Keywords
resident - training - intravitreal - retina
Intravitreal injections have been commonplace in ophthalmology clinics for over a
decade and yet there remains a need for standardization of protocols surrounding the
administration of these injections. Having standardized protocols provides a consistent
framework for medical training and improves the quality of health care delivered.
It will also strengthen the medico-legal aspects of the health care provided. In 2001
there were 4,215 intravitreal injections performed across the entire United States.[1] Ten years later that number rose to 2.5 million and it has only continued to grow
since then.[2] Complications, though rare, associated with intravitreal injections include endophthalmitis,
retinal tear and detachment, cataracts, and ocular hypertension.[3]
[4]
[5]
[6]
[7] Traditionally, intravitreal injections are performed by retina specialists and retina
fellows; however, their prevalence in practice has led to ophthalmology residents
being trained on injection techniques.[8] Prior studies have already reported on the practice patterns among retina specialists
in several countries.[9]
[10]
[11]
[12] In 2014, Avery et al published updated recommendations for intravitreal injections
which were soon followed by the American Academy of Ophthalmology (AAO) consensus
statement in 2015.[13]
[14] The aim of this study is to determine the training patterns of ophthalmology residents
performing intravitreal injections and to compare these patterns with the practice
patterns of retina specialists, guidelines provided by Avery et al in 2014, and the
AAO consensus statement in 2015.
Methods
Ophthalmology residency program directors from each of the 119 ophthalmology residency
training programs accredited by the Accreditation Council for Graduate Medical Education
(ACGME) were contacted via email with a link to an anonymous, 29 question, internet-based
survey. The questions were created by a faculty member and medical student of Penn
State College of Medicine. Historical surveys in conjunction with clinical insight
were used to determine the specifics of each question. The survey was created by adopting
questions from prior surveys that were already validated and published.[9]
[10]
[11]
[12] The new collection of questions was then reviewed by a faculty/medical student team
to assess accuracy and relevance. A complete list of the survey questions can be found
in [Appendix A]. The residency directors were asked to disseminate the survey link to the residents
at their respective programs. The survey link was first sent out on April 17, 2018
with two reminder emails that followed in the next month.
Appendix A
Results
By June 18, 2018, 117 residents (7.89%) completed the survey. All respondents that
completed the survey stated that they are trained in how to preform intravitreal injections
during residency. [Fig. 1] shows the regional representation of residents. The majority of residents that responded
are being trained in the South and Midwest. Most residents performed injections at
a university or veterans medical center ([Fig. 2]). One-hundred and eight (92.3%) residents stated that they began performing injections
during their PGY 2 year. Only seven (6.0%) and two (1.7%) residents stated that they
did not begin performing injections until the PGY 3 and PGY 4 years, respectively.
The majority of residents (43.6%) performed greater than 50 injections per year. Thirty
five percent performed 25 to 50 injections per year, 19.7% performed 10 to 25 injections
per year, and 1.7% performed less than 10 injections per year ([Fig. 3]).
Fig. 1 Regional representation of residents.
Fig. 2 Breakdown of resident training environments.
Fig. 3 Average number of intravitreal injections performed per residency year.
Preinjection Considerations
[Table 1] shows the results of this study compared with prior studies, along with the recommendations
offered by the 2014 expert panel and the 2015 AAO consensus statement.[13]
[14] The majority of residents, 91 (77.8%), routinely dilate the eye that is being injected
prior to the injection. Ninety-eight (83.8%) residents wear gloves when performing
injections. Of the 98 residents that wear gloves, 57 (58.2%) use sterile gloves and
41(41.8%) use clean gloves. Ninety-nine (84.6%) residents stated that they are trained
to avoid talking in the procedure room during injections. The majority of residents,
110 (94.0%) and 111 (94.9%), use an eyelid speculum and povidone-iodine preinjection
on the conjunctiva, respectively. Six (5.1%) residents use sodium hypochlorite as
conjunctival antiseptic instead of betadine. A small number of residents, 6.0%, use
postinjection prophylactic antibiotics.
Table 1
Comparison of IVI techniques between this study and prior studies
IVI technique
|
U.S. residents 2018 (N = 117)
|
U.S. physicians 2010[9]
|
Canadian physicians 2012[10]
|
Recommended by 2014 guidelines[13]
|
Recommended by 2015 consensus statement[14]
|
Pupil dilation
|
78% (91)
|
−
|
83%
|
Optional
|
Yes
|
Gloves
|
84% (98)
|
58%
|
39%
|
Yes
|
−
|
Sterile drape
|
10% (12)
|
12%
|
9%
|
No
|
−
|
Mask
|
34% (40)
|
29%
|
−
|
Yes
|
−
|
Lid speculum
|
94% (110)
|
92%
|
91%
|
Optional
|
−
|
Conjunctival povidone-iodine
|
95% (111)
|
100%
|
100%
|
Yes
|
Yes
|
Conjunctival sodium hypochlorite
|
5% (6)
|
0%
|
0%
|
−
|
−
|
Periocular povidone-iodine
|
45% (63)
|
−
|
48%
|
Optional
|
−
|
Preinjection antibiotics
|
10% (12/)
|
|
|
No
|
No
|
Measured injection site
|
91% (106)
|
56%
|
55%
|
−
|
−
|
Optic nerve perfusion
|
77% (89)
|
72%
|
48%
|
Yes
|
Yes
|
Postinjection antibiotics
|
6% (7)
|
81%
|
20%
|
No
|
No
|
Bilateral IVI
|
52% (61)
|
46%
|
57%
|
Optional
|
−
|
Abbreviation: IVI, intravitreal injection.
Injection Technique
[Table 2] shows ophthalmology resident injection technique preferences. Sixty-nine (59.0%)
of residents are trained to prefer a superotemporal injection site while forty-eight
(41.0%) are trained to inject inferotemporally. The majority of residents, 106 (90.6%),
use measurements to determine the location of the injection site. Of these 106 residents,
77 (72.6%) use calipers and 26 (24.5%) use a tuberculin syringe. One-hundred and six
(90.6%) residents use a 30- and 27-gauge needle when performing injections with antivascular
endothelial growth factor drugs and triamcinolone acetonide, respectively. Twenty-two
(19.0%) residents stated that they are trained to consider the speed of the jet of
fluid during the injection. Of these residents, 11 (50.0%) stated that they are trained
to inject slowly and 11 (50.0%) stated that they are trained to inject quickly.
Table 2
Other injection techniques taught to U.S. ophthalmology residents
IVI technique
|
U.S. residents 2018 (N = 117)
|
Clean gloves
|
35% (41)
|
Sterile gloves
|
49% (57)
|
Avoid talking
|
85% (99)
|
Superotemporal injection site
|
59% (69)
|
Inferotemporal injection site
|
41% (48)
|
Use calipers to measure injection site
|
66% (77)
|
Use TB syringe to measure injection site
|
22% (26)
|
Use other device to measure injection site
|
2.6% (3)
|
30-gauge needle for anti-VEGF injections
|
91% (106)
|
27-gauge needle for triamcinolone injections
|
71% (77)
|
Postinjection gross visual acuity
|
59% (69)
|
Postinjection optic nerve visualization
|
14% (16)
|
Postinjection IOP measurement
|
3.4% (4)
|
Abbreviations: IOP, intraocular pressure; IVI, intravitreal injection; TB, tuberculin;
VEGF, vascular endothelial growth factor.
Postinjection Considerations
Most residents, 89 (76.7%), assess postinjection optic nerve perfusion. Of these residents,
69 (77.5%) use visual acuity and a few, 16 (18.0%) and 4 (4.5%), are trained to use
visualization of the optic disc and measurement of intraocular pressure (IOP), respectively.
These results are summarized in [Table 2].
Discussion
Our study is the first to report the practice patterns of intravitreal injections
among residents trained in the United States. The rate of resident-performed injections
has continued to rise over the past 10 years. In 2006, a survey of program directors
found that 75% of the program directors that responded were having their residents
perform fewer than 10 injections prior to graduating.[8] Resident responders to this survey show that 78.6% are performing, on average, at
least 25 of these injections each year of their training. This survey also demonstrates
that there are discrepancies among the intravitreal injection training techniques
of U.S.-based ophthalmology residents and retina specialists. Prior studies looked
at practice patterns among retina providers but, until now, had never addressed the
potential cause of these discrepancies which may lie in the training received during
residency. The largest of these studies was conducted in 2010, prior to new guideline
recommendations that were published in 2014 and the 2015 AAO consensus statement.[9]
[13]
[14] It is important to determine whether or not these new recommendations have begun
to affect the way in which ophthalmology residents are currently being trained.
Preinjection Considerations
Ninety-one (77.8%) residents stated that they are trained to routinely dilate the
eye that is to be injected. This number is consistent with reports from Canadian physicians
in 2012.[11] The 2014 guidelines suggested this may be optional and the 2015 AAO consensus statement
suggested using mydriatics before injections.[13]
[14] There is no clear consensus on whether or not patients need to be dilated. The survey
did not distinguish between visits to make the diagnosis and injection visits; therefore,
we are unable to determine from this study if patients were dilated for a planned
injection visit. Dilation may be done as a part of the routine clinic visit to make
the diagnosis, find other retina problems, as well as to examine postinjection optic
nerve perfusion and complications. Our survey found that 57 (48.7%) residents are
trained to use sterile gloves, which reflects a 15.7% increase when compared with
providers surveyed in 2010. The use of gloves (sterile or unsterile) does not impact
the overall rate of postinjection endophthalmitis.[15]
[16] Glove use may be varied based on the culture of the practice. Ninety-four percent
of residents are trained to use a lid speculum which is consistent with the 2014 guidelines.[13] The guidelines recommend the use of some form of lid retraction if a lid speculum
is not utilized.[13] Our study specifically asked about lid speculum use so we do not know if any other
method of lid retraction was used by the remaining 6% of residents. One of the new
recommendations that came with the 2014 guidelines was to either utilize a surgical
face mask or minimize talking during the in-office procedures.[13] This recommendation is based off of the findings that endophthalmitis secondary
to intravitreal injection has a disproportionately higher rate of streptococcus isolates
when compared to endophthalmitis secondary to intraocular surgery.[17]
[18]
[19] This survey found that this new recommendation has been heeded as 84.6% of residents
stated that they are trained to avoid talking while performing injections. The 2014
panel did not recommend the use of a sterile drape as evidence showed that its use
did not affect the overall rate of postinjection endophthalmitis.[13]
[16]
[20] Residency training is consistent with this recommendation as only 10.3% of residents
stated that they are trained to use a sterile drape. One of the most notable differences
between the U.S.-based providers that were surveyed in 2010 and current U.S. residents
is that 34% of providers claimed to use prophylactic, preinjection, topical antibiotics,
while only 10.3% of current residents are trained to maintain this practice. This
is consistent with numerous studies which found that not only did preinjection antibiotics
provide no real benefit but may actually pose some harm in the form of promoting antibiotic-resistant
bacteria.[16]
[21]
[22]
[23]
[24]
[25] This discrepancy between our survey and the 2010 U.S. provider survey is most likely
because the U.S. provider survey was done before the antibiotic recommendations were
changed.
Injection Technique
This survey found that 90.6% of residents are trained to measure the distance from
the limbus to the injection site with the majority of these residents (72.6%) utilizing
calipers to make the measurement. This is in stark contrast to what was found when
active providers were surveyed in 2010. At that time only 56% of retina specialists
claimed to measure the distance between the limbus and their injection site.[9] It is possible that calipers are used preferentially in training because it provides
a more exact measurement. Retina physicians with years of experience may no longer
feel the need to measure prior to preforming their injections. Another difference
between providers in 2010 and current residents is the consideration of the speed
of the actual injection. In 2010, 59% of surveyed physicians claimed to consider the
speed at which they did their injections.[9] This survey found that only 19% of current residents are trained to make this same
consideration. In 2010, among physicians that considered injection speed, 76% chose
to inject quickly. This survey found that the speed of injection was equivocal. Guidelines
in 2004 recommended injecting slowly to avoid excessive drug dispersion.[26] New evidence, however, indicates that injection speed makes no difference in drug
dispersion.[27] The discrepancy in the literature may be the reason for our equivocal survey results.
Postinjection Technique
This survey found that 76.7% of residents are trained to assess postinjection optic
nerve perfusion. This is similar to the 72% of practicing retina specialists that
assess postinjection optic nerve perfusion in 2010.[9] It is difficult to directly compare the methods for assessing optic nerve perfusion
between 2010 and now because in 2010 the respondents were offered a choice that combined
more than one method whereas this survey only allowed respondents to select one choice.
In spite of that, it is fairly certain that measuring IOP to assess optic nerve perfusion
is less common now than it was then. Only 4.5% of residents stated that they are trained
to utilize this method of assessment, while 15% of retinal specialists selected this
as their sole method of assessment in 2010.[9] The new 2014 guidelines suggest that IOP should be checked prior to administration
of an injection but give no recommendation on IOP versus gross visual acuity for postinjection
assessment.[13] The use of postinjection prophylactic antibiotics is another area where new guidelines
have made a clear impact. The 2010 survey of U.S.-based retinal specialists found
that 81% used postinjection antibiotics. This survey found that only 6.0% of residents
are being trained to continue this practice. The 2014 guidelines reflect this shift
in practice and no longer recommend postinjection antibiotics, citing strong evidence
from the DRCR Network LRT trials.[15] One area that still remains divided is the practice of performing same-day, bilateral
injections. This survey found that 52.1% of residents are being trained to perform
simultaneous, bilateral injections. In 2010, the rate among practicing physicians
was reported as 46%.[9] The new guidelines do not take a strong stance against this practice but do make
sure to offer certain precautions. They recommend that each eye be treated as its
own separate procedure and that, when possible, different lots of medications should
be used for each eye.[13] It has been shown that bilateral injections can be done in a safe manner that provides
convenience for both the patient and the physician.[28]
Limitations of this study include difficulty in assessing how many residents received
the survey because our method of delivery depended on the residency Program Director
or Coordinator sending the survey to the ophthalmology residents. If we assume that
all residents received the survey then our response rate is low. While our survey
utilized previously validated questions, it was not newly validated for the resident
population. A future study would benefit from a validation process that takes into
consideration potential differences between the interpretations of questions by residents
and attending physicians. Current practice patterns of U.S.-trained ophthalmology
residents are reported in this study and it is clear that practice patterns have shifted
between physicians in 2010 and residents in 2018. It is still unclear if these changes
only exist in the vacuum of academic centers or if these changes have made their way
to the retina community at large. Future improvements would seek to expand the response
rate as well as to survey retina fellows at the beginning of their fellowship training
and then again after 5 years of being in practice on their own. Ultimately, the ideal
study would include residents, fellows, and physicians who have been in practice for
several years. This would allow for the capture of practice patterns at the same time,
and under the same set of guidelines, but across a myriad of different career points.
Conclusion
Unfortunately, there are no prior studies that assess residency training patterns
that our data can be directly compared with.
The only reference points currently available are the past trends of practicing retina
specialists. While this is not a direct comparison, it is the best comparison that
can be drawn. It is our belief that large differences between how residents are currently
trained and how physicians previously practiced can be attributed to changing guidelines.
Finally, this study provides a much needed baseline to assess trends in residency
training for intravitreal injections.