Persistent hydronephrosis after pyeloplasty: Is it a true obstruction? The role of endourology

Introduction: Anderson-Hynes pyeloplasty remains the gold standard in the treatment of ureteropelvic junction obstruction. The diagnostic criteria for defi ning the failure of pyeloplasty are not well-defi ned or even arbitrary. Likewise, the ideal treatment of persistent hydronephrosis after pyeloplasty is not well established. We tested an innovative endourological procedure, which simultaneously allows a diagnostic defi nition of failure and treatment when necessary.


Introduction
Dismembered pyeloplasty remains the gold standard in the treatment of ureteropelvic junction (UPJ) obstruction. Although the success rate is more than 80% [1], the reasons underlying persistent hydronephrosis after pyeloplasty and its management remain controversial. The causes of failure are not clear. Some authors have hypothesized risk factors such as young age at operation, the short duration of ureteral stenting, or the presence of hypoplastic ureter [2]. Moreover , the diagnostic criteria for de ining the failure of pyeloplasty are not well-de ined or even arbitrary. The de inition of failure is based on the persistent or increased pyelic dilation and scintigraphic results (persistence of obstructive pattern, decreased function of the affected kidney). Likewise, there is no consensus on the type of treatment for failed of pyeloplasty; possibilities include surgery (open or laparoscopic) or endoscopy.
The aim of our study was the evaluation of an innovative endourological procedure for failed pyeloplasty, which simultaneously allows a diagnostic de inition and treatment if necessary.

Material and methods
We considered for our study 13 patients with persistent hydronephrosis after dismembered pyeloplasty, treated consecutively between January 2006 and December 2015. Only patients with persistent obstructive hydronephrosis at one year follow-up were included in the present study.
We excluded patients with hydronephrosis due to crossing vessels. The pyeloplasty technique used in all patients was OPRAP, as described in previous studies [3]. The initial follow-up after pyeloplasty (Follow-up 1) consisted of ultrasound at 3 and 6 months and MAG3-scintigraphie at 9 -12 months. Persistent hydronephrosis was de ined as dilation unchanged or worsened with a persistent obstructive pattern at scintigraphie. All the 13 patients with persistent hydronephrosis underwent retrograde pyelography and ureteral calibration/dilation with a balloon catheter (Passeo-18).
The balloon catheter, originally designed for angioplasty, is currently used by our team for the treatment of obstructive megaureter [4]. It has a caliber of 3,8 French; the balloon can be in lated with contrast medium up to a diameter of 4 mm (patients 6 -24 months of age) to 6 mm (over 2 years) and up to a pressure of 12 atmospheres.
The patients in this study underwent cystoscopy under general anesthesia, using an 8-9.8 Ch cystoscope. The catheter was inserted through the ureter into the renal pelvis, the balloon was in lated with contrast medium until it reached a diameter of 4 -6 mm. The catheter was then withdrawn gradually under luoroscopic guidance. Where it was possible to withdraw the catheter without dif iculty, patency of the ureteropelvic junction was demonstrated; the diameter was of 4 -6 mm. In the case of dif iculty in advancing the balloon across the UPJ or where it was not possible to advance it, real persistence of stenosis was shown. In such cases, the balloon was de lated and repositioned at the UPJ level, and high pressure dilation was employed at 12 atm for 5 min. At the end of the dilation procedure, a double J ureteral stent was positioned for 6 weeks. In the case of simple calibration without dilation, no stent was required. All patients with persistent hydronephrosis were again followed-up (Follow-up 2) with ultrasound at 3 and 6 months and MAG3-scintigraphie (scinti 2) at 12 months; thereafter, with ultrasound yearly.

Results
Over a period of 9 years, 487 consecutive pyeloplasties were performed with a success rate of 97.4%. Prenatal or perinatal diagnosis was documented in all patients. The indications at surgery were mean hydronephrosis of 3.5 cm (range 2.5 cm -7.5 cm) and an obstructive pattern at nuclear MAG3 renogram. Median age at surgery was 21 months (range 6 -84). Of our patients, 13 children (2.6%) failed initial procedures, and presented within 10 to 15 months (mean 12) postoperatively.
Of the 13 patients with persistent hydronephrosis, all were found to have a pre-pyeloplasty dilation mean of more than 3 cm and an obstructive pattern on scintigraphie. In one case, reduction of split renal function was observed. Two patients showed symptoms (recurrent lank pain). The endoscopic procedure observed (Table 1) anastomotic stenosis in 3 cases (Figure 1a). This ring was dilated with the balloon catheter for 5 minutes at 12 atm and the stenosis was resolved (Figure 1b). The procedure was easy, well tolerated, without any technical dif icult.
In 10 patients, no stenosis was found during the endoscopic procedure ( Figure 2). No complications were seen intra or postoperatively and no recurrence in the structure after balloon dilation has seen.
At Follow-up 2, split renal function remained stable in all patients (including the one with the initial loss of function); the diuretic test showed an improvement in urine drainage; ultrasonography documented a progressive reduction in pelvic dilation, but not of calices. The 2 symptomatic patients continued to complain of lank pain, although they resulted A B    "not obstructed" at balloon calibration. They were found to have decreased split renal function at scinti 2. For this reason, they underwent a further pyeloplasty, with the operative inding of scar tissue compressing the UPJ.

Discussion
Dismembered pyeloplasty is considered the treatment of choice in cases of obstructive hydronephrosis, with a success rate of about 98% [1]. The literature offers no de initive conclusions regarding the causes of failure [5,6]. Of the possible risk factors, some authors suggest young age at intervention (less than 6 months) [6] and urinary leakage as the initial event leading to scar formation and ibrosis around the UPJ [7,8]. The incidence of crossing vessels associated with obstructive hydronephrosis is reported as being between 11% and 49% and is responsible for 10% of failed pyeloplasty owing to the lack of recognition of this condition at the irst operation [5]. Baniel et al suggest that persistent obstruction can be attributed to in lammatory edema consequent to surgical manipulation [9]. Other risk factors could be a very dilated pelvis, insuf iciently reduced at the surgical procedure, and the possibility of urinary iltration with secondary ibrosis around the anastomosis [8]. For these reasons, the authors stress the importance of the correct positioning of a stent with prolonged drainage to prevent iltration and subsequent obstruction. In fact, redundancy of the pelvis, hypoplasia of the ureter or inadequate spatulation of the ureter during intervention or any kinking may result in leakage of urine. This can lead to potential local in lammation and recurrence of stenosis [2,10].
Regarding the management of persistent/recurrent hydronephrosis, opinions differ among researchers. Nicholls et al. suggest endopyelotomy for secondary symptomatic hydronephrosis in older children [11]. Figureenshau reports a success rate of 91% with endopyelotomy for secondary obstructions of the UPJ [12]. Kinn stresses the importance of the symptoms, in the absence of which hydronephrosis in adults should be considered a benign condition, to be managed conservatively. On the contrary, the same author suggest that the presence of symptoms is an indication for surgery [13]. Open surgery is also preferred by Thomas, et al. in view of the poor results of endoscopic dilation [5].
In our series, particular attention was paid to the abovementioned risk factors. As far as possible the same surgical technique was employed, which included prolonged postoperative drainage. The search for crossing vessels was meticulous and where they were identi ied the patients were excluded from the series. Interestingly, in 3 cases, the initial pelvic dilation was greater than the average of our overall series. The large dimensions of the pelvis could then be the main risk factor for persistent hydronephrosis after pyeloplasty. In our cases, the endourological procedure allowed con irmation that, in the majority of cases (10/13), there was no real recurrence of stenosis. There was instead a residual hypotonia of the pelvis and especially of the calices (already very dilated at the start), which were responsible for the persistence of the "apparently obstructive" pattern on scintigraphie. The ultrasound at Follow-up 2 con irmed a trend of the reduction of the pelvis, whereas the calices, for obvious anatomical reasons, tended to remain dilated. A further con irmation of the hypothesis that no real stenosis exists can be seen from the stability of the split renal function and the improvement (spontaneous, with no dilation procedure) of urinary drainage at scintigraphie 2.
The calibration of the UPJ with a balloon catheter, proposed here, is essentially a diagnostic procedure, which excludes true recurrence of the obstruction in the majority of cases. In a few cases (3/13), where true stenosis was documented, the procedure also became therapeutic, with the use of high pressure dilation. One potential limitation of the endourological procedure is that it can fail to identify extrinsic compression of the UPJ by scar tissue, as a consequence of the initial operation. In this instance, the balloon catheter fails to reveal a stenosis. Clinical follow-up is essential in this situation. The 2 cases in our series with extrinsic compression, despite having a normal caliber at endoscopy and a stable split renal function, were both symptomatic and therefore had to undergo surgery.

Conclusion
Dismembered pyeloplasty remains the gold standard for treating obstructive hydronephrosis, achieving a very high success rate. In cases of failure, the endourologic calibration/ dilation of the UPJ is a minimally invasive procedure, simple, safe and with no potential risks and complications for the patients. It can be considered both as a useful diagnostic tool and as a irst therapeutic approach in cases of persistent hydronephrosis after pyeloplasty. The causes of failed pyeloplasty are not completely clear, in most cases the persistence of hydronephrosis is due to prolonged hypotonia of the pelvis rather than to recurrence of stenosis. In the absence of symptoms, the assessment of an adequate caliber of the UPJ and stability of split renal function are suf icient reasons to avoid further surgical procedures.