Surg Endosc (2007) 21: 980–984 DOI: 10.1007/s00464-007-9360-3 " Springer Science+Business Media, LLC 2007

Optimizing laparoscopic task efficiency: the role of camera and monitor positions Liam A. Haveran,1 Yuri W. Novitsky,1,2 Donald R. Czerniach,1 Gordie K. Kaban,1 Melinda Taylor,1 Karen Gallagher-Dorval,1 Richard Schmidt,3 John J. Kelly,1 Demetrius E. M. Litwin1 1 2 3

Department of Surgery, University of Massachusetts Medical Center, Worcester, Massachusetts, USA Department of Surgery, University of Connecticut Health Center, Farmington, CT, USA Department of Psychology, College of the Holy Cross, Worcester, Massachusetts, USA

Received: 26 January 2007/Accepted: 13 February 2007/Online publication: 12 April 2007

Abstract Background: Alterations of video monitor and laparoscopic camera position may create perceptual distortion of the operative field, possibly leading to decreased laparoscopic efficiency. We aimed to determine the influence of monitor/camera position on the laparoscopic performance of surgeons of varying skill levels. Methods: Twelve experienced and 12 novice participants performed a one-handed task with their dominant hand in a modified laparoscopic trainer. Initially, the camera was fixed directly in front of the participant (0!) and the monitor location was varied between three positions, to the left of midline (120!), directly across from the participant (180!), and to the right of the midline (240!). In the second experiment monitor position was constant straight across from the participant (180!) while the camera position was adjusted between the center position (0!), to the left of midline (60!), and to the right of midline (300!). Participants completed five trials in each monitor/camera setting. The significance of the effects of skill level and combinations of camera and monitor angle were evaluated by analysis of variance (ANOVA) for repeated measures using restricted maximum likelihood estimation. Results: Experienced surgeons completed the task significantly faster at all monitor/camera positions. The best performance in both groups was observed when the monitor and camera were located at 180! and 0!, respectively. Monitor positioning to the right of midline (240!) resulted in significantly worse performance compared to 180! for both experienced and novice surgeons. Compared to 0! (center), camera position to the left or

Correspondence to: Yuri W. Novitsky

the right resulted in significantly prolonged task times for both groups. Novice subjects also demonstrated a significantly lower ability to adjust to suboptimal camera/monitor positions. Conclusion: Experienced subjects demonstrated superior performance under all study conditions. Optimally, the camera should be directly in front and the monitor should be directly across from a surgeon. Alternatively, the monitor/camera could be placed opposite to the surgeon!s non-dominant hand. The suboptimal camera/ monitor conditions are especially difficult to overcome for inexperienced subjects. Monitor and camera positioning must be emphasized to ensure optimal laparoscopic performance. Key words: Laparoscopy — Task efficiency — Camera positioning — Monitor positioning — Surgical education

Video monitor orientation and endoscopic camera position are critical variables in minimally invasive surgery as alterations of either or both of these may create perceptual distortion of the operative field. Perceptual confusion may subsequently affect laparoscopic task efficiency. Several studies suggest that altered monitor and camera location in relation to the position of the surgeon may lead to sensory-motor disorientation with deleterious effects on laparoscopic task performance [1– 10]. The degree to which camera and monitor location may influence the performance of laparoscopic tasks, however, is not well established. The purpose of this study was to determine the impact of various camera and monitor positions on performance of laparoscopic tasks. We also aimed to compare the impact such changes may have on surgeons of varying skill levels.

981

Fig. 2. View from under the canopy showing our set-up for a onehanded laparoscopic task. Using a laparoscopic grasper in the dominant (right) hand, the participants moved styrophoam balls from the rectangles into the glass cup. Fig. 1. The modified laparoscopic trainer with camera positioned at midline position (0!), and the monitor located straight across from the surgeon (180!).

Materials and Methods Participants After institutional review board approval, 12 experienced (surgical residents between their fourth and sixth year of training) and 12 novice (medical students) subjects were enrolled to participate in this study. Each participant had normal or corrected-to-normal vision and all were right-hand dominant.

Instruments and set-up An endosurgical simulator designed for the study (Figure 1) included a Sony 13’’ color monitor placed on a 53’’ mobile instrument tower (eye level of a standing participant) with the plane of the screen directed toward the subject!s face. A 30! Smith & Nephew DyoCam# 750 video camera and a Dyonics$ Xenon light source were utilized for all tasks. The camera was mounted on a camera holder at a fixed angle in relationship to the field. During all experiments the laparoscope was pointed directly at the operative field with a light handle pointing directly upwards. The trainer consisted of a neoprene canopy stretched across a plexiglass rectangular frame supported by adjustable height posts. The nontransparent canopy was secured to a circular piece of plywood. Around the circumference of the plywood, holes were drilled every 30! into which the camera holder could be inserted, whereby the camera could be fastened and positioned at any vertical or horizontal angle. The laparoscopic trainer was positioned on top of a table at a height so that each participant kept the shoulder adducted and the elbow at right angles. Under the nontransparent canopy, two wooden rectangles (2’’ x 2½’’) were fixed on either side of a small, large-mouth glass cup. The rectangles and glass cup were situated in the middle of the task-space under the canopy (Figure 2). Four golf-ball-sized Styrofoam balls were placed in either rectangle. A single laparoscopic grasper was used for each task and introduced into the endoscopic trainer to subtend an azimuth angle of 30!.

Procedure Using their right hand, subjects were required to grasp a Styrofoam ball from the wooden box in the contra-lateral position to the

Fig. 3. The set-up for experiments testing the impact of monitor location. The camera was fixed at 0!.

operating hand and place it into the glass cup (Figure 2). A total of four balls were placed into the cup. The participant was first introduced to the task of moving the balls to the cup under direct viewing. Once the participant was familiar with the task using direct viewing, they were positioned in front of the endosurgical trainer and monitor. The participants performed each task using visual feedback from the video monitor. To control for improvement in task performance across the trials of the experiment, the order of presentation of various camera or monitor positions was randomized according to a Latin square design. In the first exercise, the camera remained fixed parallel to the participant midline (0!) and the angular location of the monitor position with respect to the participant midline was varied. The monitor angles used were 120! (to the left of midline), 180! (directly across from the participant), and 240! (to the right of midline) (Figure 3). In the second exercise, the monitor remained fixed directly across from the participant (180!) and the angular location of the camera with respect to the participants! midline was varied. The camera angles used were 0! (parallel to midline), 60! (to the left of midline) and 300! (to the right of midline) (Figure 4). The participants remained in the midline position throughout the exercises and were allowed to turn their heads toward the monitor. Participants completed five trials for each of the six monitor-camera

982 Table 1. Comparison between task completion times for experienced and novice participants at various camera and monitor positions. Data are mean ± standard deviation. As expected, experienced subjects completed all tasks significantly faster (p < 0.001) Camera position

Monitor position

Skill level

Time (seconds)

0!

120!

Experienced Novice Experienced Novice Experienced Novice Experienced Novice Experienced Novice

7.4 10.9 7.2 9.9 7.9 11.4 9.2 20.1 11.5 23.5

180! 240! 60!

180!

300!

Fig. 4. The set-up for experiments testing the impact of camera location. The monitor was positioned at 180! throughout the experiments.

positions for a total of 30 trials. For each trial, the length of time needed to complete the task (from the time the participant grasped the instrument handle, to when all four balls were successfully placed into the cup) was recorded in seconds. The average time between five attempts was used for analysis. In all trials, illumination in the task field remained constant, and camera rotation was maintained so that the plane of the image corresponded to the actual target plane (i.e., isoplanar display of the operative field).

± ± ± ± ± ± ± ± ± ±

1.7 4.7 3.8 6.3 2.5 4.9 3.2 15.7 5.1 21.9

sec, p = 0.002), (7.2 vs. 9.9 sec, p = 0.003), and (7.9 vs. 11.4 sec, p < 0.0001), respectively. Regardless of skill level, the worst performance was observed when the monitor was located to the right of midline (240!). Time to task completion with the monitor at that location was significantly prolonged when compared to the 180! position for both the experienced (7.9 vs. 7.2 sec, p = 0.04) and the novice (11.4 vs. 9.9 sec, p = 0.03) groups. There were no significant differences in completion times between 180! and 120!, or 120! and 240! monitor locations for either group. Camera Location

Statistical analysis

Results

Experienced subjects completed the task significantly faster than novice at all camera locations. The best performance for both groups was observed when the camera was located at 0!. Average times to task completion for experienced versus novice subjects at 0!, 60! and 300! were (7.2 vs. 9.9 sec, p < 0.0001; 9.2 vs. 20.1 sec, p < 0.0001; and 11.5 vs. 23.5 sec, p < 0.0001, respectively). Performance diminished significantly for both groups when the camera was positioned at 60! or 300!. For both groups the worst performance was observed when the camera was located at 300!. Task completion times at 60! and 300! were significantly prolonged when compared to 0! for the experienced (9.2 vs. 7.2 sec, p = 0.04 and 11.5 vs. 7.2 sec, p < 0.001, respectively), and the novice (20.1 vs. 9.9 sec, p <0.001 and 23.5 vs. 9.9 sec, p < 0.001, respectively) subjects.

The times to complete the tasks at the various monitor and camera positions are summarized in Table 1.

Adaptation

The significance of the effects of camera and monitor angle as well as skill level were evaluated by analysis of variance (ANOVA) for repeated measures using restricted maximum likelihood estimation. In the presence of significant main and/or interaction effects, pairwise comparisons of interest were evaluated using Tukey!s test using the estimated nondiagonal covariance matrix. The distributional characteristics of the procedure times were evaluated both graphically by inspection of frequency histograms and by the Kolmogorov–Smirnov goodness-of-fit test for normality. The data were transformed using successive logarithmic transformations to achieve compliance with the normal distribution. Values are expressed as mean ± standard deviation. Results were considered significant when associated p values were <0.05. All computations were performed using the Proc MIXED procedure in the SAS version 9.13 statistical software package.

Monitor Location Experienced participants completed the task significantly faster than the novice at all monitor positions. The best performance in both groups was observed when the monitor was positioned directly across from the subjects (at 180!). The average times to task completion for experienced versus novice participants at 120!, 180!, and 240! monitor positions were (7.4 vs. 10.9

To determine the ability of experienced and novice subjects to adapt to suboptimal camera and monitor positions, a difficulty coefficient was calculated by dividing the task time of task completion under suboptimal conditions (monitor at 120! or 240!; camera at 300! or 60!) to the time of task completion under the best conditions (camera at 0! or monitor at 180!). A higher difficulty coefficient was indicative of a lower degree of adaptation. As seen in Tables 2 and 3, the times of task completion for the novice subjects were

983 Table 2. The rates of adaptation to suboptimal camera positions. The 0! camera position was found to be the most optimal in our experiments. The difficulty coefficient was calculated by dividing the time of task completion at the suboptimal condition by the task time at the 0! camera position. A higher difficulty coefficient is indicative of a lower degree of adaptation Difficulty coefficient Camera position

Experienced subject

Novice subjects

P value

0! 60! 300!

1 1.19 1.32

1 1.75 1.84

0.024 0.035

Table 3. The rates of adaptation to suboptimal monitor positions. The 180! monitor position was found to be the most optimal in our experiments. The difficulty coefficient was calculated by dividing the time of task completion at the suboptimal condition by the task time at the 180! monitor position. A higher difficulty coefficient is indicative of the lower degree of adaptation Difficulty coefficient Monitor position

Experienced subject

Novice subjects

P value

180! 120! 240!

1 1.07 1.12

1 1.25 1.32

0.048 0.034

longer by the factor of 1.25–1.32 for suboptimal monitor locations and by the factor of 1.75–1.84 for the suboptimal camera locations. Overall, the experienced subjects had statistically higher rates of adjusting to suboptimal camera (p = 0.024 and 0.035) and monitor (p = 0.048 and 0.034) conditions (Tables 2 and 3).

Discussion Laparoscopic surgery presents several obstacles not encountered in open surgery, such as decreased tactile feedback, loss of binocular depth cues and perceptual distortion [11]. These shortcomings may be due to paradoxical movement of the instrument tips with respect to hand movement or loss of coaxial visual-motor alignment [12–15]. Therefore, optimal laparoscopic performance is more likely to be achieved when a surgeon achieves appropriate triangulation of the instrument and camera ports. In addition, a surgeon can benefit from maintaining a visual-motor alignment, such that a straight line exists between the monitor, camera and the surgeons! head and hands [7–10]. In reality, however, this set-up cannot always be realized. Due to spatial equipment constraints and the complexity of advanced laparoscopic operations, the monitor and camera are often positioned in an oblique manner with regard to the surgeon. Clinically, this may result in increased mental and physical stress, prolonged operative times and ultimately diminished operative performance [16]. The purpose of the current study was to establish the impact of various camera and monitor locations on laparoscopic task performance of surgeons of varying

skill levels. Our results support the importance of optimal monitor and camera positioning in laparoscopic performance regardless of skill level. Task performance degraded significantly for both groups when coaxial visual-motor alignment was disrupted. That occurred when the camera and monitor were not in line with the surgeons! hands and head. The best performance was achieved in both groups when the monitor was located directly opposite the surgeon. In contrast, however, task completion for both experienced and novice participants degraded significantly when the monitor was located to the right of midline, so that the right-handed study participants had to turn away from the direction of hand movement. Our data also supports two previous studies that demonstrated how monitor location can adversely affect optimal laparoscopic task performance, especially when the surgeon has to turn their head away from the direction where the work is being performed. Matern et al. [4] evaluated endoscopic task performance and neck muscle strain in different monitor positions and demonstrated that monitor location to the right (240! in our study) resulted in significantly worse task performance with significantly more muscle fatigue. In a similar study, Hanna et al. [9] investigated the relation between endoscopic task performance and image display. Task performance degraded significantly when the participants had to rotate their head to the left or right from the midline (120! or 240! in our study). Similar to the current investigation, optimal performance was achieved with a frontal (180!) monitor position in both the Matern et al. [4] and Hanna et al. [9] studies. We also found that monitor positions at 120! or 240! presented a statistically more challenging scenario for the novice group. Overall, if a monitor location of 180! cannot be realized because of the operative scenario, then the next best position for the monitor seems to be to the left of the right-handed surgeon. Our results substantiate previous investigations demonstrating how changes in camera positioning have dramatic effects on coordinated movement in laparoscopic surgery. Even minor changes can have a dramatic effect on the interpretation of the operative field and the spatial relationship between instruments and tissues. Similar to monitor location, the best laparoscopic performance in the current study was observed when the camera was located at 0!, resulting in coaxial alignment of the visual-motor axes. The second fastest performance for both experienced and novice participants was observed when the camera was located to the left of the midline. Interestingly, the worst performance for both experienced and novice surgeons was observed when the camera was located to the right of the midline, so that the surgeons were, in essence, looking down the axis of their dominant (and manipulating) hand toward the target. These results are in accordance with the findings of Eman et al. [3] who studied task performance and muscle work with camera positions located in the center, to the left, and to the right of the midline. The study demonstrated significant degradation of performance when the camera was located to the right (300! in our study) on the same side as the dominant hand. The

984

authors concluded that off-optical axis manipulation is especially impaired when the camera is located on the same side as the dominant hand because the display angles are decidedly different than the actual physical angles. While operating against the laparoscope at 60! may be difficult, it did not adversely affect task efficiency as much as an inline camera position (to the right along the axis of the manipulating hand). Similarly to the suboptimal monitor locations, the experienced subjects in this series had statistically easier time adjusting to the suboptimal camera position. In other words, proper camera (and monitor) placement appears to be especially important for surgeons-in-training or at the beginning of the learning curve. Given a minimal morbidity of an additional 5-mm trocar, surgeons must keep in mind potential benefits of utilizing an additional access port with resultant improved laparoscopic efficiency when the operative field is viewed at optimal camera angles. As the complexity of laparoscopic procedures continues to increase, optimization of monitor and camera position will become even more important. While proper triangulation dictates placement of the camera in the 0! position, situated between the surgeons! hands, this location may not always be possible. The same can be said for monitor position. Although the best-case scenario would be for the monitor to be located directly across from the surgeon, in reality this position is rarely realized. Since laparoscopy has become a predominant component of modern surgical training, the surgeons must understand and aim towards the most effective monitor-camera alignments to achieve maximal performance.

Conclusion Both camera and monitor locations may have significant effects on laparoscopic performance. Our study demonstrates that camera location is more critical for optimizing laparoscopic task performance than monitor position, and that the best performance is achieved when the camera is placed in the 0! location, situated between the surgeons! hands. If the camera is to be repositioned away from this location, the next orientation of choice should be with the camera to the left of the right-handed surgeon. Monitor position directly across the subjects was found to be superior to other locations. However, if a monitor location of 180! cannot be realized because of the operative scenario, then the next best position for the monitor should also be to the left of the right-handed surgeon. In this study,

experienced subjects not only were able to complete all tasks faster, but also had a significantly easier time adjusting to suboptimal camera and monitor locations. It appears that both camera and monitor positioning must be emphasized as an important aspect of laparoscopy, especially for less-experienced laparoscopic surgeons.

References 1. Holden JG, Flach JM, Donchin Y (1999) Perceptual-motor coordination in an endoscopic surgery simulation. Surg Endosc 13: 127–132 2. Omar AM, Wade NJ, Brown SI, Cuschieri A (2005) Assessing the benefits of ‘‘gaze-down’’ display location in complex tasks. Surg Endosc 19: 105–108 3. Emam TA, Hanna G, Cuschieri A (2002) Comparison of orthodox vs off-optical axis endoscopic manipulations: importance of monitor display angles. Surg Endosc 16: 401–405 4. Matern U, Faist M, Kehl K, Giebmeyer C, Buess G (2005) Monitor position in laparoscopic surgery. Surg Endosc 19: 436– 440 5. Conrad J, Shah AH, Divino CM, Schluender S, Gurland B, Shlasko E, Szold A (2006) The role of mental rotation and memory scanning on the performance of laparoscopic skills: a study on the effect of camera rotational angle. Surg Endosc 20: 504–510 6. Smith WD, Berguer R, Nguyen NT (2005) Monitor height affects surgeons! stress level and performance on minimally invasive surgery tasks. Stud Health Technol Inform 111: 498–501 7. Zehetner J, Kaltenbacher A, Wayand W, Shamiyeh A (2006) Screen height as an ergonomic factor in laparoscopic surgery. Surg Endosc 20: 139–141 8. Moschos E, Coleman RL (2004) Acquiring laparoscopic skill proficiency: does orientation matter. Am J Obstet Gynecol 191: 1782–1787 9. Hanna GB, Shimi SM, Cuschieri A (1998) Task performance in endoscopic surgery is influenced by location of the image display. Ann Surg 227: 481–484 10. Medina M (1997) Image rotation and reversal: major obstacles in learning intracorporeal suturing and knot-tying. JSLS 1: 331–336 11. Breedveld P, Stassen HG, Meijer DW, Jakimowicz JJ (2000) Observation in laparoscopic surgery: overview of impeding effects and supporting aids. J Laparoendosc Adv Surg Tech A 10: 231–41 12. Gallagher AG, McClure N, McGuigan J, Ritchie K, Sheehy NP (1998) An ergonomic analysis of the fulcrum effect in the acquisition of endoscopic skills. Endoscopy 30: 617–620 13. Gallagher AG, Ritter EM, Lederman AB, McClusky DA, Smith CD (2005) Video-assisted surgery represents more than a loss of three-dimensional vision. The American Journal of Surgery 189: 76–80 14. Crothers I, Gallagher A, McClure N, et al. (1999) Experienced laparoscopic surgeons are automated to the ‘‘fulcrum effect’’: an ergonomic demonstration. Endoscopy 31: 365–369 15. Emam TA, Hanna G, Cuschieri A (2002) Ergonomic principles of task alignment, visual display, and direction of execution of laparoscopic bowel suturing. Surg Endosc 16: 267–271 16. Zheng B, Janmohamed Z, MacKenzie CL (2003) Reactio times and the decision-making process in endoscopic surgery: an experimental study. Surg Endosc 17: 1475–1480