Risk of Stroke, Myocardial Infarction, and Death Among Patients With Retinal Artery Occlusion and the Effect of Antithrombotic Treatment

Abstract

Purpose: To evaluate the risk of future stroke, myocardial infarction (MI), and death of patients with retinal artery occlusion (RAO) and the effect of various antithrombotic treatments as secondary prevention.

Methods: This cohort study was based on nationwide health registries and included the entire Danish population from 2000 to 2018. All patients with RAO were identified and their adjusted risks of stroke, MI, or death in time periods since RAO were compared with those of the Danish population. Furthermore, antithrombotic treatment of patients with RAO was determined by prescription claims, and the association with the risk of stroke, MI, or death was assessed using multivariate Poisson regression models and expressed as rate ratios (RR) with 95% confidence intervals (95% CIs).

Results: After inclusion, 6628 individuals experienced a first-time RAO, of whom 391 had a stroke, 66 had a MI, and 402 died within the first year after RAO. RAO was associated with an increased risk of stroke, MI, or death which persisted for more than 1 year for all three outcomes but was highest on days 3 to 14 after RAO for stroke, with an adjusted RR of 50.71 (95% CI, 41.55-61.87), and on days 14 to 90 after RAO for MI and death, with adjusted RRs of 1.98 (95% CI, 1.25-3.15) and 1.64 (95% CI, 1.28-189), respectively. Overall, antithrombotic treatment was not associated with any protective effect the first year.

Conclusions: Patients with RAO had an increased risk of stroke, MI, or death. No protective effect of antithrombotic treatment was shown.

Translational relevance: These findings are relevant to the management of patients with RAO.

Figure 1.

(A–C) Cumulative risk of stroke and myocardial infarction accounting for competing risk and Kaplan–Meier curves for all-cause mortality. For patients with RAO, time corresponds to days after the RAO, starting at day 3.

Figure 2.

Incidence and adjusted RRs for stroke, MI, and death among patients with RAO compared with the general Danish population stratified in time periods after RAO. RRs are adjusted for sex, age, calendar time, diabetes mellitus, hypertension, heart failure, chronic kidney disease, cancer, atrial fibrillation, ischemic heart disease (only the analysis on stroke and death), and stroke (only the analysis on death and MI). Incidences are per 1000 person years. PY, person years; CI-95, 95% confidence interval.

Figure 3.

Incidence and adjusted RRs of stroke among patients with RAO in time periods after RAO stratified by treatment with antithrombotic medication, with no treatment as the reference. Rate ratios are adjusted for sex, age, calendar time, diabetes mellitus, hypertension, heart failure, chronic kidney disease, cancer, atrial fibrillation, ischemic heart disease, and previous antithrombotic treatment. Incidences are per 1000 person years. AC, anticoagulant treatment.

Figure 4.

Incidence and adjusted RRs of myocardial infarction among patients with RAO in time periods after RAO stratified by treatment with antithrombotic medication, with no treatment as the reference. RRs are adjusted for sex, age, calendar time, diabetes mellitus, hypertension, heart failure, chronic kidney disease, cancer, atrial fibrillation, stroke, and previous antithrombotic treatment. Incidences are per 1000 person years.

Figure 5.

Incidence and adjusted RRs of all-cause death among patients with RAO in time periods after RAO stratified by treatment with antithrombotic medication, with no treatment as the reference. RRs are adjusted for sex, age, calendar time, diabetes mellitus, hypertension, heart failure, chronic kidney disease, cancer, atrial fibrillation, ischemic heart disease, stroke, and previous antithrombotic treatment. Incidences are per 1000 person years.