Abstract
Purpose :
Anti-vascular endothelial growth factor (VEGF) intravitreal injections have become the mainstay treatment for patients with wet macular degeneration (wAMD). However, an increased risk of adverse cardiovascular events, such as myocardial infarction (MI), may be associated with the use of these injections. Few studies have examined the relationship between anti-VEGF injections and MI. This single-institution, retrospective observational study aims to assess the relationship between initial anti-VEGF injection treatment and the incidence of MI in patients being treated for wAMD.
Methods :
Patients who had a history of anti-VEGF injections for treatment of wAMD and geographic atrophy (GA) patients who did not receive injections were obtained from the electronic medical records. A propensity analysis utilizing greedy 1:1 replacement was performed to match wAMD treated anti-VEGF patients to GA patients who had no history of injections. Patients were matched across age, sex, BMI, race/ethnicity, diabetes status, and tobacco use to evaluate the binary outcome of having an MI or not, with significance set at p<0.05.
Results :
A total of 809 wAMD patients were matched to a GA control group of 316 patients. The prevalence of an MI was 2.8% in the GA population and 4.2% in the wAMD cohort. Patients had an average age of 80 in the wAMD group with 64% female and 82 in the GA control with 65% female. After matching, standardized differences indicated a successful match, defined as less than 0.10 or greater than -0.10. 809 distinct wAMD patients and 253 GA patients were utilized. A risk ratio of 0.74 (95% CI 0.46, 1.17) was found when comparing the intervention to the control group. The average time patients were followed was 29.47 months. The majority of patients continued to receive multiple injections throughout this observation period.
Conclusions :
Overall, no difference was found in the risk of MI when comparing wAMD treated with anti-VEGF to GA patients not treated with anti-VEGF. More evidence-based research is needed to determine if a time-to-event association exists between the two variables.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.