Upper eyelid height is primarily determined by the actions of the levator palpebrae superioris and Muller’s muscle. Muller’s muscle, a smooth muscle with a predominance of α-1D adrenergic receptors, contracts in response to sympathetic stimulation or adrenergic agonists [1,2]. Compounds such as apraclonidine, a medication with mixed α and α-1 receptor agonist properties, have been shown to reverse ptosis caused by Horner’s syndrome and botulinum toxin-induced ptosis, and to help assess blepharoptosis preoperatively [3-7].

Brimonidine is a medication with a similar mechanism of action; it is a selective α-2 adrenergic agonist with an additional, weaker affinity for the α receptor [8]. Brimonidine 0.2% is used in the treatment of glaucoma and ocular hypertension, as it decreases intraocular pressure by increasing uveoscleral outflow and reducing aqueous humor inflow [9]. As an alpha-adrenergic agonist, brimonidine may cause contraction of Muller’s muscle and elevate patient’s eyelids [1]. It is also used to induce vasoconstriction of both skin and ocular tissue, and as a mydriatic agent, which can help camouflage anisocoria associated with Horner’s syndrome [8,10,11].

Recently, brimonidine tartrate 0.025% (Lumify^TM, Bausch&Lomb Inc., Rochester, NY, U.S.A) has been introduced as an over-the-counter ocular redness relief drop. No serious adverse effects or tachyphylaxis were reported in a study of brimonidine 0.025% for relief of ocular redness; the most common side effects were instillation site pain and dry eye [8]. To the authors’ knowledge, there has been no prior study designed to observe the effects of brimonidine 0.025% on upper eyelid position and palpebral fissure height. In this randomized, controlled, double blind study, we aim to detect changes in palpebral fissure height after administration of brimonidine 0.025% in hopes to gain a deeper understanding of the drug’s potential.

This single-center, randomized, double-masked study (ClinicalTrials.gov identifier: NCT03782701) received full approval by the Institutional Review Board (IRB) of the University of Miami and complied in accordance with the ethical principles originating from the Declaration of Helsinki and Health Insurance Portability and Accountability act. All subjects were given full disclosure of study information prior to enrolment and provided informed consent before testing was conducted. All subjects also completed a photographic consent form prior to participating in the study. The data was collected at the Bascom Palmer Eye Institute oculoplastic clinic.

Study design and randomization

This was a parallel group study which compared two treatments, brimonidine 0.025% and saline. The sample size was determined using data from Kirpatrick et al. [3] and sample size estimation was performed using SAS (Statistical Analysis Software 9.4, SAS Institute, Cary, North Carolina, USA). Given the short duration of the study, 40 people in the recruitment dataset was considered to be adequate to account for any possible loss to follow up. The type of randomization was a simple randomization. The left and right eyes of each subject were randomized to receive a drop of either brimonidine 0.025% or saline solution, based on the excel workbook provided by the biostatistician. A third party (WL) repackaged and relabeled the products. Based on the randomization results, two examiners (SH, YP) who were blinded to the intervention applied the products to subject’s eyes.

Study population

Forty-three healthy adult (age ≥ 18 years) subjects were recruited by SH and YP between March and October 2019 by way of voluntary response to participate in this study. Exclusion criteria included pregnancy, known contraindications or sensitivities to brimonidine 0.025%, ocular surgery within the past 3 months or refractive surgery within the past 6 months, and previous eyelid or orbital surgery. Other conditions that excluded patients from the study included grossly abnormal lid margins, anatomical abnormalities, variable ptosis or eyelid position (e.g., myasthenia gravis, thyroid eye disease, or blepharospasm), significant pre-existing ptosis of any cause (defined as marginal reflex distance 1 <1mm), or presence of an ocular infection. Additional historical exclusion criteria included recent use (within 5 days of study participation) of ocular vasoconstrictors (redness relievers), decongestants, antihistamines, phenylephrine dilating drops, or other topical ophthalmic agents with the exception of artificial tears.

Image collection

All images were taken by the same examiners (SH, YP) in a room with consistent lighting conditions. An apparatus affixed to the slit-lamp was used to take standardized pictures of the subjects. A clamp affixed to the slit lamp’s vertical side bar was used to hold the camera (iPhone, Apple Inc., Cupertino, CA, U.S.A) at a standardized distance from the subjects and aligned vertically with respect to their lateral canthus. A small portable light (mini-LED selfie light, CHSMONB) was affixed to the camera to create the pupillary light reflex. Subjects’ heads were placed on the chin rest, with forehead resting on the brow bar, to position them on a consistent plane in relation to the camera. Subjects were instructed to relax and focus on the center of the camera lens with their eyes in primary gaze. As a standard reference of spatial information, a ruler was attached to the brow bar of the slit lamp to calculate the physical spacing between pixels and the position of the image in space through the calibration of the pixel/millimeter ratio (Figure 1). Photographs of the subjects were taken before application of the brimonidine 0.025% and saline drops, as well as 5 minutes, 15 minutes, and 30 minutes after application. At each time point, conjunctival redness was observed, and sensations of pain, dryness, itchiness, and burning were recorded. Conjunctival redness was measured subjectively by the same examiners on a scale of 0 (no redness) to 3 (severe redness). Prior to grading subjects, the two examiners examined a series of variably injected eyes to establish a consensus on what level of redness corresponds to none, mild, moderate, and severe in order to ensure consistency in conjunctival redness grading. Photographs were stored on a secure computer and saved as JPEG files. Subjects’ intraocular pressure was measured with a tonopen prior to the application of the brimonidine 0.025% and saline drops and after the last photograph was taken at the 30-minute time point.

Eyelid measurements

Palpebral fissure height (PFH) was measured as the distance between the superior lid margin to the inferior lid margin passing through the center of the pupil. Marginal reflex distance 1 (MRD1) and marginal reflex distance 2 (MRD2) were measured as the distances from the center of the pupil as indicated by the pupillary light reflex to the upper and lower eyelid margins respectively, in the mid-pupillary line.

All measurements were performed on digitally captured images calibrated for size using image editing software (Adobe Photoshop 21.0.2 release, Adobe Systems Inc., San Jose, CA, USA) using a ruler attached to the slit lamp as a measurement scale of reference. The number of pixels per millimeter in each photograph was set by averaging the number of pixels per 10 mm of the calibrating ruler in the photograph on the slit lamp. This scale was then used to measure the palpebral fissure height, MRD1, and MRD2 using a caliper tool as described above.

Data analysis

Data was recorded and stored in a secure Microsoft Office Excel (Microsoft Corporation, Redmond, WA, U.S.A.) spreadsheet. Data was graphically analyzed by one examiner (JL) with Adobe Photoshop. This examiner was blinded to which eye received study drug.

The results were expressed as mean ± standard deviation (SD). Statistical analysis was carried using the SAS software version 9.4 (SAS, Cary, NC, U.S.A). Statistical tests used were chi square test, Wilcoxon signed rank test, and the paired t-test where appropriate. A p-value of less than 0.05 was considered statistically significant.

Analysis was performed on 86 eyes of 43 subjects. The subjects’ average age (mean ± SD) was 36 ± 13.84 years. There were 24 (55.8%) female subjects and 19 male subjects (44.2%). Three subjects were excluded due to brow ptosis that prevented accurate palpebral fissure height measurements. 

PFH measurements in brimonidine versus saline-treated groups are described in Table 1. No statistically significant difference in PFH was found between the brimonidine 0.025% and saline-treated groups at any time point.

Ocular redness measurements are presented in Table 1. Severity of ocular redness at baseline did not differ significantly between treatment groups (p=1.00). Ocular redness was significantly lower in brimonidine 0.025%-treated eyes compared to saline-treated eyes at all post-treatment time points (p=0.001).

The mean IOP pre-treatment was 16.54 ± 3.63mmHg (10-25mmHg) in brimonidine 0.025%-treated eyes and 16.70 ± 3.42mHg (10-26mmHg) in saline-treated eyes. There was no statistically significant difference in IOP between treatment groups at baseline. At the 30-minute time point, IOP did not differ between treatment groups (p>0.05) (Table 1).

There were no serious adverse events during the study. The most common symptoms reported were mild pain, discomfort, itchiness, and dryness. There was no significant difference in drop comfort at any time point between treatment groups (p>0.05).

An “as treated” sensitivity analysis was done in which the subjects were analyzed in their treatment groups, rather than the randomization group as was done in the primary, intention-to-treat analysis. The results of this sensitivity analysis were similar to and would lead to the same conclusions as the results of primary analysis.

Brimonidine tartrate is an adrenergic agonist with 1000-fold more selectivity for the α-2 receptor than the α-1 receptor [12]. Brimonidine 0.025% is now available over the counter to treat ocular redness, and its potential effect on eyelid height has not yet been investigated. Like other alpha-adrenergic agonistic agents, such as those present in decongestant or glaucoma eyedrops [1], we hypothesized that brimonidine 0.025% could also increase palpebral fissure height by stimulating Muller’s muscle, which maintains upper eyelid tone and provides up to 2mm of additional elevation as part of the startle response [13]. Previous studies of higher doses of brimonidine found no significant effect on eyelid height, but the effects of brimonidine can vary based on dosage. For example, higher doses are associated with hyperemia [14] while low doses cause conjunctival whitening [15]. It is possible that brimonidine’s effect on PFH is also dose-dependent, and that higher or lower concentration may be required to achieve a significant increase in eyelid height.

In this randomized controlled trial, we found no significant difference in palpebral fissure height between eyes treated with brimonidine 0.025% versus saline solution. Other studies reporting on the effects of brimonidine 0.2% on palpebral fissure size vary in their findings, and similar to our study, some report no change in upper or lower eyelid position [11,16,17]. One of these studies by Mendonca et al. found no significant effect, but used fewer subjects and only started recording eyelid height 30 minutes after drop instillation [17]. The other studies included patients affected by Horner’s syndrome [11] and elevated intraocular pressure [16] in contrast to our cohort without ocular comorbidities. Rehmani et al. found that brimonidine 0.1% dramatically improved ptosis in their Horner’s patients [12], which could be explained by α-1 receptor upregulation due to chronic deficiency of norepinephrine in the synapse [18]. In healthy individuals, it is possible that selective action of brimonidine 0.025% on the α-2 receptor counteracts some of the effects at the α-1 receptor.

The lack of significant difference in palpebral fissure height or intraocular pressure between brimonidine 0.025% and saline treated eyes could also be due to systemic absorption of the drug. Of note, there was a mean increase in palpebral fissure height of approximately 0.3mm in both groups and a mean decrease in IOP by 1mmHg in both groups. When comparing brimonidine and apraclonidine use for elevated intraocular pressure, Yuksel et al. found a contralateral effect of brimonidine on IOP in the untreated fellow eyes, and observed this consensual effect to be more powerful with brimonidine than apraclonidine [16]. Although they did not evaluate eyelid height, it is possible that brimonidine could be absorbed systemically and affect Muller’s muscle on the contralateral eye.

Similar to other studies evaluating brimonidine 0.025% [8], we did not find a significant difference in IOP between brimonidine and saline-treated eyes, suggesting that its effect on IOP is limited at lower doses [15,16].

Although brimonidine is associated with hyperemia at the higher doses used for intraocular pressure lowering [14], at low doses it has been associated with conjunctival whitening through vasoconstriction [15], and has been used in ophthalmic surgery to control bleeding [19,20]. Similarly, to other studies of low-dose brimonidine [8,10], we found a significant decrease in ocular redness in patients treated with brimonidine 0.025% compared to saline. Conjunctival whitening was observed at our latest time point (30 minutes post instillation) and has been reported to last up to 4 hours [8,10].

This study has important limitations. Potential systemic absorption of the drug was not accounted for in this study. However, there was a clear difference in eye redness between the brimonidine and saline eyes, suggesting that if systemic absorption of the drug did occur, it was minimal. Still, future studies may consider treating eyes with a saline drop first, taking height measurements, and instill a brimonidine drop after an appropriate wash out period. We attempted to standardize each subject’s facial expressions when taking the photographs by asking subjects to rest their foreheads against the slit lamp brow-bar, to look directly at the camera lens, to relax their foreheads, and to not smile. Still, we noticed small differences in facial expressions that may have affected palpebral fissure height measurements at different time points. In general, these appeared to affect both brimonidine 0.025% and saline-treated eyes similarly. Another limitation is the redness grading scale, which is subjective in nature. We attempted to minimize variability in conjunctival redness measures by utilizing the same examiners to perform the redness grading. Additionally, it is possible that the placebo saline solution caused irritation and increased ocular redness, especially at the 5-minute time point. However, evaluation of ocular redness was not the primary goal of this study and has been well described previously. Some of the study’s strengths are its randomized prospective design, and the standardized, blinded assessment of subjects’ pictures.

In summary, in a cohort of healthy subjects, application of brimonidine 0.025% led to significant conjunctival whitening, but did not generate a significant increase in PFH or MRD1. No significant adverse effects were seen, and patients tolerated the medication well. Further studies may be helpful to further assess brimonidine 0.025%’s effect on different patient populations, including those with pre-existing ptosis.

Authors have no conflict of interests to report.

No funding source was used for this study.

SH, WL, LV, AL, AA were responsible for designing and editing the study protocol, interpreting results, and editing the manuscript. SH was responsible for writing the IRB protocol, conducting the search, extracting and analyzing data, interpreting results, writing and editing the manuscript report. JL, YP, OO were responsible for conducting the research, extracting data, and editing the manuscript. JL was also responsible for analyzing data. LV was responsible for statistical analyses and creating ‘Summary of findings’ tables.