短期中低剂量糖皮质激素对甲型流行性感冒社区获得性肺炎患者预后的影响

Impact of short-term low-medium dose of corticosteroids on the clinical outcomes of patients with community-acquired pneumonia due to influenza A

目的探讨短期中低剂量糖皮质激素治疗对甲型流行性感冒(以下简称流感)社区获得性肺炎住院患者临床结局的影响。方法本研究为多中心、回顾性分析。纳入北京积水潭医院、青岛市市立医院、北京回民医院、北京朝阳医院和云南省第二人民医院2013年1月1日至2018年12月31日收治的693例甲型流感社区获得性肺炎患者。比较应用糖皮质激素治疗组和未应用糖皮质激素治疗组患者的临床特征。首剂糖皮质激素均于入院后72 h内使用,剂量为甲泼尼龙(0.6±0.3)mg/(kg·d),疗程为(4.0±1.2)d。应用矫正的logistic回归模型,分析糖皮质激素对患者临床结局(无创机械通气、有创机械通气、需要使用血管活性药物、入住重症监护病房、入院后30 d的病死率、需要胰岛素治疗的高血糖和消化道出血)的影响。统计学处理采用Mann-Whitney检验和χ^2检验。结果693例患者中,132例使用糖皮质激素。logistic回归分析结果显示,合并哮喘[比值比(odds ratio,OR)=15.528,95%可信区间(confidevce interval,CI)1.953~123.484,P=0.01]、慢性阻塞性肺疾病(OR=21.904,95%CI 4.548~105.504,P<0.01)和动脉血氧合指数<300 mmHg(1 mmHg=0.133 kPa;OR=2.701,95%CI 1.513~4.822,P<0.01)是甲型流感社区获得性肺炎患者接受糖皮质激素治疗的独立危险因素。矫正的logistic回归分析结果显示,中低剂量糖皮质激素治疗可降低甲型流感社区获得性肺炎患者早期(首剂糖皮质激素治疗后0~3 d)无创机械通气的风险(OR=0.342,95%CI 0.156~0.750,P<0.01),但增加晚期(首剂糖皮质激素治疗后4~14 d)需要使用血管活性药物(OR=2.651,95%CI 1.913~6.306,P<0.01)、晚期需要胰岛素治疗的高血糖(OR=9.739,95%CI 2.174~21.769,P=0.019)、入住重症监护病房(OR=3.075,95%CI 1.166~8.143,P<0.01)和入院后30 d死亡(OR=2.372,95%CI 1.337~4.549,P<0.01)的风险;合并哮喘或慢性阻塞性肺疾病的患者(OR=2.343,95%CI 1.145~4.073,P<0.01)和动脉血氧合指数<300 mmHg的患者(OR=1.961,95%CI 1.029~4.212,P<0.01),可增高入院后30 d死亡风险。结论短期中低剂量糖皮质激素治疗与甲型流感社区获得性肺炎患者诸多不良临床结局有关,不宜常规使用。

Objective To evaluate the impact of short-term low-medium dose of corticosteroids on the clinical outcomes of patients with community-acquired pneumonia due to influenza A(FluA-CAP).Methods This was a multicenter,retrospective study,including 693 patients hospitalized with FluA-CAP from Beijing Jishuitan Hospital,Qingdao Municipal Hospital,Beijing Huimin Hospital,Beijing Chao-Yang Hospital and the 2nd People′s Hospital of Yunnan Province during January 1,2013 to December 31,2018.The clinical characteristics of patients with or without corticosteroids administration were compared.The first dose of corticosteroids was administrated within 72 hours after admission,with the average dose of methylprednisolone(0.6±0.3)mg/(kg·d)and duration of(4.0±1.2)days.An adjusted logistic regression model was performed to assess the impact of corticosteroids treatment on the clinical outcomes(noninvasive ventilation,invasive ventilation,vasopressor use,admittance to intensive care unit(ICU),30-day mortality,hyperglycemia needing insulin treatment and gastrointestinal bleeding).Mann-Whitney test andχ^2 test were used for the statistical analysis.Results Among the 693 patients,132 patients received corticosteroids.Logistic regression analysis revealed that asthma(odd ratios(OR)=15.528,95%confidence interval(CI)1.953-123.484,P=0.01),chronic obstructive pulmonary disease(OR=21.904,95%CI 4.548-105.504,P<0.01)and arterial partial pressure of oxygen(PaO2)/fraction of inspired oxygen(FiO2)<300 mmHg(1 mmHg=0.133 kPa,OR=2.701,95%CI 1.513-4.822,P<0.01)were independent risk factors for corticosteroids use in the FluA-CAP patients.An adjusted logistic regression model showed that low-medium dose corticosteroids administration was associated with decreased risks for early(defined as zero to three days after the first dose of corticosteroids)noninvasive ventilation(OR=0.342,95%CI 0.156-0.750,P<0.01),and increased risk for late(defined as four to 14 days after the first dose of corticosteroids)vasopressor use(OR=2.651,95%CI 1.913-6.306,P<0.01),late hyperglycemia which needed insulin treatment(OR=9.739,95%CI 2.174-21.769,P=0.019),ICU admission(OR=3.075,95%CI 1.166-8.143,P<0.01)and the 30-day mortality(OR=2.372,95%CI 1.337-4.549,P<0.01).In patients with asthma or chronic obstructive pulmonary disease(OR=2.343,95%CI 1.145-4.073,P<0.01)and PaO2/FiO2<300 mmHg(OR=1.961,95%CI 1.029-4.212,P<0.01),corticosteroids administration increased the risk of 30-day mortality.Conclusion Low-medium corticosteroids treatment is associated with poor outcomes of FluA-CAP patients,and is not recommended to be used routinely.