Identifying Preoperative Predictors of Operative Time and Their Impact on Outcomes in Robot-Assisted Partial Nephrectomy

Abstract

Objective:

To identify preoperative characteristics in patients with renal masses that influence operative time during robot-assisted partial nephrectomy (RAPN) and evaluate the relationship between operative time and length of stay (LOS), complication rates, and overall outcome.

Materials and Methods:

We queried our institutional database to identify a cohort of patients who underwent RAPN by two experienced robotic surgeons between 2012 and 2019. A multivariable regression model was developed to analyze operative time, LOS, and any grade complication within 30 days postoperatively using the bootstrap resampling technique.

Results:

A total of 392 patients were included. On multivariable analyses, prior abdominal surgery (p = 0.001) was associated with 22 minutes of increase in operating room time, as well as adhesive perirenal fat (22 minutes, p = 0.001). For each one unit increase in nephrometry score, there was a 4-minute increase in operating room time (p = 0.028), and for each one-cm increase in tumor size, there was an associated 12-minute increase in operating room time (p < 0.001). For each 1 year increase in age, there was an associated 0.024-day increase in LOS [odds ratio (OR) (0.013–0.035)]; in addition, for every one-cm increase in tumor size there was a 0.18-day associated increase in LOS [OR (0.070–0.28)]. Each 1-hour increase in operating room time was associated with a 0.25-day increased LOS [OR (0.092–0.41)]. Only tumor size was found to be associated with any grade complication.

Conclusions:

Patients with a history of abdominal surgery, larger complex tumors, and significant Gerota's fat undergoing robotic partial nephrectomy should anticipate longer operative times. Older patients with larger tumors and longer operative times can anticipate a longer LOS. Tumor size appears to be the common determinant of all three outcomes: operative time, LOS, and any grade Clavien complication.

Introduction

Over the last decade, robot-assisted laparoscopic partial nephrectomy has become the primary treatment modality for clinical T1 renal tumors.^1
–3 The use of nephron-sparing surgery leads to improved preservation of renal function, reduction of metabolic and cardiovascular sequelae, and acceptable oncologic outcome.^2,4

–9 Although partial nephrectomy is the most commonly performed procedure for small renal tumors, the impact of predictable preoperative variables has not been well evaluated.^4,10

–13

A number of surrogates have been postulated as important for perioperative outcomes following minimally invasive nephron-sparing surgery. These include patient comorbidities, tumor size and complexity (nephrometry score), surgeon experience, operative time, estimated blood loss (EBL), perinephric to subcutaneous fat, hospital length of stay (LOS), as well as a host of other prognostic indicators.^12

–18 Although numerous studies have evaluated these characteristics using a multitude of methodologies, to date, there is a dearth of evidence assessing how these factors may affect operative time and subsequently complication rates. We sought to examine this relationship and the relationship between operative time and LOS for patients undergoing robotic partial nephrectomy at our institution.

Materials and Methods

Data source

After obtaining approval from the Institutional Review Board, we queried our prospectively maintained partial nephrectomy database to examine the records of patients who underwent robotic partial nephrectomy for clinically localized renal tumors by two experienced fellowship-trained surgeons using the da Vinci Si or Xi system (Intuitive Surgical, Inc., Sunnyvale, CA) between 2012 and 2019. These two surgeons had performed over 500 partial nephrectomies by 2012 and were considered to be beyond their learning curve for the analyzed patients.

We excluded any patients who underwent conversion to radical nephrectomy, had aberrant anatomy such as a horseshoe kidney, and those whom preoperative imaging (computed tomography or magnetic resonance imaging) could not be located and reviewed. All data and imaging were reviewed independently by three of the study's authors (N.K.K., J.Z., and R.T.N.).

Operative technique

All patients underwent robotic partial nephrectomy via the transperitoneal approach using the da Vinci robotic system (Intuitive Surgical, Sunnyvale, CA). Patients were placed in the flank position with typically a 12 mm AirSeal port and an additional five mm assistant port along with four eight mm robotic ports. Details of the procedures followed similar steps as outlined in prior robotic partial nephrectomy technique literature.²⁰

Outcomes of interest

The outcomes of interest of this study were predetermined. The primary outcome of interest was operative time among patients undergoing robotic partial nephrectomy. Operating room time was reported in hours and minutes and is defined as time of first incision to completion of skin closure. The secondary outcomes examined the association between preoperative predictors of increased operative time and downstream perioperative outcomes such as LOS and all grade complications. LOS was measured in days as day of admission to day of discharge when the patient physically exited the hospital. Complications were recorded within 90 days postoperatively.

Covariables

Patient, tumor, and perioperative characteristics were reported and compared. This included age, sex, body mass index (BMI), prior abdominal surgery, R.E.N.A.L nephrometry score, greatest tumor dimension on imaging, perirenal fat thickness, the presence/absence of adhesive perirenal fat, Charlson comorbidity index, and median ischemia time. Adhesive perirenal fat was retrospectively determined using the imaging scoring system known as the Mayo adhesive probability (MAP) score. All patients with a maximum MAP score of five were determined to have adhesive perirenal fat preoperatively.^15,20 All other variables were prospectively captured in our database.

Patients with prior abdominal surgery were compared with those without a history of surgery, and patients determined to have adhesive perirenal fat preoperatively were compared with those without. The study cohort was further split into three groups based on operative time; the fastest group was classified as the 25th percentile or less, middle group was the 25th–75th percentile, and the slowest group included 75th percentile or longer. The cohort was then analyzed using these three operative time groups (slow, middle, and fast) to compare the impact of patient and tumor characteristics.

Statistical analysis

Patient characteristics and unadjusted outcomes were compared using Fisher's exact and chi-square tests for categorical variables and t-statistics for continuous variables. Multivariable logistic regression models were built a priori using the following: age, BMI, sex, prior abdominal surgery, presence of adhesive perirenal fat, perirenal fat thickness, R.E.N.A.L nephrometry score, and greatest tumor dimension on imaging. We included operating room time in the multivariable model for LOS and any grade Clavien–Dindo complication.²¹ To perform internal validation of our final model, we used nonparametric bootstrapping (n = 1000) with stratified resampling for sensitivity analyses. We then generated a logistic regression model with bias-corrected 95% confidence intervals (CIs) using the 1000 bootstrapped samples.

All analyses were performed using Statistical Analysis System (SAS) version 9.4 software (SAS Institute, Cary, NC). All testing was two-sided, and the probability of a type I error was set at 0.05.

Results

Unadjusted analysis

Our study included 392 patients who underwent robotic partial nephrectomy (Table 1). Patients in the fastest 25th percentile had a median operation time of 146.5 minutes; patients in the middle 25th—75th percentile group had a median operative time of 206 minutes; and patients in the slowest 75th percentile group demonstrated median times of roughly 288 minutes (range 94–453 minutes). On unadjusted analysis, prior abdominal surgery (p = 0.02), higher nephrometry score (p = 0.001), larger tumor sizes (p = 0.002), increased perirenal fat thickness (p = 0.01), longer LOS (p < 0.0001), ischemia time (p = 0.0002), and increased EBL (p < 0.0001) were all statistically significant across the three groups (Table 1). Charlson comorbidity index was not significant (p = 0.27).

Table 1.

Patient Demographic, Tumor, and Perioperative Characteristics

	Fastest 25th percentile			Middle 25th–75th percentile			Slowest 75th percentile			p
n	104			190			98
Median operative time, minute (SD)	146.5 (18)			206 (23)			288 (39)
Age, years (SD)	59 (13)			62 (12)			61 (12)			0.20
Female (%)	32 (30.8)			66 (34.7)			64 (65.3)
BMI, kg/m² (SD)	31 (7.1)			33 (10)			32 (7.3)			0.40
Prior abdominal surgery										0.02
Yes (%)	42 (40.4)			98 (51.6)			56 (57.1)
No (%)	62 (59.6)			92 (48.4)			42 (42.9)
Nephrometry score (%)	4–6		7–10	4–6		7–10	4–6		7–10	0.001
Nephrometry score (%)	83 (79.8)		21 (20.2)	132 (69.5)		58 (30.5)	55 (56.1)		43 (43.9)
Tumor greatest dimension, cm (SD)	2.9 (1.9)			3.1 (1.4)			3.7 (1.5)			0.002
Perirenal fat thickness, cm (SD)	2.2 (0.9)			2.4 (0.9)			2.7 (1.1)			0.01
Adhesive perirenal fat										0.13
Yes (%)	73 (70.2)			122 (64.2)			74 (75.5)
No (%)	31 (29.8)			68 (35.8)			24 (24.5)
Charlson comorbidity index										0.27
0–3	38			58			42
4–7	10			29			21
8–11	1			1			2
Median ischemia time, minute (SD)	15			16			19			0.0002
Length of stay, days (SD)	2.2 (1.1)			2.6 (1.2)			3.2 (1.9)			<0.0001
Estimated blood loss, mL (SD)	106 (55)			164 (148)			218 (223)			<0.0001
Clavien–Dindo complication (%)	1–2	3a-3b		1–2	3a-3b		1–2	3a-3b		0.06
Clavien–Dindo complication (%)	4 (3.8)	6 (5.8)		17 (8.9)	5 (2.6)		8 (8.2)	3 (3.1)

BMI = body mass index; SD = standard deviation.

Nephrometry score: 4–6 = Low complexity; 7–10 = Intermediate complexity.

Operative time

We examined operative time as a continuous variable measured in hours using a bootstrapped multivariable logistic model with n = 1000 iterations (Table 2). After adjusting for the covariates, we found patients with prior abdominal surgery were associated with a 22-minute increase in operative time and adhesive perirenal fat was also associated with a 22-minute increase (p = 0.001, both). These variables had the most significant impact on operating room time. With each one-unit increase of nephrometry score, there was a 4-minute increase in the operative time (p = 0.028). Regarding surrounding renal fat, for each 1-cm increase in fat thickness, there was a 0.8-minute increase in operative time (p = 0.0028). Lastly, for every centimeter increase in tumor size, there was a 12-minute increase in the operative time (p < 0.001).

Table 2.

Association of Patient Demographics, Tumor, and Perioperative Factors on Operating Room Time

	Coefficient	Additional operating room time (minute)	95% adjusted CI	p
Age, years	−0.005	−0.3	−0.14 to 0.0038	0.25
BMI, kg/m²	0.0029	0.17	−0.010 to 0.016	0.43
Sex	−0.20	−12	−0.42 to 0.019	0.07
Prior abdominal surgery	0.36	22	0.15 to 0.56	0.001
Adhesive perirenal fat	0.37	22	0.15 to 0.59	0.001
Nephrometry score	0.068	4	0.0073 to 0.13	0.028
Perirenal fat thickness, cm	0.013	0.8	0.0015 to 0.025	0.0028
Greatest tumor dimension, cm	0.20	12	0.12 to 0.29	<0.001

CI, confidence interval; OR = odds ratio.

Length of stay

LOS measured in days was then analyzed using a multivariable regression model (Table 3). After adjusting for the covariates, we found for each 1-year increase in age, there was a 0.024-day increase in LOS [95% CIs (0.013–0.035)]. In addition, for every one-cm increase in tumor size, there was a 0.18-day associated increase in LOS [95% CI (0.070–0.28)]. Each 1-hour increase in operating room time was associated with a 0.25-day increased LOS [95% CI (0.092–0.41)]. The other covariates, including BMI, sex, prior abdominal surgery, adhesive perirenal fat, nephrometry score, and perirenal fat thickness, were not significant to impact hospitalization time. Figure 1 demonstrates the unadjusted association between LOS and operative room time. As the operating room time increased, there was a gradual rise in the associated LOS.

FIG. 1.

Line chart showing association of operative room time on LOS for robotic partial nephrectomy. LOS, length of stay.

Table 3.

Association of Patient Demographics, Tumor, and Perioperative Factors on Length of Stay

	Coefficient	95% adjusted CI	p
Age, years	0.024	0.013 to 0.035	<0.001
BMI, kg/m²	−0.0029	−0.022 to 0.015	0.76
Sex	0.24	−0.052 to 0.53	0.11
Prior abdominal surgery	0.041	−0.27 to 0.35	0.80
Adhesive perirenal fat	0.25	−0.046 to 0.55	0.097
Nephrometry score	0.037	−0.037 to 0.11	0.98
Perirenal fat thickness, cm	0.0045	−0.011 to 0.020	0.56
Greatest tumor dimension, cm	0.18	0.070 to 0.28	0.001
Operating room time	0.25	0.092 to 0.41	0.002

Complications

Any grade Clavien–Dindo complication model was similarly examined using a multivariable logistic regression model with n = 1000 bootstrap iterations (Table 4). After adjusting for potential confounding variables, only increased tumor dimension was statistically significant. There was 1.4 times greater odds (p = 0.009) associated with any grade complication for each one-cm increase in tumor size. Although adhesive perirenal fat neared statistical significance (p = 0.6), no other analyzed variable was demonstrated to be associated with increasing complication risk.

Table 4.

Association of Patient Demographics, Tumor, and Perioperative Factors on Any Grade Complication

	Bias corrected OR	95% adjusted CI	p
Age, years	1.0	0.96 to 1.0	0.97
BMI, kg/m²	1.0	0.95 to 1.1	0.96
Female sex	1.6	0.69 to 3.8	0.27
Prior abdominal surgery	1.7	0.79 to 3.5	0.18
Adhesive perirenal fat	2.9	0.98 to 8.7	0.06
Nephrometry score	1.1	0.92 to 1.3	0.30
Perirenal fat thickness, cm	1.0	0.98 to 1.1	0.16
Greatest tumor dimension, cm	1.4	1.1 to 1.9	0.009
Operating room time	0.78	0.52 to 1.2	0.22

Discussion

Our experience with robotic partial nephrectomy for localized renal masses suggests the following: (1) if the surgical patient has one or more of the following, including a large complex mass, significant perirenal fat, or prior abdominal surgery, the surgeon should anticipate significantly longer operative times; (2) longer operative times are associated with increased LOS for this procedure; and (3) larger tumor size is associated with increased complication rates. Because modifications to mitigate operative time appear most critical when these preoperative variables are identified, the exercise of identifying the presence or absence of these factors before partial nephrectomy should be readily available, simple to do, and useful.

Operative time has long been recognized as an important surrogate in perioperative outcomes across all surgical subspecialties.^22

–27 However, prior studies have not found an association between tumor characteristics (specifically nephrometry score) and longer operating room time for robotic partial nephrectomy.²⁸ Interestingly, our study found that tumor size and complexity appear to be the largest determinates that impact operative time, LOS, and overall complication rate.

Explanation of this relationship may be multifactorial; first, partial nephrectomy with larger tumors may consistently require more of the allowable 30 minutes for renal reconstruction, increasing operative time as well as increasing complication risk secondary to increased rates of bleeding and urinary extravasation; second, because clamp time is relatively predictable (<30 minutes in almost all cases), we would anticipate that most of the accumulated operative time may in fact be related to the preparation of the tumor before ischemia. This may be secondary to increased renal mobilization to visualize tumor edges and the surrounding parenchyma for anticipated suture placement as well as more expansive use of quality intraoperative ultrasound. These various iterations demonstrate why the tumor size itself may be the biggest predictor of time and morbidity from the procedure.

The presence of adhesive perirenal fat and increased perirenal fat thickness leads to increased operative complexity and time during robotic partial nephrectomy.^{16

–20,29,30} The MAP score is an image-based scoring system ranging from zero to five that helps to identify patients with “sticky” perirenal fat, which results in increased operative complexity during robotic partial nephrectomies.²⁰ Similarly, our findings found that adhesive fat and perirenal thickness are major predictors of operative time for these surgeries. Interestingly, BMI itself was not predictive of longer operating room time in contrast to the fat characteristics of the kidney itself.

Our study demonstrates the nuances of managing obese patients unique to robotic kidney surgery. Objective measures such as distance from fascia to skin, amount of visceral fat, perirenal fat thickness, in addition to subjective factors such as laxity of the abdomen, presence of a significant pannus, and the shape of the flank, can all impact operating room time. Although outside the scope of the present study, it appears that retroperitoneal fat more so than extrafascial obesity may be the more important driver of difficulty when considering robotic partial nephrectomy.

LOS is an increasingly scrutinized surrogate for cost-efficient care. Similar to our findings, prior studies have demonstrated the relationship between operative time and LOS for partial nephrectomy.^31
–33 The fastest operative time patients were hospitalized one less day on average than the slowest operative time patients (2.2 vs 3.3 days). Adding an hour of surgery increased hospital stay by one quarter of a day. Moreover, prior literature has demonstrated an association between longer LOS and increased complication rates.³³

In an era in which Medicare reimbursement may be classified as “outpatient,” “outpatient in a bed,” and “inpatient,” shifts between categories may be based on one half day of stay in either direction. These subtle differences can have huge cost implications and are important nuances to understand when assessing timing for patient discharge. In addition, as LOS is increasingly being used as a benchmark for evaluating quality of care, identifying preoperative variables that may impact this determination is important.

These findings should be considered in the context of several limitations. First, the retrospective methodology of this study creates intrinsic limitations consistent with any observational study.³⁴ To our knowledge, however, no longitudinal administrative data exist that collect all of the variables that we are interested in, such as nephrometry score, adhesive perirenal fat, and operative time. We also utilized the bootstrap resampling technique to provide internal validation of our model and as certain variables may have non-normal distribution.^35,36

Furthermore, our study relied on data collected in our prospective database, which relies on the electronic medical record. These data are obtained from a robust electronic medical record system used for clinical documentation and billing purposes. To minimize this risk, we had three authors independently collect the data and one author recheck the database to reduce the possibility of data collection errors. Furthermore, we did not specifically examine anterior vs posterior tumors and their relation to operating time and LOS. Given these factors are included in the nephrometry score, this would likely confound our results.

Our renal database, although constantly being refined, does not reliably reflect all the technical evolutions of robotic partial nephrectomy dating back to 2012. Although these nuances can certainly influence operative time, the focus of our analysis was to identify and analyze clear and objective preoperative factors that can be utilized to create a roadmap of surgical expectations. The da Vinci Xi robotic system was introduced in 2014 during our study period. Both the da Vinci Si and Xi systems were used in this study although our database does not include this information. The benefits of the da Vinci Xi system would likely be minimal given the experience of the surgeons in this study.

Moreover, in our study, LOS was measured in days rather than hours and minutes. Our model is neither granular nor statistically powerful enough to account for patient factors and their changes on LOS in terms of hours. Nevertheless, one increased day in the hospital is a significant period of time and our model accounts for these changes in association with operating room time. These limitations not withstanding this work represent the first to identify and quantify specific variables that affect operative time for robotic partial nephrectomy and underscore their impact on both LOS and overall complication rates.

Operative time in robotic partial nephrectomy is important. The impact between operative time and LOS is consistently demonstrated. Our findings highlight several factors, especially renal tumor and perirenal fat characteristics, are measurable with quantifiable objective impact on outcome. The ability to predict operative time, LOS, and complication rate is useful both for patient counseling and surgical planning such as operative schedule management, operative room support, and/or modification of resident or fellow participation. As hospitals and third-party payers attempt to create fair measures to compare surgeons on a global scale, identifying predictive variables for each operation lays the groundwork to establishing benchmarks that are individualized and achievable.

Conclusion

In patients undergoing robotic partial nephrectomy, prior abdominal surgery, larger and more complex tumors, and patients with increased Gerota's fat with or without inflammation can anticipate increased operative time. Tumor size itself appears to be the biggest predictor of operative time, LOS, and complication rates. It is important to integrate and quantify these important preoperative variables into operative planning to create fair and reasonable expectations for both patient and surgeon in regard to overall outcome.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No outside or institutional funding was provided for this research.

Abbreviations Used

References

Renal Cancer: Renal Mass & Localized Renal Cancer Guideline—American Urological Association. Available at https://www.auanet.org/guidelines/renal-cancer-renal-mass-and-localized-renal-cancer-guideline (last accessed March 30, 2020 ).

Gill

, Aron

, Gervais

, Jewett

MAS

. Clinical practice. Small renal mass. N Engl J Med, 2010; 362:624–634.

Ljungberg

, Albiges

, Bensalah

, et al. European Association of Urology Guidelines. 2018. Edition. In: Vol presented. European Association of Urology Guidelines Office, 2018. Available at http://uroweb.org/guideline/renal-cell-carcinoma/ (last accessed December 18, 2020 ).

Banegas

, Harlan

, Mann

, Yabroff

. Toward greater adoption of minimally invasive and nephron-sparing surgical techniques for renal cell cancer in the United States. Urol Oncol Semin Orig Investig, 2016; 34:433.e9–433.e17.

Wasserman

, Sobel

, Pareek

. Choice of surgical options in kidney cancer and surgical complications. Semin Nephrol, 2020; 40:42–48.

Capitanio

, Terrone

, Antonelli

, et al. Nephron-sparing techniques independently decrease the risk of cardiovascular events relative to radical nephrectomy in patients with a T1a-T1b renal mass and normal preoperative renal function. Eur Urol, 2015; 67:683–689.

Deng

, Chen

, Wang

, Liu

, Wang

, Fu

. Partial nephrectomy versus radical nephrectomy for large (≥ 7 cm) renal tumors: A systematic review and meta-analysis. Urol Oncol Semin Orig Investig, 2019; 37:263–272.

Kunath

, Schmidt

, Krabbe

, et al. Partial nephrectomy versus radical nephrectomy for clinical localised renal masses. Cochrane Database Syst Rev, 2017; 2017:CD012045.

Choi

, You

, Kim

, Rha

, Lee

. Comparison of perioperative outcomes between robotic and laparoscopic partial nephrectomy: A systematic review and meta-analysis. Eur Urol, 2015; 67:891–901.

10.

Larcher

, Muttin

, Peyronnet

, et al. The learning curve for robot-assisted partial nephrectomy: Impact of surgical experience on perioperative outcomes. Eur Urol, 2019; 75:253–256.

11.

Dagenais

, Bertolo

, Garisto

, et al. Variability in partial nephrectomy outcomes: Does your surgeon matter?. Eur Urol, 2019; 75:628–634.

12.

Cacciamani

, Medina

, Gill

, et al. Impact of surgical factors on robotic partial nephrectomy outcomes: Comprehensive systematic review and meta-analysis. J Urol, 2018; 200:258–274.

13.

Cacciamani

, Gill

, Medina

, et al. Impact of host factors on robotic partial nephrectomy outcomes: Comprehensive systematic review and meta-analysis. J Urol, 2018; 200:716–730.

14.

Taub

, Miller

, Cowan

, Dimick

, Montie

, Wei

. Impact of surgical volume on mortality and length of stay after nephrectomy. Urology, 2004; 63:862–867.

15.

Ishiyama

, Kondo

, Takagi

, et al. Impact of the mayo adhesive probability score on the complexity of robot-assisted partial nephrectomy. J Endourol, 2018; 32:928–933.

16.

Ioffe

, Hakimi

, Oh

, et al. Effect of visceral obesity on minimally invasive partial nephrectomy. Urology, 2013; 82:612–619.

17.

Raman

, Reynolds

, Hannon

. An increasing proportion of perinephric to subcutaneous fat is associated with adverse perioperative outcomes of robotic partial nephrectomy. J Robot Surg, 2016; 10:255–259.

18.

Motoyama

, Matsushita

, Watanabe

, et al. Significant impact of three-dimensional volumetry of perinephric fat on the console time during robot-assisted partial nephrectomy. BMC Urol, 2019; 19:132.

19.

, Schwentner

, Neururer

, Steiner

, Bartsch

, Peschel

. Robotic-assisted laparoscopic partial nephrectomy: Surgical technique and clinical outcomes at 1 year. BJU Int, 2009; 103:663–668.

20.

Davidiuk

, Parker

, Thomas

, et al. Mayo adhesive probability score: An accurate image-based scoring system to predict adherent perinephric fat in partial nephrectomy. Eur Urol, 2014; 66:1165–1171.

21.

Dindo

, Demartines

, Clavien

P-A

. Classification of surgical complications. Ann Surg, 2004; 240:205–213.

22.

Cheng

, Clymer

, Po-Han Chen

, et al. Prolonged operative duration is associated with complications: A systematic review and meta-analysis. J Surg Res, 2018; 229:134–144.

23.

Huang

K-H

, Kaplan

, Carter

, Lipsitz

, Hu

. The impact of radical prostatectomy operative time on outcomes and costs. Urology, 2014; 83:1265–1272.

24.

Bohl

, Ondeck

, Darrith

, Hannon

, Fillingham

, Della Valle

. Impact of operative time on adverse events following primary total joint arthroplasty. J Arthroplasty, 2018; 33:2256.e4–2262.e4.

25.

Hardy

, Davis

, Constantine

, et al. The impact of operative time on complications after plastic surgery: A multivariate regression analysis of 1753 cases. Aesthetic Surg J, 2014; 34:614–622.

26.

Jackson

, Wannares

, Lancaster

, Rattner

, Hutter

. Does speed matter? The impact of operative time on outcome in laparoscopic surgery. Surg Endosc, 2011; 25:2288–2295.

27.

Maggino

, Liu

, Ecker

, Pitt

, Vollmer

. Impact of operative time on outcomes after pancreatic resection: A risk-adjusted analysis using the American College of Surgeons NSQIP database. J Am Coll Surg, 2018; 226:844–857.e3.

28.

Mufarrij

, Krane

, Rajamahanty

, Hemal

. Does nephrometry scoring of renal tumors predict outcomes in patients selected for robot-assisted partial nephrectomy?. J Endourol, 2011; 25:1649–1653.

29.

Macleod

, Hsi

, Gore

, Wright

, Harper

. Perinephric fat thickness is an independent predictor of operative complexity during robot-assisted partial nephrectomy. J Endourol, 2013; 28:587–591.

30.

Borregales

, Adibi

, Thomas

, et al. Predicting adherent perinephric fat using preoperative clinical and radiological factors in patients undergoing partial nephrectomy. Eur Urol Focus, 2021; 7:397–403.

31.

Bazzi

, Sjoberg

, Grasso

AAC

, Bernstein

, Parra

, Coleman

. Predicting length of stay after robotic partial nephrectomy. Int Urol Nephrol, 2015; 47:1321–1325.

32.

Shumate

, Roth

, Ball

, Custer

, Thiel

. Factors associated with prolonged length of stay following robotic-assisted partial nephrectomy. Can J Urol, 2019; 26:9726–9732.

33.

Monn

, Jain

, Kaimakliotis

, Flack

, Koch

, Boris

. Examining the relationship between operative time and hospitalization time in minimally invasive and open urologic procedures. J Endourol, 2014; 28:1132–1137.

34.

Carlson

MDA

, Morrison

. Study design, precision, and validity in observational studies. J Palliat Med, 2009; 12:77–82.

35.

Efron

, Tibshirani

. Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy. Stat Sci, 1986; 1:54–75.

36.

Efron

, Tibshirani

. An Introduction to the Bootstrap. New York, NY; London: Chapman & Hall, 1993.