- Email: [email protected]
Comparison of training effect on tremor using 2 training modules
656
ARTICLE
Comparison of training effect on tremor using 2 training modules Abhishek R. Payal, MB BS, MPH, Yonwook J. Kim, BS, Luis A. Gonzalez Gonzalez, MD, MPH, Mary K. Daly, MD
Purpose: To compare training effect of 2 training modelsda surgical simulator anti-tremor module and a paper versiondon tremor and time-to-task completion. Setting: Ophthalmology Department, Veterans Affairs Boston Healthcare System, Jamaica Plain, Massachusetts, USA. Design: Prospective crossover study. Methods: Trainees completed simulator and paper training modules (baseline test, 3 training sessions, posttraining test, and final test) with their dominant and nondominant hands. The change in tremor, number of paper errors, and time-to-task completion in dominant and nondominant hands were compared. The 2 training modules were compared using nonparametric tests.
3-dimensional module) and paper errors (paper, 2-dimensional module) (Spearman ⍴ Z 0.35, P < .0001). Practice on the simulator or paper modules did not reduce tremor significantly from baseline to final tasks for both hands combined (P Z .12, simulator; P Z .2, paper). Practice on the training modules improved time-to-task completion in the simulator module and paper module (both P < .0001). The improvement in time from baseline to final tasks was greater in the nondominant hands in the simulator module (improvement 64.5% over baseline time) than in the paper module (53.6% over baseline time).
Conclusion: Practice might not reduce tremor but improved the outcome measure of time, and results suggest that trainees can learn to compensate for tremor in both hands, which is important in bimanual microsurgery.
Results: The study comprised 19 trainees. There was a moderate correlation between average tremor values (simulator,
O
phthalmic surgeries are bimanual, requiring proficient use of both the dominant hand and the nondominant hand. Hand steadiness is a critical factor in ophthalmic surgeries because operative tremor can influence the outcomes of microsurgical procedures requiring precision.1 A national survey of 58 ophthalmology program directors2 showed that 199 residents (9%) had trouble mastering surgical skills, which was attributed to poor hand–eye coordination in 78 (24%) and tremor in 45 (14%). Physiologic tremor is a high-frequency low-amplitude tremor present in healthy individuals and is affected by factors such as caffeine, exercise, operating time, and fasting.3–6 It might disrupt the process of learning microsurgery, which makes it important for trainees to
J Cataract Refract Surg 2017; 43:656–661 Q 2017 ASCRS and ESCRS
understand the nature of the tremor and how to control urbe et al.8 found that experienced otosurgeons it.7 M€ showed smaller tremor amplitude, suggesting that hand steadiness could be improved with training and experience. To our knowledge, the effect of structured training on improving tremor, especially in nondominant hands, has not been studied in ophthalmic surgeries. The current training methods in ophthalmology for surgical competency include the traditional apprentice model (learning by doing), wet labs, and simulators. Given the limitations of the current education methods in surgery,9,10 virtual reality–based simulators hold tremendous potential for improving resident cataract surgical training.10–12 The 2016 Accreditation Council for Graduate Medical Education (ACGME) program requirements for
Submitted: September 28, 2016 | Final revision submitted: January 27, 2017 | Accepted: January 29, 2017 From the Departments of Ophthalmology, Veterans Affairs Boston Healthcare System (Payal, Kim, Gonzalez Gonzalez, Daly), Jamaica Plain, and Boston University School of Medicine (Kim, Daly) and Harvard Medical School (Payal, Daly), Boston, Massachusetts, USA. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States Government. Supported with resources and the use of facilities at the Veterans Affairs Boston Healthcare System, Boston, Massachusetts, USA. Jose Efren Gonzalez Monroy, MD, assisted with initial institutional review board documentation and initiating study enrollment. Presented as a poster at the annual meeting of the Association for Research in Vision and Ophthalmology, Seattle, Washington, USA, May 2016. Corresponding author: Mary K. Daly, MD, Veteran Affairs Boston Healthcare System, 150 South Huntington Avenue, Suite 8C-29, Boston, Massachusetts 02130, USA. E-mail: [email protected]. Q 2017 ASCRS and ESCRS Published by Elsevier Inc.
0886-3350/$ - see frontmatter http://dx.doi.org/10.1016/j.jcrs.2017.01.021
TRAINING EFFECT ON TREMOR
ophthalmology mandate institutions to provide “a surgical skills development resource (a wet lab or simulators)” for resident education.A However, the ACGME stops short and does not mandate specific wet-lab or surgicalsimulation skills development curricula. Systematic training programs using virtual reality simulator modules can improve cataract surgery skills13 and can be as effective as wet-lab training.14 The anterior segment anti-tremor and forceps training modules of the Eyesi surgical simulator (VRMagic Holding AG) have been shown to have construct validity.15,16 It is not known whether there is a difference in improvement of tremor in dominant hands and nondominant hands after training on such modules. Alternatively, paper-based tests, such as the Archimedean spiral, are widely used by neurologists to evaluate essential tremor (a neurological disorder with postural and kinetic tremors of the upper limbs). In patients with essential tremor, dominant hands perform better than nondominant hands in spirography.17 However, the effect of paper-based training on dominant hands and nondominant hands with respect to physiologic tremor has not been studied in healthy individuals in the context of ophthalmic surgery. We compared the change in tremor and time-to-task completion in dominant hands and nondominant hands between a high-fidelity 3-dimensional (3-D) structured simulator training module and a paper-based training module (an adjunct measure of tremor). SUBJECTS AND METHODS The study was approved by the Veterans Affairs Boston Healthcare System Institutional Review Board. Informed consent was obtained from all participants. All procedures performed in the studies were in accordance with the ethical standards of the institutional research committee and with the tenets of the Declaration of Helsinki. Medical students and an ophthalmology resident reporting for their clinical elective/rotation consecutively at the Ophthalmology Department of the Veteran Affairs Boston Healthcare System participated in the study. Medical students were recruited irrespective of their career interest in surgery or ophthalmology because no difference in microsurgery performance has been reported between those with and those without a strong interest in ophthalmology.18 All participants completed a demographic questionnaire and the Waterloo Handedness Questionnaire (WHQ). The demographic questionnaire included questions on training level, surgical experience, previous simulator exposure, and lifestyle habits. The WHQ has 20 questions that assess the hand preference for daily tasks.19 Available responses were as follows: left always, left usually, equal, right usually, and right always. Each response was respectively scored as 2, 1, 0, 1, or 2 and added up to yield a number ranging from 40 (strong left-hand preference) to 40 (strong right-hand preference). Right hand was designated as dominant for those with a positive score and left hand as dominant for those with a negative score. After hand dominance was assessed, the participants were randomized into 1 of 4 groups that determined the sequence of modules they would complete and the hand they would use first (Figure 1). The groups were as follows: paper module first–dominant hand first, paper module first–nondominant hand first, simulator module first–dominant hand first, and simulator module first– nondominant hand first. Randomization was performed to equally distribute the learning effect of 1 hand on the other, following the hypothesis of Park et al.20 that the hand that performs the task first might
657
confer a learning effect on the other hand that will perform the task later. Participants received a demonstration and completed a practice module to familiarize themselves with the simulator and paper tasks. Subsequently, they completed both the paper module and simulator module as per the group to which they were randomized. Training Modules The simulator modules were completed using the Eyesi surgical simulator (software version 2.8.12). It includes a microscope that provides a stereoscopic image to the user and handheld probes that resemble surgical instruments. For this study, antitremor module level 6 was used on a cataract mannequin head (Figure 2, left). The module consisted of moving a sphere counterclockwise from center to periphery along a spiral path in the anterior segment with the tip of an instrument without damaging the lens or cornea. These abstract simulation tasks do not require specific surgical knowledge but do require discreet movements and a degree of manual dexterity. The simulator reported average tremor values and overall scores, among other parameters. Higher scores indicated better performance within a range of 0 to 100. One hundred points were given for completion, and points were deducted for errors. Possible error parameters included an outof-tolerance percentage, injury to lens or cornea, out-of-focus manipulation, working without red reflex, and excessive time use. For the paper module, an Archimedean spiral with separation distance of 2.5 mm between spiral lines was used (Figure 2, right). Participants traced a continuous line along the spiral path with a pen starting at the center while staying in between the lines. Performance was measured by counting the number of times the participant’s line crossed or touched the spiral line and the time required to complete the task. Three training sessions in each module were chosen based on the recommendation by Saleh et al.21 that at least 3 attempts should be incorporated when designing a relevant training syllabus to account for the variability in performance of trainees. Each module required approximately 1 to 2 hours for completion and a total of 4 to 5 hours for the entire study. Participants completed the study within 2 weeks. The primary outcome was the change in tremor and time-totask completion from baseline to final task in the dominant hand and the nondominant hand in the 2 training models. Secondarily, the 2 training models were compared for effectiveness in improving tremor and time-to-task completion. A subgroup analyses was performed to compare simulator scores for “strong dominants” (ie, higher than C30 and lower than 30 on the WHQ scale) with “weak dominants” (ie, lower than C10 and higher than 10). Statistical Analysis Statistical analyses were performed using JMP Pro software (version 12.0, SAS Institute, Inc.). Simple averages, standard deviations, and descriptive statistics were used for demographic data. The Wilcoxon/Kruskal-Wallis (rank sums) test was used for comparison of quantitative data between dominant hands and nondominant hands and between different steps of the training modules. The mean change in outcomes from baseline to the final test was measured by the interaction term in the repeated-measures analysis of variance. The Spearman coefficient (⍴) was used to assess the correlation between outcomes. A P value less than 0.05 was considered statistically significant.
RESULTS Twenty participants (19 fourth-year medical students, 1 final year ophthalmology resident) were enrolled in the study; 19 (95%) completed the entire study. One medical Volume 43 Issue 5 May 2017
658
TRAINING EFFECT ON TREMOR
Figure 1. Flow diagram of study steps for each participant. The letter a represents each training module (paper or simulator) comprising 3 training sessions and a posttraining test (62 spirals/module). The letter b represents the baseline test consisting of 1 spiral each for dominant hand and nondominant hand on paper and simulator (4 spirals) with each training session comprising 10 spirals for each dominant hand and nondominant hand, totaling 60 spirals (10 spirals 2 hands 3 sessions). The posttraining test consisting of 1 spiral for each hand. The letter c represents after the trainees completed the first module, when they crossed over to the other module. Upon completion of both modules, they took a final test that included 1 paper and simulator spiral per hand (4 spirals). Thus each participant, except 1 that was lost to follow-up, completed 132 spirals (62 spirals 2 modules C 4 baseline test spirals C 4 final test spirals).
student could not complete the study because of scheduling issues but completed 22 paper spirals and 2 simulator spirals. Two participants (10%) were left-handed. Three participants had experience with an ophthalmic surgical simulator, all for less than 2 hours. Ten participants (50%) played video games regularly, and 6 (30%) had experience playing a musical instrument. Average Tremor on Simulator and Errors on Paper Modules
There was moderate correlation between average tremor on the simulator anti-tremor module (3-D) and paper errors (2-dimensional) (Spearman ⍴ Z 0.35, P ! .0001). Practice on the simulator or paper modules did not reduce tremor significantly from baseline to final tasks (P Z .12, simulator; P Z .2, paper). This pattern of no reduction in tremor or paper errors from baseline to final tasks was also observed when the dominant
Figure 2. Images of simulator anti-tremor module (left) and paper module (right).
Volume 43 Issue 5 May 2017
hand and the nondominant hand were visualized separately (Figure 3). A subgroup analysis exploring tremor and paper errors in weak dominant hands (WHQ score lower than C10 and higher than 10), moderately dominant hands (between C10 and C30 and between 10 and 30), and strong dominant hands (higher than C30 and lower than 30) showed greater variability in tremor values in the weak dominant hands (Figure 4). Time-to-Task Completion
There was improvement in the time-to-task completion from baseline to final task in the simulator and paper modules. Practice on the training modules improved time-to-task completion in both simulator and paper modules (both P ! .0001). The improvement in time (time to complete final task – time to complete baseline task) was greater in the simulator module (87.6 seconds, improvement of 63% over baseline time) than in the paper module (18.71 seconds, 48%). The improvement in time from baseline to final tasks was greater in the nondominant hands in the simulator module (improvement of 64.5% over baseline time) than in the paper module (53.6% over baseline time) (Figure 5). Dominant hands also seemed to improve more in the simulator module (improvement of 60.2%) than in the paper module (52.1%). Overall, in both modules, nondominant hands performed slower than dominant hands (both P ! .0001) (Figure 5). DISCUSSION We compared change in tremor and time-to-task completion in dominant hands and nondominant hands on a surgical
TRAINING EFFECT ON TREMOR
659
Figure 3. Mean tremor (SimAvg) on simulator versus paper errors by dominant hands (D) and nondominant hands (ND).
simulator and on paper-based tasks and found that after structured training, participants were not able to reduce tremor significantly but were able to improve their times. In addition to simulator training, we included a paper module that could potentially be an inexpensive alternative for dexterity or tremor assessment. However, we did not find a strong correlation for the primary outcomes (tremor and paper errors) in the 2 training modules. This is not a
surprise because the surgical simulator traces, with high precision, the exact path a participant’s probe travels in the simulated eye and assigns a numerical value (the tremor value) to the deviation from the prescribed spiral path. In contrast, the paper module provides a much less sophisticated method of measuring tremor. Nevertheless, the correlation between the 2 modules was moderate (r Z 0.35) and statistically significant.
Figure 4. Mean tremor (SimAvg) and paper errors in weak dominant hands (WeakDom), moderately dominant hands (ModDom), and strong dominant hands (StrDom) (D Z dominant hands; ND Z nondominant hands). Volume 43 Issue 5 May 2017
660
TRAINING EFFECT ON TREMOR
Figure 5. Time-to-task completion in simulator and paper modules by hand dominance (D Z dominant hands; ND Z nondominant hands).
Practicing the structured training program on both the simulator module and the paper module did not result in significant reduction in tremor, and this effect was true for dominant hands and nondominant hands separately. This suggests that an individual baseline physiologic tremor is not reduced through training. This finding is consistent with studies that show no differences in baseline tremor between residents and attending-level physicians.5,15 In our cohort, nondominant hands had more errors than dominant hands in the paper module at the final test (P Z .02) than at baseline (P Z .23). This discrepancy in results between the simulator module and paper module could be explained as follows: In the paper module, participants were able to see the time taken to complete each spiral task but were not able to immediately see the number of errors committed. It is possible that to improve their performance, they attempted to finish the spirals quicker, thus improving their time, while inadvertently committing more errors, which they could not see immediately. This is reflected in the significant negative correlation between time and paper error at final test (r Z 0.45, P Z .004). In addition, there is a fundamental difference in the designs of the paper module and the simulator module. In the simulator anti-tremor module, the user cannot advance on the spiral unless the instrument tip is within a tolerance range. If the tip is off the spiral, the simulator essentially forces the user to return to the correct position to proceed with task completion. There is no such feedback feature in the paper module, so it would be possible for the participants to finish quicker while committing more errors. Because the simulator places more emphasis on precision, it might be more suitable than paper training for improving ophthalmic surgical skills. Real-time feedback might play an important role in improving the effectiveness of training.22 Volume 43 Issue 5 May 2017
In a subgroup analysis, we classified handedness into weak, moderate, and strong dominant based on WHQ scores and compared these categories on tremor. We found that participants with strong dominant handedness tended to have less tremor overall and less variability in tremor from baseline to final tasks than those with weak dominant handedness (Figure 4). Participants with moderate hand dominance had the lowest number of errors and lowest average tremor of the 3 hand-dominance groups. This could be related to ambidexterity and merits further investigation. In addition, on multiple comparison of tremor on all 3 levels (comparing strong to moderate, strong to weak, and moderate to weak), we found a statistically significant difference between the 3 levels (all P ! .0001). This suggests that the strength of handedness might be an important consideration in training residents. Although tremor might not decrease significantly after training, it might be worthwhile for those with WHQ scores between 10 and C10 to practice more to achieve tremor levels of their peers. Participants improved time-to-task completion in both modules in our study. Although nondominant hands were slower overall than dominant hands in simulator tasks and paper tasks, as expected, the improvement in time from baseline to the final task was remarkable in nondominant hands in the simulator module (64.5% over baseline time). This was greater than the improvement by nondominant hands in the paper module or by dominant hands in the paper or simulator module. This finding warrants further study. The limitations of our study include those related to its design and population and those related to the equipment used. The sample in our study was small; however, it was well within the range of other ophthalmic simulation studies.15,21,23–25 The time required to complete the study was a main factor contributing to fewer participants and the
TRAINING EFFECT ON TREMOR
preponderance of medical students over residents. The finalyear ophthalmology resident included in the study was likely to have more microsurgical surgery experience than most medical students. The resident was part of Group A-b in the randomization. It is unlikely that the interpretation of results would be affected by this because the comparison of outcomes between groups was not performed; rather, they were compared between type of module, hand dominance, and strength of hand dominance. In conclusion, our study found that structured training and practice might not appreciably reduce tremor but can improve speed in dominant hands and nondominant hands. A structured simulator program that focuses on improving dexterity in both hands should be considered for ophthalmology residents and fellows.
WHAT WAS KNOWN Poor hand–eye coordination and tremor are recognized issues during resident surgical training.
WHAT THIS PAPER ADDS A structured training program helped improve time-to-task completion on both the simulator module and the paper module but did not appreciably reduce tremor. Trainees with weak dominant hands (WHQ scores between 10 and C10) might need more practice and training than their peers with strong dominant hands (scores OC30, ! 30).
REFERENCES 1. Coulson CJ, Slack PS, Ma X. The effect of supporting a surgeon’s wrist on their hand tremor. Microsurgery 2010; 30:565–568 2. Binenbaum G, Volpe NJ. Ophthalmology resident surgical competency; a national survey. Ophthalmology 2006; 113:1237–1244 3. Holmes JM, Toleikis SC, Jay WM. The effect of arm exercise and ocular massage on postural hand tremor. Ann Ophthalmol 1992; 24:156–158 4. Hsu PA, Cooley BC. Effect of exercise on microsurgical hand tremor. Microsurgery 2003; 23:323–327 5. Slack PS, Coulson CJ, Ma X, Pracy P, Parmar S, Webster K. The effect of operating time on surgeon’s hand tremor. Eur Arch Otorhinolaryngol 2009; 266:137–141 6. Wharrad HJ, Birmingham AT, Macdonald IA, Inch PJ, Mead JL. The influence of fasting and of caffeine intake on finger tremor. Eur J Clin Pharmacol 1985; 29:37–43 7. Harwell RC, Ferguson RL. Physiologic tremor and microsurgery. Microsurgery 1983; 4:187–192 €rbe D, Hu €ttenbrink KB, Zahnert T, Vogel U, Tassabehji M, Kuhlisch E, 8. Mu Hofmann G. Tremor in otosurgery: influence of physical strain on hand steadiness. Otol Neurotol 2001; 22:672–677 9. Smith JH. Teaching phacoemulsification in US ophthalmology residencies: can the quality be maintained? Curr Opin Ophthalmol 2005; 16:27–32 10. Khalifa YM, Bogorad D, Gibson V, Peifer J, Nussbaum J. Virtual reality in ophthalmology training. Surv Ophthalmol 2006; 51:259–273
661
11. Belyea DA, Brown SE, Rajjoub LZ. Influence of surgery simulator training on ophthalmology resident phacoemulsification performance. J Cataract Refract Surg 2011; 37:1756–1761 12. Grodin MH, Johnson TM, Acree JL, Glaser BM. Ophthalmic surgical training: a curriculum to enhance surgical simulation. Retina 2008; 28:1509–1514 13. Bergqvist J, Person A, Vestergaard A, Grauslund J. Establishment of a validated training programme on the Eyesi cataract simulator. A prospective randomized study. Acta Ophthalmol 2014; 92:629–634. Available at: http://onlinelibrary.wiley.com/doi/10.1111/aos.12383/epdf. Accessed March 17, 2017 14. Daly MK, Gonzalez E, Siracuse-Lee D, Legutko PA. Efficacy of surgical simulator training versus traditional wet-lab training on operating room performance of ophthalmology residents during the capsulorhexis in cataract surgery. J Cataract Refract Surg 2013; 39:1734– 1741 15. Mahr MA, Hodge DO. Construct validity of anterior segment antitremor and forceps surgical simulator training modules: attending versus resident surgeon performance. J Cataract Refract Surg 2008; 34:980–985 16. Le TDB, Adatia FA, Lam W-C. Virtual reality ophthalmic surgical simulation as a feasible training and assessment tool: results of a multicentre study. Can J Ophthalmol 2011; 46:56–60 17. Ondo WG, Wang A, Thomas M, Vuong KD. Evaluating factors that can influence spirography ratings in patients with essential tremor. Parkinsonism Relat Disord 2005; 11:45–48 18. Gillan SN, Okhravi N, O’Sullivan F, Sullivan P, Viswanathan A, Saleh GM. Influence of medical student career aims on ophthalmic surgical simulator performance (part of the international forum for ophthalmic simulation studies). Br J Ophthalmol 2016; 100:411–414 19. Steenhuis RE, Bryden MP, Schwartz M, Lawson S. Reliability of hand preference items and factors. J Clin Exp Neuropsychol 1990; 12:921–930. Available at: https://www.ucl.ac.uk/medical-education/resources/Waterloo/ WaterlooHandQuese32itemsSteenhuis1990.pdf. Accessed March 17, 2017 20. Park J, Williams O, Waqar S, Modi N, Kersey T, Sleep T. Safety of nondominant-hand ophthalmic surgery. J Cataract Refract Surg 2012; 38:2112–2116 21. Saleh GM, Theodoraki K, Gillan S, Sullivan P, O’Sullivan F, Hussain B, Bunce C, Athanasiadis I. The development of a virtual reality training programme for ophthalmology: repeatability and reproducibility (part of the International Forum for Ophthalmic Simulation Studies). Eye 2013; 27:1269–1274. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3831124/pdf/eye2013166a.pdf. Accessed March 17, 2017 €lscher AH, 22. Kleinert R, Plum P, Heiermann N, Wahba R, Chang DH, Ho Stippel DL. Embedding a virtual patient simulator in an interactive surgical lecture. J Surg Educ 2016; 73:433–441 23. Saleh S, Uppal S, Sharma V. Greater nondominant hand proficiency is not associated with enhanced simulated surgical performance. Can J Ophthalmol 2015; 50:350–353 24. Selvander M, Åsman P. Cataract surgeons outperform medical students in Eyesi virtual reality cataract surgery: evidence for construct validity. Acta Ophthalmol 2013; 91:469–474. Available at: http://onlinelibrary.wiley. com/doi/10.1111/j.1755-3768.2012.02440.x/pdf. Accessed March 17, 2017 25. Privett B, Greenlee E, Rogers G, Oetting TA. Construct validity of a surgical simulator as a valid model for capsulorhexis training. J Cataract Refract Surg 2010; 36:1835–1838 OTHER CITED MATERIAL A. Accreditation Council for Graduate Medical Education. ACGME Program Requirements for Graduate Medical Education in Ophthalmology. Available at: http://www.acgme.org/acgmeweb/Portals/0/PFAssets/ProgramRequi rements/240_ophthalmology_2016.pdf. Accessed March 17, 2017
Disclosure: None of the authors has a financial or proprietary interest in any material or method mentioned.
Volume 43 Issue 5 May 2017
ARTICLE
Comparison of training effect on tremor using 2 training modules Abhishek R. Payal, MB BS, MPH, Yonwook J. Kim, BS, Luis A. Gonzalez Gonzalez, MD, MPH, Mary K. Daly, MD
Purpose: To compare training effect of 2 training modelsda surgical simulator anti-tremor module and a paper versiondon tremor and time-to-task completion. Setting: Ophthalmology Department, Veterans Affairs Boston Healthcare System, Jamaica Plain, Massachusetts, USA. Design: Prospective crossover study. Methods: Trainees completed simulator and paper training modules (baseline test, 3 training sessions, posttraining test, and final test) with their dominant and nondominant hands. The change in tremor, number of paper errors, and time-to-task completion in dominant and nondominant hands were compared. The 2 training modules were compared using nonparametric tests.
3-dimensional module) and paper errors (paper, 2-dimensional module) (Spearman ⍴ Z 0.35, P < .0001). Practice on the simulator or paper modules did not reduce tremor significantly from baseline to final tasks for both hands combined (P Z .12, simulator; P Z .2, paper). Practice on the training modules improved time-to-task completion in the simulator module and paper module (both P < .0001). The improvement in time from baseline to final tasks was greater in the nondominant hands in the simulator module (improvement 64.5% over baseline time) than in the paper module (53.6% over baseline time).
Conclusion: Practice might not reduce tremor but improved the outcome measure of time, and results suggest that trainees can learn to compensate for tremor in both hands, which is important in bimanual microsurgery.
Results: The study comprised 19 trainees. There was a moderate correlation between average tremor values (simulator,
O
phthalmic surgeries are bimanual, requiring proficient use of both the dominant hand and the nondominant hand. Hand steadiness is a critical factor in ophthalmic surgeries because operative tremor can influence the outcomes of microsurgical procedures requiring precision.1 A national survey of 58 ophthalmology program directors2 showed that 199 residents (9%) had trouble mastering surgical skills, which was attributed to poor hand–eye coordination in 78 (24%) and tremor in 45 (14%). Physiologic tremor is a high-frequency low-amplitude tremor present in healthy individuals and is affected by factors such as caffeine, exercise, operating time, and fasting.3–6 It might disrupt the process of learning microsurgery, which makes it important for trainees to
J Cataract Refract Surg 2017; 43:656–661 Q 2017 ASCRS and ESCRS
understand the nature of the tremor and how to control urbe et al.8 found that experienced otosurgeons it.7 M€ showed smaller tremor amplitude, suggesting that hand steadiness could be improved with training and experience. To our knowledge, the effect of structured training on improving tremor, especially in nondominant hands, has not been studied in ophthalmic surgeries. The current training methods in ophthalmology for surgical competency include the traditional apprentice model (learning by doing), wet labs, and simulators. Given the limitations of the current education methods in surgery,9,10 virtual reality–based simulators hold tremendous potential for improving resident cataract surgical training.10–12 The 2016 Accreditation Council for Graduate Medical Education (ACGME) program requirements for
Submitted: September 28, 2016 | Final revision submitted: January 27, 2017 | Accepted: January 29, 2017 From the Departments of Ophthalmology, Veterans Affairs Boston Healthcare System (Payal, Kim, Gonzalez Gonzalez, Daly), Jamaica Plain, and Boston University School of Medicine (Kim, Daly) and Harvard Medical School (Payal, Daly), Boston, Massachusetts, USA. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States Government. Supported with resources and the use of facilities at the Veterans Affairs Boston Healthcare System, Boston, Massachusetts, USA. Jose Efren Gonzalez Monroy, MD, assisted with initial institutional review board documentation and initiating study enrollment. Presented as a poster at the annual meeting of the Association for Research in Vision and Ophthalmology, Seattle, Washington, USA, May 2016. Corresponding author: Mary K. Daly, MD, Veteran Affairs Boston Healthcare System, 150 South Huntington Avenue, Suite 8C-29, Boston, Massachusetts 02130, USA. E-mail: [email protected]. Q 2017 ASCRS and ESCRS Published by Elsevier Inc.
0886-3350/$ - see frontmatter http://dx.doi.org/10.1016/j.jcrs.2017.01.021
TRAINING EFFECT ON TREMOR
ophthalmology mandate institutions to provide “a surgical skills development resource (a wet lab or simulators)” for resident education.A However, the ACGME stops short and does not mandate specific wet-lab or surgicalsimulation skills development curricula. Systematic training programs using virtual reality simulator modules can improve cataract surgery skills13 and can be as effective as wet-lab training.14 The anterior segment anti-tremor and forceps training modules of the Eyesi surgical simulator (VRMagic Holding AG) have been shown to have construct validity.15,16 It is not known whether there is a difference in improvement of tremor in dominant hands and nondominant hands after training on such modules. Alternatively, paper-based tests, such as the Archimedean spiral, are widely used by neurologists to evaluate essential tremor (a neurological disorder with postural and kinetic tremors of the upper limbs). In patients with essential tremor, dominant hands perform better than nondominant hands in spirography.17 However, the effect of paper-based training on dominant hands and nondominant hands with respect to physiologic tremor has not been studied in healthy individuals in the context of ophthalmic surgery. We compared the change in tremor and time-to-task completion in dominant hands and nondominant hands between a high-fidelity 3-dimensional (3-D) structured simulator training module and a paper-based training module (an adjunct measure of tremor). SUBJECTS AND METHODS The study was approved by the Veterans Affairs Boston Healthcare System Institutional Review Board. Informed consent was obtained from all participants. All procedures performed in the studies were in accordance with the ethical standards of the institutional research committee and with the tenets of the Declaration of Helsinki. Medical students and an ophthalmology resident reporting for their clinical elective/rotation consecutively at the Ophthalmology Department of the Veteran Affairs Boston Healthcare System participated in the study. Medical students were recruited irrespective of their career interest in surgery or ophthalmology because no difference in microsurgery performance has been reported between those with and those without a strong interest in ophthalmology.18 All participants completed a demographic questionnaire and the Waterloo Handedness Questionnaire (WHQ). The demographic questionnaire included questions on training level, surgical experience, previous simulator exposure, and lifestyle habits. The WHQ has 20 questions that assess the hand preference for daily tasks.19 Available responses were as follows: left always, left usually, equal, right usually, and right always. Each response was respectively scored as 2, 1, 0, 1, or 2 and added up to yield a number ranging from 40 (strong left-hand preference) to 40 (strong right-hand preference). Right hand was designated as dominant for those with a positive score and left hand as dominant for those with a negative score. After hand dominance was assessed, the participants were randomized into 1 of 4 groups that determined the sequence of modules they would complete and the hand they would use first (Figure 1). The groups were as follows: paper module first–dominant hand first, paper module first–nondominant hand first, simulator module first–dominant hand first, and simulator module first– nondominant hand first. Randomization was performed to equally distribute the learning effect of 1 hand on the other, following the hypothesis of Park et al.20 that the hand that performs the task first might
657
confer a learning effect on the other hand that will perform the task later. Participants received a demonstration and completed a practice module to familiarize themselves with the simulator and paper tasks. Subsequently, they completed both the paper module and simulator module as per the group to which they were randomized. Training Modules The simulator modules were completed using the Eyesi surgical simulator (software version 2.8.12). It includes a microscope that provides a stereoscopic image to the user and handheld probes that resemble surgical instruments. For this study, antitremor module level 6 was used on a cataract mannequin head (Figure 2, left). The module consisted of moving a sphere counterclockwise from center to periphery along a spiral path in the anterior segment with the tip of an instrument without damaging the lens or cornea. These abstract simulation tasks do not require specific surgical knowledge but do require discreet movements and a degree of manual dexterity. The simulator reported average tremor values and overall scores, among other parameters. Higher scores indicated better performance within a range of 0 to 100. One hundred points were given for completion, and points were deducted for errors. Possible error parameters included an outof-tolerance percentage, injury to lens or cornea, out-of-focus manipulation, working without red reflex, and excessive time use. For the paper module, an Archimedean spiral with separation distance of 2.5 mm between spiral lines was used (Figure 2, right). Participants traced a continuous line along the spiral path with a pen starting at the center while staying in between the lines. Performance was measured by counting the number of times the participant’s line crossed or touched the spiral line and the time required to complete the task. Three training sessions in each module were chosen based on the recommendation by Saleh et al.21 that at least 3 attempts should be incorporated when designing a relevant training syllabus to account for the variability in performance of trainees. Each module required approximately 1 to 2 hours for completion and a total of 4 to 5 hours for the entire study. Participants completed the study within 2 weeks. The primary outcome was the change in tremor and time-totask completion from baseline to final task in the dominant hand and the nondominant hand in the 2 training models. Secondarily, the 2 training models were compared for effectiveness in improving tremor and time-to-task completion. A subgroup analyses was performed to compare simulator scores for “strong dominants” (ie, higher than C30 and lower than 30 on the WHQ scale) with “weak dominants” (ie, lower than C10 and higher than 10). Statistical Analysis Statistical analyses were performed using JMP Pro software (version 12.0, SAS Institute, Inc.). Simple averages, standard deviations, and descriptive statistics were used for demographic data. The Wilcoxon/Kruskal-Wallis (rank sums) test was used for comparison of quantitative data between dominant hands and nondominant hands and between different steps of the training modules. The mean change in outcomes from baseline to the final test was measured by the interaction term in the repeated-measures analysis of variance. The Spearman coefficient (⍴) was used to assess the correlation between outcomes. A P value less than 0.05 was considered statistically significant.
RESULTS Twenty participants (19 fourth-year medical students, 1 final year ophthalmology resident) were enrolled in the study; 19 (95%) completed the entire study. One medical Volume 43 Issue 5 May 2017
658
TRAINING EFFECT ON TREMOR
Figure 1. Flow diagram of study steps for each participant. The letter a represents each training module (paper or simulator) comprising 3 training sessions and a posttraining test (62 spirals/module). The letter b represents the baseline test consisting of 1 spiral each for dominant hand and nondominant hand on paper and simulator (4 spirals) with each training session comprising 10 spirals for each dominant hand and nondominant hand, totaling 60 spirals (10 spirals 2 hands 3 sessions). The posttraining test consisting of 1 spiral for each hand. The letter c represents after the trainees completed the first module, when they crossed over to the other module. Upon completion of both modules, they took a final test that included 1 paper and simulator spiral per hand (4 spirals). Thus each participant, except 1 that was lost to follow-up, completed 132 spirals (62 spirals 2 modules C 4 baseline test spirals C 4 final test spirals).
student could not complete the study because of scheduling issues but completed 22 paper spirals and 2 simulator spirals. Two participants (10%) were left-handed. Three participants had experience with an ophthalmic surgical simulator, all for less than 2 hours. Ten participants (50%) played video games regularly, and 6 (30%) had experience playing a musical instrument. Average Tremor on Simulator and Errors on Paper Modules
There was moderate correlation between average tremor on the simulator anti-tremor module (3-D) and paper errors (2-dimensional) (Spearman ⍴ Z 0.35, P ! .0001). Practice on the simulator or paper modules did not reduce tremor significantly from baseline to final tasks (P Z .12, simulator; P Z .2, paper). This pattern of no reduction in tremor or paper errors from baseline to final tasks was also observed when the dominant
Figure 2. Images of simulator anti-tremor module (left) and paper module (right).
Volume 43 Issue 5 May 2017
hand and the nondominant hand were visualized separately (Figure 3). A subgroup analysis exploring tremor and paper errors in weak dominant hands (WHQ score lower than C10 and higher than 10), moderately dominant hands (between C10 and C30 and between 10 and 30), and strong dominant hands (higher than C30 and lower than 30) showed greater variability in tremor values in the weak dominant hands (Figure 4). Time-to-Task Completion
There was improvement in the time-to-task completion from baseline to final task in the simulator and paper modules. Practice on the training modules improved time-to-task completion in both simulator and paper modules (both P ! .0001). The improvement in time (time to complete final task – time to complete baseline task) was greater in the simulator module (87.6 seconds, improvement of 63% over baseline time) than in the paper module (18.71 seconds, 48%). The improvement in time from baseline to final tasks was greater in the nondominant hands in the simulator module (improvement of 64.5% over baseline time) than in the paper module (53.6% over baseline time) (Figure 5). Dominant hands also seemed to improve more in the simulator module (improvement of 60.2%) than in the paper module (52.1%). Overall, in both modules, nondominant hands performed slower than dominant hands (both P ! .0001) (Figure 5). DISCUSSION We compared change in tremor and time-to-task completion in dominant hands and nondominant hands on a surgical
TRAINING EFFECT ON TREMOR
659
Figure 3. Mean tremor (SimAvg) on simulator versus paper errors by dominant hands (D) and nondominant hands (ND).
simulator and on paper-based tasks and found that after structured training, participants were not able to reduce tremor significantly but were able to improve their times. In addition to simulator training, we included a paper module that could potentially be an inexpensive alternative for dexterity or tremor assessment. However, we did not find a strong correlation for the primary outcomes (tremor and paper errors) in the 2 training modules. This is not a
surprise because the surgical simulator traces, with high precision, the exact path a participant’s probe travels in the simulated eye and assigns a numerical value (the tremor value) to the deviation from the prescribed spiral path. In contrast, the paper module provides a much less sophisticated method of measuring tremor. Nevertheless, the correlation between the 2 modules was moderate (r Z 0.35) and statistically significant.
Figure 4. Mean tremor (SimAvg) and paper errors in weak dominant hands (WeakDom), moderately dominant hands (ModDom), and strong dominant hands (StrDom) (D Z dominant hands; ND Z nondominant hands). Volume 43 Issue 5 May 2017
660
TRAINING EFFECT ON TREMOR
Figure 5. Time-to-task completion in simulator and paper modules by hand dominance (D Z dominant hands; ND Z nondominant hands).
Practicing the structured training program on both the simulator module and the paper module did not result in significant reduction in tremor, and this effect was true for dominant hands and nondominant hands separately. This suggests that an individual baseline physiologic tremor is not reduced through training. This finding is consistent with studies that show no differences in baseline tremor between residents and attending-level physicians.5,15 In our cohort, nondominant hands had more errors than dominant hands in the paper module at the final test (P Z .02) than at baseline (P Z .23). This discrepancy in results between the simulator module and paper module could be explained as follows: In the paper module, participants were able to see the time taken to complete each spiral task but were not able to immediately see the number of errors committed. It is possible that to improve their performance, they attempted to finish the spirals quicker, thus improving their time, while inadvertently committing more errors, which they could not see immediately. This is reflected in the significant negative correlation between time and paper error at final test (r Z 0.45, P Z .004). In addition, there is a fundamental difference in the designs of the paper module and the simulator module. In the simulator anti-tremor module, the user cannot advance on the spiral unless the instrument tip is within a tolerance range. If the tip is off the spiral, the simulator essentially forces the user to return to the correct position to proceed with task completion. There is no such feedback feature in the paper module, so it would be possible for the participants to finish quicker while committing more errors. Because the simulator places more emphasis on precision, it might be more suitable than paper training for improving ophthalmic surgical skills. Real-time feedback might play an important role in improving the effectiveness of training.22 Volume 43 Issue 5 May 2017
In a subgroup analysis, we classified handedness into weak, moderate, and strong dominant based on WHQ scores and compared these categories on tremor. We found that participants with strong dominant handedness tended to have less tremor overall and less variability in tremor from baseline to final tasks than those with weak dominant handedness (Figure 4). Participants with moderate hand dominance had the lowest number of errors and lowest average tremor of the 3 hand-dominance groups. This could be related to ambidexterity and merits further investigation. In addition, on multiple comparison of tremor on all 3 levels (comparing strong to moderate, strong to weak, and moderate to weak), we found a statistically significant difference between the 3 levels (all P ! .0001). This suggests that the strength of handedness might be an important consideration in training residents. Although tremor might not decrease significantly after training, it might be worthwhile for those with WHQ scores between 10 and C10 to practice more to achieve tremor levels of their peers. Participants improved time-to-task completion in both modules in our study. Although nondominant hands were slower overall than dominant hands in simulator tasks and paper tasks, as expected, the improvement in time from baseline to the final task was remarkable in nondominant hands in the simulator module (64.5% over baseline time). This was greater than the improvement by nondominant hands in the paper module or by dominant hands in the paper or simulator module. This finding warrants further study. The limitations of our study include those related to its design and population and those related to the equipment used. The sample in our study was small; however, it was well within the range of other ophthalmic simulation studies.15,21,23–25 The time required to complete the study was a main factor contributing to fewer participants and the
TRAINING EFFECT ON TREMOR
preponderance of medical students over residents. The finalyear ophthalmology resident included in the study was likely to have more microsurgical surgery experience than most medical students. The resident was part of Group A-b in the randomization. It is unlikely that the interpretation of results would be affected by this because the comparison of outcomes between groups was not performed; rather, they were compared between type of module, hand dominance, and strength of hand dominance. In conclusion, our study found that structured training and practice might not appreciably reduce tremor but can improve speed in dominant hands and nondominant hands. A structured simulator program that focuses on improving dexterity in both hands should be considered for ophthalmology residents and fellows.
WHAT WAS KNOWN Poor hand–eye coordination and tremor are recognized issues during resident surgical training.
WHAT THIS PAPER ADDS A structured training program helped improve time-to-task completion on both the simulator module and the paper module but did not appreciably reduce tremor. Trainees with weak dominant hands (WHQ scores between 10 and C10) might need more practice and training than their peers with strong dominant hands (scores OC30, ! 30).
REFERENCES 1. Coulson CJ, Slack PS, Ma X. The effect of supporting a surgeon’s wrist on their hand tremor. Microsurgery 2010; 30:565–568 2. Binenbaum G, Volpe NJ. Ophthalmology resident surgical competency; a national survey. Ophthalmology 2006; 113:1237–1244 3. Holmes JM, Toleikis SC, Jay WM. The effect of arm exercise and ocular massage on postural hand tremor. Ann Ophthalmol 1992; 24:156–158 4. Hsu PA, Cooley BC. Effect of exercise on microsurgical hand tremor. Microsurgery 2003; 23:323–327 5. Slack PS, Coulson CJ, Ma X, Pracy P, Parmar S, Webster K. The effect of operating time on surgeon’s hand tremor. Eur Arch Otorhinolaryngol 2009; 266:137–141 6. Wharrad HJ, Birmingham AT, Macdonald IA, Inch PJ, Mead JL. The influence of fasting and of caffeine intake on finger tremor. Eur J Clin Pharmacol 1985; 29:37–43 7. Harwell RC, Ferguson RL. Physiologic tremor and microsurgery. Microsurgery 1983; 4:187–192 €rbe D, Hu €ttenbrink KB, Zahnert T, Vogel U, Tassabehji M, Kuhlisch E, 8. Mu Hofmann G. Tremor in otosurgery: influence of physical strain on hand steadiness. Otol Neurotol 2001; 22:672–677 9. Smith JH. Teaching phacoemulsification in US ophthalmology residencies: can the quality be maintained? Curr Opin Ophthalmol 2005; 16:27–32 10. Khalifa YM, Bogorad D, Gibson V, Peifer J, Nussbaum J. Virtual reality in ophthalmology training. Surv Ophthalmol 2006; 51:259–273
661
11. Belyea DA, Brown SE, Rajjoub LZ. Influence of surgery simulator training on ophthalmology resident phacoemulsification performance. J Cataract Refract Surg 2011; 37:1756–1761 12. Grodin MH, Johnson TM, Acree JL, Glaser BM. Ophthalmic surgical training: a curriculum to enhance surgical simulation. Retina 2008; 28:1509–1514 13. Bergqvist J, Person A, Vestergaard A, Grauslund J. Establishment of a validated training programme on the Eyesi cataract simulator. A prospective randomized study. Acta Ophthalmol 2014; 92:629–634. Available at: http://onlinelibrary.wiley.com/doi/10.1111/aos.12383/epdf. Accessed March 17, 2017 14. Daly MK, Gonzalez E, Siracuse-Lee D, Legutko PA. Efficacy of surgical simulator training versus traditional wet-lab training on operating room performance of ophthalmology residents during the capsulorhexis in cataract surgery. J Cataract Refract Surg 2013; 39:1734– 1741 15. Mahr MA, Hodge DO. Construct validity of anterior segment antitremor and forceps surgical simulator training modules: attending versus resident surgeon performance. J Cataract Refract Surg 2008; 34:980–985 16. Le TDB, Adatia FA, Lam W-C. Virtual reality ophthalmic surgical simulation as a feasible training and assessment tool: results of a multicentre study. Can J Ophthalmol 2011; 46:56–60 17. Ondo WG, Wang A, Thomas M, Vuong KD. Evaluating factors that can influence spirography ratings in patients with essential tremor. Parkinsonism Relat Disord 2005; 11:45–48 18. Gillan SN, Okhravi N, O’Sullivan F, Sullivan P, Viswanathan A, Saleh GM. Influence of medical student career aims on ophthalmic surgical simulator performance (part of the international forum for ophthalmic simulation studies). Br J Ophthalmol 2016; 100:411–414 19. Steenhuis RE, Bryden MP, Schwartz M, Lawson S. Reliability of hand preference items and factors. J Clin Exp Neuropsychol 1990; 12:921–930. Available at: https://www.ucl.ac.uk/medical-education/resources/Waterloo/ WaterlooHandQuese32itemsSteenhuis1990.pdf. Accessed March 17, 2017 20. Park J, Williams O, Waqar S, Modi N, Kersey T, Sleep T. Safety of nondominant-hand ophthalmic surgery. J Cataract Refract Surg 2012; 38:2112–2116 21. Saleh GM, Theodoraki K, Gillan S, Sullivan P, O’Sullivan F, Hussain B, Bunce C, Athanasiadis I. The development of a virtual reality training programme for ophthalmology: repeatability and reproducibility (part of the International Forum for Ophthalmic Simulation Studies). Eye 2013; 27:1269–1274. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3831124/pdf/eye2013166a.pdf. Accessed March 17, 2017 €lscher AH, 22. Kleinert R, Plum P, Heiermann N, Wahba R, Chang DH, Ho Stippel DL. Embedding a virtual patient simulator in an interactive surgical lecture. J Surg Educ 2016; 73:433–441 23. Saleh S, Uppal S, Sharma V. Greater nondominant hand proficiency is not associated with enhanced simulated surgical performance. Can J Ophthalmol 2015; 50:350–353 24. Selvander M, Åsman P. Cataract surgeons outperform medical students in Eyesi virtual reality cataract surgery: evidence for construct validity. Acta Ophthalmol 2013; 91:469–474. Available at: http://onlinelibrary.wiley. com/doi/10.1111/j.1755-3768.2012.02440.x/pdf. Accessed March 17, 2017 25. Privett B, Greenlee E, Rogers G, Oetting TA. Construct validity of a surgical simulator as a valid model for capsulorhexis training. J Cataract Refract Surg 2010; 36:1835–1838 OTHER CITED MATERIAL A. Accreditation Council for Graduate Medical Education. ACGME Program Requirements for Graduate Medical Education in Ophthalmology. Available at: http://www.acgme.org/acgmeweb/Portals/0/PFAssets/ProgramRequi rements/240_ophthalmology_2016.pdf. Accessed March 17, 2017
Disclosure: None of the authors has a financial or proprietary interest in any material or method mentioned.
Volume 43 Issue 5 May 2017