Outcomes of Extraperitoneal Robot-Assisted Radical Prostatectomy in the Morbidly Obese: A Propensity Score-Matched Study

Abstract

Introduction:

An increasing number of obese patients (body mass index [BMI] >30 kg/m²]) with localized prostate cancer are presenting as candidates for robot-assisted radical prostatectomy (RARP), which can be carried out using the transperitoneal or the extraperitoneal (EP) approach. Morbidly obese (BMI >40 kg/m²) patients present as an especially challenging surgical cohort. We sought to evaluate the perioperative and pathologic outcomes associated with EP-RARP in morbidly obese men.

Materials and Methods:

In this institutional review board-approved study, our prospectively collected database (Cancer Information Systems [CAISIS]) was reviewed. One thousand six hundred sixty-three patients underwent EP-RARP for localized prostate cancer at our institution between July 2003 and December 2013 by a single surgeon. Forty patients were considered morbidly obese. A propensity score-matched analysis was performed using multivariate analysis incorporating 10 covariates to identify the comparable group of patients with a BMI of >40 kg/m² and <40 kg/m².

Results:

Apart from BMI, the two groups were matched (all p-values >0.05). Despite a higher total operating time and estimated blood loss (EBL) in the morbidly obese (238 vs 176 minutes, p<0.0001, and 235 vs 192 mL, p=0.003, respectively), there were no differences in the ability to perform nerve-sparing or pelvic lymphadenectomy, or the length of stay. While the morbidly obese had a higher rate of harboring more aggressive disease on final pathology (pT3 rates, 27.5 vs 7.5%, respectively), there were no differences in other postoperative pathologic parameters, such as prostate specimen weight, positive surgical margin status, and Gleason score sum. Moreover, there were no differences in intra- or postoperative complications between the two groups.

Conclusions:

The morbidly obese cohort harbored more aggressive disease with the difference in the proportion of pathologic T3 disease statistically significant. Apart from increased total operating time and EBL in the morbid obese, EP-RARP leads to comparable perioperative and pathologic outcomes to the nonmorbidly obese. Consideration should be given to added operating room time when operating on the morbidly obese.

Introduction

Obesity (body mass index [BMI] >30 kg/m²) is regarded as an epidemic with almost one third of the US adult population classified as obese.¹ It is associated with chronic diseases such as diabetes, coronary artery disease, sleep apnea, and some cancers.² The association between prostate cancer, the most common solid organ malignancy in men, and obesity is complex with obese men shown to have more aggressive prostate cancer than nonobese men.^3,4

Radical prostatectomy is the gold standard treatment for localized prostate cancer.⁵ With the high prevalence of both obesity and prostate cancer, as well as the wide acceptance of the robot-assisted approach to performing a prostatectomy (robot-assisted radical prostatectomy [RARP]) among the urologic community and patients,⁶ an increasing number of obese and morbidly obese patients (BMI ≥40 kg/m²) are presenting as candidates for RARP. The procedure can be carried out using a transperitoneal (TP) or an extraperitoneal (EP) approach.

Several studies have compared the outcomes of obese and nonobese patients undergoing a RARP.^{7

–18} However, only two studies have focused specifically on outcome in the morbidly obese.^16,17 These studies only utilized the TP approach. The outcome of EP-RARP in the morbidly obese was studied in a previous report as part of a larger cohort, but the number of morbidly obese patients was limited to three.¹⁸ Therefore, the outcomes of EP-RARP in a morbidly obese patient population have not been previously examined to a significant extent.

Herein, we report the perioperative, postoperative, and pathologic outcomes, and complications associated with EP-RARP in the morbidly obese, compared with a propensity score-matched group of nonmorbidly obese men.

Materials and Methods

The institutional review board approval was obtained for this study. We reviewed data from our prospectively collected database (Cancer Information Systems [CAISIS]) maintained from the beginning of our experience with RARP. All surgeries were performed by one surgeon (J.J.) at a tertiary academic center between July 2003 and December 2013. Two thousand six hundred sixty-three RARP (1663 EP, 1000 TP) were performed during this time period. A total of 40 morbidly obese (BMI >40 kg/m²) males who had undergone EP-RARP were identified.

Propensity score matching was used to identify a comparative group of 40 nonmorbidly obese patients (BMI <40 kg/m²) who had also undergone EP-RARP in the same time period. This was performed to minimize the bias inherent in comparing a nonmatched cohort of nonmorbidly obese men, who may have a different risk profile, and different cancer characteristics, such as a lower comorbidity status or less aggressive cancer, respectively.

The following patient and tumor characteristics were extracted: age, BMI, American Society of Anesthesia (ASA) score, comorbidities, previous abdominal surgeries, prostate-specific antigen (PSA), biopsy Gleason score sum, and clinical tumor stage (T-stage). The operative and perioperative outcome variables measured were total operative time (creation of pneumoperitoneum, trocar placement, specimen retrieval, closure, and console times), estimated blood loss (EBL), transfusion rates, length of stay, whether a pelvic lymphadenectomy was carried out, and the nerve-sparing status. Postoperative pathologic variables measured were prostate specimen weight, Gleason score sum, T-stage, and surgical margin status. Intraoperative or postoperative complications were classified according to the modified Clavien–Dindo classification.¹⁹

The World Health Organization definition of BMI was used (<25 kg/m²—normal, 25 to 30 kg/m²—overweight, 30 to 35 kg/m²—class 1 obesity, 35 to 40 kg/m²—class 2 obesity, and >40 kg/m²—class 3 or morbid obesity). Blood loss was estimated by subtracting the irrigation fluid, and estimated urine output from the blood in the suction canister at the end of the procedure. Length of stay was calculated from the time of the admission until patient discharge. Patients with intermediate and high-risk disease according to the D'Amico risk classification underwent bilateral pelvic lymphadenectomy Nerve sparing was scored as none, partial, or complete, and noted if unilateral or bilateral. It was deemed to be complete only if carried out bilaterally.

Histology was confirmed at our institution in cases of outside referral. The pathologic assessment of the prostatectomy specimens was performed according to the Stanford protocol. The presence of cancer cells at the inked margin was regarded as evidence of positive surgical margin.

We used the successive generations of the da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA) as they became available. In the beginning of our experience, we used a three-arm daVinci system.

The procedures were carried out using our standardized EP technique of a “W” configuration using six trocars.²⁰ Patients are placed in the Trendelenburg position after bladder drainage. The EP space is created under direct vision using a 0-degree laparoscope, placed through the balloon dilator (OMS-XB2 Extraview™; Covidien, Mansfield, MA), and inserted in a 1-cm paraumbilical fascial incision. Postoperative care was standardized for all patients with the removal of the drain when the output was less than 40 to 60 mL in an 8-hour shift. The Foley catheter was removed in 8 to 10 days. Unless there were contraindications, mechanical and chemical prophylaxes for the prevention of deep vein thrombosis/pulmonary embolism were used postoperatively until discharge from the hospital.

Statistical analysis

Group 1 and 2 included patients with a BMI of >40 kg/m² and <40 kg/m², respectively. The mean and standard deviation of continuous variables, and the frequency of categorical variables were calculated. Two-sample t-test was used to compare the mean values of continuous variables, and the Pearson's chi-squared or Fisher's exact test was used to compare the distributions of categorical variables in the both groups. All p-values are two-sided, with the significance level of each comparison set at 0.05. The analysis was implemented with SAS 9.3 (SAS Institute, Inc., Cary, NC).

Since the data are not from a randomized clinical trial, we used the propensity score method to select patients for group 2, so that patients in the two groups were comparable. To generate the propensity score matching, multiple logistic regression was used based on 10 covariates as independent variables: age, diabetes mellitus, hypertension, dyslipidemia, coronary artery disease, ASA score, prior abdominal surgery, overall Gleason score, PSA level, clinical T-staging, and group membership (BMI >40 kg/m² and <40 kg/m²) as binary dependent variables. A score of 0 to 1 was generated. A 1:1 propensity score matching was performed to identify the comparable group of patients with a BMI of >40 kg/m² and <40 kg/m².

Results

The mean BMI of the morbidly obese and the nonmorbidly group were 42.9 kg/m² and 29.4 kg/m², respectively. Among the nonmorbidly obese cohort, 7.5%, 42.5%, 37.5%, and 5% had a BMI of <25 kg/m² (normal), 25 to 30 kg/m² (overweight), 30 to 35 kg/m² (class 1 obesity), and 35 to 40 kg/m² (class 2 obesity) respectively. Apart from BMI, groups 1 and 2 were matched with respect to all other variables measured, with all p-values >0.05 (Table 1). Specifically, the groups were identical in their risk of perioperative complications (mean ASA 2.6 in both groups), which is a reflection of the propensity score matching.

Table 1.

Preoperative Variables for 40 Morbidly Obese Patients Compared with a Propensity Score-Matched Cohort of 40 Nonmorbidly Patients Undergoing Extraperitoneal–Robot-Assisted Radical Prostatectomy

Variable	Morbidly obese BMI >40 kg/m²	Not morbidly obese BMI <40 kg/m²	p-Value
Mean (SD; median)
BMI	42.9 (5.4; 41.8)	29.39 (1.2; 31.8)	<0.0001
Age,^a years	58.2 (7; 59)	59 (4.2; 60)	0.54
PSA^a level, ng/mL	5.5 (0.9; 5.1)	5.6 (0.5; 4.2)	0.54
n (%)
Diabetes^a	14 (35)	8 (20)	0.21
Dyslipidemia^a	19 (47.5)	19 (47.5)	1.00
Coronary artery disease^a	4 (10)	4 (10)	1.00
Hypertension^a	33 (82.5)	27 (67.5)	0.20
Prior abdominal surgery^a	10 (25)	19 (47.5)	0.06
ASA^a	2.6 (0; 3)	2.6 (1.4; 3)	1.00
Preoperative Gleason score sum^a			0.93
6	27 (67.5)	9 (72.5)
7	9 (22.5)	9 (22.5)
≥8	3 (7.5)	2 (5)
Clinical stage^a			0.53
T1	33 (82.5)	36 (90)
T2	7 (17.5)	4 (10)
T3	0	0

Variables used for propensity score matching.

ASA=American Society of Anesthesia; BMI=body mass index; PSA=prostate-specific antigen; SD=standard deviation.

Table 2 outlines the outcomes. While the total mean operating time (creation of pneumoperitoneum, trocar placement, specimen retrieval, closure, and console times) and EBL were higher in the morbidly obese (238 vs 176 minutes, p<0.0001, and 235 vs 192 mL, p=0.003 respectively), this did not impact the length of stay, with no difference in this parameter between the two groups. Majority of patients were discharged the day following their procedure.

Table 2.

Peri- and Postoperative Variables, and Complications for 40 Morbidly Obese Patients Compared with a Propensity Score-Matched Cohort of 40 Nonmorbidly Patients Undergoing Extraperitoneal–Robot-Assisted Radical Prostatectomy

Variable	Morbidly obese BMI >40 kg/m²	Not morbidly obese BMI <40 kg/m²	p-Value
Mean (SD; median)
OT, minutes	238 (70.5; 224)	176.4 (46.8; 167.5)	<0.0001
Console time	175 (49.8; 199)	138 (40.9; 132.5)	0.0008
EBL	235 (70.7; 250)	192.4 (56.5; 160)	0.003
n (%)
Nerve sparing			0.85
None	14 (35)	16 (40)
Unilateral	11 (27.5)	9 (22.5)
Bilateral	15 (37.5)	15 (37.5)
PLND			0.36
Performed	18 (45)	13 (32.5)
Not performed	22 (55)	27 (67.5)
Mean (SD; median)			0.06
Prostate weight	61.2 (12.0; 61)	65.6 (8.4; 58.5)
n (%)			0.36
PSM	4 (10)	1 (2.5)
Pathologic stage			0.04
pT2	29 (72.5)	37 (92.5)
pT3	11 (27.5)	3 (7.5)
Postoperative Gleason score sum			0.50
6	19 (47.5)	24 (60)
7	19 (47.5)	14 (35)
≥8	2 (5)	2 (5)
Mean (median)			1.00
LOS, days	1.2 (1)	1.1 (1)
n (%)
Intraoperative complications			1.00
None	39 (97.5)	40 (100)
Transfusion	0	0
Transient hypotension	1 (2.5)	0
Clavien-classified complications
Grades 1–2	1 (2.5)	0
Grades 3–4	0	0
Postoperative complications			0.71
None	35 (87.5)	37 (92.5)
Transfusion	1 (2.5)	0
UTI	1 (2.5)	0
Lower limb cellulitis	1 (2.5)	0
Wound infection	1 (2.5)	0
Bladder neck contracture	1 (2.5)	0
Lymphocele	0	1 (2.5)
Clot retention	0	1 (2.5)
Urine leak	0	1 (2.5)
Clavien-classified complications
Grades 1–2	4 (10)	1 (2.5)
Grades 3–4	1 (2.5)	2 (5)

EBL=estimated blood loss; LOS=length of stay; OT=operation time; PLND=pelvic lymph node dissection; PSM=positive surgical margin.

There were no significant differences in the ability to perform nerve-sparing (uni- or bilateral) or pelvic lymphadenectomy. There was no difference in lymph node yield. The median (range) number of lymph nodes was 5 (2–21) and 5 (1–13) in the two groups (BMI >40 kg/m² and BMI <40 kg/m²), respectively.

While the morbidly obese had a higher rate of harboring more aggressive disease on final pathology than the nonmorbidly obese group (pT3 rates 27.5% vs 7.5%, respectively, p=0.04), there were no significant differences in other postoperative pathologic parameters, such as prostate weight, positive surgical margin status, and Gleason score sum.

The only transfusion required was in a patient in group 1 postoperatively. There were no significant differences in intra- or postoperative complications between the two groups.

No cases underwent conversion to either the TP-RARP approach or an open procedure.

Discussion

RARP can be carried out by the EP or the TP route, the latter being more commonly employed. Advantages and disadvantages have been reported for both approaches.²¹ Patient with previous abdominal surgeries may benefit from an EP approach. Intra-abdominal adhesions can be avoided, thereby potentially decreasing surgical time spent in adhesiolysis and the associated risk of bowel injury.²² In the EP approach, the peritoneum serves as a natural barrier or retractor, obviating the need for bowel retraction from the operative surgical field. This lessens the need for a steep Trendelenburg positioning (which can increase anesthetic risks in a patient with poor pulmonary reserve). The EP route potentially allows containment of bleeding, and urine leak outside of peritoneum, should they occur. On the other hand, the perceived disadvantages associated with the EP approach include unfamiliarity with access and instruments, relatively limited working space, difficulty in spacing the trocars especially with the use of the fourth arm, increased risk of lymphocele formation following pelvic lymph node dissection, and increased difficulty in patients with previous laparoscopic inguinal herniorrhaphy with mesh. In an extended pelvic lymphadenectomy, the TP approach may afford better exposure of the upper limits of the template of dissection.^23,24

Our study has shown that apart from operative time and EBL, all other perioperative and postoperative outcomes, and complications were comparable between morbidly obese and nonmorbidly obese patients undergoing EP-RARP.

The mean total operating time from initial incision to dressing placement of just <4 hours in the morbidly obese can potentially be explained by the incremental effects of the additional times spent during creation of the EP space (relatively more dissection of the fatty tissue is required to reach the anterior rectus fascia before its incision and subsequent placement of the balloon dilator and the laparoscope), placement of the trocars (especially the 5 mm assistant trocar), and closure of rectus fascia, which can be relatively difficult to reach. The currently available instruments, which are only 15 cm long, can be limiting in the morbidly obese group. There is greater difficulty encountered in pushing the peritoneum cephalad, during the creation of the EP space laterally, to place the assistant trocars and the fourth arm. In our study, there was a mean difference of 63 minutes between the total operating time and console time in the morbidly obese group, compared with a mean difference of 38 minutes in the nonmorbidly obese group. However, our study has also shown a longer mean console time for the morbidly obese (175 vs 138 minutes). This is likely due to increased inherent surgical difficulty in the morbidly obese patient. Large amount of intra-abdominal or perivesical fat may affect visualization of the operative field, increasing the difficulty level, and console times as a result. Extra time is also necessary to ensure adequate positioning on the operating table, ensuring that the patient does not slide during the Trendelenburg positioning.

Although statistically significant, the mean EBL in the morbidly obese compared with the nonmorbidly obese was higher by about 40 mL only, which is not clinically significant. Only one patient in the morbidly obese cohort required a transfusion. Intraoperative blood loss was estimated at 150 mL. He had bleeding in the postoperative period, which was successfully managed with the Foley traction, and 1 unit of packed red cell transfusion.

A greater number of patients in the morbidly obese cohort underwent pelvic lymphadenectomy. This, however, was not statistically significant. Our decision to perform a lymphadenectomy is based on the characteristics of the cancer in each patient. Lymphadenectomy through the EP approach has been described as challenging due to the perceived lesser operative space, and this issue can be compounded in the morbidly obese.^23,24 The average lymph node yield was similar in both cohorts.

It has been suggested that tumor biology is impacted negatively in the obese, with such patients harboring more aggressive prostate cancer (increased rates of pathologic stage T3 and T4 disease, higher grade disease, and higher volume tumors).³ Freedland concluded that the association between obesity, higher grade cancer, and advanced stage has become stronger with time.⁴ In keeping with such previous reports,^3,4 our morbidly obese cohort harbored more aggressive disease with the difference in the proportion of pathologic T3 disease statistically significant (p=0.04).

Contrary to what was expected, the incidence of Clavien grade 3 to 4 postoperative complications was less in the morbidly obese group (although statistically insignificant).

The cases were spread almost equally over the duration of 11 years of this study, with only a slightly higher number performed over the last 6 years (n=23) compared with the first 5 years (n=17). This is reflective of the increasing experience and referral of challenging cases to the surgeon. We did not find a difference in risk classification, complications, or positive surgical margin rates between patients who underwent surgeries in the first 5 years compared with those treated in the last 6 years. However, it cannot be denied that increased experience of the surgeon and the team can account for the overall low complication rates seen in our series.

Two previous studies have explored the outcome of TP-RARP in the morbidly obese. Yates et al. reviewed outcomes in a small cohort of 15 morbidly obese patients.¹⁶ They concluded that RARP is feasible in the morbidly obese population. Abdul-Muhsin et al. reported increased EBL in a cohort of 44 morbidly obese patients, findings similar to ours.¹⁷ However, unlike our study, they found no difference in operative times and an increased rate of positive surgical margin in the morbidly obese.

Although there were a number of morbidly obese patients in our TP group of 1000 patients, our objective was not to compare the outcomes of EP-RARP and TP-RARP in the morbid obese due to the inherent bias associated with such a comparison. Our practice is to perform TP-RARP in patients with high-grade disease, to facilitate access to the cephalad boundary of the pelvic lymphadenectomy dissection template, and to lessen the risk of lymphocele. As such, the patient population in the TP-RARP cohort of the morbidly obese has worse preoperative pathologic features when compared with those who had an EP-RARP. Any attempt at propensity matching would lead to a very small number of patients in the final analysis.

The limitation of the present study is its retrospective nature, although the database was prospectively maintained. In addition, this is a single high-volume surgeon series from a tertiary academic center and the results are not generalizable.

The strength of our study is that, to our knowledge, it is the first such report, which explores the outcome of EP-RARP in the morbidly obese. In addition, bias was minimized by the utilization of propensity score matching to establish a cohort of nonmorbidly obese men.

Conclusion

Based on our results, in the hands of an experienced robotic surgeon, EP-RARP can be safely offered to the morbidly obese men with acceptable peri- and postoperative clinical and pathologic outcomes, with low morbidity. Apart from increased operative time and EBL, the outcomes of EP-RARP in morbidly obese are similar to nonmorbidly obese men. Consideration should be given to added operating room time when operating on the morbidly obese.

Footnotes

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Ogden

, Carroll

, Kit

, et al. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA, 2014; 11:806–814.

Guh

, Zhang

, Bansback

, et al. The incidence of co-morbidities related to obesity and overweight: A systematic review and meta-analysis. BMC Public Health, 2009; 9:88.

Allott

, Masko

, Freedland

. Obesity and prostate cancer: Weighing the evidence. Eur Urol, 2013; 63:800–809.

Freedland

, Aronson

. Obesity and prostate cancer. Urology, 2005; 65:433–439.

Bill-Axelson

, Holmberg

, Garmo

, et al. Radical prostatectomy or watchful waiting in early prostate cancer. N Engl J Med, 2014; 370:932–942.

Chang

, Kibel

, Brooks

, Chung

. The impact of robotic surgery on the surgical management of prostate cancer in the USA. BJU Int, 2014 [Epub ahead of print]; DOI: 10.1111/bju.12850.

Wiltz

, Shikanov

, Eggener

, et al. Robotic radical prostatectomy in overweight and obese patients: Oncological and validated-functional outcomes. Urology, 2009; 73:316–322.

Boczko

, Madeb

, Golijanin

, et al. Robot-assisted radical prostatectomy in obese patients. Can J Urol, 2006; 13:3169–3173.

Chalasani

, Martinez

, Lim

, et al. Impact of body mass index on perioperative outcomes during the learning curve for robot-assisted radical prostatectomy. Can Urol Assoc J, 2010; 4:250–254.

10.

Moskovic

, Lavery

, Rehman

, et al. High body mass index does not affect outcomes following robotic assisted laparoscopic prostatectomy. Can J Urol, 2010; 17:5291–5298.

11.

Crocitto

, Satterthwaite

, Wilson

, et al. Can robotic assisted laparoscopic prostatectomy be recommended to obese patients?. J Robotic Surg, 2008; 1:297–302.

12.

Boorjian

, Crispen

, Carlson

, et al. Impact of obesity on clinicopathologic outcomes after robot-assisted laparoscopic prostatectomy. J Endourol, 2008; 22:1471–1476.

13.

Castle

, Atug

, Woods

, et al. Impact of body mass index on outcomes after robot assisted radical prostatectomy. World J Urol, 2008; 26:91–95.

14.

Khaira

, Bruyere

, O'Malley

, et al. Does obesity influence the operative course or complications of robot-assisted laparoscopic prostatectomy?. BJU Int, 2006; 98:1275–1278.

15.

Ahlering

, Eichel

, Edwards

, et al. Impact of obesity on clinical outcomes in robotic prostatectomy. Urology, 2005; 65:740–744.

16.

Yates

, Munver

, Sawczuk

. Robot-assisted laparoscopic radical prostatectomy in the morbidly obese patient. Prostate Cancer, 2011; 2011:618–623.

17.

Abdul-Muhsin

, Giedelman

, Samavedi

, et al. Perioperative and early oncological outcomes after robot-assisted radical prostatectomy (RARP) in morbidly obese patients: A propensity score-matched study. BJU Int, 2014; 113:84–91.

18.

Sundi

, Reese

, Mettee

, Trock

, Pavlovich

. Laparoscopic and robotic radical prostatectomy outcomes in obese and extremely obese men. Urology, 2013; 82:600–605.

19.

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.

20.

Joseph

, Rosenbaum

, Madeb

, et al. Robotic extraperitoneal radical prostatectomy: An alternative approach. J Urol, 2006; 175(3 Pt. 1):945–950.

21.

Lee

, Diaz

, Cho

, Choi

. Meta-analysis of transperitoneal versus extraperitoneal robot-assisted radical prostatectomy for prostate cancer. J Laparoendosc Adv Surg Tech A, 2013; 23:919–925.

22.

Xylinas

, Ploussard

, Durand

, de la Taille

. Robot-assisted extraperitoneal laparoscopic radical prostatectomy: A review of the current literature. Urol Oncol, 2013; 31:288–293.

23.

Lee

, Diaz

, Cho

, Yu

, Chung

, Ham

, Choi

. Lymphocele after extraperitoneal robot-assisted radical prostatectomy: A propensity score-matching study. Int J Urol, 2013; 20:1169–1176.

24.

Danuser

, Di Pierro

, Stucki

, Mattei

. Extended pelvic lymphadenectomy and various radical prostatectomy techniques: Is pelvic drainage necessary?. BJU Int, 2013; 111:963–969.