临床指标与ICU老年肺部感染患者预后的关系

目的观察临床指标在预测ICU老年肺部感染患者预后中的作用。 方法收集2017年1月1日至2018年12月31日在青海省人民医院ICU住院治疗的222例老年肺部感染患者的临床资料。统计患者入科时的性别、年龄、民族、急性生理与慢性健康状况（APACHE Ⅱ）评分、序贯器官衰竭评估（SOFA）评分、CURB-65评分、降钙素原（PCT）、B型利钠肽（BNP）、C反应蛋白（CRP）、白细胞计数（WBC）、血小板（PLT），以及出科时的最近一次的PCT、BNP、CRP、WBC及PLT。记录患者的ICU住院时间及出ICU时的预后情况。采用二元Logistic回归模型分析临床预后的危险因素。采用受试者工作特征曲线（ROC）分析各指标在预测患者预后中的敏感度、特异度以及Youden指数。 结果存活组与死亡组的民族构成、性别构成、APACHE Ⅱ评分［（10（8，13）分 vs 19（16，23）分］、入科PCT［2.39（0.88，4.66）ng/ml vs 6.02（2.69，12.94）ng/ml］、入科BNP［702（367，1172）pg/ml vs 954（460，1496）pg/ml］、入科CRP［5.84（2.31，14.98）mg/dl vs 10.47（4.86，18.22）mg/dl］、出科PCT［1.47（0.56，2.65）ng/ml vs 6.33（1.98，12.54）ng/ml］、出科BNP［464（215，713）pg/ml vs 1414（412，2377）pg/ml］、出科CRP［3.35（1.40，7.09）mg/dl vs 5.10（2.75，11.98）mg/dl］比较，差异均有统计学意义（P＜0.05）。自动出院和死亡的比例为39.19%。APACHE Ⅱ评分＞15分组的病死率更高（75.51% vs 10.48%，P＜0.001）；PCT在出科时下降50%以上的患者的病死率更低（30.30% vs 46.34%，P＜0.05）；BNP在出科时下降50%以上的患者的病死率明显下降（18.75% vs 50.70%，P＜0.001）。二元Logistic回归模型分析结果显示，SOFA评分（OR=2.775，P=0.001）、CURB-65评分（OR=10.052，P=0.001）、出科PCT（OR=1.189，P=0.029）、出科BNP（OR=1.002，P=0.001）为预后不良的危险因素。 结论入院时APACHE Ⅱ评分>14.5分，SOFA评分＞8.5分，CURB-65评分＞3.5分，PCT＞4.35 ng/ml，BNP＞1155 pg/ml以及CRP＞6.53 mg/dl可以明显增加ICU老年肺部感染患者的死亡风险。

Relationship between clinical biomarkers and prognosis of geriatric patients with pneumonia in intensive care unit

ObjectiveTo investigate predictive value of clinical markers in geriatric patients with pneumonia in ICU. MethodsThe clinical data of 222 patients admitted to the ICU of Qinghai Provincial People's Hospital from January 2017 to December 2018 was collected. Gender, age, ethnicity, APACHE Ⅱ score, SOFA score, CURB-65 score, PCT, BNP, CRP, WBC and PLT were recorded. The length of stay in ICU and prognosis of the patients were also recorded. The risk factors of clinical prognosis were analyzed by binary Logistic regression model. The sensitivity, specificity and Youden index of clinical predicting indicators were analyzed with the receiver operating characteristic curve (ROC). ResultsThe APACHE Ⅱ 10(8, 13) vs 19(16, 23)], ethnicity, gender, PCT on admission 2.39(0.88, 4.66) vs 6.02(2.69, 12.94)], BNP on admission 702(367, 1172) vs 954(460, 1496)], CRP on admission 5.84(2.31, 14.98) vs 10.47(4.86, 18.22)], PCT at discharge 1.47(0.56, 2.65) vs 6.33(1.98, 12.54)], BNP at discharge 464(215, 713) vs 1414(412, 2377)], CRP at discharge 3.35(1.40, 7.09) vs 5.10(2.75, 11.98)] were significantly different between survival group and death group (P＜0.05). The percentage of patients with against-advice discharge and death was 39.19%. Patients with APACHE Ⅱ score over 15 had a higher mortality rate (75.51% vs 10.48%). The mortality of patients who had a half decreased PCT level at discharge were lower (30.30% vs 46.34%). The mortality of patients who had a half decreased BNP at discharge were lower (18.75% vs 50.70%). Binary Logistic regression model analysis results showed that SOFA score (OR=2.775, P=0.001), CURB-65 score (OR=10.052, P=0.001), PCT at discharge (OR=1.189, P=0.029), BNP at discharge (OR=1.002, P=0.001) were independent risk factors for poor outcome. ConclusionThe APACHE Ⅱ score over 14.5, SOFA score over 8.5, CURB-65 score over 3.5, PCT over 4.35 ng/ml, BNP over 1155 pg/ml, and CRP over 6.53 mg/dl can significantly increase the risk of death in geriatric patients with pneumonia in ICU.