复苏后综合征的死亡危险因素及预后数学模型

Study on Death Risk Factors and a Prognostic Mathematical Model in Patients with Post Resuscitation Syndrome

【目的】探讨复苏后综合征(PRS)的死亡危险因素,利用有统计学意义的因素建立数学模型,对PRS的预后进行预测,针对性地指导临床治疗。【方法】回顾性分析总结我院近7年来确诊的63例PRS患者的临床资料,按复苏后30d时的预后情况分组,生存组24例,死亡组39例,基于相关影响因素的单因素统计分析,进而行Logistic多元回归分析及Fisher判别分析。【结果】单因素分析结果提示自主呼吸、谷草转氨酶、谷丙转氨酶、消化道出血、凝血酶原时间、瞳孔对光反射消失、格拉斯哥评分、自主循环恢复(ROSC)时间、衰竭器官数目等9项指标在两组间差异有统计学意义(P<0.05);Logistic多元回归分析显示,自主循环恢复时间、衰竭器官数目、消化道出血、瞳孔对光反射消失为PRS患者发生死亡的独立危险因素(P<0.01);根据4项独立危险因素建立的Fisher判别函数总判别正确率为92.1%。【结论】自主循环恢复时间、衰竭器官数目、消化道出血、瞳孔对光反射消失为PRS患者发生死亡的独立危险因素,据此建立的判别函数可较准确地预测PRS预后,积极处理可干预的死亡危险因素是提高临床疗效的关键。

[Objective] To identify the death risk factors in patients with post resuscitation syndrome （PRS）, establish a mathematical model to predict the prognosis with those statistically significant factors so as to improve therapeutic efficiency. [Methods] We retrospectively analyzed the clinical material of 63 patients with PRS, who were diagnosed in our hospital in the last seven years. All the patients were divided into two groups according to the prognosis on the 30th day after resuscitation. There were 24 patients in the survival group and 39 patients in the death group. Based on the results by univariate analysis, the data were processed by Logistic multiple regression and Fisher＇s discriminant analysis. [ Results] Univariate analysis revealed that there were nine variables, including spontaneous respiration, glutamic-oxaloacetic transaminase, glutamate-pyruvate transaminase, gastrointestinal bleeding, prothrombin time, abolition of pupillary light reflex, Glasgow coma scale, time of ROSC and numbers of failure organs had a significant difference between two groups （P 〈 0.05）. Logistic multiple regression analysis showed that the recovery time of spontaneous circulation, the number of failure organs, gastrointestinal bleeding and abolition of pupillary light reflex were independent death risks of post resuscitation syndrome （P 〈 0.01）. Fisher＇s linear discriminant function was set up based on these 4 factors. The overall correct percentage for prediction of prognosis was 92.1% according to this mathematical model. [Conclusion] The recovery time of spontaneous circulation, the number of failure organs, gastrointestinal bleeding and abolition of pupillary light reflex were 4 independent death risks of post resuscitation syndrome. The mathematical model of discriminant function could be applied to predicting the prognosis of PRS accurately. Pertinent treatment on these death risk factors is the key to the improvement of the clinical curative effect.