Assessing the Impact of Surgeon Experience on Urinary Continence Recovery After Robot-Assisted Radical Prostatectomy: Results of Four High-Volume Surgeons

Abstract

Purpose:

To test the impact of surgeon experience on urinary continence (UC) recovery after robot-assisted radical prostatectomy (RARP).

Materials and Methods:

The study included 1477 consecutive patients treated with RARP by four surgeons between 2006 and 2014. UC recovery was defined as being completely dry over a 24-hour period at follow-up. Surgeon experience was coded as the total number of RARP performed by the surgeon before the patient's operation. Multivariable analysis tested the association between surgeon experience and UC recovery. Covariates consisted of patient age, Charlson comorbidity index, preoperative International Index of Erectile Function-Erectile Function domain (IIEF-EF), nerve-sparing surgery (none vs unilateral vs bilateral), and preoperative risk groups (low- vs intermediate- vs high risk).

Results:

The number of cases performed by each surgeon was 541, 413, 411, and 112, respectively. Median follow-up was 24 months (inter-quartile range: 18, 40). The UC recovery rate at 1 year after surgery was 82%. At multivariable analyses, surgeon experience represented an independent predictor of UC recovery (hazard ratio: 1.02, p < 0.001). The surgical learning curve was similar among surgeons, moving linearly from ∼60% of UC rate at the initial cases to almost 90% after more than 400 procedures.

Conclusions:

In patients undergoing RARP, surgeon experience is a significant predictor of UC recovery. The surgical learning curve of UC recovery does not reach a plateau even after more than 100 cases, suggesting a continuous improvement of the surgical technique. These findings deserve attention for patient counseling and future comparative studies evaluating functional outcomes after RARP.

Introduction

Radical prostatectomy represents one of the recommended first-line treatment options for patients with clinically localized prostate cancer and a life expectancy of at least 10 years.^1,2 However, this surgical treatment is associated with a non-negligible rate of long-term side effects. Among these, urinary incontinence and erectile dysfunction are the most bothersome, with a detrimental effect on health-related quality of life.^3,4

After the introduction of robot-assisted radical prostatectomy (RARP), several studies reported better functional outcomes compared with other surgical routes.^4

–8 Urinary continence (UC) recovery depends on different factors such as patient age, body mass index (BMI), nerve-sparing technique, and patient comorbidities.^9
–11 However, no previous studies evaluated the impact of surgeon experience on postoperative UC recovery. The number of previous operations may improve the surgeon's skills with a robotic system, leading to better surgical outcomes.

The correlation between surgeon experience and UC recovery after RARP would be important for patient counseling, quality control of surgical outcomes, and future comparative studies evaluating functional outcomes after RARP. In this study, we aimed at assessing the surgical learning curve of RARP in terms of UC recovery. We hypothesized that surgeon experience was significantly associated with UC recovery after RARP.

Materials and Methods

Study population

The study included 1477 consecutive patients treated with RARP by four surgeons at a single tertiary referral center between 2006 and 2014. All patients included in the study had complete pre-operative, pathological, functional, and oncological data that were prospectively collected.

Surgical technique

The vesico-urethral anastomosis was usually performed with posterior reconstruction.¹² Specifically, posterior reconstruction was performed by using two 3-0 Monocryl sutures, tied together at the free ends. The first suture was run from right to left, approximating Denonvilliers' fascia beneath the bladder proximally to the posterior fascia near the urethral stump distally and tied to itself. The second suture was then passed proximally to the posterior bladder neck in the midline and distally to the posterior fascia adjacent to the urethral stump caudal to the previous row of sutures. Finally, the vesico-urethral anastomosis was performed with the single-knot method.¹³

Indication for neurovascular bundle preservation was given according to the judgment of each treating surgeon. When carried out, the nerve-sparing technique was performed as previously described.^14,15 In particular, athermal early retrograde release of the neurovascular bundle was performed.¹⁶

Lymph node (LN) dissection was performed when the risk of LN invasion exceeded 5% according to the Briganti nomogram.^17,18 When performed, pelvic LN dissection was extended,¹⁹ consisting of the excision of fibro-fatty tissue along the external iliac vein, the distal limit being the deep circumflex vein, and the femoral canal. Proximally, extended pelvic lymph node dissection was performed up to, and including, the bifurcation of the common iliac artery. Further, all fibrofatty tissue within the obturator fossa was removed to completely skeletonize the obturator nerve. The lateral limit consisted of the pelvic sidewall, and the medial dissection limit was defined by peri-vesical fat. LNs along the internal iliac vessels were invariably dissected. In some cases, LNs located in the pre-sacral and common iliac areas were also removed. For each patient, six separate LN packets were sent, one for each of the following anatomical regions: the obturator (right and left), the external iliac (right and left), and the internal iliac (right and left) area. Additional packets were sent when the pre-sacral and/or common iliac nodes were dissected.

Variables definition

Surgeon experience was coded as the total number of RARP performed by the surgeon before the patient's operation.^20
–22 Patient characteristics included age at surgery, BMI, preoperative International Index of Erectile Function-Erectile Function domain (IIEF-EF), and Charlson comorbidity index (CCI).²³ Clinical characteristics of the tumor consisted of pre-operative prostate-specific antigen level, clinical T stage (T1 vs T2 vs T3), and biopsy Gleason score (≤6 vs 7 vs ≥8). Pre-operative risk group classification (low- vs intermediate- vs high risk) was also reported.^1,2 Pathological characteristics of the tumor consisted of pathological stage (pT2 vs pT3a vs pT3b/pT4), pathological Gleason score (≤6 vs 7 vs ≥8), LN status (pNx/pN0 vs pN1), and surgical margins status (negative vs positive).

Outcomes

The study outcome consisted of UC recovery at 1 year after surgery that was defined as being completely dry over a 24-hour period. Continence rates were assessed by patient-reported pad usage over 24 hours. Patients were followed up at 1, 3, 6, and 12 months postoperatively and every 6 months thereafter.

Statistical analyses

Statistical analyses consisted of two main steps. First, multivariable Cox-regression analysis tested the association between surgeon experience and UC recovery. Covariates consisted of patient age, CCI, pre-operative IIEF-EF, nerve-sparing approach (none vs unilateral vs bilateral), and pre-operative risk groups (low- vs intermediate- vs high risk).

Second, the probability of UC recovery at 1 year after RARP was calculated based on the multivariable model. Such probability was plotted against surgeon experience to obtain the surgical learning curve of each surgeon.

All statistical analyses were performed by using Stata (StataCorp. LP, College Station, TX), version 12.0.

Results

Patient characteristics

The number of cases performed by the four surgeons was 541, 413, 411, and 112, respectively. Descriptive characteristics of the population are illustrated in Table 1. Comorbidity profile was not significantly different among the four surgeons (p = 0.3), as well as the rate of bilateral nerve-sparing procedure (p = 0.12) (Table 1A). Overall, 43% and 48% of the cohort were classified as a low- and intermediate risk, whereas 4% of the patients were high risk (Table 1B). At final pathology, 76% of the cohort harbored T2 disease. A Gleason pattern equal to six was found in 35% of the cohort, whereas 57% had a Gleason score of 7, and 8% harbored a Gleason score ≥8. Nodal metastases were found in 5% of the patients, whereas positive surgical margins were observed in 17% of the cohort (Table 1C).

Table 1.

Descriptive Characteristics of 1477 Consecutive Patients Treated with Robot-Assisted Radical Prostatectomy by Four Different Surgeons Between 2006 and 2014

Variable	Overall population (n = 1477)	Surgeon #1 (n = 541)	Surgeon #2 (n = 413)	Surgeon #3 (n = 411)	Surgeon #4 (n = 112)	p
A. Patient and intervention characteristics
Age at surgery, year	63 (58, 68)	63 (58, 68)	63 (58, 68)	62 (57, 67)	64 (60, 68)	0.036
Pre-op IIEF-EF	25 (15, 29)	26 (14, 29)	25 (15, 28)	25 (19, 29)	25 (14, 29)	0.7
CCI						0.3
0, n (%)	1442 (86)	477 (88)	350 (85)	352 (86)	89 (79)
1, n (%)	199 (12)	50 (9)	50 (12)	47 (11)	20 (18)
≥2, n (%)	45 (3)	14 (3)	13 (3)	12 (3)	3 (3)
Nerve sparing						0.12
No, n (%)	191 (11)	62 (11)	45 (11)	45 (11)	19 (17)
Mono-lateral, n (%)	144 (9)	38 (7)	32 (8)	47 (11)	8 (7)
Bilateral, n (%)	1351 (80)	441 (82)	336 (81)	319 (78)	85 (76)
B. Clinical characteristics of the tumor
Total PSA, ng/mL	6.0 (4.6, 8.3)	6.0 (4.7, 8.6)	6.1 (4.8, 8.0)	5.8 (4.5, 8.0)	6.1 (4.7, 8.6)	0.4
Clinical stage						<0.0001
T1, n (%)	1011 (68)	335 (62)	305 (74)	299 (73)	72 (64)
T2, n (%)	420 (28)	173 (32)	106 (26)	106 (26)	35 (31)
T3, n (%)	46 (3)	33 (6)	2 (0)	6 (1)	5 (4)
Biopsy Gleason						<0.0001
≤6, n (%)	960 (65)	300 (55)	302 (73)	296 (72)	62 (55)
7, n (%)	448 (30)	194 (36)	104 (25)	104 (25)	46 (41)
≥8, n (%)	69 (5)	47 (9)	7 (2)	11 (3)	4 (4)
Risk group						<0.0001
Low, n (%)	654 (44)	204 (38)	213 (52)	198 (48)	39 (35)
Intermediate, n (%)	685 (46)	251 (46)	183 (44)	191 (46)	60 (54)
High, n (%)	138 (9)	86 (16)	17 (4)	22 (5)	13 (12)
C. Pathologic characteristics
Pathologic stage						<0.0001
pT2, n (%)	1129 (76)	370 (68)	338 (82)	336 (82)	85 (76)
pT3a, n (%)	257 (17)	127 (23)	52 (13)	56 (14)	22 (20)
pT3b/pT4, n (%)	91 (6)	44 (8)	23 (6)	19 (5)	5 (4)
Pathologic Gleason						<0.0001
≤6, n (%)	528 (36)	149 (28)	182 (44)	168 (41)	29 (26)
7, n (%)	833 (56)	323 (60)	211 (51)	227 (55)	72 (64)
≥8, n (%)	116 (8)	69 (13)	20 (5)	16 (4)	11 (10)
Lymph node status						<0.0001
pN0/pNx, n (%)	1404 (95)	495 (91)	398 (96)	403 (98)	108 (96)
pN1, n (%)	73 (5)	46 (9)	15 (4)	8 (2)	4 (4)
Surgical margins						0.4
Negative, n (%)	1241 (84)	445 (82)	347 (84)	350 (85)	99 (88)
Positive, n (%)	236 (16)	96 (18)	66 (16)	61 (15)	13 (12)

All numbers are medians (inter-quartile range) or frequencies (proportions).

CCI = Charlson comorbidity index; IIEF-EF = International Index of Erectile Function-Erectile Function domain; PSA = prostate-specific antigen.

Median follow-up for patients who did not recover UC after RARP was 24 months (inter-quartile range: 18, 40). The UC recovery rate at 1 year after surgery was 82%. At multivariable analysis, surgeon experience represented an independent predictor of UC recovery (hazard ratio [HR] per 10 procedures: 1.02; 95% confidence interval: 1.01, 1.02; p < 0.001) (Table 2). The surgical learning curve of RARP in terms of UC recovery is represented in Figure 1. Specifically, after accounting for confounders, a patient operated by a surgeon who performed <100 RARP has a probability of being pad free at 12 months of ∼70%. Conversely, the same patient operated by a surgeon with a previous experience of 500 RARP has a probability of being pad free at 12 months of 90%. When stratified according to the surgeon, the surgical learning curves were similar, moving linearly from ∼60% of UC rate at the initial cases to almost 90% after more than 400 procedures (Fig. 2). The annual caseload of the surgeons over the study period is reported in Table 3. The relatively low annual caseload was mainly due to the low number of cases performed in the first years at the beginning of the learning curve. Moreover, this study included only patients in whom the full operation was performed by one of the four surgeons, whereas cases in which residents or fellows performed few steps of the operation were excluded.

FIG. 1.

The surgical learning curve for RARP evaluating the probability of UC recovery. RARP = robot-assisted radical prostatectomy; UC = urinary continence.

FIG. 2.

The surgical learning curve for RARP evaluating the probability of UC recovery by different surgeons.

Table 2.

Multivariable Cox Regression Analysis Predicting Urinary Continence Recovery After Robot-Assisted Radical Prostatectomy in 1477 Patients Treated Between 2006 and 2014

Predictor	HR	95% CI	p
Age at surgery	0.98	0.97, 1.00	0.018
Pre-operative IIEF-EF	1.00	1.00, 1.01	0.2
CCI	1.02	1.01, 1.02	<0.0001
Nerve-sparing technique
Not performed	1.00	Ref.	—
Monolateral	1.71	1.03, 2.84	0.038
Bilateral	1.59	1.04, 2.41	0.030
Risk group
Low	1.00	Ref.	—
Intermediate	0.93	0.75, 1.15	0.5
High	0.75	0.47, 1.20	0.2
Surgeon experience, per 10 procedures	1.02	1.01, 1.02	<0.0001

CI = confidence interval; HR = hazard ratio.

Table 3.

Annual Caseload of the Four Surgeons Over the Study Period

Year of surgery	Surgeon #1	Surgeon #2	Surgeon #3	Surgeon #4	Total
2006	0	0	18	0	18
2007	0	0	37	0	37
2008	18	21	62	0	101
2009	45	57	57	1	160
2010	74	99	53	5	231
2011	102	114	74	22	312
2012	80	108	75	25	288
2013	108	14	31	51	204
2014	114	0	4	8	126
Total	541	413	411	112	1477

Discussion

In prostate cancer patients treated with RARP, UC recovery represents one of the main challenging outcomes significantly affecting patients' quality of life after surgery. It is well known that UC recovery depends on a multitude of factors related both to the patient's characteristics and to the surgical approach.^3,4 For this reason, over the years, several techniques have been suggested, aiming at optimizing this aspect.^10,24
–26 Moreover, the outcomes of radical prostatectomy have been shown to depend on surgical techniques and surgical volume, but no clear data on the correlation between surgical experience and UC recovery are available to date. Therefore, we hypothesized that the surgeon's experience might affect time to full UC recovery after surgery. To address this issue, we relied on prospectively collected data of patients treated with RARP by four different surgeons in a single referral center, evaluating time to UC recovery according to the progressive number of surgical procedures for each surgeon.

Our results clearly show that surgical experience represents an independent predictor of UC recovery at 12 months of follow-up (HR per 10 procedures: 1.02; p < 0.001). Moreover, no plateau on UC outcomes was reached even after an extremely high number of procedures performed by a single surgeon. Indeed, the per-surgeon sub-analysis showed that even for the most experienced surgeons who completed more than 200 RARPs there is a constant, progressive improvement toward improved continence outcomes. Some authors hypothesized the need of a learning curve of more than 100 cases before obtaining a plateau on early UC.²⁷ On the contrary, Thompson et al. showed progressive improvement in UC recovery according to the number of surgeries performed.²⁸ However, their study compared functional and pathological outcomes between open radical prostatectomy (230 patients) and RARPs (379 patients) focusing on a single surgeon experience, whereas our study included the analyses of the results of four different surgeons, and it is interesting to see that the results are similar among all of them.

The possible explanation of our results might be related to the influence of the surgical technique on UC recovery. For example, it has been shown that both the adoption of a nerve-sparing procedure and the grade of nerve sparing are significantly related to UC recovery.^10,29

Our results are important, since the application of such findings may be used to evaluate the results of new surgical techniques. For example, the results of a recently published randomized clinical trial reported no statistically significant differences in domain-specific quality of life parameters (including UC) in patients treated with RRP relative to their counterparts treated with radical retropubic prostatectomy.³⁰ Despite the necessity for the authors to be commended for their work, it is noteworthy to underline that at the beginning of the trial the RRP surgeon completed more than 1500 procedures whereas the RARP surgeon completed only 200 RARPs. According to our findings that provide evidence that UC recovery after 200 robotic cases is roughly 75%, consequently, RARP's UC outcomes could be underestimated in this trial.

On the other hand, our study has several limitations. First, it has recently been shown that there is an improvement in functional outcomes even beyond the 24-month follow-up, and, therefore, it could be argued that our results might not be complete. However, short-term UC recovery represents an important outcome for patients' quality of life and, therefore, the results of our analyses should be taken into account when proposing RARP as a therapeutic approach. Second, although this is a multi-surgeon study, it was performed at a single high-volume institution, and the presence of younger, less experienced surgeons may itself limit our findings and their generalizability. Further, the single-institution nature of the study may influence its result, as all the surgeons used the same technique and the same postoperative physical therapy. Third, the definition of UC recovery as being completely dry over a 24-hour period at follow-up has some limitations, because some patients may be dry for more than 24 hours and still be incontinent as they lose urine the next day, depending on position, liquid intake, and physical activity. Finally, the retrospective nature of this observational report, even if corrected for confounders, may limit our results.

Despite all these limitations, our study represents one of the largest single-center series specifically addressing the association between surgical experience and UC recovery after RARP by using a proper methodological approach. This study proved an increasing improvement on UC recovery rates even for most experienced urologists. At the same time, we believe that this should push young robotic urologists toward improving their proficiency by structured training to limit the impact of the learning curve on patients' quality of life.

Conclusions

In patients undergoing RARP, surgeon experience represents a significant predictor of UC recovery. The surgical learning curve of UC recovery does not reach a plateau even after more than 200 cases, suggesting a continuous improvement in surgical technique. These findings deserve attention for patient counseling and for the design of comparative studies evaluating functional outcomes after RARP. Finally, the introduction of structured training programs may reduce the learning curve of RARP.

Footnotes

Acknowledgment

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Mottet

, Bellmunt

, Bolla

, et al. EAU-ESTRO-SIOG guidelines on prostate cancer. Part 1: Screening, diagnosis, and local treatment with curative intent. Eur Urol, 2017; 71:618–629.

Mohler

, Armstrong

, Bahnson

, et al. Prostate cancer, Version 1.2016. J Natl Compr Canc Netw, 2016; 14:19–30.

Resnick

, Koyama

, Fan

K-H

, et al. Long-term functional outcomes after treatment for localized prostate cancer. N Engl J Med, 2013; 368:436–445.

Ficarra

, Novara

, Rosen

, et al. Systematic review and meta-analysis of studies reporting urinary continence recovery after robot-assisted radical prostatectomy. Eur Urol, 2012; 62:405–417.

Geraerts

, Van Poppel

, Devoogdt

, et al. Prospective evaluation of urinary incontinence, voiding symptoms and quality of life after open and robot-assisted radical prostatectomy. BJU Int, 2013; 112:936–943.

Tewari

, Srivasatava

, Menon

, et al. A prospective comparison of radical retropubic and robot-assisted prostatectomy: Experience in one institution. BJU Int, 2003; 92:205–210.

Kim

, Song

, Kim

, et al. Factors determining functional outcomes after radical prostatectomy: Robot-assisted versus retropubic. Eur Urol, 2011; 60:413–419.

Ficarra

, Novara

, Ahlering

, et al. Systematic review and meta-analysis of studies reporting potency rates after robot-assisted radical prostatectomy. Eur Urol, 2012; 62:418–430.

Kumar

, Samavedi

, Bates

, et al. Continence outcomes of robot-assisted radical prostatectomy in patients with adverse urinary continence risk factors. BJU Int, 2015; 116:764–770.

10.

Suardi

, Moschini

, Gallina

, et al. Nerve-sparing approach during radical prostatectomy is strongly associated with the rate of postoperative urinary continence recovery. BJU Int, 2013; 111:717–722.

11.

Reeves

, Preece

, Kapoor

, et al. Preservation of the neurovascular bundles is associated with improved time to continence after radical prostatectomy but not long-term continence rates: Results of a systematic review and meta-analysis. Eur Urol, 2015; 68:692–704.

12.

Brien

, Barone

, Fabrizio

, et al. Posterior reconstruction before vesicourethral anastomosis in patients undergoing robot-assisted laparoscopic prostatectomy leads to earlier return to baseline continence. J Endourol, 2011; 25:441–445.

13.

Van Velthoven

, Ahlering

, Peltier

, et al. Technique for laparoscopic running urethrovesical anastomosis: The single knot method. Urology, 2003; 61:699–702.

14.

Montorsi

, Salonia

, Suardi

, et al. Improving the preservation of the urethral sphincter and neurovascular bundles during open radical retropubic prostatectomy. Eur Urol, 2005; 48:938–945.

15.

Montorsi

, Wilson

, Rosen

, et al. Best practices in robot-assisted radical prostatectomy: Recommendations of the Pasadena Consensus Panel. Eur Urol, 2012; 62:368–381.

16.

Coughlin

, Dangle

, Palmer

, et al. Athermal early retrograde release of the neurovascular bundle during nerve-sparing robotic-assisted laparoscopic radical prostatectomy. J Robot Surg, 2009; 3:13–17.

17.

Briganti

, Chun

FKH

, Salonia

, et al. Validation of a nomogram predicting the probability of lymph node invasion among patients undergoing radical prostatectomy and an extended pelvic lymphadenectomy. Eur Urol, 2006; 49:1019–1027.

18.

Briganti

, Larcher

, Abdollah

, et al. Updated nomogram predicting lymph node invasion in patients with prostate cancer undergoing extended pelvic lymph node dissection: The essential importance of percentage of positive cores. Eur Urol, 2012; 61:480–487.

19.

Abdollah

, Suardi

, Gallina

, et al. Extended pelvic lymph node dissection in prostate cancer: A 20-year audit in a single center. Ann Oncol, 2013; 24:1459–1466.

20.

Vickers

, Bianco

, Serio

, et al. The surgical learning curve for prostate cancer control after radical prostatectomy. J Natl Cancer Inst, 2007; 99:1171–1177.

21.

Vickers

, Savage

, Hruza

, et al. The surgical learning curve for laparoscopic radical prostatectomy: A retrospective cohort study. Lancet Oncol, 2009; 10:475–480.

22.

Fossati

, Barbagli

, Larcher

, et al. The surgical learning curve for one-stage anterior urethroplasty: A prospective single-surgeon study. Eur Urol, 2016; 69:686–690.

23.

Charlson

, Pompei

, Ales

, et al. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis, 1987; 40:373–383.

24.

Gandaglia

, Albersen

, Suardi

, et al. Postoperative phosphodiesterase type 5 inhibitor administration increases the rate of urinary continence recovery after bilateral nerve-sparing radical prostatectomy. Int J Urol, 2013; 20:413–419.

25.

Rocco

, Gregori

, Stener

, et al. Posterior reconstruction of the rhabdosphincter allows a rapid recovery of continence after transperitoneal videolaparoscopic radical prostatectomy. Eur Urol, 2007; 51:996–1003.

26.

Heesakkers

, Farag

, Bauer

, et al. Pathophysiology and contributing factors in postprostatectomy incontinence: A review. Eur Urol, 2017; 71:936–944.

27.

Good

, Stewart

, Laird

, et al. A critical analysis of the learning curve and postlearning curve outcomes of two experience- and volume-matched surgeons for laparoscopic and robot-assisted radical prostatectomy. J Endourol, 2015; 29:939–947.

28.

Thompson

, Egger

, Böhm

, et al. Superior quality of life and improved surgical margins are achievable with robotic radical prostatectomy after a long learning curve: A prospective single-surgeon study of 1552 consecutive cases. Eur Urol, 2014; 65:521–531.

29.

Carlsson

, Jäderling

, Wallerstedt

, et al. Oncological and functional outcomes 1 year after radical prostatectomy for very-low-risk prostate cancer: Results from the prospective LAPPRO trial. BJU Int, 2016; 118:205–212.

30.

Yaxley

, Coughlin

, Chambers

, et al. Robot-assisted laparoscopic prostatectomy versus open radical retropubic prostatectomy: Early outcomes from a randomised controlled phase 3 study. Lancet, 2016; 388:1057–1086.