Drug-induced retinal vein occlusion: a disproportionality analysis from the FDA adverse event reporting system (2004-2023)

Abstract

Introduction: Retinal vein occlusion (RVO) often causes irreversible visual impairment, making early prevention crucial. This study aims to identify associations between different medications and RVO and provide information for clinical practice.

Method: This study included reports of RVO from the FDA Adverse Event Reporting System (FAERS) database from the first quarter (Q1) of 2004 to the fourth quarter (Q4) of 2023. The reported drugs were analyzed for adverse drug reaction (ADR) signals using four disproportionality algorithms. Kaplan-Meier curves and median time to onset were used to evaluate the drugs.

Results: From 2004 to 2023, the FAERS database recorded 6,151 reports associated with RVO. Disproportionality analyses identified 25 drugs significantly associated with RVO. Mirabegron showed the highest risk signal, followed by Raloxifene, Tadalafil, Fingolimod, and Bimatoprost. These high-risk drugs are distributed across different therapeutic areas, including urogenital system and sex hormones, ophthalmic drugs, nervous system drugs, musculoskeletal system drugs, anti-tumor and immune-modulating drugs, and anti-parasitic drugs. Specific drug targets such as adrenergic receptor agonists, hormone regulators, and PDE5 inhibitors were identified as high risk. Ophthalmic drugs exhibited the longest median time to adverse ocular reactions at 532.01 days, followed by anti-parasitic drugs, nervous system drugs, urogenital system and sex hormone drugs, anti-tumor and immune-modulating drugs, and musculoskeletal system drugs.

Conclusion: This study provides an overview of drug-induced RVO, identifying potential culprit drugs and their distribution characteristics. These findings enhance understanding of medication safety and help optimize clinical practice.

Keywords: FDA adverse event reporting system; adverse events; disproportionality analysis; pharmacovigilance; retinal vein occlusion.

FIGURE 1

Flowchart of patient selection and data cleaning for drug-induced retinal vein occlusion in the FAERS database. Notes: Process of screening to obtain information on patients with retinal vein occlusion and their medication use from the FAERS database.

FIGURE 2

Distribution of baseline data for patients reporting adverse events of retinal vein occlusion in the FAERS database. Notes: Baseline characteristics of 2,304 patients with drug-induced RVO (A) Age and gender distribution of patients (B) Temporal trend of reported cases (C) Routes of administration (D) Patient outcomes (E) Occupational distribution of reporters (F) Number of reports by country (G) Geographic distribution of reports.

FIGURE 3

Ranking drug risk based on the BCPNN algorithm. Notes: Ranking drugs based on their risk of causing retinal vein occlusion using the BCPNN algorithm.

FIGURE 4

Forest plots and heat maps of drugs with positive signals for drug-induced retinal vein occlusion based on disproportionality analysis methods from the FAERS database. Notes: Disproportionality analysis of RVO risk drugs based on ATC classification (A) and target classification (B). Abbreviation: ROR, reporting odds ratio; PRR, proportional reported ratio; BCPNN, Bayesian confidence propagation neural network; MGPS, multi-item gamma poisson shrinker. Abbreviation: BCPNN, Bayesian confidence propagation neural network.

FIGURE 5

Cumulative risk curves for ocular adverse reactions by drug classification. Notes: The timeline of ocular adverse reactions induced by related drugs: (A) Kaplan-Meier curve (B) Median induction time. Statistical differences are labeled with letters; groups with the same letter indicate no significant difference.