Pediatric Ureteropelvic Junction Obstruction

Updated: Jun 30, 2022
  • Author: Paul R Bowlin, MD; Chief Editor: Marc Cendron, MD  more...
  • Print

Practice Essentials

Ureteropelvic junction (UPJ) obstruction is by far the most common cause of pediatric hydronephrosis, occurring in 1 per 1000-2000 newborns. Widespread use of antenatal ultrasonography (US) and the advent of modern imaging techniques have resulted in earlier and more common diagnosis of hydronephrosis. Before the advent of US, congenital hydronephrosis presented throughout childhood and even adulthood with various symptoms such as abdominal or flank pain, recurrent urinary tract infection (UTI), abdominal mass, renal stones, and hematuria.

Since the first reconstruction of an obstructed kidney in the late 1800s by Trendelenburg, surgery for UPJ obstruction has evolved significantly. In 1936, Foley described the results of 20 pyeloplasties using the so-called YV-plasty repair. [1]  In 1946, Anderson and Hynes published their experience with an operation that included complete transection of the upper ureter, subsequent spatulation of the distal ureter, and trimming of the redundant pelvis. [2]  This highly successful technique has become the standard for surgical repair used today, even in robotic pyeloplasties.

There has been an explosion of information relating directly to the pathogenesis, diagnosis, and treatment of UPJ obstruction. The decision between observation, medical prophylaxis, and surgical intervention mostly depends on how inefficient ureteropelvic urine transport is; determining this inefficiency may not be straightforward, because various biologic, mechanical, and clinical factors must be considered. The effects of the obstructive process on renal function must also be considered.

Upon diagnosis of a UPJ obstruction, prompt intervention is appropriate to prevent or minimize renal damage. First, prophylactic antibiotic therapy is warranted in cases of moderate-to-severe dilatations because any UTI, especially in the neonatal period, dramatically increases the chance of fibrosis and parenchymal damage.

The timing of surgical correction of hydronephrosis suggestive of UPJ obstruction in newborns is highly controversial. Those who support delayed management contend that for most newborns with differential renal function (DRF) that is relatively preserved (>35% of DRF), hydronephrosis is a relatively benign disease without proof of progression. Renal function does not deteriorate; thus, immediate surgery is not necessary.



The UPJ is formed during week 5 of embryogenesis; by weeks 10-12 of gestation, the initial tubular lumen of the ureteric bud becomes canalized, with the UPJ area being the last to canalize. Inadequate canalization of this area is the main embryologic explanation for UPJ obstruction. Östling folds, which resemble UPJ obstruction, are remnants of fetal renal development. [3] No significant renal pelvic dilation or hydronephrosis is associated with this condition; however, most hydronephrotic kidneys now are detected antenatally.



The urinary drainage from renal pelvis to ureter is determined by many factors. Pressure within the renal pelvis is determined by the volume of urine produced, the internal diameter of the UPJ and collecting system, and the compliance of the renal pelvis, as well as the peristaltic activity of the ureter.

In response to the increased volume and pressure, the renal pelvis dilates. Initially, the smooth muscle of the renal pelvis may thin out, but over time, it may become hypertrophied to varying degrees. The effects on the developing renal parenchyma may be quite variable, owing to the compliance of the renal collecting system. Despite massive dilation, preservation of renal function may occur.

Koff et al proposed the concepts of pressure-dependent and volume-dependent flow. [4] In instances of intrinsic obstruction, at low urinary flow rates, no obstruction exists; however, as the flow rate increases, the urinary bolus is not conducted, which causes the renal pelvis to distend. This pattern is referred to as pressure-dependent or volume-independent flow.

On the other hand, in cases of extrinsic compression usually caused by aberrant vessels, urine flow is impeded only after a definite amount of urine is collected in the renal pelvis. This is an example of volume-dependent flow, and the pressure damage is only evident intermittently. Because the pressure elevations occur infrequently, the degree of damage generally is less than that observed with intrinsic obstruction .

Experimental results show that the release of complete, acute ureteral obstruction of 24 hours’ duration results in a near-normal glomerular filtration rate (GFR), but the total number of filtering nephrons is decreased, causing an increase in the single-nephron GFR. Local production of prostacyclin and prostaglandin E2 (PGE2) seems to affect glomerular arteriolar resistance. Tubular functions also are changed by acute obstruction, but the effects of obstruction on distal tubular and collecting duct physiology may not be as quantitatively significant as the effect on deep nephron function.

Most clinically observed obstructions are partial, and models of partial ureteral obstruction have been developed. In this case, the total GFR of the obstructed kidney is decreased, but the single-nephron GFR increases. This process probably is the result of changes in the local production of vasoactive peptides and cytokines. [5]

Significant urinary obstruction may result in tubular dilation, glomerulosclerosis, [6] inflammation, and fibrosis. Although it is not absolute, a good correlation exists between the severity of these histologic changes and the function remaining in the affected kidneys. Sclerotic glomeruli and fibrosis are reliably localized to areas of the kidney that demonstrate the most inflammatory infiltrate. The infiltrate consists mostly of mononuclear cells in both the cortex and the medulla. Macrophages are the predominant cell type, though a small number of T cells are present.

Elevated levels of interleukin (IL)-5 and eotaxin-2 have been shown in partially obstructed kidneys, which suggests that chemokines produced by the urothelium may be the chemoattractants for leukocytes, leading to further inflammatory cell infiltration, mast cell migration, and activation. The arrival of the infiltrating cells closely corresponds with the decrease in renal blood flow and GFR after obstruction. Obstructed kidneys also demonstrate increases in cyclooxygenase activity and increased thromboxane synthetase. [7, 8, 9] The monocytic infiltration has a role in changing the eicosanoid elaboration in the kidney, which, in turn, acts locally to decrease the GFR.

Detailed knowledge of the cytokine responses in the urothelium is necessary for a better understanding of the role of urothelial cellular reactions in congenital UPJ obstruction. This may provide the basis for early intervention to prevent exacerbation of congenital hydronephrosis in the future.

The activation of the renin-angiotensin system has been thought to be a major factor in partial ureteral obstruction. Administration of angiotensin-converting enzyme (ACE) inhibitors has been shown to maintain renal blood flow in partially obstructed kidneys at 3 weeks post obstruction and to prevent the histologic changes of glomerulosclerosis. [10] The effects of obstruction are not all ischemic. Ureteral obstruction can mimic renal artery stenosis, and because of its intense vasoconstrictor action, the resulting increase in angiotensin II (AII) leads to decreases in GFR. [11]

It has become clear, however, that AII profoundly affects the expression of growth factors in the developing kidney. Upregulation of transforming growth factor beta (TGF-β) has been demonstrated in obstructed kidneys, and the degree of upregulation correlates directly with the amount fibrosis and collagen deposition seen in chronically obstructed kidneys. [12]

How some of the experimental information translates into clinical practice remains to be clarified. The long-term effects of a chronic obstructive process on the kidney are highly variable, being affected by differences in the degree of obstruction, the timing of obstruction, and the ability of the renal collecting system and renal parenchyma to adjust to the change in environment. Early obstruction is felt to cause severe malformation of the kidney (dysplasia), whereas late obstruction may not affect the kidney as severely.



UPJ obstruction can be broadly divided into the following two categories:

  • Lesions that involve the UPJ intrinsically
  • Lesions that are extrinsic

Intrinsic abnormalities

During embryogenesis, the ureter arises from the ureteral bud and extends towards the area of parenchyma that will become the kidney. The UPJ forms during week 5. Induction of the metanephric blastema has been thought to be mediated by the ureteral bud through transcription factors such a Pax-2 and by growth factors such as c-ret, kdn-1, and wt1, as well as by TGF-β. [13] By weeks 10-12 of gestation, the initial tubular lumen of the ureteric bud becomes canalized.

It has been suggested that the ureteropelvic and ureterovesical portions of the ureter are the last to canalize; thus, failure of the process to complete would lead to partial canalization. Another theory for the development of an obstructive process suggests premature arrest of ureteral wall musculature development leading to the persistence of an aperistaltic segment at the level of the UPJ, thus preventing normal propulsion of urine down the ureter. The role of some of the transcription and growth factors in the etiology of UPJ obstruction remains to be elucidated.

Another theory proposes that improper innervation with diminished synaptic vesicles may be a factor in the development of UPJ obstruction. Factors involved in neuronal development, such as protein gene product (PGP) 9.5 (a general neuronal marker), S-100 protein (a nerve supporting cell marker), synaptophysin (a synapse vesical marker), and nerve growth factor receptor have been reported in decreased amounts in resected specimens of UPJ.

Intrinsic obstruction is noted as ureteral narrowing with angulation during surgical exploration. In these cases, the ureter is solely involved, and the lumen generally is narrowed but open, allowing the retrograde passage of a small feeding tube or ureteral catheter. Early in development, the proximal ureter is folded on itself, and persistence of the infolding may contribute to the kinked appearance of the proximal ureter.

The most attractive theory is that the obstruction is secondary to muscular discontinuity, which disrupts the coordinated motion of smooth-muscle cells and may result in impeded peristalsis propagation across the UPJ and interference with urine bolus formation in the proximal ureter. This absence or disorientation of smooth-muscle fibers at the UPJ is clearly evident on electron microscopy, with the findings of rearrangement and widely separated smooth-muscle cells, excessive collagen fibers, and increased elastin in the adventitia, combined with diminution of nerve terminals and nerves at the stenotic portion.

Problems with this narrow but structurally patent lumen of the UPJ may not be evident under low workloads (ie, low urine volumes into low bladder pressures) but become apparent as a consequence of inability to adjust efficiently to increased workloads (high urine volumes or high bladder pressure).

In all likelihood, intrinsic UPJ obstruction may be caused by a variety of factors, which may have both biochemical and mechanical etiologies.

Extrinsic abnormalities

Extrinsic obstructions secondary to fibrous bands, kinks, and aberrant crossing vessels are also encountered. It has been reported that the proximal ureter is angulated, distorted, and compressed by vessels that supply the lower pole of the kidney in about 25% of UPJ obstructions. During ascent and rotation of the fetal kidney, the kidney has segmented vessels from the aorta arranged in a ladder pattern. As the kidney ascends further, its blood supply is derived from higher vessels, while the lower ones disappear.

Thus, abnormal progression of renal ascent and rotation combined with renovascular formation may result in an unfavorable ureterovascular configuration. In this case, with the increased urine volume, the UPJ angulation with intrapelvic volume expansion causes increased resistance and obstruction. Further angulation may occur if it is associated with an inflammatory process. However, whether these crossing or anomalous vessels can primarily cause significant UPJ obstruction in the absence of a significant intrinsic lesion is controversial.

Patients with extrinsic obstructions tend to present late in childhood, with intermittent abdominal or flank pain. Horseshoe or pelvic kidney, duplex collecting systems, and other rotational abnormalities also may cause UPJ obstruction. Cases of so-called high inserted ureter–to–renal pelvis exist, but this is presumed to be a secondary phenomenon to obstruction because the ureteral insertion seems to be higher in cases of dilated renal pelvis.



Obstruction at the UPJ is found more commonly in boys than in girls, especially in the newborn period, with up to twice as many cases diagnosed in boys. As many as 67% of cases involve the left kidney in the newborn period, and bilateral cases (synchronous and asynchronous) are observed in 10-40% of cases, most of which are noted in infants younger than 6 months; however, fewer than 5% of patients require bilateral repair because of spontaneous resolution in a significant number of cases.

Congenital renal malformation anomalies may also be noted in association with UPJ obstruction. Specifically, renal dysplasia and multicystic dysplastic kidney have been observed in the contralateral kidneys, with almost 5% of children having renal agenesis. Other anomalies include duplicated renal collecting system, in which case the lower-pole UPJ is usually the obstructed segment; horseshoe kidney; and ectopic kidney.

A fairly high (up to 40%) rate of associated vesicoureteral reflux (VUR) has also been reported. The reflux is usually of relatively low grade and may resolve spontaneously.



The overall success rate with the dismembered repair (see Treatment) is quite satisfactory; most series report a success rate in excess of 90-95%. Long-term obstruction at the anastomosis can occur; but reoperation for this is low, occurring in 2-5% of cases. However, there are no set criteria for follow-up or even for success.

Although there is no protocol for follow-up, postoperative evaluation includes US at 1-3 months and renal scanning or excretory pyelography at 2-3 months. Late problems occurring beyond 12-24 months are uncommon in the absence of symptoms. A successful outcome does not always mean an improvement in DRF as measured by renography. In most cases, dismembered pyeloplasty improves the degree of hydronephrosis and washout on the renogram. Symptoms of pain, infection, and hematuria, if present before surgery, resolve along with the improvement of hydronephrosis.

There is little debate that mild degrees of hydronephrosis, such as Society of Fetal Urology (SFU) grades 1 or 2, have a favorable outcome and are unlikely to cause a problem necessitating surgical intervention. However, the percentage of cases that show progressive renal deterioration during observation is the main concern. Most series of serial observation in unilateral hydronephrosis with split renal function greater than 35-40% reported that 15-33% showed progressive renal deterioration, with the exception of the report by Koff et al, which reported the lowest deterioration rate, 7%. [4]

Another reason for delayed surgery, apart from progressive deterioration of renal function, was recurrent febrile UTIs. Most observation series report that about 50% did not regain lost renal function after delayed pyeloplasty, again with the exception of the report by Koff et al, in which the recovery rate was 100%. The timing of surgery seems to play an important role in the recovery of renal function, with most reports suggesting that repair before age 1 year permits maximal improvement of renal function.

Patients with lower percentages of elastin in the renal pelvis, the UPJ proper, or the ureter tended to show better resolution of hydronephrosis 6 months after pyeloplasty. [14] Increased elastin of the renal pelvis and ureter might result in inelasticity and low compliance, which delays the improvement of hydronephrosis following pyeloplasty.