Characterizing Current Attitudes and Practices for Human Subject Safety in Studies Involving Pupil Dilation

Introduction

For research taking place in the United States, part of the criteria for Institutional Review Board (IRB) approval and oversight of human subjects research is to ensure that the risks to subjects are minimized in accordance with the Common Rule (45 CFR 46). As part of their review of a new research protocol, it would be typical for an IRB to request that the study team describe the features of the project intended to minimize risks to subjects. These features could include detailing recruitment practices, features of the project design, or specific information about the conduct of procedures. Even in the review of NIH grants, the adequacy of protection against risks is a factor in assessing the appropriateness of the overall plan for the involvement of human subjects in the research. As such, the principal investigator is asked to “…describe planned strategies for protecting against or minimizing all potential risks identified…” (Department of Health & Human Services, 2018). In addition, IRBs need to ensure that potential subjects are informed of the possible risks from the research procedures. Explicitly, 45 CFR 46.116(b)(2) requires that the elements of informed consent include “a description of any reasonably foreseeable risks or discomforts to the subject’.

Importantly, the above requirements apply even to studies that only involve minimal risk procedures – as minimal risk does not equate to an absence of risk. According to federal regulations (45 CFR 46.102(j)), minimal risk studies are those for which the “…probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests” (Office for Human Research Protections, 2018). Often, minimal risk procedures are familiar to individuals responsible for overseeing the protection of human subjects at an institution and the protocol features intended to minimize risk may be thought of as standard or “common sense”. For example, in a study that requires a small blood sample, the investigator might explain that a trained phlebotomist will be used to collect the sample. Not only might this seem like “common sense”, but it is specified as a best practice for phlebotomy by the World Health Organization (World Health Organization, 2010). Researchers may elect to add additional “common sense” features that may not be standard clinical practice, but nevertheless are intended to further minimize risk – such as collecting as small a volume as possible and giving the subject water prior to the blood draw (to reduce the chances of becoming lightheaded or fainting). Whether an IRB requires that these “common sense” features be detailed in a research protocol is up to the individual IRB. However, the availability of standards and familiarity with the procedure aides the IRB in their oversight role.

However, there are a number of other procedures that meet the definition of minimal risk, but which are not as common and thus may not be broadly familiar to IRB reviewers and staff members. As such, the expected or “common sense” steps to be taken to minimize risk may not be intuitive, which introduces the potential for inconsistent oversight across sites and as a result possibly increases risk to participants. One such procedure is pupil dilation, which is used during routine eye examination in ophthalmic and optometric settings. Thus, when pupil dilation is used in the research setting, it is often considered to meet the definition of a minimal risk procedure. While the likelihood of a serious complication is low, there is not an absence of risk (Applebaum & Jaanus, 1983; Pandit & Taylor, 2000; Portello & Krumholz, 2008; Stavert et al. 2015; Vuori et al. 1994; Yolton et al., 1980). Potential serious complications (either directly or through interaction with other drugs) include acute angle closure glaucoma, allergic reactions, increased blood pressure, arrhythmias, tachycardia, or neurological effects (Ah-Kee et al., 2015; Brooks et al., 1986; Farkouh et al., 2016; Fraunfelder & Meyer, 1984; Labetoulle et al., 2005; Lachkar & Bouassida, 2007; van Minderhout et al., 2015). While there are some guidelines available to healthcare professionals regarding dilation (Lee, 2007; Rushing, 2007), there is no such guidance as to which risks to disclose to the research subject. Even if there were clear clinical standards, these may not necessarily be sufficient in the research setting as the risk:benefit ratio is often vastly different in the eye care clinic compared to a vision science lab doing retinal imaging studies. In an effort to initiate discussion around recommendations for safer practices in the dilation of human subjects for research purposes, we set out to define the current state of dilation protocols and attitudes for the assessment of pupil dilation risk present among vision researchers and individuals involved in the institutional oversight of human subjects research. Future dialogue could use our data as a starting point to the development of guidelines or best practices – both for use by researchers in designing their protocols and by IRBs in evaluating the adequacy of protections in research studies.

Methods

Two surveys were developed to identify attitudes and practices associated with pupil dilation risk: one for vision researchers and the other for individuals associated with the oversight of human subjects research at institutions (see Supplemental Material). The survey for vision researchers focused on characterizing the extent and methods by which a human subject is assessed for risk prior to pupil dilation and attitudes toward dilation risk. For individuals associated with institutional research oversight, the survey assessed the extent of their IRB template language, policies, guidance and overall institutional perception of risks associated with pupil dilation. Question format consisted mainly of yes/no questions with the opportunity for open ended responses after select questions. For researchers, the questions were designed to assess their use of dilation drops and their current practices regarding screening subjects, informing subjects of risks, and administering eye drops. We used published information on best dilation practices as well as reports of adverse events to develop many of the questions related to screening and administration of the drops (Joint Commission on Allied Health Personnel in Ophthalmology, 2013; Lachkar & Bouassida, 2007; Lavanya et al., 2008; Mapstone, 1977; Rushing, 2007; Wolfs et. al., 1997; Yolton et al., 1980).

Forty-one questionnaires were distributed to vision researchers via email. Contact information was obtained through publicly available databases such as PubMed, NIH Reporter, institutional websites, etc. Individuals associated with human subjects research oversight were obtained through advertisement on a well-known IRB discussion forum. Confidentiality of information and anonymity of survey data to investigators was achieved using REDCap (Research Electronic Data Capture) servers located in a local data center at the Medical College of Wisconsin together with the Clinical & Translational Science Institute of Southeast Wisconsin Biomedical Informatics Core.

Inclusion criteria for this study consisted of vision researchers and individuals associated with research oversight over the age of 18 who worked at academic medical centers, colleges, or universities. Prior to gaining access to their respective surveys, respondents were presented with a short summary of the project. Subsequent participation was granted if respondents acknowledged they were over the age of 18, understood that participation was voluntary, and that responses were to be solely used for research purposes. Respondents were informed that they did not need to answer any questions in which they did not feel comfortable doing so. This study was approved by the IRB of the Medical College of Wisconsin as an exempt study under United States Federal Regulation 45 CFR 46.104(d)(2).

Results

Of the 24 vision researchers that opened their invitations to participate and 44 individuals associated with institutional oversight of human subjects research that viewed the invitation on an IRB discussion forum, 14 (58%) and 21 (48%) completed their respective surveys. Responses were collected from April 17 - May 20, 2020. The positions of respondents for the two surveys are provided in Table 1. Among vision researchers, a majority of respondents were associate professors (43%) or professors (36%). For those involved in institutional oversight of human subjects research, the majority of responses came from those involved with IRB and research administration (57%).

Table 1.

Survey Participant Positions.

Vision Researcher (n = 14)	n (%)	Research Regulator and IRB representative (n = 21)	n (%)
Associate Professor	6 (43)	Director - IRB	3 (14)
Professor	5 (36)	Research/Compliance/Regulatory Analyst	3 (14)
Assistant Professor	1 (7)	IRB Chair	2 (10)
Research Engineer	1 (7)	IRB Administrator	2 (10)
No Answer	1 (7)	IRB Manager	2 (10)
		Director – Office for Protection of Research Subjects	1 (5)
		Director – Research Regulatory Affairs	1 (5)
		IRB Reliance Coordinator	1 (5)
		IRB Coordinator	1 (5)
		IRB Member (MD)	1 (5)
		Associate Professor	1 (5)
		Research Education and Compliance Manager	1 (5)
		No Answer	2 (10)

Note. IRB = Institutional Review Board; MD = Doctor of Medicine

Researcher Practices

General Conceptions Regarding Pupil Dilation

Attitudes and beliefs towards dilation among vision researchers are described in Table 2. In our survey, researchers reported using phenylephrine (86%), cyclopentolate (21%) and tropicamide (93%) when dilating a subject. A limitation is that we did not specifically ask about drug concentration, which does impact the safety profile. While pupil dilation was used unanimously, researcher reporting of institutional attitudes towards dilation varied. Seventy-nine percent noted that pupil dilation constituted minimal risk at their institution. For those who provided explanation into their response, reasons included having subjects coming in for a study having been already dilated from a clinical care visit (n=3), research/ethics guidelines (n=2), pupil dilation being considered standard of care in ophthalmology (n=2), screening for narrow angles prior to dilation (n=2), having professional staff dilating subjects specifically for research purposes (n=1), dilating patients who have historical experience with dilation (n=1), due to the dilation of a single eye (n=1), and/or due to the researcher explaining risk in protocol and to the subject (n=1). Among the two respondents whose institutions considered dilation to be above minimal risk, reasons revolved around their need to submit a risk assessment for dilation drop use in a study (n=1) and a requirement of a slit lamp exam prior to dilation for non-MD personnel (n=1). Nevertheless, none of the researcher respondents reported adverse events associated with the use of dilation drops in their studies.

Table 2.

General Use and Attitudes Regarding Pupil Dilation Among Researchers.

Variable	Yes (%)	No (%)	Unsure (%)
Dilation Drop Medications Used in Human Subjects
Tropicamide	13/14 (93)	1/14 (7)
Phenylephrine	12/14 (86)	2/14 (14)
Cyclopentolate	3/14 (21)	11/14 (79)
Atropine	0/14 (0)	14/14 (100)
Other	0/14 (0)	14/14 (100)
Does your institution consider the use of dilation drops a minimal risk procedure?	11/14 (79)	2/14 (14)	1/14 (7)
Do you use dilation drops in the conduct of one or more of your research studies?	14/14 (100)	0/14 (0)
Are you aware of any adverse events associated with pupil dilation in your research?	0/14 (0)	14/14 (100)

Informing Human Subjects of Pupil Dilation Risk

Table 3 provides a summary of reported practices by researchers in informing human subjects on risks associated with pupil dilation. Ninety-three percent of researchers included some mention of risks associated with dilation drops in their informed consent form, with the other respondent mentioning that they verbally explained risk in their inclusion meeting and included it in their protocol. However, there was some variability as to which specific risks were included in the informed consent documents. Ninety-two percent included discomfort (stinging or burning) associated with the drops and 77% included both increased light sensitivity and temporary blurred vision as a risk. Eighty-five percent include acute angle closure as a risk. Sixty-nine percent mentioned redness or irritation as a risk, with two respondents adding that they specifically include allergic reaction. Furthermore, researcher protocols allowing the dilation of women who are pregnant or nursing, and children was split at 50% for both populations. Of the seven researchers whose protocols permitted pupil dilation in women who were pregnant or nursing, none reported having extra language in their informed consent addressing added risk to the subject or infant/fetus. For those who provided explanations, reasons included not having this population in their protocols but basing their response off of permissive clinical dilation practices for vision threatening conditions (n=1) and being unaware of any risks to the mother or infant (n=1).

Table 3.

Researcher Use of Informed Consent and Education of Human Subject on Pupil Dilation Risk.

Variable	Yes (%)	No (%)
Are possible risks associated with use of dilation drops provided in the informed consent document to the subject prior to enrollments?	13/14 (93)	1/14 (7)
Specific risks associated with dilation drops listed in informed consent documents of investigators who used them.
Discomfort with drop administration (itching/burning)	12/13 (92)	1/13 (8)
Redness or Irritation	9/13 (69)	4/13 (31)
Acute Angle Closure Glaucoma	11/13 (85)	2/13 (15)
Increased Light Sensitivity	10/13 (77)	3/13 (23)
Temporary Blurred Vision	10/13 (77)	3/13 (23)
Other: (allergic reaction) a	2/13 (15)	11/13 (85)
Do you educate the subject about signs and symptoms associated with angle closure complications from dilation?	9/14 (64)	5/14 (36)
For those that educate on angle closure complications, what medium of education is used?
Verbal Alone	4/9 (44)	5/9 (56)
Verbally and in Writing	5/9 (56)	4/9 (44)
Writing Alone	0/9 (0)	9/9 (100)
Does your protocol permit dilating women who are pregnant or nursing?	7/14 (50)	7/14 (50)
For protocols that permit dilating women who are pregnant or nursing, does your informed consent address extra risks to the pregnant subject or fetus/baby?	0/7 (0)	7/7 (100)
Does your protocol permit dilating children under the age of 18?	7/14 (50)	7/14 (50)

^aOne respondent included allergic reaction as an “other” risk in this question and a second respondent included this same risk in the text they provided to the research participants.

Variation in researcher practices was also seen regarding the education of human subjects on signs and symptoms of angle closure complication. Of the 64% who noted educating subjects on this matter, the medium of education was split among those who educated just verbally (44%) and those who provided written information alongside discussion (56%). Three respondents provided the language/text they shared with the subject. These varied in the level of detail but all three included a discussion of acute angle closure and two mentioned the precautions/exclusions that would be taken for acute angle closure suspects and that care would be needed/available should acute angle closure arise. The specific responses (with identifying information redacted) for these three respondents are provided in the Supplemental Material. Also included is a fourth respondent who provided informed consent language as part of the “other comments” open text field at the end of the survey.

Researcher Pupil Dilation Practices and Risk Assessment

Table 4 summarizes the research practices regarding minimizing risks associated with pupil dilation in human subjects before, during and after dilation. Prior to dilation, researcher methods for assessing dilation risk were variable. Seventy-one percent conducted a systemic review of a subject including their current medications and 57% of researchers asked about a family history of glaucoma. Ninety-three percent asked whether or not a subject had been previously dilated. In terms of examining the patient’s eye prior to dilation, 79% measured refractive error, and 64% examined the iridocorneal angle (an anatomical risk factor for acute angle closure). Methods used to examine the angle also varied with 78% using slit lamp (Van Herick Technique), 22% using gonioscopy, 0% using anterior segment optical coherence tomography (AS-OCT) and 67% using the penlight method (oblique temporal illumination of iris). Personnel tasked with examining the angle also varied. This was reported as being carried out by ophthalmologists (67%), optometrists (56%), ophthalmology residents/fellows (22%), graduate students/postdoctoral fellows (22%), and research scientists (PhD) (11%).

Table 4.

Researcher Practices Through Process of Dilating Human Subjects.

Variable	Yes (%)	No (%)	Variable	Yes (%)	No (%)
Before dilation do investigators…			For investigators who examine the iridocorneal angle before dilation, do you use…
conduct a systemic review including current medications?	10/14 (71)	4/14 (29)	slit lamp (Van Herick Technique)?	7/9 (78)	2/9 (22)
ask about family history of glaucoma?	8/14 (57)	6/14 (43)	gonioscopy?	2/9 (22)	7/9 (78)
measure refractive error?	11/14 (79)	3/14 (21)	anterior segment OCT?	0/9 (0)	9/9 (100)
ask whether the subject has been previously dilated?	13/14 (93)	1/14 (7)	penlight method (oblique temporal illumination of iris)?	6/9 (67)	3/9 (33)
examine the iridocorneal angle?	9/14 (64)	5/14 (36)	other?	0/9 (0)	9/9 (100)
Regarding protocol, do you do anything different based on the subjects…			For investigators who examine the iridocorneal angle, who measures the angle?
race/ethnicity?	0/14 (0)	14/14 (100)	ophthalmologist	6/9 (67)	3/9 (33)
age?	2/14 (14)	12/14 (86)	ophthalmology resident/fellow	2/9 (22)	7/9 (78)
sex?	0/14 (0)	14/14 (100)	optometrist	5/9 (56)	4/9 (44)
refractive error?	0/14 (0)	14/14 (100)	graduate student/postdoctoral fellow	2/9 (22)	7/9 (78)
current medications?	1/14 (7)	13/14 (93)	research scientist (PhD)	1/9 (11)	8/9 (89)
other?	0/14 (0)	14/14 (100)	other	0/9 (0)	9/9 (100)
When dilating a subject…			After dilation…
are gloves used when administering dilation drops? a	4/13 (31)	9/13 (69)	do you provide sunglasses to subjects before they leave?	10/14 (71)	4/14 (29)
is punctal occlusion used to reduce systemic absorption of dilation drops? a	1/13 (8)	12/13 (92)	do you examine the iridocorneal angle?	0/14 (0)	14/14 (100)
			do you check intraocular pressure?	0/14 (0)	14/14 (100)

Note. OCT = Optical Coherence Tomography

^aOne investigator noted exclusive use of subjects dilated from clinical setting; therefore, their responses were not Included for these questions leading to n=13.

While 71% of researchers reported conducting a systemic review of a subject that included addressing current medications, only one respondent noted that their protocol would change based on a subject’s medications. Two respondents indicated that they modified their protocol based on the age of the subject. None of the researchers noted changing their dilation protocol based on the subject’s refractive error, race/ethnicity, or sex. During the dilation of subjects, only 8% of researchers using punctal occlusion to reduce systemic absorption and 31% reporting they used gloves when administering drops for their studies. After dilation, 71% of researchers reported that they provided sunglasses to human subjects. In contrast, none of the researcher respondents reported assessing the patients angle or intraocular pressure post dilation. It is worth noting that despite all respondents answering no to this question, one respondent did share the information provided to the subject which in fact contained language stating that post-dilation examination of the eye would take place to ensure that effects were within acceptable limits.

Institutional Practices

General Conceptions Regarding Pupil Dilation Risk

Table 5 summarizes the responses from individuals associated with the oversight of human subjects research at their respective institutions. Forty-three percent of respondents indicated that dilation was considered to be a minimal risk procedure by their institution, with 14% reporting that the associated risk was considered greater than minimal, and the other 43% of respondents being unsure of the risk classification. Among the nine institutions that regarded pupil dilation as a minimal risk, a major theme among provided explanations based the response on the use of dilation drops in a standard ophthalmology visit (n=5). One respondent noted that minimal risk was given only for studies not involving children. One institution mentioned that pupil dilation qualified for expedited IRB review, while another required a convened meeting to decide. Conversely, institutions that felt dilation did not constitute minimal risk provided explanations reasoning that the use of dilation drops was not an activity of everyday life (n=1), was not a part of a routine physical or psychological exam (n=1), had inherent risks of adverse events due to medication that was used (n=2), that FDA regulations associated with drop use constituted elevated risk (n=1), and/or as a result of ongoing trials regarding dilation drop use in special populations (n=1). For respondents that were unsure on whether dilation constituted minimal risk, provided reasons included being unfamiliar with policies regarding dilation risk as a result of not having reviewed studies that used them (n=7), viewing specific variables of a study involving dilation to make a holistic determination of risk (n=5), and/or as a result of being in a position that did not make those considerations (n=1).

Table 5.

Current Attitudes Regarding Dilation Risk Among IRB Members and Research Regulators.

Question	Yes (%)	No (%)	Unsure (%)
Does your institution consider use of dilation a minimal risk procedure?	9/21 (43)	3/21 (14)	9/21 (43)
Does the IRB have any standard/template language for use in informed consent documents discussing the risks relating to the use of dilation drops in research studies?	2/21 (10)	19/21 (90)
Does the IRB offer other guidance to researchers as to how to minimize risks associated with the use of dilation drops in research studies?	3/21 (14)	18/21 (86)
Does the IRB have any policies/guidance to the use of dilation drops in research studies involving women who are pregnant or nursing?	1/21 (5)	20/21 (95)
Does the IRB have any policies/guidance related to the use of dilation drops in research studies involving children/minors?	1/21 (5)	20/21 (95)
Are there any other groups that the IRB considers to be at increased risk of complications from dilation drops? a	4/21 (19)	17/21 (81)
Has a researcher at your institution reported any adverse events associated with the use of dilation drops in a research setting?	0/21 (0)	21/21 (100)

Note. IRB = Institutional Review Board

^aOne individual marked “Yes” for this question, but when explaining answer notes: “There aren’t any specific groups, it would be based on population being included.” Their response in this table is marked under “No”.

IRB Assistance for Researchers Using Pupil Dilation

When asked about IRB assistance for researchers using pupil dilation in studies, 90% of respondents reported that they did not provide researchers with standard template language for informed consent forms. Additionally, 86%, 95%, and 95% of respondents did not offer guidance/policies for minimizing general associated risks of pupil dilation, risks associated with women who are pregnant or nursing, and risks involving children/minors respectively (Table 5).

For the respondent that provided language for discussing pupil dilation risks in the informed consent, it was noted that they worked to ensure risk language was present that addresses risks associated with photosensitivity, blurred vision and transportation concerns post-dilation. Among institutions that offered guidance for minimizing general risk, provided language reminded researchers to be thoughtful with inclusion and exclusion criteria (n=1) and/or to include all risks associated with mydriatic eye drops (n=2).

With policies and guidance addressing pupil dilation in pregnant or nursing women, the institution that addressed this noted that it was general policy and made sure associated risks are considered when appropriate. For policy and guidance that was offered for the dilation of children/minors, the one institution that addressed this noted that studies involving minors are automatically reviewed by the full IRB board. Other provided groups that institutions considered to be at elevated risk included individuals with cardiac issues or those with a history of psychosis (n=1), those that had undergone recent eye surgery (n=1), subjects that are generally in poor health (n=1), and/or subjects who drive (n=1). Nevertheless, research regulators and IRB representatives unanimously did not report any reported adverse events associated with the use of pupil dilation at their institutions.

Discussion

All topical mydriatics, while generally considered safe and used extensively for clinical and research purposes, have the potential for significant adverse effects. Probably the most well-recognized risk is the chance of an acute iridocorneal angle closure event. While rare, acute angle closure can lead to vision loss if left untreated (Pandit & Taylor, 2000; Prum et al., 2016; Lai & Gangwani, 2012). There are anatomical factors that have been associated with increased incidence of an acute angle closure event, including narrow iridotrabecular angles, a shallow anterior chamber, thicker crystalline lens, cataracts, and shorter axial length (Lai & Gangwani, 2012; Prum et al., 2016; Yoon et al., 2003). While our survey noted that 64% of researchers examined the iridocorneal angle prior to dilation, none said that they examined the angle after dilation. Other demographic variables such as female sex, Asian and Inuit ethnicity, increased age, hypermetropia, and those with a family history of glaucoma are reported to be at an elevated risk of developing acute angle closure glaucoma (Ah-Kee et al., 2015, Prum et al., 2016). Despite this, our survey found that a vast majority of researchers do not change their protocols around variables such as race/ethnicity (0%), age (14%), refractive error (0%) or sex (0%).

With respect to systemic absorption, phenylephrine drug labelling notes significant cardiovascular effects such as, “…hypertension, syncope, myocardial infarction, tachycardia and subarachnoid hemorrhage” (Paragon BioTeck, 2014). However, phenylephrine’s effect on the blood pressure and heart rate may not be as significant as previously thought, with 2.5% preparations being recommended for safe clinical use (Stavert et al., 2015). Nevertheless, there is a presence of significant adverse events associated with phenylephrine at higher concentrations, such as with the 10% preparation (Lai, 1989). This results in caution or avoidance for clinical use in individuals predisposed to systemic cardiovascular effects, such as those with cardiovascular disease and thyrotoxicosis (Coulter et al., 2008; Nevitt, 2013; Paragon BioTeck, 2014; Lai, 1989). For the antimuscarinic mydriatic cyclopentolate, significant adverse effects including, but not limited to ataxia, drowsiness, incoherent speech, seizures, and hallucinations have been reported with topical use, especially with children (Bausch & Lomb, 2016; Mirshahi & Kohnen, 2003; Rajeev et al., 2010; Wakayama et al., 2018). Tropicamide has also been reported to cause effects including but not limited to central nervous system disturbances, psychotic reactions and vasomotor or cardiorespiratory collapse, but with less propensity than cyclopentolate (Bausch & Lomb, 2019; van Minderhout et al., 2015; Vuori et al., 1994). Furthermore, all three of the aforementioned medications have been reported to precipitate allergic reactions upon administration in some patients (Camarasa & Pla, 1996; Haroun-Díaz et al., 2014; Madsen & Andersen, 2015). The pharmaceutical label for phenylephrine mentions that topical phenylephrine should be avoided in patients taking “atropine-like” drugs (Paragon BioTeck, 2014). Phenylephrine may potentially precipitate interactions when combined with drugs including, but not limited to, monoamine oxidase inhibitors and tricyclic antidepressants (Nevitt, 2013; Coulter et al., 2008; Meyer & Fraunfelder, 1980; Shur & Checkley, 1982). Shur and Checkley specifically mention that the use of desipramine, a tricyclic antidepressant, can limit the mydriatic effect of phenylephrine when used alone; therefore, leading to the potential temptation of repeated doses and subsequent systemic absorption (Shur & Checkley, 1982). In terms of the antimuscarinic mydriatics discussed, both tropicamide and cyclopentolate labelling notes that they can interfere with, “…anti-hypertensive action of carbachol, pilocarpine, or ophthalmic cholinesterase inhibitors” (Bausch & Lomb, 2019; Bausch & Lomb 2016). Elderly populations may be more likely to experience adverse events as a result of their increased propensity for chronic conditions and associated medications (Coulter et al., 2008). While our survey notes that 71% of researchers performed a systemic review and screened a subject’s medications prior to dilation, only 7% noted that their protocols changed as a result.

Given an understanding of the associated risks, the widespread use of dilation drops in vision research, and the inconsistencies identified in our survey with informing and managing these risks, there is a need to develop guidelines that IRBs and other individuals involved in overseeing research can reference when trying to determine if appropriate steps have been taken to minimize risks for the specific population being studied. Fortunately, it is possible to identify individuals at increased risk as well as methods to minimize general risks associated with administering dilation drops in general. For example, with respect to limiting adverse events associated with the physical administration of eye drops, infection risks can be reduced through proper hand hygiene (Lee, 2007). In our survey, only 31% of researchers reported using gloves with the administration of eyedrops, though we recognize that our question was flawed in that we only asked about gloves and did not ask about hand washing. In addition, the angle of instillation and geometry of the bottle are factors that contribute to the spread of microbial disease (Costa et al., 2020). Thus, proper training of the individuals administering the dilation drops could be a fairly simple and low-cost method to reduce the likelihood of adverse events. It is important to keep in mind that the guidelines may need to vary according to the specific population being studied, as some demographic groups are at elevated risk than others.

When minimizing risks associated with dilation in a research study, perhaps the most effective means would involve using subjects that are already dilated (research visit following clinical visit) as it essentially eliminates risks associated with dilation in a research study. In fact, 3/14 (21%) of our researcher respondents indicated they usually or always use dilation in patients who were already getting dilated for clinical care purposes. While it would not be practical to only recruit such subjects, this is specifically included in 45 CFR 46.111(a)(1) as a means to minimize risks to subjects: “whenever appropriate, using procedures already being performed on the subjects for diagnostic or treatment purposes.” However, the informed consent process should ideally take place prior to dilation for clinical purposes as dilation may interfere with the ability to read and not all researchers may have this option available to them. Other means of addressing risk can involve questioning to identify individuals at risk for complications from systemic absorption. As drug interactions with dilation drops are known, a review of current medications can identify individuals who should not be dilated unless medically necessary. A systems review can identify individuals with medical conditions associated with an increased risk of adverse effects from dilation, such as those with Down syndrome and those with an increased propensity for ectopia lentis, such as individuals with Marfan’s syndrome or homocystinuria. Specifically, avoidance of cyclopentolate in individuals with Down syndrome is clinically advised due to the propensity for hypersensitivity and for those with ectopia lentis, general caution with dilation as a result of an increased propensity for angle closure complications (Portello & Krumholz, 2008). Similarly, as the risks in women who are pregnant or nursing are unknown (Bausch & Lomb, 2019; Bausch & Lomb 2016; Paragon BioTeck, 2014), women of child-bearing age should be asked if they are pregnant or nursing and the potential for specific risks disclosed. In our study, 50% of researcher protocols allowed for the dilation of women who were pregnant or nursing. Of those that did, none offered additional information in their informed consent discussing potential risks to the mother and fetus/baby. While it is up to individual IRBs to determine the specific screening that is appropriate (i.e., self-reporting or pregnancy testing), self-reporting is likely sufficient. Measures can also be taken with the administration of dilation drops to limit systemic absorption. Punctal occlusion (accomplished by applying moderate pressure to the lacrimal puncta) is noted to effectively reduce the systemic absorption of topically administered medications, leading to decreased potential for adverse effects (Coulter et al., 2008; Müller et al., 2020). Punctal occlusion is included as a best practice in some dilation literature (Rushing, 2007). However, in our study only 8% of researchers used punctal occlusion with the administration of dilation drops.

The greatest risk factor for an acute angle closure event is the anatomy of the anterior segment. Individuals with a narrow iridocorneal angle are at increased risk of an acute angle closure event (Lai & Gangwani, 2012). As such, an effective approach to minimize the likelihood of an adverse event associated with this risk is to examine the anterior segment prior to dilation. For methods of viewing the angle, gonioscopy is considered to be the gold standard (Smith et al., 2013). However, Van Herick slit lamp and AS-OCT imaging methods have both provided reliable detection capability of narrow angles, which could be beneficial to researchers without access to professionally trained personnel (Dabasia et al, 2015; Park et al., 2011; Smith et al., 2013). In contrast, the oblique temporal illumination of the iris (penlight method), while better than no method, has yielded specificity concerns that have prevented its recommendation for use in community-based screening (He et al., 2007). Among researchers that examined the angle, the Van Herick method via slit lamp (78%) and the penlight method (67%) were the most utilized, with gonioscopy (22%) and AS-OCT (0%) being the least. An effective screening technique can be to ask if the subject has been dilated previously without an adverse event, as individuals who have been dilated safely before would be less likely to have an ocular anatomy that predisposes them to acute angle closure. Beyond screening, monitoring the eye after dilation can play an important role in minimizing risk in research studies (especially those that require dilation of higher risk subjects). In the event of an acute angle closure event, examination of IOP after dilation would in theory reduce the chance of vision loss as this would provide earlier detection and allow more rapid treatment (Mapstone, 1977). However, none of the researchers who responded indicated that they measured IOP or examined the anterior segment following dilation.

Regardless of the specific steps in a protocol intended to minimize adverse events, it is critical that research subjects are informed of potential risks of dilation as part of the informed consent process. Our study found that 64% of researchers reported educating their subjects on angle closure complications. Among them, 44% explicitly used a verbal form of communication. In the healthcare setting, 15-80% patients have issue with recalling information explained to them at a visit (Kessels, 2003; Laws et al., 2018). The percentages were influenced by variables including the method of delivery, age and importance of information (Kessels, 2003; Laws et al., 2018). Importantly, other than acute angle closure (85%) and allergic reaction (8%), the rare but serious side effects of mydriatic medications reviewed above were not reported to be covered in their informed consent form, though two respondents mentioned providing subjects with educational material that did mention the rarer complications. Additionally, our survey reported that only 10% of IRBs offered template language addressing risk for use in the informed consent process for studies in which dilation drops are used. Equally important to describing the potential risks is taking the time to ensure patients are educated on signs and symptoms of potential adverse outcomes as well as actions to be taken if they do occur. This is another important aspect of study design that can serve to minimize the potential for serious adverse events.

Best Practices

Our study highlights a variance in safety measures and guidance that are utilized in the dilation of human subjects. Without a standard in pupil dilation safety, variation in practice and the subsequent potential occurrence of adverse events will inevitably remain in vision research. Although the creation of a standard will require IRB collaboration among institutions, we believe that certain factors should be included in the discussion based on our findings. Primarily, (a) a greater depth of uncommon effects associated with specific mydriatics and populations outlined in an informed consent, with subsequent subject education, (b) increased use of screening measures for a better assessment of human subject risk prior to dilation, (c) an emphasis on safety measures taken both during and after the administration of dilation drops, and (d) more prevalent IRB guidance/policy on minimizing risks associated with topical mydriatic use for vision researchers.

Research Agenda

As a result of this pilot study, further investigation into practices among researchers and research regulators is needed for further characterization and expansion of the current state of safety practices. Although our study highlighted areas that could be improved upon, we were limited by the number of responses received. However, future endeavors may encounter similar concerns with response yield due to the nature of questions asked. We believe that our results, in conjunction with the absence of a current standard, widespread dilation use, and the known possibility of adverse events allow for subsequent discussion and creation of a standard in dilation practices among IRBs and researchers. With the creation of this standard, significant risks associated with human subject dilation may be further minimized in populations predisposed to experience them at a higher frequency. It is important to note that our observations only refer to studies that are subject to US laws and regulations regarding human subjects research. Whether similar issues exist in other countries was not assessed, though we believe there may be universal value in developing clearer guidance on the use of dilation agents for research use.

Educational Implications

For a procedure that is commonly viewed as minimal risk, potential pitfalls associated with specific mydriatics and subject populations may be overlooked or unknown to researchers and IRBs. Education of these individuals is therefore necessary to prevent adverse outcomes in predisposed subjects. Education can take many forms including but not limited to the formation of guidelines, presentations, and departmental announcements.