This cohort study investigates the association of semaglutide and tirzepatide with nonarteritic anterior ischemic optic neuropathy and other optic nerve and visual pathway disorders in patients with type 2 diabetes.
Key Points
Question
Is treatment with either semaglutide or tirzepatide associated with an increased risk of nonarteritic anterior ischemic optic neuropathy and other optic nerve and visual pathway disorders?
Findings
This cohort study emulated target trials among 1 511 637 eligible patients with type 2 diabetes by comparing treatment with either semaglutide or tirzepatide vs other antidiabetic medications. Semaglutide or tirzepatide was associated with increased risk of nonarteritic anterior ischemic optic neuropathy and other optic nerve disorders, but the overall risk was low.
Meaning
This analysis of patient electronic health record data suggests an increased risk of optic nerve disorders associated with semaglutide or tirzepatide in patients with type 2 diabetes, highlighting the need for close monitoring of these conditions.
Abstract
Importance
Semaglutide and tirzepatide, the new generation of glucagon-like peptide-1 receptor agonist used to treat type 2 diabetes and obesity, were recently reported to be associated with increased risk of nonarteritic anterior ischemic optic neuropathy (NAION). However, reported associations were inconsistent, and it remains unknown if semaglutide or tirzepatide is associated with other disorders of the optic nerve and visual pathways.
Objective
To examine associations of treatment with either semaglutide or tirzepatide with first-time diagnoses of optic nerve and visual pathway disorders in patients with type 2 diabetes.
Design, Setting, and Participants
This cohort study emulated target trials based on a nationwide, population-based database of US patient electronic health records from December 2017 to January 2023. Models were adjusted by propensity score matching. Eligible patients had type 2 diabetes and no prior diagnosis of eye disorders and were prescribed semaglutide, tirzepatide, or other antidiabetic medications.
Exposures
Prescription of semaglutide or tirzepatide vs other antidiabetic medications (insulins, metformin, dipeptidyl-peptidase-4 inhibitors, sodium-glucose cotransporter-2 inhibitors, sulfonylureas, thiazolidinediones, and other glucagon-like peptide 1 receptor agonists).
Main Outcomes and Measures
First-time diagnoses of NAION and other optic nerve disorders during a 2-year follow-up were examined using Cox proportional hazards and Kaplan-Meier survival analyses.
Results
After propensity score matching among 1 511 637 eligible patients, the study population comprised 159 398 matched patients with type 2 diabetes (mean [SD] age, 56.5 [13.3] years; 83 123 women [52.15%]), including 79 699 patients in the semaglutide or tirzepatide group and 79 699 patients in the comparison group. During the 2-year follow-up, there were 35 patients (0.04%) with NAION in the semaglutide or tirzepatide group and 19 patients (0.02%) with NAION in the matched comparison group (hazard ratio, 1.76 [95% CI, 1.01-3.07]). There were 93 patients (0.12%) with other optic nerve disorders in the semaglutide or tirzepatide group and 54 patients (0.07%) with other optic nerve disorders in the matched comparison group (hazard ratio, 1.65 [95% CI, 1.18-2.31]). No association was found with other disorders of the optic nerve or visual pathways.
Conclusions and Relevance
In this study of patients with type 2 diabetes who had no prior eye disorders, patients prescribed semaglutide or tirzepatide had an increased risk of NAION and other optic nerve disorders, although the overall risk was low. These findings highlight the need for close monitoring of these conditions.
Introduction
Glucagon-like peptide-1 receptor agonist (GLP-1RA) medications have transformed the treatment of type 2 diabetes and obesity, with associated reductions in cardiovascular and nephrological complications.1,2 Semaglutide and tirzepatide are second-generation GLP-1RA medications approved by the Food and Drug Administration for the treatment of diabetes and obesity.3,4,5,6 The main associated adverse outcomes are gastrointestinal outcomes, including nausea, vomiting, diarrhea, gastroparesis, and constipation. Studies also reported potential associations of GLP-1RAs with thyroid C-cell tumors and pancreatitis.7,8
Nonarteritic anterior ischemic optic neuropathy (NAION) is characterized by loss of vision caused by loss of blood flow to the optic nerve.9 Its prevalence is higher in patients with diabetes, obesity, hypertension, hyperlipidemia, cerebrovascular disease, and obstructive sleep apnea.10 Although the etiopathogenesis of NAION is not fully understood, decreased arterial perfusion of the optic nerve and the associated swelling and compression of the optic nerve are implicated.9 Several retrospective studies from 2024 and 2025 reported a potential association of semaglutide with NAION in patients with diabetes and patients with obesity.11,12,13,14,15 However, a meta-analysis of randomized clinical trials16 did not detect an association of GLP1-RA therapy, including semaglutide, with NAION. A rare condition, NAION has an incidence rate of 2.5 to 11.8 incidents per 100 000 individuals.9 These inconsistent findings may be due to limited sample sizes and different study designs. Importantly, it remains unknown if semaglutide or tirzepatide is associated with other optic nerve and visual pathway disorders. This study leveraged a nationwide, multicenter database of electronic health records (EHRs) of more than 118 million US patients to conduct rigorous target trial emulation in patients with type 2 diabetes to examine associations of semaglutide or tirzepatide with optic nerve and visual pathway disorders, including NAION.
Methods
This retrospective cohort study was exempt from ethical review and informed consent because it was a secondary analysis of existing data, did not involve intervention or interaction with human participants, and used deidentified data per the deidentification standard defined in section §164.514(a) of the HIPAA Privacy Rule. The process by which the data were deidentified is attested to through a formal determination by a qualified expert as defined in section §164.514(b)(1) of the HIPAA Privacy Rule. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Specification of Target Trials
Study Overview
We compared semaglutide or tirzepatide with other antidiabetic medications for the associated risk of optic nerve and visual pathway disorders in patients with type 2 diabetes who had no prior diagnosis of any eye disorder (determined by the absence of International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] codes H00-H59, “Diseases of the eye and adnexa,” documented in EHRs) using a target trial emulation framework.17,18 Other antidiabetic medications included insulin, metformin, dipeptidyl-peptidase-4 inhibitors, sodium-glucose cotransporter-2 inhibitors, sulfonylureas, thiazolidinediones, and other GLP-1RAs (albiglutide, dulaglutide, exenatide, liraglutide, and lixisenatide). Key protocol components are listed in eTable 1 in Supplement 1.
Semaglutide and tirzepatide are newer and more potent GLP-RAs to treat type 2 diabetes and obesity.19 To examine differences between semaglutide or tirzepatide and first-generation GLP-1RAs in their associated risk for NAION and other optic nerve and visual pathway disorders, we performed a head-to-head comparison of semaglutide or tirzepatide with other GLP-1RAs and a head-to-head comparison of semaglutide with tirzepatide.
Eligibility Criteria
Eligibility criteria for all target trials included a prior diagnosis of type 2 diabetes, a recent medical encounter for type 2 diabetes diagnosis in the past year (active type 2 diabetes), prescriptions of antidiabetic medications (semaglutide, tirzepatide, or other antidiabetic medications) between December 2017 and January 2023, and diagnosis with at least 1 condition among the prescription guidelines for semaglutide and tirzepatide (eg, obesity, hypertension, hypercholesterolemia, heart disease, stroke, kidney disease, or an A1C level ≥8.5%).20,21 Exclusion criteria included a prior diagnosis of any eye disorder, including optic nerve and visual pathway disorders; initiation of semaglutide or tirzepatide and comparison antidiabetic medications at the same time; and certain medical conditions (ie, pancreatitis, type 1 diabetes, thyroid cancer, and gastroparesis) based on contraindications, warnings, and limited use information for semaglutide and tirzepatide (eTable 2 in Supplement 1).20,21
Treatment Strategies
In each target trial, treatment strategies were the initiation of semaglutide or tirzepatide use at baseline (time zero, index event, or time of randomization) or the initiation of comparison antidiabetic medication use at baseline but not both strategies. The treatment strategy was assigned at baseline, regardless of medication use adherence, medication switch, or add-on.
Study Outcomes
Main outcomes were the first-time diagnosis of overall and subcategories of disorders of optic nerve and visual pathways: (1) disorders of optic nerve and visual pathways (ICD-10 code H46-H47), (2) optic neuritis (H46), (3) other disorders of optic nerve and visual pathways (H47), (4) disorders of optic nerve (H47.0), (5) NAION (H47.01), (6) other optic nerve disorders (H47.02, H47.03, H47.09), (7) papilledema (H47.1), (8) optic atrophy (H47.2), and (9) other disorders of optic disc (H47.3). Each outcome was analyzed separately, with no multiple comparisons or competing outcomes. For each outcome, eligible patients were followed up starting after the index event until the occurrence of the outcome, death, loss to follow-up, or 2 years after the index event, whichever occurred first. Details of diagnosis codes for study outcomes are in eTable 3 in Supplement 1.
Statistical Analysis
The estimated outcomes of interest represent the association of treatment strategy assignments with outcomes in the intention-to-treat population. Cumulative incidences were estimated using Kaplan-Meier survival analysis. Cox proportional hazard analyses were used to compare rates of time to events daily during follow-up after the index event. Censoring was applied in Kaplan-Meier survival and Cox proportional hazard analyses. Hazard ratios (HRs) and 95% CIs were calculated; HRs were considered statistically significant at the P < .05 level if the 95% CIs did not include 1.
We used TriNetX, a global federated health research network providing access to statistics on EHRs (diagnoses, procedures, medications, laboratory values, and genomic information) from approximately 118 million patients in 69 large health care organizations covering diverse geographic regions, age, race and ethnicity groups, income and insurance groups, and clinical settings.22
We explicitly emulated target trials described previously using data and built-in analytic functions on the TriNetX Analytics platform. We previously performed emulation target trials and cohort studies using the TriNetX platform to examine the association of GLP-1RAs, including semaglutide, with Alzheimer disease,23 substance use disorders,24,25,26,27 suicidal ideation,28 and cancers.29,30,31,32
Available data elements of EHRs included extensive information on demographics, diagnoses, medications, procedures, laboratory tests, visits, and socioeconomic and lifestyle information. All covariates were binary, categorical, or continuous but were essentially guaranteed to exist (more details on TriNetX are in the eAppendix in Supplement 1). Sex, race, and ethnicity data were self-reported from contributing health care systems. Race and ethnicity data were mapped according to Office of Management and Budget standards into race (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, White, and unknown race) and ethnicity (Hispanic or Latinx, not Hispanic or Latinx, or unknown ethnicity). Missing sex values are represented using unknown sex. Missing data for race and ethnicity are presented as unknown race or unknown ethnicity. Race and ethnicity as social constructs were assessed due to their association with health disparities and the differential impact of structural disadvantages.
Each component of the target trial was emulated using EHRs from the TriNetX Analytics platform (more details of target trial emulation components are in eTables 1-4 and eFigure 1 in Supplement 1). Patients were classified into drug treatment groups (semaglutide or tirzepatide vs other antidiabetic medications, including insulin, metformin, dipeptidyl-peptidase-4 inhibitors, sodium-glucose cotransporter-2 inhibitors, sulfonylureas, thiazolidinediones, and other GLP-1RAs) based on the first prescription in the study period (December 2017 to January 2023), which was the baseline, time zero, or index event. Eligibility criteria and more than 50 baseline covariates were evaluated at baseline based on their status on or before time zero. Baseline covariates included demographics, adverse determinants of socioeconomic health, lifestyle factors, medical conditions and medications that are risk factors associated with NAION and optic nerve and visual pathway disorders,33,34 prior prescriptions of antidiabetic medications, and medical encounters. For each comparison, exposure and comparison groups were propensity score matched for covariates at the baseline to emulate randomization. We used the TriNetX Analytics built-in 1:1 propensity score–matching function, which uses nearest neighbor greedy matching with a caliper of 0.1 times the SD for baseline covariates.17 This means that for 2 patients to match, the difference between their propensity scores must be no greater than a tolerance level based on a predefined caliper of 0.1 times the pooled SDs of the propensity scores. For each variable, standardized mean difference was calculated. A threshold of a standardized mean difference less than 0.1 was used to indicate balanced variables.
Sensitivity Analyses
Several negative control outcomes were used for quantifying potential residual bias. They were first diagnosis of congenital eye diseases (Q10-Q15), first-time diagnosis of any eye diseases (H00-H59), outpatient medical encounters, and overall medical encounters. Congenital eye diseases can be caused by genetic mutations, exposure to drugs or alcohol during pregnancy, or maternal infections, and our study populations comprised mostly adult patients with type 2 diabetes (mean age, 56.5 years); therefore, we assumed that semaglutide or tirzepatide was not associated with an altered risk of congenital eye diseases. However, whether semaglutide or tirzepatide was associated with an altered risk of overall eye diseases remained unknown. Outpatient and overall medical encounters were used to measure overall health care use. As part of sensitivity analyses, we also calculated the E-value, which measured how susceptible the study results were to unmeasured or uncontrolled confounding.18 The E-value for the observed outcome and CIs using HR and 95% CI was calculated with the following equation:
| E-value (outcome) = HR + √(HR × [HR − 1]) |
| E-value (lower CI, or LL) = LL + √(LL × [LL − 1]) |
| E-value (upper CI, or UL) = UL + √(UL × (UL-1)). |
Analytics Platform and Statistical Software
Data were collected and analyzed in February 2025 within the TriNetX Analytics platform. All statistical analyses, including propensity score matching, Kaplan-Meier survival analysis, and Cox proportional hazards analyses, were done using built-in functions within the TriNetX Analytics platform that are implemented using the R Survival package35 version 3.2-3in R statistical software version 4.0.2 (R Project for Statistical Computing) and libraries and utilities for in Python programming language version 3.7 (Python Software Foundation) and Java programming language version 11.0.16 (Oracle Corporation).
Results
Study Populations
Figure 1 is the flowchart showing cohort construction. The study included 1 511 637 eligible patients with type 2 diabetes without prior diagnosis of any eye disorder who were prescribed antidiabetic medications, including 82 972 patients prescribed semaglutide or tirzepatide (63 926 patients [77.05%] prescribed semaglutide, 3815 patients [4.60%] prescribed tirzepatide, and 15 231 patients [18.36%] prescribed both medications) and 1 428 665 patients prescribed other antidiabetic medications. Before propensity score matching, semaglutide or tirzepatide and comparison groups differed by demographics, comorbidities, and medications. After propensity score matching, the study population comprised 159 398 matched patients with type 2 diabetes (mean [SD] age, 56.5 [13.3] years; 83 123 women [52.15%]; 11 487 Hispanic [7.21%]; 7157 Asian [4.49%], 25 121 Black [15.76%], and 102 381 White [64.23%]), including 79 699 patients in the semaglutide or tirzepatide group and 79 699 patients in the comparison group. Matched groups were balanced (Table; eFigures 2-3 in Supplement 1) and had high comorbidities, including 92 232 patients with obesity (57.80%) and 121 317 patients with hypertension (76.11%).
Figure 1. Study Flowchart.
DPP-4i indicates dipeptidyl-peptidase-4 inhibitor; GLP-1RA, glucagon-like peptide-1 receptor agonist; SGLT2i, sodium-glucose cotransporter-2 inhibitor; SU, sulfonylurea; TZD, thiazolidinedione.
aThe combined total of 1 428 665 patients in the comparison arm is not a sum of patients from each antidiabetic medication group because a patient could be prescribed more than 1 antidiabetic medication during the study period. However, there was no overlap between the semaglutide or tirzepatide group and comparison medication groups.
bOther GLP-1RAs included albiglutide (1108 patients [0.89%]), dulaglutide (83 501 patients [66.85%]), exenatide (17 046 patients [13.65%]), liraglutide (53 189 patients [42.59%]), and lixisenatide (2465 patients [1.97%]).
Table. Patient Characteristics.
| Characteristic | Patients, No. (%)a | |||||
|---|---|---|---|---|---|---|
| Before propensity score matching (N = 1 511 637) | After propensity score matching (N = 159 398) | |||||
| Semaglutide or tirzepatide (n = 82 972) | Other antidiabetic medications (n = 1 428 665) | SMD | Semaglutide or tirzepatide (n = 79 699) | Other antidiabetic medications (n = 79 699) | SMD | |
| Age at index event, mean (SD), y | 56.6 (12.4) | 63.3 (14.0) | 0.51b | 56.8 (12.4) | 56.2 (14.1) | 0.04 |
| Sex | ||||||
| Female | 42 994 (51.8) | 636 197 (44.5) | 0.15b | 41 210 (51.7) | 41 913 (52.6) | 0.02 |
| Male | 35 017 (42.2) | 738 209 (51.7) | 0.19b | 33 813 (42.4) | 33 421 (41.9) | 0.01 |
| Unknown | 4958 (6.0) | 54 288 (3.8) | 0.10b | 4676 (5.9) | 4365 (5.5) | 0.02 |
| Ethnicity | ||||||
| Hispanic or Latinx | 5849 (7.0) | 134 959 (9.4) | 0.09 | 5679 (7.1) | 5808 (7.3) | 0.006 |
| Not Hispanic or Latinx | 58 747 (70.8) | 915 044 (64.0) | 0.15b | 56 406 (70.8) | 57 119 (71.7) | 0.02 |
| Unknown | 18 375 (22.1) | 378 691 (26.5) | 0.10b | 17 614 (22.1) | 16 772 (20.8) | 0.03 |
| Race | ||||||
| Asian | 3983 (4.7) | 79 663 (5.6) | 0.04 | 3725 (4.7) | 3432 (4.3) | 0.02 |
| Black | 12 904 (15.6) | 257 078 (18.0) | 0.07 | 12 506 (15.7) | 12 615 (15.8) | 0.004 |
| White | 52 963 (63.8) | 851 259 (59.6) | 0.09 | 50 829 (63.8) | 51 552 (64.7) | 0.02 |
| Unknown | 8563 (10.3) | 158 577 (11.1) | 0.03 | 8189 (10.3) | 7830 (9.8) | 0.02 |
| Adverse socioeconomic determinants of healthc | 2460 (3.0) | 26 741 (1.9) | 0.07 | 2305 (2.9) | 2321 (2.9) | 0.001 |
| Problems related to lifestyled | 5941 (7.2) | 62 518 (4.4) | 0.12b | 5645 (7.1) | 5615 (7.0) | 0.001 |
| Type 2 diabetes complications | ||||||
| Kidney | 12 497 (15.1) | 178 331 (12.5) | 0.08 | 11 839 (14.9) | 11 609 (14.6) | 0.008 |
| Ophthalmic | 2993 (3.6) | 28 221 (2.0) | 0.10b | 2834 (3.6) | 2919 (3.7) | 0.006 |
| Neurological | 13 277 (16.0) | 143 041 (10.0) | 0.18b | 12 597 (15.8) | 12 929 (16.2) | 0.01 |
| Circulatory | 6337 (7.6) | 67 035 (4.7) | 0.12b | 5933 (7.4) | 5839 (7.3) | 0.005 |
| Other specified | 40 942 (49.3) | 350 065 (24.5) | 0.53b | 38 446 (48.2) | 38 877 (48.8) | 0.01 |
| Unspecified | 9368 (11.3) | 82 479 (5.8) | 0.20b | 8830 (11.1) | 9001 (11.3) | 0.007 |
| Preexisting medical conditions | ||||||
| Overweight or obesity | 48 286 (58.2) | 353 004 (24.7) | 0.72b | 45 550 (57.2) | 46 682 (58.6) | 0.03 |
| Metabolic disorder | 65 428 (78.9) | 798 209 (55.9) | 0.51b | 62 445 (78.4) | 63 538 (79.7) | 0.03 |
| Hypertension | 62 888 (75.8) | 868 192 (60.8) | 0.33b | 60 213 (75.6) | 61 104 (76.7) | 0.03 |
| Hypotension | 3444 (4.2) | 68 372 (4.8) | 0.03 | 3336 (4.2) | 3284 (4.1) | 0.003 |
| Disorder of lipoprotein metabolism or other lipidemia | 62 182 (74.9) | 703 727 (49.3) | 0.55b | 59 250 (74.3) | 60 506 (75.9) | 0.04 |
| Ischemic heart disease | 15 120 (18.2) | 296 180 (20.7) | 0.06 | 14 632 (18.4) | 14 346 (18.0) | 0.009 |
| Other form of heart disease | 21 779 (26.2) | 390 040 (27.3) | 0.02 | 20 984 (26.3) | 20 465 (25.7) | 0.02 |
| Cerebrovascular disease | 5583 (6.7) | 135 799 (9.5) | 0.10b | 5418 (6.8) | 5285 (6.6) | 0.007 |
| Kidney disease | 13 680 (16.5) | 277 510 (19.4) | 0.08 | 13 222 (16.6) | 12 919 (16.2) | 0.01 |
| Atherosclerosis | 3491 (4.2) | 63 244 (4.4) | 0.01 | 3366 (4.2) | 3364 (4.2) | <.001 |
| Sleep apnea | 24 168 (29.1) | 170 150 (11.9) | 0.44b | 22 660 (28.4) | 23 114 (29.0) | 0.01 |
| Migraine | 5853 (7.1) | 32 631 (2.3) | 0.23b | 5411 (6.8) | 5443 (6.8) | 0.002 |
| Coagulation defect, purpura, or other hemorrhagic condition | 3934 (4.7) | 70 392 (4.9) | 0.009 | 3790 (4.8) | 3715 (4.7) | 0.004 |
| Alcohol use disorder | 1728 (2.1) | 40 670 (2.8) | 0.05 | 1689 (2.1) | 1657 (2.1) | 0.003 |
| Tobacco use disorder | 10 215 (12.3) | 162 128 (11.3) | 0.03 | 9853 (12.4) | 9819 (12.3) | 0.001 |
| Prior medications | ||||||
| Antihypertensive | 12 413 (15.0) | 172 321 (12.1) | 0.09 | 11 907 (14.9) | 11 812 (14.8) | 0.003 |
| Antihypertensive, other | 17 331 (20.9) | 197 027 (13.8) | 0.19b | 16 482 (20.7) | 16 456 (20.6) | 0.001 |
| Antiarrhythmic | 37 024 (44.6) | 399 862 (28.0) | 0.35b | 35 081 (44.0) | 34 735 (43.6) | 0.009 |
| Amphetamine | 2441 (2.9) | 7785 (0.5) | 0.18b | 2169 (2.7) | 2093 (2.6) | 0.006 |
| Aspirin | 25 052 (30.2) | 316 568 (22.2) | 0.18b | 24 024 (30.1) | 24 164 (30.3) | 0.004 |
| Sildenafil | 5157 (6.2) | 30 323 (2.1) | 0.21b | 4801 (6.0) | 4802 (6.0) | <.001 |
| Tadalafil | 3204 (3.9) | 17 719 (1.2) | 0.17b | 2985 (3.7) | 3021 (3.8) | 0.002 |
| Amiodarone | 1685 (2.0) | 26 057 (1.8) | 0.02 | 1629 (2.0) | 1604 (2.0) | 0.002 |
| Interferon | 77 (0.1) | 1222 (0.1) | 0.002 | 72 (0.1) | 116 (0.1) | 0.02 |
| Insulin | 34 773 (41.9) | 251 055 (17.6) | 0.55b | 32 650 (41.0) | 33 969 (42.6) | 0.03 |
| Metformin | 57 029 (69.0) | 295 891 (20.7) | 1.11b | 54 032 (67.8) | 56 665 (71.1) | 0.07 |
| DPP-4 inhibitor | 15 991 (19.3) | 73 846 (5.2) | 0.44b | 14 865 (18.7) | 15 281 (19.2) | 0.01 |
| SGLT2 inhibitor | 21 453 (25.9) | 24 567 (1.7) | 0.75b | 18 578 (23.3) | 15 947 (20.0) | 0.08 |
| Sulfonylurea | 23 155 (27.9) | 141 814 (9.9) | 0.47b | 21 741 (27.3) | 22 623 (28.4) | 0.03 |
| Other GLP-1RA | 20 960 (25.3) | 28 164 (2.0) | 0.72b | 18 289 (23.3) | 16 112 (20.0) | 0.08 |
| Medical encounters | ||||||
| Outpatient medical visit | 69 369 (83.6) | 918 705 (64.3) | 0.45b | 66 567 (83.5) | 69 213 (86.8) | 0.09 |
| Inpatient medical visit | 22 515 (27.1) | 431 388 (30.2) | 0.07 | 21 785 (27.3) | 21 835 (27.4) | 0.001 |
| Emergency visit | 31 489 (38.0) | 456 293 (31.9) | 0.13b | 30 057 (37.7) | 29 617 (37.2) | 0.01 |
Abbreviations: DPP-4, dipeptidyl peptidase 4; GLP-1RA, glucagon-like peptide 1 agonist; SGLT2, sodium-glucose cotransporter 2; SMD, standardized mean difference.
Shown are groups before and after propensity score matching for the listed variables. The status of variables was based on the presence of related clinical codes any time up to 1 day before the index event (first prescription of semaglutide or insulin during December 2017 to January 2023).
SMD greater than 0.1, a threshold indicating cohort imbalance.
Adverse socioeconomic determinants of health included housing and economic circumstances, upbringing, education, physical environment, and social environment.
Problems with lifestyle included tobacco use, lack of physical exercise, inappropriate diet and eating habits, and others.
Associations of Semaglutide or Tirzepatide With Optic Nerve and Visual Pathway Disorders
During the 2-year follow-up, there were 35 patients (0.04%) with NAION in the semaglutide or tirzepatide group and 19 patients with NAION (0.02%) in the matched comparison group (HR, 1.76 [95% CI, 1.01-3.07]). There were 93 patients (0.12%) with other optic nerve disorders (H47.02, H47.03, and H47.09) in the semaglutide or tirzepatide group and 54 patients (0.07%) with these disorders in the matched comparison group (HR, 1.65 [95% CI, 1.18-2.31]). There was no association of semaglutide or tirzepatide with optic neuritis, papilledema, or optic atrophy. Patients in the semaglutide or tirzepatide group had an increased incidence of other optic disc disorders, but this difference was not statistically significant (Figure 2).
Figure 2. Risk of Optic Nerve and Visual Pathway Disorders for Semaglutide or Tirzepatide vs Other Antidiabetic Medications.
Exposure and comparison groups were propensity score matched for variables listed in Table 1, and the status of variables was based on the presence of related clinical codes any time up to 1 day before the index event (first prescription of semaglutide or tirzepatide vs comparison medication classes during December 2017 to January 2023). Individuals in matched groups were followed up after the index event until the occurrence of the outcome, death, loss to follow-up, or 2 years after the index event, whichever occurred first. Hazard rates were calculated using a Cox proportional hazards model with censoring applied. Overall risk was the number of patients with outcomes during the follow-up time window divided by the number of patients in the cohort at the beginning of the time window. HR indicates hazard ratio; NAION, nonarteritic anterior ischemic optic neuropathy.
For sensitivity analysis, semaglutide or tirzepatide was not associated with any of 4 negative control outcomes, including medical encounters for congenital eye diseases diagnosis, medical encounters for any eye disease diagnosis, outpatient medical encounters, and all medical encounters (Figure 2). The E-value was 2.92 (95% CI, 1.11-5.59) for NAION, greater than the previously reported HRs of 1.65 for hypertension and 2.46 for hypercoagulable states, 2 major risk factors associated with NAION.36 However, reported associations of risk factors with NAION varied among studies.33,37 The E-value was 2.69 (95% CI, 1.64-4.05) for other optic nerve disorders.
The 2-year cumulative incidence curves of optic nerve disorders, NAION, and other optic nerve disorders are shown in Figure 3. For the outcome of optic nerve disorders, the mean (SD) follow-up time was 671.3 (50.9) days for the semaglutide or tirzepatide group and 643.0 (77.3) days for the comparison group. For the outcome of NAION, the mean (SD) follow-up time was 674.8 (49.2) days for the semaglutide or tirzepatide group and 647.5 (74.3) days for the comparison group. For the outcome of other optic nerve disorders, the mean (SD) follow-up time was 671.9 (50.5) days for the semaglutide or tirzepatide group and 643.8 (76.6) days for the comparison group. The slope for the semaglutide or tirzepatide group was consistently higher than that for the comparison group during the 2-year follow-up, indicating a sustained association of semaglutide or tirzepatide with the risk of developing optic nerve disorders.
Figure 3. Cumulative Incidence of Optic Nerve and Visual Pathway Disorders.
The 2-year cumulative incidence curves are shown for A, optic nerve disorders, B, nonarteritic anterior ischemic optic neuropathy (NAION), and C, other optic nerve disorders for the propensity score matched semaglutide or tirzepatide group and other antidiabetic medication groups in patients with type 2 diabetes who had no prior diagnosis of eye disorders.
Head-to-Head Comparisons With Other GLP-1RAs and Semaglutide vs Tirzepatide
To examine potential differential associations of new and older-generation GLP-1RAs with optic nerve and visual pathway disorders, we performed head-to-head comparison of semaglutide or tirzepatide with other GLP-1RAs (albiglutide, dulaglutide, exenatide, liraglutide, and lixisenatide). Before propensity score matching, the semaglutide or tirzepatide cohort and other GLP-1RA cohorts included 82 972 and 124 908 patients, respectively. The cohorts differed in demographics, comorbidities, and medications. After matching, each cohort included 71 459 patients and the 2 cohorts were balanced (eTable 5 in Supplement 1). Compared with other GLP-1RAs, treatment with semaglutide or tirzepatide was not associated with risk of NAION (HR, 1.75 [95% CI, 0.98-3.10]) but was associated with increased risk of other optic nerve disorders (HR, 1.94 [95% CI, 1.32-2.85]) and other optical disc orders (HR, 1.39 [95% CI, 1.00-1.93]) (Figure 4).
Figure 4. Risk of Optic Nerve and Visual Pathway Disorders for Semaglutide or Tirzepatide vs Other Glucagon-Like Peptide 1 Agonists (GLP-1RAs).
Exposure and comparison groups were propensity score matched. HR indicates hazard ratio; NAION, nonarteritic anterior ischemic optic neuropathy.
Semaglutide was then directly compared with tirzepatide. After matching, there were 3806 patients in each cohort. There were 6 patients with optic nerve disorders in the semaglutide cohort, including 2 patients with NAION, and 4 patients with optic nerve disorders in the tirzepatide cohort, including 1 patient with NAION.
Discussion
In a population of patients with type 2 diabetes who had no prior diagnosis of eye diseases, this cohort study found that semaglutide or tirzepatide compared with other antidiabetic medications was associated with a differential risk of optic nerve and visual pathways, including increased risk of NAION and other optic nerve disorders, but not optic neuritis, papilledema, optic atrophy, or optic disc orders. Cumulative incidence curves began to diverge immediately after medication initiation and continued to separate thereafter. The slope of the cumulative incidence curve for the semaglutide or tirzepatide group was consistently larger than that for the comparison group, indicating the potential sustained association of semaglutide or tirzepatide with risk of NAION and other optic nerve disorders.
Our finding of increased risk of NAION is consistent with findings from cohort studies from 2024 and 2025. Hathaway et al11 showed in a single-center 2024 cohort study comparing semaglutide with non-GLP-1RA antidiabetic medications that semaglutide was associated with increased risk of NAION in 403 matched patients with type 2 diabetes (HR, 4.28 [95% CI, 1.62-11.29]) and 613 matched patients with obesity (HR, 7.64 [95% CI, 2.21-26.36]), with a temporal association observed within the first year. In a cohort study13 that was conducted in 2024 using TriNetX network, semaglutide was not associated with an increased risk of NAION in patients with type 2 diabetes compared with non-GLP-1RA antidiabetic medications but had an HR of 2.32 (95% CI, 0.60-8.97). In a 2025 multicenter cohort study,15 semaglutide was not associated with an increased NAION risk among individuals with type 2 diabetes compared with non-GLP-1RA antidiabetics but had an HR of 1.44 (95% CI, 0.78-2.68). In a 2024 Danish cohort study38 of 424 152 patients with type 2 diabetes, semaglutide was associated with an increased risk of NAION (HR, 2.19 [95% CI, 1.54-3.12]). We replicated the previous finding of an increased risk of NAION in our large cohort of 159 398 propensity score matched patients with type 2 diabetes (79 699 patients in each matched cohort). The increased risk (HR, 1.76 [95% CI, 1.01-3.07]) was significant, although the HR was smaller than in most previously reported associations. Results were consistent regardless of comparison groups (GLP-1RAs and non-GLP-1RA antidiabetic medications). In addition, we showed that the associations were sustained over the entire 2-year follow-up period. However, the CI was still wide, and future studies with larger sample sizes are needed.
In our study, we found that patients with type 2 diabetes who were prescribed semaglutide or tirzepatide had various levels of risk of optic nerve and visual pathway disorders, with increased risk of NAION and other optic nerve disorders and no associations with optic neuritis, papilledema, optic atrophy, or optic disc disorders. Each optic nerve and visual pathway disorder has distinct causes. NAION is due to the lack of blood flow to the optic nerve caused by hypertension, hypercholesterolemia, diabetes, cardiovascular and cerebrovascular disease, and obstructive sleep apnea.34,39 Optic neuritis is an inflammatory demyelinating disorder of the optic nerve due to multiple sclerosis, autoimmune disorders, or infections.40 Optic atrophy is the degeneration of the optic nerve due to ischemia, trauma, genetic conditions, glaucoma, optic neuritis, or toxic effects.41 Papilledema is optic disc swelling due to high intracranial pressure caused by intracerebral mass lesions, cerebral hemorrhage, head trauma, hydrocephalus, and idiopathic intracranial hypertension.42 Given that each optic nerve and visual pathway disorder has distinct causes, the observed differential risks associated with semaglutide and tirzepatide are expected, although future work is needed to understand the underlying mechanisms.
GLP-1 receptors are expressed in the optic nerve43 and as such, could be directly affected by GLP1-R agonist drugs. We performed a head-to-head comparison of semaglutide or tirzepatide with other GLP-1RAs and found that semaglutide or tirzepatide was associated with increased risk of NAION and other optic nerve disorders. For the direct comparison of semaglutide with tirzepatide, the sample size was limited and the number of patients with outcomes in both cohorts was small, preventing statistical significance testing. These results suggest that underlying mechanisms may be specific to these 2 medications not the GLP-1RA class, which may include their indirect associations with abrupt changes in metabolic parameters. While subgroup analysis based on risk factors, such as hypertension, could offer relevant insights, we could not stratify patients due to imbalanced distribution of these risk factors in the study population. For example, 76% of individuals in the study population had hypertension so the sample size of patients without hypertension was limited. Future research into mechanisms driving the increased risk will inform clinical strategies to mitigate it. Additional work to examine potential modifications by race and ethnicity and gender may be important to identify populations at increased risk. In the future, with routine clinical access to genetic analyses, it may be possible to identify individuals at increased risk for these potential complications.44
Limitations
Our study has several limitations. First, retrospective observational studies using patient EHRs can have inherent limitations, including unmeasured or uncontrolled confounding and biases. We conducted rigorous target trial emulation, performed sensitivity analyses using negative control outcomes, quantified E-values to assess the potential impact of unmeasured confounding, and validated previous findings of the association of semaglutide or tirzepatide with increased risk of NAION.11,15 However, inherent limitations of retrospective EHR data analyses could not be eliminated and this was an observational study, so causal conclusions cannot be drawn from our findings. Second, our patient cohort was sourced from the TriNetX network. Validation of results in other EHR databases and analytics platforms is warranted. Third, this study focused on patients with type 2 diabetes for a 2-year follow-up. Future studies should explore longer follow-up periods and different populations. Fourth, ICD-10–based diagnoses for NAION and other diseases could have limitations due to overdiagnosis, misdiagnosis, and underdiagnosis. Because there is no specific ICD-10 code for NAION, we used ICD-10 code H47.01, “ischemic optic neuropathy” for NAION, which could have overdiagnosed NAION.11 Hathaway et al11 used this code plus manual medical record review for NAION confirmation. Cai et al15 used the clinical code for ischemic optic neuropathy for sensitive definition and a second confirmatory ischemic optic neuropathy diagnosis code within 90 days for a specific definition and showed that findings were largely consistent across the 2 NAION definitions. In our study, we had no access to patient-level data and could not perform manual review to confirm NAION diagnoses. We used the TriNetX analytics platform and its built-in statistical functions, which do not allow users to specify 2 outcomes within certain days. Nonetheless, our findings were consistent with previously reported findings.11,15 Fifth, outcomes in this study were a mixture of general and specific disease terms. This study had 2 goals: to validate previous findings about an association of semaglutide with NAION in a larger sample size and a target trial emulation framework and to examine the association of semaglutide and tirzepatide with other, related optic nerve and visual pathway disorders, which have not been studied, to our knowledge. We used ICD-10 code ontology, which is a tree-structured organization of disease terms and their hierarchical relationships. In a hypothesis-free manner, we started with more general term, disorders of optic nerve and visual pathways (ICD-10 code H46-H47, and then focused on more specific codes along the tree. We used this strategy to systematically identify where semaglutide or tirzepatide may be associated with optic nerve and visual pathways. We found that these drugs preferably had associations with optic nerves but not other visual pathways. Findings for this study could set a foundation for future hypothesis-driven approaches to focus on specific eye disorders. Sixth, EHRs in TriNetX lack data on medication adherence. Practice pattern variations among health care organizations, and patient health care use could not explicitly be controlled. However, sensitivity analyses indicated similar health care use levels between exposure and comparison groups, minimizing potential surveillance bias.
Conclusions
This cohort study found that treatment with semaglutide or tirzepatide was associated with increased risk of optic nerve disorders, including NAION, but no other optic nerve or visual pathway disorder in patients with diabetes. However, the overall risk remained very low. Future studies are needed to replicate these findings, explore underlying mechanisms, identify individuals at increased risk for these potential complications, and examine other eye disorders.
eAppendix. TriNetX analytics platform
eTable 1. Specification and emulation of pragmatic target trials
eTable 2. Eligibility criteria and exposure definitions
eTable 3. Outcome definitions
eTable 4. Definitions of covariates
eTable 5. Characteristics of semaglutide group and comparison group before and after propensity score matching
eFigure 1. Graphical illustration of study design
eFigure 2. Propensity score matching and density function before and after matching semaglutide or tirzepatide and other antidiabetic medication groups on listed variables in eTable 4
eFigure 3. Propensity score matching and density function before and after matching semaglutide or tirzepatide and other GLP-1RA groups on listed variables in eTable 4
Data Sharing Statement
References
- 1.Hinnen D. Glucagon-like peptide 1 receptor agonists for type 2 diabetes. Diabetes Spectr. 2017;30(3):202-210. doi: 10.2337/ds16-0026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Vilsbøll T, Christensen M, Junker AE, Knop FK, Gluud LL. Effects of glucagon-like peptide-1 receptor agonists on weight loss: systematic review and meta-analyses of randomised controlled trials. BMJ. 2012;344:d7771. doi: 10.1136/bmj.d7771 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Frías JP, Davies MJ, Rosenstock J, et al. ; SURPASS-2 Investigators . Tirzepatide versus semaglutide once weekly in patients with type 2 diabetes. N Engl J Med. 2021;385(6):503-515. doi: 10.1056/NEJMoa2107519 [DOI] [PubMed] [Google Scholar]
- 4.Garvey WT, Frias JP, Jastreboff AM, et al. ; SURMOUNT-2 investigators . Tirzepatide once weekly for the treatment of obesity in people with type 2 diabetes (SURMOUNT-2): a double-blind, randomised, multicentre, placebo-controlled, phase 3 trial. Lancet. 2023;402(10402):613-626. doi: 10.1016/S0140-6736(23)01200-X [DOI] [PubMed] [Google Scholar]
- 5.Wilding JPH, Batterham RL, Calanna S, et al. ; STEP 1 Study Group . Once-weekly semaglutide in adults with overweight or obesity. N Engl J Med. 2021;384(11):989-1002. doi: 10.1056/NEJMoa2032183 [DOI] [PubMed] [Google Scholar]
- 6.Willard FS, Douros JD, Gabe MB, et al. Tirzepatide is an imbalanced and biased dual GIP and GLP-1 receptor agonist. JCI Insight. 2020;5(17):e140532. doi: 10.1172/jci.insight.140532 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Silverii GA, Monami M, Gallo M, et al. Glucagon-like peptide-1 receptor agonists and risk of thyroid cancer: a systematic review and meta-analysis of randomized controlled trials. Diabetes Obes Metab. 2024;26(3):891-900. doi: 10.1111/dom.15382 [DOI] [PubMed] [Google Scholar]
- 8.Smits MM, Van Raalte DH. Safety of semaglutide. Front Endocrinol (Lausanne). 2021;12:645563. doi: 10.3389/fendo.2021.645563 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Salvetat ML, Pellegrini F, Spadea L, Salati C, Zeppieri M. Non-arteritic anterior ischemic optic neuropathy (NA-AION): a comprehensive overview. Vision (Basel). 2023;7(4):72. doi: 10.3390/vision7040072 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Martin-Gutierrez MP, Petzold A, Saihan Z. Correction: NAION or not NAION: a literature review of pathogenesis and differential diagnosis of anterior ischaemic optic neuropathies. Eye (Lond). 2024;38(3):631. doi: 10.1038/s41433-023-02873-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hathaway JT, Shah MP, Hathaway DB, et al. Risk of nonarteritic anterior ischemic optic neuropathy in patients prescribed semaglutide. JAMA Ophthalmol. 2024;142(8):732-739. doi: 10.1001/jamaophthalmol.2024.2296 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mollan SP. Semaglutide and nonarteritic anterior ischemic optic neuropathy. JAMA Ophthalmol. 2024;142(8):740-741. doi: 10.1001/jamaophthalmol.2024.2514 [DOI] [PubMed] [Google Scholar]
- 13.Chou CC, Pan SY, Sheen YJ, et al. Association between semaglutide and nonarteritic anterior ischemic optic neuropathy: a multinational population-based study. Ophthalmology. 2025;132(4):381-388. doi: 10.1016/j.ophtha.2024.10.030 [DOI] [PubMed] [Google Scholar]
- 14.Katz BJ, Lee MS, Lincoff NS, et al. Ophthalmic complications associated with the antidiabetic drugs semaglutide and tirzepatide. JAMA Ophthalmol. 2025;143(3):215-220. doi: 10.1001/jamaophthalmol.2024.6058 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cai CX, Hribar M, Baxter S, et al. Semaglutide and nonarteritic anterior ischemic optic neuropathy. JAMA Ophthalmol. 2025;143(4):304-314. doi: 10.1001/jamaophthalmol.2024.6555 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Silverii GA, Pala L, Cresci B, Mannucci E. Glucagon-like peptide 1 (GLP1) receptor agonists and risk for ischemic optic neuropathy: a meta-analysis of randomised controlled trials. Diabetes Obes Metab. 2025;27(2):1005-1009. doi: 10.1111/dom.16076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.TriNetX Engineering . Function propensity_score_matching. Accessed December 17, 2024. https://lucid.trinetx.com/r/v1.0/reference/propensity_score_matching.html
- 18.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med. 2017;167(4):268-274. doi: 10.7326/M16-2607 [DOI] [PubMed] [Google Scholar]
- 19.Lenharo M. How rival weight-loss drugs fare at treating obesity, diabetes and more. Nature. 2024. Published online September 3, 2024. doi: 10.1038/d41586-024-02716-8 [DOI] [PubMed] [Google Scholar]
- 20.Novo Nordisk . Ozempic prescribing information. Accessed June 27, 2024. https://www.ozempic.com/prescribing-information.html
- 21.Lilly . Mounjaro (tirzepatide) injection and Zepbound (tirzepatide) injection. Accessed March 3, 2025. https://tirzepatide.lilly.com/
- 22.TriNetX . Real-world data. TriNetX. Accessed January 16, 2024. https://trinetx.com/real-world-data/
- 23.Wang W, Wang Q, Qi X, et al. Associations of semaglutide with first-time diagnosis of Alzheimer’s disease in patients with type 2 diabetes: target trial emulation using nationwide real-world data in the US. Alzheimers Dement. 2024;20(12):8661-8672. doi: 10.1002/alz.14313 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Association of semaglutide with tobacco use disorder in patients with type 2 diabetes. target trial emulation using real-world data. Ann Intern Med. 2024;177(8):1016-1027. doi: 10.7326/M23-2718 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wang W, Volkow ND, Wang Q, et al. Semaglutide and opioid overdose risk in patients with type 2 diabetes and opioid use disorder. JAMA Netw Open. 2024;7(9):e2435247. doi: 10.1001/jamanetworkopen.2024.35247 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Associations of semaglutide with incidence and recurrence of alcohol use disorder in real-world population. Nat Commun. 2024;15(1):4548. doi: 10.1038/s41467-024-48780-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Association of semaglutide with reduced incidence and relapse of cannabis use disorder in real-world populations: a retrospective cohort study. Mol Psychiatry. 2024;29(8):2587-2598. doi: 10.1038/s41380-024-02498-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Association of semaglutide with risk of suicidal ideation in a real-world cohort. Nat Med. 2024;30(1):168-176. doi: 10.1038/s41591-023-02672-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wang L, Wang Q, Li L, Kaelber DC, Xu R. Glucagon-like peptide-1 receptor agonists and pancreatic cancer risk: target trial emulation using real-world data. J Natl Cancer Inst. 2025;117(3):476-485. doi: 10.1093/jnci/djae260 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wang L, Wang W, Kaelber DC, Xu R, Berger NA. GLP-1 receptor agonists and colorectal cancer risk in drug-naive patients with type 2 diabetes, with and without overweight/obesity. JAMA Oncol. 2024;10(2):256-258. doi: 10.1001/jamaoncol.2023.5573 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wang L, Berger NA, Kaelber DC, Xu R. Association of GLP-1 receptor agonists and hepatocellular carcinoma incidence and hepatic decompensation in patients with type 2 diabetes. Gastroenterology. 2024;167(4):689-703. doi: 10.1053/j.gastro.2024.04.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wang L, Xu R, Kaelber DC, Berger NA. Glucagon-like peptide 1 receptor agonists and 13 obesity-associated cancers in patients with type 2 diabetes. JAMA Netw Open. 2024;7(7):e2421305. doi: 10.1001/jamanetworkopen.2024.21305 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Liu B, Yu Y, Liu W, Deng T, Xiang D. Risk factors for non-arteritic anterior ischemic optic neuropathy: a large scale meta-analysis. Front Med (Lausanne). 2021;8:618353. doi: 10.3389/fmed.2021.618353 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Raizada K, Margolin E. Non-arteritic anterior ischemic optic neuropathy. In: Abdelsattar M, Abernethy LT, Ackley WB, et al. , eds. StatPearls. StatPearls Publishing; 2025. Accessed July 9, 2025. https://www.ncbi.nlm.nih.gov/books/NBK559045/ [PubMed] [Google Scholar]
- 35.Therneau T, Lumley T. A Package for Survival Analysis in R: R Package: Version 2. R Project for Statistical Computing; 2015. [Google Scholar]
- 36.Cestari DM, Gaier ED, Bouzika P, et al. Demographic, systemic, and ocular factors associated with nonarteritic anterior ischemic optic neuropathy. Ophthalmology. 2016;123(12):2446-2455. doi: 10.1016/j.ophtha.2016.08.017 [DOI] [PubMed] [Google Scholar]
- 37.Tamhankar M, Volpe NJ. Nonarteritic anterior ischemic optic neuropathy: epidemiology, pathogenesis, and etiologies. UpToDate. Accessed July 5, 2024. https://www.uptodate.com/contents/nonarteritic-anterior-ischemic-optic-neuropathy-epidemiology-pathogenesis-and-etiologies
- 38.Grauslund J, Taha AA, Molander LD, et al. Once-weekly semaglutide doubles the five-year risk of nonarteritic anterior ischemic optic neuropathy in a Danish cohort of 424,152 persons with type 2 diabetes. Int J Retina Vitreous. 2024;10(1):97. doi: 10.1186/s40942-024-00620-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Arnold AC. Pathogenesis of nonarteritic anterior ischemic optic neuropathy. J Neuroophthalmol. 2003;23(2):157-163. doi: 10.1097/00041327-200306000-00012 [DOI] [PubMed] [Google Scholar]
- 40.Toosy AT, Mason DF, Miller DH. Optic neuritis. Lancet Neurol. 2014;13(1):83-99. doi: 10.1016/S1474-4422(13)70259-X [DOI] [PubMed] [Google Scholar]
- 41.Votruba M, Moore AT, Bhattacharya SS. Clinical features, molecular genetics, and pathophysiology of dominant optic atrophy. J Med Genet. 1998;35(10):793-800. doi: 10.1136/jmg.35.10.793 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Rigi M, Almarzouqi SJ, Morgan ML, Lee AG. Papilledema: epidemiology, etiology, and clinical management. Eye Brain. 2015;7:47-57. doi: 10.2147/EB.S69174 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Hebsgaard JB, Pyke C, Yildirim E, Knudsen LB, Heegaard S, Kvist PH. Glucagon-like peptide-1 receptor expression in the human eye. Diabetes Obes Metab. 2018;20(9):2304-2308. doi: 10.1111/dom.13339 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Chen T, Ma J, Shan G, Zhong Y. The polymorphisms of ATOH 7, ET-1 and ACE in non-arteritic anterior ischemic optic neuropathy. Exp Eye Res. 2018;174:147-151. doi: 10.1016/j.exer.2018.05.021 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eAppendix. TriNetX analytics platform
eTable 1. Specification and emulation of pragmatic target trials
eTable 2. Eligibility criteria and exposure definitions
eTable 3. Outcome definitions
eTable 4. Definitions of covariates
eTable 5. Characteristics of semaglutide group and comparison group before and after propensity score matching
eFigure 1. Graphical illustration of study design
eFigure 2. Propensity score matching and density function before and after matching semaglutide or tirzepatide and other antidiabetic medication groups on listed variables in eTable 4
eFigure 3. Propensity score matching and density function before and after matching semaglutide or tirzepatide and other GLP-1RA groups on listed variables in eTable 4
Data Sharing Statement
