创伤患者大量输血预测评分方案建立与诊断性能评价

Establishment and diagnostic performance evaluation of massive blood transfusion prediction scoring scheme for trauma patients

目的通过回顾性分析创伤输血患者临床病例资料,分析创伤输血患者红细胞输注的独立影响因素,以建立大量输血预测评分方案并评价其临床预测性能。方法收集南昌大学第一附属医院2013年1月1日—2016年12月31日创伤输血患者265例,根据患者入院后24 h红细胞输血量分为大量输血组（输血量≥18 U）和非大量输血组（输血量〈18 U）,分析2组患者的临床资料,使用单因素和多因素回归筛选红细胞输注的相关影响因素,以建立与大量输血预测评分方案,并使用受试者操作曲线（ROC）评价其临床预测性能。结果 1）本研究共筛选纳入265例患者,其中非大量输血患者229例,大量输血患者36例,2组患者年龄（47. 21±14. 50 vs 43. 86±14. 95）、性别（170/59vs 28/8）和致伤原因间差异均不具统计学意义（P〉0. 05）。非大量输血患者创伤严重度评分（ISS）、心率（HR）、凝血酶原时间（PT）、活化部分凝血活酶时间（APTT）、剩余碱（BE）值均明显高于大量输血组（P〈0. 05）,格拉斯哥昏迷评分（GCS）、收缩压（SBP）、血红蛋白（Hb）、纤维蛋白原（Fbg）均明显低于大量输血组（P〈0. 05）; 2）多因素回归分析显示创伤类型、ISS、HR、Hb、PT、Fbg、BE是创伤输血患者RBC输注的独立影响因素（P〈0. 05）; 3）建立了大量输血预测评分方案,大量输血预测评分总分为0—8分,ROC分析结果显示曲线下面积为0. 914;当评分为4分时,敏感度和特异度分别为88. 9%和79. 9%。结论大量输血预测评分方案具有较好的敏感度和特异度,当创伤患者评分大于4分时应积极启动大量输血方案,给予患者输血治疗,以改善患者预后,提高患者生存率。

Objective Through retrospective analysis for clinical case data of traumatic transfusion patients,the independent influencing factors of red blood cell（ RBC） transfusion in traumatic transfusion patients were analyzed to establish a massive transfusion prediction scoring program and evaluate its clinical predictive performance. Methods The data of 265 traumatic transfusion patients were collected dating from January 1,2013 to December 31,2016 in the First Affiliated Hospital of Nanchang University. According to the RBC transfusion volume within 24 h after admission,the patients were divided into the massive transfusion group（ the units of transfusion ≥18 U） and the non-massive transfusion group（ the units of transfusion 〈18 U）. The clinical data of the two groups were compared,and the related factors of RBC transfusion were screened by single factor and multi-factor regression to establish a massive transfusion prediction scoring model. Its clinical predictive performance were evaluated by using the receiver operating characteristic curve. Results 1） There were no significant differences in age,gender and cause of injury between the two groups（ P〉0. 05）. The differences among other clinical indicators were statistically significant（ P〈0. 05）. 2） Multivariate regression analysis showed that trauma type,injury severity score,heart rate,hemoglobin value,prothrombin time,fibrinogen,and base excess were independent factors of RBC transfusion in traumatic transfusion patients（ P〈0. 05）. 3） A massive transfusion prediction scoring model has been established and the prediction score ranges from 0 to 8 points. The ROC analysis showed that the area under the curve was 0. 914.At a score of 4 points,the sensitivity and specificity were 88. 9% and 79. 9%,respectively. Conclusion The massive transfusion prediction scoring model we constructed presented improved sensitivity and specificity. When the trauma patient scores more than 4 points,the massive transfusion protocol should be immediately initiated,and patients should be given blood transfusion treatment to improve the treatment outcome and the survival rate of trauma patients.