糖尿病视网膜病变被引次数最多的100篇文章分析
该研究通过Scopus数据库检索1960年至2020年6月发表的糖尿病视网膜病变文献,分析被引次数最多的100篇文章的影响力与特征。结果显示这些文章发表于1961至2017年,中位被引503次,随机对照试验虽仅占24%但发表于更高影响因子期刊且被引更多,33篇被国际临床指南引用,人工智能筛查DR为新兴领域。
要点
- 百篇高被引DR文章中84%发表于1990年后,中位被引503次。
- 随机对照试验仅占24%但发表于更高影响因子期刊且被引次数更多。
- 33篇高被引文章被多个国际DR临床指南引用。
核心结论
- 百篇高被引DR文章中随机对照试验被引最多,且发表于非眼科期刊的文章多于眼科期刊。
- 该分析凸显了百篇高被引DR文章对现有DR管理指南的影响,并识别出AI筛查DR为新兴领域及亚洲贡献的当代上升。
研究设计与样本
- 由两名独立研究者检索Scopus数据库,纳入1960年至2020年6月发表的糖尿病视网膜病变文章,分析被引次数最多的100篇。
主要发现
- 59%发表于综合医学/糖尿病期刊,37%发表于眼科期刊;前六种期刊贡献56%。
- 观察性研究占33%最常见,随机对照试验占24%但发表于更高影响因子期刊且被引更多(中位影响因子和被引分别为7.113、503 vs 21.437、696.5,均P<0.05)。
- 33篇被多个国际DR临床指南引用;美国贡献63%,但2000年后18%来自亚洲;第一和最后作者中男性均超过80%。
- AI筛查DR的两篇文章分别排名第14和第99,分别发表于2016和2017年。
关键数值与推荐
- 随机对照试验与观察性研究比较:中位影响因子和被引分别为7.113、503 vs 21.437、696.5,均P<0.05。
- 33篇被多个国际DR临床指南引用。
临床怎么用
- 百篇高被引DR文章多数发表于非眼科期刊,提示DR管理证据需跨学科关注综合医学与糖尿病期刊。
- 随机对照试验被引更高且影响因子更高,提示高质量RCT对指南和临床实践影响更大。
- AI筛查DR为新兴高被引领域,提示临床可关注AI在DR筛查中的应用证据。
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
Diabetic retinopathy (DR) research has had significant advancements over the past decades. We analysed the impact and characteristics of the top 100 (T100) most-cited articles in DR research. The Scopus database was searched for articles published from 1960 to June 2020 by two independent investigators. The T100 DR articles were published between 1961 and 2017 with median citations of 503 (range: 306-20 100); 84% were published after 1990. More than half (59%) were published in general medical/diabetes journals while 37% in ophthalmology journals. The top six journals contributed to 56% of the T100: Ophthalmology (n=13), Archives of Ophthalmology (n=12), Diabetes (n=9), New England Journal of Medicine (n=8), Journal of the American Medical Association (n=7) and The Lancet (n=7). Although observational studies were most popular (33%), randomised controlled trials (RCTs, 24%) published in journals with higher impact factor (IF) and citations (median IF and citations=7.113, 503 vs 21.437, 696.5, both p-value<0.05). 33 of the T100 were cited by several international DR clinical guidelines. The USA contributed to 63% of T100, but 18% of articles published after 2000 came from Asia. More than 80% of both first and last authors were men. Artificial intelligence (AI) to screen for DR ranked 14th and 99th despite recent publications in 2016 and 2017, respectively. To conclude, our T100 analysis showed that RCTs were most-cited and more articles were published in non-ophthalmology than ophthalmology journals. It highlights the impact the T100 DR has in shaping guidelines used to date in DR management, identifies AI for DR screening as an emerging area and shows a contemporary rise of Asian contribution in DR research.
引用
- Koh BMQR, Banu R, Nusinovici S, et al. 100 most-cited articles on diabetic retinopathy. Br J Ophthalmol. 2020. PMID: 32855165. doi:10.1136/bjophthalmol-2020-316609
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。