Dismembered Anderson-Hynes pyeloplasty performed via open or minimally invasive approach is the gold standard technique for the surgical treatment of UPJO in children [17, 18]. There is still discussion as to whether intraoperative drainage is needed and what drainage method to use for an anastomosis. Some practitioners advocated for a stent-less repair and do not advise any form of drainage, and no stent was placed [7]. Others argued that the stent played an important role in supporting the anastomosis and reduced the formation of urinary leakage [19]. The preference for drainage is divided into external drainage and internal drainage. No matter what kind of drainage method, the ultimate goal of drainage is to reduce the occurrence of complications and ensure the success of the operation.

This study compared the efficacy of DJ stent drainage and DJ stent drainage combined with a pyelostomy tube, so the placement of the DJ stent was an important process.

Reasonable selection of the length of the double-J tube is the basis for successful implantation. The ureter length of patients of different ages is different. Too long of a DJ stent easily causes more bladder irritation to patients, while too short of a DJ stent can easily move up or down due to the patients’ physical activity. Palmer et al. [20] provided a simple formula: ureter length = age (years) + 10 cm. Our experience is that an F3 DJ stent with a length of 120-140 mm can be used within one year of age. An F4 DJ stent with a length of 140-200 mm can be used depending on the intraoperative situation for older than 1-year old patients.

The possible risks from a DJ stent implantation include displacement, fracture, stone formation, blockage, hematuria, urinary tract infection, and low back pain. Once the ureteral stent is displaced, serious problems may occur [14]. The displacement of a DJ stent to the posterior urethra will produce serious lower urinary tract symptoms, including discomforts such as frequent urination and urgency of urination. Once the DJ stent is displaced the generated urine will cause pressure on the newly formed anastomosis, and the postoperative tissue edema may also be obstructed, resulting in further increase of postoperative hydronephrosis.

In this study, there was one case of the DJ stent ascending in the DJ stent group. Because the patient had no obvious symptoms, we found that the DJ stent was not in the bladder by renal ultrasound 2-months later. During the operation, the DJ stent was found in the lower ureter of the child through ureteroscopy, and it was diagnosed and removed successfully. This may have been caused by insufficient intraoperative experience and improper selection of the DJ stent. In the combination group there was one case of DJ stent downward movement that caused frequent and urgent urination and was automatically discharged through the urethra. However, there was no urine leakage and no special treatment.

Some studies had reported that the use of DJ stents may lead to postoperative iatrogenic vesicoureteral reflux [21], which may be related to the occurrence of postoperative pyelonephritis. In the studies of Zhu et al. and Zhang et al. [19, 22], the common postoperative complication was pyelonephritis. In our study the incidence of pyelonephritis was similar (two in each group) despite DJ stent being in place for a longer time. This may be due to the routine antibiotic prophylaxis given to all patients after discharge.

In a meta-analysis by Liu et al. in 2019 [23], the average operative time of internal drainage versus external drainage was 147 min versus 155 min for the DJ versus external PU stents. The duration of surgery for the DJ stent and the combination groups were 157.08 ± 17.36 min and 160.88 ± 14.71 min in our study. The difference was not statistically significant(p > 0.05). The hospital stay for the DJ stent group was 6.46 ± 2.66 days, and that of the combination group was 5.22 ± 1.63 days, which was similar to what has been reported in previous literature [24], but there was a statistically significant difference between the two groups (p < 0.05). The use of a pyelostomy tube combined with DJ stent reduced perirenal exudation,which allowed us to remove the perirenal drainage tube faster, which is a factor in faster discharge.

In some studies, the external drainage was from the renal parenchyma [25], which may cause damage to the renal parenchyma and aggravate bleeding. Our pyelostomy tube passed through the renal pelvis and should reduce the damage to the renal parenchyma. There was no incidence of continuous hematuria in the pyelostomy tube or catheter in this study.

Both the DJ stent drainage and the DJ stent combined with pyelostomy tube had a certain improvement effect on postoperative renal function and anterior posterior diameter of the renal pelvis. The two drainage methods were safe and effective, and the success rate of the surgery in the DJ stent group was 92.3% and that of the combination group was 100%, which was not different from previous literature reports [26]. There were 14 cases (26.9%) of complications in the DJ stent group, while only 8 cases (19.5%) in the combined stent group, all of which were Clavien grade I and II that were easy to manage. Non-stent related complications were 10 cases (19.2%) in the DJ stent group and one case (2.4%) in the combination group, and the difference was statistically significant. We believe the reason is the use of a DJ stent combined with a pyelostomy tube can significantly reduce the occurrence of urinary leakage and unplanned reoperation, as it acts as a double insurance drainage method. Park et al. [27] proposed that there was a negative correlation between the improvement of hydronephrosis and immediate postoperative obstruction, and the existence of such an obstruction, even if only temporary, would affect the postoperative outcome. This indicated the importance of adequate drainage for postoperative recovery. Due to the existence of the pyelostomy tube, even if the DJ stent is displaced, the urine can be fully drained to reduce the tension on the anastomotic stoma, which is conducive to the recovery of the anastomotic stoma after the operation. We can wash the renal pelvis to reduce the occurrence of blood clots, and we can also verify whether there is obstruction by antegrade pyelography through the pyelostomy tube [28].

In the DJ stent group, three patients developed abdominal distention and vomit after the removal of the stent, two patients recovered after conservative treatment, and one patient experienced further aggravation of hydronephrosis. Finally, their symptoms improved after the DJ stent was placed again. In the combination group one patient suffered from abdominal distension, vomit, and aggravation of hydronephrosis after removal of the DJ stent and then clamping the pyelostomy tube. The symptoms disappeared after we opened the pyelostomy tube. Finally, by prolonging the indwelling time of the pyelostomy tube to 3-months, the hydronephrosis was significantly reduced and reoperation was avoided. These cases indicate the presence of postoperative anastomotic obstruction, which mean longer hospital stays or the possibility of reoperation. This is unacceptable for children and parents, and may cause conflicts between doctors and patients, which brings great challenges to clinical work. Due to the presence of a pyelostomy tube, the likelihood of adverse reactions in children after extubation is low, and the risk of reoperation is also reduced. Meanwhile, we speculate that some patients may need more time to recover their anastomotic stomas after surgery, but this requires more research to verify.

We do not advocate blindly adding drainage tubes. We should fully consider the patient’s preoperative and intraoperative conditions and choose appropriate drainage methods. We think that for UPJO children with recurrent urinary tract infections, increased intraoperative bleeding, or complex surgical methods, the DJ stent combined with a pyelostomy tube drainage may have a good effect. However, Due to the addition of an additional drainage tube, daily care taken to clean the skin around the fistula to reduce the risk of retrograde infection. We used an F8 catheter made of silica gel. We had two cases of pyelostomy tube falling off after the operation. This is mainly because the volume of water in the catheter balloon will slowly decrease and lose its fixation function over time. Therefore, we suggest replenishing the balloon once every 2–3 weeks and regularly cleaning and disinfecting the skin around the pyelostomy tube. In addition, a pyelostomy increases the incision in the lateral abdominal wall, increases trauma, and can negatively affect the favorable cosmetic outcome of the minimally invasive procedure.

There are still several limitations to this study. First, this study was a single-center study and more studies from multiple centers are needed to further evaluate the efficacy and complications of this drainage method. Second, this study was a retrospective analysis and lacked a prospective randomized controlled arm. Third, the number of patients in this study was relatively limited, and large-scale studies are needed in the future. Fourth, the follow-up period of this study was short, and a longer follow-up is required.