Coexisting Ureteropelvic Junction Obstruction and Vesicoureteral Reflux: Diagnostic and Therapeutic Implications

0022-534 7/89/1422-0490$02.00/0 THE JOURNAL OF UROLOGY Copyright© 1989 by AMERICAN UROLOGICAL ASSOCIATION, INC.

Vol. 142, August

Printed in U.S.A.

Vesicoureteral Reflux COEXISTING URETEROPELVIC JUNCTION OBSTRUCTION AND VESICOURETERAL REFLUX: DIAGNOSTIC AND THERAPEUTIC IMPLICATIONS JEAN G. HOLLOWELL, HOWARD G. ALTMAN, HOWARD McC. SNYDER, JOHN W. DUCKETT

IIIAND

From the Division of Urology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

ABSTRACT

The coexistence of ureteropelvic junction obstruction and vesicoureteral reflux was seen in 17 (14 per cent) of 147 consecutive patients undergoing pyeloplasty. These patients were analyzed to determine the therapeutic implications of this association. They fell into 3 clinical groups: group 1-primary ureteropelvic junction obstruction, group 2-ureteropelvic junction obstruction secondary to high grade reflux and group 3-pseudo-ureteropelvic junction obstruction. There were 11 patients in group 1. These patients had incidental low grade reflux. Pyeloplasty should be the initial procedure with use of a nephrostomy tube or Foley catheter postoperatively. Vesicoureteral reflux resolved spontaneously with linear growth in approximately half of the cases. There were 4 patients in group 2. The obstructive uropathy should be treated initially, since primary ureteral reimplantation may provoke acute ureteropelvic junction decompensation. There were 2 patients in group 3. Pseudo-ureteropelvic junction obstruction is suggested when pelvic dilatation on the voiding cystourethrogram suggests obstruction but drainage films or antegrade studies document good drainage. The recognition of pseudo-ureteropelvic junction obstruction is important to avoid surgery on a ureteropelvic junction that is not obstructed. Primary correction of the reflux is appropriate. However, it must be recalled that a fixed kink may rarely develop later leading to true secondary ureteropelvic junction obstruction, which will require surgical correction. We recommend that a voiding cystourethrogram be part of the routine evaluation of children with suspected ureteropelvic junction obstruction. (J. Urol., part 2, 142: 490-493, 1989) Physicians have become increasingly aware of the need to evaluate radiologically children with urinary tract infections. This evaluation should include ultrasound or an excretory urogram (IVP) and a voiding cystourethrogram. However, if ureteropelvic junction obstruction presents with symptoms and signs other than urinary tract infection, a voiding cystourethrogram is not always routinely performed. While several articles have indicated the coexistence of ureteropelvic junction obstruction and vesicoureteral reflux, this patient population has not been well defined. l-4 A review of our records of 147 patients (1973 to 1986) with ureteropelvic junction obstruction revealed coexisting vesicoureteral reflux in 17 (14 per cent). An analysis of these patients indicates that this population is not homogeneous but it falls into 3 clinical categories: group 1-primary ureteropelvic junction obstruction with incidental low grade reflux into ureters without dilatation, kinking or tortuosity; group 2-secondary ureteropelvic junction obstruction documented by hydronephrosis and delayed drainage on an IVP or a diuretic renal scan with high grade reflux into ureters that are dilated, tortuous and have a fixed kink at the ureteropelvic junction, which leads to true obstruction (fig. 1), and group 3-pseudo-ureteropelvic junction obstruction defined as reflux into a distended pelvis with a kink at the ureteropelvic junction radiographically mimicking the upper tract seen in ureteropelvic junction obstruction. However, drainage films, an IVP or a renal scan documents good drainage, showing that this group does not repre-

sent an obstructive uropathy (fig. 2). This review does not include all of our patients with reflux and pseudo-ureteropelvic junction obstruction but this entity is presented to emphasize the importance of its recognition and to indicate that after correction of reflux, a delayed fixed kink can develop at the ureteropelvic junction leading to true obstruction requiring pyeloplasty. We have seen 2 such cases. Understanding the relationships of vesicoureteral reflux to ureteropelvic junction obstruction in each of these groups is important because of different therapeutic implications that result. MATERIALS

The records of 147 consecutive patients who underwent pyeloplasty for ureteropelvic junction obstruction at our hospital from 1973 to 1986 were reviewed. Of these patients 120 underwent a voiding cystourethrogram, 17 (14 per cent) of which revealed reflux. The incidence was 21 per cent (9 of 42) among female and 10 per cent (8 of 78) among male patients. To define better this patient population, their records were reviewed with regard to presentation, appearance of the ureter on voiding cystourethrogram, whether the reflux was ipsilateral or contralateral to the ureteropelvic junction obstruction, the initial surgery and the followup. RESULTS

The 17 patients with coexistent vesicoureteral reflux and ureteropelvic junction obstruction ranged in age from newborn

490

lJR,gTEH,OPELViC JUI~C'TIO>J OBSTRtJCTIO!'-J AND \TES!(_;ouRETERAL REFLDX

491

FIG, L A, voiding cystourethrogram demonstrates kink at left urete:ropelvic junction suggesting obstruction, B, IVP with catheter in bladder shows delayed drainage from left renal pelvis consistent with true secondary ureteropelvic junction obstruction,

FIG, 2, Voiding cystourethrog:ram demonstrates right reflux with kink at ureteropelvic junction. Since renal scan revealed good washout diagnosis was pseudo-ureteropelvic junction obstruction.

to 9 years. Of the patients 6 presented with urinary tract infection, 3 had an antenatal diagnosis of hydronephrosis, 1 had a palpable abdominal mass, 4 had flank and/or abdominal pain, 2 had the prune belly syndrome and 1 had undergone an abdominal ultrasound for unknown reasons. While the pre-

senting diagnosis (urinary tract infection, antenatal diagnosis of hydronephrosis and prune belly vri,mm,,,-, would have led to a voiding cystourethrogram in 11 (65 per cent) of the 17 patients, the remaining 6 (35 per cent) would not have had the reflux diagnosed had it not been our practice to obtain routinely a voiding cystourethrogram in patients with ureteropelvic junction obstruction. U:reteropelvic junction obstruction was diagnosed by an IVP in the majority of cases. Ultrasound was used in those patients less than 1 month old. Diuretic :renograms were used frequently as an adjunct in the diagnosis of obstruction in more recent years. Reflux was documented by a contrast voiding cystomethrogram in 27 ureters in these 17 patients. Of the 17 patients 9 had unilateral ureteropelvic junction obstruction and bilateral 4 had unilateral reflux ipsilateral to the ureteropelvic junction obstruction, 3 had unilateral reflux contralateral to the ureteropelvic junction obstruction, and 1 had bilateral obstruction and bilateral reflux. Reflux graded in the presence of ureteropelvic junction the grading system for primary reflux is only ,u,µucwu,e if reflux is the cause of the dilated collecting system. VVe categorized reflux as low grade if the ureters were normal size, and high if the ureters were dilated and tortuous, Of the patients 11 had low grade and 6 had high grade reflux, Patients were classified into 3 groups. The 11 patients with primary ureteropelvic junction obstruction and iil.cidental low grade reflux into nondilated ureters were included in group 1. The other 6 patients with high grade reflux into tortuous dilated ureters with kinking at the level of the ureteropelvic junction were divided, depending on whether the kink at the ureteropelvic junction represented true or merely pseudo-ureteropelvic junction obstruction. There are 2 radiographic presentations for true secondary ureteropelvic junction obstruction (group 2, 4 patients): 1) a voiding cystourethrogram showing high grade reflux into dilated ureters up to the level of the ureteropelvic junction but not into the pelvis and an IVP or diuretic renogram documenting obstruction, and 2) a voiding cystourethrogram demonstrating high grade reflux into a dilated renal pelvis and the contrast material in the renal pelvis does not drain. An IVP or diuretic renogram confirms the presence of obstruction.

492

HOLLOWELL AND ASSOCIATES

In pseudo-ureteropelvic junction obstruction (group 3) the distension of the renal pelvis seen on a voiding cystourethrogram mimics ureteropelvic junction obstruction (fig. 2) but drainage films show no hold-up at the ureteropelvic junction and antegrade studies (an IVP or diuretic renogram) confirm the absence of obstruction. Rarely, a fixed kink may later develop at the ureteropelvic junction, resulting in a true secondary uropathy that usually presents with intermittent flank pain. 5 • 6 This occurred in 2 of our patients and pyeloplasty was required. Of the 17 patients 11 in group 1 and 4 in group 2 underwent initial dismembered pyeloplasty. Group 3 patients underwent initial correction of the reflux. Followup of the 11 patients in group 1 revealed that reflux ceased in 6 within an average period of 18 months, reflux persisted in 2 at 2½ to 10 years and bilateral reimplantation was performed in 1, 6 years after pyeloplasty. The remaining 2 patients were lost to followup. Among these 11 patients 9 ureters with reflux were ipsilateral and 9 were contralateral to the ureteropelvicjunction obstruction. Reflux resolved spontaneously in 5 ipsilateral and 5 contralateral ureters. Followup of the 4 patients in group 2 revealed that reflux persisted 2½ years later in 2, bilateral reimplantation was performed 3 years later in the patient with the prune belly syndrome and unilateral reimplantation was performed 6 months later for a paraureteral diverticulum on the side of reflux in 1. The 2 patients in group 3 underwent ureteral reimplantation as the initial procedure. One patient presented 4 years later with flank pain, and an IVP revealed hydronephrosis and delayed drainage at the ureteropelvic junction that had previously drained well. The other child originally presented with left flank pain, and evaluation revealed high grade reflux and kinking at the ureteropelvic junction but good drainage on a diuretic renal scan. The flank pain resolved after ureteral reimplantation but then it recurred 6 months later. A repeat voiding cystourethrogram revealed no reflux but a repeat diuretic renal scan demonstrated obstruction. Both of these patients underwent successful secondary dismembered pyeloplasty. A nephrostomy tube and/or ureteral stent was used in 10 of the 16 patients with reflux undergoing pyeloplasty. All of the patients had a catheter in the bladder postoperatively. There were no cases of prolonged drainage from the flank drain whether or not a nephrostomy tube was used. There were no surgical complications of the pyeloplasty or ureteral reimplantations. All patients were placed on prophylactic antibiotics at the time of diagnosis of vesicoureteral reflux and it was continued until reflux ceased. DISCUSSION

The presentation of 35 per cent (6 of 17) of our patients would not have necessitated a voiding cystourethrogram and, thus, reflux might not have been diagnosed. We emphasize the importance of a voiding cystourethrogram as part of the evaluation of all patients with ureteropelvic junction obstruction, since reflux has important therapeutic implications at the time of pyeloplasty, as well as it requires independent long-term followup. We recommend that a catheter be placed in the bladder at pyeloplasty to prevent increased pressure at the anastomosis from refluxing urine during voiding. The additional use of a nephrostomy tube should be at the discretion of the surgeon but it is not required. After pyeloplasty the patients should remain on prophylactic antibiotics and the vesicoureteral reflux should be followed. Pyeloplasty as the initial procedure generally is recommended for group 1 patients. There are different opinions regarding the initial surgery in patients with true ureteropelvic junction obstruction secondary to high grade reflux. Propo-

Followup of reflux after pyeloplasty in ureters ipsilateral and contralateral to the ureteropelvic junction obstruction Degree of Reflux

lpsilat. Contralat.

Low

High

9 (5) 9 (5)

6 (0) 3 (0)

Numbers in parentheses represent spontaneous resolution.

nents of performing a pyeloplasty initially indicate that ureteral reimplantation may cause acute worsening of the ureteropelvic junction obstruction because of edema at the ureterovesical junction after reimplantation. 1 Johnston and Farkas discussed reflux causing "angulation and obstruction", which may be "relieved by curing the reflux" as opposed to when the "angulation has become fixed" and pyeloplasty is then required. 7 Williams reported treating patients with ureteropelvic junction obstruction secondary to high grade reflux with reimplantation initially to avoid overdistension of the renal pelvis and alleviate kinking at the ureteropelvic junction. 8 We would classify these "obstructions" that are expected to be cured by treating the reflux as pseudo-ureteropelvic junction obstruction. True ureteropelvic junction obstruction would not be expected to resolve after ureteral reimplantation. Maizels and associates discussed evaluation of the ureteropelvic junction obstruction to differentiate between "severe, fixed kinks or strictures or kinks with strictures" as opposed to "most tortuosities that are not obstruction" and they recommended pyeloplasty initially for cases of true obstruction. 2 Whitaker used pressure/flow studies to differentiate between true obstruction and pseudo-obstruction. He believed that the etiology of true secondary obstruction is related to the easy distensibility of the pelvis changing its shape and interfering with the ability of the renal pelvis to form a bolus of urine, so that during a downward propagating pelvic contraction, the circular component obstructs the lumen. 3 It would appear that the varying recommendations regarding the initial surgical procedure are due to a problem differentiating between true secondary and pseudo-ureteropelvic junction obstruction. Careful study of the voiding cystourethrogram drainage films, and an IVP or renal scan with a catheter in the bladder should resolve this dilemma. True obstruction cannot be expected to resolve after ureteral reimplantation and we agree the obstructive uropathy should be treated first. All of our patients with true ureteropelvic junction obstruction (whether primary or secondary) were treated with pyeloplasty as the initial procedure (see table). In 10 (55 per cent) of 18 ureters low grade reflux resolved spontaneously. The rate of spontaneous resolution of reflux was the same for ureters ipsilateral and contralateral to the pyeloplasty. This result does not support the belief of some authors that the amount of urine flowing down the ureter increases after pyeloplasty and, thus, enhances spontaneous resolution of reflux. 9 All 4 ureters with high grade reflux continue to reflux or they have been reimplanted. Although it is not possible to grade precisely the reflux, we see no indication that the natural history of reflux coexisting with ureteropelvic junction obstruction is different from reflux occurring independently. Our 2 patients with pseudo-ureteropelvic junction obstruction and high grade reflux were managed initially with reimplantation. Due to later presentation with flank pain, repeat upper tract studies were obtained, which revealed obstruction. Although previously kinks at the ureteropelvic junction had been noted, there had been good drainage. DeKlerk and associates also described 2 similar patients in whom ureteropelvic junction obstruction developed after successful ureteral reimplantation.4 Thus, patients who have had the ureteral anatomy altered due to reflux can have kinks at the ureteropelvic junction, which later in life become fixed and obstructing.

URE'TEROPELVIC JUNCTION OBSTRUCTION AND VESICOURETERAL REFLUX CONCLUSIONS

A voiding cystourethrogram should be a routine part of the evaluation of a child with ureteropelvic junction obstruction. High grade reflux can lead to kinks at the ureteropelvic junction and it must be determined whether there is true secondary (group 2) or merely pseudo-ureteropelvic junction (group 3) obstruction. If a patient has true ureteropelvic junction obstruction, whether it is primary with low grade reflux (group 1) or secondary to high grade reflux (group 2), pyeloplasty should be the initial surgical procedure and a catheter should be left in the bladder postoperatively. Patients with pseudo-ureteropelvic junction obstruction (group 3) should undergo ureteral reimplantation initially but it must be kept in mind that kinks at the ureteropelvic junction may later become fixed and obstructing. Thus, the family should be warned of the possible significance of later flank or abdominal pain. REFERENCES 1. Lebowitz, R. L. and Blickman, J. G.: The coexistence of uretero-

pelvic junction obstruction and reflux. Amer. J. Roentgen., 140:

493

231, 1983. 2. Maizels, M., Smith, C. K. and Firlit, C. F.: The management of children with vesicoureteral reflux and ureteropelvic junction obstruction. J. Urol., 131: 722, 1984. 3. Whitaker, R. H.: Reflux induced pelvic-ureteric obstruction. Brit. J. Urol., 48: 555, 1976. 4. DeKlerk, D. P., Reiner, W. G. and Jeffs, R. D.: Vesicoureteral reflux and ureteropelvic junction obstruction: late occurrence of ureteropelvic obstruction after successful ureteroneocystostomy. J. Urol., 121: 816, 1979. 5. Snyder, H. M., III, Lebowitz, R. L., Colodny, A. H., Bauer, S. B. and Retik, A. B.: Ureteropelvic junction obstruction in children. Urol. Clin. N. Amer., 7: 273, 1980. 6. Koff, S. A., Hayden, L. J., Cirulli, C. and Shore, R.: Pathophysiology of ureteropelvic junction obstruction: experimental and clinical observations. J. Uro!., 136: 336, 1986. 7. Johnston, J. H. and Farkas, A.: The congenital refluxing megaureter: experiences with surgical reconstruction. Brit. J. Urol., 47: 153, 1975. 8. Williams, D. I.: The natural history of reflux: a review. Urol. Int., 26: 350, 1971. 9. Teele, R. L., Lebowitz, R. L. and Co!odny, A. H.: Reflux into the unused ureter. J. Urol., 115: 310, 1976.