Congenital Ureteropelvic Junction (UPJ) Obstruction Imaging and Diagnosis

Updated: Oct 17, 2020
  • Author: John S Wiener, MD, FACS, FAAP; Chief Editor: Eugene C Lin, MD  more...
  • Print

Practice Essentials

The widespread practice of prenatal ultrasonography imaging has made ultrasonographic findings the primary presentation of congenital ureteropelvic junction (UPJ) obstruction. Routine prenatal assessment typically occurs at 16-20 weeks' gestation. On the basis of those findings, an appropriate series of in utero and postnatal studies is performed. [1, 2, 3, 4]  Radiologic imaging is crucial in diagnosing UPJ obstructionCongenital UPJ obstruction is the most common cause of upper urinary tract obstruction in children. [5, 6, 7]

By definition, the diagnosis of UPJ obstruction signifies functionally impaired transport of urine from the renal pelvis into the ureter. Because the increased renal pelvic pressure from obstruction may lead to progressive renal injury and embarrassment, correct diagnosis is clinically important. The impairment may be congenital or acquired in nature. This, along with the chronicity and severity of the condition, dictates the course of management. [1, 8, 9, 10, 11]

There are multiple criteria to define fetal hydronephrosis. The most commonly used standard defines hydronephrotic kidneys as those with an anteroposterior (AP) diameter at the renal pelvis of greater than 4 mm at a gestational age of less than 33 weeks and an AP diameter of greater than 7 mm at a gestational age of 33 weeks or older. [12] An abnormal initial ultrasonogram should be followed up with another ultrasonogram after 4 weeks in severe cases or after 33-34 weeks in mild to moderate cases.

(See the images below.)

Prenatal longitudinal ultrasonogram of the right k Prenatal longitudinal ultrasonogram of the right kidney demonstrates right hydronephrosis. The right kidney is enlarged compared with the left and measures 55.3 mm in length. The renal cortex appears to be thinned.
Prenatal longitudinal ultrasonogram of a normal le Prenatal longitudinal ultrasonogram of a normal left kidney. The renal pelvis is minimally split by a small amount of urine, which is anechoic. Renal length is 34.8 mm.
Prenatal transverse ultrasonograms of the abdomen. Prenatal transverse ultrasonograms of the abdomen. (The spine at the top of the image is causing the dark shadowing.) Transverse views of the kidneys demonstrate hydronephrosis involving the right kidney. The anteroposterior (AP) diameter of the right renal pelvis is 21.9 mm. A normal-appearing left kidney is also shown, with a normal-sized left renal pelvis (AP diameter = 3.7 mm).


Criteria for fetal hydronephrosis

The most common criteria used to grade fetal hydronephrosis are the Society of Fetal Urology (SFU) consensus guidelines, which are based on pelvic dilation and caliectasis [13, 14] . They include the following:

  • Grade 0 — Normal kidney

  • Grade 1 — Minimal pelvic dilation

  • Grade 2 — Greater pelvic dilation without caliectasis

  • Grade 3 — Pelviectasis and caliectasis without cortical thinning

  • Grade 4 — Hydronephrosis with cortical thinning

Grade 3-4 hydronephrosis is 88% sensitive and 95% specific for obstruction on diuretic renograms. However, because of interobserver and intraobserver variability in assessing the degree of dilatation in ultrasonography, many centers continue to rely on the AP diameter of the renal pelvis to diagnose hydronephrosis. The hydration status of patients naturally has a bearing on the measurement and assessment of dilatation in the collecting system, and any extreme outlying states should be duly noted.

In 2014, the Soceity of Fetal Urology, along with the American College of Radiology, the American Institute of Ultrasound in Medicine, the American Society of Pediatric Nephrology, the Society for Maternal Fetal Medicine, the Society for Pediatric Urology, the Society for Pediatric Radiology, and the Society of Radiologists in Ultrasound convened to develop a unified description of urinary tract dilation (UTD). [12] The UTD classification system utilizes additional factors to describe hydronephrosis, including ureteral and bladder abnormalites; stratifies based on gestional age and detection prenatally or post-natally; and offers follow-up schemata. Measurements are based on anterior-posterior renal pelvic dilation.

Prenatal presentation is as follows [12] :

  • A1 (low risk): 4–< 7 mm at 16–27 weeks' gestation or 7–< 10 mm at >28 weeks' gestation.
  • A2-3 (increased risk): >7 mm at 16–27 weeks' gestation or >10 mm at >28 weeks' gestation or with peripheral calyceal dilation, abnormal parenchymal thickness, abnormal ureters, abnormal bladder, or unexplained oligohydramnios.

Postnatal presentation is as follows [12] :

  • P1 (low risk): 10–< 15 mm after 48 hours of life.
  • P2 (intermediate risk): >15 mm or with peripheral calyceal dilation or abnormal ureters.
  • P3 (high risk): >15 mm or with peripheral calyceal dilation, abnormal parenchymal thickness, abnormal parenchymal appearance, abnormal ureters, or abnormal bladder.

Follow-up postnatal monitoring

Much controversy exists over the optimum timing of follow-up postnatal imaging. Some suggest delaying imaging at least 48 hours after birth to minimize false-negative findings, owing to the infant's relative state of dehydration and decreased glomerular filtration rate (GFR). Others have found no difference between early and delayed ultrasonography. This issue is of some practical significance, because an early ultrasonogram may be obtained before the infant is discharged home from the nursery, which helps prevent noncompliance with follow-up.

Typically, more severe dilation requires routine serial monitoring every 3-4 months for the first year of life; hydronephrosis of SFU grades 1 and 2 may be monitored with less frequency (every 6-9 months). For hydronephrosis of SFU grade 4, administering a diuretic nuclear renogram to assess renal function is usually delayed until 1 month of age to allow for the physiologic increase and stabilization of GFR. For SFU grade 3 hydronephrosis, follow-up imaging is less defined. The need for repeat nuclear renography is controversial; the decision is made largely on the basis of physician preference, ultrasonographic findings, or both. Definitive treatment is initiated with the finding of reduced or worsening renal function, worsening hydronephrosis, or both. Because reflux may cause hydronephrosis and coexists in 13-42% of cases of congenital UPJ obstruction, cystourethrography was previously recommended in all patients to assess vesicoureteral reflux (VUR). More recently, more selected use of cystourethroscopy has been utilized. [15]

Asymptomatic UPJ in older children and adults

Not uncommonly, asymptomatic UPJ obstruction is discovered in older children or adults when radiologic studies, such as ultrasonography, intravenous pyelography (IVP), computed tomography (CT) scanning, or magnetic resonance imaging (MRI), are performed for other reasons. Often, the initial examination is a CT scan, because CT has become the study of choice at many institutions for evaluating acute flank or abdominal pain. The use of intravenous contrast material—with nephrogenic and delayed excretion phases—during CT scanning may provide qualitative information regarding obstruction, but in general, CT should be avoided because of the inherent high radiation dose. When hydronephrosis is seen and UPJ obstruction is suspected, diuretic renography is more accurate than CT.

Intravenous pyelography was historically the primary study used to diagnose UPJ obstruction in adults, because it also provided anatomic and functional information. However, ultrasonography is preferred as the initial study in children because of its nonionizing and noninvasive nature. Once an obstruction is suspected on the basis of anatomic studies, the diagnosis of UPJ obstruction is made by means of diuretic renography.

Limitations of techniques

To diagnose ureteropelvic junction obstruction, anatomic and functional studies are necessary. As a standalone study, ultrasonography cannot definitively diagnose UPJ obstruction. Ultrasonography provides no information regarding physiologic or functional status, and in many cases, despite the presence of moderate or severe hydronephrosis, urine passage is unobstructed. Obstruction must be demonstrated by means of functional studies, such as diuretic nuclear renography. Even with nuclear renography, the definition of true obstruction is a subject of debate; some proponents rely on decreased differential renal function as an indication of obstruction, whereas others rely on a decrease in washout time. Many features of the infant kidney, including its immature state and its variable appearance (depending on volume status), may lead to inaccurate ultrasonographic findings.



Intravenous pyelography (IVP) has traditionally been the primary study for evaluating hydronephrosis in older children and adults. It provides functional and anatomic detail about the renal parenchyma and collecting system. IVP findings suggestive of ureteropelvic junction obstruction include marked dilatation of the renal calyces and pelvis, funneling down to a narrow beak end, with nonvisualization of the ipsilateral ureter (see the images below).

Anteroposterior intravenous pyelogram of the abdom Anteroposterior intravenous pyelogram of the abdomen in the early excretion phase demonstrates normal excretion of contrast material from the left kidney. Contrast material is beginning to accumulate in the dilated calyces of the right kidney, but it has not replaced the large amount of nonenhancing urine in the right renal pelvis. Therefore, the dilated right renal pelvis is not yet seen.
Anteroposterior intravenous pyelogram of the abdom Anteroposterior intravenous pyelogram of the abdomen later in the excretory phase. Contrast material fills the dilated calyces and pelvis of the right kidney. No contrast enhancement is seen in the right ureter on this or other images, suggesting ureteropelvic junction obstruction. The right kidney demonstrates relatively increased renal length in comparison with the left kidney; this is consistent with obstruction.

Delayed imaging is essential to maximize visualization of the urinary tract, because the higher volume of urine in a hydronephrotic pelvis dilutes the contrast medium and delays its passage into the ureter, even in the absence of functional obstruction. The administration of intravenous furosemide (0.5-1.0 mg/kg) during IVP may further assist in differentiating true UPJ obstruction from nonobstructive hydronephrosis.

Retrograde pyelography may provide the greatest degree of anatomic detail in delineating UPJ anatomy when results of antegrade studies (eg, IVP) are inconclusive or equivocal. However, retrograde pyelography is the most invasive study; cystoscopy and ureteral catheterization in children require general anesthesia. In UPJ obstruction, findings may include a kink of the proximal ureter near the UPJ, with a tight jet of contrast passing into the pelvis upon administration or the complete failure of contrast material to flow into the pelvis.

Although cystourethrography is traditionally an unreliable test for diagnosing UPJ obstruction itself, this test may be useful in the workup of UPJ obstruction in children because VUR can cause hydronephrosis. Cystourethrography may demonstrate VUR and other anomalies of the lower urinary tract that may alter renal function. About 10% of patients with UPJ obstruction also have VUR.

Cystographic signs suggestive of UPJ obstruction in the presence of reflux include the following: (1) poorer visualization of the pelvis in comparison with the ureter, which results from the dilution of contrast material by urine in the dilated pelvis, and (2) dilatation of the pelvis and calyces out of proportion to that of the ureter. Contrast-enhanced cystourethrography should be used instead of radionuclide cystography because the latter provides inadequate anatomic detail.

Degree of confidence

IVP can reliably demonstrate existing hydronephrosis and reveal the site of obstruction, provided that function of the ipsilateral kidney is adequate for the excretion of contrast material. IVP is of limited utility in infants, whose kidneys are immature, as well as in patients with poor kidney function. Additionally, IVP is limited in the diagnosis of UPJ obstruction because it cannot provide adequate quantitative assessment of function in the obstructed kidney.

Retrograde pyelography is useful for further delineating the anatomy of the urinary tract when the results of IVP are equivocal. IVP may help in defining the exact site and nature of the obstruction. Often, retrograde pyelography is performed under general anesthesia just before surgery for definitive repair to confirm the pathology.

Cystourethrography is highly sensitive and specific for VUR and is occasionally helpful in diagnosing UPJ obstruction.

False-positive IVP results may occur in cases involving a dilated, nonobstructing collecting system that is slow to drain. False-positive results may also occur as a result of underlying renal insufficiency or when excretion is reduced because of immaturity. An acquired proximal ureteral obstruction, as might occur through the presence of a stone or by extrinsic compression, may also lead to a false-positive finding.


Computed Tomography

CT may depict ureteropelvic junction (UPJ) obstruction when it is used as a primary study for evaluating common presenting symptoms (see the image below). In older children, the modality is useful for assessing causes of acquired UPJ and ureteral obstruction (eg, stones, tumors, retroperitoneal processes). Often, hydronephrosis is found incidentally on CT scans, and further studies are needed to distinguish UPJ obstruction from nonobstructive hydronephrosis. Computerized Tomography angiography with multiplanar reformation and three-dimensional images may be used to depict suspected crossing vessels as a cause of UPJ obstruction in older children and adults.  [6]

Excretory phase image from an abdominal computed t Excretory phase image from an abdominal computed tomography– (CT-) scan demonstrates massive dilation of the left renal pelvis and calyces, with thinning of the left renal cortex. The right kidney is normal in size and appearance.

Cortical thinning in a hydronephrotic kidney may be seen on CT scans and may be predictive of ipsilateral renal function. The use of intravenous contrast material with nephrogenic and delayed excretory-phase images also may be helpful in determining whether renal function and excretion are impaired.

Degree of confidence

CT provides superior anatomic delineation of the urinary tract system; the use of intravenous contrast material offers some degree of functional information. CT reliably reveals existing hydronephrosis, and the chronicity of obstruction may be correlated with renal cortical thinning. CT is particularly useful for diagnosing acquired UPJ obstructions, such as those caused by stones or extrinsic compression.

False-positive results may occur in cases involving a massively dilated collecting system in the absence of true functional obstruction, as depicted by nuclear renography. False-negative results are unlikely.


Magnetic Resonance Imaging

As with CT, MRI provides excellent details of anatomy in different planes, but it does not offer significant benefit over other, simpler modalities in the evaluation of hydronephrosis. Although MRI is generally not used in the workup of UPJ obstruction, it is as reliable as CT in its ability to depict hydronephrosis and the site of urinary obstruction. False-positive and false-negative findings with MRI are similar to those with CT. [16, 17, 18, 19]

The use of gadolinium-based contrast material may provide some information about renal function. Delayed images may show the presumed site of obstruction with multiplanar imaging in a manner not possible with CT. However, the requirements of MRI—prolonged time and complete stillness—often necessitate the use of sedation in the pediatric patient, making this study less desirable than isotope studies. [20]

Gadolinium-based contrast agents have been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD). The disease has occurred in patients with moderate to end-stage renal disease after being given a gadolinium-based contrast agent to enhance MRI or magnetic resonance angiography scans. NSF/NFD is a debilitating and sometimes fatal disease. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness.



Ultrasonography is the primary study for evaluating hydronephrosis in children (see the images below). Hydronephrosis appears as an anechoic or hypoechoic cavity that splits the bright, central echo pattern of the renal sinus; this may assume the configuration of the pelvis and calyces. [2, 3]  Although the AP diameter of the renal pelvis may be correlated with the likelihood of obstruction, it is not helpful in diagnosing obstruction. Without functional information, renal ultrasonographic results can only suggest and not confirm ureteropelvic junction (UPJ) obstruction. Furthermore, because ultrasonography is poor for imaging retroperitoneal structures (including the ureters), assessing the site of obstruction is often difficult. In UPJ obstruction, upper-tract dilatation is constant; hydronephrosis that varies intermittently or that increases with voiding suggests the presence of vesicoureteral reflux (VUR). [7]

Doppler ultrasound may identify a crossing vessel, when present. This imaging method also has been used differentiating obstructive from non-obstructive hydronephrosis by renal arterial resistive index measurements. [6]  

Longitudinal ultrasonogram of the right kidney in Longitudinal ultrasonogram of the right kidney in a 10-day-old male neonate. The right kidney is larger than the contralateral kidney (6.49 cm in length). Several large, rounded, anechoic areas represent marked caliectasis. Note the thinning of the renal cortex.
Longitudinal image of the right kidney demonstrate Longitudinal image of the right kidney demonstrates that the anechoic areas seen on the previous image all interconnect, differentiating these from renal cysts.
Transverse view of the right kidney demonstrates t Transverse view of the right kidney demonstrates the anechoic areas all interconnecting at the renal pelvis; this is consistent with hydronephrosis from a ureteropelvic junction obstruction.
Longitudinal view of the left kidney in the same p Longitudinal view of the left kidney in the same patient as in the previous image, with Society of Fetal Urology (SFU) grade 1 hydronephrosis and normal renal cortical thickness. Renal length is normal for a newborn, at 4.21 cm.
Transverse images of the right kidney demonstrate Transverse images of the right kidney demonstrate significant pelviectasis, with thinning of the surrounding renal parenchyma.
Transverse images of the right kidney demonstrate Transverse images of the right kidney demonstrate significant pelviectasis. The pelviectasis is so large that the renal pelvis is visualized adjacent to the dome of the distended bladder.

Degree of confidence

Ultrasonography can depict fluid-filled renal collections with greater than 98% accuracy; therefore, it is reliable in identifying hydronephrosis prenatally and postnatally. Ultrasonography is limited in identifying the site of obstruction and in distinguishing between obstructed and nonobstructed dilated systems. The presence of a dilated ureter suggests a more distal obstruction or high-grade VUR. Therefore, ultrasonography serves as an effective screening and monitoring study for hydronephrosis, but its results cannot confirm the diagnosis of UPJ obstruction.

False-positive results may occur in cases involving a large extrarenal pelvis, a peripelvic renal cyst, nonobstructive hydronephrosis, or VUR. Because of medullary immaturity in patients younger than 3 months of age, the renal pyramids may look sonolucent; this may lead to an erroneous diagnosis of caliectasis. Other anatomic and functional causes of hydronephrosis also may lead to a misdiagnosis of UPJ obstruction.

Patient dehydration may also lead to false-negative results, especially in the newborn. Skin defects, exceptional body habitus, intervening bone, and patterns of bowel gas may limit the transduction of sound waves. These factors, as well as operator experience, may lead to a missed diagnosis of hydronephrosis.


Nuclear Imaging

Nuclear medicine scanning may be used to quantitatively assess differential renal function, and it has become a primary study for defining ureteropelvic junction (UPJ) obstruction. In most centers, mercaptoacetyltriglycine (MAG3) has replaced diethylenetriamine pentaacetic acid (DTPA) as the radionuclide of choice (see the images below). Because MAG3 is both filtered and secreted by the renal tubules, it is more useful in immature or chronically insufficient kidneys than is DTPA, which is filtered only by the glomerulus and is not actively secreted. [21]

(Click Image to enlarge.) Renal mercaptoacetyltrig (Click Image to enlarge.) Renal mercaptoacetyltriglycine (MAG3) study in a patient with right ureteropelvic junction obstruction. Initial blood-flow images demonstrate normal perfusion to the kidneys.
(Click Image to enlarge.) Posteroanterior renal sc (Click Image to enlarge.) Posteroanterior renal scan in a patient with right ureteropelvic junction obstruction shows normal uptake and excretion of radiotracer from the left kidney into the left ureter and bladder (and out of the Foley catheter). The progressive uptake of contrast material into the right renal collecting system without excretion is consistent with proximal obstruction.
(Click Image to enlarge.) Continued from the previ (Click Image to enlarge.) Continued from the previous image, this image shows normal uptake and excretion of radiotracer from the left kidney into the left ureter and bladder. The progressive uptake of contrast material in the right renal collecting system without excretion is consistent with proximal obstruction.
(Click Image to enlarge.) The top images demonstra (Click Image to enlarge.) The top images demonstrate the accumulation of tracer without excretion, in the right kidney. The bottom renogram demonstrates renal tracer accumulation and excretion. The normal left kidney demonstrates prompt uptake and excretion. The kidney with ureteropelvic junction obstruction demonstrates continued tracer accumulation without excretion.

Information obtained from a renal scan includes the relative differential renal function; the clearance rate of the radioisotope, with and without a diuretic; and the gross morphology of the collecting system sufficient to distinguish dilatation of the pelvis alone or of both the pelvis and the ureter.

Most centers use the well-tempered renogram, [22] which was designed to standardize prestudy hydration, as well as the dose and timing of the tracer and diuretic. Its purpose is to allow easy comparison between studies and between institutions. In UPJ obstruction, the clearance rate of a radioisotope, often referred to as the washout half-life, or the time required for the radionuclide activity to be reduced by 50%, is greater than 20 minutes. (Ten minutes is normal; 10-20 minutes is indeterminate.)

The drainage curve of an obstructed kidney is characterized by a prolonged accumulation of radioisotope in the kidney without a subsequent declining slope after the administration of furosemide. As previously mentioned, there is debate as to what defines a true obstruction. Many urologists use only decreased differential renal function (< 40%) as a sign of ureteropelvic junction obstruction, disregarding the half-life and the washout curve. Others perform serial scanning; these practitioners intervene only when the differential renal function declines by greater than 10%, regardless of the initial differential function.

Degree of confidence

In general, the rate of urinary flow (measured by washout half-life) is inversely related to the likelihood of obstruction. However, a more accurate representation of obstruction is the resistance to flow, which takes into consideration the pressure needed to generate the flow rate. The best study for measuring resistance is the Whitaker test, which is an invasive, percutaneous pressure-flow study that allows the measurement of renal pelvic pressures. Although there is no true criterion standard for defining UPJ obstruction, the Whitaker test comes closest. Because of its invasive nature, this test is now rarely performed. Nuclear renography measures only urinary flow; thus, its usefulness is limited to determining the severity of the obstruction in certain cases in which the compliance and capacity of the collecting system is abnormal.

False positives/negatives

False-positive results may occur in cases involving a large and highly compliant collecting system in which pelvic volume is high but pelvic pressure is not. This condition may lead to a relatively long washout half-life in the absence of significant obstruction. False-positive studies may also occur in patients with poor renal function in whom the washout half-life is prolonged secondary to poor radioisotope excretion. Another cause of false-positive results is a full bladder, which leads to delayed ureteral emptying; therefore, some centers routinely utilize a change in posture, encourage voiding, or employ a bladder catheter.

False-negative results may occur in cases involving an obstructed collecting system of small capacity and low compliance.



Angiography of the renal artery may be performed before surgery to delineate crossing or supernumerary arteries that may be causing extrinsic ureteropelvic junction obstruction. In patients undergoing minimally invasive endoscopic repair, the presence of such vessels should be excluded to ensure a higher rate of success.

Angiography is reliable in identifying supernumerary arteries, but it provides no information as to whether these arteries are causing mechanical obstruction. CT or MR angiography has largely replaced this invasive study.

False-positive and false-negative findings are not an issue, because angiography is not used to diagnose ureteropelvic junction obstruction.