National Practice and Outcomes of Laparoscopic Pyeloplasty in the United Kingdom

Abstract

Background and Purpose:

Laparoscopic pyeloplasty is now widely practiced in the United Kingdom and considered the gold standard in the treatment of patients with ureteropelvic junction obstruction. The aim of this audit was to determine the national practice and outcomes for this procedure.

Patients and Methods:

The British Association of Urological Surgeons sent out standardized audit proformas in May 2008 to units across the United Kingdom inviting surgeons who were performing laparoscopic pyeloplasties to submit their results over the last 4 years. Data on the presentation, preoperative investigations, intra-perative details, and postoperative follow-up were collected centrally and inserted into a national database for analysis.

Results:

There were 323 returns from a total of 30 surgeons. At a median follow-up of 4 months (1–24), the overall symptomatic and renographic failure rates were 10.3% and 8.7%, respectively. Mean operative time was 181 minutes (3–-425 min); there were 18 (6%) conversions, 33 (10.5%) complications, and one (0.3%) mortality. Surgeons who submitted 10 or more returns had a lower conversion rate than surgeons submitting fewer than 10 (2.9% vs 14.7%). The median hospital stay was 3 days (1–34 d). There was no difference in failure and complication rate for the retroperitoneal and transperitoneal approaches, although the conversion rate was higher with the retroperitoneal approach.

Conclusions:

The results show that laparoscopic pyeloplasty, although achieving acceptable outcomes at a national level in the United Kingdom, had areas of practice that could be improved. It highlights the importance of a high-volume practice in achieving optimum results and the potential problems associated with the retroperitoneal approach.

Introduction

L aparoscopic pyeloplasty has become an accepted form of treatment for patients with ureteropelvic junction obstruction (UPJO) with equivalent outcomes to open pyeloplasty and the additional benefits attributed to minimally invasive surgery.^{1

–10} It is now almost two decades since its introduction to urological practice.⁹ In developed countries, it can now be regarded as the treatment of choice, superseding open surgery as the gold standard. Nevertheless, the laparoscopic approach necessitates a different set of complex skills that, although they may now be familiar to the urology community in the United Kingdom, are still relatively new to nonspecialized units by comparison with open surgery. It is therefore desirable to have information about the national practice of this widespread technique and the outcomes assessed regularly.

The Section of Endourology of the British Association of Urological Surgeons (BAUS) carried out this audit with the objective of obtaining such data to provide a baseline for future monitoring of practice and a benchmark by which surgeons can compare their own practice and outcomes. This may allow the identification of potential discrepancies that may need to be addressed, thus ensuring safe use of the technique and compliance with internationally accepted standards.

Patients and Methods

Standards for laparoscopic pyeloplasty were initially determined from contemporary national and international data published in the literature. The outcomes from the larger series are summarized in Table 1. Standardized BAUS laparoscopic pyeloplasty audit proformas were sent out in May 2008 to all adult urology units within the United Kingdom (UK) inviting surgeons to provide data on each case performed over the last 4 years. Data about the patient's demographics, preoperative assessment, intraoperative details, and postoperative follow-up and complications were obtained. The information was then entered into a national database for analysis, and comparisons were then made with the accepted standards. Statistical analyses were performed with the Student t test and the Fisher exact test where appropriate. A P value of less than 0.05 was considered statistically significant.

Table 1.

Outcome of Laparoscopic Pyeloplasty Series in the Literature

Study	Number (n)	Follow-up (months)	Success rate (renographic)	Mean operative time (min)	Conversion rate	Complication rate
Pouliot¹	111	–	86%	128	0.9%	19%
Srivastava²	185	39	94.3%	180	4.8%	9.6%
Symons³	118	12.4	94.5%	205	0%	0%
Yanke⁴	116	20	88.8%	N/A	N/A	N/A
Maynes¹⁰	92	13	92%	250	–	3%
Rassweiler⁷	143	63	94.4%	124	0.6%	6.3%
Moon⁸	170	12	96.2%	140	0.6%	7.1%
Davenport⁵	83	15–35	88%	209–224	2%	14%
Inagaki⁶	147	24	95%	246	0%	8.8%

Results

Demographics

There were 323 completed returns from a total of 30 surgeons over the audit period. Median number of returns per surgeon was seven (range 1–35). Some information was missing from a number of these patients, and, as a consequence, in the analysis below, not all figures add up to 323.

The mean age of the patient was 38.3 years (range 12–84 y). The sex distribution was 137 (42%) males to 172 (53%) females.

Clinical presentation

Table 2 displays the presenting symptoms of the patients. Of 282, 281 (99.6%) patients underwent preoperative diuresis renography. The mean preoperative relative function was 39.4% (range 4–100%). In addition, 163 (50.4%) patients had intravenous urography, 34 (10.5%) CT, and 26 (8%) ultrasonography.

Table 2.

Presenting Symptoms of All Patients

Presenting symptoms	Number (%)
Loin pain	282 (87%)
Urinary tract infection	28 (9%)
Incidental	17 (5%)
Stones	12 (4%)
Hematuria	9 (3%)
Other	17 (5%)

Ten patients (3%) had a previous therapeutic procedure for UPJO on the same side. Of these patients, three (1%) had a previous pyeloplasty, four (1.2%) a balloon dilation, and three (1%) an endopyelotomy. In a further 12 (3.7%) patients, a stent was inserted previously, while three (1%) had a nephrostomy tube inserted.

Operative details

Of the 308 patients, 304 (98.7%) had a laparoscopic Anderson-Hynes dismembered anastomotic pyeloplasty. Three of the remaining had a V-Y pyeloplasty, while one had a nondismembered unspecified procedure. The transperitoneal approach was used in 228 (73%) of cases while 85 (27%) were approached through retroperitoneal access.

There was a mentor present in 10 (3%) of the procedures. One hundred and thirty-six (43%) procedures were performed on the left kidney and 181 (57%) on the right. A crossing vessel was encountered in 137 (45.6%) of patients.

A drain was inserted in 311 (98%) of cases for a median of 2 days (range 1–14 d). All patients had a urethral catheter for a median of 2 days (range 1–14 d). One patient did not have a stent inserted because of technical difficulties with inserting it antegradely before completion of the anastomosis; this patient subsequently needed a nephrostomy tube postsurgery. For the rest of the patients, the stent was left in for a median period of 6 weeks (range 3–28 wks). The single patient who had a stent in for 28 weeks had a postoperative urinoma.

The mean “skin-to-skin” operative duration was 181 minutes (range 35–420 min). There were 18 (6%) conversions to an open procedure. Table 3 displays the indications stated for the conversion.

Table 3.

Documented Reasons for Conversion

Reasons for conversion	Frequency
Difficulty with anastomosis	3
Adhesions	2
Failure to progress	1
Kidney too medial	1
Distortion of anatomy from previous PCNL	1
Fibrotic pelvis from stones	1
Pyonephrosis	1
Portal hypertension and varices	1
Perineal fibrosis	1
Achondroplasia	1
Not stated	5
Total	18

PCNL=percutaneous nephrolithotomy.

Complications

Morbidity was documented in 46 (14%) cases. Of these, 15 (5%) were attributed to the stent. Ten patients were stated to have had a postoperative urinary leak, diagnosed by either excessive urine in the drain or the development of an urinoma. Six (2%) of these were classified as “simple,” resolving spontaneously, while four (1%) were classified as “complex,” necessitating additional intervention. The interventions were a postoperative nephrostomy and anterograde stent insertion in one patient in whom intraoperative anterograde stent insertion had failed, and a repositioning of the stent for another patient. In a further two patients, no description of the intervention was provided, despite a request. There was one (0.3%) mortality from pseudomembranous colitis secondary to Clostridium difficile within 30 days of the operation. Table 4 shows the total major and minor complications, excluding stent-related morbidity.

Table 4.

Frequency of Major and Minor Complications

Major	Frequency	Minor	Frequency
Complex urine leak	4	Simple urine leak	6
Port site hernia	2	DVT	2
Duodenal perforation	1	Pyrexia	2
Lost needle	1	Hematuria	1
Perinephric abscess	1	Ileus	1
Liver failure, ascites	1	Paresthesia—spontaneous recovery	1
Retroperitoneal bleed	1	Port site wound infection	1
Myonecrosis of gluteal	1	Renal colic from VUJ stone	1
Bleeding necessitating exploration	1	Nephrostomy for failed stent	1
Death (pseudomembranous colitis)	1	Repositioning of stent	1
		Shortness of breath	1
		Subcutaneous emphysema	1
		Urinary tract infection	1
Total	14 (4.3%)		20 (6.2%)

DVT=deep vein thrombosis; VUJ=vesicoureteral junction.

Tables 5 and 6 compare the retroperitoneal and transperitoneal approaches in this audit. The operative durations were similar (181 vs 176 min, P=0.22), but the conversion rate was higher for the retroperitoneal approach (10.5% vs 3.9%, P=0.028). The complication rate was also higher, but this was not statistically significant (14% vs 9.6%, P=0.18). There was no difference between the failure rates according to postoperative renographic results and reported symptoms.

Table 5.

Comparing the Retroperitoneal and Transperitoneal Approaches

	Retroperitoneal	Transperitoneal	P value
Operative duration	181 min	176 min	0.22
Conversion rate	9/85 (10.5%)	9/228 (3.9%)	0.028
Minor morbidity	6/85 (7%)	14/228 (6.1%)	0.47
Major morbidity	6/85 (7%)	8/228 (3.5%)	0.15
Obstructed postoperative renogram	4/61 (6.5%)	15/194 (7.7%)	0.51
Persistent symptoms	4/61 (6.5%)	14/194 (7.2%)	0.56

Table 6.

Complications According to Approach

Complication	Retroperitoneal	Number	Transperitoneal	Number
Minor	Simple urine leak	1	Simple urine leak	5
	Ileus	1	Deep vein thrombosis	2
	Port site infection	1	Hematuria	1
	Pyrexia	2	Paresthesia	1
	Subcutaneous emphysema	1	Renal colic	1
			Nephrostomy	1
			Repositioning of stent	1
			Shortness of breath	1
			Urinary tract infection	1
Total		6		14
Major	Complex urine leak	3	Complex urine leak	1
	Retroperitoneal bleed	1	Port site hernia	2
	Myonecrosis of gluteal	1	Duodenal perforation	1
	Bleeding necessitating exploration	1	Lost needle	1
			Perinephric abscess	1
			Liver failure, ascites	1
			Death (pseudomembranous colitis)	1
Total		6		8

Postoperative details

The median length of hospital stay was 3 days (range 1–34 d). The patient with the 34-day hospital stay had pseudomembranous colitis and unfortunately died. At a median follow-up of 4 months (range 1–24 mos), 182/203 (89.7%) were pain free, and of the 217 patients who had at least one diuresis renography result postoperation, 144 (66.3%) were reported as unobstructed, 54 (24.8%) as equivocal, and 19 (8.7%) as obstructed.

Discussion

Large series of laparoscopic pyeloplasties are available nationally as well as internationally, and all report excellent results equivalent if not superior to those of open pyeloplasties.^{1

–10} The majority of these come from single specialized centers and cannot be assumed to reflect our national practice in the UK because not all units possess the same degree of expertise, support, or caseload. The Section of Endourology of BAUS conducted this audit with the aim of providing a “snapshot” of national practices and outcomes for this procedure to ensure that acceptable standards were being achieved and maintained in everyday practice.

The results were encouraging, with only 10.3% reporting symptomatic failure and 8.7% of follow-up renograms showing evidence of persistent obstruction. There were 24.8% of the postoperative diuresis renograms that were equivocal, however. Although concerning, this is mitigated by the fact that only 5 of the 54 equivocal renograms were in patients who were still symptomatic, which was also true for 7 of the 20 obstructed renograms. This may reflect the difficulties in interpreting postoperative renograms with pooling of the tracer in a “baggy” renal pelvis possibly leading to an equivocal or false-positive result for obstruction despite operative and clinical success. Nonetheless, asymptomatic equivocal as well as obstructed renograms need close and careful follow-up to avoid missing disease progression that may manifest itself by symptom development or a deterioration of function. In these cases, postoperative glomerular filtration rate measures may also be informative.

Inagaki and colleagues⁶ found that most failures occurred within 6 months. In our audit, a subanalysis of 37 patients with more than 6 months follow-up showed only 3.7% of renograms were still obstructed while 11.1% were equivocal. Although the numbers were small, this represented an improvement on the overall renographic outcomes. Of patients in this group, however, 16.1% were still symptomatic.

The mean skin-to-skin operative time (181 min), complication rate (10.5%), and median length of hospital stay (3 d) were also as expected and comparable to those reported in the literature. Conversely, the conversion rate (6%), although acceptable, was higher than those reported from specialized centers. This is likely to reflect the lesser experience with the technique when compared to these centers and perhaps the less than ideal patient selection. The conversion rate was 2.9% for surgeons who returned 10 or more cases over the audit period, while those who returned fewer than 10 had a conversion rate of 14.7% (P<0.009). Notwithstanding the concerns about the completeness of data capture, this supports the importance of a high-volume practice in maintaining standards, and it would have been interesting to have the practice volume of each center for comparison, but unfortunately this was not provided.

It is unclear how some of the reasons given, such as portal hypertension, perineal fibrosis, and achandroplasia led to a conversion. In one patient, the reason given was that “the kidney was too medial.” It is possible that this may have been related to unsatisfactory port placement that might have been resolved by port repositioning. It was not known whether this was the causative factor in this case. Despite these conversions, median hospital stay was still acceptable at 5 days (2–28), and outcomes were no worse with 3 of the 18 conversions showing renographic obstruction and only one patient reporting persistent pain.

In virtually all of the cases, the Anderson-Hynes dismembered pyeloplasty was performed. Only four others had an alternative type of reconstruction. This is unsurprising given that the Anderson-Hynes pyeloplasty is the favored technique for the classic type of UPJO, with alternative techniques reserved for cases with atypical etiology or anatomy. Given the rarity, a relative lack of familiarity may also be deterring those more uncommon techniques from being performed laparoscopically.

There was also a relatively small number (n=10, 3%) of secondary laparoscopic pyeloplasties reported in this audit. Other series in the literature have shown this treatment to be efficacious in secondary cases, albeit less so than a primary procedure.^4,6,8,11 The reasons for this low number of secondary procedures in this audit is likely because of underreporting, but the favoring of a secondary open or endoscopic procedure because of the still relative lack of experience may have also been a factor.

The symptomatic and renographic success rates were no different between the transperitoneal and retroperitoneal approaches. The majority of the operations were performed transperitoneally (73%). This is likely to be a reflection of the more anatomically familiar orientation seen on the screen with this approach, which makes the learning process easier. There is also a bigger “work area” that may allow greater suturing and knotting space and accuracy. The retroperitoneal mode does have its proponents with the cited advantage of a more direct approach to the UPJ with less dissection needed. This should, in theory, reduce the operative time and lessen the chances of potential bowel injury. Together with the absence of peritoneal contamination from blood, urine, and carbonic acid, there should also be less postoperative ileus. In this audit, there was no difference in operative time between the two approaches (181 vs 176 min, P=0.22), but there was one duodenal injury and one postoperative ileus, both in cases performed via the transperitoneal approach. Although the overall reported complication rate for the retroperitoneal approach was higher (14% vs 9.6%), this was not statistically significant (P=0.18).

Recognized disadvantages of the retroperitoneal approach include restricted instrument movement because of the relatively smaller retroperitoneal space and also the unfamiliar orientation that requires getting used to. This may have been a factor in the higher conversion rate seen with this approach (10.5% vs 3.9%, P=0.028). On closer inspection, however, four of the nine retroperitoneal conversions were attributed to one surgeon skewing the results. When this surgeon's figures were omitted, the recalculated conversion rate was still higher but more respectable (6.2%). This highlights the importance of auditing individual surgeon's results to ensure that acceptable standards are met. Despite the possible reporting bias accounting for this higher conversion rate, the retroperitoneal approach may be less straightforward to master, and surgeons wishing to use this approach should recognize this, be appropriately trained and mentored, with their results closely audited. It is noteworthy that only 10 (3%) of the cases (one in the retroperitoneal group) were performed with a mentor. Unfortunately the surgeons' previous experiences were not determined for the purposes of this audit. Although not the primary objective of the audit, it would certainly have been interesting to correlate such information with the outcomes and therefore be able to advise on training requirements.

Whether the procedure was performed retroperitoneally or transperitoneally did not affect the reporting frequency of a crossing vessel. This was encountered in 45.6% of cases, which was at the lower limit of the range reported by others.^4,6
–8,12 Unfortunately, no information was available on whether the crossing vessel was transposed or not.

In seven patients, the procedure was performed on a kidney with a preoperative relative function of <15%, generally regarded as the cutoff value for which a pyeloplasty to preserve the function would not be considered worthwhile. The exception to this would be in cases with poor overall starting renal function, where the small contribution from the affected kidney could be significant, making preservation desirable. There was no indication that this was the reason for a pyeloplasty in these seven patients. None of these patients reported pain postsurger, and in two of the five who had follow-up renography, the function improved to above >20%. Clearly there are circumstances where postoperative recovery of function to a useful level is possible, and in equivocal cases, a trial of preoperative stent placement before another functional measurement may be worthwhile. This was only performed in one of the seven patients in this group.

Other reported technical details of the procedure were consistent with standard practice. Virtually all patients had a wound drain and urethral catheter inserted that remained for a median duration of 2 days. All patients had a stent for a median period of 6 weeks, apart from one patient who did not because of technical difficulties trying to insert it anterogradely. The reason for this technical failure was not given. In the anterograde approach, the stent is inserted percutaneously and manipulated through the dismembered ureter before the anastomosis is completed. This has the advantage of saving on the time usually needed to insert the stent retrogradely at the start of the procedure. In addition some may find it easier to dismember and prepare the UPJ without a stent in place, especially when the ureter is narrow. Although successful stentless laparoscopic pyeloplasties have been reported,¹³ the single case in which the patient did not have a stent in this audit was complicated by a significant urinary leak, necessitating nephrostomy tube placement.

A retrospective audit of this nature has its deficiencies, with the usual incompleteness and inconsistencies of data capture. Because returns were voluntary, there was potential for selection bias, and it is unknown as to the true number of procedures performed during this period. This was also recognized in the BAUS laparoscopic nephrectomy audit, which, although performed prospectively, was also voluntary.¹⁴ With the UK Departments of Health and General Medical Council's agenda of professional appraisal and revalidation, the collection of such data may become a requirement, which would improve the submission rate and accuracy of the data.

Nevertheless, this audit provides the only evidence so far as to how laparoscopic pyeloplasty is being performed at a national level and provides a yardstick by which surgeons can compare their own practice and results. Although overall outcomes were acceptable, there were areas that could be improved, such as the conversion rate, which was higher than those published from specialized centers. It could be inferred from the results that this procedure be best performed in these specialized centers with high-volume practices. The results also suggest that the transperitoneal approach should be the technique of choice in the UK and recommends that surgeons wishing to perform the procedure retroperitoneally be aware of the potential technical difficulties, receive the appropriate training, and audit their results closely.

Further national prospective audits are recommended to help raise and maintain standards. The Section of Endourology of BAUS has agreed to launch a prospective national audit of the management of UPJO using the recently introduced on-line database accessible through the BAUS Web site, which will also allow more valuable longer term data to be collected.

Footnotes

Acknowledgments

The authors wish to thank Paddy O'Reilly, consultant urological surgeon; Stephen Brown, consultant urological surgeon; Julie Morris, head medical statistician at University Hospital of South Manchester; Sarah Fowler, data and audit manager for BAUS; the BAUS Section of Endourology executive committee; Paul Downey, consultant urological surgeon; and the 30 surgeons who kindly submitted their results.

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Pouliot

, Lebel

, Audet

, Dujardin

. Determination of success by objective scintigraphic criteria after laparoscopic pyeloplasty. J Endourol, 2010; 24:299–304.

Srivastava

, Singh

, Maheshwari

et al. Laparoscopic pyeloplasty: A versatile alternative to open pyeloplasty. Urol Int, 2009; 83:420–424.

Symons

, Bhirud

, Jain

et al. Laparoscopic pyeloplasty: Our new gold standard. J Endourol, 2009; 23:463–467.

Yanke

, Lallas

, Pagnani

et al. The minimally invasive treatment of ureteropelvic junction obstruction: A review of our experience during the last decade. J Urol, 2008; 180:1397–1402.

Davenport

, Minervini

, Timoney

, Keeley

FX.

Jr . Our experience with retroperitoneal and transperitoneal laparoscopic pyeloplasty for pelvi-ureteric junction obstruction. Eur Urol, 2005; 48:973–977.

Inagaki

, Rha

, Ong

et al. Laparoscopic pyeloplasty: Current status. BJU Int, 2005; 95,suppl 2:102–105.

Rassweiler

, Subotic

, Feist-Schwenk

et al. Minimally invasive treatment of ureteropelvic junction obstruction: Long-term experience with an algorithm for laser endopyelotomy and laparoscopic retroperitoneal pyeloplasty. J Urol, 2007; 177:1000–1005.

Moon

, El-Shazly

, Chang

et al. Laparoscopic pyeloplasty: Evolution of a new gold standard. Urology, 2006; 67:932–936.

Adeyoju

, Hrouda

, Gill

. Laparoscopic pyeloplasty: The first decade. BJU Int, 2004; 94:264–267.

10.

Maynes

, Levin

, Webster

et al. Measuring the true success of laparoscopic pyeloplasty. J Endourol, 2008; 22:1193–1198.

11.

Sundaram

, Grubb RL

III

, Rehman

et al. Laparoscopic pyeloplasty for secondary ureteropelvic junction obstruction. J Urol, 2003; 169:2037–2040.

12.

Sampaio

. Vascular anatomy at the ureteropelvic junction. Urol Clin North Am, 1998; 25:251–258.

13.

Shalhav

, Mikhail

, Orvieto

et al. Adult stentless laparoscopic pyeloplasty. JSLS, 2007; 11:8–13.

14.

Davenport

, Timoney

, Keeley

et al. A 3-year review of The British Association of Urological Surgeons Section of Endourology Laparoscopic Nephrectomy Audit. BJU Int, 2006; 97:333–337.