Suppr超能文献

昏迷颅脑损伤患者的预后及结局预测

Prognosis and prediction of outcome in comatose head injured patients.

作者信息

Braakman R, Habbema J D, Gelpke G J

出版信息

Acta Neurochir Suppl (Wien). 1986;36:112-7. doi: 10.1007/978-3-7091-8859-0_30.

Abstract

Recent studies on the prognosis of comatose head injured patients have identified single powerful prognostic features at various time points during the first month after onset of coma. Using appropriate statistical methods even more powerful combinations of prognostic features can be selected. At each time point, optimal prediction requires sets of only 3 to 5 features. These features include depth and duration of coma as assessed by the Glasgow Coma Scale, pupil reactivity to light, age in decades, and spontaneous and reflex eye movements. In individual new patients, bedside predictions are now possible, e.g. using a booklet with prognosis tables like the one used in Rotterdam. Doctors actually learn by using these tables as they retain some of the information. However, the main application is that these tables permit one to evaluate whether differences in survival rates in different centres with different management regimes are due to a difference in management efficacy or to a difference in initial severity of injury.

摘要

近期关于昏迷颅脑损伤患者预后的研究已确定,在昏迷发作后的第一个月内,不同时间点有单一强大的预后特征。使用适当的统计方法,甚至可以选择更强大的预后特征组合。在每个时间点,仅需3至5个特征集就能实现最佳预测。这些特征包括通过格拉斯哥昏迷量表评估的昏迷深度和持续时间、瞳孔对光反应、年龄(以十岁为单位)以及自发和反射性眼球运动。对于个体新患者,现在可以进行床边预测,例如使用像鹿特丹所使用的带有预后表格的手册。医生在使用这些表格的过程中确实能学到东西,因为他们记住了其中一些信息。然而,这些表格的主要用途是,能让人评估不同管理方案的不同中心的存活率差异是由于管理效果的差异还是初始损伤严重程度的差异。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验