Crossing Vessels in Retroperitoneal Laparoscopic Pyeloplasty: Anatomy and Outcomes

Abstract

Purpose:

Crossing vessels are commonly encountered during pyeloplasty for ureteropelvic junction obstruction (UPJO) and may influence surgical strategy. We aimed to describe the anatomical distribution of crossing vessels and their management options during retroperitoneal laparoscopic pyeloplasty (RLP) and report the long-term outcomes.

Materials and Methods:

We retrospectively reviewed consecutive adult patients who underwent RLP for UPJO between 2006 and 2025 by a single surgeon. Preoperative computed tomography (CT) or magnetic resonance (MR) urography was routinely performed to assess vascular anatomy. Surgical technique emphasized identification of crossing vessels, routine UPJ reconstruction, mobilization and/or transposition of arteries, selective ligation of nonessential veins, and creation of a dependent tension-free anastomosis.

Results:

The study included 165 patients with mean age of 58.2 years. Crossing vessels were identified in 114 patients (69.1%), most commonly gonadal vein variants (45.6%) and accessory renal arteries (31.6%). Mean operative time was 175 minutes, with no conversions to open surgery. Only 19 (16.67%) patients required ureteral transposition. Postoperative Clavien-Dindo grade III complications occurred in 5 (3%) patients, with no grade IV or V events. With median follow-up of 64 months, overall success rate was 96.1%. None of the variables were statistically linked to failed pyeloplasty (P > .5 for all variables).

Conclusions:

RLP with systematic identification of crossing vessels is associated with excellent long-term functional outcomes and low morbidity. Management of crossing vessels should be individualized according to the type and relation of the crossing vessel to the UPJ. Ureteral transposition should not be a routine part of RLP.

INTRODUCTION

Ureteropelvic junction (UPJ) obstruction (UPJO) represents the most common congenital anomaly of the ureter.¹ The management paradigm for UPJO has undergone substantial transformation over the past 3 decades. Open dismembered pyeloplasty, first described by Anderson and Hynes in 1949, long remained the gold standard with success rates exceeding 90%.² The advent of minimally invasive approaches beginning with laparoscopic techniques and evolving into robotic-assisted platforms has revolutionized surgical management while maintaining comparable success rates.³

Approximately 50% of cases of UPJO are associated with crossing vessels. The clinical significance of crossing vessels has evolved from incidental anatomical findings to recognized etiological factors in both primary and secondary UPJO.^1,4 Despite technological advancements in management of UOPJO, the optimal management of crossing vessels during pyeloplasty remains a subject of ongoing investigation, particularly regarding vessel transposition versus division and the impact on long-term outcomes.⁵

Recent literature has emphasized the importance of meticulous vascular anatomy assessment using preoperative computed tomography angiography (CTA) or magnetic resonance urography (MRU), with reported sensitivity exceeding 90% for detecting crossing vessels.^6,7 However, controversy persists regarding whether all crossing vessels require transposition and whether their presence affects surgical outcomes.^5,8,9 For better understanding of the landscape for treatment of UPJO with crossing vessels, there emerges a critical need for a comprehensive understanding of the anatomical intricacies surrounding crossing vessels. While previous investigators touched upon vessel orientation in relation to the UPJ, a detailed anatomical exploration has been notably absent from the discourse.¹⁰

While many laparoscopic series, especially those favoring the transperitoneal approach, advocate for ureteral transposition following the Anderson-Hynes principle, the advent of retroperitoneal laparoscopy provided a unique perspective.¹¹ The retroperitoneal approach provides a mirror image of the vessel and UPJ arrangement seen in transperitoneal approach. Notably, given the typical anterior location of crossing vessels, the retroperitoneal approach often avoids their obstructive presence, sparking ongoing debates on the necessity of ureteral transposition during laparoscopic pyeloplasty.^9,12

Addressing these gaps, we aimed to meticulously review a large single surgeon's experience with retroperitoneal laparoscopic pyeloplasty (RLP) to determine the incidence of crossing vessels and unravel their anatomical origins and their management in the context of the retroperitoneal approach, shedding light on its potential advantages over the traditional transperitoneal approach in laparoscopic pyeloplasty.

MATERIALS AND METHODS

We conducted a comprehensive retrospective analysis of consecutive cases of RLP performed by a single surgeon between 2006 and 2025. Institutional review board approval was obtained.

• Patient Selection and Preoperative Evaluation: All patients were evaluated with cross-sectional imaging (CT urography or MRU) to confirm UPJO, delineate crossing vessel anatomy, and identify concomitant pathology (e.g., stones, parapelvic cysts). Diuretic renal scintigraphy was routinely used for functional assessment.

• Surgical Procedure: Patients were initially placed in the modified dorsal lithotomy position for routine cystoscopy, retrograde pyelography, and placement of a double-pigtail ureteral stent. They were then repositioned in the flank position. A standard retroperitoneal, either 3-port or 4-port, approach was employed. Retroperitoneal access was achieved via a 10-mm incision off the tip of the 12th rib. A balloon dilator was used to develop the working space, followed by placement of a posterior 10-mm and an anterior 5-mm port. A fourth assistant port was used selectively.

• Surgical Tenets: The surgical principles guiding our approach included meticulous dissection of the UPJ to define crossing vasculature, preservation and repositioning of all crossing renal arteries, division of unessential tethering venous anomalies, and the creation of a dependent collecting system funnel beneath the crossing vasculature. Pyeloplasty (reconstruction of the UPJ) was routinely performed while ureteral transposition (from anterior to posterior) was not a routine practice unless necessitated by the presence of a posterior crossing vessel. Our ultimate reconstructive goal was the establishment of the classic renal hilar arrangement, with the renal pelvis positioned as the most posterior dependent structure.

• Postoperative Management and Follow-Up: The foley catheter was typically removed on postoperative day 1 or 2, and surgical drains were removed before discharge, usually on day 2. Ureteric stent was usually removed after 4 to 6 weeks. Follow-up evaluations were conducted at 3, 6, 12, and 24 months postoperatively, involving postoperative ultrasonography and diuretic renal scans.

• Outcome Assessment: Treatment success was defined as both symptomatic relief and improved drainage on diuretic renography. All complications within 90 days were graded using the Clavien-Dindo classification.

Statistical Analysis

Sample characteristics were described using descriptive statistics; categorical variables were presented as absolute numbers and percentages while continuous variables were presented as means and standard deviations (or medians and ranges when abnormally distributed). Two-sample t-test (or Wilcoxon rank-sum test when appropriate) were used to assess the differences in continuous variables while χ² tests or Fisher’s exact tests were used for categorical variables according to the expected cell counts. Multivariate logistic regression models were used in multivariate analysis. The significance level was set at a P-value <.05.

RESULTS

The study included 165 patients who underwent laparoscopic pyeloplasty for UPJO between 2006 and 2025. The baseline characteristics including demographics, clinical presentations and urological history of the study population are shown in Table 1.

Table 1.

Baseline Patient Characteristics and Perioperative Data

Parameter	Overall Cohort (N = 165)
Age (Years) Mean ± SD	58.2 ± 15.3
Sex, n (%)
Male	70 (42.4%)
Female	95 (57.6%)
Affected side, n (%)
Right	83 (50.3%)
Left	82 (49.7%)
UPJ pathology classification, n (%)
Primary	149 (90.3%)
Secondary	16 (9.7%)
Presenting symptoms, n (%)*
Flank pain	110 (66.7%)
Incidental discovery	31 (18.8%)
Hematuria (micro/macro)	19 (11.5%)
Recurrent pyelonephritis /UTI	30 (18.1%)
Renal insufficiency/failure	5 (3%)
Previous urologic interventions, n (%)*
None	45 (27.5%)
Preoperative stent placement	38 (22.9%)
Previous endopyelotomy	8 (4.6%)
Previous nephrostomy tube	10 (6%)
Contralateral pyeloplasty	3 (1.8%)
Contralateral nephrectomy	5 (3%)
Retroperitoneal lymph node dissection	1 (0.6%)

*Patients may be present with multiple symptoms; percentage do not sum 100%.

All procedures were performed retroperitoneally, using 3 ports (77%) or 4 ports (23%). Mean operative time was 175.4 ± 25 minutes. No open conversions occurred. Crossing vessels were identified in 114 (69.1%) patients. Concomitant procedures included stone extraction (9.1%), cyst excision (3.64%), partial nephrectomy (3%), and lymph node dissection (1.8%). Median (interquartile range [IQR]) hospital stay was 2 (1–3) days while median (IQR) stent duration was 4 (4–6) weeks. Calvien-Dindo Grade III complications occurred in 5 (3%) patients; with no grade IV/V complications (Table 2).

Table 2.

Perioperative Outcomes of the Study Population

Operative Parameter	Value (N = 165)
Port configuration, n (%)
3-port technique	127 (77%)
4-port technique	38 (23%)
Operative time, minutes, mean ± SD	175.4 ± 25
Concomitant procedures, n (%)
Ipsilateral stone extraction	15 (9.1%)
Parapelvic cyst excision	6 (3.64%)
Partial nephrectomy	5 (3%)
Lymph node dissection	3 (1.8%)
Open conversion rate, n (%)	0 (0%)
Intraoperative complication, n (%)	2 (1.2%)
Estimated blood loss, mL, median (IQR)	100 (50–150)
Hospital-stay, days, median (IQR)	2 (1–3)
Overall complication rate, n (%)	21 (12.7%)
Clavien-Dindo classification, n (%)
Grade I	8 (4.8%)
Grade II	8 (4.8%)
Grade III	5 (3%)
Grade IV	0 (0%)
Grade V	0 (0%)

Among patients with crossing vessels (n = 114), 140 vessels were documented (mean 1.23 ± 0.45 per patient). The list of vessels encountered and their respective incidences are shown in Table 3. Gonadal vein variants were most common (45.6%), followed by accessory renal artery (31.6%) and accessory renal veins (28.9%). Unique variants included right gonadal vein draining to accessory renal vein (n = 4) and trifurcating left gonadal vein complex (n = 6). Single vessels were found in 59.6%, multiple vessels in 40.4%.

Table 3.

Anatomical Distribution and Characterization of Crossing Vessels (N = 114)

Vessel Type
Venous variants, n (%)
Gonadal Vein Anomalies*	52 (45.6%)
Accessory renal vein	33 (28.9%)
Main renal vein	3 (2.6%)
Mesenteric vein	4 (3.5%)
Lumbar vein	3 (2.6%)
Arterial variants, n (%)
Accessory renal artery	36 (31.6%)
Main renal artery	6 (5.3%)
Gonadal artery	3 (2.6%)
Lumbar artery	1 (0.9%)
Complex anatomical patterns
Single crossing vessel	68 (59.6%)
Multiple crossing vessel	46 (40.4%)

We divided the gonadal vein group into several entities: complex gonadal veins (CGV), gonadal vein plexuses (GVP) and accessory/multiple gonadal veins (AGV). Amongst the CGV subset, we noted singly branching gonadal veins with the branch causing a UPJO. We also noted gonadal veins branching into lumbar veins as well as branches to nearby mesenteric veins. Trifurcating gonadal veins, an anomaly that we identified in 6 patients often included renal vein branches, as well as mesenteric and lumbar branches. GVP anomalies describe multiple, small, spidery branches from the gonadal vein, inferior vena cava, or occasionally from mesenteric vessels all coalescing at the UPJ. A venous anomaly of particular interest was the right accessory gonadal vein arising from an accessory renal vein on the right side. This vascular anomaly, which we noted in 4 patients, certainly represents a departure from conventionally described anatomy where the gonadal arises directly from the vena cava. An arterial anomaly of interest, and a subset of the accessory renal artery, was a lower pole artery arising from the level of the iliac bifurcation. This was encountered in a single patient.

Among the patients with crossing vessels (no = 114), only 19 (16.67%) patients had ureteral transposition through Anderson-Hynes dismembered pyeloplasty while other patients did not require transposition either due to crossing vessels being already anteriorly located or UPJ repositioned downward with the vessel positioned upwards away from the UPJ level.

On outcome assessment, 12 patients were excluded due to lack of follow-up. On assessment of outcome for a total of 153 patients with median follow-up of 64 months, overall success rate was 96.1% (147/153). When analyzing the patient with failed repair, univariate analysis showed none of the factors including age, crossing vessels, operative time, or concomitant procedures predicted failure was associated with failed pyeloplasty (all P > .05) (Table 4).

Table 4.

Predictors of Surgical Outcome: Univariate Analysis (N = 153)

Variable	Success Group (n = 147)	Failure Group (n = 6)	P-Value
Age ≥60 years, n (%)	53 (36.1%)	3 (50.0%)	.45
Female sex, n (%)	85 (57.8%)	3 (50.0%)	.72
Secondary UPJ, n (%)	13 (8.8%)	1 (16.7%)	.44
Presence of crossing vessel, n (%)	108 (73.5%)	6 (100%)	.34
Multiple crossing vessel, n (%)	44 (29.9%)	2 (33.3%)	.87
OR time >180 minutes, n (%)	58 (39.5%)	4 (66.7%)	.21
Concomitant procedures, n (%)	15 (10.2%)	2 (33.3%)	.21
Major complications (grade ≥III), n (%)	4 (2.72%)	1 (16.7%)	.12
Stent duration >4 weeks, n (%)	78 (53.1%)	4 (66.7%)	.69

DISCUSSION

Laparoscopic and robotic assisted pyeloplasty have been the gold standard minimally invasive treatment option for PUJO. In a systemic review study, Uhlig et al reported that laparoscopic and robotic assisted pyeloplasty have shown higher success rate and lower complication rate in comparison to open pyeloplasty and endopyelotomy.¹³ The comparison between retroperitoneal and transperitoneal laparoscopy in the context of pyeloplasty has sparked debates, particularly regarding their efficacy in defining crossing vessels.¹⁴ In a recent meta-analysis, Li et al reported that RLP fared better than transperitoneal approach in terms of higher success rate and lower complication rate.¹⁵ Another meta-analysis reported that transperitoneal and retroperitoneal approaches had similar success rate, hospital stay, overall complication rate however, retroperitoneal approach was associated with longer operative time.¹⁶

Our findings align with the notion that retroperitoneal approach enhances sensitivity by providing a clear anatomical view of the UPJ and related vessels through the elevation of the kidney’s lower pole. This heightened sensitivity, we posit, contributes to our observed high incidence of crossing vessels, potentially influenced by evolving preoperative imaging practices that increasingly identify these structures. Our study echoes the work of Rassweiler et al, who reported an overall crossing vessel incidence of 52% in UPJ treatment.⁹ However, when employing retroperitoneal laparoscopy, the incidence surged to 71%, a figure remarkably close to our own 69.1%. Our relatively high incidence of crossing vessels may reflect the inherent advantages of the retroperitoneal approach where prompt access to the hilum may help in defining vascular anomalies. Existing anatomical studies, quoting a 65% incidence of vessels adjacent to the ventral surface of the UPJ, align with the varied rates reported in different laparoscopic pyeloplasty series, ranging from 42% to 57%.^9,17–19

In our investigation, gonadal venous origin emerged as the most prevalent type of crossing vessel, followed closely by lower pole arteries, the archetypal crossing vessel. Intriguingly, we uncovered several unique venous anomalies, notably the trifurcating gonadal vein, giving rise to renal, mesenteric, and lumbar branches. Despite advancements in preoperative imaging's sensitivity to crossing vessels, venous anatomy remains less predictable than arterial structures. This highlights the critical role of surgical exploration in comprehending the intricate venous arrangement surrounding the UPJ.

Our team’s extensive experience in retroperitoneal renal surgery, garnered through nephrectomy and partial nephrectomy procedures, has honed our familiarity with retroperitoneal vascular dissection. Guided by the tenets of our repair, we emphasize mobilizing the neo-UPJ away from preserved arteries rather than routinely transposing crossing vessels. This nuanced approach requires meticulous dissection and division of tethering venous anomalies to achieve effective mobilization.

In line with other retroperitoneal surgeons, our inclination has been to avoid transposition of anterior crossing vessels. Notably, instances of UPJ transposition around a vessel were reserved for posterior crossing vessels, transposed ventrally. This strategy aligns with findings by other investigators who noted a 12% incidence of vessels crossing posteriorly, recommending ventral transposition.

Rassweiler et al’s algorithmic approach to UPJ obstruction, distinguishing between anterior and posterior crossing vessels, has implications for surgical management. While posterior vessels may warrant dismembered pyeloplasty with anterior transposition, anterior vessels often align with Y-V pyeloplasty, unless specific anatomical considerations necessitate a different approach.⁹ In our series, anterior vessels were predominant, a trend consistent with Rassweiler et al’s findings and potentially influenced by the choice between transperitoneal and retroperitoneal approaches.

In essence, the retroperitoneal approach's distinct advantages, especially in defining vascular anatomy, positions favorably for pyeloplasty involving crossing vasculature. The classical renal hilar arrangement, with vessels anterior and the renal pelvis posterior, remains our reconstructive goal, emphasizing the importance of surgical approach in achieving optimal outcomes.

The current study reports an overall success rate of 96.1%, which is consistent with the reported success rate for laparoscopic pyeloplasty that varies between 87% and 98%.^20–23 Whereas some studies reported risk factors associated with failed pyeloplasty that include low preoperative split renal functions (<25%), old age, prior failed endopyelotomy, and early urine leak, none of the variables predicted failure in the current study group.²⁴

This study has significant advantages that include a large cohort performed by a single surgeon with large experience, a standardized technique of RLP and long-term follow-up that exceeds 5 years. Also, the current study is the first to describe the anatomical details of the crossing vasculature associated with UPJO. The limitations of the study include its retrospective design with inherent selection and information biases. Additionally, the single-surgeon design may limit generalizability.

CONCLUSIONS

RLP offers durable and reproducible outcomes for UPJO with inherent advantages where prompt access to the hilum helps defining vascular anomalies and identifying crossing vessels. Meticulous retroperitoneal vascular dissection and varied schemes for UPJ reconstruction facilitate long-term success and help selecting patients for ureteral transposition.