PURPOSE Intravitreal anti-vascular endothelial growth factor (VEGF) pharmacotherapy has become standard of care for the management of diabetic macular edema (DME). The systemic safety profile of this treatment in routine clinical practice remains incompletely understood. We used a large claims database to investigate the risk of systemic serious adverse events (SAEs) in patients receiving anti-VEGF for DME compared with controls treated with macular laser photocoagulation or intravitreal corticosteroid. DESIGN Retrospective cohort study. PARTICIPANTS By using a large U.S. insurance database, we identified privately insured and Medicare Advantage patients aged ≥18 years treated with anti-VEGF for DME between January 1, 2006, and December 31, 2015, along with control patients receiving macular laser or corticosteroid. We included patients with 1 year of medical coverage before initial DME treatment. METHODS We assessed associations between treatment modalities and predefined systemic outcomes using Cox proportional hazards regression. We performed 2 separate comparisons, one between anti-VEGF and macular laser and one between anti-VEGF and corticosteroid. We used inverse propensity score weighting for the first comparison to account for treatment selection bias. For the second, we used 2:1 propensity score matching on demographics, year, and baseline comorbidities because of the smaller number of corticosteroid-treated patients. MAIN OUTCOME MEASURES Risk of cerebrovascular disease, myocardial infarction, major bleeding, and all-cause hospitalization occurring within 6 months of initial DME treatment as hazard ratios (HRs) with 95% confidence intervals (CIs). RESULTS A total of 23 348 patients receiving treatment for DME met inclusion criteria; 13 365 received macular laser, 9219 received intravitreal anti-VEGF, and 764 received intravitreal corticosteroid as initial treatment. Anti-VEGF pharmacotherapy was not associated with an increased hazard of cerebrovascular disease (HR, 0.96; 95% CI, 0.65-1.41; P = 0.83), major bleeding (HR, 1.23; 95% CI, 0.76-1.99; P = 0.41), or myocardial infarction (HR, 1.03; 95% CI, 0.73-1.44; P = 0.88) when compared with macular laser for DME; however, there was an increased hazard of post-treatment all-cause hospital admission (HR, 1.17; 95% CI, 1.05-1.30; P = 0.01). The rates of all primary systemic SAE outcomes were similar after treatment with anti-VEGF versus corticosteroid (P > 0.05 for all). CONCLUSIONS We identified no increased risk of cerebrovascular disease, myocardial infarction, or major bleeding within 6 months after intravitreal anti-VEGF pharmacotherapy for the treatment of DME in routine clinical practice. A potential difference in all-cause hospitalization may merit further investigation.

中文翻译：

---

在常规临床实践中，与玻璃体内抗VEGF疗法治疗糖尿病性黄斑水肿相关的全身性不良事件的风险。

目的玻璃体内抗血管内皮生长因子（VEGF）药物治疗已成为治疗糖尿病性黄斑水肿（DME）的护理标准。在常规临床实践中，这种治疗方法的全身安全性尚不完全清楚。我们使用了一个大型索赔数据库来研究接受抗VEGF DME治疗的患者与经黄斑激光光凝或玻璃体内注射糖皮质激素治疗的对照组相比，发生全身性严重不良事件（SAE）的风险。设计回顾性队列研究。参与者通过使用大型美国保险数据库，我们确定了2006年1月1日至2015年12月31日期间接受DME抗VEGF治疗的年龄在≥18岁的私人保险和Medicare Advantage患者，以及接受黄斑激光或皮质类固醇治疗的对照患者。在首次DME治疗之前，我们纳入了1年医疗保险的患者。方法我们使用Cox比例风险回归评估了治疗方式与预定义的系统结果之间的关联。我们进行了2个单独的比较，一个在抗VEGF和黄斑激光之间，一个在抗VEGF和皮质类固醇之间。我们使用逆倾向得分加权进行第一次比较，以说明治疗选择的偏倚。第二，由于皮质类固醇治疗的患者人数较少，我们在人口统计学，年份和基线合并症中使用了2：1倾向评分匹配。主要观察指标在初次DME治疗后6个月内发生脑血管疾病，心肌梗塞，大出血和全因住院的风险为危险比（HRs），置信区间（CIs）为95％。结果共有23 348名接受DME治疗的患者符合入选标准。13 365例接受黄斑激光治疗，9219例接受玻璃体内抗VEGF，764例接受玻璃体内皮质类固醇治疗。抗VEGF药物疗法与增加脑血管疾病的危险性无关（HR，0.96; 95％CI，0.65-1.41; P = 0.83），大出血（HR，1.23; 95％CI，0.76-1.99; P = 0.41） ）或与黄斑激光治疗DME相比，心肌梗塞（HR，1.03； 95％CI，0.73-1.44； P = 0.88）；但是，全因医院治疗后的危险性增加（HR，1.17； 95％CI，1.05-1.30； P = 0.01）。抗VEGF和皮质类固醇治疗后，所有主要系统性SAE结局发生率均相似（所有P均> 0.05）。结论在常规临床实践中，在玻璃体内抗VEGF药物治疗DME后的6个月内，我们没有发现脑血管疾病，心肌梗塞或大出血的风险增加。全因住院的潜在差异可能值得进一步调查。