文献速递烧伤

烧伤患者非烧伤原因死亡影响死亡率预测模型:一项全国性探索性研究

Burned patients who die from causes other than the burn affect the model used to predict mortality: a national exploratory study.
PubMed 文献 · Burns · 2017-08-19

该回顾性描述性探索性研究纳入1993至2015年单中心收治的1946例烧伤患者,重点分析Baux评分<100的死亡病例(n=23)的死因。结果显示,全部死亡病例中四分之一Baux评分<100,其中17/23死于非烧伤原因(主要为脑血管病、心血管病和呼吸衰竭)。排除非烧伤原因死亡后,Baux评分的死亡预测AUC从0.9733升至0.9888,敏感性从45.2%升至53.9%。

要点

  • Baux评分<100的死亡患者占全部死亡的四分之一(23/93)。
  • 该组中17/23死于非烧伤原因,以脑血管病(9例)和心血管病(6例)为主。
  • 排除非烧伤原因死亡后,Baux评分预测敏感性从45.2%升至53.9%。

核心结论

  • Baux评分<100的烧伤死亡患者中,多数死于非烧伤原因(尤其是脑血管病),这可能是Baux评分敏感性低的原因。
  • 排除非烧伤原因死亡后,Baux评分的死亡预测价值提高(AUC从0.9733升至0.9888)。

研究设计与样本

  • 回顾性、描述性、探索性研究,纳入1993至2015年单中心收治的全部烧伤患者(n=1946)。
  • 研究组为Baux评分<100的死亡患者(n=23),查阅病历分析死因及共存疾病。

主要发现

  • 总体粗死亡率为5%(93/1946),其中23/93(四分之一)死亡患者Baux评分<100(范围64-99)。
  • Baux评分<100死亡组中,火焰烧伤最常见(18/23),年龄中位数70岁(10th-90th百分位46-86),TBSA中位数21%(5.0-40.5%),全层面积中位数7%(0-27.0%)。
  • 23例中17例死因归类为「非烧伤原因」:脑血管病9例、心血管病6例、呼吸衰竭2例;其余6例为烧伤相关死亡,以多器官衰竭(主要为肾衰竭)为主。
  • 排除非烧伤原因死亡后,Baux评分预测死亡的AUC从0.9733(95%CI 0.9633-0.9834)升至0.9888(95%CI 0.9839-0.9936),敏感性从45.2%升至53.9%。

临床怎么用

  • 对Baux评分<100的烧伤患者,应警惕非烧伤原因(尤其是脑血管病)导致的死亡,完善相关评估。
  • 在利用Baux评分预测死亡率时,需考虑非烧伤原因死亡对模型敏感性的影响。

局限性

  • 本研究为单中心回顾性研究,样本量有限(Baux评分<100死亡患者仅23例)。
  • 研究设计为描述性、探索性,未进行多因素校正或外部验证。

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: The Baux score - the sum of age and total body surface area burned (TBSA %) - is a good predictor of mortality has a high specificity but low sensitivity. Our aim was to examine the causes of death in patients who die with Baux scores of <100, which may explain the lower sensitivity and possibly affect the prediction of mortality. METHODS: All patients admitted to our centre for burn care from 1993 to 2015 (n=1946) were included in this retrospective, descriptive, exploratory study. The study group comprised those patients who died with Baux scores of <100 (n=23), and their medical charts were examined for the cause of death and for coexisting diseases. RESULTS: Crude mortality was 5% (93/1946) for the overall cohort, and a quarter of the patients who died (23/93) had Baux scores of less than 100 (range 64-99). In this latter group, flame burns were the most common (18/23), the median (10th-90th centile) age was 70 (46-86) years and for TBSA 21 (5.0-40.5) %, of which 7 (0-27.0) % of the area was full thickness. The main causes of death in 17 of the 23 were classified as "other than burn", being cerebral disease (n=9), cardiovascular disease (n=6), and respiratory failure (n=2). Among the remaining six (burn-related) deaths, multiple organ failure (predominantly renal failure) was responsible. When we excluded the cases in which the cause of death was not related to the burn, the Baux mortality prediction value improved (receiver operating characteristics area under the curve, AUC) from 0.9733 (95% CI 0.9633-0.9834) to 0.9888 (95% CI 0.9839-0.9936) and the sensitivity estimate increased from 45.2% to 53.9%. CONCLUSION: Patients with burns who died with a Baux score <100 were a quarter of all the patients who died. An important finding is that most of these deaths were caused by reasons other than the burn, usually cerebrovascular disease. This may be the explanation why the sensitivity of the Baux score is low, as factors other than age and TBSA % explain the fatal outcome.

来源:Burns,查看原文。版权归原作者所有。

引用

  1. Pompermaier L, Steinvall I, Elmasry M, et al. Burned patients who die from causes other than the burn affect the model used to predict mortality: a national exploratory study. Burns. 2017. PMID: 28830698. doi:10.1016/j.burns.2017.07.014

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。