Education Laparoscopic Training on Bench Models: Better and More Cost Effective than Operating Room Experience? Daniel J Scott, MD, Patricia C Bergen, MD, FACS, Robert V Rege, MD, FACS, Royce Laycock, MD, FACS, Seifu T Tesfay, RN, R James Valentine, MD, FACS, David M Euhus, MD, FACS, D Rohan Jeyarajah, MD, William M Thompson, MD, FACS, Daniel B Jones, MD, FACS provement was adjusted for differences in baseline performance. Background: Developing technical skill is essential to surgical training, but using the operating room for basic skill acquisition may be inefficient and expensive, especially for laparoscopic operations. This study determines if laparoscopic skills training using simulated tasks on a video-trainer improves the operative performance of surgery residents. Results: Five residents were unable to participate because of scheduling problems; 9 residents in the training group and 13 residents in the control group completed the study. Baseline laparoscopic experience, video-trainer scores, and global assessments were not significantly different between the two groups. The training group on average practiced the video-trainer tasks 138 times (range 94 to 171 times); the control group did not practice any task. The trained group achieved significantly greater adjusted improvement in video-trainer scores (five of five tasks) and global assessments (four of eight criteria) over the course of the four-week curriculum, compared with controls. Study Design: Second- and third-year residents (nⴝ 27) were prospectively randomized to receive formal laparoscopic skills training or to a control group. At baseline, residents had a validated global assessment of their ability to perform a laparoscopic cholecystectomy based on direct observation by three evaluators who were blinded to the residents’ randomization status. Residents were also tested on five standardized videotrainer tasks. The training group practiced the videotrainer tasks as a group for 30 minutes daily for 10 days. The control group received no formal training. All residents repeated the video-trainer test and underwent a second global assessment by the same three blinded evaluators at the end of the 1-month rotation. Within-person improvement was determined; im- Conclusions: Intense training improves video-eyehand skills and translates into improved operative performance for junior surgery residents. Surgical curricula should contain laparoscopic skills training. (J Am Coll Surg 2000;191:272–283. © 2000 by the American College of Surgeons) Developing technical skills is essential to surgical resident training. William Halsted1,2 introduced the surgical residency system in the United States almost a century ago, whereby residents learn in the operating room through graded responsibility under direct supervision. Teaching residents in the operating room is effective but may be inefficient, costly, and may increase patient morbidity.3,4 Managed care has placed increasing financial constraints on hospital and physician reimbursements. With more pressure on physicians to maximize efficiency, faculty may have less time available for teaching. No competing interests declared. Funding was provided by the Southwestern Center for Minimally Invasive Surgery as supported in part by an educational grant from United States Surgical—A Division of Tyco Healthcare Group. The videotrainer was provided by Karl Storz Endoscopy. Presented, in part, at the American College of Surgeons 85th Annual Clinical Congress, Surgical Forum, San Francisco, CA, October 13, 1999. Received September 14, 1999; Revised January 12, 2000; Accepted April 3, 2000. From the Department of Surgery, University of Texas Southwestern Medical Center, Dallas, TX. Correspondence address: Daniel B Jones, MD, Department of Surgery, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75235-9092. © 2000 by the American College of Surgeons Published by Elsevier Science Inc. 272 ISSN 1072-7515/00/$21.00 PII S1072-7515(00)00339-2 Vol. 191, No. 3, September 2000 Scott et al Financial constraints may further compromise the availability of operating room time for teaching purposes. The cost of using operating room time for training surgical residents in the United States is an estimated $53 million per year.3 Laparoscopy has emerged as a very useful surgical modality but complicates the problems of teaching residents in the operating room.5,6 Laparoscopy poses a new obstacle to skill acquisition because significant experience is required before competency is achieved.7-10 Depth perception is altered by a two-dimensional video imaging system and new cues must be learned before spatial relationships can be reliably established. Long instruments diminish tactile feedback and can be awkward to use. Range of motion is limited by trocars, and video-eye-hand coordination must be developed to correctly position instruments in the operative field. Teaching junior residents basic laparoscopic skills in the operating room can be frustrating for both the attending and resident surgeons, as well as time consuming and inefficient.11 Several options exist for teaching surgical skill outside the operating room. Cadavers offer a high degree of fidelity to the living patient and a nonpressured learning atmosphere. But cadavers are costly, of limited availability, and have noncompliant tissue that may be difficult to use for operations.11 Live animal models may also be useful.6 But animals differ in anatomy from humans, can be costly, require appropriate facilities and personnel, and raise ethical concerns.12 Inanimate models have recently become popular, especially for laparoscopic training in residency programs.4,7,9,13-18 They are safe, reproducible, readily available, offer unlimited practice, and require no supervision. Compared with cadavers and animals, they are more cost effective.4,9,11 But bench models may not realistically mimic human anatomy and living tissue. Even so, bench models may be the best option for training residents outside of the operating room. Although training programs using inanimate models have become increasingly widespread, to date there is no evidence that such programs provide residents with skills that are transferable to the operating room.11,19 The purpose of this study was to develop a model to provide intense laparoscopic skills training to residents on surgical rotations and to determine if improvement of skill level on a video-trainer Laparoscopic Training on Bench Models 273 translates into an improvement in operative performance. METHODS A total of 27 second- and third-year surgery residents rotating for one-month periods on the general surgery services at Parkland Memorial Hospital were available to participate in the study from August 1998 through January 1999. A nonparametric power analysis20 was performed to ascertain if a meaningful training effect could be detected with a high probability using a sample of this size. It was determined that this sample would provide a power of at least 0.8 with a type I error of 0.05 if the equivalent effect size21 equalled or exceeded unity. Residents gave informed consent under a protocol approved by the University of Texas Southwestern Medical Center Institutional Review Board. All patients gave consent for photographs to be taken (operations were videotaped for investigations outside the scope of this study). Residents were randomized (Fig. 1) at the beginning of the study to either a training or a control group using a random digits table. The original randomization scheme included 13 residents assigned to the training group and 14 residents assigned to the control group. During the first week of their rotation, all residents were asked to complete a baseline questionnaire regarding earlier laparoscopic experience and competency in laparoscopic skills. Global assessments (Table 1) were used to measure baseline operative performance as each resident performed a laparoscopic cholecystectomy in the operating room. Patients with the diagnosis of symptomatic cholelithiasis for whom an elective cholecystectomy was indicated were scheduled for the observed cases. All operations were supervised by one of three designated faculty surgeons who were blinded to the training status of the residents. The designated faculty member served as first assistant during the entire operation. Residents performed the laparoscopic cholecystectomy in a one-handed or twohanded fashion, according to the faculty surgeon’s preference. The assistants were instructed to allow the resident to perform the operation with as much independence as possible, while assuring patient safety. The resident was to make key decisions re- 274 Scott et al Laparoscopic Training on Bench Models J Am Coll Surg Figure 1. Algorithm for testing and randomization. Second- and third-year surgery residents were randomized to a training or a control group and underwent testing at the beginning and end of the rotation. garding the sequence of dissection and was told to direct the assistant to provide adequate retraction. The assistants were instructed to quiz the resident on the locations of key anatomic landmarks, such as the cystic-common bile duct junction, the borders of the triangle of Calot, and the cystic artery. The assistants were also instructed to quiz the resident on key points of the procedure and to ask “what step Vol. 191, No. 3, September 2000 Scott et al Laparoscopic Training on Bench Models 275 Table 1. Global Rating Scale of Operative Performance* Performance Characteristic Scale 1 Respect for tissue Frequently used unnecessary force on tissue or caused damage by inappropriate use of instruments Time and motion Many unnecessary moves Instrument handling Repeatedly makes tentative awkward or inappropriate moves with instruments Knowledge of instruments Frequently asked for wrong instrument or used inappropriate instrument Flow of operation Frequently stopped operating and seemed unsure of next move Use of assistants Failed to use assistants Knowledge of specific procedure Overall performance Required specific instruction at most steps Unable to perform operation independently 2 3 4 5 Careful handling of tissue but occasionally caused inadvertent damage Consistently handled tissues appropriately with minimal damage Efficient time/motion but some unnecessary moves Competent use of instruments but occasionally stiff or awkward Knew names of most instruments and used appropriate tool for task Demonstrated some forward planning with reasonable progression of procedure Appropriate use of assistants most of the time Knew all important steps of the operation Clear economy of movement and maximum efficiency Fluid moves with instruments and no awkwardness Competent, could perform operation with minimal teaching assistance Clearly superior, able to perform operation independently with confidence Obviously familiar with the instruments and their names Planned course of operation effortless from one move to the next Strategically used assistants to the best advantage at all times Familiar with all aspects of the operation *Modified from Reznick and colleagues.22 1 ⫽ worst possible score, 5 ⫽ best possible score. is next,” so that residents would vocalize their operative plan. In this way, the information the evaluators could use for global assessments was maximized. Residents were briefed on the nature of the experiment but were not knowledgeable regarding the specific content of the global assessment. Confidentiality of the results was guaranteed. Global assessments were performed by three additional faculty surgeons who were independent observers and did not participate in the operation. The evaluators were also blinded to the training status of the resident. The evaluators were present in the operating room during the key parts of the case and rated the resident according to eight criteria, each relating to some aspect of operative performance. The eight performance criteria included “respect for tissue,” “time and motion,” “instrument handling,” “knowledge of instruments,” “flow of operation,” “use of assistants,” “knowledge of specific procedure,” and “overall performance.” Each area of performance was rated on a scale of 1 (worst) to 5 (best), with the middle and extreme endpoints of the scale anchored by explicit descriptors, as described by Reznick and colleagues.11,22,23 We based our curriculum on five established laparoscopic drills suitable for novice surgeons that could be performed on a video-trainer (Fig. 2).7,13,15,24 The five tasks included Checkerboard (Fig. 3), Bean Drop (Fig. 4), Running String (Fig. 5), Block Move (Fig. 6), and Suture Foam (Fig. 7). The Checkerboard drill involves arranging 16 metal letters and numbers in the appropriate squares on a flat surface. The Bean Drop drill consists of individually grasping five beans and moving the beans 15cm to place them in a 1-cm hole at the top of an elevated cup. The dominant hand is used to grasp the beans while the nondominant hand moves the laparoscope to provide adequate visualization during the procedure. The Running String drill mimics running bowel; two graspers are used to run a 140-cm string from one end to the other, grasping the string only at colored sections marked at 12-cm intervals. The Block Move drill consists of individ- 276 Scott et al Laparoscopic Training on Bench Models Figure 2. Southwestern Center for Minimally Invasive Surgery Guided Endoscopic Module (GEM). This six-station videotrainer (Karl Storz Endoscopy, Culver City, CA) was used for training residents. ually lifting four blocks using a curved needle (held in a grasper) to hook a metal loop on the top of each block. The dominant hand manipulates the grasper to move the blocks 15cm and to lower them onto a designated space on a flat surface. The nondominant hand moves the laparoscope to provide adequate visualization during the procedure. The Suture Foam drill consists of using an Endostitch device (United States Surgical Corporation, Norwalk, CT) to suture two foam organs together and tie a single intracorporeal square knot. During week 1, all residents were tested on the five video-trainer tasks. No resident had earlier exposure to video-trainer drills. All residents were briefed on the nature of the testing and each task was demonstrated once. No practice was allowed before testing except for the Suture Foam drill, for which a single practice was allowed so that the resident could become familiar with the mechanical workings of the device. Each task was set up at only one station. During testing, each resident performed each of the five tasks three times. In order to allow up to five residents to be tested simultaneously, residents were instructed to start at any unoccupied station and to perform three repetitions J Am Coll Surg Figure 3. The Checkerboard drill consists of arranging 16 metal letters and numbers in the appropriate squares on a flat surface. Modified from Jones and colleagues.7 Figure 4. The Bean Drop drill consists of individually grasping five beans and moving the beans 15 cm to place them in a 1-cm hole at the top of an elevated cup. The dominant hand is used to grasp the beans while the nondominant hand moves the laparoscope to provide adequate visualization during the procedure. Modified from Rosser and associates.15 Vol. 191, No. 3, September 2000 Scott et al Laparoscopic Training on Bench Models 277 Figure 5. The Running String drill mimics running bowel; two graspers are used to run a 140-cm string from one end to the other, grasping the string only at colored sections marked at 12-cm intervals. Modified from Rosser and associates.15 of the task set up at that station. When finished, residents were instructed to rotate to another unoccupied station. If more than one station was unoccupied, the resident could choose which station to go to next. Scores were recorded as the average time necessary for task completion. After baseline testing was completed, the residents were told to which group they had been randomized. In the second and third weeks of their rotation, residents randomized to training met as a group for at least 30 minutes daily for 10 days. All residents were excused from clinical duties to attend training sessions, which were held at 7:00 AM, Monday through Friday, for 2 weeks. During this structured training time, residents practiced the video-trainer tasks; the choice of which tasks to practice was left up to the resident, but residents were encouraged to practice all five tasks during each session. Addi- Figure 6. The Block Move drill consists of individually lifting four blocks using a curved needle fixed to a grasper in the dominant hand; the blocks are then moved 15 cm and lowered onto a designated space on a flat surface. The nondominant hand moves the laparoscope to provide adequate visualization during the procedure. Modified from Rosser and associates.15 Figure 7. The Suture Foam drill consists of using an Endostitch device (United States Surgical Corporation, Norwalk, CT) to suture two foam organs together and tie a single intracorporeal knot. 278 Scott et al Laparoscopic Training on Bench Models J Am Coll Surg Table 2. Baseline Video-trainer Scores: Time (seconds) for Task Completion Task Checkerboard Bean Drop Running String Block Move Suture Foam Control (nⴝ13) Trained (nⴝ9) p Value* 144 (122–152) 56 (48–69) 62 (46–74) 50 (38–55) 58 (49–90) 146 (126–187) 53 (45–66) 74 (67–84) 48 (39–58) 56 (42–94) 0.616 0.471 0.096 0.815 0.695 Values are medians with 25th–75th percentiles in parentheses. *Trained versus control groups, two-tailed Wilcoxon rank-sum test. tional demonstration of tasks was rarely given and only on specific request by the resident. Residents randomized to the control group received no structured skills training outside of the operating room and were not allowed access to the video-trainer during the study. Didactic lectures on how to perform a laparoscopic cholecystectomy were not given to either group during the study interval. In the fourth week, all residents were again tested on the five video-trainer tasks. Each task was performed three times and the average time for completion of each task was recorded. All residents performed a second laparoscopic cholecystectomy in the operating room with the same first assistant surgeon, who was blinded to the resident’s training status. The same three independent faculty evaluators performed global assessments based on direct observation; the evaluators were also blinded to the resident’s training status. At the completion of the rotation, all residents were asked to complete questionnaires regarding their laparoscopic experience; those who underwent training were asked if they perceived improvement in their laparoscopic abilities. After their rotation, residents randomized to the control group were offered, for their own educational benefit, the same training that the training group received; data from repeat trainees were not included as part of this study. Questionnaire data regarding comfort with laparoscopic surgery were analyzed using Fisher’s exact test. Baseline performance was analyzed by comparing video-trainer and global assessment scores for the control and trained groups. To test the hypothesis that there was no difference between the control and trained groups at baseline testing, a two-tailed Wilcoxon rank-sum test was used. To determine if training was beneficial, withinperson changes in performance were compared for the control and trained groups. Improvement was defined as the difference in performance at the preand posttesting intervals. Improvement was determined for both video-trainer and global assessment scores. Because the amount of improvement varied with baseline performance, a linear covariance adjustment was used to compensate for differences in baseline scores. The covariance-adjusted improvements for residents in the control and trained groups were compared using a Wilcoxon rank-sum test. To test the hypothesis that the trained group achieved greater adjusted improvement than the control group, a one-tailed test was used. Tests were considered significant at pⱕ0.05. RESULTS Over the 6-month course of the study, one resident in the control group and four residents in the trained group were unable to participate because of scheduling problems, caused by vacations and changes in rotation schedules. So, 22 residents completed the study, with 13 in the control group and 9 in the trained group. There was no crossover between groups, except for 6 of 13 control group residents who subsequently underwent training for their own benefit. Data from repeat trainees were not included in this study. Faculty assistant surgeons and evaluators were interviewed to verify blinding. All of the assistant surgeons and evaluators reported being blinded to the randomization status of the residents 100% of the time during the study. On the baseline questionnaire, both groups reported comparable laparoscopic experience; the mean number of cases per resident as surgeon or first assistant was 18 for control versus 15 for trained (p⫽0.501). Baseline scores for the video-trainer test and global assessment during laparoscopic cholecystec- Vol. 191, No. 3, September 2000 Scott et al Laparoscopic Training on Bench Models 279 Table 3. Baseline Laparoscopic Cholecystectomy Global Assessment Scores Assessment area Respect for tissue Time and motion Instrument handling Knowledge of instruments Flow of operation Use of assistants Knowledge of specific procedure Overall performance Control (nⴝ13) Trained (nⴝ9) p Value* 3.0 (2.4–3.2) 2.5 (2.0–3.0) 3.0 (2.5–3.0) 3.0 (2.5–3.5) 3.0 (2.0–3.2) 2.0 (1.5–3.0) 3.0 (2.5–3.6) 3.0 (2.3–3.0) 3.0 (2.8–3.2) 3.0 (2.0–3.0) 3.0 (2.0–3.2) 3.0 (2.3–3.3) 2.0 (2.0–3.2) 2.0 (1.5–2.8) 3.0 (2.5–3.2) 2.5 (2.0–3.0) 0.445 0.781 0.782 0.433 0.356 0.785 0.632 0.581 Values are medians with 25th–75th percentiles in parentheses. 1 ⫽ worst possible score, 5 ⫽ best possible score. *Trained versus control groups, two-tailed Wilcoxon rank-sum test. tomy are listed in Tables 2 and 3. There were no significant differences between the trained and control groups. All nine residents in the training group completed 10 practice sessions lasting 30 minutes. On average, trained residents practiced 138 videotrainer tasks (range 94 to 171 tasks). Each of the five tasks was practiced 28 times (range 19 to 34 times). Residents randomized to the control group did not practice any video-trainer task. Adjusted improvements in performance are listed in Tables 4 and 5. The trained group had significantly larger median time reductions for all five video-trainer tasks compared with the control group. On global assessment, the trained group realized significantly larger median increases in four of eight performance criteria, compared with the control group. Laparoscopic experience in the operating room during the study interval was comparable for both groups; the mean number of cases per resident as surgeon or first assistant was five for both the control and trained groups (p⫽0.612). When initially asked if they felt comfortable with their laparoscopic skills, 3 of 13 control residents and 5 of 9 trained residents replied “yes.” On the completion questionnaire, 6 of 13 control residents and 8 of 9 trained residents felt comfortable with their laparoscopic skills at the end of the rotation. Of those individuals who were not comfortable with their laparoscopic skills at baseline, 3 of 10 in the control group were comfortable at the end of the rotation, in contrast to 3 of 4 in the trained group (p⫽0.175). After training, nine of nine residents believed that the video-trainer practice had improved their video-eye-hand coordination and eight of nine felt that the training had improved their skills in the operating room. DISCUSSION Although cognitive knowledge is paramount, acquisition of technical skill plays a pivotal role in surgical education. Teaching surgical residents outside of the operating room has become increasingly popular, mainly because of financial constraints on teaching in the operating room. No consensus exists as to what type of training is appropriate and how much training is necessary to effectively impact operative performance. In this study, we developed a technical skills curriculum for laparoscopic surgery. Our curricu- Table 4. Adjusted Improvement* in Video-trainer Scores: Time (seconds) for Task Completion Task Checkerboard Bean Drop Running String Block Move Suture Foam Control (nⴝ13) Trained (nⴝ9) p Value† 31 (⫺1–37) 14 (10–18) 3 (⫺13–16) 9 (⫺2–14) 26 (18–38) 63 (23–75) 24 (18–26) 26 (21–38) 22 (11–25) 48 (44–50) 0.014 0.002 0.001 0.015 0.001 Values are medians with 25th–75th percentiles in parentheses. *Improvement defined as baseline minus posttraining score, calculated individually for each resident, adjusted by linear analysis of covariance for differences in baseline scores. † Trained versus control groups, one-tailed Wilcoxon rank-sum test. 280 Scott et al Laparoscopic Training on Bench Models J Am Coll Surg Table 5. Adjusted Improvement* in Laparoscopic Cholecystectomy Global Assessment Scores Assessment area Control (nⴝ13) Trained (nⴝ9) p Value† Respect for tissue Time and motion Instrument handling Knowledge of instruments Flow of operation Use of assistants Knowledge of specific procedure Overall performance 0.1 (⫺0.6–0.5) ⫺0.3 (⫺0.5–0.6) 0.3 (⫺0.4–0.3) 0.4 (0.0–1.0) 0.4 (⫺0.5–1.2) 0.7 (⫺0.4–1.0) 0.4 (0.0–1.1) 0.2 (⫺0.5–0.6) 0.3 (0.3–0.5) 0.3 (0.1–0.8) 0.6 (0.4–0.8) 0.6 (0.5–1.5) 1.0 (0.6–1.2) 1.0 (0.8–1.6) 1.0 (0.4–1.3) 0.7 (0.6–1.0) 0.035 0.075 0.005 0.058 0.090 0.035 0.100 0.007 Values are medians with 25th–75th percentiles in parentheses. *Improvement defined as posttraining minus baseline score, calculated individually for each resident, adjusted by linear analysis of covariance for differences in baseline scores. † Trained versus control groups, one-tailed Wilcoxon rank-sum test. lum was aimed at teaching basic skills to junior surgery residents when they were beginning to perform laparoscopic operations. We chose a curriculum based on five established laparoscopic drills that foster the development of video-eye-hand coordination.7,13,15,24 The Checkerboard drill is designed to develop spatial relationships on a planar surface and to facilitate accurate motor skills. The Bean Drop and Block Move drills both require using the nondominant hand to move the laparoscope to provide adequate visualization during the procedure; so, two-handed video-eye-hand coordination is developed. The Bean Drop drill requires fine motor skill to accurately grasp the beans. The Block Move drill requires supination and pronation to hook the metal ring with the curved needle during block lifting and releasing; depth perception and wrist articulation skills are developed. The Running String drill requires two-handed coordination and develops spatial relationships along a linear structure. The Suture Foam drill requires manual dexterity to manipulate the Endostitch device and develops two-handed coordination during suturing of the foam organs. Depth perception is also critical to successful knot tying. Compared with conventional intracorporeal suturing, using the Endostitch device is faster and preferred by surgery residents.24 Whereas conventional laparoscopic suturing is an advanced skill suitable for senior residents, the Endostitch device may be more suitable for novice surgeons. Each of the video-trainer drills could be completed quickly (in less than 3 minutes) and multiple repetitions of each task were possible during the 30-minute training sessions. Although each drill re- quired a specific subset of technical abilities, all of the drills focused on the development of video-eyehand coordination. A multitask curriculum with multiple repetitions was chosen to facilitate skill acquisition. We hypothesized that if residents mastered basic laparoscopic skills, they would be better prepared to perform actual operations. We tailored the course schedule to train residents while they were on a four-week surgical rotation. Training was conducted in group sessions on a free-standing six-station video-trainer developed by the Southwestern Center for Minimally Invasive Surgery at the University of Texas Southwestern Medical Center (Fig. 2). Multistation videotrainers have never previously been available. The advantages of group-session training may be enhanced esprit de corps, as evidence by all nine residents in the trained group completing all 10 training sessions. A healthy sense of competition may develop when residents train side by side, enhancing motivation to improve their skills. Our data indicate that the trained and control groups were not significantly different at baseline according to self-reported laparoscopic experience, video-trainer scores, and global assessments. Additionally, there was no difference in laparoscopic operative experience during the study interval between the two groups. So, the measured difference in the final skill level between the two groups can be attributed to the training received in the skills laboratory. The difference between individual baseline and posttest scores was used to assess the effect of training. An individual who performed poorly on the baseline test had a greater opportunity to improve Vol. 191, No. 3, September 2000 Scott et al than did a person who did well on the initial assessment. The statistical technique of covariance analysis25 provided a means of compensating for individual differences in baseline test scores. The net effect of this analysis was to adjust each individual’s improvement score to an equivalent score, as if each baseline test score had been equal to the overall baseline test average. Using the covariance adjustment, we detected improvement in video-trainer skills for both groups over the course of the 4-week study period (Table 4). Improvement in the control group was expected because these residents were exposed to each task three times during the initial testing session and because they were undergoing “on-the-job” training while performing operative cases on their surgical rotation. The trained group, however, achieved significantly larger median time reductions for all five video-trainer tasks. Similarly, improvement was noted on global assessments for both groups. But trained residents achieved greater median improvement in the operating room compared with control residents (Table 5). The differences in global assessment improvement for the control versus trained residents had observed significances that did not exceed 0.1 for all eight criteria, and were less than 0.05 for four of eight criteria. The difference in “overall performance” improvement for trained and control residents was significant at p⫽0.007. Our conclusion from this data is that training worked. Not only did training improve performance on the video-trainer, but more importantly training improved performance in the operating room. The questionnaire data indicate that junior level residents believe they need additional laparoscopic training. Only 36% of the residents reported feeling comfortable with their skill level at the beginning of the rotation. Of those individuals who were not comfortable with their laparoscopic skills at baseline, 3 out of 4 in the trained group were comfortable at the end of the rotation, in contrast to 3 out of 10 in the control group. Despite the larger proportion of trained residents feeling comfortable with their laparoscopic skills compared with controls, the observed significance was 0.175. The majority of trained residents felt that training had improved their skill level in the training center (100%) and in the operating room (89%), further support- Laparoscopic Training on Bench Models 281 ing the video-trainer and global assessment data that showed the same effects. Several authors have reported training programs based on inanimate models for both conventional open surgery and for laparoscopic surgery.4,7,9,11,13-16,22,23 Rosser and associates15 showed that laparoscopic skills may be taught outside of the operating room using simulators. Similarly several studies have shown that laparoscopic skills can be measured on a video-trainer and that proficiency improves with task repetition.7,18 The limitation of these studies is that the outcomes (improved skills) are measured on the same simulator on which training took place and not in the operating room. Several investigators have asked whether improved proficiency in the laboratory setting correlates with excellent surgical performance in clinical practice.11,19 Martin and colleagues23 showed that for open operations, testing skill level on a bench model is equally effective as testing on live animal models. Anastakis and coworkers11 showed that training for open operations on bench models yields an equivalent skill level as training on cadaver models, suggesting that skills acquired in the dry laboratory may be transferable to human operations. No such data exist for laparoscopic procedures. Until now, no data existed for open or laparoscopic procedures, which definitively correlates improvement in skill level on inanimate models with improved skill level in live human operations.11,19 Our study shows that skills acquired on a laparoscopic simulator are transferable to the operating room. Several obstacles made reaching this conclusion difficult. In addition to the difficulties associated with developing a training model and a schedule capable of accommodating residents while on surgical rotations, the ability to measure skill level in the operating room was a major undertaking. Several authors have reported using a procedurespecific checklist for open11,22,23,26,27 and laparoscopic28 procedures, in an effort to evaluate operative performance. Global assessments of operative performance based on direct observation have superior validity and reliability to checklist evaluations and may be used for different operations without modification.11,22,23,29 Global assessments may currently be the best tools available for evaluating skill level in the operating room. 282 Scott et al Laparoscopic Training on Bench Models Although standardizing the operating room experience proved difficult, we controlled for potential problems related to differing operative conditions by using three designated staff members blinded to the resident’s randomization status as first assistants for all cases. All staff assistants were given standardized instructions regarding their role. Residents were instructed to use either a onehanded or a two-handed technique at the attending physician’s discretion. Some faculty members were not comfortable allowing junior residents to use a two-handed technique. The choice of which technique to use was left up to the faculty surgeons, because they were ultimately responsible for patient safety. The use of either a one-handed or a twohanded technique may have created a bias in our methodology; fortunately, evaluators were able to perform global assessments regardless of technique. A prospective randomized design was chosen to overcome variability between groups. Evaluators were blinded to the resident’s randomization status and were independent observers who did not participate in the operations. Despite our efforts to include only elective cholecystectomies in the study, the severity of underlying disease varied. Outcomes such as operative time, length of stay, and complications were not prospectively measured. Despite the variability present in the operating room, global assessments were able to reliably measure differences in operative performance. Proving that skills acquired on a video-trainer are transferable to the operating room has significant ramifications. If residents first master basic laparoscopic skills in the laboratory setting, they may be better prepared to enter the operating room to perform laparoscopic procedures. The resident’s first few laparoscopic cases will no longer be nerveracking experiences in which a staff surgeon painfully tries to teach the resident basic video-eye-hand coordination, resulting in frustration for all those involved. Once the learning curve of using long laparoscopic instruments in a disorienting twodimensional environment is overcome, residents no longer will use up vast amounts of expensive operating room time learning basic skills. Instead, residents will be able to concentrate on learning anatomic details, nuances of surgical technique, and discussing patient management. So, the educa- J Am Coll Surg tional experience in the operating room is enhanced by laboratory skills training. For hospitals, improved operator skill may decrease operative times and lower hospital costs. Better-trained residents may mean improved patient care. Training residents before they actually enter the operating room seems credible. How long the benefit of training will last is not yet known. How long it would take for the control residents to “catch up” with their trained peers is also not known. A followup study on the residents who underwent our curriculum would be interesting. Because a number of control residents subsequently underwent training for their own educational benefit, such a study may not be feasible. It is intuitive that providing residents with basic videoeye-hand coordination boosts their skills to a level that is necessary to adequately perform laparoscopic cases and should be longlasting. Once basic skills are learned, it is unlikely that they will be forgotten. A final note concerning costs. The list price of the Guided Endoscopic Module (GEM, Karl Storz Endoscopy, Culver City, CA) ranges from $215,000 to $285,000, depending on the quality of video-imaging equipment installed. At the University of Texas Southwestern Medical Center, 186 residents train in general surgery, urology, and gynecology. The cost of training residents using the video-trainer is $270 per graduating resident. In comparison, Bridges and Diamond,3 at the University of Tennessee Medical Center—Knoxville, estimate that using operating room time to train residents costs about $48,000 per graduating resident. Training outside of the operating room, using a video-trainer such as the GEM, seems cost effective. Acknowledgment: The authors gratefully acknowledge funding and research design assistance from the Association for Surgical Education. Funding was provided by the Southwestern Center for Minimally Invasive Surgery as supported in part by an educational grant from United States Surgical—A Division of Tyco Healthcare Group. The video-trainer was provided by Karl Storz Endoscopy. Statistical analysis was performed by William H Frawley, PhD, at the Department of Academic Computing, University of Texas Southwestern Medical Center. Vol. 191, No. 3, September 2000 Scott et al References 1. Halsted WS. The training of the surgeon. Bull Johns Hopkins Hosp 1904;15:267. 2. Barnes RW, Lang NP, Whitesede MF. Halstedian technique revisited: innovations in teaching surgical skills. Ann Surg 1989; 210:118–121. 3. Bridges M, Diamond DL. The financial impact of teaching surgical residents in the operating room. Am J Surg 1999;177: 28–32. 4. Hawasli A, Featherstone R, Lloyd L, Vorhees M. Laparoscopic training in residency program. J Laparoendosc Surg 1996;6:171–174. 5. Scott-Conner CEH, Hall TJ, Anglin BL, et al. The integration of laparoscopy into a surgical residency and implications for the training environment. Surg Endosc 1994;8:1054–1057. 6. Hunter JG, Sackier JM, Berci G. Training in laparoscopic cholecystectomy: quantifying the learning curve. Surg Endosc 1994; 8:28–31. 7. Jones DB, Brewer JD, Soper NJ. The influence of threedimensional video systems on laparoscopic task performance. Surg Laparosc Endosc 1996;6:191–197. 8. Wolfe BM, Szabo Z, Moran ME, et al. Training for minimally invasive surgery: need for surgical skills. Surg Endosc 1993;7: 93–95. 9. Melvin WS, Johnson JA, Ellison EC. Laparoscopic skills enhancement. Am J Surg 1996;172:377–379. 10. Dent TL. Training, credentialing, and evaluation in laparoscopic surgery. Surg Clin North Am 1992;72:1003–1010. 11. Anastakis DJ, Regehr G, Reznick RK, et al. Assessment of technical skills transfer from the bench training model to the human model. Am J Surg 1999;177:167–170. 12. Rowan AN. Is justification of animal research necessary? JAMA 1993;269:1114. 13. Rosser JC, Rosser LE, Savalgi RS. Skill acquisition and assessment for laparoscopic surgery. Arch Surg 1997;132:200–204. 14. Shapiro SJ, Paz-Partlow M, Daykhovsky L, et al. The use of a modular skills center for the maintenance of laparoscopic skills. Surg Endosc 1996;10:816–819. Laparoscopic Training on Bench Models 283 15. Rosser JC, Rosser LE, Savalgi RS. Objective evaluation of a laparoscopic surgical skill program for residents and senior surgeons. Arch Surg 1998;1333:657–661. 16. Derossis AM, Fried GM, Abrahamowicz M, et al. Development of a model for training and evaluation of laparoscopic skills. Am J Surg 1998;175:482–487. 17. Macmillan AIM, Cuschieri A. Assessment of innate ability and skills for endoscopic manipulations by the advanced Dundee endoscopic psychomotor tester: predictive and concurrent validity. Am J Surg 1999;177:274–277. 18. Derossis AM, Bothwell J, Sigman HH, Fried GM. The effect of practice on performance in a laparoscopic simulator. Surg Endosc 1998;12:1117–1120. 19. Gagner M. Objective evaluation of a laparoscopic surgical skill program [letter]. Arch Surg 1998;133:911–912. 20. Hollander M, Wolfe DA. Nonparametric statistical methods. New York: Wiley; 1999:119–120. 21. Kirk RE. Statistics, an introduction. Fort Worth: Holt, Rinehart & Winston; 1990:334. 22. Reznick R, Regehr G, MacRae H, et al. Testing technical skill via an innovative “bench station” examination. Am J Surg 1996; 173:226–230. 23. Martin JA, Regehr G, Reznick R, et al. Objective structured assessment of technical skills (OSATS) for surgical residents. Br J Surg 1997;84:273–278. 24. Nguyen NT, Ho HS, Mayer KL, et al. Laparoscopic knot tying techniques: a comparison of performance and preference among surgery residents. J Surg Res 1999;86:291. 25. Fleiss JL. Design and analysis of clinical experiments. New York: Wiley; 1986. 26. Kopta JA. An approach to the evaluation of operative skill. Surg 1971;70:297–303. 27. Lossing A, Gretsch G. A prospective controlled trial of teaching basic surgical skills with 4th year medical students. Med Teacher 1992;14:49–52. 28. Eubanks TR, Clements R, Pohl D, et al. A valid and reliable scoring system for assessing technical skills during laparoscopic cholecystectomy. Surg Endosc 1999;13[suppl]:S32. 29. Winckel CP, Reznick RK, Cohen R, Taylor B. Reliability and construct validity of a structured technical skills assessment form. Am J Surg 1994;167:423–427.