文献速递神经创伤神经重症

Marshall CT评分、CRASH及IMPACT模型对重型颅脑损伤行早期去骨瓣减压术患者不良神经预后的预测值过高

Predicted Unfavorable Neurologic Outcome Is Overestimated by the Marshall Computed Tomography Score, Corticosteroid Randomization After Significant Head Injury (CRASH), and International Mission for Prognosis and Analysis of Clinical Trials in Traumatic Brain Injury (IMPACT) Models in Patients with Severe Traumatic Brain Injury Managed with Early Decompressive Craniectomy.
PubMed 文献 · World Neurosurg · 2017-02-20

在哥伦比亚一家大学医院,127例重型颅脑损伤患者接受早期去骨瓣减压术,评估Marshall CT评分、CRASH和IMPACT模型对14天死亡率、6个月死亡率及6个月结局的预测效能。结果显示三种模型均高估不良神经预后,6个月实际死亡率为29.13%,实际不良结局为37%。

要点

  • Marshall CT评分预测死亡率为52.8%,实际6个月死亡率为29.13%。
  • CRASH计算器预测死亡率为59.9%,实际6个月死亡率为29.13%。
  • IMPACT预测不良结局为77%,实际6个月不良结局为37%。

核心结论

  • 在哥伦比亚一家大学医院,Marshall CT评分、IMPACT和CRASH模型对重型颅脑损伤患者的不良神经预后预测值过高。

研究设计与样本

  • 研究为队列研究,纳入哥伦比亚一家区域创伤中心2年内收治的127例颅脑损伤患者,采用双变量和多变量分析。
  • 通过逻辑回归和受试者工作特征曲线下面积(AUC)评估模型的区分度、准确性和精确度,并用Shapiro-Wilk检验、χ2检验和Wilcoxon检验比较实际结局与预测结局。

主要发现

  • 患者中位年龄33岁,男性占84.25%,损伤严重度评分中位数25,格拉斯哥昏迷量表运动评分中位数3。
  • 6个月死亡率为29.13%,6个月不良结局为37%。
  • Marshall CT评分预测死亡率为52.8%,P=0.104(AUC 0.585;95%CI 0.489-0.681)。
  • CRASH预后计算器预测死亡率为59.9%,P<0.001(AUC 0.706;95%CI 0.590-0.821)。
  • IMPACT预测不良结局为77%,P<0.048(AUC 0.670;95%CI 0.575-0.763)。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

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

INTRODUCTION: Traumatic brain injury (TBI) is of public health interest and produces significant mortality and disability in Colombia. Calculators and prognostic models have been developed to establish neurologic outcomes. We tested prognostic models (the Marshall computed tomography [CT] score, International Mission for Prognosis and Analysis of Clinical Trials in Traumatic Brain Injury (IMPACT), and Corticosteroid Randomization After Significant Head Injury) for 14-day mortality, 6-month mortality, and 6-month outcome in patients with TBI at a university hospital in Colombia. METHODS: A 127-patient cohort with TBI was treated in a regional trauma center in Colombia over 2 years and bivariate and multivariate analyses were used. Discriminatory power of the models, their accuracy, and precision was assessed by both logistic regression and area under the receiver operating characteristic curve (AUC). Shapiro-Wilk, χ2, and Wilcoxon test were used to compare real outcomes in the cohort against predicted outcomes. RESULTS: The group's median age was 33 years, and 84.25% were male. The injury severity score median was 25, and median Glasgow Coma Scale motor score was 3. Six-month mortality was 29.13%. Six-month unfavorable outcome was 37%. Mortality prediction by Marshall CT score was 52.8%, P = 0.104 (AUC 0.585; 95% confidence interval [CI] 0 0.489-0.681), the mortality prediction by CRASH prognosis calculator was 59.9%, P < 0.001 (AUC 0.706; 95% CI 0.590-0.821), and the unfavorable outcome prediction by IMPACT was 77%, P < 0.048 (AUC 0.670; 95% CI 0.575-0.763). CONCLUSIONS: In a university hospital in Colombia, the Marshall CT score, IMPACT, and Corticosteroid Randomization After Significant Head Injury models overestimated the adverse neurologic outcome in patients with severe head trauma.

来源:World Neurosurg,查看原文。版权归原作者所有。

引用

  1. Charry JD, Tejada JH, Pinzon MA, et al. Predicted Unfavorable Neurologic Outcome Is Overestimated by the Marshall Computed Tomography Score, Corticosteroid Randomization After Significant Head Injury (CRASH), and International Mission for Prognosis and Analysis of Clinical Trials in Traumatic Brain Injury (IMPACT) Models in Patients with Severe Traumatic Brain Injury Managed with Early Decompressive Craniectomy. World Neurosurg. 2017. PMID: 28223249. doi:10.1016/j.wneu.2017.02.051

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。