摘要目的 分析南昌市空气污染的主要特征,并探讨空气污染的暴露〔尤其是细颗粒物(PM2.5)〕与重症医学科(ICU)重症肺炎患者病情发展的相关性及其滞后效应.方法 选择2014年1月1日至2016年12月31日江西省人民医院ICU收治的居住地为南昌市的2454例患者,根据诊断将患者分为肺炎组(156例)和非肺炎组(2298例).记录患者一般临床特征及研究期间南昌市的空气污染情况;采用多元回归模型分析气象因素对肺炎患者病情的影响程度,用优势比(OR)反映空气污染的暴露与肺炎患者病情发展的关联强度,用95％可信区间(95％CI)反映关联强度的可信程度;建立分布滞后非线性模型(DLNM),以评估空气质量参数对时间滞后效应的影响.结果 2014至2016年南昌市空气污染情况,总体来看:一氧化碳(CO)、二氧化硫(SO2)和二氧化氮(NO2)年均浓度值较低,但CO和NO2年均浓度值在2015年后有所回升,而SO2年均浓度值在2015年后迅速下降;PM2.5年均浓度值在2014年下降之后趋于平缓;可吸入颗粒物(PM10)年均浓度值在2014年有所下降后2015年又继续回升;臭氧(O3)年均浓度值从2014至2016年呈持续升高的趋势.与非肺炎组比较,肺炎组患者年龄大,男性患者比例高,并具有更高的预期病死率和急性生理学与慢性健康状况评分Ⅱ(APACHEⅡ)评分;且入ICU当日平均气温低,空气污染物PM2.5、PM10均较高.多元回归模型分析显示,入ICU当日PM2.5和气温与ICU肺炎患者显著相关(PM2.5:OR=1.02,95％CI=1.00～1.03,P<0.05;气温:OR=0.96,95％CI=0.92～0.98,P<0.05),PM2.5对ICU肺炎患者的影响至少可以持续5 d(OR=1.04,95％CI=1.00～1.09,P<0.05),直到入院后7 d才消失.根据不同浓度PM2.5对ICU肺炎患者的影响分析,当PM2.5浓度达到200μg/m3时,对ICU肺炎患者的影响可持续5 d(OR=1.45,95％CI=1.07～1.76,P<0.01).结论 PM2.5和气温与ICU肺炎患者病情存在显著相关,并且高浓度的PM2.5对ICU肺炎患者的影响存在滞后效应.

abstractsObjective To analyze the main characteristics of air pollution in Nanchang City, and discuss the correlation between air pollution exposure (especially PM2.5) and the development of pneumonia related intensive care unit (ICU) patients and their lag effect.Methods 2454 patients who lived in Nanchang City admitted to ICU of Jiangxi Provincial People's Hospital from January 1st, 2014 to December 31st, 2016 were enrolled. According to the diagnosis, the patients were divided into pneumonia group (156 cases) and non-pneumonia group (2298 cases). The general clinical characteristics of patients and air pollution concentration in Nanchang in the year between 2014-2016 were collected. Multiple regression model was used to analyze the influence of meteorological factors on the condition of ICU patients associated with pneumonia. Using odds ratio (OR), the correlation intensity of air pollution exposure and pneumonia related ICU patients' development was reflected, and the confidence level of association intensity was reflected by the 95％ confidence interval (95％CI). The distribution lag nonlinear model (DLNM) was established to evaluate the effect of air mass parameters on the time lag effect.Results The results of air pollution analysis in Nanchang City in the year between 2014-2016 showed that the annual average concentration of carbon monoxide (CO), sulfur dioxide (SO2) and nitrogen dioxide (NO2) was low and maintained at the same level in the year between 2014-2016. The annual average concentration of CO and NO2 increased in the year between 2014-2016, but the average annual concentration of SO2 decreased rapidly in the year between 2014-2016, and the average annual concentration of PM2.5 tended to slow down after the year between 2014-2016. The annual average concentration of PM10 decreased in the year between 2014-2016, but continued to rise in the year between 2014-2016. The annual mean concentration of O3 showed a trend of continuous increase from the year between 2014-2016. The age of pneumonia group was generally higher than that of non-pneumonia group, most of them were male, and had higher expected mortality and acute physiology and chronic health evaluation Ⅱ(APACHEⅡ) score. The average air temperature in the pneumonia group was lower than that in the non-pneumonia group on the day of entering the group, and the air pollutants such as PM2.5 and PM10 were significantly higher than those in the non-pneumonia group. The analysis of multiple regression models showed that PM2.5 and air temperature were significantly related to patients with ICU pneumonia on the day of entry (PM2.5:OR = 1.02, 95％CI =1.00-1.03,P < 0.05; air temperature:OR = 0.96, 95％CI = 0.92-0.98,P < 0.05), and the effect of PM2.5 on patients with ICU pneumonia could last for at least 5 days (OR = 1.04, 95％CI = 1.00-1.09,P < 0.05), and the effect disappeared until the 7th day. According to the analysis of the influence of different concentrations of PM2.5 on ICU pneumonia patients, when the PM2.5 concentration reached 200μg/m3, its effect on ICU pneumonia patients would last for 5 days (OR = 1.45, 95％CI = 1.07-1.76,P < 0.01).Conclusion PM2.5 and air temperature are significantly related to the condition of ICU patients with pneumonia, and the influence of high concentration of PM2.5 on ICU patients with pneumonia has a lag effect.