Background
The role of transesophageal echocardiography (TEE) simulation in cardiology fellows’ learning is unknown. Standard TEE training at the authors’ institution occurs during the second of 3 clinical years. Fellows spend 2 months in the TEE laboratory learning through hands-on experience. The addition of TEE simulation to this experience may improve proficiency, speed learning, and increase fellows’ comfort with TEE. This study was designed to compare methods of TEE simulator training with standard training.
Methods
Group A ( n = 8) consisted of fellows who had completed standard TEE training. Fellows starting their second clinical year were randomly assigned to group B ( n = 10), simulator training during month 1, or group C ( n = 9), simulator training during month 2. All groups completed 2 months of standard TEE training. All groups underwent assessment of TEE performance and a self-assessment of ability and comfort level with TEE.
Results
Groups B and C had higher total assessment scores than group A. Groups B and C had higher numbers of views achieved without assistance ( P = .01). After month 1, group B had higher total scores and number of views achieved without assistance compared with group C ( P = .02 and P = .02, respectively). The length of time of the examination tended to be lower for group B, and fellows in group B had greater comfort with TEE than those in group C ( P = .01).
Conclusions
These data suggest that TEE simulator training improves proficiency and helps speed learning and comfort with TEE.
Simulation-based training has emerged as a valuable tool in the medical field for the development and assessment of technical skills and knowledge among trainees. Initially developed for the training of pilots and military personnel, simulation is increasingly being used by residency and fellowship programs in areas such as surgery, anesthesiology, emergency medicine, and cardiology. Trainees in these procedural specialties have traditionally acquired the requisite technical skills by performing procedures on patients under the supervision of experienced physicians. Simulation-based training enables these trainees to develop and practice their skills before applying these techniques in a patient setting. In cardiovascular medicine, simulation training has been developed for coronary angiography, electrophysiologic applications such as transseptal catheterization, and echocardiography. Procedural competency in transesophageal echocardiography (TEE) is a required component of our cardiovascular medicine fellowship training. Our observation has been that the learning curve for the acquisition of technical expertise varies substantially among fellows, and a method of speeding and standardizing this process would be beneficial.
The role of TEE simulation in cardiovascular medicine trainees’ learning and performance is unknown. We hypothesized that TEE simulator training among cardiovascular medicine fellows is superior to standard training and will improve proficiency in the performance of TEE, speed learning, and increase trainee comfort with the procedure.
Methods
TEE Simulator
The HeartWorks TEE simulator (Inventive Medical, Ltd., London, UK) was used in this study ( Figure 1 ). The simulator consists of a manikin upper-body torso with a transesophageal echocardiographic probe inserted into the mouth of the manikin; the probe is connected to a dedicated computer with proprietary software. The probe has authentic controls with standard flexion, rotation, and angulation capabilities of a clinical multiplane probe. As the cardiovascular medicine fellow performs an examination, a simulated ultrasound image of the heart is generated on one part of the screen with a three-dimensional virtual heart adjacent to it. The three-dimensional heart model can be manipulated to display “slices” through the heart that correspond to the accompanying ultrasound-generated images. The simultaneous viewing of the two-dimensional ultrasound image and the three-dimensional anatomy of the heart promotes development of the requisite spatial-anatomic knowledge for transesophageal echocardiographic examinations.
Study Design
The study was approved by the institutional review board at Vanderbilt University Medical Center. Participation was voluntary, and all eligible cardiovascular medicine fellows provided consent to participate in the study. The study flowchart is shown in Figure 2 . The standard TEE training at our institution occurs during the second of 3 clinical years. During this year, the fellows rotate through the TEE laboratory for 2 months, learning the procedure through direct hands-on experience. Group A was composed of fellows who were completing the second year of clinical training and had completed our standard TEE curriculum ( n = 8). This group completed this training before our access to the TEE simulator and served as historical controls. Upon completion of standard TEE training, fellows in group A underwent a one-time assessment of their performance by conducting a comprehensive transesophageal echocardiographic examination on an actual patient as described below.
Cardiovascular medicine fellows who were beginning their second clinical year and had no prior formal TEE instruction were randomized into two groups, group B ( n = 10) and group C ( n = 9). These trainees completed our standard TEE training program but also had an opportunity to train with the TEE simulator during 1 of their 2 rotation months. At the beginning of their first TEE rotation month, group B received instruction on the use of the TEE simulator by an attending physician. Fellows in group B were expected to practice their procedural skills using the simulator for ≥2 hours per week in addition to their clinical duties of performing TEE on scheduled inpatients and outpatients. During their second TEE rotation month, fellows in group B performed their usual clinical duties and standard training. Group C completed standard TEE training during their first rotation month. They were then instructed on the TEE simulator and asked to use this tool for training during their second rotation month.
At the conclusion of each of their two rotation months, fellows in groups B and C underwent a formal assessment of their ability to perform a complete transesophageal echocardiographic examination on an actual patient and also completed a self-assessment questionnaire. The performance evaluation ( Appendix A ) and the self-assessment questionnaire ( Appendix B ) were the same across groups A, B, and C.
The transesophageal echocardiographic examination consisted of 37 views or structures and awarded 2, 1, or 0 points for each view obtained without instruction, obtained with instruction, or not obtained, respectively. The maximum score achievable was 74 points ( Figure 3 ). Fellows were allowed to perform the transesophageal echocardiographic examinations in their preferred sequences. If they attempted and were unable to achieve a view, verbal instruction was given. If achieved, this was deemed to be “achieved with instruction.” If the view could still not be obtained, this was deemed to be “not achieved.” If at the completion of the study, a view had not been attempted (not included in the fellow’s complete study), the fellow was asked to obtain the view, and this was deemed to be “not achieved.” Doppler assessment was required for the view to be deemed “achieved” for structures in which this assessment would be considered part of a complete evaluation (i.e., color Doppler of the mitral valve). The time to completion of the entire examination was recorded by the attending physician supervising the study. The skills assessments were performed by the same two faculty members to increase the consistency of the examinations, and the format of the examination as described above was followed for each assessment. The assessments were performed on patients referred for TEE, who provided a high probability of being able to complete a full examination. Because of logistic constraints, the faculty was not blinded as to use of the simulator. Fellows also completed a self-assessment questionnaire rating their ability and comfort levels with the procedure. The self-assessment questionnaire was graded on a scale of 1 to 5 and included questions regarding confidence level with the performance and interpretation of the examination (5 being most confident) and the utility of simulator training (5 being most useful).
The primary end points of our study were the difference in total and view-specific transesophageal echocardiographic examination scores between fellows who completed our standard training (group A) and simulator-based training (groups B and C) and the difference in total and view-specific examination scores between groups B and C after TEE-based simulator training during the first month versus the second month of their clinical training.
Statistical Analysis
We compared TEE simulator view assessment scores as the outcomes. Categorical variables are expressed as percentages. Wilcoxon’s tests were used for comparison of continuous variables among groups; all results are expressed as median scores. P values < .05 were considered statistically significant. All statistical analysis was conducted using R version 2.13.1 (R Foundation for Statistical Computing, Vienna, Austria).
Results
After completing TEE training, groups B and C (those with simulator training) had higher total assessment scores than group A (those without simulator training) (median, 72 vs 68.5; Figure 4 A). Groups B and C also had higher numbers of views achieved without assistance and lower numbers of views that were not achieved (median, 35 vs 33 [ P = .01] and 0 vs 2 [ P = .03], respectively; Figure 4 B). There was a significant difference in score for groups B and C for deep esophageal views as well as transgastric views compared with group A (median, 6 vs 5.5 [ P = .02] and 13 vs 9.5 [ P = .04], respectively; Figure 4 B). There was no significant difference in scores for midesophageal or aortic views. There was also no significant difference between the fellows’ perceptions of the TEE training, comfort with performing TEE, or perception of ability to perform or interpret the results of TEE. Although not statistically significant, the mean time for completion of TEE tended to be lower for groups B and C (15.9 min) compared with group A (17 min).
When groups B and C were compared, group B had higher total scores after month 1 compared with group C ( P = .02), which may have been due to higher midesophageal and gastric view scores ( P = .01 and P = .02, respectively) ( Figure 5 A). The number of views achieved without assistance was different ( P = .02), and the number of views achieved with instruction was lower for group B compared with group C ( P = .01) ( Figure 5 B). The length of time of the examination also tended to be lower for group B. In addition, group B trainees, who had been exposed to simulator-based training during their first month, were able to maintain their skill set from month 1 to month 2, as measured by a smaller change in total score compared with group C ( Figure 6 ). Fellows in group B appeared to have greater comfort with TEE and with the interpretation of normal transesophageal echocardiographic results than those in group C ( P = .01 and P = .01, respectively). Although not statistically significant, group B tended to have greater comfort with interpretation of abnormal transesophageal echocardiographic results and more confidence in their ability to perform TEE.
The amount of time actually spent on the simulator was self-reported and averaged 1 hour per week, with a range of 0.25 to 2 hours per week. The mean number of transesophageal echocardiographic studies performed did not vary significantly among groups (group A, 92; group B, 104; and group C, 86). No relationship was found between assessment scores and prior physics education, interest in a career in imaging, number of months between TEE rotation months, or number of months spent on transthoracic echocardiography training.
Results
After completing TEE training, groups B and C (those with simulator training) had higher total assessment scores than group A (those without simulator training) (median, 72 vs 68.5; Figure 4 A). Groups B and C also had higher numbers of views achieved without assistance and lower numbers of views that were not achieved (median, 35 vs 33 [ P = .01] and 0 vs 2 [ P = .03], respectively; Figure 4 B). There was a significant difference in score for groups B and C for deep esophageal views as well as transgastric views compared with group A (median, 6 vs 5.5 [ P = .02] and 13 vs 9.5 [ P = .04], respectively; Figure 4 B). There was no significant difference in scores for midesophageal or aortic views. There was also no significant difference between the fellows’ perceptions of the TEE training, comfort with performing TEE, or perception of ability to perform or interpret the results of TEE. Although not statistically significant, the mean time for completion of TEE tended to be lower for groups B and C (15.9 min) compared with group A (17 min).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
