Evaluating the Learning Curve of Experienced Laparoscopic Surgeons in Robot-Assisted Radical Prostatectomy

Abstract

Background and Purpose:

Radical prostatectomy is the gold standard surgical treatment for organ-confined prostate cancer. There is no consensus on the impact of previous laparoscopic experience on the learning curve of robot-assisted laparoscopic prostatectomy (RALP). We compared the perioperative complications and early patient outcomes from our initial 100 cases of RALP with laparoscopic prostatectomy (LRP) cases performed well beyond the learning curve.

Patients and Methods:

Between July 2011 and January 2012, 110 RALP were performed by one of two surgeons, each with previous experience of more than 1000 LRP. The cases were pair matched from among the last 208 patients who had undergone LRP by the same surgeons at the same time. The clinical parameters, operative details, postoperative complications, and short-term outcomes from these patients, collected prospectively, were compared between the two groups.

Results:

The prostate-specific antigen (PSA) level and age of the two groups was similar. The operative time (128.4 vs 153.9 min; P=0.01) and blood loss (200 vs 254 mL; P=0.01) was significantly less for the LRP group, but the duration of catheterization was similar (5.89 vs 6.2 days). The complication rate was low. No procedures needed conversions, and no patient had a visceral injury or blood transfusion. Twenty-three patients in the LRP group and 33 patients in the RALP group had extraprostatic disease, and the positive margin rate was 14% and 19% for these respective groups. At 3 months, PSA level was undetectable in 94% of LRP and 92% RALP patients, while 56% and 65% (P=0.062) patients in these groups were using 0 to 2 pads per day.

Conclusions:

The initial results of the outcome of RALP are at least at par with those of LRP and with those of previously published RALP series. This suggests the lack of a steep learning curve for experienced laparoscopic surgeons in performing RALP.

Introduction

P rostate cancer is the most commonly detected cancer in men and the second leading cause of male cancer deaths in the United States and Europe.^1,2 With improvements in awareness and in screening and detection, it is increasingly diagnosed at an earlier stage, in younger, healthier men.³ Because the stage of the disease at diagnosis is the most important prognostic factor, this earlier diagnosis has led to a fall in prostate cancer-related deaths in recent years.²

There is no consensus on the best treatment for organ-confined prostate cancer. Management options include active surveillance, radical prostatectomy (RP), external beam radiotherapy, brachytherapy, and newer focal therapies such as cryoablation and high-intensity focused ultrasound, the oncogenic effectiveness of which remains uncertain.⁴ RP is the accepted surgical treatment for localized prostate cancer. The procedure was first described by Walsh⁵ in 1980 and was refined over the next two decades.

Laparoscopic radical prostatectomy (LRP) was first described by Schuessler and associates⁶ who concluded that it offered no advantage with regard to continence, potency, length of stay, or convalescence compared with open prostatectomy. It was Guillonneau and Vallancien⁷ who first demonstrated that the procedure could be performed safely and efficaciously with the advantage of decreased postoperative patient convalescence over the open procedure. Similar results were reported by others, which established LRP as the standard operative procedure in many institutions.^8
–10 With increasing awareness of the anatomy of the bladder neck and urethra, numerous refinements have been presented by various groups to better achieve the “trifecta” outcomes after RP.^11
–13

More recently, with the introduction of the da Vinci Surgical System (Intuitive Surgical, Inc, Sunnyvale, CA), robotic-assisted radical prostatectomy (RARP) has gained popularity since first described more than a decade ago. This has the advantages of LRP with perceived ergonomic and technical advantages, but it is comes at a premium cost.¹⁴ There are no randomized trials comparing LRP with robot-assisted laparoscopic prostatectomy (RALP). While the uptake of RALP has been rapid,¹⁵ there is no consensus on the learning curve for the procedure.

At our institution, LRP was initiated in 2000, and by January 2011, more than 3000 cases were performed. We have previously reported our learning curve with developing a modular training program and have published data for oncologic and functional outcomes as the operative technique has evolved.^16,17 We acquired a da Vinci Si robotic system in July 2011, and since then, more than 100 procedures have been performed by two surgeons, who operate independently. During this period, we have continued to perform LRP, of which more than 200 procedures have been performed. To our knowledge, our institution had more experience in conventional laparoscopy than any other institution before commencing robotic surgery. We evaluate whether our significant previous experience has resulted in an accelerated learning curve.

Patients and Methods

Between July 2011 and January 2012, a total of 110 RALP were performed by two surgeons on a single daVinci Si robot system at the University of Leipzig. Data were collected prospectively, including patient parameters, surgical and pathologic results, and perioperative complications.

We excluded the first 10 cases because they were performed under supervision of an experienced robotic surgeon.

Participating surgeons

Two surgeons used the daVinci system to perform RALP in these 100 patients. Each of these surgeons had performed more than 1000 LRPs. Neither underwent a mentored fellowship program, but attended a formal training workshop.

Patient selection

The first 10 RALP cases were performed under the supervision of an experienced mentor and were excluded. The next consecutive 100 patients were included in this study. They were pair matched with equivalent patients from among the last 208 cases who underwent LRP. The criteria were age (74% perfect matching, only differences until 3 years) and type of nerve-sparing procedure (1. bilateral, 2. unilateral, 3. nonnerve-sparing surgery). Patient matching was conducted independently by a trial statistician from the University of Leipzig.

Statistical analysis

Patients were characterized by means (ranges) and counts (percents), respectively. The intervention groups were compared with regard to continuous variables by t test/Mann-Whitney U test. Proportions were compared by chi-square test/Fisher exact test, where appropriate. A two-tailed significance level of 5% was defined. All analyses were performed by IBM SPSS Statistics, version 16.0.2.

Data collection

Data were collected prospectively. The recorded parameters included patient demographics including age and body mass index (BMI), preoperative parameters (prostate-specific antigen (PSA) level, biopsy Gleason score, clinical stage), pathologic parameters (prostate size, prostatectomy Gleason score, pathologic stage, margin status, and location of positive margins), and details of the operation including operative times, estimated blood loss, complications, including blood transfusions. Morbidities and complications during hospital stay, the durations of catheterization and patient stay, and the margin status were recorded on discharge. All complications were graded according to Clavien classification.

Three months after surgery, a questionnaire was sent to all patients to evaluate continence and potency. Continence was assessed by number of pads over a 24-hour period and potency by International Index of Erectile Function questionnaire.

Surgical technique

An extraperitoneal, intrafascial approach was used for all patients, unless they were high risk, in which case a transperitoneal approach was used. The transperitoneal approach facilitates extended lymph node dissection by enabling better access to the proximal iliac lymph nodes and has a lower rate of lymphoceles. This was performed in patients with a PSA>20 ng/mL.

A nerve-sparing procedure was performed for all preoperatively potent patients with a PSA of 10 ng/mL or less, Gleason score≤7, or when the patient specifically demonstrated a desire for nerve preservation, provided the prostate margin was negative at the neurovascular bundle. A lymphadenectomy was performed for all patients with a PSA>10 ng/mL or a Gleason score≥7.

All samples were sent for frozen section, and a positive surgical margin was defined as the presence of tumor at the inked surface of the specimen. A further resection was performed if necessitated by the frozen section result. The final pathology report was subsequently noted and recorded into the patient database.

Results

Preoperative data

Patients numbers 11 to 110 who underwent RALP were match-paired to 100 patients from among the 208 who underwent LRP between October 2010 and November 2011. Patient characteristics for the LRP and RALP groups are compared in Table 1.

Table 1.

Preoperative Patient Characteristics

	LRP (n=100)	RALP (n=100)
Mean age (range, SD)	61.33 (45–76, 7.4)	61.21 (45–77, 7.7)
PSA (range, SD) ng/mL	10.7 (1.2–80, 11.49)	8.75 (0.29–53, 7.1)
Nerve sparing
Unilateral	16	16
Bilateral	45	45
Lymphadenectomy	58	50

LRP=laparoscopic radical prostatectomy; RALP=robot-assisted laparoscopic prostatectomy; SD=standard deviation; PSA=prostate-specific antigen.

Intraoperative and early postoperative data

The intraoperative and postoperative data can be seen in Table 2. The mean operative time for LRP and RALP was 128.4 and 153.95 minutes, respectively (P=0.001), while the mean blood loss was 200 and 254 mL, respectively (P=0.01). The duration of catheterization for the two groups was similar, and no significant intraoperative complications were noted. Only one patient from each group had a wound infection that was managed medically (Clavien grade II).

Table 2.

Intraoperative and Early Postoperative Data

	LRP (n=100)	RALP (n=100)	P value
Mean operative time in minutes (range, SD)	128.4 (70–240, 35.4)	153.95 (80–270, 38.5)	0.001
Mean blood loss in mL (range, SD)	200 (50–900, 125)	254 (50–650, 126.9)	0.01
Transfusion rate %	0	0
Duration of catheterization in days	5.89 (4–18)	6.2 (5–20)	0.36
Intraoperative complications
Rectal/visceral injury	0	0
Intraoperative transfusion	0	0
Early postoperative complications
Wound infection/abscess	1	1
Urinary infection	0	0
Urinary retention	0	0
Pulmonary embolism/DVT	0	0

LRP=laparoscopic radical prostatectomy; RALP=robot-assisted laparoscopic prostatectomy; SD=standard deviation; DVT=deep vein thrombosis.

Pathologic data

Pathologic data obtained from the surgical specimens are summarized in Table 3. Overall, the groups were similar in Gleason grade or pathologic stage. A higher number of patients in the RALP group had pT₃ disease, but this was not statistically significant. Consequently, the number of patients in the RALP group with margin positive disease was higher. These differences were not statistically significant.

Table 3.

Pathologic Data

	LRP (n=100)	RALP (n=100)	P value
Gleason grade (%)
≤6	35	33
7	48	52
>7	17	15
Pathologic stage (%)
pT_2a	5	10
pT_2b	2	2
pT_2c	70	55
pT_{3 (a,b)}	20 (10,10)	33 (17,16)
pT₄	3	0
Positive surgical margins (%)
Overall	14	19	0.343
pT₂	6.5% (5/77)	9.0% (6/67)
pT₃	30% (6/20)	39.4% (13/33)
pT₄	100% (3/3)

LRP=laparoscopic radical prostatectomy; RALP=robot-assisted laparoscopic prostatectomy.

Early complications and outcomes at 3 months

Complications within the first 30 days after surgery, the PSA status, and continence and potency, as obtained by validated questionnaires at 3 months, are summarized in Table 4. The number of patients in the RALP group needing prolonged catheterization was higher, but this was not statistically significant.

Table 4.

Postoperative Data and Outcomes

	LRP	RALP	P value
Early complications (within 1 month)
Anastomotic leakage
Catheter >7 days	2%	7%
Catheter >14 days	1%	4%
Symptomatic lymphocele	3.5% (2/58)	2% (1/50)
Bleeding/hematoma	0%	1%
Oncologic and functional outcomes at 3 months
PSA<0.2 ng/mL	93.6%	91.4	0.346
Continence (number of pads/day)
0–1	56	65	0.062
2	19	21
>2	25	14
Potency
Able to achieve intercourse	8.8 %	13.6%	0.58

LRP=laparoscopic radical prostatectomy; RALP=robot-assisted laparoscopic prostatectomy; PSA=prostate-specific antigen.

At 3 months, patients in the RALP group tended to have better continence, but this tendency was not statistically significant (P=0.062)

Of patients who underwent a nerve-sparing procedure, 8.8% and 13.6% were able to achieve an erection satisfactory for intercourse in the LRP and RALP groups, but this was statistically insignificant.

Discussion

LRP, while efficacious, is a challenging operation, and the technical demands of the operation and the learning curve prevented the adoption of LRP by most urologic surgeons.⁴ Specifically, the reduction in the range of motion, two-dimensional vision, counterintuitive movement, and reduced haptic sense are the main obstacles associated with a long learning curve.⁴

It was initially reported that it required nearly 60 cases to attain proficiency.¹⁸ With improvement in the technique of the procedure, structured training programs have been developed to enable safe and efficacious training of surgeons with no previous experience of open or laparoscopic surgery.¹⁶

The challenging issues facing skill acquisition were overcome in part by the introduction of the da Vinci Surgical System. While it is unclear as to who first performed the procedure, RALP was first described by Binder and Kramer¹⁹ and Abbou and associates²⁰ in 2000 and popularized by Menon and colleagues^21,22; it has since rapidly gained popularity, particularly in the United States, where by the end of 2008, three of four RPs were completed using the robotic approach.²³ Robot-assisted surgery offers the additional advantages of X10 magnified, binocular, three-dimensional vision with tremor filtration and miniature wristed, articulating instruments with seven degrees of freedom, which result in significantly improved surgical ergonomics. Because of improved precision and comfort, visualization and dissection of surgical planes is improved, particularly in areas such as the prostatic apex, where space is very limited. Suturing is also made significantly easier and does not necessitate the often complex maneuvering needed with rigid conventional laparoscopic instruments used in urethrovesical anastomosis.

Robotic surgery does, however, come at a premium: The initial cost of the system is in excess of 2 million euros, and it is estimated that when considering purchase and maintenance of the robot, the financial burden would increase by $2,698 per patient given an average of 126 cases per year.¹⁴ Menon and coworkers¹⁵ estimate that surgeons in an institution must perform 75 cases per year with an average operating time of 3 hours per case to be cost-effective in the United States.

The superiority of RALP over LRP has yet to be demonstrated; while RALP is elegant and poses significant ease to the surgeon, this has not been shown to translate into patient benefit in oncologic control or functional outcome from RP. A double-blind randomized controlled trial comparing LRP with RALP poses several difficulties: First, there is increasing patient perception that robotic surgery is perhaps superior to conventional laparoscopy, a fact partly attributable to the increased cost of RALP. It would be unlikely that patients would be willing to be randomized into the conventional laparoscopy arm. Second, institutions that have acquired the da Vinci system may not be able to justify “reverting” to conventional laparoscopy, because that may undermine the case for acquisition of the robot in the first instance. Third, not many institutions have equivalent or similar expertise in LRP and RALP beyond the learning curve; because surgical expertise is known to be a key factor in outcome, one of the techniques would have a disadvantage to start with.

There is just one report in the literature of a direct comparison of RALP vs LRP. Rozet and colleagues²⁴ compared 133 patients undergoing RALP with an equivalent cohort undergoing LRP, matched with respect to age, BMI, PSA, Gleason score, and clinical stage. They concluded that LRP is equivalent to RALP with respect to operative time, blood loss, hospital stay, length of bladder catheterization, and positive margin rate in the hands of skilled urologic surgeons at their institution. Other studies have demonstrated a statistically significant benefit in favor of RALP in operative blood loss, operative time, transfusion rate, conversion rate, and length of hospital stay and bladder catheterization.^15,25

The current study assesses the operative characteristics and early patient outcomes from initial RALP cases and compares them with LRP cases performed by the same surgeons well beyond their learning curve. Both groups in this series had broadly similar outcomes. We have previously published our results from 2000 LRP performed at our institution. The current series comparing early RALP experience with cases late in our LRP experience demonstrates some important points. While LRP had a significantly lower operative time and blood loss, the RALP group results were broadly similar. While the perioperative complication rate and outcomes at 3 months from the LRP group were satisfactory (not unexpected, given our significant previous experience), we have also achieved similar outcomes from our initial RALP group. Also, our preliminary RALP outcomes are at par with previous series published by groups significantly more established in the technique^24,25 and better than other groups that commenced RALP without such a large previous experience of LRP; earlier series published by Rozet and colleagues²⁴ and Ploussard and coworkers,²⁶evaluating RALP outcomes after the initial 133 and 206 patients, respectively, have reported mean blood loss of 609 and 504 mL, respectively, and a complication rate of 19.4% and 16.5%, respectively. This is significantly higher than this current series. Badani and associates²⁷ published their experience after 2766 cases, in which their mean operative blood loss was 100 mL, and the mean operative time was 154 minutes. These are superior, yet comparable to our initial series.

We had an overall margin positive rate of 9% for patients with organ-confined prostate cancer who underwent RALP. This compares favorably with previous series where this figure has been between 12.3% and 17.2%. Patients in the RALP group had a higher margin positive rate than those in the LRP group (19% vs 14%), but this was attributable to the higher number of patients in the RALP group with pT₃ disease (33% vs 20%). In patients with organ-confined disease, the positive margin rate was similar and low (6.5% and 9% for LRP and RALP, respectively).

This study is not without limitations. It has been designed to evaluate only the initial data from patients undergoing RALP and LRP. Obviously, the “trifecta” of outcome are best assessed later. Functional outcomes improve considerably and cannot be assessed at 3 months to draw final conclusions. We believe, however, that operative parameters and early postoperative complications are reflective of surgical expertise and parameters such as margin positivity are known to be key in oncologic outcome from the procedure.

Earlier series evaluating the learning curve of RALP have suggested that the learning process is expedited particularly for prostate cancer surgeons or those with previous experience of open surgery; this is of particular value for laparoscopy-naïve surgeons. Experienced laparoscopy surgeons, however, show an improvement only in the economy of movement and are not necessarily innately advantaged.^28,29

In our experience, the uptake of RALP with a background experience in LRP should be accelerated for a number of reasons: First, the steps of the robot-assisted procedure are similar to those of the conventional laparoscopic technique. Second, the absence of tactile feedback for RALP can be compensated for by the visual familiarity of the technique acquired with LRP. An example is the plane between the bladder and the prostate. While the feel of the bladder and the prostate (hollow vs firm) is different and may help identify the point of incision for the beginner, it becomes less important for experienced laparoscopists who would not necessarily rely on palpating the structures before incision. In this series, there were two positive surgical margins at the bladder neck in both the LRP group and the RALP groups; three of these patients had pT_3b disease while one had pT₄ disease, and all had a Gleason score of 9. In addition, there were no cases of “buttonholing” posteriorly, necessitating bladder neck reconstruction, suggesting that a lack of tactile sensation did not compromise dissection or oncologic outcome. This significant visual familiarity is carried over as is the knowledge of anatomic landmarks, which would provide an innate advantage.

This study demonstrates that surgeons with significant experience in laparoscopy are able to deliver early operative outcomes that are comparable to those of surgeons much more experienced in RALP. Also, the initial results from RALP are satisfactory and similar to those produced by LRP after significant experience. This translates into benefit for patients who are not disadvantaged by surgeons in the learning curve of their experience.

Footnotes

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Ferlay

, Parkin

, Steliarova-Foucher

. Estimates of cancer incidence and mortality in Europe in 2008. Eur J Cancer, 2010; 46:765–781.

Jemal

, Siegel

, Xu

, Ward

. Cancer statistics, 2010. CA Cancer J Clin, 2010; 60:277–300.

Walsh

. Cancer surveillance series: Interpreting trends in prostate cancer—part I: Evidence of the effects of screening in recent prostate cancer incidence, mortality, and survival rates. J Urol, 2000; 163:364–365.

Xylinas

, Ploussard

, Durand

, de la Taille

. Robot-assisted extraperitoneal laparoscopic radical prostatectomy: A review of the current literature. Urol Oncol, 2010E-pub ahead of print.

Walsh

. Radical prostatectomy for the treatment of localized prostatic carcinoma. Urol Clin North Am, 1980; 7:583–591.

Schuessler

, Schuam

, Clayman

, Kavoussi

. Laparoscopic radical prostatectomy: Initial short-term experience. Urology, 1997; 50:854–857.

Guillonneau

, Vallancien

. Laparoscopic radical prostatectomy: The Montsouris technique. J Urol, 2000; 163:1643–1649.

Rassweiler

, Sentker

, Seemann

et al. Laparoscopic radical prostatectomy with the Heilbronn technique: An analysis of the first 180 cases. J Urol, 2001; 166:2101–2108.

Türk

, Deger

, Winkelmann

et al. Laparoscopic radical prostatectomy. Technical aspects and experience with 125 cases. Eur Urol, 2001; 40:46–53.

10.

Stolzenburg

, Do

, Pfeiffer

et al. The endoscopic extraperitoneal radical prostatectomy (EERPE): Technique and initial experience. World J Urol, 2002; 20:48–55.

11.

Stolzenburg

, Schalenberg

, Horn

et al. Anatomical landmarks of radical prostatecomy. Eur Urol, 2007; 51:629–639.

12.

Stolzenburg

, Rabenalt

, Do

et al. Intrafascial nerve-sparing endoscopic extraperitoneal radical prostatectomy. Eur Urol, 2008; 53:931–940.

13.

Kaul

, Bhandari

, Hemal

et al. Robotic radical prostatectomy with preservation of the prostatic fascia: A feasibility study. Urology, 2005; 66:1261–1265.

14.

Bolenz

, Gupta

, Hotze

et al. Cost comparison of robotic, laparoscopic, and open radical prostatectomy for prostate cancer. Eur Urol, 2010; 57453–458.

15.

Menon

, Shrivastava

, Tewari

. Laparoscopic radical prostatectomy: Conventional and robotic. Urology, 2005; 66,suppl 5:101–104.

16.

Stolzenburg

, Rabenalt

, Do

et al. Modular training for residents with no prior experience with open pelvic surgery in endoscopic extraperitoneal radical prostatectomy. Eur Urol, 2006; 49:491–500.

17.

Stolzenburg

, Rabenalt

, Do

et al. Endoscopic extraperitoneal radical prostatectomy: The University of Leipzig experience of 2000 cases. J Endourol, 2008; 22:2319–2325.

18.

Guillonneau

, Rozet

, Barret

et al. Laparoscopic radical prostatectomy: Assessment after 240 procedures. Urol Clin North Am, 2001; 28:189–202.

19.

Binder

, Kramer

. Robotically-assisted laparoscopic radical prostatectomy. BJU Int, 2001; 87:408–410.

20.

Abbou

, Hoznek

, Salomon

et al.

[Remote laparoscopic radical prostatectomy carried out with a robot. Report of a case]

(Fre) Prog Urol, 2000; 10:520–523.

21.

Menon

, Shrivastava

, Tewasri

et al. Laparoscopic and robot assisted radical prostatectomy: Establishment of a structured program and preliminary analysis of outcomes. J Urol, 2002; 168:945–949.

22.

Menon

, Tewari

, Baize

et al. Prospective comparison of radical retropubic prostatectomy and robot-assisted anatomic prostatectomy: The Vattikuti Urology Institute experience. Urology, 2002; 60:864–868.

23.

Kwon

, Bautista

, Jung

et al. Impact of robotic training on surgical and pathologic outcomes during robot-assisted laparoscopic radical prostatectomy. Urology, 2010; 76:363–368.

24.

Rozet

, Jaffe

, Braud

et al. A direct comparison of robotic assisted versus pure laparoscopic radical prostatectomy: A single institution experience. J Urol, 2007; 178:478–482.

25.

Joseph

et al.

Robot-assisted vs pure laparoscopic radical prostatectomy: Are there any differences?

BJU Int, 2005; 96:39–42.

26.

Ploussard

, Xylinas

, Salomon

et al. Robot-assisted extraperitoneal laparoscopic radical prostatectomy: Experience in a high-volume laparoscopy reference centre. BJU Int, 2010; 105:1155–1160.

27.

Badani

, Kaul

, Menon

. Evolution of robotic radical prostatectomy: Assessment after 2766 procedures. Cancer, 2007; 110:1951–1958.

28.

Chandra

, Nehra

, Parent

et al. A comparison of laparoscopic and robotic assisted suturing performance by experts and novices. Surgery, 2010; 147:830–839.

29.

Ahlering

, Skarecky

, Lee

, Clayman

. Successful transfer of open surgical skills to a laparoscopic environment using a robotic interface: Initial experience with laparoscopic radical prostatectomy. J Urol, 2003; 170:1738–1741.