Ranibizumab and Risk of Hospitalisation for Ischaemic Stroke and Myocardial Infarction in Patients with Age-Related Macular Degeneration: A Self-Controlled Case-Series Analysis

Abstract

Background: Ranibizumab, a vascular endothelial growth factor (VEGF) inhibitor, is used in the treatment of age-related macular degeneration. Inhibition of VEGF has an anti-angiogenic action and is associated with thrombogenicity, thus, myocardial infarction and ischaemic stroke are potential side effects of VEGF inhibitors. Objective: Our objective was to assess the association between use of ranibizumab and risk of hospitalisation for ischaemic stroke (IS) and myocardial infarction (MI). Conclusion: This case-series analysis suggests an increased risk of hospitalisation for ischaemic stroke for patients receiving ranibizumab in the 31–60 days risk period. Studies with larger populations are required to confirm the risk in the 1–30 days risk period. No evidence of increased risk of hospitalisation for MI was observed. Methods: The self-controlled case series design was used, including subjects exposed to ranibizumab (Anatomical Therapeutic Chemical [ATC] code S01LA04) who were hospitalized for IS (International Classification of Diseases, tenth edition [ICD-10] code I63) or the combined endpoint of stroke or transient ischaemic attack (TIA) (ICD-10 code G45) or MI (ICD-10 code I21) were identified between August 2007 and March 2013. Rate ratios in exposed periods compared with unexposed periods were calculated using conditional Poisson regression. Results: A total of 323 subjects received ranibizumab and were hospitalized for IS, 490 for IS or TIA, and 391 for MI. Median period of exposure was 8–9 months with follow-up times of approximately 2.8 years. No elevated risk of IS was seen in the 1–30 days post initiation (incidence rate ratio [IRR] 1.36; 95 % confidence interval [CI] 0.98–1.88); however, elevated risk was observed for those who received therapy for 31–60 days (IRR 1.91; 95 % CI 1.13–3.24). Sensitivity analyses adjusting for time-varying confounders found elevated risk in both the 1–30 days and 31–60 days periods. Similar results to those for IS were observed for the combined endpoint of IS or TIA. No association was seen for MI in either time period (1–30 days IRR 0.90, 95 % CI 0.65–1.23; 31–60 days IRR 0.98, 95 % CI 0.54–1.79).