查尔森合并症指数评分、急诊脓毒症病死率评分及两者联合应用预测急诊肺部感染患者预后能力的对比研究

Comparative study between Charlison＇ s weighted index of comorbidities score, the diagnostic criteria for emergency sepsis score and combination of the two scoring systems to predict the emergencypulmonary infection prognosis

目的比较查尔森合并症指数（WIC）、急诊脓毒症病死率（MEDS）及两者联合应用预测急诊肺部感染患者预后的预测效果。 方法山东省青岛市海慈医疗集团急诊室于2016年1月至2017年1月收治肺部感染患者327例，入院时记录WIC评分、MEDS评分，并依据WIC评分、MEDS评分进行危险分层，依据28 d治疗转归情况分为存活组和死亡组，采用ROC曲线寻找WIC评分、MEDS评分预测急诊肺部感染患者预后的最佳临界值，并比较WIC评分、MEDS评分、WIC评分与MEDS评分联用及急性生理学与慢性健康状况评分Ⅱ（APACHEⅡ）预测急诊肺部感染患者预后能力。 结果WIC评分低危、中危和高危患者死亡率分别为13.7%（29/212）、48.7%（38/78）和78.4%（29/37），差异有统计学意义（χ2=82.097，P=0.000），MEDS评分低危、中危和高危患者死亡率分别为11.3%（23/203）、50.6%（40/77）和73.3%（33/45），差异有统计学意义（χ2=145.526，P=0.000）；存活组与死亡组的WIC评分分别为（1.3 ± 0.9）分、（2.7 ± 1.1）分，差异有统计学意义（t=11.030，P=0.000），存活组的MEDS评分[（6.1 ± 4.0）分]显著低于死亡组[（12.6 ± 4.9）分]，差异有统计学意义（t=11.502，P=0.000），单独的WIC、MEDS评分预测急诊肺部感染预后的最佳临界值分别为1.7分、11.6分，WIC、MEDS评分及联合应用预测预后的ROC曲线下面积分别为0.632、0.798、0.897，联合预测的灵敏度和准确度[93.8%（212/226）和89.9%（294/327）]显著高于单独的WIC[72.7%（168/231）和75.2%（246/327）]、MEDS评分[67.5%（156/231）和72.2%（236/327）]预测，差异有统计学意义（χ2=0.562~42.954，P〈0.05），且联合应用和APACHEⅡ预测预后的灵敏度、准确度差异无统计学意义（P〉0.05）。 结论WIC评分与MEDS评分联合应用预测急诊肺部感染患者预后的灵敏度和准确度较单独的WIC评分、MEDS评分高，预测效果更好。

ObjectiveTo compare the predictive effect of Charlison＇s weighted index of comorbidities （WIC）, the diagnostic criteria for emergency sepsis （MEDS） and combination of the two scoring systems to predict the emergency pulmonary infection prognosis. MethodsA total of 327 patients with pulmonary infection admitted from January 2016 to January 2017 were enrolled in this study whose WIC score, MEDS score and risk stratification were recorded at admission. They were divided into survival group and death group according to the 28 d treatment outcome, the optimal cutoff of WIC score and MEDS score to predict the prognosis were found by ROC curve, and the prediction effect of WIC score, MEDS score, the combined use of both and APACHE Ⅱ to predict the prognosis were compared. ResultsThe mortality of low, middle and high risk of WIC score were 13.7% （29/212）, 48.7% （38/78） and 78.4% （29/37） with significant difference （χ2= 82.097, P = 0.000）, mortality of low, middle and high risk of MEDS score were 11.3%（23/203）, 50.6%（40/77） and 73.3%（33/45） with significant difference （χ2= 145.526, P = 0.000）. The WIC scores in survival group and death group were 1.3 ± 0.9 and 2.7 ± 1.1 with significant difference （t = 11.030, P = 0.000）. The MEDS score of live group （6.1 ± 4.0） was significantly lower than death group （12.6 ± 4.9） （t = 11.502, P = 0.000）. the optimal cutoff values of WIC and MEDS to predict prognosis were 1.7 points, 11.6 points, the ROC curve area between WIC, MEDS score and combined application to predict prognosis were 0.632, 0.798, 0.897, and the sensitivity and accuracy of the combined prediction[93.8% （212/226）/89.9% （294/327）] were significantly higher than those of the individual WIC[72.7% （168/231）/75.2% （246/327）] and MEDS[67.5% （156/232）/72.2% （236/327）] （χ2=0.562-42.594, P 〈 0.05）. The sensitivity and accuracy of the combined application and APACHE Ⅱ to predict of prognosis had no statistical significant difference （P 〉 0.05）. ConclusionsThe sensitivity and accuracy of WIC score combined with MEDS score to predict the prognosis of patients with acute lung infection is higher than the individual WIC score and MEDS score, and its prediction effect is more better.