Abstract
Introduction and Objectives:
Photoselective vaporization of the prostate (PVP) is a frequently performed procedure by postgraduate trainees (PGTs). However, there is no PVP-specific objective assessment tool to evaluate the acquisition of PVP skills. The aim of the present study was to develop and validate an objective structured assessment of technical skills tool for the PVP procedure (PVP-OSATS).
Methods:
This study was conducted in two phases. Phase I included the development of PVP-OSATS and assessment of its reliability and construct validity. Panel discussion among experts led to the development of the PVP-OSATS tool with 12 parameters, each scored from 1 (worst) to 5 (best) with a maximum score of 60. Laser prostatectomy experts and PGTs from postgraduate years (PGY) 4 and 5 were recruited. Inter-rater reliability, using Cohen's and Fleiss's kappa, was calculated for all parameters. To assess for construct validity, PGTs were compared with experts. Phase II included assessment of the concurrent validity of this novel tool. This was performed by recruiting Quebec urology PGTs between PGY-3 and −5 to test their PVP skills during the semiannual objective structured clinical examination using the validated GreenLight™ simulator.
Results:
During phase I,116 intraoperative PVP-OSATS assessments were collected; 102 for PGTs and 14 for experts. Cohen's and Fleiss's kappa was adequate (k ≥ 0.6) for all 12 parameters, confirming adequate inter-rater reliability. There was significant difference between PGTs and experts in all PVP-OSATS parameters (p ≤ 0.01) except in respect to anatomical landmarks and instrument damage. During phase II, there was significant positive correlation between PVP-OSATS scores inside the operating room and global scores obtained by the GreenLight simulator (r = 0.814; p < 0.001).
Conclusion:
This study reports inter-rater reliability, construct, and concurrent validity of PVP-OSATS as a novel PVP-specific objective assessment tool.
Introduction
A
Materials and Methods
Study design
This study was conducted in two phases. Phase I included the development of PVP-OSATS and assessment of its reliability and construct validity, and phase II included assessment of the concurrent validity of this novel tool.
Phase I
PVP-OSATS: how was it devised?
The original OSATS was modified by a panel of three laser prostatectomy experts to make it PVP specific in addition to incorporating patient safety parameters. 17 Therefore, the modified OSATS was labeled PVP-OSATS and it consisted of 12 parameters, each being scored on a Likert scale from 1 to 5, where 1 is the worst and 5 is the best, bringing the maximal score to 60 (Table 1). Parameters included were based on literature review of PVP outcomes, expert opinion, and parameters used by the GL-SIM to obtain the global score. 10,13,18 –21 The first parameter is safety consciousness. It is important for PGTs to be aware of the hazards of the 532 nm laser and what would happen when there is fiber damage. Recent studies have reported serious visual complications ranging from corneal abrasions to total blindness from 532 nm lasers used without appropriate eye protection. 22,23 Therefore, it was agreed that PGTs should follow all safety precautions, including ensuring that appropriate eye protection is worn by all OR personnel, including the patient. In addition, the 532 nm laser should not be activated until all operating doors have been closed. Since this parameter is important, video recording of endoscopic images could not be used to assess using the PVP-OSATS tool. Therefore, evaluators needed to be present at the time of the PVP procedure. The second parameter is instrument handling. The panel agreed that PGTs should be familiar with all instruments used in the procedure. The third parameter is the use of appropriate energy levels at the bladder neck (BN) and the external urinary sphincter. The energy level should not exceed 80 W at these areas. The fourth parameter is in respect to anatomical landmarks. PGTs should clearly identify the verumontanum, BN, ureteral orifices, and the two lateral lobes. In addition, the procedure should start with a transverse mark in front of the verumontanum using coagulation settings to mark the distal limit of the procedure. The fifth parameter is the knowledge of the specific procedure. It is expected that PGTs performing the PVP procedure should know all aspects of this procedure, dealing gently with tissues, and avoiding shifting from one point to another except after achieving the same depth of vaporization. Distance from tissue is the sixth parameter. PGTs should not touch the tissue during the PVP procedure. Ideally, they should keep the fiber at 0.5 to 2.5 mm from the tissue. The seventh parameter is the speed of rotation, which should be neither too fast nor too slow. Ideally, it should rotate at a speed of 3 mm/second throughout vaporization. The eighth parameter is efficiency of vaporization, which is usually evident by the amount of air bubbles created during vaporization. Control of bleeding is vital throughout the procedures. Therefore, this constitutes the ninth parameter. The next is the flow of operation and forward planning. PGTs should have a plan throughout the procedure and it is not acceptable to proceed haphazardly. Economy of the operative time is one of the most important parameters during PVP where the lasing time should correlate appropriately with the operative time, optimally from 66% to 80%, and this could not be achieved without economy of motion. Finally, the fiber should be kept in good condition throughout the procedure as long as the PGTs follow the rules of the PVP technique in terms of maintaining optimal distance from tissue and efficiency of vaporization. These PVP-OSATS parameters were assessed subjectively by raters using experts perform PVP procedure both on the GL-SIM and in the OR as reference.
Inter-rater reliability and construct validity
McGill University Urology PGTs between postgraduate years (PGY-4 and −5) were recruited to perform PVP under direct supervision of a PVP expert. PGTs were evaluated either by two or three raters, including one laser prostatectomy expert. Inter-rater reliability of all PVP-OSATS parameters was calculated. The chance-adjusted measure of agreement was calculated using the “Cohen's kappa” for two raters and the “Fleiss's kappa” for three raters. In addition, two PVP experts were assessed using the same PVP-OSATS tool. Comparisons between experts and PGTs were performed to establish construct validity of the PVP-OSATS tool.
Phase II
Concurrent validity
Concurrent validity of the PVP-OSATS tool was tested by recruiting Quebec urology PGTs to perform the PVP procedure on the right lobe of a 30 g prostate on the GreenLight SIM during a semiannual urology objective structured clinical examination (OSCE) that took place on December 5, 2015. The OSCE consisted of 11 stations, each lasting 15 minutes. One of the stations was a simulation station to assess PVP skills of PGTs using the GreenLight (GL-SIM) developed by Dr. Robert Sweet (American Medical Systems). 13 Performance reports generated by the GL-SIM were collected. The concurrent validity of the PVP-OSATS tool was assessed by correlating PVP-OSATS scores and global score from the GL-SIM.
Statistical analysis
The chance-adjusted measure of agreement was calculated using the “Cohen's kappa” for two raters and the “Fleiss's kappa” for three raters. The online kappa calculator was used for calculation of inter-rater agreements at
Results
Inter-rater reliability and construct validity
During the period between December 2014 and October 2015, 116 PVP-OSATS evaluations were collected. All procedures were performed by six PGTs from PGY-4 and −5 under direct supervision of two laser prostatectomy experts. There were 54 evaluations by three evaluators and 62 evaluations by two evaluators, with the laser prostatectomy expert being one of these evaluators. Of the 34 PVP cases performed, 17.6% of patients were diabetic, 29.4% were on alpha blockers, and 70.6% were on combined alpha blockers and 5-alpha reductase inhibitors. In addition, 5.9% of patients were on antiplatelet therapy, which was continued perioperatively. The median patient age was 79 years (IQR: 68.5–83), with the median American Society of Anesthesiologists (ASA) score of 2 (IQR: 1.5–3). The preoperative median prostate size estimated on transrectal ultrasound was 31 cc (IQR: 25–47) with median prostate specific antigen of 1.8 ng/mL (IQR: 1.3–3.1). Preoperative flowmetric assessment revealed a median maximum flow rate (Q-max) of 7 mL/second (IQR: 4.8–10.3), median postvoid residual (PVR) of 221 cc (IQR: 82.5–278.5), median International Prostate Symptom Score (IPSS) of 18 (IQR: 14–27), and median quality-of-life (QoL) score of 3 (IQR: 3–5.5). Indications for surgery were severe LUTS (70.6%) and refractory urinary retention (29.4%). The median operative time was 60 minutes (IQR: 35–78) with median lasing time of 30 minutes (IQR: 16–47) and median total energy use of 147 kJ (IQR: 84–260). All procedures were effectively completed without intraoperative complications and without the need for blood transfusion. Only one case required auxiliary procedure in the form of TURP at the end of the procedure for control of bleeding. All patients were admitted overnight and had trial of void on the following morning. The rate of de novo stress urinary incontinence within the first postoperative month was 5.9%. When compared with preoperative values, there were significant improvements in the lower urinary tract parameters in terms of the Q-max (7.5 ± 2.9 vs 20.7 ± 13.9; p = 0.003), PVR (218 ± 157 vs 89 ± 130; p = 0.01), IPSS (20.3 ± 5.9 vs 7.5 ± 4.0; p = 0.001), and QoL (4 ± 1.4 vs 1.3 ± 1.5; p = 0.001) within the first month postoperatively. Cohen's and Fleiss's kappa calculations showed adequate inter-rater reliability for all 12 PVP-OSATS parameters (k ≥ 0.6) (Table 2). Furthermore, there was significant difference between PGTs and experts in all PVP-OSATS parameters (p < 0.01), except in respect to anatomical landmarks (p = 0.95) and instrument damage (p = 0.42) (Table 3).
CI = confidence interval; K = kappa or Fleiss coefficient; PGTs = postgraduate trainees; PVP-OSATS = photoselective vaporization of the prostate–objective structured assessment of technical skills; SE = standard error.
Concurrent validity
Twenty-nine PGTs from all four urology training programs in the province of Quebec participated in the study during the OSCE. There were 12 (41.4%) from McGill University, 9 (31%) from University of Montreal, 6 (20.7%) from Laval University, and 2 (6.9%) from Sherbrooke University. Furthermore, 11 participants (38%) were from PGY-3 and 9 (31%) from each of PGY-4 and -5 years.
Participants had performed an average of 5.3 ± 1.3 (0–30) PVP cases before this OSCE. During a 15-minute station, they performed the PVP procedure of the right lobe of a 30-g normal prostate on the GL-SIM with a mean global score of 146.7 ± 10.1 and mean PVP-OSATS score of 41.8 ± 1.6. There was significant correlation between the PVP-OSATS scores and the global scores generated by the validated GL-SIM, thus confirming concurrent validity of the PVP-OSATS tool (r = 0.814; p < 0.001) (Fig. 1).

Correlation between GL-SIM global score and PVP-OSATS score. GL-SIM, GreenLight simulator; PVP-OSATS, photoselective vaporization of the prostate–objective structured assessment of technical skills.
Discussion
Competency-based medical education (CBME) has become the basis for postgraduate surgical education in different training frameworks in the developed world. 12,24 –26 In addition to providing a safe environment for PGTs to practice, high-fidelity VR simulators objectively assess competency of PGTs. For example, the GL-SIM offers training for the PVP procedure without the need to wear eye protection. Therefore, trainees could acquire the early part of their PVP learning curve without exposing themselves to the risk of 532 nm lasers, which has been reported to cause blindness if appropriate eye protection is not worn. 22 The GL-SIM has been recently used for training and assessment of PVP procedure. 14 –16 However, the main concern is that whether skills obtained through practicing on the GL-SIM are transferable to the OR with live PVP cases. To asses this, task-specific objective assessment tools are needed to assess performance on both the GL-SIM and during live PVP cases in the OR. In the present study, the newly developed PVP-OSATS showed adequate inter-rater reliability. As shown in the Results section, Cohen's and Fleiss's kappa was adequate for all PVP-OSATS parameters (k ≥ 0.6), which means that PVP-OSATS has a good inter-rater reliability even among three raters. Furthermore, the significant difference between laser prostatectomy experts and PGTs meant that PVP-OSATS was able to differentiate between PGTs and experts (46.4 vs 58.7; p < 0.001), thus establishing the construct validity of this novel PVP-OSATS tool (Table 3). However, there was no significant difference between experts and PGTs in terms of respect to anatomical landmarks and instrument damage. This could be due to several reasons. First is the meticulous supervision by a laser prostatectomy expert throughout each PVP procedure. Second, the GL-SIM is available in our institution where PGTs have the chance to practice before performing the PVP procedure in the OR. Third, the PVP procedure was performed by senior PGTs from PGY-4 and −5. Therefore, these senior PGTs had similar scores with experts in respect to anatomical landmarks and instrument damage parameters (p > 0.05). It is possible that if more junior PGTs were to perform the PVP procedure, a significant difference may be seen. However, for patient safety reasons, it would not be ethical to ask medical students to perform the PVP procedure as novices.
Concurrent validity was established by demonstrating a significant correlation between global scores on the GL-SIM and the PVP-OSATS scores (r = 0.814; p < 0.001) (Fig. 1). Previously, global scores obtained from the GL-SIM significantly correlated with the number of previously performed PVPs. 16 This means that the newly developed PVP-OSATS tool could be used in the future to assess PGTs both on the GL-SIM and inside the OR to assess transfer of PVP skills from the GL-SIM to the OR.
Limitations of the current study could be addressed in the following points. First, one of the three raters was not available for a period of time, therefore, there were some evaluations performed by two and others by three raters. However, this gave the opportunity to calculate both the Cohen's and Fleiss's kappa. Second is the halo effect, which is a cognitive bias on which an evaluator's overall impression of a PGT influences his/her feelings and thoughts about that PGTs performance. 27 Since evaluators were not blinded to PGTs nor to their level of training, it may have introduced bias. Blinded endoscopic images were not used in the present study since it would not have been possible to assess patient safety parameters of the PVP-OSATS tool. Third, although this study included 116 evaluations, there was a small sample size of experts (2) and PGTs (6) in the first phase of the study. Despite including all urology PGTs from PGY-3 to −5 in the province of Quebec, there were 29 PGTs during the second phase of the study. Finally, OSATS 17 was used as a template rather than the O-SCORE 28 to create the PVP-OSATS tool. Whereas the O-SCORE is a tool to assess the independent performance of a procedure, the PVP-OSATS is a PVP-specific tool taking into account patient safety issues. Therefore, one of the limitations of PVP-OSATS is that it lacks assessment of independent performance of the PVP procedure. In the future, the PVP-OSATS tool could be modified to include assessment of independent performance. Nonetheless, this is the first study to develop and validate a new PVP-specific objective structured assessment tool. Since PVP is a frequently performed procedure by PGTs in urology training programs, the newly devised PVP-OSATS could be used by program directors to assess the competency of PGTs in performing the PVP procedure safely. These results were recently presented at a podium session of the American Urological Association meeting. 29
Conclusions
This study reports the inter-rater reliability, construct, and concurrent validity of the PVP-OSATS tool as a novel PVP-specific objective assessment tool. Future studies are needed to externally validate the PVP-OSATS and use it to assess the learning curve of PVP and compare the performance of PGTs on both the GL-SIM and in the OR.
Ethical statement
This study was conducted in concordance with the Declaration of Helsinki 2013 and after obtaining McGill University ethics approval (No. A07-E59-14B) and informed consent.
Footnotes
Acknowledgments
This work was partially sponsored by Fonds de la Recherche Santé du Québec (FRSQ) Chercheur Boursier grant to Dr. Sero Andonian and by a grant from the Urology Care Foundation Research Scholars Program and the Boston Scientific Corporation, The Endourological Society, and the “Friends of Joe” to Dr. Y.A.N. and the Canadian Urological Association Scholarship Foundation—Société Internationale d'Urologie (CUASF-SIU) International Scholarship grant to Dr. Y.A.N.
Author Disclosure Statement
No financial interests with American Medical Systems. The GreenLight simulator was lent gratuitously by American Medical Systems.
