文献速递眼底病与葡萄膜炎

日本抗VEGF治疗nAMD期间需手术干预的黄斑下出血发生率:JTY三中心真实世界研究

Submacular Hemorrhage Requiring Procedural Intervention During Anti-VEGF Therapy for Neovascular Age-Related Macular Degeneration in Japan: The JTY Three-Center Real-World Incidence Study.
PubMed 文献 · Retina · 2026-10-30

该多中心回顾性真实世界研究纳入47,358次抗VEGF注射,识别出54例需气体注入或玻璃体切割术的黄斑下出血事件(每1000次注射1.14例)。PCV占事件72.2%,程序治疗后最终BCVA中位数1.00 logMAR(20/200),较出血前0.46 logMAR(约20/60)明显下降。探索性药物分析未显示各抗VEGF药物间调整后发生率有显著差异。

要点

  • 抗VEGF治疗nAMD期间需手术干预的黄斑下出血发生率为每1000次注射1.14例。
  • PCV患者占黄斑下出血事件的72.2%,既往有黄斑下出血史者占22.2%。
  • 程序治疗后最终视力中位数降至1.00 logMAR(20/200),视力预后差。

核心结论

  • 抗VEGF治疗nAMD期间需手术干预的黄斑下出血不常见但视力后果严重。
  • PCV是黄斑下出血事件的主要亚型,占72.2%。
  • 探索性药物特异性分析未显示各抗VEGF药物调整后发生率有显著差异。

研究设计与样本

  • 日本三中心回顾性真实世界发生率研究,审查手术记录识别抗VEGF治疗期间需气体注入或玻璃体切割术的黄斑下出血事件。
  • 分母为同期nAMD抗VEGF注射总次数47,358次。
  • 采用准泊松回归估计每1000次注射发生率和站点调整发生率比(IRR)。

主要发现

  • 共识别54例需程序干预的黄斑下出血事件,发生率为每1000次注射1.14例。
  • 程序治疗后最终BCVA中位数1.00 logMAR(Snellen等效20/200),出血前为0.46 logMAR(约20/60)。
  • PCV占39例(72.2%),既往黄斑下出血史见于12眼(22.2%)。
  • 粗发生率:雷珠单抗2.20、布西珠单抗1.09、阿柏西普1.08、法瑞西单抗0.24(每1000次注射)。
  • 以阿柏西普为参照,IRR:雷珠单抗1.83(95%CI 0.73-4.59;P=0.195),布西珠单抗0.91(95%CI 0.25-3.28;P=0.883),法瑞西单抗0.20(95%CI 0.01-3.16;P=0.251)。

临床怎么用

  • 对nAMD抗VEGF治疗患者应警惕黄斑下出血风险,尤其是PCV亚型及有出血史者。
  • 发生需手术干预的黄斑下出血后视力预后较差,需及时识别并处理。
  • 药物选择对黄斑下出血发生率的影响尚不明确,需个体化评估。

局限性

  • 回顾性真实世界研究设计,可能存在记录偏倚。
  • 药物特异性分析为探索性,置信区间宽,未显示显著差异。
  • 需未来患者层面研究评估亚型特异性发生率、出血复发及PCV靶向治疗策略。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

PURPOSE: To estimate the incidence and clinical characteristics of submacular hemorrhage (SMH) requiring procedural intervention during anti-VEGF therapy for neovascular age-related macular degeneration (nAMD) in a Japanese multicenter real-world setting, with exploratory comparison according to the preceding anti-VEGF agent. METHODS: This multicenter retrospective real-world incidence study reviewed surgical records from three Japanese centers to identify SMH requiring gas injection or vitrectomy during ongoing anti-VEGF treatment. The denominator was the total number of anti-VEGF injections for nAMD during the same period. Incidence per 1,000 injections and site-adjusted incidence rate ratios (IRRs) were estimated using quasi-Poisson regression. RESULTS: Among 47,358 injections, 54 SMH events requiring procedural intervention were identified (1.14 per 1,000 injections). Median final BCVA after procedural treatment was 1.00 logMAR (Snellen equivalent, 20/200), compared with 0.46 logMAR before SMH (approximately 20/60). PCV accounted for 39 events (72.2%), and prior SMH was documented in 12 eyes (22.2%). Crude incidence was 2.20 for ranibizumab, 1.09 for brolucizumab, 1.08 for aflibercept, and 0.24 for faricimab. Using aflibercept as the reference, IRRs were 1.83 for ranibizumab (95% confidence interval [CI], 0.73-4.59; P = 0.195), 0.91 for brolucizumab (95% CI, 0.25-3.28; P = 0.883), and 0.20 for faricimab (95% CI, 0.01-3.16; P = 0.251). CONCLUSIONS: SMH requiring procedural intervention during anti-VEGF therapy for nAMD was uncommon but visually severe; among affected eyes, PCV accounted for 72.2% of events. Exploratory drug-specific analyses did not demonstrate significant adjusted differences among anti-VEGF agents. Future patient-level studies should evaluate subtype-specific incidence, hemorrhagic recurrence, and polyp-directed treatment strategies in PCV-rich populations.

来源:Retina,查看原文。版权归原作者所有。

引用

  1. Inoda S, Takahashi H, Azuma K, et al. Submacular Hemorrhage Requiring Procedural Intervention During Anti-VEGF Therapy for Neovascular Age-Related Macular Degeneration in Japan: The JTY Three-Center Real-World Incidence Study. Retina. 2026. PMID: 42961146. doi:10.1097/IAE.0000000000005044

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。