreground and background tissues, shows an early-branching left renal vein (arrow) anterior to the renal artery (arrowhead).

 


View larger version (188K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b.  (a) Coronal volume-rendered multi-detector row CT angiogram shows the renal arteries. Note the two right renal arteries (arrowheads), with the lower-pole segmental artery (a common variant) arising directly from the inferior aorta. Long arrow = left renal artery, short arrow = left renal vein. (b) Coronal volume-rendered image shows the renal arteries (arrowheads), left renal vein (short arrow), and left gonadal vein (long arrow). (c) Coronal volume-rendered multi-detector row CT angiogram, obtained with slab editing to remove foreground and background tissues, shows an early-branching left renal vein (arrow) anterior to the renal artery (arrowhead).

 


View larger version (174K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1c.  (a) Coronal volume-rendered multi-detector row CT angiogram shows the renal arteries. Note the two right renal arteries (arrowheads), with the lower-pole segmental artery (a common variant) arising directly from the inferior aorta. Long arrow = left renal artery, short arrow = left renal vein. (b) Coronal volume-rendered image shows the renal arteries (arrowheads), left renal vein (short arrow), and left gonadal vein (long arrow). (c) Coronal volume-rendered multi-detector row CT angiogram, obtained with slab editing to remove foreground and background tissues, shows an early-branching left renal vein (arrow) anterior to the renal artery (arrowhead).

 
Usually, the low-attenuation ureter may be discerned without contrast material opacification and is conspicuous against the opacified vasculature. When required, delayed images obtained at 3–4 minutes will show an opacified collecting system, ureter, and bladder, which can be used to generate supplemental CT urograms. It is true that with a very large and redundant extrarenal pelvis, the collecting system and ureter may not be fully opacified in some cases. Multi–detector row CT urography is superior to excretory urography in depicting the complex anterior and posterior anatomic relationships of UPJO to the renal vasculature, the renal position, and the ureter. Although excretory urograms still have greater line pair and urothelial resolution, this is not usually of consequence for assessment of benign UPJO. Large parapelvic cysts closely positioned to the ureter may be difficult to distinguish from UPJO at early-phase contrast-enhanced imaging, but they will fail to opacify at CT urography.

Data Display
Postprocessing Tools.— Multiplanar reformation (MPR) images represent a simple reordering of image voxels. They can provide much of the additional information required for assessment of UPJO and largely preserve the spatial and contrast resolution of the original high-quality multi–detector row CT data set. They are preferred by the author to assess asymmetric function or parenchymal thinning (Fig 2) and have been found superior in their depiction of the unopacified ureteropelvic junction and its exact relationship to the crossing vessel at the anatomic point of obstruction.



View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2.  Left UPJO and hydronephrosis in a 49-year-old woman. Coronal MPR image of the kidneys shows moderate hydronephrosis (L). Note how the image displays the mild asymmetric cortical thinning of the left kidney.

 
Maximum intensity projection (MIP), which displays the attenuation values above a certain defined threshold, is similar in principle to projectional angiography. With its high contrast, it has value in depicting vessel branch patterns and small intrarenal vessels (Fig 3) but does require a good contrast material bolus. However, it lacks the attenuation value range to display the unopacified ureteropelvic junction and the depth information required to assess the ureterovascular relationships of the UPJO.



View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3.  Coronal MIP image of a normal left kidney shows the left renal artery (long arrow) and intrarenal branch vessels, the left renal vein (short arrow), and the left gonadal vein (arrowhead). Note how the image depicts the small intrarenal vasculature.

 
Volume rendering requires a lot of computer power but is a 3D tool of high fidelity to the acquired data and can be employed to generate CT angiograms, urograms, and a 3D perspective for laparoscopic surgical planning. It allows images to be customized to the patient of interest through infinite planes, perspectives, and trapezoid segmentation. Trapezoids are histogram representations of the attenuation values present in the image that allow the user to separate individual anatomy by attenuation value, opacity, brightness, color, and other parameters. Since volume rendering preserves all attenuation values within a voxel, information on the vascular pedicle, parenchyma, and collecting system is also preserved in the final 3D image. The strength of this technique lies in the depth perception at the renal hilum, and it is employed to generate CT angiograms, urograms, and a 3D perspective for laparoscopic approach planning (2,3) (Fig 4).



View larger version (190K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4.  Coronal volume-rendered multi-detector row CT angiogram of a normal kidney shows the depth and 3D relationships of the renal artery (long arrow) and renal vein (short arrow), both proximally and at the hilum.

 
Viewing Planes.— Axial 3–5-mm reconstructed planar sections allow a comprehensive review of the kidneys and their ureterovascular relationships and usually suffice for assessment of the presence and severity of hydronephrosis and other unrelated pathologic conditions. Current near-isotropic multi–detector row CT data allow interpretation to be independent of the plane of acquisition, and although axial images will demonstrate renal rotation, precise assessment of the UPJO itself requires alternate viewing projections. An oblique coronal perspective in the plane of the vascular pedicle and kidney depicts branch vessel patterns as well as providing a better perspective of the severity of hydronephrosis, the symmetry of perfusion, and coronal-plane kidney rotation (Fig 5). After therapeutic intervention, such images also better depict stent position and map any segmental renal perfusion injury.



View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5.  Coronal volume-rendered multi-detector row CT angiogram shows a horseshoe kidney with four right renal arteries (white arrowheads) and three left renal arteries (black arrowheads, black arrow). Note the lateral rotation of the upper poles relative to the medial lower poles.

 
A key perspective that assimilates many of the anatomic relations is the "hilar clock-face view." This is an oblique sagittal MPR plane fashioned orthogonal to the renal hilar vessels as they surround the ureter and enter the hilum (Fig 6). Sliding MPR in this plane may follow the course of the vessels and depict their relationship to the collecting system and ureter to best effect. The ureteropelvic junction transition point site and its angulation are highly variable and usually require imaging planes customized to the anatomy of the particular patient under study.



View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6.  Hilar clock-face view. Oblique sagittal image obtained at the right renal hilum shows the radial distribution of the renal arteries and veins (arrows) as they enter the hilum around the exiting enhanced ureter.

 
CT Angiography, CT Urography, and Functional Studies.— CT angiography has eliminated the need for conventional renal arteriography to depict arterial anatomy, and it is superior for venous anatomic evaluation (4). Helical CT angiography has shown conventional angiography to be an imperfect standard of reference. It has demonstrated good surgical correlation and high accuracy for renal donor vascular assessment (1,5–9). Similarly high sensitivity and specificity have been shown for crossing vessel detection in UPJO (2,4,10–20), although CT angiography has not been used to date to question the etiologic role of the vasculature. We have been satisfied with its sensitivity in enumerating and characterizing the course and patency of the renal vasculature and have found that 3D multi–detector row CT offers new avenues for exploring etiologic relationships in UPJO. Specifically, one can demonstrate not just the presence of crossing vessels but whether they are at the same level as and in direct apposition to the focal site of ureteral obstruction.

The symmetry of parenchymal and ureteric opacification after contrast material administration and features of parenchymal loss are crude surrogate markers of renal function at CT. However, whole-kidney CT perfusion studies carry a radiation dose that is not justified and the processing software remains limited in applications for routine use in noncerebral organs. The more sensitive and specific nuclear medicine studies (eg, diethylenetriaminepentaacetic acid [DTPA] or mercaptoacetyltriglycine [MAG3]) remain the standard in objective, quantitative measurement of relative function or obstruction of the kidneys. A single, selective-injection conventional arteriogram remains of value when it is important to quantify the amount of tissue perfused by a crossing or branch vessel.


   Background of UPJO
Top
 Abstract
 LEARNING OBJECTIVES
 Introduction
 Acquisition and Postprocessing...
 Background of UPJO
Interpretation of CT Images...
 Conclusions
 References
 
UPJO is defined as functional or anatomic obstruction to urine flow from the renal pelvis into the ureter at their anatomic junction, which, if left untreated, results in symptoms, renal damage, or both (21,22). UPJO generally implies a congenital partial proximal ureteric obstruction detected in utero or in later life, although the exact cause and possible embryologic source of UPJO are not known at this time. Failure of recanalization and the presence of valves are not likely causes. However, the problem is more likely due to an intrinsic abnormality of collagen or muscle than to an extrinsic cause. Secondary UPJO strictures from iatrogenic causes, inflammation, or tumor are less common (21). Multicystic dysplastic kidney may represent the end result of complete UPJO, although the natural history and so-called renal destiny of partial UPJO remain unclear, and therapeutic intervention is largely based on symptoms or image-based evidence of asymmetric dysfunction or the morphologic change of hydronephrosis (21).

The Ureterovascular Tangle
The ureterovascular tangle is a term that embraces the renal pelvis, ureter, and adjacent vessels (both arterial and venous), all of which alone or in combination have been implicated as potential causes of UPJO (23–25). The nature of this anatomy is important to the urologic surgeon, as it will dictate a laparoscopic or open surgical approach or guide the endoluminal pyelotomy incision to minimize the risk of vascular injury.

The Vasculature.— In the context of UPJO, "crossing vessels" are those renal arteries or veins found in the region of the ureteric transition point. The normal renal arteries may be single or multiple and give rise to anterior and posterior branches. The right renal artery normally crosses posterior to the vena cava. Arteries crossing anterior to the vena cava have been implicated in UPJO (26). Most crossing vessels are renal arteries and are anterior. The anterior vessel branches supply the superior and middle renal segments with a lower-segment branch to the anterior and posterior lower pole (Figs 7, 8). The posterior branch arches over the renal pelvis to supply the smaller corresponding superior and middle posterior segments. All renal arteries are end arteries and thus should be sacrificed for UPJO treatment only after thorough consideration of their role in the process of obstruction. When crossing vessels are observed, it is tempting to implicate them in the etiology of UPJO, although in many cases they may be "innocent bystanders."



View larger version (91K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7.  Drawings show the vascular anatomy of the renal hilum and the anterior and posterior segmental divisions of renal perfusion. The common arterial variation implicated in UPJO—an inferior segmental artery arising directly from the aorta—is also shown.

 


View larger version (81K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8.  Drawing shows the anatomy of the renal hilum and the ureterovascular relationships. This perspective is useful in assessment of UPJO and may be reproduced by means of an oblique sagittal clock-face MPR image. Note also the segmental renal perfusion and the common site of insertion of the lower-pole segmental artery variant. A = anterior, P = posterior.

 
The normal embryologic kidney ascends from the pelvis, receiving a ladder-like supply of arteries from the aorta. As vessels are gained superiorly, those more inferiorly are shed. During its ascent, the kidney also rotates into its final adult anatomic position (21). Interruption of this process can lead to variant vascular anatomy or unfavorable ureterovascular relationships. Variant arterial and venous anatomy including variation of origin, branch pattern, and course is common (10%–30%) although largely not pathologic (1,5,7–9,13,27). Crossing vessels in UPJO may be arteries or veins. Supernumerary or accessory renal arteries are frequently not additional vessels but rather are normal segmental vessels of aberrant origin. The point of origin of such vessels may be located anywhere along the renal artery, aorta, or iliac arterial system (23). The closer the origin of crossing vessels is to the aorta and the further away it is from the renal hilum, the more likely they will be implicated in the pathogenesis of UPJO (Figs 7, 8). Renal veins are commonly multiple or have variant early branching patterns and may be retro- or circumaortic in location and anterior or posterior to the ureter (6,7,28).

The lower-pole segmental artery or vein in particular has been implicated in UPJO (23). It has been thought to cause or aggravate obstruction, complicate therapy, or limit its successful outcome. The distended renal pelvis secondary to UPJO is largely extrarenal and of necessity will balloon over a lower renal segmental vessel when a potential space is created between it and the adjacent main renal artery (Fig 9). The closer the origin of the lower segmental vessel is to the iliac bifurcation vessels, the larger the space created and the greater the potential for such ballooning to occur secondary to obstruction (Figs 7, 8, 10). MPR and volume-rendered 3D multi–detector row CT reveals that the pelvis herniates both anteriorly and posteriorly over this vessel. Our experience with 3D multi–detector row CT of UPJO would support suggestions that these vessels may not be etiologically related but may be in the correct place at the incorrect time, giving an appearance that they contribute to the transition point. Also in support of this, our experience reflects the literature, showing a high prevalence of nonobstructive crossing vessels in those with a normal ureteropelvic junction (Figs 1a, 5, 6) and non-UPJO obstruction, and of course UPJO has been documented in the absence of crossing vasculature (1,5–7,28) (Fig 11).



View larger version (77K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9a.  Drawings show left UPJO with hydronephrosis in which the dilated extrarenal pelvis and proximal ureter bow anteriorly (a) or posteriorly (b) over the lower-pole segmental artery. Note that, when there is significant hydronephrosis, the UPJO site is frequently more cephalad behind the overgrown pelvis. UPJ = ureteropelvic junction.

 


View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9b.  Drawings show left UPJO with hydronephrosis in which the dilated extrarenal pelvis and proximal ureter bow anteriorly (a) or posteriorly (b) over the lower-pole segmental artery. Note that, when there is significant hydronephrosis, the UPJO site is frequently more cephalad behind the overgrown pelvis. UPJ = ureteropelvic junction.

 


View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10a.  UPJO in a 24-year-old man. (a) Coronal MPR image shows the inverted teardrop shape of right UPJO with a dilated extrarenal pelvis (arrowhead) and a transition point in the region of the ureteropelvic junction (arrow). (b) Coronal MIP image shows two right renal arteries (arrowheads) with the dilated extrarenal pelvis in between (arrow). (c) Oblique sagittal hilar clock-face view shows the relationship of the renal vein and main renal artery (arrowhead), the dilated extrarenal pelvis (short arrow), and the variant lower-pole artery (long arrow). Note that the crossing vessel is above the point of caliber transition.

 


View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10b.  UPJO in a 24-year-old man. (a) Coronal MPR image shows the inverted teardrop shape of right UPJO with a dilated extrarenal pelvis (arrowhead) and a transition point in the region of the ureteropelvic junction (arrow). (b) Coronal MIP image shows two right renal arteries (arrowheads) with the dilated extrarenal pelvis in between (arrow). (c) Oblique sagittal hilar clock-face view shows the relationship of the renal vein and main renal artery (arrowhead), the dilated extrarenal pelvis (short arrow), and the variant lower-pole artery (long arrow). Note that the crossing vessel is above the point of caliber transition.

 


View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10c.  UPJO in a 24-year-old man. (a) Coronal MPR image shows the inverted teardrop shape of right UPJO with a dilated extrarenal pelvis (arrowhead) and a transition point in the region of the ureteropelvic junction (arrow). (b) Coronal MIP image shows two right renal arteries (arrowheads) with the dilated extrarenal pelvis in between (arrow). (c) Oblique sagittal hilar clock-face view shows the relationship of the renal vein and main renal artery (arrowhead), the dilated extrarenal pelvis (short arrow), and the variant lower-pole artery (long arrow). Note that the crossing vessel is above the point of caliber transition.

 


View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 11a.  Incidentally discovered UPJO in a 36-year-old man without crossing vessels at CT. (a) Axial planar image shows left hydronephrosis and a dilated extrarenal pelvis (arrow). (b) Coronal volume-rendered image shows the inverted teardrop shape of left UPJO, which tapers inferiorly (long arrow). The left renal artery and vein are noted (short arrows).

 


View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 11b.  Incidentally discovered UPJO in a 36-year-old man without crossing vessels at CT. (a) Axial planar image shows left hydronephrosis and a dilated extrarenal pelvis (arrow). (b) Coronal volume-rendered image shows the inverted teardrop shape of left UPJO, which tapers inferiorly (long arrow). The left renal artery and vein are noted (short arrows).

 
The Ureter.— The normal ureter inserts at the inferior aspect of the renal pelvis at an oblique angle and with gradual taper. It has been noted that in the setting of UPJO, the ureter tends to have a more cephalad or acute insertion (29). This may be a primary event or secondary to the renal upper-pole lateral rotation caused by the dilating extrarenal pelvis in the mobile kidney. Although this site of insertion may not be a precipitating event in UPJO, its acute angle of insertion may aggravate it by creating a flap-valve as the extrarenal pelvis dilates. With continued renal pelvis dilatation, a "pulley effect" on the ureter over the lower-pole segmental vessel has been thought to cause it to kink. The ureteral segment at the site of UPJO does have decreased peristalsis, hypertrophied muscle cells, and normal transitional epithelium. The changes are thought to be primary and may result in a functional abnormality of peristalsis and a physical abnormality of inadequate distensibility (21).

Therapy
Characteristic symptoms with morphologic evidence of UPJO may indicate a need for therapy. Such symptoms include intermittent loin pain after drinking large volumes of fluid or fluids with a diuretic effect. Asymptomatic UPJO may be treated if there is evidence of asymmetric function or deterioration in renal function or hydronephrosis. Three-dimensional CT has already shown value for current treatment options and may influence selective management choices of retrograde endopyelotomy or surgical dismembered pyeloplasty (either open or laparoscopic) (11,12, 30–34). In select cases, a small crossing vessel is sacrificed or vasculopexy is performed (34), as such vessels may lead to recurrent obstruction (12). Precise documentation of the crossing vasculature remains important to avoid the complications of hemorrhage, pseudoaneurysm, and fistula formation (34–36).


   Interpretation of CT Images of UPJO
Top
 Abstract
 LEARNING OBJECTIVES
 Introduction
 Acquisition and Postprocessing...
 Background of UPJO
 Interpretation of CT Images...
Conclusions
 References
 
In this section, we discuss applications of the principles presented earlier and the understanding of UPJO for interpretation of studies that uses all the information available in postprocessed multi–detector row CT examinations. A series of case examples illustrate the practical applications of the principles.

Renal Parenchyma, Malrotation, Hydronephrosis, and Vasculature
The affected kidney may be rotated in the axial plane with the hilum facing anteriorly and in the coronal plane with the upper pole deviated laterally, and the lack of perinephric stranding reflects the chronic state. The report should document asymmetry of corticomedullary and ureteropelvic opacification and any cortical thinning present, which give an indication of chronicity and renal impairment.

It is the characteristic pattern of hydronephrosis on coronal reformatted images that first leads to the suggestion of UPJO. It tends to involve the extrarenal collecting system more than the intrarenal portion. It assumes an inverted teardrop shape tapering to the point of transition with the normal ureter distal to the artery. The capacity of the extrarenal portion may limit caliceal dilatation, and a nonobstructive extrarenal pelvis may be differentiated by less abrupt transition at the ureteropelvic junction. The severity of hydronephrosis should be noted, as this does affect surgical outcome. One should document from hilar clock-face views whether the ureter balloons anteriorly or posteriorly over the crossing vessels, as each is possible. The region of transition may be deduced from review of axial planar images, but its exact site, length, and angulation are more accurately depicted with MPR or volume rendering customized to the patient under study.

Although these are not necessarily involved in the pathophysiology of UPJO, one should determine the number, branch pattern, origin, size, and course of crossing vessels that may affect therapy (12,36). Any significant atherosclerotic change or stenosis can be documented. Comprehensive interpretation of multi–detector row CT images in 2D and 3D should describe the exact superior-inferior and anterior-posterior relation to the ureteric site of UPJO. We estimate the vertical distance of the crossing vessel from the discrete point of transition and document whether the vessel is in direct apposition to this portion of the ureter and whether there is a fat plane present. One may be able to suggest that the vessel is not etiologically related while documenting its effect on the surgical approach. After therapy, as well as documenting possible improvement in hydronephrosis, CT studies should document any stent placement and any segmental cortical areas of poor perfusion, indicating a procedure complicated by vascular injury.

Review of Cases
Case 1: Evaluation of UPJO.— A 24-year-old man presented with multiple episodes of right flank ache that were worse with alcohol consumption. Symmetric function was documented with a technetium-99m DTPA study. This case demonstrates the value of MPR alone in laying out the UPJO transition point, grading the hydronephrosis, and showing preservation of renal parenchyma (Fig 10a) as well as demonstrating the variant arterial anatomy (Fig 10b, 10c). The hilar clock-face MPR view showed that the variant artery was not anatomically related to the site of obstruction (Fig 10c), and this was confirmed at surgery. The vessel was preserved, and the patient was treated with laparoscopic pyeloplasty.

Case 2: Evaluation of UPJO.— A 36-year-old man was incidentally found to have UPJO at CT (Fig 11). The axial (Fig 11a) and volume-rendered (Fig 11b) images showed the classic inverted teardrop shape of the hydronephrosis. There was mild asymmetry of enhancement, intrarenal hydronephrosis, and thinning of the left kidney parenchyma. No crossing vasculature was implicated, and the patient was treated conservatively.

Case 3: Evaluation of UPJO.— A 71-year-old woman was incidentally found to have hydronephrosis at ultrasound in 1992. Intravenous pyelography demonstrated no obstruction at that time. Subsequently, in 2002, a severe left renal obstruction was documented with Tc-99m DTPA scanning. This case shows how a single coronal volume-rendered image can reveal much of the ureterovascular tangle, demonstrating both variant arterial and venous anatomy (Fig 12a). The hilar clock-face view did show the direct anterior apposition of the vessels to the short-segment site of transition and the extrarenal pelvis ballooning anteriorly (Fig 12b). The patient was treated with laparoscopic pyeloplasty.



View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 12a.  UPJO in a 71-year-old woman. (a) Coronal volume-rendered CT image shows the main renal artery (long black arrow) and a variant lower segmental renal artery (short black arrow). The lower segmental artery crosses posteriorly to the UPJO. An early-branching retroaortic left renal vein is noted, with the lower-pole branch crossing anteriorly to the dilated pelvis in the region of the UPJO (white arrow). (b) Oblique sagittal hilar clock-face view shows a renal vein branch and the main renal artery above the dilated renal pelvis (short arrow). The early-branch lower-pole renal vein is noted below the dilated renal pelvis (long arrow). The distal normal ureter is seen (white arrowhead). Note how the dilated extrarenal pelvis balloons anteriorly (black arrowhead) in the potential space created between the vasculature above and below.

 


View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 12b.  UPJO in a 71-year-old woman. (a) Coronal volume-rendered CT image shows the main renal artery (long black arrow) and a variant lower segmental renal artery (short black arrow). The lower segmental artery crosses posteriorly to the UPJO. An early-branching retroaortic left renal vein is noted, with the lower-pole branch crossing anteriorly to the dilated pelvis in the region of the UPJO (white arrow). (b) Oblique sagittal hilar clock-face view shows a renal vein branch and the main renal artery above the dilated renal pelvis (short arrow). The early-branch lower-pole renal vein is noted below the dilated renal pelvis (long arrow). The distal normal ureter is seen (white arrowhead). Note how the dilated extrarenal pelvis balloons anteriorly (black arrowhead) in the potential space created between the vasculature above and below.

 
Case 4: Evaluation of UPJO.— A 54-year-old man was documented to have high-grade UPJO at excretory urography and Tc-99m DTPA studies in 2002. The coronal MPR image demonstrated the mild hydronephrosis (Fig 13a). The coronal volume-rendered images showed the ureteropelvic junction traversing the crossing vasculature (Fig 13b), and the MIP image demonstrated the crossing arteries and vein (Fig 13c). However, the hilar clock-face view suggested that, though posterior to the ureteropelvic junction, the variant crossing vasculature was remote anatomically from the transition point (Fig 13d). The patient was treated with laparoscopic pyeloplasty.



View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 13a.  UPJO in a 54-year-old man. (a) Coronal MPR image shows mild right hydronephrosis with a transition point in the region of the ureteropelvic junction (arrow). (b) Coronal volume-rendered view shows right UPJO. The dilated renal pelvis is noted to balloon anteriorly (black arrow) over a second lower-pole segmental vein draining directly to the inferior vena cava (white arrowhead). Black arrowhead = main renal vein, white arrow = distal ureter beyond the UPJO. (c) Coronal MIP image shows the right UPJO. The main renal artery (short arrow) and a variant lower-pole segmental artery (long arrow) are noted. (d) Oblique sagittal MPR image shows that the enlarged extrarenal pelvis balloons posteriorly over the lower-pole artery and vein (short arrow). Note that the vessels are above the transition zone of the UPJO (long arrow).

 


View larger version (182K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 13b.  UPJO in a 54-year-old man. (a) Coronal MPR image shows mild right hydronephrosis with a transition point in the region of the ureteropelvic junction (arrow). (b) Coronal volume-rendered view shows right UPJO. The dilated renal pelvis is noted to balloon anteriorly (black arrow) over a second lower-pole segmental vein draining directly to the inferior vena cava (white arrowhead). Black arrowhead = main renal vein, white arrow = distal ureter beyond the UPJO. (c) Coronal MIP image shows the right UPJO. The main renal artery (short arrow) and a variant lower-pole segmental artery (long arrow) are noted. (d) Oblique sagittal MPR image shows that the enlarged extrarenal pelvis balloons posteriorly over the lower-pole artery and vein (short arrow). Note that the vessels are above the transition zone of the UPJO (long arrow).

 


View larger version (176K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 13c.  UPJO in a 54-year-old man. (a) Coronal MPR image shows mild right hydronephrosis with a transition point in the region of the ureteropelvic junction (arrow). (b) Coronal volume-rendered view shows right UPJO. The dilated renal pelvis is noted to balloon anteriorly (black arrow) over a second lower-pole segmental vein draining directly to the inferior vena cava (white arrowhead). Black arrowhead = main renal vein, white arrow = distal ureter beyond the UPJO. (c) Coronal MIP image shows the right UPJO. The main renal artery (short arrow) and a variant lower-pole segmental artery (long arrow) are noted. (d) Oblique sagittal MPR image shows that the enlarged extrarenal pelvis balloons posteriorly over the lower-pole artery and vein (short arrow). Note that the vessels are above the transition zone of the UPJO (long arrow).

 


View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 13d.  UPJO in a 54-year-old man. (a) Coronal MPR image shows mild right hydronephrosis with a transition point in the region of the ureteropelvic junction (arrow). (b) Coronal volume-rendered view shows right UPJO. The dilated renal pelvis is noted to balloon anteriorly (black arrow) over a second lower-pole segmental vein draining directly to the inferior vena cava (white arrowhead). Black arrowhead = main renal vein, white arrow = distal ureter beyond the UPJO. (c) Coronal MIP image shows the right UPJO. The main renal artery (short arrow) and a variant lower-pole segmental artery (long arrow) are noted. (d) Oblique sagittal MPR image shows that the enlarged extrarenal pelvis balloons posteriorly over the lower-pole artery and vein (short arrow). Note that the vessels are above the transition zone of the UPJO (long arrow).

 
Case 5: Evaluation of UPJO.— A 49-year-old patient with flank pain and hematuria was shown to have UPJO at excretory urography. This case shows how the CT study may reveal more than just the UPJO (Fig 14). Variant arterial anatomy was shown with volume rendering (Fig 14a), though the coronal (Fig 14b) and hilar clock-face (Fig 14c) MPR images showed the UPJO site to be above the level of the crossing vasculature. Note too how customized projections are required to show the S-shaped curvature of the UPJO (Fig 14b). The soft tissue sensitivity of CT also allowed an associated transitional cell carcinoma to be revealed (Fig 14b). The patient was treated with nephrectomy, and pathologic analysis revealed a transitional papillary cell carcinoma.



View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 14a.  UPJO in a 49-year-old patient. The presence of a soft tissue mass in the renal pelvis was suggested at retrograde pyelography. (a) Coronal volume-rendered view shows three right renal arteries (arrows). (b) Coronal MPR image shows a curved UPJO (long arrow) above the level of the variant lower-pole renal artery (short arrow). Arrowhead = soft tissue mass in the superior renal pelvis. (c) Oblique coronal hilar clock-face view shows a dilated extrarenal pelvis (black arrowhead) ballooning anteriorly over lower-pole variant vessels (short arrow). Note that the transition zone of the UPJO is above the level of the variant vasculature (long arrow). White arrowhead = soft tissue mass in the superior renal pelvis. Right nephrectomy revealed a low-grade papillary carcinoma.

 


View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 14b.  UPJO in a 49-year-old patient. The presence of a soft tissue mass in the renal pelvis was suggested at retrograde pyelography. (a) Coronal volume-rendered view shows three right renal arteries (arrows). (b) Coronal MPR image shows a curved UPJO (long arrow) above the level of the variant lower-pole renal artery (short arrow). Arrowhead = soft tissue mass in the superior renal pelvis. (c) Oblique coronal hilar clock-face view shows a dilated extrarenal pelvis (black arrowhead) ballooning anteriorly over lower-pole variant vessels (short arrow). Note that the transition zone of the UPJO is above the level of the variant vasculature (long arrow). White arrowhead = soft tissue mass in the superior renal pelvis. Right nephrectomy revealed a low-grade papillary carcinoma.

 


View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 14c.  UPJO in a 49-year-old patient. The presence of a soft tissue mass in the renal pelvis was suggested at retrograde pyelography. (a) Coronal volume-rendered view shows three right renal arteries (arrows). (b) Coronal MPR image shows a curved UPJO (long arrow) above the level of the variant lower-pole renal artery (short arrow). Arrowhead = soft tissue mass in the superior renal pelvis. (c) Oblique coronal hilar clock-face view shows a dilated extrarenal pelvis (black arrowhead) ballooning anteriorly over lower-pole variant vessels (short arrow). Note that the transition zone of the UPJO is above the level of the variant vasculature (long arrow). White arrowhead = soft tissue mass in the superior renal pelvis. Right nephrectomy revealed a low-grade papillary carcinoma.

 
Case 6: Follow-up after Pyeloplasty for UPJO.— A postpyeloplasty case shows the value of CT in posttherapy evaluation (Fig 15). This case shows that the coronal MPR image is favored for detection and quantification of any segmental perfusion abnormalities, residual hydronephrosis, or asymmetric dysfunction.



View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 15.  UPJO after pyeloplasty. Coronal MPR image obtained after pyeloplasty shows residual hydronephrosis (P). Note the segmental perfusion defect of the right inferior renal cortex (arrow) and the variant lower-pole artery (arrowhead).

 

   Conclusions
Top
 Abstract
 LEARNING OBJECTIVES
 Introduction
 Acquisition and Postprocessing...
 Background of UPJO
 Interpretation of CT Images...
 Conclusions
References
 
The origin and natural history of UPJO remain unclear, but precise anatomic studies of primary abnormalities and secondary effects may be achieved with multi–detector row CT by using patient-specific MPR and volume rendering postprocessing for CT angiography and CT urography. As well as documenting the presence of crossing vasculature, 3D multi–detector row CT now allows questioning of causal relationships. CT images can provide only circumstantial or supportive evidence of the pathophysiology of UPJO, and alone they do not make the case for surgical intervention. However, they do greatly facilitate therapeutic intervention when it is clinically indicated and may eliminate the need for it in select cases. Careful attention to postprocessing of images may show that crossing vasculature bears no direct relation to the point of transition in many cases.


   Footnotes
 
Abbreviations: DTPA = diethylenetriaminepentaacetic acid, MIP = maximum intensity projection, MPR = multiplanar reformation, 3D = three-dimensional, 2D = two-dimensional, UPJO = ureteropelvic junction obstruction


   References
Top
 Abstract
 LEARNING OBJECTIVES
 Introduction
 Acquisition and Postprocessing...
 Background of UPJO
 Interpretation of CT Images...
 Conclusions
 References
 

  1. Kawamoto S, Montgomery RA, Lawler LP, Horton KM, Fishman EK. Multidetector CT angiography for preoperative evaluation of living laparoscopic kidney donors. AJR Am J Roentgenol 2003; 180:1633-1638.[Abstract/Free Full Text]
  2. Herts BR. Helical CT and CT angiography for the identification of crossing vessels at the ureteropelvic junction. Urol Clin North Am 1998; 25:259-269.[CrossRef][Medline]
  3. Stabile Ianora AA, Scardapane A, Chiumarullo L, Calbi R, Rotondo A, Angelelli G. Congenital stenosis of ureteropelvic junction: assessment with multislice CT. Radiol Med (Torino) 2003; 105:315-325.
  4. Rouviere O, Lyonnet D, Berger P, Pangaud C, Gelet A, Martin X. Ureteropelvic junction obstruction: use of helical CT for preoperative assessment—comparison with intraarterial angiography. Radiology 1999; 213:668-673.[Abstract/Free Full Text]
  5. Rubin GD, Alfrey EJ, Dake MD, et al. Assessment of living renal donors with spiral CT. Radiology 1995; 195:457-462.[Abstract/Free Full Text]
  6. Rydberg J, Kopecky KK, Tann M, et al. Evaluation of prospective living renal donors for laparoscopic nephrectomy with multisection CT: the marriage of minimally invasive imaging with minimally invasive surgery. RadioGraphics 2001; 21(spec no):S223-S236.[Abstract/Free Full Text]
  7. Pozniak MA, Balison DJ, Lee FT, Jr, Tambeaux RH, Uehling DT, Moon TD. CT angiography of potential renal transplant donors. RadioGraphics 1998; 18:565-587.[Abstract]
  8. Del Pizzo JJ, Sklar GN, You-Cheong JW, Levin B, Krebs T, Jacobs SC. Helical computerized tomography arteriography for evaluation of live renal donors undergoing laparoscopic nephrectomy. J Urol 1999; 162:31-34.[CrossRef][Medline]
  9. Cochran ST, Krasny RM, Danovitch GM, et al. Helical CT angiography for examination of living renal donors. AJR Am J Roentgenol 1997; 168:1569-1573.[Abstract/Free Full Text]
  10. Farres MT, Pedron P, Gattegno B, et al. Helical CT and 3D reconstruction of ureteropelvic junction obstruction: accuracy in detection of crossing vessels. J Comput Assist Tomogr 1998; 22:300-303.[CrossRef][Medline]
  11. Kumon H, Tsugawa M, Hashimoto H, Yasui K, Hiraki Y, Ohmori H. Impact of 3-dimensional helical computerized tomography on selection of operative methods for ureteropelvic junction obstruction. J Urol 1997; 158:1696-1700.[CrossRef][Medline]
  12. Van Cangh PJ, Wilmart JF, Opsomer RJ, Abi-Aad A, Wese FX, Lorge F. Long-term results and late recurrence after endoureteropyelotomy: a critical analysis of prognostic factors. J Urol 1994; 151:934-937.[Medline]
  13. Smith PA, Fishman EK. Three-dimensional CT angiography: renal applications. Semin Ultrasound CT MR 1998; 19:413-424.[CrossRef][Medline]
  14. Siegel CL, McDougall EM, Middleton WD, et al. Preoperative assessment of ureteropelvic junction obstruction with endoluminal sonography and helical CT. AJR Am J Roentgenol 1997; 168:623-626.[Abstract/Free Full Text]
  15. Khaira HS, Platt JF, Cohan RH, Wolf JS, Faerber GJ. Helical computed tomography for identification of crossing vessels in ureteropelvic junction obstruction: comparison with operative findings. Urology 2003; 62:35-39.[CrossRef][Medline]
  16. Mitsumori A, Yasui K, Akaki S, et al. Evaluation of crossing vessels in patients with ureteropelvic junction obstruction by means of helical CT. RadioGraphics 2000; 20:1383-1393; discussion 1393–1395.[Abstract/Free Full Text]
  17. Pfister C, Thoumas D, Simon I, Benozio M, Grise P. Value of helical CT scan in the preoperative assessment of the ureteropelvic junction syndrome. Prog Urol 1997; 7:594-599.[Medline]
  18. Quillin SP, Brink JA, Heiken JP, Siegel CL, McClennan BL, Clayman RV. Helical (spiral) CT angiography for identification of crossing vessels at the ureteropelvic junction. AJR Am J Roentgenol 1996; 166:1125-1130.[Abstract/Free Full Text]
  19. Wolf JS, Jr, Siegel CL, Brink JA, Clayman RV. Imaging for ureteropelvic junction obstruction in adults. J Endourol 1996; 10:93-104.[Medline]
  20. Lacey NA, Massouh H. Use of helical CT in assessment of crossing vessels in pelviureteric junction obstruction. Clin Radiol 2000; 55:212-216.[CrossRef][Medline]
  21. Park JM, Bloom DA. The pathophysiology of UPJ obstruction: current concepts. Urol Clin North Am 1998; 25:161-169.[CrossRef][Medline]
  22. Gupta M, Moore RG, Nadler RB, Chandhoke PJ, Das S. Symptomatic ureteropelvic junction (UPJ) obstruction. J Endourol 1999; 13:413-416.[Medline]
  23. Stephens FD. Ureterovascular hydronephrosis and the "aberrant" renal vessels. J Urol 1982; 128:984-987.[Medline]
  24. Sampaio FJ. The dilemma of the crossing vessel at the ureteropelvic junction: precise anatomic study. J Endourol 1996; 10:411-415.[Medline]
  25. Sampaio FJ. Vascular anatomy at the ureteropelvic junction. Urol Clin North Am 1998; 25:251-258.[CrossRef][Medline]
  26. Yeh BM, Coakley FV, Meng MV, Breiman RS, Stoller ML. Precaval right renal arteries: prevalence and morphologic associations at spiral CT. Radiology 2004; 230:429-433.[Abstract/Free Full Text]
  27. Smith PA, Ratner LE, Lynch FC, Corl FM, Fishman EK. Role of CT angiography in the preoperative evaluation for laparoscopic nephrectomy. RadioGraphics 1998; 18:589-601.[Abstract]
  28. Mathews R, Smith PA, Fishman EK, Marshall FF. Anomalies of the inferior vena cava and renal veins: embryologic and surgical considerations. Urology 1999; 53:873-880.[CrossRef][Medline]
  29. Koff SA. Pathophysiology of ureteropelvic junction obstruction: clinical and experimental observations. Urol Clin North Am 1990; 17:263-272.[Medline]
  30. Turk IA, Davis JW, Winkelmann B, et al. Laparoscopic dismembered pyeloplasty: the method of choice in the presence of an enlarged renal pelvis and crossing vessels. Eur Urol 2002; 42:268-275.[CrossRef][Medline]
  31. Conlin MJ. Results of selective management of ureteropelvic junction obstruction. J Endourol 2002; 16:233-236.[Medline]
  32. Deger S, Roigas J, Wille A, Giessing M, Schonberger B, Turk IA. Laparoscopic dismembered pyeloplasty with Anderson-Hynes technique. Urologe A 2003; 42:347-353. [German].[Medline]
  33. Martin X, Rouviere O. Radiologic evaluations affecting surgical technique in ureteropelvic junction obstruction. Curr Opin Urol 2001; 11:193-196.[CrossRef][Medline]
  34. Perlberg S, Pfau A. Management of ureteropelvic junction obstruction associated with lower polar vessels. Urology 1984; 23:13-18.[CrossRef][Medline]
  35. Hoffer FA, Lebowitz RL. Intermittent hydronephrosis: a unique feature of ureteropelvic junction obstruction caused by a crossing renal vessel. Radiology 1985; 156:655-658.[Abstract/Free Full Text]
  36. Walz J, Lecamus C, Lechevallier E, et al. Complications of "Acucise" balloon endopyelotomy. Prog Urol 2003; 13:39-45. [French].[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow CME Test (opens in a new window)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lawler, L. P.
Right arrow Articles by Fishman, E. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lawler, L. P.
Right arrow Articles by Fishman, E. K.
Related Collections
Right arrow Computed Tomography
Right arrow Genitourinary Radiology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOGRAPHICS RADIOLOGY RSNA JOURNALS ONLINE