Keywords
ophthalmology - simulation - training
Training in surgical repair of ocular trauma is of utmost importance to the ophthalmologist.
This is true in the civilian world, but of even more critical importance in the deployed
military ophthalmologist. Data suggest involvement of the eye is relatively commonplace
in traumatic incidents. Twenty-six percent of survivors of the terrorist attack on
the World Trade Center in New York City in 2001 and 8% of survivors of the Oklahoma
City bombing in 1995 sustained ocular injuries.[1]
[2] Among the military population, ocular injuries accounted for the fourth most common
injury during Operation Iraqi Freedom and Operation Enduring Freedom.[3] The most frequently encountered ocular injuries included lid and brow lacerations,
open globe injuries, and orbital fractures.
Training in techniques to repair such injuries has typically been accomplished with
the use of animal tissue models. Porcine eye and adnexal tissue have been demonstrated
to be histologically similar to human eye tissue and generally suitable for surgical
training.[4] However, histologic similarity does not necessarily correlate directly with fidelity
to human tissue, particularly in anatomic similitude and tissue characteristics. This
applies not just to the eyelid but also to globe models as well. Advances in surgical
simulation systems in the last several years have allowed training physicians the
opportunity to hone skills in a simulated environment, without the risk of injury
to patients, or the reliance on living tissue to facilitate training.[5]
[6] Ideally, the most desirable feature should be high simulator-to-human similarity,
which is a goal of simulation training. As the Department of Defense has directed
the reduction in use of animals for medical education training when “alternative methods
produce scientifically or educationally valid or equivalent results,”[7] simulation-based systems have become increasingly necessary as alternate training
platforms. Studies into the effectiveness of surgical simulators in phacoemulsification
training for resident physicians have previously demonstrated improvements in speed
of skill progression, and decreased errant capsulorrhexis rates.[8]
[9] Here, we describe a study into the efficacy of one such simulation system in comparison
to living tissue models in training of repair to marginal eyelid lacerations.
Methods
The Uniformed Services University of the Health Sciences (USUHS) Institutional Review
Board (IRB), USUHS Institutional Animal Care and Use Committee and the United States
Army Medical Research and Material Command IRB reviewed and approved the study prior
to initiation.
The study comprised novice surgeons training in military and civilian ophthalmology
residency programs, as well as board-certified and board-eligible ophthalmologists,
all chosen from a convenience sample of residents and staff physicians taking part
in the annual Tri-Service Ocular Trauma Surgery Laboratory (OTSL) over a period of
3 years. The OTSL is an annual training course for military ophthalmologists incorporating
didactic, Socratic, and hands-on methods to teach established strategies for diagnosing
and treating ocular trauma. In each of the 3 years, half of the participating residents
were randomly assigned to receive training in surgical repair of either eyelid margin
lacerations or corneal lacerations ([Fig. 1]). In this article, we focus on data pertaining to those in the eyelid repair group,
as the corneal trial is ongoing. Residents (n = 43) were second- and third-year ophthalmology residents, whereas staff physicians
(n = 16) were either general comprehensive ophthalmologists or subspecialty trained
and certified in oculoplastic surgery. Staff physicians taking part were evaluated
in the second and third years of the study in an effort to evaluate simulator effect
on skill level of presumed expert surgeons.
Fig. 1 Resident study structure, years 1–3.
Participants performed an initial evaluation of surgical skill (before training repair
[BTR]) by performing an unassisted repair of a 10-mm linear, full-thickness laceration
involving the upper lid margin on an exenterated pig eye model. After BTR time to
completion was recorded, resident participants were stratified according to the median
time, which was deemed the group's baseline “time required to close wound.” This measure
served as a surrogate for baseline surgical skill level. Half of each group of residents
was randomly allocated using a random number generator to receive instruction via
traditional live tissue-based instruction (LIVE, n = 21), or to a surgical simulator (SIM, n = 22). Subjects were graded by three masked observers using the median score as the
final score for analysis on several metrics. Residents then received instruction limited
to 90 minutes while using either traditional LIVE materials or SIM material. Following
the instruction period, residents then repeated the assessment on the pig eye with
evaluation of metrics once again (post-training repair [PTR]).
Staff physicians were randomly divided into two groups of eight. Each group consisted
of two general ophthalmologists and two oculoplastic fellowship-trained ophthalmologists.
Both groups performed BTR. One group randomly received SIM training, while the control
group received LIVE training. Each group was then assessed again for their PTR results
using the same metrics as for the resident group.
Materials
Training for repair of marginal eyelid lacerations took place on either a traditional
animal tissue model (LIVE) or a surgical simulator (SIM). LIVE training was accomplished
using swines (Sus scrofa domestica), male or female, weighing 35 to 50 lb. Each animal was placed under general anesthesia
and maintained under a surgical plane of anesthesia for the entire duration of the
procedure. Animals were positioned and a retrobulbar block (1:1 mixture of 2% lidocaine
and 0.75% bupivacaine) was performed prior to the procedure. Two percent (2%) lidocaine
with 1:100,000 epinephrine was injected into the eyelid for hemostasis and a 10-mm
full-thickness superior eyelid laceration was created. Students were then instructed
by staff prior to and during the surgical repair with regard to proper preparation
and technique. Participants selected to receive training with a surgical simulator
used the Ocular and Craniofacial Trauma Treatment Training System (Medical Device
and Simulation Laboratory [formerly Simulation Group] at Massachusetts General Hospital,
Boston, MA). The Ocular and Craniofacial Trauma Treatment Training System is a silicone-based
mannequin simulator based on physically accurate anatomy and uses authentic surgical
instruments to repair typical eyelid wounds. The device is composed of a mannequin
head with replaceable eye and lid modules ([Fig. 2]). It uses optical and magnetic tracking systems to record instrument and hand motions,
and algorithms to compare user performance with reference expert gestures. The simulator
is fully developed to simulate eyelid injuries and subsequent repairs. Use of the
microscope head is optional; surgical loupes can also be utilized during training.
Normative data exist for expert surgeons to allow for direct comparisons. The simulator
captures the following output metrics for repairing eyelid lacerations: time of procedure,
suture path length, and compactness of motion. Instruction was provided to the individual
learners via an oculoplastic-trained ophthalmologist following BTR analysis.
Fig. 2 Ocular and Craniofacial Trauma Treatment Training System with optional microscope
attachment.
Metrics
Metrics of BTR and PTR analysis included successful repair of the eyelid laceration,
number of sutures required to close the laceration, time to repair the wound, tarsal
plate reconstruction, and gray line approximation. Successful repair of the laceration
was scored as a categorical value, yes or no. Participants either achieved or did
not achieve surgically appropriate approximation of the laceration as evaluated by
the staff grader. Number of sutures needed to close the laceration was recorded as
a discrete ordinal value. While there is no correct number of sutures needed to close
any wound, the number of sutures necessary to close a wound is an independent variable
in efficacy of a wound closure. More sutures require more time and potentially more
variability in other objective measures, while fewer sutures may lead to higher rate
of wound reopening and/or increased suture tension, which would cause increased eyelid
deformity and loss of function. Time required to repair the laceration was recorded
as a continuous variable measured in seconds. This metric is intended to measure a
surgeon's efficiency in surgical repair. Tarsal plate reconstruction was graded on
an ordinal scale with four grades: 1 = poor, 2 = fair, 3 = good, 4 = excellent as
graded by the course instructors. A poor grade (1) would consist of failure to oppose
tarsal tissue from end-to-end. A fair grade (2) would result from a satisfactory repair
with an end-to-end repair, but suture pass inconsistency (irregular spacing, irregular
depths, etc.) was evident in one or more sutures. A good grade (3) demonstrates satisfactory
repair and reveals more than 50% consistency in suture passes for all sutures placed.
An excellent grade (4) demonstrates adequate tarsal plate closure with 100% consistency
in suture passes for all sutures placed. Finally, gray line approximation was measured
in a similar fashion to tarsal plate reconstruction as an ordinal scale with four
grades: 1 = poor, 2 = fair, 3 = good, 4 = excellent as graded by the course instructors.
A poor grade (1) demonstrates failure of the student to perform the task. A fair grade
(2) reveals success in closing the eyelid skin, but the eyelid edges are not opposed
at the eyelid margin. A good grade (3) would consist of appropriate approximation
of the gray line, but skin is not opposed on either side. An excellent grade (4) reveals
100% consistency in suture placement with no noticeable deficiencies and careful approximation
of the gray line.
Grading was completed by three independent board-certified ophthalmologists. Each
was provided standard descriptions correlating with each level of repair. Two graders
were used as the primary scorers, while the third grader served as a tiebreaker in
the event of disparate scoring between graders.
In addition to graded metrics, participants in the course were given a survey to assess
level of training, prior simulator experience, and prior surgical experience. Participants
were also given a selection of questions to assess their experience with the simulator
and its usefulness in training and maintenance of skills.
Results
Participant Demographics
Demographics and survey of prior simulator and eyelid repair experience of resident
study participants are demonstrated in [Table 1]. Between the SIM and LIVE groups, gender, age, and number of prior eyelid repairs
were largely similar among residents. However, simulator experience was variable between
groups. In general, prior volume of experience favored the high end of the scale with
59% of SIM and 65% of LIVE participants having spent more than 11 hours on the simulator,
and 54.5% of SIM and 50% of LIVE participants having repaired more than four lids
in the past.
Table 1
Resident group demographics and survey of prior simulation and eyelid repair experience
|
Characteristics of resident oculoplastic training group
|
|
Demographics
|
SIM training (n = 22)
|
LIVE training
(n = 21)
|
|
Males
|
11
|
13
|
|
Females
|
11
|
8
|
|
Age
|
|
26–35
|
22
|
19
|
|
36–45
|
0
|
2
|
|
Status
|
|
2nd year resident
|
18
|
17
|
|
3rd year resident
|
4
|
4
|
|
Prior SIM use (hours)
|
|
0–3
|
6
|
1
|
|
4–10
|
3
|
6
|
|
11–20
|
9
|
10
|
|
> 20
|
4
|
3
|
|
No. of prior eyelid repairs
|
|
0
|
1
|
1
|
|
1–3
|
9
|
9
|
|
4–10
|
9
|
7
|
|
> 10
|
3
|
3
|
Abbreviations: LIVE, live tissue; SIM, simulator.
Demographics and experience data for expert participants are presented in [Table 2]. With the exception of a larger age range, demographic and experience data are largely
similar between SIM and LIVE groups.
Table 2
Expert group demographics and survey of prior simulation and eyelid repair experience
|
Characteristics of expert oculoplastic training group
|
|
Demographics
|
SIM training (n = 8)
|
LIVE training (n = 8)
|
|
Males
|
7
|
7
|
|
Females
|
1
|
1
|
|
Age
|
|
26–35
|
1
|
3
|
|
36–45
|
3
|
1
|
|
46–54
|
3
|
2
|
|
55–64
|
1
|
1
|
|
> 65
|
0
|
1
|
|
Status
|
|
Comprehensive
|
0
|
1
|
|
Subspecialty-trained
|
8
|
7
|
|
Prior SIM use (hours)
|
|
0–3
|
6
|
4
|
|
4–10
|
0
|
0
|
|
11–20
|
0
|
2
|
|
> 20
|
2
|
2
|
|
No. of prior eyelid repairs
|
|
0
|
0
|
1
|
|
1–3
|
0
|
3
|
|
4–10
|
3
|
1
|
|
> 10
|
5
|
3
|
Abbreviations: LIVE, live tissue; SIM, simulator.
Resident Improvement with Training (within-Group Analysis)
Pre- and post-training metrics for SIM and LIVE groups were compared and are presented
in [Table 3]. There was a significant difference in margin approximation among the SIM group
(BTR: 2.0, PTR: 3.0; p = 0.03). SIM group metrics for residents were not significantly different from pre-
to post-training evaluation in number of sutures, time to completion, or tarsal plate
reconstruction. There was no difference in successful completion of the exercise in
either SIM or LIVE group when comparing pre- and post-training metrics.
Table 3
Within-group analysis of resident scoring metrics, baseline versus post-training,
SIM and LIVE groups
|
Resident group evaluation: within-group metrics
|
|
Measures
|
SIM (n = 22)
|
p-Value
|
LIVE (n = 21)
|
p-Value
|
|
Median (IQR)
|
Baseline
|
Post-training
|
Baseline
|
Post-training
|
|
No. of sutures
|
5.0
(4.0–6.3)
|
5.0
(5.0–6.0)
|
0.28
|
6.0
(5.0–6.0)
|
5.0
(5.0–6.0)
|
0.86
|
|
Time (s)
|
1,577
(1,070–1,892)
|
1,740
(1,110–2,067)
|
0.77
|
1,310
(1,147–2,076)
|
1,412
(1,266–1,953)
|
0.50
|
|
Tarsal plate reconstruction[a]
|
1.0
(1.0–2.3)
|
1.0
(1.0–2.0)
|
0.48
|
1.0
(1.0–2.0)
|
1.0
(1.0–2.0)
|
0.99
|
|
Margin approximation[a]
|
2.0
(1.0–3.0)
|
3.0
(2.0–3.0)
|
0.03
|
2.0
(1.0–3.0)
|
2.0
(2.0–3.0)
|
0.26
|
|
Completed successfully
|
54.7% (12)
|
68.2% (15)
|
0.54
|
38.1% (8)
|
47.6% (10)
|
0.76
|
Abbreviations: IQR, interquartile range; LIVE, live tissue; SIM, simulator.
a Scale: 1 (poor) to 4 (excellent).
LIVE group metrics were not significantly different from pre- to post-training evaluation
in any metric studied. There was no significant difference in margin approximation
in the LIVE group as was demonstrated in the SIM group.
Resident LIVE versus SIM (between-Group Analysis)
When comparing pre-training evaluation of SIM versus LIVE participants ([Table 4]), there was no significant difference noted in any of the measured metrics. When
the same metrics were compared between groups following training, there was a significant
difference in margin approximation (SIM: 3.0, LIVE: 2.0, p = 0.03). Post-training evaluation of number of sutures, time to completion, and tarsal
plate reconstruction failed to demonstrate significant differences. Overall, there
was no significant difference in successful exercise completion when comparing SIM
versus LIVE groups in post-training metrics.
Table 4
Between-group analysis of resident pre-training and post-training scoring metrics,
SIM versus LIVE
|
Resident group evaluation: between-group metrics
|
|
SIM (n = 22)
|
LIVE (n = 21)
|
p-Value
|
|
Pre-training
|
|
No. of sutures required to close wound
|
5.0 (4.0–6.3)
|
6.0 (5.0–6.0)
|
0.48
|
|
Time (s)
|
1,577 (1,070–1,892)
|
1,310 (1,147–2,076)
|
0.90
|
|
Tarsal plate reconstruction[a]
|
1.0 (1.0–2.3)
|
1.0 (1.0–2.3)
|
0.86
|
|
Margin approximation[a]
|
2.0 (1.0–3.0)
|
2.0 (1.0–3.0)
|
0.45
|
|
Completed successfully
|
54.5% (12)
|
38.1% (8)
|
0.36
|
|
Post-training
|
|
No. of sutures required to close wound
|
5.0 (5.0–6.0)
|
5.0 (5.0–6.0)
|
0.85
|
|
Time (s)
|
1,713 (1,100–2,067)
|
1,668 (1,266–1,953)
|
0.50
|
|
Tarsal plate reconstruction[a]
|
1.0 (1.0–2.0)
|
1.0 (1.0–2.0)
|
0.58
|
|
Margin approximation[a]
|
3.0 (2.0–3.0)
|
2.0 (2.0–3.0)
|
0.03
|
|
Completed successfully
|
68.2% (15)
|
47.6% (10)
|
0.22
|
Abbreviations: LIVE, live tissue; SIM, simulator.
a Scale: 1 (poor) to 4 (excellent).
Expert Progression (within-Group Analysis)
[Table 5] presents pre- and post-training metrics for expert participants within each SIM
or LIVE group. SIM group metrics for expert participants were not significantly different
from pre- to post-training evaluation in any metric studied.
Table 5
Within-group analysis of expert scoring metrics, baseline versus post-training, SIM
and LIVE groups
|
Oculoplastic expert group evaluation: within-group metrics
|
|
Measures
|
SIM (n = 8)
|
p-Value
|
LIVE (n = 8)
|
p-Value
|
|
Median (IQR)
|
Baseline
|
Post-training
|
Baseline
|
Post-training
|
|
No. of sutures
|
5.0
(3.3–5.0)
|
5.0
(4.0–5.0)
|
0.99
|
5.0
(4.3–5.8)
|
5.0
(5.0–6.0)
|
0.32
|
|
Time (s)
|
1,085
(605–1,582)
|
1,122
(555–1,620)
|
0.67
|
1,334
(929–2,160)
|
1,342
(906–1,838)
|
0.09
|
|
Tarsal plate reconstruction[a]
|
1.5
(1.0–4.0)
|
2.0
(1.0–4.0)
|
0.32
|
1.5
(1.0–2.0)
|
1.5
(1.0–3.8)
|
0.16
|
|
Margin approximation[a]
|
2.5
(1.3–3.8)
|
2.0
(1.0–3.8)
|
0.16
|
2.5
(2.0–3.0)
|
3.0
(2.3–3.8)
|
0.32
|
|
Completed successfully
|
62.5% (5)
|
50% (4)
|
0.99
|
62.5% (5)
|
75% (6)
|
0.99
|
Abbreviations: IQR, interquartile range; LIVE, live tissue; SIM, simulator.
a Scale: 1 (poor) to 4 (excellent).
Similarly, LIVE group metrics were not significantly different from pre- to post-training
evaluation in any metric studied.
Expert Progression (between-Group Analysis)
Pre- and post-training metrics for SIM versus LIVE groups in expert participants are
presented in [Table 6].
Table 6
Between-group analysis of expert pre-training and post-training scoring metrics, SIM
versus LIVE
|
Oculoplastic expert group evaluation: between-group metrics
|
|
SIM (n = 8)
|
LIVE (n = 8)
|
p-Value
|
|
Pre-training
|
|
No. of sutures required to close wound
|
5.0 (3.3–5.0)
|
5.0 (4.3–5.8)
|
0.44
|
|
Time (s)
|
1,085 (605–1,582)
|
1,334 (929–2,160)
|
0.20
|
|
Tarsal plate reconstruction[a]
|
1.5 (1.0–4.0)
|
1.5 (1.0–2.0)
|
0.57
|
|
Margin approximation[a]
|
2.5 (1.3–3.8)
|
2.5 (2.0–3.0)
|
0.88
|
|
Completed successfully
|
62.5% (5)
|
62.5% (5)
|
0.99
|
|
Post-training
|
|
No. of sutures required to close wound
|
5.0 (4.0–5.0)
|
5.0 (5.0–6.0)
|
0.13
|
|
Time (s)
|
1,122 (555–1,620)
|
1,342 (906–1,838)
|
0.44
|
|
Tarsal plate reconstruction[a]
|
2.0 (1.0–4.0)
|
1.5 (1.0–3.8)
|
0.80
|
|
Margin approximation[a]
|
2.0 (1.0–3.8)
|
3.0 (2.3–3.8)
|
0.33
|
|
Completed successfully
|
50% (4)
|
75% (6)
|
0.61
|
Abbreviations: LIVE, live tissue; SIM, simulator.
a Scale: 1 (poor) to 4 (excellent).
Subjective Participant Scores and Comments
[Table 7] shows OTSL participant responses to the end of training survey. Responses were generally
favorable toward the simulator modality. Seventy-nine percent of respondents felt
the simulator was helpful in teaching skills, and 83% stated they would use the simulator
to maintain skills. Fifty-five percent of respondents felt the simulator was comparable
for some or all of the tasks. However, when asked if the simulator could substitute
for animal tissue on some or all metrics, 41% responded favorably, while 59% collectively
replied “somewhat” or “no.”
Table 7
OTSL participant survey responses
|
OTSL survey responses
|
|
Question
|
Number of OTSL survey respondents (n = 29)
|
|
Yes
|
Somewhat
|
No
|
No response
|
|
Was the simulator comparable to animal tissue for all/some of the tasks?
|
16
(55%)
|
9
(31%)
|
4
(14%)
|
0
|
|
Was simulator training helpful in teaching all/some of the skills?
|
23
(79%)
|
6
(21%)
|
0
|
0
|
|
Can simulator training substitute animal tissue training on all/some of the metrics?
|
12
(41%)
|
6
(21%)
|
11
(38%)
|
0
|
|
Are you likely to use the simulator, if it were available, to maintain your skills?
|
24
(83%)
|
–
|
4
(14%)
|
1
|
Abbreviation: OTSL, Ocular Trauma Surgery Laboratory.
Discussion
As advancements in technology allow ever-increasing accuracy in simulation of real-life
surgical situations, continual evaluation of training methodology is of the utmost
importance. Force readiness in today's combat environment necessitates well-trained
ophthalmic surgeons capable of handling large volumes of complicated eye trauma. In
an effort to continually improve readiness, ophthalmologists need to have reliable,
ethical, reproducible, and measureable methods to practice complex surgical repair.
In this study, we have demonstrated that simulation technology is noninferior to live
animal tissue for training ophthalmologists in marginal eyelid laceration repair.
With the exception of marginal approximation in the SIM group, within-group analysis
among resident participants in the OTSL course demonstrated no significant difference
in metric and task completion within SIM or LIVE groups when pre- and post-training
scores were compared.
Between-group analysis of SIM versus LIVE training when comparing pre- and post-training
scores for resident course participants demonstrated no significant difference between
metric and task completion scoring, with exception of margin approximation.
These findings demonstrate efficacy of SIM training for marginal eyelid laceration
repair when viewed in light of results obtained for the gold standard LIVE training.
Indeed, our results even suggest possible superiority in the marginal approximation
metric. Given our small sample size, it is possible further study with larger numbers
of participants may yield additional data supporting metric-based superiority of SIM
training. However, at this stage, we can reasonably conclude that SIM training is
noninferior to LIVE training for marginal eyelid laceration repair among novice surgeons,
and is therefore warranted in its continued use and development.
Within-group analysis of SIM and LIVE training, and between-group analysis of SIM
versus LIVE training, among board-certified ophthalmologists and oculoplastic experts
demonstrated no significant difference when comparing pre- and post-training metrics
and task completion. Again, these results are supportive of continued use and development
of simulated training systems for marginal eyelid repair. Given the presumed high
level of surgical skill among comprehensive ophthalmologists and oculoplastic experts,
significant pre- and post-training scoring differences are not expected. As LIVE training
is the gold standard, our results show that SIM training is not significantly different
as a training methodology when limitations in basic surgical skill are eliminated
with expert participants.
Interestingly, among both expert and resident groups, tarsal plate reconstruction
and margin approximation scores were low. Indeed, tarsal plate approximation was never
graded as good or excellent on the four-point scoring metric in either the resident
or expert group. This may reflect inadequacy of the porcine model as the final test
model, though it remains the gold standard. Anatomical variation between species may
account for this inadequacy, even if the tissue characteristics of porcine skin are
more similar to human. Given these concerns for porcine to human fidelity, future
studies could examine outcomes with the final test model as the mannequin lid, which
is designed to be more anatomically correct, regardless of other tissue characteristics.
OTSL participants largely reported a positive experience with simulator training,
and a desire for its continued use in skill maintenance. Among the survey questions
asked, the two most positively responded to questions were “Was simulator training
helpful in teaching all/some of the skills?” and “Are you likely to use the simulator,
if it were available, to maintain your skills?” Participants' response was less enthusiastic
when asked about simulator training as a substitute for animal training, and the comparable
quality of simulator tissue to animal tissue. Indeed, 59% of respondents stated that
simulator training was somewhat or not able to substitute for animal training.
These results indicate that while simulator technology needs improvement for more
lifelike tissue and subsequent use as a substitute for animal training, simulator
technology is viewed favorably and likely to have continued use. As both simulator
and artificial tissue technologies continue to improve, their ability to substitute
for traditional animal-based training is expected to improve as well. Given the desire
to move away from animal-based training, it is encouraging that learners are enthusiastic
about simulator training, and willing to incorporate its continued use for their own
skill maintenance.
Limitations for this study include small sample size, isolated participant demographics,
individual participant variability in surgical skillset, and narrow window of participant
follow-up. Sample size was relatively limited both in participants and participant
affiliation. This is a result of OTSL course limitations in number of possible participants,
and the restricted nature of a military training environment. Indeed, given the small
sample size of the staff physicians taking part in the training, there was not enough
statistical power to derive correlation of expert progression in measured metrics
as compared with resident progression. To achieve the power necessary to provide a
statistically significant result, sample size would need to approach approximately
400, not feasible for a study this size. Individual variability in surgical skills
was attempted to be controlled against in initial grouping of participants according
to their time to completion using a median split of the group's baseline “time required
to close wound.” While this should have cut down on baseline skill variability interference
with results, it is still possible it may have had an effect. Finally, longer term
follow-up of participants (months to years) following training could give insight
into maintenance of skills learned, as well as long-term perceptions of simulation
training.
Addressing these limitations would involve primarily a larger scope of study. Increased
institutional participation, both military and civilian, would increase participant
sample size and improve heterogeneity of sample demographics.
As the ultimate goal of any ophthalmic training program is to enhance patient safety
and improve outcomes, task performance should be standardized and ideal simulator
technology that provides realistic tissue characteristics, measurable and reproducible
trainee tasks, and formative feedback should become the standard in training.
