Abstract
Background
Improved cancer control with increasing surgical experience (the “learning curve”) has been demonstrated for open and laparoscopic prostatectomy. We aim to assess the relationship between surgical experience and oncologic outcomes of robot-assisted radical prostatectomy.
Methods
The study cohort included 2857 prostate cancer patients treated with robot-assisted radical prostatectomy at a single tertiary care center. For each patient, surgical experience was coded as the total number of robotic prostatectomies performed by the surgeon before the patient’s operation. The relationship between a surgeon’s prior experience and the probability of positive surgical margins and biochemical recurrence was evaluated in regression models, adjusting for stage, grade and PSA.
Results
After adjusting for case mix, greater surgeon’s experience was associated with a lower probability of positive surgical margins (p=0.035). The risk of positive margins decreased from 15.3% to 6.7% for a patient treated by a surgeon with 10 and 250 prior procedures, respectively (risk difference between 10 and 250 procedures 8.6%, 95%CI 2.3 to 12.8). In non-organ confined disease, the predicted probabilities of positive margins were 41.5% for patients treated by surgeons with 10 prior operations and 21.1% for patients treated by surgeons with 250 prior operations (absolute risk reduction 20.4%, 95%CI: −10.8 to 23.7). The relationship between surgical experience and the risk of biochemical recurrence after surgery was not significant (p=0.8).
Conclusions
More experienced surgeons have lower risk of positive margins, while the probability of recurrence after robotic prostatectomy is not affected by experience. Aspects of surgical technique associated with improved margins rate need consideration. The impact of experience on cancer control after robotic prostatectomy should be further investigated in larger, multi-institutional studies.
Keywords: Prostate Cancer, Robot-Assisted Radical Prostatectomy, Robotic Surgery, Learning Curve, Positive Surgical Margins, Biochemical Recurrence, Oncologic Outcomes, Cancer Control
INTRODUCTION
It is widely acknowledged that the outcome of surgery is related to the experience of the surgeon, what is commonly referred as the “learning curve”. The learning curve pertains to both general surgical technique, and mastery of new operating procedures. Studies on the learning curve in urology have become widespread in recent years[1].
The relationship between experience and the outcomes of radical prostatectomy has been previously evaluated[2]. Although it has commonly been shown that the results of a surgeon improve with experience, the majority of learning curve studies focused on technical aspects such as transfusion and operative time[3–5]. These outcomes are undoubtedly important, particular with respect to understanding how surgeons master the surgical techniques, but are less relevant to patients than the main goal of oncologic surgery, that is, cancer control.
The impact of experience on oncologic efficacy of radical prostatectomy has been assessed in open[6] and laparoscopic[7] series. Although the improvement in outcomes was slower for laparoscopy than for the open approach, both the studies found a learning curve for cancer control. It is noteworthy that such papers included many surgeons from several institutions. By contrast, the evidences on the learning curve for robotic prostatectomy are limited. In one of the few studies in the literature, the association between experience and the risk of recurrence was assessed for a single surgeon converting from open to robotic surgery[8]. Other investigators assessed the learning curve for minimally invasive radical prostatectomy performed by 9 surgeons[9]. However, instead of calculating a learning curve, the authors split patients into different categories of experience, which has been demonstrated to be a suboptimal method[10].
Accordingly, we used individual data from our institutional database to assess how a surgeon’s prior experience is related to the oncologic efficacy of robotic radical prostatectomy.
MATERIALS AND METHODS
Our study population consisted of 2857 patients with clinically localized prostate cancer who were treated by robot-assisted radical prostatectomy (RARP) between 2006 and 2017 at San Raffaele Hospital (Milan, Italy). We excluded patients who received neo-adjuvant (n=83) or adjuvant therapies (n=184), who had missing data for the covariates (n=243) or were treated by surgeons with a total experience of less than 50 RARPs (n=116) at the time their last patient was included in this study, leaving 2231patients eligible for analysis. Eligible patients were treated at our institution by one of nine surgeons. Details of surgical experience prior to the first RARP at our institution were requested for each surgeon. All but one of the surgeons in our series performed their first robotic procedure at our hospital; the surgeon who had previous robotic experience reported having done 50 procedures before moving to our institution.
The aim of the study was to evaluate the impact of surgical experience on the rate of positive surgical margin (PSM) and biochemical recurrence (BCR) after robot-assisted radical prostatectomy. PSM status was defined as a tumor involving the inked margin of resection in the surgical specimen. Follow up consisted of measuring serum PSA levels every 3 months during the first year after surgery, every 6 months during the second year, and annually thereafter. Biochemical recurrence was defined as a PSA concentration of more than 0.2 ng/ml in two consecutive measurements.
Surgical technique
Surgery was performed using a conventional surgical approach for robot-assisted radical prostatectomy, as previously described[11]. An extended PLND was performed in patients with a preoperative risk of nodal involvement greater than 5% according to the Briganti nomogram[12]. Dedicated uro-pathologists examined the surgical specimens. For grading and staging purposes, the most updated ISUP grading system[13] and TNM classification[14] at the time of evaluation were used.
Statistical analysis
Surgical experience was coded as the number of prior robotic prostatectomies performed by the surgeon at the time of the index patient’s operation. Procedures were considered as part of a surgeon’s prior experience only if the surgeon was documented as having primary responsibility, while those at which the surgeon assisted, for instance, during training, were not included. The number of prior operations includes those involving patients unsuitable for the analysis as well as those performed at a different institution and thus reflects the actual robotic experience for each surgeon prior to a given patient’s surgery. During our preliminary analysis, we found that one surgeon had almost twice the number of cases as the next most experienced surgeon. As previously described[10] this can distort estimation of learning curves. Accordingly, we curtailed surgical experience at 500 procedures. Removing these 404 cases, our final cohort for analysis consisted of 1827 records.
Our analysis consisted of three main steps. First, a multivariable logistic regression model was used to assess the association between positive margin status and surgical experience (entered as a continuous variable). The adjustment for case mix among surgeons was based on the following covariates: total preoperative PSA (continuous), pathological ISUP grade (1 vs 2 vs 3 vs 4–5), seminal vesicle involvement (yes/no), extra-prostatic extension (yes/no), nodal status (pN1 vs pN0 vs pNx) and at least 50 open RPs prior to the first RARP (yes/no). As there is evidence of a non-linear relationship between surgical experience and the PSM in previous literature[15], we included RARP surgical experience as a non-linear term using restricted cubic splines with knots at quartiles. Moreover, since data from different patients treated by the same surgeon may be correlated, we incorporated surgeon clustering in our analysis using a generalized estimating equations approach through the use of the cluster option in Stata statistical software. To visualize our findings, we calculated the probability of PSM and its confidence intervals for each level of surgical experience by setting variables at the mean. Since the local stage may influence the rate of positive surgical margins, we stratified and plotted our results according to whether the tumor was organ confined on analysis of the surgical pathology.
Second, we assessed the relationship between the surgical experience and biochemical recurrence. Since data on biochemical recurrence were available only for 1283 (71%) patients, we investigated whether patients with available or missing BCR data had similar disease characteristics by Wilcoxon rank-sum and Chi-squared.
Finally, to evaluate the association between a surgeon’s robotic experience and recurrence after radical prostatectomy we created a multivariable Cox regression model. The identified covariates were the same described above and, similarly, we included restricted cubic splines for modeling the effects of surgical experience on BCR. Then, we calculated the 5-year probability of freedom from BCR and the corresponding confidence intervals and used that likelihood to produce a learning curve according to the surgical experience.
RESULTS
The distribution of surgeons by their total number of procedures performed and by the median annual caseload is shown in Table 1a and 1b. Although many of the surgeons have an annual caseload of <30 procedures, roughly half of the patients were treated by a high-volume (≥80 procedures/year) surgeon. Table 2 represents the clinical and pathological characteristics of our cohort, according to different levels of surgical experience. There were significant differences between the groups, with less experienced surgeons treating patients with more advanced disease than surgeons with a greater prior experience.
Table 1a.
Distribution of surgeons and number of patients according to total lifetime number of robotic procedures performed per surgeon
| Total lifetime number of robotic prostatectomies performed | Number of surgeons (%) | Number of patients (%) |
|---|---|---|
| 50–149 | 4 (44) | 312 (17) |
| 150–299 | 2 (23) | 288 (16) |
| 300+ | 3 (33) | 1227 (67) |
| Total | 9 (100) | 1827 (100) |
Table 1b.
Distribution of surgeons and number of patients according to the annual caseload per surgeon
| Annual number of robotic prostatectomies performed | Number of surgeons (%) | Number of patients (%) |
|---|---|---|
| <30 | 3 (33) | 294 (16) |
| 30–79 | 4 (44) | 718 (39) |
| ≥80 | 2 (23) | 815 (45) |
| Total | 9 (100) | 1827 (100) |
Table 2.
Clinical and pathologic characteristic of the study cohort by level of experience (prior surgeries) of the surgeon at the time of the index patient’s operation.
| 0–99 (N=685; 37%) | 100–249 (N=564; 31%) | 250+(N=578; 32%) | P value | |
|---|---|---|---|---|
| Total PSA, ng/ml | 6.0 (4.7, 8.2) | 6.1 (4.8, 8.2) | 5.9 (4.6, 8.0) | 0.5 |
| Seminal Vesicles Involvement | 18 (2.6%) | 22 (3.9%) | 20 (3.5%) | 0.4 |
| Extracapsular Extension | 155 (23%) | 134 (24%) | 121 (21%) | 0.5 |
| Pathological ISUP grade | ||||
| 1 | 226 (33%) | 168 (30%) | 210 (36%) | 0.006 |
| 2/3 | 412 (60%) | 356 (63%) | 350 (61%) | |
| >3 | 47 (6.9%) | 40 (7.1%) | 18 (3.1%) | |
| Nodal Status | ||||
| pN0 | 412 (60%) | 398 (71%) | 489 (85%) | <0.0001 |
| pN1 | 13 (1.9%) | 15 (2.7%) | 10 (1.7%) | |
| pNx | 260 (38%) | 151 (27%) | 79 (14%) |
Our multivariable model for the prediction of positive surgical margin showed significant, non-linear association with surgical experience (p=0.035). The PSM learning curve is shown in Figure 1. The rate of positive surgical margins lowers with increasing surgical experience and starts to plateau at the 200th procedure. The risk of positive margins for a surgeon with 10 and 250 prior operations was 15.3% and 6.7%, respectively (absolute difference 8.6%, 95%CI 2.3, 12.8). Stratified by pathologic stage (T2 vs T3), the curve for extra-prostatic disease was steeper. More specifically, the risk of positive margins for a patient with non-organ confined disease treated by a surgeon with 10 vs 250 prior operations was 41.5% and 21.1% (absolute risk reduction 20.4%; 95%CI: −10.8 to 23.7). Interestingly, prior open experience did not significantly affect the PSM rate (OR: 0.83; 95%CI: 0.55, 1.25; p=0.4).
Figure 1.
PSM learning curve in patients with typical cancer severity. Predicted probability of positive surgical margins (solid) and 95% CI (dash) by increasing level of surgical experience. The results are presented for the entire cohort (A) and stratified by pathologic stage (B).
Since it is plausible that our results concerning surgical margins might have been influenced by a few number of surgeons who developed a high level of experience, we restricted the analysis to 447 patients treated by surgeons who had performed <200 total procedures. The results were similar to our main analyses in this subgroup of patients, with greater prior experience associated with lower risk of positive margins (p<0.0001). Similarly, our findings were unaffected when we restricted the analysis to patients whose surgeons had performed at least 200 procedures in their career to date, testing the hypothesis that a learning curve is still present in highly experienced surgeons. The risk of positive margins decreased from 10.2% to 7.1% for a surgeon with 10 and 250 prior procedures, respectively (absolute reduction 3.1%, 95%CI: 0.6 to 6.8).
Data on biochemical recurrence were available for 1283 (71%) patients. When compared with patients lost at follow-up, the group with available BCR data had slightly lower preoperative PSA level (Median: 5.9 vs 6.2 ng/ml; p=0.02) and a lower rate of ISUP 2–3 tumors (58% vs 67%, p=0.001). There were 118 biochemical recurrences with a 3- and 5-years BCR-free probability of 92% (95%CI: 90, 93) and 85% (95%CI: 82, 88), respectively. The number of patients who did not experience a BCR with available follow-up data of 3 and 5 years was 496 and 172. The median follow-up for patients without biochemical recurrence was 32 months (IQR 16, 50).
Figure 2 shows the 5-year recurrence-free probability according to different levels of surgical experience at the time of the patient’s surgery. On multivariable Cox regression, the relationship between surgical experience and biochemical recurrence risk was not significant (p=0.8). To graphically evaluate our analyses, we plotted the 5-years probability of freedom from biochemical recurrence after robotic prostatectomy by increasing surgical experience in Figure 3.
Figure 2.
Probability of freedom from biochemical recurrence after robotic radical prostatectomy. The data are stratified by surgeon experience (number of prior procedures) at the time of the patient’s surgery
Figure 3.
Probability of freedom from biochemical recurrence after robotic prostatectomy by increasing surgical experience. Solid line: predicted probability. Dashed lines: 95% CIs
DISCUSSION
We noted a learning curve for surgical margins after robotic radical prostatectomy. As such, the risk of positive margins after RARP for more experienced surgeons was lower than that of surgeons with less prior experience. We also found that surgical experience and biochemical recurrence after robotic prostatectomy are not significantly associated. Accordingly, our results suggest that robotic prostatectomy has no learning curve for cancer control.
Previous studies have examined the relationship between surgical margins and surgeon’s experience in robotic prostatectomy[4, 9, 16]. However, such evidences were limited by their comparative[9, 16] and single-surgeon nature[4, 9] and did not provide insight on the shape of the learning curve for surgical margins. To date, this is the first study that evaluates the relationship between surgical experience and the probability of PSM after robotic radical prostatectomy in a multi-surgeon series.
We found a significant lower risk of positive margins with increasing experience. A surgeon with a prior experience of 250 operations compared to a surgeon who has done 10 prior surgeries had a PSM absolute risk reduction of 8.6%. We also reported that a prior experience in open surgery was not associated with the risk of positive margins during robotic prostatectomy. A number of possible explanations may be postulated. For example, the absence of haptic feedback may limit the translation of important skills from open surgery, during the dissection of the gland as well as during challenging steps that affect the risk of positive margins (i.e. nerve-sparing). It is also possible that the magnified vision allows a different surgical anatomy compared to the open technique[17, 18]. Thus, surgeons with prior open experience may not be able to apply their technique in such new surgical field. Similar findings resulted from a laparoscopic series[7], supporting the evidence that each surgical approach has its own operating technique.
Our results also showed that surgical experience and the probability of biochemical recurrence after robotic prostatectomy were not significantly associated. Being radical prostatectomy a highly skilled procedure, better outcomes for improved experience would be not surprising, as previously shown for open[6] and laparoscopic[7] techniques. However, it is plausible that aspects of the robotic technique affect cancer control differently than in other surgical approaches. For example, the high dexterity provided by robotic surgery may ease the dissection of the prostate or facilitate the surgeon in thoroughly remove the lymph nodes. It is thus possible that robotic prostatectomy ensures a proper cancer excision during the early procedures, with oncologic efficacy even for less experienced surgeons. Such hypothesis seems to contradict our finding of a learning curve for surgical margins. However, the magnitude of the relationship between margin status and recurrence is still debatable[15]. Although positive margins are associated with higher risk of recurrence[19], improved PSM rate for experienced surgeons does not necessarily translate in lower risk of biochemical recurrence. Therefore, it is plausible that the process of learning for surgical margins might be partially independent from the oncologic efficacy of robotic surgery.
The non-significant association between experience and cancer control may also be explained by the relatively limited number of surgeons in our series (9). Prior evidence of a learning curve for laparoscopic prostatectomy resulted from 29 surgeons from 7 institutions[7]. Although the single center nature of our study limited differences in surgical technique, disease management (e.g., indication for lymphadenectomy) or definition of biochemical recurrence adopted[20], there is a clear need to further investigate the surgical learning of robotic prostatectomy in larger, multi-institutional studies.
Other studies observed a learning curve for recurrence rates after robotic prostatectomy when compared to those of open[8] and laparoscopic[9] technique. However, these findings may reflect a different methodology. Thompson et al[8] described a single surgeon series while Sivaranam et al[9] compared recurrence rates of surgeons grouped according to different levels of experience. As previously described[10], these represent flaws of learning curve studies. By contrast, our approach to the surgical learning curve likely reflect the true relationship between prior robotic experience and cancer control. Although our findings may be taken to suggest that the oncologic efficacy of robotic prostatectomy is not affected by experience, the graphical representation of our results showed confidence interval that may be consistent with different-shapes curves, including a learning curve. As such, further research is needed to draw definite conclusions.
Our study has several limitations. First, our cohort did not have complete recurrence data. This reflects the nature of our institution that, as a tertiary center, provides care to external patients who may not be willing to follow-up after surgery. However, the comparison of patients with available and missing data of biochemical recurrence showed only small differences. Thus, we are confident that this limitation did not affect our results. Second, although we adjusted for pathologic tumor characteristics we cannot completely rule out residual confounding factors. For example, the grading system for surgical specimen changed during the study period. It seems unlikely, though, that this might have affected our results since the definition of positive surgical margins did not changed. A third limitation is that biochemical recurrence is not of direct clinical relevance to patients. However, it invariably precedes stronger oncologic endpoints such as metastasis or survival and often triggers postoperative treatments that may be associated with side effects. Moreover, being less influenced by external factors than clinical recurrence or survival outcomes, it is a reasonable endpoint for learning curve studies[10].
Our findings have implications for surgical practice. For example, empirical research should determine how more experienced robotic surgeons avoid the occurrence of positive margins. The fact that greater experience might result in more accurate dissection of the gland would not be surprising. However, other factors (i.e. anatomical landmarks) may be crucial to achieve negative margins in delicate steps of the surgical technique, especially at the apex[21]. In this contest, surgical videos may facilitate the comparison between the technique of more and less experienced surgeons in such demanding steps. With respect to research, the magnitude of the association between margin status and biochemical recurrence is currently unclear. Although our results reinforce prior evidence that such outcomes are weakly related[15], additional research is needed to draw definite conclusions on the reliability of margins status as oncologic surrogate. Moreover, future studies should examine the effect of experience on cancer control after robotic prostatectomy. It seems counterintuitive that a surgeon’s results do not improve with experience. However, the relationship between experience and the oncologic efficacy may be different for the robotic technique than other approaches. Potential reasons may include cases’ selection[22] (do patients treated robotically have better baseline characteristics?), procedure complexity[23] (is the robotic technique easier than other approaches?) and surgical education[24–26] (does robotic surgery allow better surgical training?). Accordingly, more attention should be paid to the whole process of learning in order to improve the surgical quality of robotic radical prostatectomy.
REFERENCES
- 1.Abboudi H, Khan MS, Guru KA, et al. (2013) Learning curves for urological procedures: a systematic review. BJU Int 114:617–629. doi: 10.1111/bju.12315 [DOI] [PubMed] [Google Scholar]
- 2.Hu JC, Gold KF, Pashos CL, et al. (2003) Role of Surgeon Volume in Radical Prostatectomy Outcomes. JCO 21:401–405. doi: 10.1200/JCO.2003.05.169 [DOI] [PubMed] [Google Scholar]
- 3.Sharma N, A P, D L (2011) First 500 cases of robotic-assisted laparoscopic radical prostatectomy from a single UK centre: learning curves of two surgeons. BJU Int 1–9. doi: 10.1111/j.1464-410X.2010.09941,10500,10501.x [DOI] [PubMed] [Google Scholar]
- 4.Yoon KJ, Koo HH (2015) Learning Curve of Robot-Assisted Laparoscopic Radical Prostatectomy for a Single Experienced Surgeon: Comparison with Simultaneous Laparoscopic Radical Prostatectomy. World J Mens Health 1–6. doi: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jaffe J, Castellucci S, Cathelineau X, et al. (2009) Robot-Assisted Laparoscopic Prostatectomy: A Single-Institutions Learning Curve. URL 73:127–133. doi: 10.1016/j.urology.2008.08.482 [DOI] [PubMed] [Google Scholar]
- 6.Vickers AJ, Bianco FJ, Serio AM, et al. (2007) The Surgical Learning Curve for Prostate Cancer Control After Radical Prostatectomy. JNCI Journal of the National Cancer Institute 99:1171–1177. doi: 10.1093/jnci/djm060 [DOI] [PubMed] [Google Scholar]
- 7.Vickers AJ, Savage CJ, Hruza M, et al. (2009) The surgical learning curve for laparoscopic radical prostatectomy: a retrospective cohort study. Lancet Oncology 10:475–480. doi: 10.1016/S1470-2045(09)70079-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Thompson JE, Egger S, Böhm M, et al. (2018) Superior Biochemical Recurrence and Long-term Quality-of-life Outcomes Are Achievable with Robotic Radical Prostatectomy After a Long Learning Curve—Updated Analysis of a Prospective Single-surgeon Cohort of 2206 Consecutive Cases. European Urology 73:664–671. doi: 10.1016/j.eururo.2017.11.035 [DOI] [PubMed] [Google Scholar]
- 9.Sivaraman A, Sanchez-Salas R, Prapotnich D, et al. (2017) Learning curve of minimally invasive radical prostatectomy: Comprehensive evaluation and cumulative summation analysis of oncological outcomes. Urologic Oncology: Seminars and Original Investigations 35:149.e1–149.e6. doi: 10.1016/j.urolonc.2016.10.015 [DOI] [PubMed] [Google Scholar]
- 10.Vickers A, Maschino A, Savage C (2012) Assessing the learning curve for prostate cancer surgery. Robotic Urologic Surgery [Google Scholar]
- 11.Gandaglia G, De Lorenzis E, Novara G, et al. (2017) Robot-assisted Radical Prostatectomy and Extended Pelvic Lymph Node Dissection in Patients with Locally-advanced Prostate Cancer. European Urology 71:249–256. doi: 10.1016/j.eururo.2016.05.008 [DOI] [PubMed] [Google Scholar]
- 12.Gandaglia G, Fossati N, Zaffuto E, et al. (2017) Development and Internal Validation of a Novel Model to Identify the Candidates for Extended Pelvic Lymph Node Dissection in Prostate Cancer. European Urology 72:1–9. doi: 10.1016/j.eururo.2017.03.049 [DOI] [PubMed] [Google Scholar]
- 13.Epstein J, Egevad L, Egevad, Amin MB (2015) The 2014 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma 1–9. [DOI] [PubMed] [Google Scholar]
- 14.Amin MB, Greene FL, Edge S, et al. , eds. AJCC Cancer Staging Manual (ed 8th Edition). New York: Springer; 2017. [Google Scholar]
- 15.Vickers A, Bianco F, Cronin A, et al. (2010) The Learning Curve for Surgical Margins After Open Radical Prostatectomy: Implications for Margin Status as an Oncological End Point. Journal of Urology 183:1360–1365. doi: 10.1016/j.juro.2009.12.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Thompson JE, Egger S, Böhm M, et al. (2014) 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. European Urology 65:521–531. doi: 10.1016/j.eururo.2013.10.030 [DOI] [PubMed] [Google Scholar]
- 17.Walz J, Epstein JI, Ganzer R, et al. (2016) A Critical Analysis of the Current Knowledge of Surgical Anatomy of the Prostate Related to Optimisation of Cancer Control and Preservation of Continence and Erection in Candidates for Radical Prostatectomy: An Update. European Urology 70:301–311. doi: 10.1016/j.eururo.2016.01.026 [DOI] [PubMed] [Google Scholar]
- 18.Moro FD (2018) How robotic surgery is changing our understanding of anatomy. Arab Journal of Urology 16:297–301. doi: 10.1016/j.aju.2017.10.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhang L, Bin Wu, Zha Z, et al. (2018) Surgical margin status and its impact on prostate cancer prognosis after radical prostatectomy: a meta-analysis. World Journal of Urology 1–13. doi: 10.1007/s00345-018-2333-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Novara G, Ficarra V, Mocellin S, et al. (2012) Systematic Review and Meta-analysis of Studies Reporting Oncologic Outcome After Robot-assisted Radical Prostatectomy. European Urology 62:382–404. doi: 10.1016/j.eururo.2012.05.047 [DOI] [PubMed] [Google Scholar]
- 21.Albadine R E Hyndman M, Chaux A, et al. (2012) Characteristics of positive surgical margins in robotic- assisted radical prostatectomy, open retropubic radical prostatectomy, and laparoscopic radical prostatectomy: a comparative histopathologic study from a single academic center. Human Pathology 43:254–260. doi: 10.1016/j.humpath.2011.04.029 [DOI] [PubMed] [Google Scholar]
- 22.Briganti A, Bianchi M, Sun M, et al. (2012) Impact of the introduction of a robotic training programme on prostate cancer stage migration at a single tertiary referral centre. BJU Int 111:1222–1230. doi: 10.1111/j.1464-410X.2012.11464.x [DOI] [PubMed] [Google Scholar]
- 23.Andolfi C, Umanskiy K (2017) Mastering Robotic Surgery: Where Does the Learning Curve Lead Us? Journal of Laparoendoscopic & Advanced Surgical Techniques 27:470–474. doi: 10.1089/lap.2016.0641 [DOI] [PubMed] [Google Scholar]
- 24.Vickers AJ, Sjoberg D, Basch E, et al. (2012) How Do You Know If You Are Any Good? A Surgeon Performance Feedback System for the Outcomes of Radical Prostatectomy. European Urology 61:284–289. doi: 10.1016/j.eururo.2011.10.039 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Schommer E, Patel VR, Mouraviev V, et al. (2017) Diffusion of Robotic Technology Into Urologic Practice has Led to Improved Resident Physician Robotic Skills. Journal of Surgical Education 74:55–60. doi: 10.1016/j.jsurg.2016.06.006 [DOI] [PubMed] [Google Scholar]
- 26.Guzzo TJ, Gonzalgo ML (2009) Robotic surgical training of the urologic oncologist. Urologic Oncology: Seminars and Original Investigations 27:214–217. doi: 10.1016/j.urolonc.2008.09.019 [DOI] [PubMed] [Google Scholar]