摘要目的:研究社区获得性肺炎（CAP）住院患者心血管事件（CVE）的发生率、危险因素，以及CVE对患者入院后30 d死亡风险的影响。方法:该研究为多中心、回顾性研究。收集2013年1月1日至2015年12月31日在北京市、山东省和云南省的5家教学医院住院的所有CAP患者的病历资料，并将入选患者按是否发生CVE分为CVE组和无CVE组。通过电子病历系统收集入选患者的年龄、性别、基础疾病、入院当天肺炎严重性指数（PSI）/CURB-65评分、血常规、生化检查和影像学资料，以及入院后30 d死亡人数。研究主要终点为住院期间发生急性CVE，次要研究终点为入院后30 d死亡。采用多因素Cox回归模型分析CAP患者发生CVE的独立危险因素。采用Kaplan-Meier生存曲线评估发生和未发生CVE的CAP患者入院后30 d的累计生存率，生存曲线之间的比较采用Log-rank检验。采用多因素Cox回归模型分析CVE对CAP患者入院后30 d死亡的影响。结果:共有3 561例住院CAP患者纳入该研究，其中210例（5.9%）发生了CVE（CVE组），3 351例（94.1%）未发生CVE（无CVE组）。与无CVE组比较，CVE组患者年龄较大（ P<0.001），合并高血压、冠心病、慢性心力衰竭（心衰）、脑血管疾病、慢性阻塞性肺病、慢性肾脏病、吸入因素、长期卧床者比例较高（ P均 <0.001），CURB-65评分3~5分和PSI分级Ⅳ~Ⅴ级者比例较高（ P均 <0.001）。与无CVE组比较，CVE组患者腋温<36 ℃、呼吸频率≥30次/min、意识改变、血白细胞计数>10×10 9/L、血红蛋白<100 g/L、血小板>300×10 9/L、血白蛋白<35 g/L、血尿素氮>7 mmol/L、空腹血糖>11 mmol/L、血C反应蛋白>100 mg/L、血降钙素原≥2 μg/L、动脉血pH值<7.35、动脉血氧合指数≤300 mmHg（1 mmHg=0.133 kPa）以及胸部X线或CT显示多肺叶浸润和胸水者比例较高（ P均 <0.05），入院后30 d病死率也较高（ P<0.001）。有心脑血管基础疾病的患者CVE发生率高于无心脑血管基础病的患者[13.9%（150/1 079）比2.4%（60/2 482），χ 2=178.737， P<0.001）]。PSI分级Ⅰ/Ⅱ级的患者CVE发生率为1.7%（22/1 273），Ⅲ级的患者为7.8%（40/512），Ⅳ/Ⅴ级的患者为16.9%（72/426），依次升高（χ 2=228.350， P<0.001）。CURB-65评分0~1分的患者CVE发生率为3.8%（110/2 924），2分的患者为17.1%（78/457），3~5分的患者为18.9%（14/74），依次升高（χ 2=387.154， P<0.001）。多因素Cox回归分析结果显示，年龄（ HR=1.05，95 %CI 1.02~1.09， P=0.002）、基础疾病为冠心病（ HR=1.88，95 %CI 1.01~3.51， P=0.048）和慢性心衰（ HR=4.25，95 %CI 1.89~9.52， P<0.001）、PSI分级（ HR=1.66，95 %CI 1.50~2.62， P=0.029）以及血降钙素原≥ 2 μg/L（ HR=3.72，95 %CI 1.60~8.66， P=0.002）是CAP患者发生CVE的独立危险因素。Kaplan-Meier生存曲线结果显示，发生CVE的CAP患者入院后累计30 d生存率低于未发生CVE的CAP患者（ P<0.001）。校正了年龄、性别、基础疾病、CURB-65评分和PSI分级后多因素Cox回归分析结果显示，发生CVE增加CAP患者入院后30 d死亡风险（ HR=6.05，95 %CI 3.11~11.76， P<0.001）。 结论:虽然CVE在我国CAP住院患者中的发生率不高，但在重症肺炎和原有心血管基础疾病的患者中则较为常见。年龄、心血管基础疾病和PSI分级以及血降钙素原是住院CAP患者发生CVE的重要危险因素。CVE可增加CAP住院患者30 d死亡风险。

abstractsObjective:To explore the incidence, risk factors of cardiovascular events (CVE) and their impact on 30-day mortality in patients hospitalized with community-acquired pneumonia (CAP).Methods:This is a multicenter, retrospective study. Patients hospitalized with CAP from 5 teaching hospitals in Beijing, Shandong and Yunnan provinces during 1 January 2013 to 31 December 2015 were included and clinical data were retrieved from the Hospital Information System (HIS), and patients were divided into CVE group and non-CVE group. Age, sex, comorbidities, pneumonia severity index(PSI)/CURB-65 score, routine blood test, biochemical examinations, radiological findings on admission and mortality on 30-day after admission were analyzed. The primary endpoint was acute CVE during hospitalization, the secondary endpoint was 30-day death after admission. Multivariate Cox regression analysis was used to explore the risk factors for CVE. Kaplan-Meier survival curve was used to compare the difference on 30-day mortality between CVE patients and non-CVE patients by Log-rank test. Multivariate Cox regression model was used to assess the impact of CVE on the 30-day mortality among CAP patients after adjustment with age, sex, comorbidities, PSI/CURB-65 score.Results:A total of 3 561 CAP patients were included into the final analysis, including 210 (5.9%) patients in CVE group and 3 351 (94.1%) patients in non-CVE group. Compared with patients in non-CVE group, patients in CVE group were older ( P<0.001), prevalence of hypertension, coronary heart disease, chronic heart failure, cerebrovascular disease, chronic obstructive pulmonary disease, chronic kidney disease, aspiration risk and bedrid were significantly higher (all P<0.001); prevalence of CURB-65 score 3-5 and PSI risk class Ⅳ/Ⅴ were also significantly higher (both P<0.001). The proportion of axillary temperature<36 ℃, respiratory rate≥30 beats/minutes, confusion, leukocytes>10×10 9/L, hemoglobin <100 g/L, platelets>300×10 9/L, albumin <35 g/L, blood urea nitrogen >7 mmol/L, fasting blood glucose>11 mmol/L, serum C-reaction protein >100 mg/L, serum procalcitonin≥2 μg/L, arterial pH<7.35, arterial PO 2/FiO 2≤300 mmHg (1 mmHg=0.133 kPa), and multilobar infiltrates and pleural effusion on chest X-ray or CT scan were significantly higher in CVE group than in non-CVE group(all P<0.05); the 30-day mortality was significantly higher in CVE group than in non-CVE group( P<0.001). The incidence of CVE was significantly higher in patients with cardiovascular and cerebrovascular disease(CVD) than in patients without CVD (13.9%(150/1 079) vs. 2.4%(60/2 482), χ 2=178.737, P<0.001). Meanwhile, the incidence of CVE increased with PSI in patients with Ⅰ/Ⅱ, Ⅲ and Ⅳ/Ⅴ class, respectively(χ 2=228.350, P<0.001); and CURB-65 score 0-1, 2 and 3-5, respectively (χ 2=387.154, P<0.001). Cox regression analysis revealed that age ( HR=1.05, 95 %CI 1.02-1.09, P=0.002), coronary heart disease ( HR=1.88, 95 %CI 1.01-3.51, P=0.048), chronic heart failure ( HR=4.25, 95 %CI 1.89-9.52, P<0.001), PSI risk class ( HR=1.66, 95 %CI 1.50-2.62, P=0.029) and serum procalcitonin≥ 2 μg/L ( HR=3.72, 95 %CI 1.60-8.66, P=0.002) were independent risk factors for CVE in CAP patients. Kaplan-Meier curve showed that the survival probability of patients with CVE was significantly lower than patients without CVE ( P<0.001). After adjustment for age, sex, comorbidities and PSI/CURB-65 score, Cox regression model showed that CVE was associated with increased 30-day mortality in CAP patients ( HR=6.05, 95 %CI 3.11-11.76, P<0.001). Conclusions:Although the incidence of CVE is not high in Chinese patients hospitalized with CAP, CVE is common in patients with severe pneumonia and in patients with CVD. Age, cardiovascular disease, PSI risk class and serum procalcitonin are the risk factors for CVE in this patient cohort. CVE is related to increased 30-day mortality in CAP patients.