Postoperative Complications After Robotic Partial Nephrectomy

Abstract

Objectives:

To assess the incidence of postoperative arterial malformation (AM) and urine leak/urinoma (UL) after robotic partial nephrectomy (RPN) in a contemporary series and to evaluate risk factors for these complications.

Materials and Methods:

All RPNs were queried from Institutional Review Board-approved retrospective and prospective nephrectomy databases. Demographics, perioperative variables, and postoperative complications were collected. Differences between cohorts were analyzed using univariate analysis. Postoperative complications were graded using the Clavien–Dindo system. UL was defined in the context of signs and symptoms of a collection with supporting evidence of urine collection through drainage or aspiration. AM was identified based on postoperative imaging indicative of arteriovenous fistula or pseudoaneurysm and/or requirement for selective embolization. Predictors of AM and UL were assessed by univariate analysis.

Results:

A total of 395 RPNs were performed by four urologists between January 2014 and October 2018. Tumor complexity, defined by nephrometry score, was significantly greater in the prospective cohort (p = 0.01). Overall incidence of postoperative complications was 5.6% with cohort-specific incidences of 5.3% and 5.8%. The retrospective cohort had a greater percentage of complications classified as ≥IIIa: 8/13 (61.5%) vs 2/8 (25%). Overall incidence of AM was 2.3% with cohort-specific incidence of 3.1% (7/225) vs 1.1% (2/170). Overall incidence of UL was 0.25% with cohort-specific incidence of 0.55% (1/225) and 0.0% (0/170). The difference in incidence of both complications between cohorts was significant (p < 0.05). No significant predictors for AM were identified.

Conclusions:

The incidence of postoperative complications after RPN remains low (5.3% vs 5.8%, overall: 5.6%). UL and AM are becoming rarer with experience, despite increasing surgical complexity (0.55% vs 0%, 3.1% vs 1.1%).

Introduction

Partial nephrectomy (PN) is the established procedure of choice for treating renal masses <7 cm and in recent years, PN has been used increasingly for larger, more complex renal masses.^1,2 At present, in the United States, the number of PNs performed using a minimally invasive robotic approach has surpassed those performed through an open approach.³ Robotic procedure offers the advantages of minimally invasive procedure: faster recovery time, shorter hospital stays, and decreased intraoperative blood loss.^4,5 However, there are well-established postoperative complications to both this procedure and approach. These include postoperative arterial malformation (AM) leading to hemorrhage and urine leak/urinoma (UL) formation.

Postoperative hemorrhage after PN remains a significant concern: hemorrhagic complications are reported in as many as 10% of patients and these bleeds may result in life-threatening hemorrhage.^6,7 AM, resulting from pseudoaneurysm (PA) or arteriovenous fistula (AV fistula), is the most common culprit of postoperative hemorrhage after PN. In a large series of PNs with one of the lowest reported incidences of PA, Jain and colleagues showed a significantly higher incidence (1.96% vs 1.0%) of PA after minimally invasive PN vs open PN.⁸ In light of the finding that the incidence of AM and postoperative bleeding may be greater after a minimally invasive approach coupled with the increasing utilization of this approach, further investigation of risk factors and prevention strategies for AM after robotic PN (RPN) is warranted.

UL is another postoperative complication associated with PN that often necessitates readmission and procedural intervention. UL results from the disruption of any level of the urinary collecting system.⁹ Urinoma occurs when the extravasated urine from a urine leak becomes encapsulated by a thick fibrous wall, forming a mass.^10,11 Uncomplicated small urine leaks may be managed conservatively with Foley catheterization; however, larger leaks can lead to severe complications including abscess formation and renal damage. Image-guided percutaneous intervention is often sufficient to handle these larger urinomas, although stent placement or reoperation may be necessary in severe cases.⁹ UL is also relatively rare after RPN with reported incidence of as much as 3.0%.¹² Historically, this complication has been more frequently reported among open PN and laparoscopic PN (LPN) with incidence as high as 16.5%.¹² We believe that with experience, UL after RPN is occurring less frequently, secondary to the better visualization and ability to suture afforded by the robotic technique.

Materials and Methods

Patients and data management

Data were obtained from retrospective and prospective Institutional Review Board-approved kidney cancer databases consisting of all partial and radical nephrectomies performed for renal masses at Hackensack University Medical Center (Hackensack, NJ). Inclusion criteria were as follows: all RPNs for which patient follow-up was at least 30 days in length. We used the REDCap system for data management and storage.¹³ The retrospective database encompasses cases from January 2014 to May 2017 and the prospective database includes cases completed between June 2017 and October 2018. All RPNs were queried from these databases for the purpose of this specific analysis.

Surgeons and technique

Four attending urologists (M.A., G.L., R.M., and M.D.S.) performed RPNs. All 4 surgeons had performed at least 150 RPNs before the earliest inclusion date of the retrospective study. Of note, M.S. joined the institution in spring of 2016, after the first cases included in this analysis.

All RPNs were performed using the da Vinci^® Surgical System (Intuitive, Sunnyvale, CA). The techniques used have been described previously.¹⁴ Both transperitoneal and retroperitoneal approaches were utilized. No enucleations were performed. In select cases, select clamping was utilized. All surgeons used the sliding-clip technique for renorrhaphy that has been previously described.⁵ Individual entries to the collecting system or to perforating vessels were closed using interrupted 3.0 Vicryl suture or incorporated running suture along the base of the defect. Drain placement and methods of hemostasis were at the discretion of the surgeon with the surgeons in this analysis primarily relying on barbed suture for the capsule rather than hemostatic agents in general. Ureteral stents were not utilized prophylactically. Postoperative anticoagulation was not administered.

Analysis

Postoperative complications occurring within 30 days of discharge were classified according to the Clavien–Dindo classification system.¹⁵ We distinguished minor (grade I and II) from major (grade III and IV) complications. AM was defined as any patient with postoperative imaging consistent with an AV fistula or PA and/or necessitating selective embolization. UL after RPN was defined using Russo and coworkers criteria: signs and symptoms of a perinephric collection of urine, confirmed to be urine after needle aspiration or postoperative drainage.¹⁶

Incidence of all postoperative complications necessitating readmission was calculated per cohort and incidence of AM and UL. The incidence of these complications among each individual surgeon was assessed by χ² analysis. Preoperative and intraoperative factors that may influence the occurrence of AM were assessed using univariate analysis; continuous variables were assessed by two-tailed t-test and categorical variables were assessed by χ² analysis. Mean values were utilized as point measures for normally distributed variables. Medians were used as point estimates for continuous variables with non-normal distributions.

Results

At the time of analysis, our database contained 395 PNs performed with a robot-assisted approach. A total of 225 RPNs were queried from the retrospective cohort, whereas 170 were obtained from our ongoing prospective database. Table 1 describes patient demographics in each cohort. The average age of patients in the retrospective cohort was significantly greater than the prospective cohort (64 years vs 60.1 years). There was a greater proportion of patients identified as white, 89.4% vs 69.8% (p < 0.002) and a higher tumor complexity, defined by nephrometry score (p = 0.01) in the prospective cohort. Approximately 17% of cases in the prospective cohort had an associated nephrometry score of ≥10 vs only 5% of those in the retrospective cohort. Mean operative time, warm ischemia time (WIT), and intraoperative blood loss were all significantly lower in the prospective cohort than in the retrospective cohort at p < 0.05. Mean operative time was ∼30 minutes less on average for the prospective cases vs the retrospective, with an average of 6 minutes less of WIT. In addition, a significantly greater proportion of the prospective cases involved collecting system entry (p = 0.002).

Table 1.

Characteristics of Retrospective and Prospective Cohort Robotic Partial Nephrectomy Patients

Patient characteristic	Retrospective (N = 225)	Prospective (N = 170)	p^a
Mean age (years)	64.0 (10.7)	60.1 (13.1)	0.001
Gender: male, n (%)	159 (72.6)	106 (62.4)	0.083
Race: white, n (%)	157 (69.8)	152 (89.4)	<0.002
Mean BMI (kg/m²)	29.2 (5.28)	29.2 (5.94)	1.0
Smoking status: ever, n (%)	108 (48.0)	72 (42.4)	0.26
Previous abdominal operation: yes, n (%)	98 (43.5)	89 (52.4)	0.08
Nephrometry score, %
4–6	50	31.7	0.01
7–9	45	51.2
10–12	5	17.1
Mean operative time (minutes)	141.8 (50.7)	112.8 (65.5)	0.0001
Mean warm ischemia time (minutes)	19.7 (8.61)	13.4 (8.23)	<0.0001
Mean intraoperative blood loss (mL)	182.4 (167.1)	129.4 (111.3)	0.0004
Collecting system entry: yes, n (%)	43 (19.1)	97 (57.1)	<0.002

Continuous variables: unpaired two-sided t-test; Categorical variables: χ² test.

BMI = body mass index.

Table 2 describes the complications leading to readmission within 30 days of discharge after RPN. Thirteen patients (5.8%) and nine patients (5.3%) from the retrospective and prospective studies, respectively, were readmitted within 30 days for postoperative complications related to operation. These complications ranged from Clavien I grade to Clavien IIIa. Clavien I and II complications included wound infections/disruptions, hematomas, hospital-acquired pneumonia, urinary tract infection, and fever. Clavien IIIa complications were limited to AM and UL. Procedural intervention was required for 53.3% of the retrospective postoperative readmissions and 8.3% of the prospective postoperative readmissions. Median length of stay (LOS) for readmission was 2 and 1 day for our retrospective and prospective cohorts, respectively. The nature of one of the readmission complications that occurred among the prospective RPNs was unable to be determined because of admission to an outside facility. There was no significant difference in incidence of postoperative complications among the four surgeons.

Table 2.

Postoperative Complications Necessitating Readmission Within 30 Days After Robotic Partial Nephrectomy

Grade (organ system)	Retro complications (n)	Prospective complications (n)
I
Genitourinary	Perinephric hematoma (2)	Urinary tract infection (1), bladder clot (1)
Gastrointestinal	Small bowel obstruction (1)
Dermatological		Wound disruption (1)
Systemic		Fever (1)
Cardiovascular		Chest pain (1)
II
Genitourinary
Gastrointestinal	Abdominal pain (1)
Dermatological	Wound infection (1)
Pulmonary		Hospital acquired pneumonia (1)
III
Genitourinary	Urine leak (1)
Vascular	Pseudoaneurysm (6)	Arteriovenous fistula (2)
	Arteriovenous fistula (1)

Table 3 describes the Clavien IIIa complications by cohort. We report nine AMs with an overall incidence of 2.3%. Seven of the nine AMs occurred in the retrospective cohort, whereas two occurred in the prospective cohort, with incidence of 3.1% vs 1.1%, respectively. The incidence of AM was significantly different between the retrospective and prospective time periods (p = 0.02). All observed AMs were treated with arterioembolization with an overall interventional embolization rate of 2.3% for the entire combined cohort. All cases presented with gross hematuria with a mean presentation 7.1 days after operation. Only a single UL was reported among our 395 RPN cases, resulting in an overall incidence of 0.25% and cohort-specific incidences of 0.55% and 0%, respectively. This UL was treated percutaneously but did result in a readmission LOS of 9 days, which was primarily attributed to other coexisting patient comorbidities rather than the UL itself. Because of the low incidence of UL in our cohorts, we were unable to explore potential risk factors for this using our data as originally intended. The incidence of postoperative AM does not seem to be influenced by the preoperative and perioperative variables given in Table 4.

Table 3.

Major Postoperative Complications Requiring Procedural Intervention After Robotic Partial Nephrectomy

Complication	Retrospective (N = 225)	Prospective (N = 170)	p
Urinoma, % (n)	0.55 (1)	0.0 (0)	0.004
AM, % (n)	3.1 (7)	1.1 (2)	0.02

AM = arterial malformation.

Table 4.

Potential Risk Factors for Arterial Malformation After Robotic Partial Nephrectomy

Variable	Control (N = 386)	AM (N = 9)	p (χ² or t-test)^a
Average age (years)	62.4	65.9	0.07
Sex, n (%)
Female	122 (32.1)	2 (22.2)	0.53
Male	258 (67.9)	7 (77.8)
Race, n (%)
White	305 (79.4)	6 (66.7)	0.62
Black	24 (6.3)	1 (11.1)
Asian	14 (3.6)	1 (11.1)
Other	41 (10.7)	1 (11.1)
Average BMI (kg/m²)	29.2	31.8	0.29
Hypertension
Yes	239 (62.6)	6 (66.7)	0.80
No	143 (37.4)	3 (33.3)
History of smoking, n (%)			0.44
Never	208 (54.7)	4 (44.4)
Former	136 (35.8)	3 (33.3)
Active	36 (9.5)	2 (22.2)
Average nephrometry score	7.6	7.3	0.70
Average warm ischemia time (minutes)	15.4	20.4	0.11
Entry into the collecting system, n (%)
Yes	137 (38.8)	5 (67.5)	0.17
No	216 (61.2)	3 (32.5)
Average operating time (minutes)	129.6	117.7	0.31
Average estimated operative blood loss (mL)	156.2	134.4	0.50
Previous abdominal operation, n (%)
Yes	184 (48.9)	2 (22.2)	0.11
No	192 (51.1)	7 (77.7)

Continuous variables: unpaired two-sided t-test; categorical variables: χ² test.

Discussion

In recent years, RPN has emerged as a well-accepted and increasingly utilized surgical management strategy for renal masses. Proponents of RPN suggest that a robotic approach may allow surgeons to address even relatively complex masses, offering the benefits of minimally invasive procedure while minimizing the risks associated with the steep-learning curve of traditional LPN.¹⁷ In experienced hands, overall complication rates after RPN have remained acceptably low and comparable with those of open PN and LPN with the majority of the complications classified as Clavien I and Clavien II.¹⁸ A question of particular interest is whether the benefits of a robotic approach occur at the expense of a higher rate of postoperative complications after PN than those of open PN and LPN.^8,18 Several studies have illustrated that hemorrhage and urine leak are two of the most common postoperative complications after RPN, whereas others suggest that these complications are becoming rarer in the contemporary era of RPN.^12,18,19 We sought out to both assess if and how the incidence of these complications has changed with increasing robotic experience.

Postoperative hemorrhagic complications

AM is a significant postoperative complication that can occur after PN because of the associated possibility of life-threatening hemorrhage. Minimally invasive PN has been suggested to increase the risk of AM compared with open PN.^8,20,21 Reported incidence of this complication in the literature ranges from ∼1% to 5%.^20,21 More recently, a higher incidence of asymptomatic PAs was reported among minimally invasive PN vs open PN.²² We report an overall incidence of 2.3%, which is consistent with the lower end of what has recently been reported in modern series. Despite the incidence of postoperative complications necessitating readmission remaining relatively stable, we observed an appreciable difference in the incidence of AM between the retrospective and prospective time periods of our study with an incidence of 3.1% vs 1.1%, respectively. The number of malformations per attending surgeon was not significantly different. This result should be interpreted in the context of the increasing surgical complexity of the prospective cases as more cases required collecting system entry and overall nephrometry scores were significantly higher than those among the retrospective cases. Despite the increased surgical complexity of the prospective RPNs, operative time, WIT, and intraoperative blood loss were all significantly lower than in the retrospective cohort. These results suggest what has been suspected by other groups: AM is becoming rarer after RPN as experience increases, even in the context of increasingly complex cases.

There are various theories as to why rates of vascular injury are higher in minimally invasive PN than in open. Larger needles utilized in minimally invasive PN may increase trauma to renal arterioles during renorrhaphy.²³ Inadequate hemostasis after renorrhaphy may also be a factor at play.¹⁹ Fardoun and colleagues assessed predictive factors for hemorrhagic complication (defined as AV fistula, PA, or hematoma necessitating transfusion) and found that intraoperative blood loss was a significant predictor; however, this may be partially explained by surgical complexity.²⁴ We did not find intraoperative blood loss, operative time, or nephrometry score to be significantly predictive of postoperative AM on univariate analysis.

Postoperative UL

Historically, UL has been considered one of the most common postoperative complications after PN; however, this complication has become particularly rare in the modern era of RPN.¹² In a multicenter analysis of 1791 PNs, Potretzke and coworkers reported an 0.8% incidence of urine leak.¹² They also reviewed the available literature to analyze trends in the incidence of urine leak after RPN and reported a range from 0.6% to 3.0% based on results from 13 studies published between 2008 and 2015.¹² Several authors attribute the decreasing incidence of postoperative urine leak to both advances in preoperative imaging and surgical technique.^12,25 Our finding of a single postoperative urine leak among 395 RPNs performed collectively by four experienced surgeons at a high-volume center supports this theory. The overall 0.25% incidence is one of the lowest reported for urine leak in large series of RPNs. Previous series have shown that risk factors for urine leak include WIT, blood loss, collecting system repair, operative time, and nephrometry score.^12,26,27 We were unable to assess these risk factors because of the extremely low incidence of urine leak in our cohort. The case however, did necessitate collecting system repair and had an associated nephrometry score of 11 × .

Study limitations

Our data suggest that over time, the incidence of postoperative complications after RPN that warrant readmission has remained both stable and low. The nature of these complications, however, has shifted over time. A significantly higher percentage of the readmission complications from our retrospective cohort were classified as Clavien III or higher and necessitated procedural intervention compared with those occurring during the prospective study period. As given in Table 1, mean operative time, blood loss, and WIT have decreased significantly between time periods of our study. These changes also underscore the increase in RPN experience that occurred throughout the time period of interest in this study.

In interpreting our data, there are several limitations to consider. Our study design is monocentric. It should be noted that all surgeons in this study have substantial experience in RPN and thus findings may not be generalizable to all institutions. About half of our cases were obtained from our retrospective kidney cancer database, thus limitations that result from retrospective design apply in these instances. In addition, data regarding preoperative, intraoperative, and postoperative variables including details regarding complications are limited to available records. The cases derived from our prospective cohort benefit from our ability to document more precise surgical and postoperative outcomes. Moreover, at the time of analysis, our retrospective data spanned a greater range of time, which theoretically could explain the lower incidence of both complications of interest in the prospective study compared with the retrospective. However, both groups contain a similar breakdown of RPNs and extremely comparable overall postoperative complication rates (5.8% vs 5.3%), rendering this possibility less likely.

Conclusions

Global readmission rates for postoperative complications after RPN remain acceptably low. UL and AM are two of the most concerning postoperative complications as they often necessitate readmission, procedural intervention, and can portend significant morbidity. Incidence of both these complications has decreased significantly with experience, and is now at an all-time reported low with AM occurring more frequently than urine leak. As a secondary end point we noted that despite an increase in tumor complexity and need for collecting system entry, operative times, WIT, blood loss, LOS, and readmission rates were lower in the more recent, more experienced cohort. Risk factors for postoperative AM warrant further investigation.

Footnotes

Author Disclosure Statement

M.D.S. is a consultant for VTI and a lecturer for Intuitive, Ethicon, and Conmed. R.M. is a speaker for Boston Scientific, Inc., and Amniox Medical, Inc.

Funding Information

No funding was received for this article.

Abbreviations Used

References

Ljungberg

, Albiges

, Abu-Ghanem

, et al. European Association of Urology Guidelines on renal cell carcinoma: The 2019 update. Eur Urol, 2019; 75:799–810.

White

, Marco

DJT

, Bolton

, et al. Trends in the surgical management of stage 1 renal cell carcinoma: Findings from a population-based study. BJU Int, 2017; 120 Suppl 3:6–14.

Cheung

, Wang

, Chang

, Khandwala

, Del Giudice

, Chung

. Adoption of robot-assisted partial nephrectomies: A population-based analysis of U.S. Surgeons from 2004 to 2013. J Endourol, 2017; 31:886–892.

Ghani

, Sukumar

, Sammon

, Rogers

, Trinh

, Menon

. Practice patterns and outcomes of open and minimally invasive partial nephrectomy since the introduction of robotic partial nephrectomy: Results from the nationwide inpatient sample. J Urol, 2014; 191:907–912.

Benway

, Bhayani

, Rogers

, Dulabon

, Patel

, Lipkin

, Wang

, Stifelman

. Robot assisted partial nephrectomy versus laparoscopic partial nephrectomy for renal tumors: A multi-institutional analysis of perioperative outcomes. J Urol, 2009; 182:866–872.

Hennessey

, Wei

, Moon

, Kinnear

, Bolton

, Lawrentschuk

, Chan

. Strategies for success: A multi-institutional study on robot-assisted partial nephrectomy for complex renal lesions. BJU Int, 2018; 121 Suppl 3:40–47.

Gill

, Kavoussi

, Lane

, et al. Comparison of 1,800 laparoscopic and open partial nephrectomies for single renal tumors. J Urol, 2007; 178:41–46.

Jain

, Nyirenda

, Yates

, Munver

. Incidence of renal artery pseudoaneurysm following open and minimally invasive partial nephrectomy: A systematic review and comparative analysis. J Urol, 2013; 189:1643–1648.

Titton

, Gervais

, Hahn

, Harisinghani

, Arellano

, Mueller

. Urine leaks and urinomas: Diagnosis and imaging-guided intervention. Radiographics, 2003; 23:1133–1147.

10.

McInerney

, Jones

, Roylance

. Urinoma. Clin Radiol, 1977; 28:345–351.

11.

Pyrah

, Smiddy

. Pararenal pseudo-hydronephrosis; a report of two cases. Br J Urol, 1953; 25:239–246.

12.

Potretzke

, Knight

, Zargar

, et al. Urinary fistula after robot-assisted partial nephrectomy: A multicentre analysis of 1791 patients. BJU Int, 2016; 117:131–137.

13.

Harris

, Taylor

, Thielke

, Payne

, Gonzalez

, Conde

. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform, 2009; 42:377–381.

14.

Rogers

, Singh

, Blatt

, Linehan

, Pinto

. Robotic partial nephrectomy for complex renal tumors: Surgical technique. Eur Urol, 2008; 53:514–521.

15.

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.

16.

Godoy

, Katz

, Adamy

, Jamal

, Bernstein

, Russo

. Routine drain placement after partial nephrectomy is not always necessary. J Urol, 2011; 186:411–415.

17.

Patton

, Salevitz

, Tyson MD

, Andrews

, Ferrigni

, Nateras

, Castle

. Robot-assisted partial nephrectomy for complex renal masses. J Robot Surg, 2016; 10:27–31.

18.

Spana

, Haber

, Dulabon

, Petros

, Rogers

, Bhayani

, Stifelman

, Kaouk

. Complications after robotic partial nephrectomy at centers of excellence: Multi-institutional analysis of 450 cases. J Urol, 2011; 186:417–421.

19.

Kondo

, Takagi

, Morita

, et al. Early unclamping might reduce the risk of renal artery pseudoaneurysm after robot-assisted laparoscopic partial nephrectomy. Int J Urol, 2015; 22:1096–1102.

20.

Ghoneim

, Thornton

, Solomon

, Adamy

, Favaretto

, Russo

. Selective arterial embolization for pseudoaneurysms and arteriovenous fistula of renal artery branches following partial nephrectomy. J Urol, 2011; 185:2061–2065.

21.

Netsch

, Brüning

, Bach

, Gross

. Management of renal artery pseudoaneurysm after partial nephrectomy. World J Urol, 2010; 28:519–524.

22.

Takagi

, Kondo

, Tajima

, Campbell

, Tanabe

. Enhanced computed tomography after partial nephrectomy in early postoperative period to detect asymptomatic renal artery pseudoaneurysm. Int J Urol, 2014; 21:880–885.

23.

Singh

, Gill

. Renal artery pseudoaneurysm following laparoscopic partial nephrectomy. J Urol, 2005; 174:2256–2259.

24.

Fardoun

, Chaste

, Oger

, et al. Predictive factors of hemorrhagic complications after partial nephrectomy. Eur J Surg Oncol, 2014; 40:85–89.

25.

Thompson

, Leibovich

, Lohse

, Zincke

, Blute

. Complications of contemporary open nephron sparing surgery: A single institution experience. J Urol, 2005; 174:855–858.

26.

Stroup

, Palazzi

, Kopp

, et al. RENAL nephrometry score is associated with operative approach for partial nephrectomy and urine leak. Urology, 2012; 80:151–156.

27.

Bruner

, Breau

, Lohse

, Leibovich

, Blute

. Renal nephrometry score is associated with urine leak after partial nephrectomy. BJU Int, 2011; 108:67–72.