This special installment of "Hands On" is presented in a point-counterpoint format, with two leading experts discussing open versus. robot-assisted radical prostatectomy.
Still standing at the station: The case for open RP
Joel B. Nelson, MD
Dr. Nelson is professor and chairman of the department of urology, University of Pittsburgh School of Medicine.
Dr. NelsonThe format of the open versus robotic radical prostatectomy debate-a piling up of uncontrolled observations to support one position or another-is regrettable. Urology has gone headlong into the use of robot-assisted radical prostatectomy (RARP) in the management of clinically localized prostate cancer with flimsy evidence for an advance over the standard approach, namely, open radical prostatectomy (ORP). As such, our field stands in distinction to our colleagues in general surgery, who conducted the prospective, randomized trials needed to examine the use of minimally invasive surgery in the treatment of colon cancer and those needed to compare the combination of lumpectomy and radiation versus radical mastectomy for breast cancer.
At a time when there is growing evidence to support the use of active surveillance in the initial management of low-risk prostate cancer, we are seeing increased use of robotic surgery in the initial management of prostate cancer-most strikingly in the elderly-driven by the economic demands inherent with RARP adoption (N Engl J Med 2010; 363:701-4). The imbalance of expectations and reality results in regret, an outcome reported significantly more frequently with RARP-treated patients (Eur Urol 2008; 54:785-93). The painful reality is that RARP and ORP can’t be compared based on the most rigorous standard; namely, an adequately powered randomized controlled trial, and it seems increasingly unlikely this will ever be achieved.
The goal of radical prostatectomy, by any approach, is to cure prostate cancer. In the absence of long-term data, one cannot claim RARP is equivalent to ORP in achieving this goal. Setting aside whether a comparison of short-term results is even meaningful when metastatic progression and prostate cancer-specific death is a remote event, the most mature published data of biochemical-free survival results of RARP are, in the words of the authors, “unexpectedly low” (Cancer 2007; 110:1951-8). In a series of 2,766 consecutive RARPs from the pioneering center for this approach, the 5-year actuarial biochemical-free survival was 84% in a cohort initially selected for low-risk features. Compare those results to 2,391 ORP patients from my practice (932 at risk at 60 months) with a biochemical recurrence–free survival rate at 5 years of 93.3% (95% CI: 92.0–94.6).
Furthermore, surgeons performing RARP perform pelvic lymph node dissections five times less frequently than those performing ORP, and this practice is independent of risk stratification (Cancer 2011; 117:3933-42). By scuttling the oncologic goal of surgery-to remove the cancer-RARP alone will never have the same results as ORP with a well-performed pelvic lymph node dissection. For example, considering 89% biochemical free-survival at 10 years after ORP with pelvic lymph node dissection reported recently by surgeons at USC (Urology 2012; 79:626-31), it is reasonable to question whether RARP will ever reach that bar.
Historical comparisons of RARP to ORP based on intermediate endpoints, such as positive surgical margins, are fraught with methodologic challenges. Many of the largest ORP series bridged the PSA era, and the prostate cancer patient of 1990 is not the prostate cancer patient of today. Indeed, year of surgery has been shown to be an independent prognostic factor favoring more contemporarily treated patients. In a series of 2,700 consecutive ORPs that I performed, the overall positive surgical margin rate is 7% (190/2,700); the positive margin rate for pT2 disease is 2.3% and 21.1% for pT3. Despite an increase in pT3 disease-from a historic rate of less than 25% up to 32% in the last 600 cases-there has been significant decline in positive margins (figure 1). In the last 1,000 cases, over 90% of the procedures were bilateral nerve-sparing surgeries.
How are these results achieved? In my opinion, ORP allows the intra-operative flexibility that simply does not exist with RARP. Decisions on a plane of dissection are based on both binocular vision (the natural state of having two functioning eyes) and the tactile cues transmitted through the instruments. Although surgeons performing RARP may be able to compensate to some degree for the lack of haptic feedback, no robotic surgeon would reject a device providing the sense of touch.
Furthermore, ORP permits the surgeon to operate in a largely bloodless field through the use of suction. Admittedly, RARP is associated with less overall blood loss, but I am struck by how often the dissection in RARP is performed in a pool of blood or with less than ideal visualization. Tenfold magnification is not the answer to a better inter-operative view; the key is actually being able to see the tissue. Furthermore, with a focus on meticulous control of bleeding, I have seen a significant decline in estimated blood loss, averaging between 300 and 350 mL over the last 1,200 cases. The straw horse argument of less blood loss with RARP is lost when one considers the equivalent transfusion requirements compared to ORP we see in our program.
Claims of reduced perioperative pain and quicker return to full activities with RARP are not supported by careful, prospective, non-randomized observational studies (Urology 2007; 70:945-9; J Urol 2005; 174:912-4) (figure 2). Average length of hospital stay in our department between RARP and ORP are 1.2 and 1.4 days, respectively-no significant difference.
A reasonable criticism of this debate is a reliance on data from high-volume centers with surgeons who do nothing else. It has been proposed that RARP allows the lower volume surgeon with average skills to achieve better results than they could with ORP. Unfortunately, a recent study does not support this perspective (J Clin Oncol 2012; 30:513-8). In a random sample of 20% Medicare recipients in a nationwide survey who have undergone either ORP or RARP, one-third said they were experiencing a moderate to big problem with continence, and nearly 90% of patients reported difficulty with sexual function.
While sexual function outcomes were equally poor among both ORP and RARP patients, there was a non-significant trend toward worse continence outcomes with RARP (odds ratio: 1.41; 95% CI: 0.97–2.05). Currently, it does not appear RARP is producing any better functional outcomes than ORP.
Reportedly, RARP accounts for 80% of all prostatectomies being performed in the U.S., a remarkable dissemination of a new technology in a decade. Why? Let’s be honest: This has been driven much more by an equally remarkable marketing campaign-fueled by the fear of being left behind-than by any real improvement in oncologic or functional outcomes. In most communities, surgeons still offering ORP are viewed as archaic, a perspective at least passively exploited by competing surgeons offering RARP (ie, Isn’t new technology inherently better?). In my opinion, the most technologically advanced device is still the human hand attached to an experienced and well-trained brain.
The wide adoption of RARP has increased the financial burden of managing localized prostate cancer. Even at a high-volume academic hospital like mine that is clearly focused on the bottom line (reporting a $351 million operating income on $9.6 billion operating revenues in FY ‘12), RALP is performed at a loss of $4,013 per case (Urology 2012; 80:126-9). In the coming days of increased scrutiny on health care expenditures, it will be wise to assure ORP is not a lost art.
As one of the few ORP surgeons in the United States who has maintained (and grown) this practice in the RARP era, I am regularly asked to defend ORP. We are told the train has left the station on radical prostatectomy in the United States, despite an absence of significant evidence-based advantages of RARP over ORP. I tell my patients that in my hands, ORP is much better than any published results for RARP, and I am happy to be still standing in the station for them.
Current status of robotic RP: Is the evidence robust?
Ananthakrishnan Sivaraman, MS, MCh (Uro)
Srinivas Samavedi, MD
Vipul R. Patel, MD
Dr. Sivaraman is a consultant urologist at Apollo Hospitals of Chennai, India; Dr. Samavedi is a clinical fellow at Florida Hospital’s Global Robotics Institute in Celebration, FL; and Dr. Patel is medical director of the Global Robotics Institute.
It has been slightly more than a decade since the full-fledged introduction of robotic surgery. From very humble beginnings, it has progressed rapidly, so much so that in the U.S., approximately 85% of all prostatectomies are performed robotically, according to Intuitive Surgical, Inc., the Sunnyvale, CA maker of the da Vinci robot.
When the technology was introduced, the expectations regarding outcomes were much different than those of today (BJU Int 2001; 87:408-10). Initial expectation was that robotic prostatectomy be done in a safe manner, in a reasonable time frame, and with lower blood loss when compared to open surgery. Now, both the referring doctors and patients hold robotic surgeons to far higher standards and expect significant differences in outcomes. Fortunately, the experience we have gained in the last 10 years allows us to deliver on these expectations. As our experience at the Global Robotics Institute has reached 5,600 robotic prostatectomies, we chronicle our journey to achieving optimal outcomes and present our view on the current debate between robotic and open prostatectomy.
Examining current evidence for robotic surgery, we found a paucity of direct, head-to-head comparisons between conventional and robotic surgery. Such randomized trials may never be possible due to patient choice. Hence, the bulk of our data is from nonrandomized studies.
In a recent landmark paper that provides a contemporary snapshot of current perioperative outcomes, Trinh et al assessed the rate of robot-assisted laparoscopic radical prostatectomy (RALP) utilization and the differences in perioperative complication rates between RALP and open radical prostatectomy (ORP) (Eur Urol 2012; 61:679-85). Of 19,462 surgeries performed, 61.1% were RALP, 38.0% were ORP, and 0.9% were laparoscopic radical prostatectomy (LRP). In multivariable analyses of propensity score–matched populations, patients undergoing RALP were less likely to receive a blood transfusion (odds ratio: 0.34; 95% confidence interval: 0.28–0.40), experience an intraoperative complication (OR: 0.47; 95% CI: 0.31–0.71), experience a postoperative complication (OR: 0.86; 95% CI: 0.77–0.96), or have a prolonged length of stay (OR: 0.28; 95% CI: 0.26–0.30).
Moreover, when individual postoperative complications were examined, cardiac, respiratory, and vascular complications were found less likely to occur in patients undergoing RALP than in patients undergoing ORP, indicating a beneficial effect of RALP on medical complications as well. Previous population-based studies showed shorter lengths of stay and fewer respiratory complications, miscellaneous surgical complications, and strictures. In our own published series of 2,500 RALPs, we encountered total complications of 5.08% (Eur Urol 2010; 57:945-52). Out of all 2,500 cases, the majority were Clavien grade 1 or 2 (4.04%).
In a systematic review and cumulative analysis of published literature up to 2009, Ficarra et al analyzed positive surgical margins (PSM) in ORP and RALP series (Eur Urol 2009; 55:1037-63). In their analysis, a statistically significant difference in favor of RALP was identified (relative risk: 1.58; 95% CI: 1.29–1.94; p<.00001).
Tewari et al conducted a systematic review and meta-analysis comparing retropubic, laparoscopic, and robotic prostatectomy abstracted data from 400 original research articles representing 167,184 ORP patients, 57,303 LRP patients, and 62,389 RALP patients (table 1) (Eur Urol 2012; 62:1-15). Primary outcomes were PSMs and complication rates. The overall PSM rates were 24.2% for ORP patients and 16.2% for RALP patients; pT2 PSM rates were 16.6% for ORP patients and 10.7% for RALP patients, but the difference did not attain significance after propensity score adjustment and Hochberg correction (overall PSM, p=.002; pT2 PSM, p=.01). At the very least, these results indicate that RALP is equal, if not superior to, ORP in terms of PSMs.
Examining oncologic outcomes, Masterson et al, in a retrospective review of 1,041 patients who underwent ORP or RALP between 1999 and 2010, compared PSMs and oncologic outcomes in 357 ORP patients and 669 RALP patients. Comparing biochemical recurrence-free survival rates according to surgical approach, no differences were seen at 24 or 60 months postoperatively between ORP patients (87% and 71%, respectively) and RALP patients (87% and 73%, respectively). However, longer-term results are required for effective comparison of oncologic outcomes. Current results indicate that oncologic outcomes are at least similar, if not better, in RALP patients than in those undergoing ORP.
With long-term survival ensured for localized prostate cancer, functional outcomes have become the focus of prostatectomy. A large systematic review by Berryhill et al found continence rates to range from 73% to 91% and 54% to 87% for RALP and ORP, respectively (Urol 2008; 72:15-23). Koehler et al, in a multicenter, longitudinal study of 350 prostate cancer patients (166 RALP and 184 ORP) in seven German hospitals, assessed early continence rates and found no significant difference between the approaches at 3 months (44% for RALP and 40% for ORP) (Urol Oncol, June 28, 2011). Hu et al found a significantly higher rate of urinary incontinence diagnoses for minimally invasive RP compared to ORP (15.9 vs. 12.2 per 100 person-years, p=.02) (JAMA 2009; 302:1557-64).
Nevertheless, with regard to analysis of the need for urinary incontinence procedures, which is a more realistic urinary continence endpoint, two population-based studies found no significant difference in the need for post-RP urinary-continence procedures (JAMA 2009; 302:1557-64; J Urol 2010; 183:1366-72). Coelho et al, in an analysis of data from high-volume centers, found that weighted mean continence rates were 80% for ORP and 92% for RALP (J Endourol 2010; 2279-83).
In a nonrandomized prospective trial comparing ORP and RALP, Tewari et al found that the median time to erectile function recovery was 440 days after ORP and 180 days after RALP (J Clin Oncol 2008; 26:4999-5000). Similarly, the median time to intercourse was 700 days after ORP and 340 days after RALP.
Ficarra et al defined potency as an International Index of Erectile Function-5 score of >17 (Eur Urol 2007; 51:45-56). Limiting their analysis to only those patients who underwent bilateral nerve-sparing RP with at least 1 year of follow-up, they found that 49% of ORP patients and 81% of RALP patients were potent by their definition (p<.001 adjusted for the effects of age, preoperative erectile function, and comorbidity, all of which may have been very different between the two groups). These results reflect the outcomes from our institution, where we have seen continence rates of 97.4% and potency rates of 91.5% in preoperatively potent men who undergo bilateral nerve sparing. Thus, in terms of functional outcomes, RALP seems again to have an advantage over ORP.
Financial aspects, limitations, learning curve
Primary limitations for robotic prostatectomy include considerable cost disadvantages. Lack of competition has resulted in a lack of incentive for manufacturer Intuitive Surgical to reduce costs. Very few centers are able to consistently make a profit from the robot. However, we believe these lacunae can be partially overcome by improving efficiency of robot utilization and increasing volume and centralization of services.
RALP results are generally measured in terms of the trifecta and the pentafecta (table 2). The trifecta is achieving cancer control, continence and potency concurrently. The pentafecta is achieving the trifecta without complications or positive surgical margins. RALP resulting in optimal outcomes (trifecta, pentafecta) requires a significant learning curve (J Urol 2010; 183:1360-5). This learning curve has to be breached for optimal outcomes, and this can only be achieved in high-volume centers. Such centers of excellence for training and performing the bulk of cases around the country represent the current need. We believe the approach (open or robotic) is not as important as the experience of the surgeon.
Although RALP has eclipsed ORP in terms of volume, there are still a few areas in which it has to catch up. One such area is training. A structured curriculum is essential for training and objectively assessing future robotic surgeons. Currently, the certification process to be a robotic surgeon rests with the manufacturer. This process needs to be more structured; hence, the process of creating a curriculum for robotic surgery has been initiated. Defining outcomes, curriculum, and specific training tasks, along with validating the training tasks and measurements and setting up passing criteria, are current goals for improving the quality of robotic surgery being performed as well as achieving a uniformity of outcomes among centers.
Robotic prostatectomy has progressed rapidly since its introduction. RALP surgical outcomes have also kept pace and currently are second to none. To improve uniformity in outcomes, a structured curriculum with well-defined tasks and measures is essential to train future robotic surgeons. High-volume centers of excellence can ideally fulfill this need for adequate training and thus ensure optimal outcomes.
Debates regarding the approach to prostatectomy (open vs. robotic) are not as important as the experience of the surgeon. More importantly, research needs to be directed toward accurately risk stratifying prostate cancer patients so that they may choose the optimal treatment option.UT
Subscribe to Urology Times to get monthly news from the leading news source for urologists.