Abstract
Objective
Systemic antihypertensive medications may influence glaucoma-related outcomes through intraocular pressure–dependent and vascular mechanisms. While topical α₂-agonists are established glaucoma therapies, the association between systemic α₂-agonists and glaucoma hazard remains unexamined; we compared their hazard with systemic β-blockers in adults with hypertension.
Methods and analysis
We conducted a multicentre retrospective cohort study using deidentified electronic health records. Adults aged ≥40 years with hypertension and chronic exposure to systemic α₂-agonists or β-blockers were included. Patients with pre-existing ocular hypertension (OHT), primary open-angle glaucoma (POAG), concomitant comparator use, or inadequate ophthalmic follow-up were excluded. We balanced cohorts using propensity score matching (PSM) and assessed the hazard of OHT and POAG at 1, 3, and 5 years. Cox models estimated adjusted HRs (aHRs) with 95% confidence intervals (CIs).
Results
Patients treated with α₂-agonists (n=4186) were matched to those treated with β-blockers (n=360 399), resulting in 4152 patients per group after PSM. In Cox models (on the unmatched cohorts) adjusted for demographics, comorbidities, medications, laboratory and biometric measures, socioeconomic indicators and selected procedures, α₂-agonist exposure was associated with a lower hazard of OHT (aHR (95% CI) 0.582 (0.473 to 0.716), 0.559 (0.468 to 0.669) and 0.534 (0.450 to 0.634)) and POAG (aHR (95% CI) 0.359 (0.300 to 0.429), 0.333 (0.282 to 0.392) and 0.334 (0.286 to 0.390)) at 1-year, 3-year, and 5-year follow-up, respectively (all p<0.001).
Conclusions
Systemic α₂-agonist use was associated with a substantially lower hazard of glaucoma-related diagnoses compared with systemic β-blocker use. These findings are limited by the observational design, potential residual confounding, and reliance on diagnostic coding. These results warrant further investigation in prospective mechanistic studies and randomised clinical trials.
Keywords: Glaucoma, Treatment Medical, Aqueous humour, Intraocular pressure
WHAT IS ALREADY KNOWN ON THIS TOPIC.
WHAT THIS STUDY ADDS
Systemic α₂-agonist use was associated with a significantly lower hazard of ocular hypertension and primary open-angle glaucoma compared with systemic β-blocker use at 1-year, 3-year, and 5-year follow-up.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY
These findings are hypothesis-generating and support further prospective and mechanistic studies; they suggest that antihypertensive medication selection may warrant consideration among patients with increased susceptibility to glaucoma.
Introduction
Glaucoma remains a leading cause of irreversible blindness worldwide,1 characterised by progressive optic neuropathy and retinal ganglion cell loss that often develops insidiously before detection. Although elevated intraocular pressure (IOP) is the most recognised and modifiable risk factor,2 glaucoma pathogenesis is multifactorial, involving vascular dysregulation, oxidative stress, and neurodegenerative pathways.3 4 Systemic conditions,5 6 such as hypertension, diabetes and dyslipidaemia as well as systemic medications,7,9 have been linked to glaucoma susceptibility, yet the directionality and strength of these associations remain inconsistent across studies.
Among patients with hypertension, long-term exposure to systemic antihypertensive agents represents a potential modifiable influence on ocular physiology. β-Blockers, for instance, can modestly lower IOP by inhibiting aqueous humour production, and topical formulations have long been used in glaucoma management.10 However, the ocular effects of systemic β-blockers are less predictable, with some studies suggesting a neutral or even adverse impact due to reduced ocular perfusion pressure or nocturnal hypotension.11,13 Other antihypertensive classes, including angiotensin-converting enzyme inhibitors, angiotensin receptor blockers (ARBs), and calcium channel blockers (CCBs), have yielded mixed or contradictory findings regarding glaucoma risk.14
α₂-Adrenergic agonists, by contrast, represent a pharmacologic class with established intraocular benefits in topical form. Agents such as brimonidine and apraclonidine lower IOP by suppressing aqueous secretion and enhancing uveoscleral outflow, while also exhibiting potential neuroprotective properties through modulation of glutamate excitotoxicity and oxidative injury.15,17 Despite these well-documented ocular effects, the potential influence of systemic α₂-agonists, such as clonidine, guanfacine, dexmedetomidine, and methyldopa, on glaucoma development has never been systematically explored. Because these drugs exert central sympatholytic actions, reduce circulating catecholamines, and may stabilise ocular blood flow, it is plausible that chronic exposure could influence pathways relevant to glaucomatous damage,18 19 although direct effects remain uncertain.
The interplay between systemic α₂-agonists and glaucoma has potential clinical relevance for hypertensive patients, many of whom are treated chronically with cardiovascular agents that may either mitigate or exacerbate the progression of ocular disease. However, to date, no large-scale population-based study has directly compared the hazard of glaucoma among users of systemic α₂-agonists versus β-blockers, two antihypertensive classes with overlapping therapeutic indications but distinct ocular mechanisms.
Therefore, the present study aimed to evaluate whether systemic α₂-agonist use is associated with an altered hazard of glaucoma-related diagnoses compared with systemic β-blocker use in a large, real-world, propensity-matched cohort. We hypothesised that α₂-agonist exposure would be associated with a lower hazard of glaucoma diagnoses, potentially reflecting both intraocular and neurovascular protective pathways.
Materials and methods
Data source
This retrospective cohort study used deidentified patient data from the TriNetX Analytics Platform (TriNetX, LLC), a global federated health research network that aggregates electronic health records (EHRs) from more than 30 countries. For this analysis, we accessed data from over 130 million patients across 67 United States healthcare organisations. Because only deidentified data were used, the study was deemed exempt from Institutional Review Board (IRB) approval by the University of Arkansas for Medical Sciences. The research adhered to the ethical principles of the Declaration of Helsinki (2013 version) and followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.20 Diagnoses were identified using International Classification of Diseases, Tenth Revision (ICD-10) codes. Diagnoses recorded before 2015 using the Ninth Revision (ICD-9) were internally mapped to ICD-10 equivalents to ensure consistency and comparability across healthcare organisations. Medications were identified using RxNorm codes, and procedures were captured using Current Procedural Terminology (CPT) codes.
Inclusion/exclusion criteria and study groups
Patients aged 40 years or older who have been diagnosed with a hypertensive disease (ICD code: I10-I1A) and treated with systemic α₂-agonists (clonidine (RxNorm: 2599), guanfacine (RxNorm: 40114), dexmedetomidine (RxNorm: 48937), or methyldopa (ATC: C02AB)) (cohort 1) were compared with those treated with β-blockers (VA: CV100) (cohort 2). Tizanidine was not assessed because, although an α₂-agonist, it is not used for hypertension management. To limit the analysis to patients with chronic medication use, only those with at least four prescriptions of the assessed medication were included. Other exclusion criteria included ocular hypertension (OHT) (H40.05) and primary open-angle glaucoma (POAG) (H40.11) at or before the index date. To ensure adequate ophthalmologic follow-up, eligible patients had a minimum of three eye examinations within 5 years following initiation of the exposure medication. We excluded patients concomitantly exposed to the other comparator medication. The index date was defined as the first occurrence when a patient met all inclusion criteria within the study period (ie, follow-up began after the fourth α2-agonists or β-blocker prescription), spanning from 1 January 2005 to 1 January 2025.
Propensity score matching
Using propensity score matching (PSM) (1:1 nearest-neighbour greedy algorithm with a calliper of 0.1 SD), patients treated with α₂-agonists were matched to those treated with β-blockers to reduce baseline confounding. Matching was performed within the TriNetX platform, and balance between cohorts was assessed by comparing standardised mean differences of covariates before and after matching, with values below 0.1 indicating adequate covariate balance. We selected covariates for matching based on clinical relevance, including variables that could plausibly confound the relationship between exposure and outcomes (systemic comorbidities, procedures, medication exposures, and laboratory measures), as well as factors accounting for variability in access to and utilisation of care within the EHR setting. PSM included socioeconomic indicators related to literacy and education (Z55), employment and unemployment (Z56), occupational exposures (Z57), and environmental factors (Z58). To mitigate detection bias and account for variability in ophthalmologic care utilisation, CPT codes for comprehensive and intermediate eye examinations (92002, 92004, 92012, 92014) were also incorporated into the matching process.
We further conducted a Cox proportional hazards analysis on the unmatched cohorts to retain the full sample size and evaluate the temporal association between exposure and outcomes. The Cox models were adjusted for the same covariates used in the PSM process to estimate adjusted hazard ratio (aHRs) and their corresponding 95% confidence intervals (CIs). Both propensity score–matched and Cox proportional hazards analyses were used as complementary approaches to assess the robustness and consistency of the findings.
Clinical endpoints and statistical analysis
The clinical outcomes assessed in this analysis were the hazard of OHT and POAG. All outcomes were evaluated at 1, 3, and 5 years after the index date. Patients were included in each time-point analysis based on the availability of sufficient follow-up duration. For example, patients with at least 3 years of follow-up were included in the 1-year and 3-year analyses. Those without continued follow-up beyond 3 years, due to death, loss to follow-up, or absence of subsequent encounters within the network, were not included in the 5-year analyses. Because TriNetX is based on routinely collected EHR data, continued follow-up depends on the presence of clinical encounters within participating healthcare organisations, and patients may be excluded from later time-point analyses if subsequent encounters are not recorded.
Baseline characteristics were summarised using descriptive statistics, with continuous variables reported as means and SD and categorical variables expressed as proportions. All statistical analyses were performed within the TriNetX analytics platform, with statistical significance defined as a two-sided p-value <0.05.
Results
Baseline characteristics
Patients treated with α₂-agonists (n=4186) were matched to those treated with β-blockers (n=360 399), resulting in 4152 patients per group after PSM. The mean age was 59.7±12.8 years in the α₂-agonist cohort and 59.7±13.0 years in the β-blocker cohort. Sex distribution was well balanced, with 38.7% male in the α₂-agonist group compared with 39.5% in the β-blocker group. The racial composition of the α₂-agonist cohort included 62.4% White, 25.7% Black or African American, and 2.4% Asian. In comparison, the β-blocker cohort consisted of 61.8% White, 26.5% Black or African American, and 2.1% Asian. Regarding ethnicity, 8.6% of the α₂-agonist cohort and 8.6% of the β-blocker cohort were Hispanic or Latino. Comorbidities, procedures, medication use, and laboratory variables were well balanced between groups after matching. A complete comparison of baseline characteristics before and after PSM is provided in table 1 and online supplemental table 1.
Table 1. Baseline characteristics (demographics and diagnoses) of patients receiving α₂-agonists vs β-blockers, before and after PSM.
| Before PSM matching | After PSM matching | |||||
|---|---|---|---|---|---|---|
| Characteristics | α₂-Agonist n=4186 (%) | β-Blockers n=360 399 (%) | Standard difference | α₂-Agonist n=4152 (%) | β-Blockers n=4152 (%) | Standard difference |
| Demographics | ||||||
| Age at index (mean±standard deviation) | 59.6±12.8 | 67.0±12.3 | 0.585 | 59.7±12.8 | 59.7±13.0 | 0.003 |
| Male | 1616 (38.6) | 166 234 (46.3) | 0.156 | 1606 (38.7) | 1639 (39.5) | 0.016 |
| Female | 2566 (61.4) | 192 658 (53.7) | 0.156 | 2546 (61.3) | 2510 (60.5) | 0.018 |
| White | 2614 (62.5) | 234 198 (65.3) | 0.057 | 2590 (62.4) | 2565 (61.8) | 0.012 |
| Black or African American | 1068 (25.5) | 74 014 (20.6) | 0.117 | 1065 (25.7) | 1101 (26.5) | 0.02 |
| Asian | 99 (2.4) | 15 697 (4.4) | 0.111 | 99 (2.4) | 87 (2.1) | 0.02 |
| American Indian | 30 (0.7) | 1228 (0.3) | 0.052 | 29 (0.7) | 34 (0.8) | 0.014 |
| Pacific Islander | 18 (0.4) | 1627 (0.5) | 0.003 | 18 (0.4) | 21 (0.5) | 0.011 |
| Other race | 139 (3.3) | 11 340 (3.2) | 0.009 | 138 (3.3) | 149 (3.6) | 0.015 |
| Unknown race | 214 (5.1) | 20 807 (5.8) | 0.03 | 213 (5.1) | 195 (4.7) | 0.02 |
| Hispanic or Latino | 357 (8.5) | 29 440 (8.2) | 0.012 | 356 (8.6) | 355 (8.6) | 0.001 |
| Not Hispanic or Latino | 3113 (74.4) | 263 085 (73.3) | 0.026 | 3088 (74.4) | 3074 (74.0) | 0.008 |
| Unknown ethnicity | 712 (17.0) | 66 386 (18.5) | 0.039 | 708 (17.1) | 723 (17.4) | 0.01 |
| Diagnoses | ||||||
| Cerebrovascular (I60–I69) | 569 (13.6) | 61 298 (17.1) | 0.096 | 565 (13.6) | 578 (13.9) | 0.009 |
| Resistant hypertension (I1A.0) | 21 (0.5) | 390 (0.1) | 0.071 | 20 (0.5) | 28 (0.7) | 0.025 |
| Heart failure (I50) | 330 (7.9) | 68 199 (19.0) | 0.33 | 329 (7.9) | 346 (8.3) | 0.015 |
| Chronic ischaemic heart (I25) | 494 (11.8) | 120 344 (33.5) | 0.537 | 494 (11.9) | 513 (12.4) | 0.014 |
| Atrial fibrillation (I48) | 192 (4.6) | 66 020 (18.4) | 0.443 | 192 (4.6) | 192 (4.6) | <0.001 |
| Atrioventricular block (I44) | 179 (4.3) | 27 862 (7.8) | 0.147 | 177 (4.3) | 197 (4.7) | 0.023 |
| Conduction disorders (I45) | 250 (6.0) | 23 121 (6.4) | 0.019 | 245 (5.9) | 261 (6.3) | 0.016 |
| Cardiac arrhythmias (I49) | 677 (16.2) | 76 234 (21.2) | 0.13 | 670 (16.1) | 677 (16.3) | 0.005 |
| Paroxysmal tachycardia (I47) | 107 (2.6) | 26 694 (7.4) | 0.225 | 107 (2.6) | 121 (2.9) | 0.021 |
| Diabetes mellitus (E08–E13) | 1925 (46.0) | 155 931 (43.4) | 0.052 | 1909 (46.0) | 1951 (47.0) | 0.02 |
| Lipidemia (E78) | 2591 (62.0) | 250 555 (69.8) | 0.166 | 2576 (62.0) | 2664 (64.2) | 0.044 |
| Overweight and obesity (E66) | 1719 (41.1) | 107 963 (30.1) | 0.232 | 1702 (41.0) | 1671 (40.2) | 0.015 |
| Sleep disorders (G47) | 1793 (42.9) | 101 795 (28.4) | 0.307 | 1764 (42.5) | 1711 (41.2) | 0.026 |
| Chronic lower respiratory diseases (J40–J4A) | 1557 (37.2) | 89 112 (24.8) | 0.271 | 1539 (37.1) | 1577 (38.0) | 0.019 |
| Thyroid disorders (E00–E07) | 1128 (27.0) | 82 706 (23.0) | 0.091 | 1116 (26.9) | 1110 (26.7) | 0.003 |
| Chronic kidney disease (N18) | 685 (16.4) | 65 856 (18.3) | 0.052 | 678 (16.3) | 750 (18.1) | 0.046 |
| Nicotine dependence (F17) | 1192 (28.5) | 49 841 (13.9) | 0.363 | 1168 (28.1) | 1181 (28.4) | 0.007 |
| Attention-deficit hyperactivity disorders (F90) | 321 (7.7) | 3021 (0.8) | 0.343 | 294 (7.1) | 272 (6.6) | 0.021 |
| Myopia (H52.1) | 770 (18.4) | 38 803 (10.8) | 0.216 | 754 (18.2) | 776 (18.7) | 0.014 |
PSM, propensity score matching.
Development of OHT and POAG
Propensity score–matched analysis
In the propensity score–matched cohorts, there were no statistically significant differences in the development of OHT diagnosis at 1 year (1.40% vs 1.07%, p=0.186) and at 3 years (2.04% vs 2.42%, p=0.242) following the index date. However, by 5 years, the development of OHT was significantly less among α₂-agonist users (2.26% vs 3.86%, p<0.001). For POAG, there was no significant difference at 1 year (1.45% vs 1.48%, p=0.903). At 3 years, the development of POAG diagnosis was significantly lower with α₂-agonists (1.92% vs 3.31%, p<0.001), and this pattern persisted at 5 years (2.24% vs 4.17%, p<0.001) (table 2).
Table 2. The development of OHT and POAG in patients treated with α₂-agonists vs β-blockers over 1-year, 3-year and 5-year follow-up periods, after propensity score matching.
| Outcome | α₂-Agonists (%) | β-Blockers (%) | P value |
|---|---|---|---|
| 1-year follow-up | |||
| OHT | 1.40 | 1.07 | 0.186 |
| POAG | 1.45 | 1.48 | 0.903 |
| 3-year follow-up | |||
| OHT | 2.04 | 2.42 | 0.242 |
| POAG | 1.92 | 3.31 | <0.001 |
| 5-year follow-up | |||
| OHT | 2.26 | 3.86 | <0.001 |
| POAG | 2.24 | 4.17 | <0.001 |
Values in bold indicate statistically significant.
OHT, ocular hypertension; POAG, primary open-angle glaucoma.
Cox proportional hazards model
A Cox proportional hazards model was conducted on the unmatched cohort, adjusting for the same set of covariates. The analysis demonstrated that the α₂-agonist group was associated with a lower hazard of OHT compared with the β-blocker group at 1 year (aHR=0.582, 95% CI 0.473 to 0.716, p<0.001), 3 years (0.559, 95% CI 0.468 to 0.669, p<0.001) and 5 years (0.534, 95% CI 0.450 to 0.634, p<0.001). POAG showed a similar trend, with lower associated hazard at 1 year (aHR=0.359, 95% CI 0.300 to 0.429, p<0.001), 3 years (0.333, 95% CI 0.282 to 0.392, p<0.001) and 5 years (0.334, 95% CI 0.286 to 0.390, p<0.001) (figure 1).
Figure 1. The adjusted HR (aHR) of ocular hypertension (OHT) and primary open-angle glaucoma (POAG) in patients treated with α₂-agonists vs β-blockers over 1-year, 3-year, and 5-year follow-up (FU) periods.
Discussion
In this retrospective cohort study leveraging a large multicentric database, we found that systemic α₂-agonist use in adults aged 40 years and older with hypertension was associated with a significantly lower hazard of OHT and POAG compared with β-blocker use. The association persisted at 1, 3 and 5 years after robust multivariable adjustment. These findings indicate an association between systemic α₂-agonist use and lower glaucoma hazard compared with β-blockers, suggesting possible differences in their ocular or neurovascular effects.
Our results must be interpreted in the context of prior studies evaluating systemic medications and glaucoma outcomes. A population-based analysis demonstrated that systemic β-blocker use was associated with slightly lower IOP but not with a reduced prevalence of glaucoma after covariate adjustment.21 A systematic review and meta-analysis similarly reported a modest reduction in glaucoma odds associated with β-blocker use (OR=0.83, 95% CI 0.75 to 0.92) across seven studies, but the underlying studies were heterogeneous.13 A large population-based analysis of antihypertensive classes further suggested that non–β-blocker agents (ARBs and CCBs) may increase glaucoma risk, highlighting variability across medication classes.14
Regarding α-adrenergic agonists, the existing literature largely focuses on topical ocular α₂-agonists such as brimonidine and apraclonidine, which are established glaucoma therapies. These agents reduce aqueous humour production and modestly enhance uveoscleral outflow, resulting in modest IOP reductions.18 22 Pharmacological studies have demonstrated the presence of α₂-adrenergic receptors in ocular tissues, including the ciliary body and retina, supporting these effects.17 23 Other reviews underscore that these drugs’ dual actions (reduced aqueous secretion and increased uveoscleral outflow) are well accepted in glaucoma therapy.18 However, direct evidence linking systemic α₂-agonist exposure to glaucoma development remains limited.
Thus, our findings suggest a potential association between systemic α₂-agonist use and lower glaucoma hazard compared with systemic β-blockers, further raising interesting mechanistic questions.
Notably, randomised trials of topical α₂-agonists versus β-blockers have suggested class differences beyond IOP. In the Low-Pressure Glaucoma Treatment Study (LoGTS), brimonidine and timolol achieved similar IOP, yet visual-field progression was less frequent with brimonidine,24 and earlier studies (eg, apraclonidine vs timolol) similarly demonstrated α₂-agonist IOP-lowering with differing profiles.25 However, these findings are based on topical administration and are not directly comparable to systemic exposure, given substantial differences in ocular bioavailability and pharmacokinetics, and therefore provide only indirect biologic context rather than direct support for the observed associations.
Systemic α₂-agonists are primarily used for central and peripheral sympathetic inhibition, yet pharmacologic data suggest plausible ocular effects.26 In anaesthesiology studies, agents such as clonidine and dexmedetomidine attenuate acute elevations in IOP associated with physiological stress.27 While these findings do not establish long-term IOP modulation, they demonstrate that systemic α₂-agonists can influence ocular pressure dynamics under stress conditions.
One potential mechanism involves suppression of sympathetic tone affecting aqueous humour dynamics or ocular vasculature. It remains uncertain whether systemic α₂-agonists achieve sufficient ocular concentrations to replicate the effects observed with topical agents. However, if present at low intraocular concentrations, they may influence aqueous humour dynamics by reducing aqueous production or modestly enhancing uveoscleral outflow. The documented presence of α₂-adrenergic receptors in ocular tissues may support the biological plausibility of such receptor-mediated effects.23 28 However, direct mechanistic links between systemic α₂-agonist exposure and glaucoma development remain to be established.
Another compelling possibility is that α₂-agonists exert neuroprotective or vasoregulatory effects independent of IOP. Experimental data suggest α₂-agonists may upregulate neurotrophic factors and reduce apoptotic signalling in retinal injury models.23 Additionally, by attenuating sympathetic-mediated vasoconstriction, these agents may stabilise ocular perfusion pressure and microcirculatory flow at the optic nerve head. Compromised ocular blood flow and vascular dysregulation are increasingly recognised contributors to glaucomatous optic neuropathy, particularly in normal-tension glaucoma.29 Flammer’s vascular dysregulation theory emphasises the role of perfusion instability, vasospasm and endothelial dysfunction in optic nerve damage.30 Agents that mitigate sympathetic drive could therefore plausibly reduce ischaemic or reperfusion-related injury to the optic nerve.
The protective role of systemic β-blockers on IOP is reasonably established: β-adrenergic receptors in the ciliary body drive aqueous humour formation, and systemic β-blockers can lower IOP modestly (although less consistently than topical β-blockers).21 31 32 Indeed, a recent report affirmed the IOP-lowering effect of systemic β-blockers, while finding no such effect among other systemic drug classes.33 However, the discrepancy between IOP lowering and glaucoma protection in observational studies has long been noted. Systemic β-blockers may possibly have unintended systemic or ocular side effects (eg, blood pressure-lowering, ocular perfusion pressure compromise, or upregulation of alternate pathways) that partially offset their benefit.34 35
In susceptible individuals, particularly those with low baseline perfusion pressure or vascular dysregulation, systemic β-blockade may further reduce ocular perfusion pressure and increase optic nerve vulnerability.36 Prior studies have raised concerns regarding glaucoma risk in patients receiving systemic antihypertensive therapy under conditions of low mean arterial pressure.14 37 Thus, the net effect of β-blockers on glaucoma risk may reflect a balance between IOP lowering and vascular compromise.
Moreover, it is conceivable that patients on systemic α₂-agonists may have different haemodynamic or autonomic profiles (eg, more sympathetic suppression) that are beneficial for vascular homeostasis, thereby reducing episodic ischaemic stress on the optic nerve.
Thus, while systemic β-blockers may confer some protective IOP effect, their net influence on glaucoma development may be moderated by haemodynamic or vascular trade-offs, whereas systemic α₂-agonists may tilt the balance towards vascular stability or neuroprotection.
Our findings suggest a potential association between systemic α₂-agonist use and glaucoma-related outcomes. Although the observed HRs are larger than typically expected for a systemic medication with indirect ocular effects, the consistency across time points and analytical approaches supports a potentially meaningful association; however, residual confounding cannot be excluded. As systemic α₂-agonists are not prescribed for ocular indications, this observation may motivate further investigation in prospective and mechanistic studies evaluating their effects on IOP, optic nerve structure (eg, retinal nerve fibre layer), visual function and glaucoma development.
In practice, these findings may have implications for antihypertensive selection in patients at elevated risk of glaucoma.38 However, given the observational design and differences between systemic and topical α₂-agonist effects, any clinical application or recommendation must await prospective confirmation and safety evaluation (ie, relative to side effects and medication tolerance).
Mechanistic studies (animal, pharmacokinetic, receptor expression) are needed to quantify ocular tissue penetration, receptor engagement and long-term effects of systemic α₂-agonists. Additionally, future observational studies should explore dose-response, duration effects, subclass analyses (eg, clonidine vs methyldopa vs guanfacine) and interactions with other modifiers of glaucoma development (eg, perfusion pressure, vascular comorbidities).
This study has several limitations. Despite extensive PSM and multivariable adjustment, residual confounding cannot be ruled out. The TriNetX database lacks detailed ophthalmic parameters such as IOP, corneal thickness, optic disc structure, and visual field data. Furthermore, the platform does not provide access to individual-level data and does not capture medication adherence or duration; moreover, defining exposure based on prescription counts may oversimplify true medication use, raising the possibility of exposure misclassification. Diagnostic accuracy is limited by reliance on ICD-10 codes without structural or functional validation, which may over-represent or under-represent true disease, introduce detection bias, and not reflect severity or progression. Although requiring at least three eye examinations within 5 years reduced surveillance bias, differential access to ophthalmic care may persist.
As an observational study, causal inference remains speculative, and confounding by indication remains a limitation, as patients receiving systemic α₂-agonists may be systematically different from those receiving β-blockers in ways that are not fully captured by available covariates. Although we adjusted for comorbidities that may reflect disease severity (eg, resistant hypertension, heart failure and chronic ischaemic heart disease), these proxies do not fully capture hypertension severity or blood pressure variability, and more granular measures are needed in future studies. Additionally, the magnitude of the observed effect sizes is notable for a systemic medication with indirect ocular effects and may reflect the influence of residual confounding, bias or exposure misclassification. Furthermore, exclusion of concomitant comparator antihypertensive medication enhances internal validity but limits generalisability to patients receiving combination therapy. Systemic α₂-agonists are typically prescribed in more selective clinical contexts (eg, resistant hypertension), and therefore, the study population may not be representative of the broader hypertensive population. Finally, differences among individual α₂-agonists and β-blockers were not analysed, and findings from this United States-based cohort may not extend to other health systems. Hence, these findings highlight a potential association that warrants further investigation in prospective and mechanistic studies to validate the hypothesis. Future work should focus on more granular observational analyses, pharmacologic and mechanistic studies, and if supported, prospective investigations to better define the relationship between systemic α₂-agonists and glaucoma risk.
In summary, this study provides the first large-scale evidence that systemic α₂-agonist use is associated with a significantly lower hazard of developing glaucoma and related diagnostic entities compared with systemic β-blocker use in adults with hypertension. The consistent reduction in hazard across multiple outcomes and time frames, coupled with biologically plausible mechanisms involving both IOP modulation and vascular-neuroprotective pathways, supports further exploration of this association. Prospective and mechanistic studies are needed to determine whether systemic α₂-agonists can meaningfully influence glaucoma pathogenesis or progression and to define the therapeutic and safety implications of these findings for clinical practice.
Supplementary material
Footnotes
Funding: RKL’s research is supported by the Camiener Foundation Glaucoma Research Fund, the Barlow Family Glaucoma Fund and the Gutierrez Family Research Fund.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Data availability free text: This retrospective cohort study used deidentified patient data from the TriNetX Analytics Platform (TriNetX, LLC), a global federated health research network that aggregates electronic health records (EHRs) from more than 30 countries.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Data availability statement
Data may be obtained from a third party and are not publicly available.
References
- 1.Tham Y-C, Li X, Wong TY, et al. Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis. Ophthalmology. 2014;121:2081–90. doi: 10.1016/j.ophtha.2014.05.013. [DOI] [PubMed] [Google Scholar]
- 2.Belamkar AV, Harris A, Wirotsko B, et al. Medical and surgical treatment management in open angle glaucoma patients of Asian descent: A narrative review. Eur J Ophthalmol. 2025;35:1883–95. doi: 10.1177/11206721251340435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Caprioli J, Varma R. Intraocular pressure: modulation as treatment for glaucoma. Am J Ophthalmol. 2011;152:340–4. doi: 10.1016/j.ajo.2011.05.029. [DOI] [PubMed] [Google Scholar]
- 4.Weinreb RN, Aung T, Medeiros FA. The pathophysiology and treatment of glaucoma: a review. JAMA. 2014;311:1901–11. doi: 10.1001/jama.2014.3192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Tielsch JM, Katz J, Sommer A. Hypertension, Perfusion Pressure, and Primary Open-angle Glaucoma. Arch Ophthalmol . 1995;113:216. doi: 10.1001/archopht.1995.01100020100038. [DOI] [PubMed] [Google Scholar]
- 6.Zhao D, Cho J, Kim MH, et al. Diabetes, fasting glucose, and the risk of glaucoma: a meta-analysis. Ophthalmology. 2015;122:72–8. doi: 10.1016/j.ophtha.2014.07.051. [DOI] [PubMed] [Google Scholar]
- 7.Amaral DC, Guedes J, Cruz MRB, et al. GLP-1 Receptor Agonists Use and Incidence of Glaucoma: A Systematic Review and Meta-Analysis. Am J Ophthalmol. 2025;271:488–97. doi: 10.1016/j.ajo.2024.12.024. [DOI] [PubMed] [Google Scholar]
- 8.Elhusseiny AM, Eleiwa TK, Darwich R, et al. Association Between Chronic Oral Nitrate Use and the Risk of Ocular Hypertension and Open-Angle Glaucoma. Am J Ophthalmol. 2025;279:118–26. doi: 10.1016/j.ajo.2025.07.014. [DOI] [PubMed] [Google Scholar]
- 9.Elhusseiny AM, Eleiwa TK, Dihan QA, et al. Racial and Ethnic Differences in the Association Between Statin Use and the Risk of Ocular Hypertension and Open-Angle Glaucoma. Am J Ophthalmol. 2025;276:117–25. doi: 10.1016/j.ajo.2025.04.003. [DOI] [PubMed] [Google Scholar]
- 10.Wang T, Cao L, Jiang Q, et al. Topical Medication Therapy for Glaucoma and Ocular Hypertension. Front Pharmacol. 2021;12:749858. doi: 10.3389/fphar.2021.749858. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Charlson ME, de Moraes CG, Link A, et al. Nocturnal systemic hypotension increases the risk of glaucoma progression. Ophthalmology. 2014;121:2004–12. doi: 10.1016/j.ophtha.2014.04.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Leske MC, Wu S-Y, Hennis A, et al. Risk factors for incident open-angle glaucoma: the Barbados Eye Studies. Ophthalmology. 2008;115:85–93. doi: 10.1016/j.ophtha.2007.03.017. [DOI] [PubMed] [Google Scholar]
- 13.Leung G, Grant A, Garas AN, et al. A Systematic Review and Meta-analysis of Systemic Antihypertensive Medications With Intraocular Pressure and Glaucoma. Am J Ophthalmol. 2023;255:7–17. doi: 10.1016/j.ajo.2023.03.014. [DOI] [PubMed] [Google Scholar]
- 14.Lee JS, Cha HR, Bae HW, et al. Effect of antihypertensive medications on the risk of open-angle glaucoma. Sci Rep. 2023;13:16224. doi: 10.1038/s41598-023-43420-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Burke J, Schwartz M. Preclinical evaluation of brimonidine. Surv Ophthalmol. 1996;41 Suppl 1:S9–18. doi: 10.1016/s0039-6257(96)82027-3. [DOI] [PubMed] [Google Scholar]
- 16.Gao H, Qiao X, Cantor LB, et al. Up-regulation of brain-derived neurotrophic factor expression by brimonidine in rat retinal ganglion cells. Arch Ophthalmol . 2002;120:797–803. doi: 10.1001/archopht.120.6.797. [DOI] [PubMed] [Google Scholar]
- 17.Woldemussie E, Wijono M, Pow D. Localization of alpha 2 receptors in ocular tissues. Vis Neurosci. 2007;24:745–56. doi: 10.1017/S0952523807070605. [DOI] [PubMed] [Google Scholar]
- 18.Arthur S, Cantor LB. Update on the role of alpha-agonists in glaucoma management. Exp Eye Res. 2011;93:271–83. doi: 10.1016/j.exer.2011.04.002. [DOI] [PubMed] [Google Scholar]
- 19.Wheeler LA, Woldemussie E. Alpha-2 adrenergic receptor agonists are neuroprotective in experimental models of glaucoma. Eur J Ophthalmol. 2001;11 Suppl 2:S30–5. doi: 10.1177/112067210101102s03. [DOI] [PubMed] [Google Scholar]
- 20.MACMAHON B, WEISS NS, POCOCK SJ, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Strobe Initiative. 2007;18 doi: 10.1097/EDE.0b013e3181577654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Vergroesen JE, Schuster AK, Stuart KV, et al. Association of Systemic Medication Use with Glaucoma and Intraocular Pressure: The European Eye Epidemiology Consortium. Ophthalmology. 2023;130:893–906. doi: 10.1016/j.ophtha.2023.05.001. [DOI] [PubMed] [Google Scholar]
- 22.Tanna AP, Rademaker AW, Stewart WC, et al. Meta-analysis of the efficacy and safety of alpha2-adrenergic agonists, beta-adrenergic antagonists, and topical carbonic anhydrase inhibitors with prostaglandin analogs. Arch Ophthalmol . 2010;128:825–33. doi: 10.1001/archophthalmol.2010.131. [DOI] [PubMed] [Google Scholar]
- 23.Bylund DB, Chacko DM. Characterization of alpha2 adrenergic receptor subtypes in human ocular tissue homogenates. Invest Ophthalmol Vis Sci. 1999;40:2299–306. [PubMed] [Google Scholar]
- 24.Krupin T, Liebmann JM, Greenfield DS, et al. A randomized trial of brimonidine versus timolol in preserving visual function: results from the Low-Pressure Glaucoma Treatment Study. Am J Ophthalmol. 2011;151:671–81. doi: 10.1016/j.ajo.2010.09.026. [DOI] [PubMed] [Google Scholar]
- 25.Nagasubramanian S, Hitchings RA, Demailly P, et al. Comparison of apraclonidine and timolol in chronic open-angle glaucoma. A three-month study. Ophthalmology. 1993;100:1318–23. doi: 10.1016/s0161-6420(13)31818-1. [DOI] [PubMed] [Google Scholar]
- 26.Nguyen V, Tiemann D, Park E, et al. Alpha-2 Agonists. Anesthesiol Clin. 2017;35:233–45. doi: 10.1016/j.anclin.2017.01.009. [DOI] [PubMed] [Google Scholar]
- 27.Rajan S, Krishnankutty SV, Nair HM. Efficacy of alpha2 agonists in obtunding rise in intraocular pressure after succinylcholine and that following laryngoscopy and intubation. Anesth Essays Res. 2015;9:219–24. doi: 10.4103/0259-1162.156312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.McAuliffe-Curtin D, Buckley C. Review of alpha adrenoceptor function in the eye. Eye (Lond) 1989;3 (Pt 4):472–6. doi: 10.1038/eye.1989.71. [DOI] [PubMed] [Google Scholar]
- 29.Binggeli T, Schoetzau A, Konieczka K. In glaucoma patients, low blood pressure is accompanied by vascular dysregulation. EPMA J. 2018;9:387–91. doi: 10.1007/s13167-018-0155-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Grieshaber MC, Mozaffarieh M, Flammer J. What is the link between vascular dysregulation and glaucoma? Surv Ophthalmol. 2007;52 Suppl 2:S144–54. doi: 10.1016/j.survophthal.2007.08.010. [DOI] [PubMed] [Google Scholar]
- 31.Laspas P, Hartmann A, Scholz I, et al. Change in Systemic Medication and its Influence on Intraocular Pressure - Results From the Gutenberg Health Study. Invest Ophthalmol Vis Sci. 2024;65:4. doi: 10.1167/iovs.65.14.4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wu A, Khawaja AP, Pasquale LR, et al. A review of systemic medications that may modulate the risk of glaucoma. Eye (Lond) 2020;34:12–28. doi: 10.1038/s41433-019-0603-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Schuster A-G, Vergroesen J, Stuart KV, et al. Association of commonly used systemic medications with glaucoma prevalence and intraocular pressure across Europe: the E3 Consortium. Invest Ophthalmol Vis Sci. 2022;63:2267–67. [Google Scholar]
- 34.van Beek LM, de Keizer RJ, Polak BC, et al. Incidence of ocular side effects of topical beta blockers in the Netherlands. Br J Ophthalmol. 2000;84:856–9. doi: 10.1136/bjo.84.8.856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Zanden JAV, Valuck RJ, Bunch CL, et al. Systemic Adverse Effects of Ophthalmic β-Blockers. Ann Pharmacother. 2001;35:1633–7. doi: 10.1345/aph.18464. [DOI] [PubMed] [Google Scholar]
- 36.Harris A, Jonescu-Cuypers CP. The impact of glaucoma medication on parameters of ocular perfusion. Curr Opin Ophthalmol. 2001;12:131–7. doi: 10.1097/00055735-200104000-00009. [DOI] [PubMed] [Google Scholar]
- 37.Lowry EA, Chansangpetch S, Lin SC. Use of Topical Intraocular Pressure-lowering Medications in the US Population: Results From the NHANES Study 1999 to 2014. J Glaucoma. 2019;28:772–6. doi: 10.1097/IJG.0000000000001315. [DOI] [PubMed] [Google Scholar]
- 38.McMonnies CW. Glaucoma history and risk factors. J Optom. 2017;10:71–8. doi: 10.1016/j.optom.2016.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data may be obtained from a third party and are not publicly available.