Open,Laparoscopic,and Robotic Ureteroneocystotomy for Benign and Malignant Ureteral Lesions: A Comparison of Over 100 Minimally Invasive Cases

Abstract

Introduction:

Laparoscopic (LAP) and robot-assisted laparoscopic (RAL) approaches have been applied to ureteroneocystostomies (UNC) although such experience has been limited to a small number of patients and limited follow-up. Herein, we detail our experience with over 100 minimally invasive UNC, the largest such series to date.

Methods:

All minimally invasive UNC performed at our institution between 1997 and 2013 and all open UNC performed between 2008 and 2013 were identified. Perioperative parameters of relevance were identified and recorded. Chi-squared and ANOVA with post hoc Tukey analysis were performed for all categorical and continuous variables, respectively.

Results:

A total of 130 patients met our study criteria. One hundred five underwent the minimally invasive approach (20 RAL and 85 LAP). Mean follow-up duration was 504 days. Patients in the RAL, LAP, and open cohorts were of similar age, gender and laterality distribution, American Society of Anesthesiologists (ASA) score, body–mass index, history of previous abdominal surgery, history of prior treatment for the ureteral lesion, and surgical indication (Table 1). Operative time was similar across all cohorts (235–257 minutes, p=0.123). Estimated blood loss (EBL) was significantly lower in the RAL and LAP cohorts (100 and 150 mL) compared to their open counterparts (300 mL, p=0.001) although a decrease in hematocrit was similar across all groups. Only four intraoperative complications (4.7%) and two (2.4%) conversions to open were identified in the LAP group, without statistical significance. No intraoperative complications or conversions were identified in the RAL or open cohorts. Median length of stay (LOS) was significantly shorter in the minimally invasive cohorts compared to open (p<0.002). Ninety-day readmission rates (18.8–20%), major complications (10–20%), and failure rates (5.9–16%) were highest in the open cohort although without statistical significance.

Conclusion:

RAL or LAP UNC is feasible, safe, and comparable to the open technique with some perioperative benefit in EBL, LOS, and stent duration.

Introduction

Ureteroneocystostomy (UNC) is the definitive surgical treatment of distal ureteral stricture disease or injury not amenable to primary repair and often preferred over primary repair. It is also indicated as the reconstruction technique following a distal ureterectomy. Classically, a UNC has been performed utilizing an open approach either through a lower midline incision or a Gibson incision to allow for the exposure necessary for this reconstruction. Adjuncts to the technique include the psoas hitch and the Boari flap; techniques that can be utilized separately or in conjunction to facilitate the creation of a more proximal tension-free anastamosis.¹

Recently, the laparoscopic (LAP) approach to the UNC has been described in several reports.^2
–4 Such series hail excellent success rates proving feasibility without comparison. One report, a matched cohort of 10 patients, revealed demonstrable improvements in estimated blood loss (EBL), morphine requirements, length of stay (LOS), and convalescence time⁵ when compared to open UNC. Similarly, the robot-assisted laparoscopic (RAL) approach has been applied to this procedure with similar enthusiasm,^6

–10 with more recent series suggesting similar improvements in EBL, morphine requirements, and LOS over the open approach.^11,12 However, such reports are frequently limited by a small number of cases with short-term follow-up. Furthermore, no report has compared the LAP approach to the RAL approach. Herein, we detail to our knowledge the largest cohort comparison of UNC performed through the open, LAP, and RAL approach.

Methods

With institutional review board approval, the records of all patients billed for UNC between January 2008 and December 2013 at our institution were identified and added to a previously created database dedicated to LAP UNC (1997–2007). Retrospective chart review was then conducted for the identification of age, gender, laterality, American Society of Anesthesiologists (ASA) score, body–mass index (BMI), a history of prior abdominal surgery, stricture etiology, history of prior ureteral stricture treatments, EBL, operative time, technique utilized (including psoas hitch, advancement flap, boari flap, distal ureterectomy), intraoperative complications, and LOS. UNC performed in setting of intraoperative consult and for ureteral stricture associated with transplant kidney were excluded from analysis. Preoperative and postoperative serum creatinine (Cr) and hematocrit (Hct) were also recorded to assess for differences. Any complications, as graded by the Clavien–Dindo classification system,¹³ experienced during the 90-day postoperative period were also noted, with major complications noted as any complication grade Clavien 3a or greater. Failures were defined as those cases with evidence of subsequent obstruction, as defined by requiring subsequent drainage, reoperation (redo UNC or nephrectomy), or with evidence of obstruction on follow-up imaging. SPSS v 16 was utilized to calculate chi-squared and ANOVA with post hoc Tukey analysis for all categorical and continuous variables, respectively, with significance deemed by a two-tailed p-value less than 0.05.

Preoperative evaluation

All patients who underwent scheduled UNC were evaluated preoperatively with appropriate blood chemistry and complete blood count, urinalysis, urine culture, and urine cytology. All scheduled patients were also evaluated with cross-sectional body imaging (CT or MRI). A diuretic mercaptoacetyltriglycine (MAG-3) renal scan was performed in cases of questionable obstruction or renal function. Often patients also underwent retrograde pyelogram. Unscheduled patients or those with intraoperative iatrogenic ureteral injury were staged intraoperatively as deemed necessary by the consultant urologist. All anastomoses were stented with double-pigtail ureteral stents and a perianastamotic drain was placed in all cases regardless of etiology or approach.

Postoperative care and evaluation

All patients who underwent UNC were evaluated postoperatively with serum chemistry and complete blood count. The perianastomotic drain was removed after the drain fluid revealed no elevation in Cr. Patients were maintained on anticholinergic medications to minimize bladder spasm for duration of Foley catheter. The Foley catheter remained for 1 week; often it was removed after a cystogram revealed no extravasation. Ureteral stents were removed 4–6 weeks postoperatively. Subsequent imaging was performed regularly, most commonly with the MAG-3 diuretic renal scan although occasionally with renal ultrasound, 1 month after ureteral stent removal. The success of procedure was defined as any patient who underwent an intended procedure without conversion and without need for subsequent decompression. Hence, patients requiring conversion to open (from LAP or RAL) or subsequent ureteral stent placement or percutaneous nephrostomy placement for obstruction were considered failures.

Surgical technique

Open UNC was performed as previously described¹⁴ by 16 different surgeons. LAP UNC was performed as previously described by Reddy and Evans¹⁵ with modifications as per Lima et al.¹⁶ utilizing a nontunneled spatulated anastomosis with interrupted 4-0 vicryl suture by three surgeons. Robotic UNC was performed by five different surgeons in an analogous manner to the LAP approach with small modifications similar to prior reports.^11,12 The surgical approach was based on individual surgeon preference.

Results

A total of 130 patients met our study criteria. One hundred five underwent minimally invasive approach (20 RAL and 85 LAP). Mean follow-up duration was 504 days. Perioperative variables are summarized in Table 1. Patients in the RAL, LAP, and open cohorts were of similar age, gender and laterality distribution, ASA score, BMI, history of abdominal surgery, history of prior treatment for the ureteral lesion, and surgical indication (Table 1).

Table 1.

Patient Demographics, Medical History

	Surgical modality
	RAL	LAP	Open
Number of cases	20	85	25	p-Value
Age, mean±SD	60±14	55±15	56±18	0.263
Body–mass index, mean±SD	27±5	28±6	28±7	0.589
Prior abdominal surgery, n (%)	14 (70.0)	46 (63.0)	19 (79.2)	0.330
ASA score, n (%)				0.140
1	0 (0.0)	14 (16.5)	1 (4.0)
2	13 (65.0)	41 (48.2)	12 (48.0)
3	7 (35.0)	30 (35.3)	11 (44.0)
4	0 (0.0)	0 (0.0)	1 (4.0)
Gender, n (%)				0.108
Men	9 (45.0)	24 (28.2)	12 (48.0)
Women	11 (55.0)	61 (71.8)	13 (52.0)
Side, n (%)				0.707
Left	11 (55.0)	39 (45.9)	11 (44.0)
Right	8 (40.0)	44 (51.8)	14 (56.0)
Bilateral	1 (5.0)	2 (2.4)	0 (0.0)
Indication, n (%)				0.106
Stricture	2 (10.0)	20 (23.8)	2 (8.0)
Malignancy	6 (30.0)	14 (16.7)	9 (36.0)
Iatrogenic	7 (35.0)	20 (23.8)	10 (40.0)
Stone	0 (0.0)	6 (7.1)	1 (4.0)
Other	5 (25.0)	24 (28.6)	3 (12.0)
Prior treatment, n (%)				0.167
None	4 (20.0)	36 (42.4)	11 (44.0)
Stent	8 (40.0)	12 (14.1)	8 (32.0)
NT	4 (20.0)	18 (21.2)	4 (16.0)
Dilation	2 (10.0)	6 (7.1)	0 (0.0)
Endo incision	1 (5.0)	4 (4.7)	2 (8.0)
Other definitive	0 (0.0)	7 (8.2)	0 (0.0)
Other nondefinitive	1 (5.0)	2 (2.4)	0 (0.0)

ASA=American Society of Anesthesiologists; RAL=robot-assisted laparoscopic; LAP=laparoscopic; SD=standard deviation.

Operative data are summarized in Table 2 and demonstrated a propensity for the LAP cohort to utilize psoas hitch less and boari flap more than the other two cohorts. Furthermore, distal ureterectomy was performed less in the LAP cohort. The operative time was similar across all cohorts (235–257 minutes, p=0.123). EBL was significantly lower in the RAL and LAP cohorts (100 and 150 mL) compared to their open counterparts (300 mL, p=0.001). A decrease in Hct (%) was by 3, 3, and 7 for the open, robotic, and LAP cohorts without any statistically significant difference (p=0.66). Only four intraoperative complications (4.7%) were identified in the LAP group, without statistical significance. Two inadvertent ureterotomies were repaired primarily, an inadvertent enterotomy required intraoperative general surgery consult and was repaired with a small bowel resection, and an inadvertent injury to the appendix required an unplanned appendectomy. No intraoperative complications were identified in the RAL or open cohorts. Two conversions (2.4%) to open were required in the LAP cohort, one for small bowel resection and the other for failure to progress; no conversions were noted in the RAL cohort.

Table 2.

Operative Data

	Surgical Modality
	RAL	LAP	Open
Number of cases	20	85	25	p-Value
Technique, n (%)
UNC only	9 (45.0)	41 (48.2)	9 (36.0)	0.558
UNC+psoas hitch	6 (30.0)	1 (1.2)	12 (48.0)	<0.001
UNC+boari flap (including advancement flap)	8 (35.0)	44 (51.8)	8 (32.0)	0.183
UNC+distal ureterectomy	11 (55.0)	12 (14.1)	10 (40.0)	<0.001
OR time, median (IQR), minutes	236 (219–305)	235 (157–297)	257 (199–425)	0.123
EBL, median (IQR), mL	100 (63–200)	150 (100–300)	300 (100–600)	0.001
Δ hematocrit, median (IQR)	−3 (−5 to −2)	−7 (−10 to −2)	−3 (−8 to −1)	0.660
Conversion to open	0 (0%)	2 (2.4%)	n/a	1.000

Bold type indicates statistically significant p-values.

EBL=estimated blood loss; OR=operative; UNC=ureteroneocystostomy.

Postoperative data are summarized in Table 3 and detail the follow-up duration, LOS, major complication rate, 90-day readmission rate, and change in Cr at latest follow-up. Median LOS was significantly shorter in the minimally invasive cohorts compared to open (p<0.002). Overall major complications (Clavien 3a or greater) were limited to 2 (9.5%), 10 (11.8%), and 5 (20%) patients in the RAL, LAP, and open cohorts. Ninety-day readmission rates ranged from 18.8% to 24% with no statistically significant difference across all three cohorts. Such complications were detected within 90 days and were often related to procedure failure, which occurred in two (10%), five (5.9%), and four (16%) patients in the RAL, LAP, and open cohorts without any statistically significant difference (p=0.269). There appeared to be no change in Cr in any of the groups with a mean follow-up of 17, 79, and 42 weeks for RAL, LAP, and open cohorts.

Table 3.

Postoperative Data

	Surgical modality
	RAL	LAP	Open
Number of cases	20	85	25	p-Value
90-day readmission, n (%)	4 (20.0%)	16 (18.8%)	6 (24.0%)	0.850
Major complication, n (%)	2 (10.0%)	10 (11.8%)	5 (20.0%)	0.509
Failure, n (%)	2 (10.0%)	5 (5.9%)	4 (16.0%)	0.269
LOS (days), median (IQR)	2 (2–4)	3 (2–4)	5 (3–7)	0.010
Δ Creatinine, median (IQR)	0.0 (−0.1 to 0.2)	0.0 (−0.1 to 0.2)	−0.2 (−0.3 to 0.0)	0.001
Follow-up (weeks), median (IQR)	16 (9–34)	79 (15–131)	58 (13–122)	0.002
Foley duration (days), median (IQR)	8 (8–10)	8 (6–13)	13 (9–15)	0.084
Stent duration (days), median (IQR)	38 (29–45)	44 (34–47)	49 (44–62)	0.080

Bold type indicates statistically significant p-values.

LOS=length of stay.

Discussion

Minimally invasive surgical approaches provide clear benefit. Associated reduction in postoperative pain, LOS, and convalescence are motivating factors for the expanding application of the LAP approach to nearly any surgical procedure. While the feasibility of LAP UNC has been demonstrated in several reports with limited number of cases and follow-up duration, only one prior study compared outcomes to open UNC (Table 4). Rassweiler et al.⁵ compared 10 cases of LAP UNC to a matched cohort of 10 open UNC demonstrating significant superiority of the LAP group in many parameters (lower EBL, shorter time to PO intake, shorter LOS, decreased narcotic requirement, and shorter convalescence time). They also demonstrated nonsignificant reduction in complications, both minor and major, despite a nonsignificantly greater mean BMI and stricture length in the LAP cohort. The limitation of the LAP approach, however, was a statistically significant increase in mean operative time (43 minutes). Hence, the main limitation of the LAP approach, it is postulated, is the technical difficulty associated with reconstruction and intracorporeal suturing.

Table 4.

Summary of Published Series

Author	Year	LAP/RAL	Compared with	No. of cases	OR time (minutes)	EBL (mL)	LOS (days)	No. of boari	No. of psoas hitch	No. of distal ureterectomy	Follow-up duration (months)	% success	Complication
Seideman²	2009	LAP	None	45	n/a	150 (median)	3	21	n/a	4	24 mo	100	15.5% (7)
Pompeo⁴	2013	LAP	None	9	207	122 (mean)	3.3	0	2	0	6 mo	100	22% (2)
Soares³	2010	LAP	None	11	166	162 (mean)	2.9	2	1	0	18 mo	100	18% (2)
Rassweiler⁵	2007	LAP	c/w open	10	228	370 (mean)	9.2	4	10	0	17 mo	100	20% (2)
Current	2014	LAP	c/w open and LAP	85	235	150 (median)	3	44	1	12	20 mo	94.10	11.8 (major)
Hemal⁹	2010	RAL	None	18	130–210 (range)	100 (mean)	3.1	0	1	1	13.3 mo	100	5% (1)
Schimpf⁷	2009	RAL	None	11	189 (mean)	82 (mean)	2.5	2	3	6	20.5 mo	100	27% (3)
Williams⁸	2009	RAL	None	8	247 (mean)	109 (mean)	2	0	0	0	18 mo	87.50	0%
Patil⁶	2008	RAL	None	12	208 (mean)	48 (mean)	4.3	0	12	0	15.5 mo	100	0%
Musch¹⁰	2013	RAL	None	16	260 (median)	n/a	7.5 (median)	5	9	8	10.2 mo (median)	94	12% (major)
Kozin¹¹	2012	RAL	c/w open	10	306 (mean)	31 (mean)	2.4	2	4	0	30 mo	100	0%
Isac¹²	2013	RAL	c/w open	25	279 (mean)	100 (mean)	3	10	4	0	11.6 mo	92	2 major
Current	2014	RAL	c/w open and RAL	20	236	100 (median)	2	8	6	11	4 mo	90.00	10% (major)

c/w=compared with; n/a=not available.

The da Vinci Surgical robot (Intuitive Surgical, Sunnyvale, CA) has similarly been applied to most urologic procedures, in the hopes of providing the benefits of decreased postoperative pain and convalescence. Moreover, it is clear that the robotic endowrist, with its seven degrees of freedom, allows for improved dexterity and facilitates intracorporeal tailoring and suturing. Such benefits may allow more urologists, without the requisite expertise in laparoscopy, to capably perform UNC in a minimally invasive manner.

Several authors have reported their experience with RAL UNC with favorable results. Such reports suggest success rates ranging from 87.5% to 100% with low complication rates, EBL, and acceptable operative times (ranging from 189 to 306 minutes). Table 4 summarizes their findings. Only two reports compare their RAL UNC experience to an open UNC cohort. Kozin et al. evaluated outcomes associated with RAL UNC performed for benign distal and midureteral strictures in 10 cases compared to 10 open UNC cases.¹¹ Both cohorts were similar in age, gender distribution, ASA, BMI, length and laterality of ureteral stricture. They reported 100% success rates utilizing four psoas hitch and two boari flap in their RAL UNC cohort with no complications reported. As expected, EBL and LOS were lower in the RAL UNC cohort. RAL UNC also demonstrated a slightly greater operative time (difference 36 minutes) and less mean morphine equivalents used, although neither of these differences was of significant difference, statistically. A similar study by Isac et al. reviewed a 10-year experience of 25 and 41 patients who underwent RAL UNC and open UNC, respectively.¹² Their analysis revealed no difference in baseline demographics (age, gender, BMI, ASA, stone disease etiology, laterality, indication) or in surgical technique distribution (primary UNC vs tapered UNC, psoas hitch, and boari flap). Predictably, the RAL UNC cohort did require a longer operative time (79 minutes) and lower EBL (and decrease in Hct) compared to the open UNC cohort. Furthermore, RAL UNC patients experienced shorter LOS and lower narcotic requirement.

We performed a similar retrospective cohort analysis evaluating RAL and LAP outcomes compared to an open cohort. Our cohorts demonstrated similar patient demographics, although benefits of minimally invasive approach were evident in decreased EBL and decreased LOS with similar operative times, complication rates, and success rates compared to open UNC. In fact, both RAL and LAP demonstrated lower 90-day readmission rates, major complication rates, and failure rates, although without statistical significance, when compared to open UNC, although some of these differences may be due to selection bias. Our analysis did reveal an unexpectedly high 90-day readmission rate across all three cohorts. We feel that these admissions were often secondary to issues with the Foley catheter and concern for urinary tract infection with low threshold for admission for IV antibiotics.

Our report has several limitations. This is a retrospective review of two chronologically disparate groups subject to selection bias. Clearly, undetected or poorly recorded variables may introduce unexpected differences and alter our results. For instance, severity of previous surgery (not accounted for in this study) may have increased the likelihood of open surgical approach. Furthermore, several different surgeons performed all of the procedures noted—which can further introduce variation based on individual surgeon skill or technique. The heterogeneous nature of our cohort can also limit the precision of findings. Nonetheless, this is the largest cohort evaluating both minimally invasive approaches (LAP and RAL) to open UNC. This allows for comparison with each other and an open cohort—a feature unique to our study. Such an approach also conceivably allows for the detection of conversions to open no matter how the case is coded. We also demonstrate the feasibility of distal ureterectomy using the LAP and RAL approaches with 23 such cases performed—the largest reported in the literature.

Conclusion

LAP and RAL UNC are feasible and comparable to open UNC. Associated procedures—boari flap, psoas hitch, and distal ureterectomy are also feasible by both LAP and RAL means, despite their increased complexity. LAP and RAL demonstrated superiority in terms of major complication rates and failure rates with p-values trending toward significance. RAL UNC was associated with the lowest EBL. RAL or LAP UNC is feasible, safe, and comparable to the open technique with some perioperative benefit.

Footnotes

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Wenske

, Olsson

, Benson

. Outcomes of distal ureteral reconstruction through reimplantation with psoas hitch, boari flap, or ureteroneocystostomy for benign or malignant ureteral obstruction or injury. Urology, 2013; 82:231–236.

Seideman

, Huckabay

, Smith

, et al. Laparoscopic ureteral reimplantation: Technique and outcomes. J Urol, 2009; 181:1742–1746.

Soares

RSQ

, de Abreu

Jr. , Tavora

JEF

. Laparoscopic ureteral reimplant for ureteral stricture. International Braz J Urol, 2010; 36:38–43.

Pompeo

, Molina

, Sehrt

, et al. Laparoscopic ureteroneocystostomy for ureteral injuries after hysterectomy. J Soc Lap Surg, 2013; 17:121–125.

Rassweiler

, Gozen

, Erdogru

, et al. Ureteral reimplantation for management of ureteral strictures: A retrospective comparison of laparoscopic and open techniques. Eur Urol, 2007; 51:512–523.

Patil

, Mottrie

, Sundaram

, Patel

. Robotic-assisted laparoscopic ureteral reimplantation with psoas hitch: A multi-institutional, multinational evaluation. Urology, 2008; 72:47–50.

Schimpf

, Wagner

. Robot-assisted laparoscopic distal ureteral surgery. JSLS, 2009; 13:44–49.

Williams

, Leveillee

. Expanding the horizons: Robot-assisted reconstructive surgery of the distal ureter. J Endourol, 2009; 23:457–461.

Hemal

, Nayyar

, Gupta

, et al. Experience with robot assisted laparoscopic surgery for upper and lower benign and malignant ureteral pathologies. Urology, 2010; 76:1387–1394.

10.

Musch

, Hohenhorst

, Pailliart

, et al. Robot-assisted reconstructive surgery of the distal ureter: Single institution experience in 16 patients. BJU Int, 2013; 111:773–783.

11.

Kozin

, Canes

, Sorcini

, et al. Robotic vs open distal ureteral reconstruction and reimplantation for benign stricture disease. J Endourol, 2012; 26:147–151.

12.

Isac

, Kaouk

, Altunrende

, et al. Robot-assisted ureteroneocystostomy: Technique and comparative outcomes. J Endourol, 2013; 27:318–323.

13.

Dindo

, Demartines

, Clavien

. Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg, 2004; 240:205–213.

14.

Nakada

, Hsu

THS

. Ureteral stricture disease, chapter 41. In: Wein

, Kavoussi

, et al., eds. Campbell-Walsh Urology, 10th ed. Philadelphia: Elsevier, pp. 1122–1168, 2012.

15.

Reddy

, Evans

. Laparoscopic ureteroneocystostomy. J Urol, 1994; 152:2057–2059.

16.

Lima

, Rais-Bahrami

, Link

, et al. Laparoscopic ureteral reimplantation: A simplified dome advancement technique. Urology, 2005; 66:1307–1309.