Individualized Laparoscopic Surgery for UPJO Patients with Crossing Vessels

Background: To characterize the anatomical subtypes of ureteropelvic junction obstruction (UPJO) caused by crossing vessels (CVs) and demonstrate the Individualized operation procedures for these cases. Methods: From March 2015 to July 2019, 51 consecutive adult patients underwent treatment of primary UPJO via a retroperitoneal laparoscopic approach. The clinical data, iconography inspection results, and surgical procedures for each patient were retrospectively reviewed by our team. The diagnosis of etiological CV was conrmed during the operation in 13 patients (25.49%), which included 7 men and 6 women. Results: The mean surgical age was 30±11.66 years. The operating time was approximately 233±62.76 minutes, and there were one open conversions. In the follow-up period (mean, 27.23±15.46 months), all patients had a full recovery in the CV group. However, 3 patients without CV did not completely recover from uronephrosis, as determined on iconography inspection, and there was no improvement in the renal colic symptoms. In the CV group, none of the patients had lithiasis whereas 25% of the patients without CV had lithiasis. Conclusion: CV accounts for approximately 25.49% of the cases. Based on the we identied and 84.62% of the CVs were located anterior to the The approach for comprehensive understanding


Background
Laparoscopic pyeloplasty was rst described in 1993 by Schuessler, and since then, it has largely replaced endopyelotomy or open pyeloplasty. It has been adopted as a feasible and reliable treatment for patients with ureteropelvic junction obstruction (UPJO) in the last two decades [1,2] . Laparoscopic pyeloplasty has greater than 90% success rates, and the outcomes of this procedure are equivalent to open pyeloplasty [3] . The advantages of this technique include short convalescence, less usage of analgesic drugs, decreased postoperative pain, low morbidity, short hospital stay, and improved cosmesis [4,5] .
The etiology of UPJO is heterogeneous, including intrinsic disorganization and/or extrinsic compression [6,7] . Moreover, the extrinsic obstruction usually refers to the presence of aberrant lower pole crossing vessels (CVs), brous band adhesions, or kinks in a normal caliber UPJ [2] . The incidence of extrinsic UPJO by CVs was reported to be 38-71% [8][9][10] . In the case of long-term endopyelotomy, the presence of CVs was concurrent with a decrease in the success rate from 82-33% [11] . This suggests that CVs play an important role in the prognosis of UPJO.
The conventional technique chosen by most urologists for the treatment of UPJO caused by pyeloureteral segment, stenosis, renal pelvis volume, or a possible extrinsic compression, was classical Anderson-Hynes (AH) plasty with dorsal relocation of the vessel. However, cases of failed vascular hitch and redo surgery with dismembered pyeloplasty or endopyelotomy have been described [2,8,11] . Depending on the anatomic situation, the new reanastomosis and CVs are not always stress-free [12] . The management of the CVs during laparoscopic pyeloplasty has become controversial and speci c, such as the division, transposition, and suspension of the CVs and the treatment of the relative UPJ.
Currently, there is little evidence in the literature that discusses the different procedures that should be used by urologists when faced with anatomic variability in the association between the renal pelvis and the CVs. Therefore, to aid in the determination of the correct procedure for a given case, we collected and reviewed our cases as well as those in the literature that focused on the treatment of lower pole vessels.
We have characterized the anatomical subtypes of UPJO caused by etiological CV and demonstrated the potential operation procedures for these cases.

Patients
A total of 51 patients with UPJO underwent retroperitoneal laparoscopic dismembered AH pyeloplasty consecutively at our hospital between 2015 and 2019. We collected the characteristics of the patients including age at surgery, preoperative imaging results, operative time, need for conversion, length of hospital stay, need for drainage, and postoperative complications (Table 1). Each patient was evaluated preoperatively with routine blood tests, renal function, urinalysis, diuretic renography, ultrasound, and abdominal computed tomography CT. The diagnosis of UPJO was based on medical history, clinical symptom, and imaging examination. In addition, we conducted angio-CT to evaluate the presence of CVs prior to surgery. There were 13 patients (25.49%) who had a pathogenic CV that compressed the UPJ, causing a potential obstruction, 7 of whom underwent a classic dismembered pyeloplasty (DP) with or without CVs' transposition, and 4 underwent DP with further treatment for CVs.

Surgical Procedure
After general anesthesia, all patients underwent retroperitoneal laparoscopic surgery. Subsequently, the surrounding tissues were meticulously separated, and the CVs were isolated. The vein was routinely clipped and divided. According to anatomical criteria, in cases with the posterior crossing artery, an AH procedure was performed after the CVs were suspended to the psoas muscle with a 3 − 0 suture. In the case of anterior CVs, the patients were selected for different techniques depending on the ndings during the procedure. First, AH pyeloplasty with or without translocation of the artery was performed. In the absence of compression, further maneuver was not needed. However, if the CVs compressed the UPJ to a certain degree even after relocation, the CVs were suspended anterior to the pelvis or posterior to the psoas at least 1 cm away from the anastomosis with a 3 − 0 suture. In case the suspension of CVs did not give satisfactory results, we transected the CVs.
The outcome was assessed during the follow-up period of 3 months after the surgery. The patients were evaluated by ultrasound and for pain 6 months after the surgery. Radiological follow-up was continued yearly with CT. Success was de ned as 80% or greater pain relief according to the VAS score, no sign of obstruction on IVU (patent UPJ), decreasing excretion curve with T1/2 < 12 min, and improved or stable differential renal function on DR.

Statistical analysis
For statistical analysis, Chi-square test and Fisher's exact test were used to compare categorical values.
The independent samples t-test was used to compare numerical variables. A p value of < 0.05 was considered statistically signi cant.

Results
Among the 51 consecutive patients, 13 patients (25.49%) had CVs during the surgery, including 7 men and 6 women. The average age of patients was 30 and 31 years for group 1 and group 2, respectively. The success rate was 92.1% and 100% for group 1 and group 2, respectively. Mean operating time was 197.85 minutes and 233.58 minutes for group 1 and group 2, respectively. All clinical data between the two groups were statistically different, except for a higher incidence of lithiasis in the patients without CVs than in those with CVs (Table 1). One patient nally converted to open surgery to avoid the cumbersome procedure. The mean hospital stay after surgery was 8.49 and 8.85 days for group 1 and group 2, respectively. In the post-operative time, the patients in group 1 had three complications including fever, perirenal effusion, and shedding of the double-J stent. In addition, group 2 had one patient with a fever. During the follow-up period, the subjective feelings of the patients, renal ultrasound at 3 months, and IVU and diuretic renogram at least 6 months after surgery showed 3 patients in the control group with evidence of obstruction. However, all patients with CVs had a substantial renal functional recovery with improvement in hydronephrosis (Fig. 1A, B, C, D). The preoperative data and success rate between the two groups was statistically insigni cant.
According to the intraoperative anatomical relationship between the CVs, the renal pelvis, the UPJ, and the ureter in the cases that we reviewed, we mainly distinguished two different types, the anterior and posterior CVs groups, and this is illustrated in Table 2. The hydronephrosis between the two groups was statistically signi cant. The patients with anterior CV usually had more severe hydronephrosis than those with posterior CVs. To ensure that there is no pressure between the CVs and reanastomosis after the operation, we further summarized at least ve surgical schemes for the rst anatomical position. Meanwhile, there was possibly only one best solution for the second anatomical type. A. After DP, there is no pressure against the UPJ or the reanastomosis: no further treatment was necessary.
B. After DP and relocation of CVs, there is no pressure against the UPJ or the reanastomosis: no further treatment was necessary.
C. CVs located in front of the UPJ can obtain a pressure-free position, but they may still exert pressure on UPJ: the CVs were suspended on the renal pelvis.
D. CVs located behind the UPJ can obtain a pressure-free position, but they may still exert pressure on UPJ: the CVs were transferred to the posterior of UPJ and xed on the psoas.
E. No satisfactory anatomical relationship between the CV and UPJ could be obtained: the CV was transected.
Type 2. The vessels were placed behind the UPJ: the CVs were suspended on the psoas.

Discussion
In the past few decades, there have been controversies whether the ectopic blood vessels is the main cause of UPJO. Due to the development of minimally invasive urological techniques, a more comprehensive understanding has been proposed by urologists regarding the etiology heterotopic perirenal vessels. Although some experts believe that the CVs may not always exert pressure on the UPJ, the decrease in the success rate of endoscopic treatment due to heterotopic blood vessels and the success of the Hellström technique con rm the importance of the heterotopic blood vessels in UPJO surgery. Leavitt et al. reported that the CVs accounted for approximately 22.7% of 601 UPJO cases. Further, they summarized six different deplaning positions between the UPJ and CVs, of which the probability of the CVs compressing the UPJ from the posterior position was 7.4% [13] . These results are consistent with the data obtained in our study (Table 1).
Although routine transposition of CVs during laparoscopic AH plasty is the gold standard for treating UPJO patients with CVs, an increasing number of urologists are questioning the validity of this traditional method. Szydelko T and Ng CS reported that the traditional dorsal transposition of the vessel would not necessarily create a tension-free anastomosis, a better planing structure may be obtained with its cephalad translocation [14,15] . Further, Janetschek G et al. claimed that dorsal displacement of the CVs may worsen the anatomic conditions [16]. In a multi-center study, Chiarenza SF et al. suggest that carefully selected patients with CVs could be treated with vascular surgery [17] . A systematic description of three types of patients according to the different locations of UPJ and ventral CVs were summarized by A. Schneider et al. They suggested that AH pyeloplasty should be performed when the CVs are present anterior to the dilated pelvis or the UPJ with intrinsic stenosis site, while vascular suspension should be performed when the vessels are present posterior to the UPJ, resulting in ureteral kinking [18] .
Although the current procedure of transposing, xing, and suspending the CVs is controversial, the consensus among experts is that after careful dissection during the operation, further selection is made on the basis of the etiology of the CVs, endogenous obstruction, laparoscopic approach, relationship with the surrounding tissues, such as the UPJ, and the potential for obstruction [12,15,19,20] . In our study, the techniques were selected according to the anatomy between UPJ and CVs. Some experts believe that transposition facilitates anastomosis in the transperitoneal approach, and it can provide better anterior access to the renal pelvis and renal CVs [21,22] . In contrast, we believe that the posterior approach can provide us with a clear understanding of whether the CVs exert pressure on the UPJ because the position of the CVs and UPJ is not affected during the surgery with the posterior approach. Moreover, we found that the degree of hydronephrosis for the dorsal CVs was relatively mild compared to that for the anterior CVs (Table 2). In the case of posteriorly located CVs, they can be suspended on the psoas muscle without affecting the subsequent operation (Fig. 2C, D).
As described previously, besides AH pyeloplasty, the tension-free renal pelvis and ureteral reanastomosis after the operation is an important aspect [12] . Zeltser IS et al. showed that CVs located within 5-6 mm of the ureteropelvic junction can cause a problem [23] . Boylu U et al. considered that transposition was performed when the anastomosis after mobilization was less than 1 cm from the CVs [12] . We suggest that the distance between the CVs and the new anastomosis should be at least 1 cm based on the changes in position during and after the operation. Therefore, according to the relationship between the new anastomosis and CVs after pyeloplasty, we have further divided treatment of the CVs into the following ve individual procedures. When there is no pressure on the anastomotic site after pyeloplasty or transposition of CVs to the posterior side, we do not provide any further treatment. When the location of the CV is too close to the new anastomosis, we consider suspending the CV anterior to the pelvis or posterior to the psoas major muscle ( Fig. 2A, B). If none of the above methods meet our requirements, we may consider transecting the vessel. Kelly cut the vessels using laparoscope in 2 cases with clinical improvement and without hypertension [24] . We had a similar case in this study, and the postoperative outcome was commendable during the follow-up. In conclusion, the nal decision was made by the surgeon after visualization and tactile sensation of the new association between the renal pelvis, ureter, and CV in each individual.
The limitations of this study are the small sample size, the subjective judgment of intrinsic pathogeny, and the possibility of complex and rare vascular anatomic position. Further studies will be carried out to validate our classi cation through intraoperative opaci cation and postoperative pathological examination and con rm the intrinsic stenosis. A prospective and multicentric study will be conducted with increased sample size. A comprehensive evaluation of the relationships between blood vessels and UPJ according to the individual situation is necessary to provide the appropriate treatment plan.

Conclusion
CVs compression is one of the common causes of ureteropelvic junction obstruction. -Consent to publish The images are entirely unidenti able and there are no details on individuals reported within the manuscript.
-Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

-Competing interests
The author declares that they have no competing interests.