Comparison of Three Approaches to Single-Port Robot-Assisted Radical Prostatectomy: Our Institution's Initial Experience

Abstract

Background:

We aimed to compare three robot-assisted radical prostatectomy (RARP) approaches—Retzius sparing (RS), extraperitoneal (EP), and transperitoneal (TP)—performed at our institution using the da Vinci^® single-port (SP) platform (Intuitive Surgical, Sunnyvale, CA).

Materials and Methods:

We retrospectively reviewed the records of 101 patients who underwent SP-RARP at our institution and stratified them into three cohorts based on the RARP approach: RS (n = 32), EP (n = 30), and TP (n = 39). Data regarding preoperative patient characteristics, perioperative characteristics, oncologic outcomes, and early functional outcomes were collected. The Fisher's exact test and chi-square tests were utilized for categorical variables, and the Kruskal–Wallis test was utilized for numerical variables. Wilcoxon rank-sum tests were utilized for pairwise comparisons. A two-tailed p < 0.05 was considered significant.

Results:

All three cohorts were largely similar in terms of preoperative patient characteristics. Operative time was significantly different between cohorts (p < 0.001), with the RS approach having a faster mean operating time than the TP approach (208 ± 40 minutes vs 248 ± 36 minutes, p < 0.001). Clinically significant margin rates did not differ significantly between cohorts (p = 0.861). Postoperative continence differed significantly between cohorts (p < 0.001); higher continence rates were observed in RS vs EP—94% (30/32) vs 52% (15/29), respectively, p < 0.001. Return of erectile function also differed significantly between cohorts (p = <0.001); higher erectile function recovery rates were observed in RS vs EP—88% (28/32) vs 41% (11/27), respectively, p < 0.001—and in RS vs TP—88% (28/32) vs 60% (22/37), respectively, p = 0.014. Median (IQR) follow-up time was 150 (88–377) days.

Conclusions:

RS SP-RARP is associated with improved early functional outcomes when compared with both EP and TP approaches. These benefits are achieved while maintaining equivalent oncologic outcomes. Further research is needed to optimize the patient selection paradigm for the SP-RARP approach.

Background

Radical prostatectomy is a standard in the management of clinically localized prostate cancer and >80% of surgeries in the United States are performed robotically.^1,2 Conventional robot-assisted radical prostatectomy (RARP) is performed using the multiport (MP) [Si, X, Xi] da Vinci^® surgical robotic platform (Intuitive Surgical, Sunnyvale, CA). Further innovation has resulted in the development of the da Vinci single-port (SP) platform (Intuitive Surgical), approved by the U.S. Food and Drug Administration in 2018.

Compared with MP-RARP, SP-RARP minimizes invasiveness while maintaining efficacious oncologic outcomes.^3,4 Surgical approaches utilized during RARP include the standard retropubic transperitoneal (TP) approach, retropubic extraperitoneal (EP) approach, and Retzius-sparing (RS) approach.^5–7

The RS approach to RARP has been extensively evaluated in the setting of MP surgery, with benefits of improved continence recovery rates and faster regain of continence.^8–10 SP-RARP approaches evaluated include TP, EP, and RS approaches.^5,6,11

In a comparative study of EP and TP approaches, Kaouk et al demonstrate that EP SP-RARP is associated with lower length of stay, lower operative time, and decreased postoperative narcotic administration.⁶ However, the RS approach to SP-RARP has not been fully evaluated and compared against EP and TP approaches to SP-RARP. We aim to provide a comparison of perioperative, oncologic, and early functional outcomes between the RS, EP, and TP approaches to SP-RARP from our institution's initial experience using the da Vinci SP system.

Materials and Methods

Subjects

Upon obtaining institutional review board approval, we conducted a retrospective review of 101 consecutive patients receiving SP-RARP performed by two surgeons at our institution from November 2019 to June 2021. The cases included represent each surgeon's initial experience with the SP platform for RARP. Our final cohort of 101 patients was stratified into three groups based on surgical approach: RS (n = 32), TP (n = 39), and EP (n = 30).

SP-RARP surgical technique

Our surgical technique utilizing the da Vinci SP platform has been previously described.^12,13 The colinear single-entry device allows for the performance of RARP through a single 2.5 cm cannula using three robotic arms and a flexible camera. In all cases, the Hasson technique was used for placement of the periumbilical SP trocar. For the RS and TP approaches, the SP robotic trocar was placed directly into the abdomen, above the umbilicus, based on distance from target anatomy. For the EP approach, the SP robotic trocar was placed through a GelPOINT^® Mini (Applied Medical, Rancho Santa Margarita, CA). Standard pelvic lymph node dissection (PLND) was performed in 90 out of 101 cases based on treating surgeon's discretion regarding the likelihood of lymph node involvement. Foley catheters were left in place for 7 days after surgery.

EP approach

A 3-cm subumbilical incision was made. The abdominal fascia was exposed and incised, and digital development of the space of Retzius was performed. “Air docking” was performed by placing a GelPOINT Mini into the EP space with the SP trocar (Fig. 1A). Before docking the SP robot, the patient was placed in <10° of Trendelenburg position. An 8-mm AirSeal^® (CONMED, Utica, NY) assistant port was placed under direct vision, taking care to avoid the epigastric vessels. Once docked, the space of Retzius was developed followed by opening of the endopelvic fasciae.

FIG. 1.

Intraoperative images of the single-port EP approach. (A) Initial view of the retropubic space after digital development and “air docking.” (B) Electrocautery dissection of the dorsal venous complex. (C) Anterior dissection of the vasa deferentia and seminal vesicles bilaterally. (D) Vesicourethral anastomosis. EP = extraperitoneal. Color images are available online.

The dorsal venous complex and puboprostatic ligaments were divided with electrocautery (Fig. 1B), and the anterior bladder neck was opened. The Foley catheter was grasped and brought midline, followed by posterior bladder neck dissection. The vasa deferentia and seminal vesicles were approached anteriorly and dissected bilaterally (Fig. 1C). The Foley catheter was retrieved and placed on tension. The posterior bladder neck was divided, and the prior seminal vesicle and vas dissections were reached. These structures were anteriorly lifted, and Denonvilliers' fascia was dissected from the posterior prostate.

Prostatic pedicles were controlled with either electrocautery or clips. Neurovascular bundles were dissected in an antegrade manner, and the urethra was transected at the apex of the prostate. Running 3-0 V-Loc suture (Medtronic, Minneapolis, MN) was utilized to accomplish vesicourethral anastomosis, taking full-thickness bites of the urethral stump and bladder neck (Fig. 1D). Cooper's ligament, the obturator nerve, and the external iliac vein were utilized as boundaries of dissection for bilateral pelvic lymph nodes. All specimens were extracted using an entrapment sac at the SP trocar site.

TP approach

First, the SP trocar was placed, and the pneumoperitoneum was established. An 8 mm AirSeal (CONMED) assistant port was then placed in the lower left quadrant. Then, the SP robot was docked, and the patient was placed in 20° to 25° of Trendelenburg position. With the exception of a posterior dissection of the seminal vesicles and vas deferens, TP SP-RARP was performed following the same general steps already described for the EP approach.

RS approach

After patients were positioned at 20° to 25° of Trendelenburg, the SP trocar was placed above the umbilicus in a standard pelvic configuration. An assistant port was placed to the left of the trocar. The camera was placed in the down position (6 o'clock), monopolar scissors were used in the right hand (3 o'clock), the bipolar Maryland forceps in the left hand (9 o'clock), and Cadiere forceps in the up position (12 o'clock). At the anterior surface of the rectovesical pouch, the parietal peritoneum was incised. Dissection of the seminal vesicles followed by ligation of the vas deferens was performed, and then carried antegrade over the posterior and posterolateral surfaces of the prostate until the prostatic apex was reached (Fig. 2A).

FIG. 2.

Intraoperative images of the single-port RS approach. (A) Electrocautery transection of the right prostatic pedicle. (B) Exposure and dissection of the bladder neck. (C) Vesicourethral anastomosis. (D) Initiation of right pelvic lymph node dissection. BN = bladder neck; RS = Retzius sparing. Color images are available online.

When determined possible from an oncologic standpoint, full nerve sparing was performed. Upon identification of the vesicoprostatic junction, this structure was divided posteriorly. After identifying the Foley catheter, division of the anterior bladder neck occurred (Fig. 2B). Next, the anterior surface of the prostate was bluntly dissected until the apex of the gland was reached. Then, the urethra was divided. Vesicourethral anastomosis was then performed using double arm 3-0 V-loc sutures (Fig. 2C). Finally, PLND was performed with the inguinal ligament, obturator nerve, and external iliac vein utilized as boundaries of dissection (Fig. 2D).

Variables

Data collected included preoperative patient and tumor characteristics, perioperative outcomes, oncologic outcomes, and early functional outcomes. Preoperative characteristics included patient age, body mass index (BMI), initial prostate specific antigen (PSA) level, biopsy Gleason grade group, American Society of Anesthesiologists (ASA) physical status classification score, and history of abdominal or pelvic surgery.

Perioperative outcomes assessed included operative time, estimated blood loss (EBL), length of stay, usage of opioid medications for pain control, postoperative morphine milligram equivalents (MMEs), prostate size, readmission within 90 days of surgery, presence of postoperative lymphocele within 90 days of surgery, and Clavien–Dindo grade II or greater complications within 90 days of surgery.

Oncologic outcomes assessed include pathologic Gleason grade group, pathologic T-stage, performance of PLND, lymph node yield if PLND was performed, biochemical recurrence, and clinically significant margin rates.¹⁴ Rates of receiving adjuvant and salvage radiation were also assessed. Biochemical recurrence was defined as a postoperative PSA >0.2 ng/mL and clinically significant margin was defined as >3 mm and Gleason subpattern 4+.^15,16

Patient-reported functional outcomes were assessed at postoperative visits and included continence defined as using zero pads per day and return of satisfactory erection adequate for sexual activity.

Statistical analysis

The Fisher's exact test and chi-square tests were utilized for categorical variables, and the Kruskal–Wallis test for numerical variables. Wilcoxon rank-sum tests were utilized for pairwise comparisons. A two-tailed p-value of <0.05 was considered significant. Statistical analyses were performed in R version 3.5.2 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Of the 101 patients receiving SP-RARP, RS was performed in 32 cases, EP in 30, and TP in 39. There were no procedural conversions to open surgery. In two EP cases, postoperative vesicourethral anastomotic leaks occurred with one case requiring drain placement.

All preoperative patient characteristics evaluated are described in Table 1 and did not differ significantly between groups. Overall mean patient age was 63.1 ± 7.66 (p = 0.523) and overall mean BMI was 29.8 ± 5.04 (p = 0.059). Regarding biopsy Gleason grade grouping, 54 patients had grade group 2 disease and 21 patients had grade group 3 disease.

Table 1.

Preoperative Characteristics

Preoperative characteristics	Total	RS	EP	TP	p	Pairwise comparisons
	n = 101	n = 32	n = 30	n = 39		RS vs EP	RS vs TP	EP vs TP
Age (years), mean ± SD	63.1 ± 7.7	62.0 ± 7.7	64.6 ± 8.6	62.7 ± 6.8	0.523	0.300	0.808	0.360
BMI (kg/m2), mean ± SD	29.8 ± 5.0	28.7 ± 3.8	32.1 ± 6.4	28.8 ± 4.3	0.059	0.043	0.958	0.034
PSA (ng/mL), mean ± SD	7.8 ± 5.3	7.1 ± 5.5	9.1 ± 5.5	7.4 ± 5.0	0.115	0.043	0.395	0.174
Surgical history, n (%)a
No	85 (84.2)	28 (87.5)	24 (80.0)	33 (84.6)	0.742	0.502	1.000	0.751
Yes	16 (15.8)	4 (12.5)	6 (20.0)	6 (15.4)
Biopsy Gleason GrGp, n (%)
1	7 (6.9)	2 (6.3)	1 (3.3)	4 (10.3)	0.413	0.737	0.366	0.204
2	54 (53.5)	17 (53.1)	14 (46.7)	23 (59.0)
3	21 (20.8)	8 (25.0)	7 (23.3)	6 (15.4)
4	9 (8.9)	1 (3.1)	3 (10.0)	5 (12.8)
5	9 (8.9)	3 (9.4)	5 (16.7)	1 (2.6)
Missing	1 (1.0)	1 (3.1)	—	—
ASA score, n (%)
1 or 2	56 (55.4)	21 (65.6)	15 (50.0)	20 (51.3)	0.372	0.213	0.223	0.916
3 or 4	45 (44.6)	11 (34.4)	15 (50.0)	19 (48.7)

The bold values represent statistical significance which we defined as a two-tailed p < 0.05.

^a Only prior abdominal and pelvic surgeries were considered.

ASA = American Society of Anesthesiologists; BMI = body mass index; EP = extraperitoneal; GrGp = grade group; PSA = prostate specific antigen; RS = Retzius sparing; SD = standard deviation; TP = transperitoneal.

All perioperative outcomes evaluated are described in Table 2. Overall mean operative time was 228 ± 42 minutes and differed significantly between cohorts (p < 0.001). The RS approach had a decreased operative time as compared with the TP approach (208 minutes vs 248 minutes, p < 0.001) and the EP approach had a decreased operative time compared with TP approach (224 minutes vs 248 minutes, p = 0.015). Our 90-day readmission rate was 9% (9/101); readmission rates did not differ significantly between cohorts (p = 0.306). The overall 90-day complication rate was 10% (10/101), complication rates did not differ significantly between cohorts (p = 0.347) either. Lymphocele occurrence was significantly different between cohorts (p = 0.05)—lymphoceles occurred in 13% (4/32) RS cases vs 0 TP cases (p = 0.037).

Table 2.

Perioperative Outcomes

Perioperative outcomes	Total	RS	EP	TP	p	Pairwise comparisons
	n = 101	n = 32	n = 30	n = 39		RS vs EP	RS vs TP	EP vs TP
Operative time (minutes), mean ± SD	228 ± 42	208 ± 40	224 ± 41	248 ± 36	<0.001	0.125	<0.001	0.015
EBL (mL), mean ± SD	126 ± 69	112 ± 46	138 ± 87	130 ± 70	0.602	0.400	0.365	0.926
Prostate size (g), mean ± SD	47.3 ± 25.0	41.1 ± 13.8	48.3 ± 17.4	51.8 ± 35.1	0.162	0.061	0.196	0.562
Postoperative MME, mean ± SD	9.3 ± 14.5	7.6 ± 9.3	10.4 ± 19.3	9.9 ± 14.0	0.970	0.955	0.910	0.735
Postoperative opioid requirement, n (%)
No	33 (32.7)	11 (34.4)	6 (20.0)	16 (41.0)	0.176	0.205	0.566	0.063
Yes	68 (67.3)	21 (65.6)	24 (80.0)	23 (59.0)
Length of stay (days), n (%)
0	25 (24.7)	20 (62.5)	5 (16.7)	—	<0.001	<0.001	<0.001	0.019
1	71 (70.3)	12 (37.5)	22 (73.3)	37 (94.9)
2	5 (5.0)	—	3 (10.0)	2 (5.1)
90-day CD-II or greater complication, n (%)
No	91 (90.1)	27 (84.4)	27 (90.0)	37 (94.9)	0.347	0.709	0.231	0.646
Yes	10 (9.9)	5 (15.6)	3 (10.0)	2 (5.1)
90-day readmission, n (%)
No	92 (91.1)	27 (84.4)	28 (93.3)	37 (94.9)	0.306	0.427	0.231	1.000
Yes	9 (8.9)	5 (15.6)	2 (6.7)	2 (5.1)
Lymphocele, n (%)
No	96 (95.0)	28 (87.5)	29 (96.7)	39 (100)	0.050	0.355	0.037	0.435
Yes	5 (5.0)	4 (12.5)	1 (3.3)	—

The bold values represent statistical significance which we defined as a two-tailed p < 0.05.

CD-II = Clavien–Dindo grade II; MMEs = morphine milligram equivalents; EBL = estimated blood loss.

Oncologic outcomes are described in Table 3. Final pathologic T-Stage was significantly different between cohorts (p = 0.017). All other oncologic outcomes assessed were not significantly different between cohorts.

Table 3.

Oncologic Outcomes

Oncologic outcomes	Total	RS	EP	TP	p	Pairwise comparisons
	n = 101	n = 32	n = 30	n = 39		RS vs EP	RS vs TP	EP vs TP
Lymph node yield, mean ± SDa	4.7 ± 3.5	4.5 ± 2.6	4.5 ± 4.6	4.9 ± 3.4	0.730	0.604	0.853	0.424
Final pathologic Gleason grade group, n (%)
1	7 (6.9)	1 (3.1)	2 (6.7)	4 (10.3)	0.955	0.857	0.749	0.915
2	55 (54.5)	20 (62.5)	15 (50.0)	20 (51.3)
3	21 (20.8)	6 (18.8)	6 (20.0)	9 (23.1)
4	8 (7.9)	2 (6.3)	3 (10.0)	3 (7.7)
5	10 (9.9)	3 (9.4)	4 (13.3)	3 (7.7)
Final pathologic T-stage, n (%)
T2	65 (64.3)	28 (87.5)	16 (53.3)	21 (53.8)	0.017	0.014	0.009	0.858
T3a	23 (22.8)	3 (9.4)	8 (26.7)	12 (30.8)
T3b	13 (12.9)	1 (3.1)	6 (20)	6 (15.4)
Clinically significant positive margin, n (%)
No	73 (72.3)	22 (68.8)	22 (73.3)	29 (74.4)	0.861	0.691	0.601	0.923
Yes	28 (27.7)	10 (31.3)	8 (26.7)	10 (25.6)
PLND, n (%)
No	11 (10.9)	2 (6.3)	7 (23.3)	2 (5.1)	0.05	0.077	1.000	0.035
Yes	90 (89.1)	30 (93.8)	23 (76.7)	37 (94.9)
Biochemical recurrence, n (%)
No	84 (86.6)	25 (80.6)	26 (96.3)	33 (84.6)	0.168	0.108	0.662	0.226
Yes	13 (13.4)	6 (19.4)	1 (3.7)	6 (15.4)
Unknown	4	1	3	—
Received salvage radiation, n (%)
No	89 (88.1)	27 (84.4)	29 (96.7)	33 (84.6)	0.228	0.197	0.978	0.128
Yes	12 (11.9)	5 (15.6)	1 (3.3)	6 (15.4)
Received adjuvant radiation, n (%)
No	96 (95.0)	32 (100)	27 (90.0)	37 (94.9)	0.223	0.107	0.498	0.646
Yes	5 (5.0)	—	3 (10.0)	2 (5.1)
Time to last PSA in days, median (IQR)	188 (88–402)

The bold values represent statistical significance which we defined as a two-tailed p < 0.05.

^a Subset of patients who received PLND.

PLND = pelvic lymph node dissection.

Functional outcomes are described in Table 4 and Figure 3. Return of continence differed significantly between cohorts (p < 0.001) during the follow-up period assessed. The RS cohort had higher continence rates than the EP cohort—94% (30/32) vs 52% (15/29), respectively, p < 0.001. Continence did not differ significantly between the RS and TP cohorts (p = 0.096), though the RS cohort did have a higher continence rate than the TP cohort—94% (30/32) vs 77% (30/39), respectively.

Table 4.

Early Functional Outcomes

Early functional outcomes	Total	RS	EP	TP	p	Pairwise comparisons
	n = 101	n = 32	n = 30	n = 39		RS vs EP	RS vs TP	EP vs TP
Continence, n (%)
No	25 (25.0)	2 (6.3)	14 (48.3)	9 (23.1)	<0.001	<0.001	0.096	0.030
Yes	75 (75.0)	30 (93.8)	15 (51.7)	30 (76.9)
Unknown	1	—	1	—
Time to continence in days, mean ± SDa	82.2 ± 69.7	47.7 ± 37.2	84.0 ± 53.1	116 ± 85.1	<0.001	0.039	<0.001	0.193
Erection, n (%)
No	35 (36.5)	4 (12.5)	16 (59.3)	15 (40.5)	<0.001	<0.001	0.014	0.139
Yes	61 (63.5)	28 (87.5)	11 (40.7)	22 (59.5)
Unknown	5	—	3	2
Erection without medication, n (%)
No	61 (63.5)	13 (40.6)	21 (77.8)	27 (73.0)	0.004	0.004	0.007	0.661
Yes	35 (36.5)	19 (59.4)	6 (22.2)	10 (27.0)
Unknown	5	—	3	2
Time to erection in days, mean ± SDa	89.1 ± 80.3	47.2 ± 29.9	105 ± 99.3	134 ± 89.1	<0.001	0.005	<0.001	0.320
Time to last follow-up in days, median (IQR)	150 (88–377)

The bold values represent statistical significance which we defined as a two-tailed p < 0.05.

^a Subset of patients endorsing recovery.

FIG. 3.

Comparison of early functional outcome rates between each type of surgical approach. Patient-reported outcomes were assessed at clinic visits as part of a routine follow-up schedule. Continence was defined as requiring—pads per day. Erectile function was defined as achieving satisfactory erections adequate for sexual activity, with or without the use of medication. *p < 0.05, **p < 0.001. TP = transperitoneal.

Mean time to continence differed significantly between cohorts (p < 0.001). In the RS cohort, mean time to continence was 48 ± 37 days compared with 84 ± 53 days in the EP cohort (p = 0.039). Compared with the RS cohort, the TP cohort had a significantly longer mean time to continence—116 ± 85 days vs 48 ± 37 days; (p < 0.001). Return of erectile function differed significantly between cohorts (p = <0.001). In the RS cohort, 88% (28/32) of patients regained erectile function compared with 41% (11/27) in the EP cohort (p < 0.001).

Compared with the RS cohort, significantly fewer patients in the TP cohort regained erectile function as well—60% (22/37) vs 88% (28/32); p = 0.014. Mean time to erection was also significantly different between cohorts (p < 0.001). In the RS cohort, mean time to erection was 47 ± 30 days compared with 105 ± 99 days in the EP cohort (p = 0.005). Compared with the RS cohort, the TP cohort had a significantly longer mean time to erection—134 ± 89 vs 47 ± 30 days (p < 0.001).

Discussion

In this retrospective cohort study, we present our institution's initial experience with SP-RARP performed through three approaches—RS, TP, and EP. To our knowledge, this study represents the largest characterization of RS SP-RARP. We show that these three approaches to SP-RARP are safe and feasible, with similarity in terms of perioperative outcomes, oncologic outcomes, and postoperative pain control. Faster and improved returns of both continence and erection were associated with RS SP-RARP.

SP-RARP has been demonstrated to have equivalent oncologic outcomes to MP-RARP.¹⁷ The benefits of the RS approach over the non-RS retropubic approaches in terms of return of continence after MP-RARP are well described in the literature.^8–10 We establish that a similar benefit is observed in SP-RARP when compared with a retropubic EP approach.

We did observe that the RS cohort had a higher continence rate than TP cohort, however, this difference was not significant. Basset et al describe RS SP-RARP methodology and report a 90-day continence rate of 89%, which is consistent with our results.¹¹ An additional benefit of the RS approach over TP and EP approaches, as demonstrated by studies of MP-RARP, is the improved recovery of erectile function.^9,18 Our findings suggest that this benefit may be replicated with the SP platform.

We observed similarity in baseline patient characteristics including age, BMI, initial PSA, history of abdominal or pelvic surgery, and biopsy Gleason grade group for all three approaches. However, we found that BMI was observed to be higher in patients undergoing EP SP-RARP. The mean BMI in the EP category was >30 kg/m², whereas the mean BMIs for the TP and the RS cohorts were both <30 kg/m². This is likely because of preferential utilization of the EP approach in obese patients. The EP approach allows for less steep Trendelenburg positioning, which has been demonstrated to increase respiratory stress during robotic pelvic surgery.¹⁹ In addition, vesicourethral anastomosis has been found to be more difficult in morbidly obese patients in TP MP-RARP.²⁰

Obesity has been previously identified as a predictor of prolonged MP-RARP operative times.²¹ Despite the EP cohort having a higher proportion of obese patients than the TP cohort, we observed a decreased operative time in the EP cohort. It is likely that that the combination of an EP approach and the SP platform mitigates the previously observed effects of patient body habitus on RARP operative times. Future research directly comparing approaches and platforms in obese patients is necessary. The operative time for RS was also significantly lower than that of TP.

Factors that may inform operative time such as patient history of previous abdominal or pelvic surgery, prostate size, and performance of PLND all did not differ significantly between cohorts; this suggests that decreased operative time is likely a result of the RS approach itself as the RS approach does not involve dropping the bladder. Our RS SP-RARP operative time is consistent with operative times reported in the literature.¹¹ Future randomized studies of the SP-RARP approaches and potential multicenter studies are necessary.

Although overall 90-day complication rates did not differ significantly between cohorts, the overall postsurgical lymphocele rate did. Lymphoceles were observed in a total of five SP-RARPs. Four were performed using the RS approach and one was performed using the EP approach. All four RS SP-RARPs that resulted in lymphocele were performed before June 2020. Our group has previously indicated that the RS approach is a factor significantly associated with development of symptomatic lymphocele.¹² Upon our realization of this association, we modified our approach to PLND by utilizing peritoneal tacking in a manner that allows PLND beds to remain open to the peritoneal space.

Limitations of this study pertain to its retrospective design and sample size. Bias was introduced because of surgeons selecting the approach for each case without randomization. Although surgeon-specific variation is a limitation of this study, both treating surgeons were fellowship trained and at the beginning of the SP procedural learning curve that may lessen the extent of this variation. Utilization of patient-reported functional outcomes may represent a source of bias in this study, and our definition of continence and erectile function as binary outcomes may limit this. Another limitation is the relatively short-term follow-up for oncologic and functional outcomes.

Conclusions

We present an effective adoption of the SP platform for performing RARP through three surgical approaches—RS, TP, and EP. Baseline patient characteristics, perioperative outcomes, and oncologic outcomes were largely similar between groups, indicating that the three approaches employed for SP-RARP are safe and feasible. Improved and faster early functional outcomes were achieved with RS SP-RARP. Ultimately, this highlights that the previously reported benefits of the RS approach on MP platforms are preserved when performed through SP. Differences between approaches for length of stay and operative time may be related to technical factors or patient selection.

Abbreviations Used

ASA

American Society of Anesthesiologists

BMI

body mass index

EBL

estimated blood loss

EP

extraperitoneal

MMEs

morphine milligram equivalents

MP

multiport

PLND

pelvic lymph node dissection

PSA

prostate specific antigen

RARP

robot-assisted radical prostatectomy

RS

Retzius sparing

SP

single port

TP

transperitoneal

Authors' Contributions

S.B., A.S., and E.H.K. provided substantial contributions to design, data acquisition, data analysis, interpretation of the data, drafting, revising, and final approval of this study, and agree to be accountable for all aspects of the study. J.M.V. provided substantial contributions to design, data analysis, drafting, revising, and final approval of this study and agrees to be accountable for all aspects of the study. G.M.H. and R.S.F. provided substantial contributions to design, interpretation of the data, drafting, revising, and final approval of this study and agree to be accountable for all aspects of the study.

Ethics Approval

This study was approved by Washington University IRB.

Consent for Publication

All authors have provided consent for publication.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was supported by the National Center for Advancing Translational Sciences at the National Institutes of Health under Grant Nos. UL1 TR002345, KL2 TR002346, and TL1 TR002344.