Objective: To compare the change of prognosis and pathogens of adult patients with severe community-acquired pneumonia (CAP) requiring admission to Respiratory Intensive Care Unit (RICU) during different period, and estimate the risk factors of severe CAP. Methods: The clinical data of 186 patients with severe CAP admission to RICU was collected and assessed the trend in outcomes of these patients during three time periods: Period Ⅰ (January 1 2000 to October 31 2006), Period Ⅱ (November 1 2006 to December 31 2010), and Period Ⅲ (January 1 2011 to December 31 2015), analyzed the risk factors by Cox Proportional Hazards regression, and compared the pathogens detection. Results: The total mortality was 34.9%(65/186)during 16 years, three periods was 41.4%, 35.3% and 27.7% respectively.After adjustment for other variables, the mortality was significant higher for patients admitted during Period Ⅰ than that of Period Ⅲ (HR: 1.900; 95.0%CI 1.053-3.429, P=0.033), and no significant difference for Period Ⅰ compared with Period Ⅱ (HR: 1.394; 95.0%CI 0.774-2.514, P=0.269), Period Ⅱ compared with Period Ⅲ(HR: 1.379; 95.0%CI 0.800-2.375, P=0.248). Cox regression analysis showed the predicting risk factors for prognosis including BUN≥7.14 mmol/L, acute heart failure, acute renal failure, multilobar infiltrates and septic infectious shock ( P<0.05). Multi-variate Cox regression analysis showed that multilobar infiltrates(HR=3.951, 95.0%CI 2.252-6.929, P=0.001) and septic infectious shock(HR=2.044, 95.0%CI 1.131-3.692, P=0.018) were the independent risk factors for prognosis.104 (55.9%) were detected pathogens in 186 patients.The common pathogens were Staphylococcus aureus (18 isolated strains), Legionella pneumophila (16 isolated strains), Streptococcus pneumonia (13 isolated strains). There was no significant difference between bacterial isolation except virus detection(P=0.033). Conclusion: There has been an improvement in the prognosis of severe CAP.Multilobar infiltrates and septic infectious shock were the independent risk factors for prognosis.Pathogens had changed in SCAP over time, and the detection rate of virus increased gradually.
目的： 比较不同时期入住重症监护病房成人重症社区获得性肺炎(CAP)患者的预后及病原体变化，并评估重症CAP患者死亡的危险因素。 方法： 回顾性分析186例北京大学人民医院呼吸与危重症医学科的重症CAP患者记录基础资料、入院时的临床资料及实验室检查、病原学检测、治疗过程、住院天数、预后等，对比3个时期患者的预后：Ⅰ期(2000年1月1日至2006年10月31日)，Ⅱ期(2006年11月1日至2010年12月31日)，Ⅲ期(2011年1月1日至2015年12月31日)，采用Cox风险回归分析死亡的危险系数及相关危险因素，并比较3期患者病原学检测情况。 结果： 16年186例重症CAP患者病死65例，总病死率34.9%，3期分别为41.4%，35.3%，27.7%；调整混杂因素后，对比3期病死的危险系数分别为1.394(Ⅰ期对比Ⅱ期，95.0%CI 0.774～2.514，P＝0.269)，1.900(Ⅰ期对比Ⅲ期，95.0%CI 1.053～3.429，P＝0.033)，1.379(Ⅱ期对比Ⅲ期，95.0%CI 0.800～2.375，P＝0.248)，Ⅰ期与Ⅲ期比较差异有统计学意义(P<0.05)，Ⅰ期与Ⅱ期、Ⅱ期与Ⅲ期相比差异无统计学意义(P>0.05)。单因素Cox回归模型分析显示，尿素氮≥7.14 mmol/L、多肺叶浸润、感染性休克及急性心力衰竭、急性肾衰为病死的预测危险因素；多因素Cox回归模型分析显示，多肺叶浸润(HR＝3.951，95.0%CI 2.252～6.929，P＝0.001)和感染性休克(HR＝2.044，95.0%CI 1.131～3.692，P＝0.018)为病死的独立危险因素。186例患者中有104例(55.9%)检测出病原体，金黄色葡萄球菌(18株)、嗜肺军团菌(16株)及肺炎链球菌(13株)为常见病原体。除流感病毒(P＝0.033)外，3期病原体分布差异无统计学意义(P>0.05)。 结论： 重症CAP患者病死率呈下降趋势，多肺叶浸润和感染性休克是死亡的独立危险因素。随着时间推移，重症CAP病原谱有所变化，病毒检出率逐渐升高。.
Keywords: Community-acquired infections; Noxae; Pneumonia; Prognosis.