Introduction

Ureteropelvic junction obstruction (UPJO) is a common urologic condition resulting in impeded urine flow from the pelvis to the proximal ureter and progressive dilation of the renal collecting system. The obstruction can be either of congenital or acquired origin [1]. The typical clinical presentation of an UPJO is renal colic and back pain, which can be exacerbated by increased fluid intake or infections [1, 2]. Indications for surgical treatment are recurrent flank pain, kidney stone formation, recurrent infections, and loss of renal function [3].

The Anderson–Hynes technique was first described in 1949 [4] and the first laparoscopic pyeloplasty was described in 1993 [5]. The advantages of the laparoscopic technique are fewer postoperative complications, less pain, and shorter hospitalizations comparable to open pyeloplasty [6]. In 2002, the first robot-assisted laparoscopic Anderson–Hynes pyeloplasty (RALP) was described [7].

Postoperative [8,9,10,11,12,13] and renal function outcomes [8,9,10,11,12,13,14,15,16] have been reported in several publications, but the effect of RALP in patients treated because of symptoms, infections, or stone formation is sparse [17]. The objective of this study was to evaluate the safety of RALP for UPJO and report outcomes 1 year after the procedure.

Material and methods

We performed a single-center retrospective cohort study where we included patients aged 16 years or older who underwent RALP performed on the DaVinci Si system (Intuitive Surgical, Inc.) from June 2016–December 2021.

The procedure was performed with incision of the renal pelvis and subsequent spatulation of the ureter along the lateral surface approximately 5 mm beyond the obstructed part. In cases with crossing vessels, the ureter was moved anteriorly to the vessels. The anastomosis was performed with vicryl 4-0 or monocryl violet 4-0 sutures. After completing the posterior wall of the ureter, a double-J ureteral stent was inserted antegraded though the trocar using a guidewire and its placement in the bladder was confirmed with reflux of blue dyed water [18]. The double-J ureteral stent was only changed in patients who had a stent prior to RALP in case of incrustations, if it had been in place for more than 4 months if the surgeon deemed it necessary. The double-J stent was routinely removed 4 weeks after RALP in the out-patient clinic using a flexible cystoscope.

In case of kidney stone, extraction was performed through a laparoscopic port using a flexible ureteroscope. All patients received one dose of prophylactic IV antibiotic perioperative.

Patients were identified from electronic records by the surgical code for RALP. The following data were recorded following electronic patient chart review: demographic information including age, gender, smoking status, BMI, ASA score, and Charlson comorbidity index (CCI); preoperative information including flank pain, infection, recurrent stone formation, preoperative decompression with double-J stent or nephrostomy, CT scan findings (presentation of crossing vessel) and results of last MAG3 renal scan before RALP. We have exclusively noted the primary indication for treatment, despite some patients naturally presenting with more than one problem relating to the UPJO. In case of preoperative decompression, this was kept until definitive RALP. Operative details included: operations time, crossing vessels, and perioperative complications. Postoperative details included: length of stay, time with double-J ureteral stent, and complications within 30 days of surgery.

Complications were classified according to the Clavien–Dindo classification (CD) [19]. Postoperative success was evaluated 1 year after the procedure. Postoperative renal function was assessed with MAG3 renal scan and recorded as improved, stable or deteriorated. Symptoms, defined as pain, were recorded as improved, unchanged or aggravated. Patients treated because of stones underwent CT scan following 1 year. Surgeries were defined as successful if patients treated because of deteriorating renal function had stable or improved renal function, if symptoms had improved in patients treated for pain, while for patients with stones or infection, success was defined as being stone- or infection-free with stable or improved kidney function at 1-year follow-up.

Finally, recurrent UPJOs and their treatments were also recorded.

The study received ethical and legal approval from the regional center for register research of the Capital Region of Denmark according to Danish law (journal number: R-22013482). The manuscript was prepared according to the STROBE statement (www.strobe-statement.org).

Statistics

Descriptive statistics were used. Median follow-up was calculated with the reverse Kaplan–Meier method [20]. Statistical analysis was performed with Excel and R version 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Overall, 208 patients were identified based on the surgical code for RALP. Of these, 194 patients were included in the analysis of operative outcome and postoperative complications, while 182 patients were available for 1 year postsurgical evaluation. The following 14 patients were excluded: 2 due to different pyeloplasty techniques being used (YV: n = 1, a.m. Heineke–Mikulicz: n = 1), 8 who underwent a re-do pyeloplasty, and 4 with no postsurgical follow-up data at all.

The median age at the time of surgery was 38 years (IQR 27–58) and 111 (57%) were female. The median BMI was 23 (IQR 21–26), the median ASA score was 1 (IQR 1–2) and the median CCI was 0 (IQR 0–2), Table 1. The median follow-up time following surgery was 4.5 (IQR 3.0–6.0) years.

Table 1 Baseline characteristics of 194 patients undergoing RALP for UPJO
Full size table

The primary indication for RALP was deteriorating renal function in 88/194 patients (45%), of which 61/88 patients (69%) had a double-J ureteral stent inserted prior to RALP to protect the renal function. Flank pain was the indication for 70/194 (36%), of which 27/70 patients (38%) had a double-J stent prior to RALP to decrease pain. Infection was the indication for 22/194 (11%) patients, of which 17 patients required a nephrostomy or double-J stent as a treatment for pyonephrosis prior to RALP. In 11/194 patients (6%), recurrent formation of kidney stones was the indication. Earlier stone surgery induced a UPJO resulting in flank pain for two patients and one patient was evaluated because of flank pain and diagnosed with a UPJO following ureteroscopy under the suspicion for upper urothelial cancer.

The median operation time was 134 min (IQR 112–159) and 119/194 (61%) had crossing vessels. The median length of stay was 2 days (IQR 2–2) and the median time with double-J stent postoperatively was 24 days (IQR 22–27).

No major perioperative complications occurred and none of the procedures were converted to open surgery. Twenty-four of one hundred ninety-four patients (12%) experienced a CD I complication, twenty-six patients (13%) experienced a CD II, and fifteen patients (8%) experienced a CD grade IIIa or IIIb complication, Table 2. Of note, none of the patients who received IV antibiotics for postoperative fever had positive blood culture. Two patients developed abscesses after 5 and 7 days, respectively, and required percutaneous drainage. Five patients required nephrostomy to treat uroplania. In six patients, the double-J ureteral stent was displaced and removed with a ureteroscope.

Table 2 Overview of complications classified according to the Clavien–Dindo score
Full size table

The renal function following 1 year could be assessed in 85 of the 88 patients who were treated because of deteriorating renal function. In total, 78/85 (92%) had stable or improved function following the procedure, while 8% lost additional renal function. Effect on symptoms following 1 year could be assessed in 63 of the 70 patients who were treated because of flank pain. In total, 78% (49/63) had improvement of their symptoms, while symptoms were unchanged in 21% (13/63) and 1% (1/63) had aggravation. Eighteen of twenty-two (82%) patients treated because of infection had not had a recurrent infection at 1-year follow-up. One patient experienced pyonephrosis within 30 days, but also had stone formation that was not removed during RAPL. After retrograde intrarenal stone removal, the patient did not experience any additional infections. Three patients still suffered from lower urinary tract infections but did not experience pyelonephritis. Seven of nine patients (78%) were stone free at the 1-year follow-up—two patients were lost to follow-up. In one patient, stone extraction was not possible due to narrow calyces. Another patient had an unchanged number of small kidney stones at follow-up.

In total, 14/194 (7%) patients had a recurrent UPJO during follow-up, of which 4 patients had a new intervention: 2 patients underwent re-do RALP, 2 patients underwent endopyelotomy, with one of these failing endopyelotomy and requiring a subsequent RALP. In total, seven patients lost all renal function on the affected kidney. Two of these underwent a subsequent nephrectomy because of pain.

Discussion

In this retrospective study of RAPL, we found 92% success rate for patients treated because of deteriorating renal function. Comparable numbers were: 78% for patients treated because of pain—with 1% experiencing aggravation of their symptoms—82% for patients treated because of infections, and 78% for patients treated because of kidney stone. The procedure was safe with few major (CD III) and no serious (CD IV-V) complications.

The main limitation of this study is the retrospective design. Renal outcomes, which were measured objectively with MAG3 renal scan, are more likely to be generalizable than the symptomatic outcomes which was measured by retrospective patient chart review and could be affected by re-call biases [21].

Treatment of UPJO has rapidly developed over the last 2 decades from open pyeloplasty to minimal invasive surgery [6, 7]. Laparoscopic pyeloplasty and RALP are associated with reduced pain, fewer postoperative complications, and shorter hospital stay compared to an open procedure [22, 23]. Other advantages of RALP are three-dimensional vision, tremor filtering, motion scaling, and articulating movements with greater degrees of freedom [23,24,25]. Because of these advantages, RALP has shown a faster learning curve, with fewer cases needed to achieve a plateau in operation time compared to laparoscopic pyeloplasty [26].

The observed CD I and II complications are higher than previous studies [9, 27], whereas the 7% major complications are in line with the 3–11% major complications in previous reports [9, 12, 17]. Minor complications such as fever and urinary tract infections are manageable and expected. Moreover, we registered postoperative pain as a CD I complication where other studies potentially have not included this as a complication [9, 28].

Following 1 year, 92% of patients treated because of deteriorating renal function had improved or stable function following the procedure, while 8% lost additional renal function. Previous studies report 69–100% renal success rates with the majority between 90–95% [8,9,10,11,12,13,14,15,16, 18, 29]. In patients treated because of pain, 78% had fewer symptoms, which is lower than the only previous study reporting on symptom outcomes [17]. This study found a 94% improvement of symptoms in a cohort of 88 patients but did not specify when symptoms were recorded postoperatively [17].

We used the last renal scintigraphy before pyeloplasty in patients who had a double-J ureteral stent prior to the procedure because we wanted to compare the renal function following 1 year with the status at the time of the procedure to be able to fairly counsel our patients as to what they can expect from undergoing the procedure. Thus, it would be interesting to investigate whether a preoperative double-J stent and duration of the decompression affect the success rate. Unfortunately, our cohort is too small to perform any reliable analysis on this topic.

We found 7% had a recurrent UPJO during follow-up, which previously has been reported between 2–4% [9, 12]. Finally, in line with previous reports, 3.6% of the patients ended up with a solitary functioning kidney [9, 13].

Conclusion

Robot-assisted laparoscopic Anderson–Hynes pyeloplasty demonstrates an acceptable safety and a high renal success rate and moderate symptomatic success rate in treating UPJO with few patients having deteriorated renal function or aggravation of symptoms. However, we were unable to find the same impressive improvement in pain-related symptoms as a previous study and more than 1/20 experienced a recurrence of UPJO. Thus, the surgical treatment of UPJO warrants further investigation preferably in prospective design with validated objective symptom score.