Abstract: Objective　To explore whether RIFLE grades in evaluation of cute kidney injury (AKI) can predict the prognosis in elderly patients with community-acquired pneumonia (CAP). Methods　Four hundred and eight no less than 65-year-old patients with definite diagnosis as CAP treated at Fuzhou Pulmonary Hospital from January 2017 to December 2021 were selected, including 313 males and 95 females. According to the RIFLE grades, there were 281 cases in the no-AKI group and 127 cases in the AKI group. The AKI group was further stratified into a risk group (61 cases), an injury group (47 cases), and a failure group (19 cases). The influences of demographic characteristics and severities on prognosis in the four groups were compared. The measurement data of normal distribution were compare between the groups by ANOV, and the measurement data not of normal distribution by rank sum test. The enumeration data were compared by chi square test. The two-way ordinal categorical data were done by Spearman correlation test. Results　In this study, 127 patients developed AKI. The incidence of AKI was 31.1%. The majority of the patients in every group was male. There were no statistical differences in gender and smoke history between the 4 groups. The AKI group was (74.8±4.8) years old, the risk group (73.9±4.1), the injury group (74.1±4.1), and the failure group (79.8±5.4), indicating that AKI grades had a positive correlation with age (r=0.359). There were no statistical differences in the ratios of coronary heart disease, chronic obstructive pulmonary disease, and cerebrovascular disease between the 4 groups, but were in the ratios of hypertension [the no-AKI group 9.25% (26/281), the risk group 19.7% (12/61), the injury group 25.5% (12/47), and the failure 31.6% group (6/19)] and diabetes [the no-AKI group 15% (42/281), the risk group 22.9% (14/61), the injury group 27.6% (13/47), and the failure group 42.1% (8/19)] (χ²=17.216, P=0.001; χ²=12.560, P=0.006). The ratios of developing AKI between the low-risk and high-risk groups calculated by PSI and CURB-65 scores were 48.0% (61/127) vs. 52.0% (66/127) and 29.9% (38/127) vs 70.1% (89/127), respectively. With the rising grade of RIFLE, the portions of the patients administrated with machinal ventilation, vasoactive agents, and continuous renal replacement therapy also went up. The hospital stays in the no-AKI group, the risk group, the injury group, and the failure group were (12.2±2.4) days, (17.1±2.4) days, (19.8±3.0) days, and (29.8±4.7) days, and the mortalities 0.4% (1/281), 1.6% (1/61), 6.4% (3/47), and 47.4% (9/19), respectively, with statistical differences between the 4 groups. Correlation analysis showed that AKI grade was positively correlated with hospital stay (r=0.747) and mortality (r=0.992). Conclusions　The incidence of AKI is high in elderly patients with CAP, and has a positive correlation with age. Severe pneumonia complications with hypertension and diabetes cline to develop AKI, resulting in prolonged hospitalization and increased mortality.

Key words: Acute kidney injury, RIFLE grades, Elderly pneumonia, Community-acquired pneumonia, Prognosis

摘要： 目的　探讨应用RIFLE分级进行急性肾损伤（acute kidney injury，AKI）的评估能否预测老年社区获得性肺炎患者预后。方法　选取2017年1月至2021年12月于福建省福州肺科医院住院的社区获得性肺炎患者共408例，年龄≥65岁，其中男313例，女95例。根据是否发生急性肾损伤分为no-AKI组281例及AKI组127例（AKI组根据AKI严重程度再分危险、损伤和衰竭亚组，各61、47和19例），比较4组患者人口学特征、病情严重程度及对预后的影响。各组间比较符合正态分布的采用方差分析，非正态分布采用秩和检验，计数资料采用χ²检验，对于双向有序计数资料同时进行Spearman相关性检验。结果　AKI组127例，AKI发生率为31.1%（127/408）。各组患者均以男性患者为主，性别、吸烟史在各组间比较差异均无统计学意义。AKI组年龄（74.8±4.8）岁，危险、损伤和衰竭组的年龄分别为（73.9±4.1）岁、（74.1±4.1）岁、（79.8±5.4）岁，AKI分级与年龄呈正相关（r=0.359）。基础疾病中冠心病、慢性阻塞性肺病、脑血管病对是否发生AKI差异均无统计学意义，但高血压病［no-AKI组9.3%（26/281），危险组19.7%（12/61），损伤组25.5%（12/47），衰竭组31.6%（6/19）］、糖尿病［no-AKI组14.9%（42/281），危险组22.9%（14/61），损伤组27.7%（13/47），衰竭组42.1%（8/19）］在4组间比较差异均有统计学意义（χ²=17.216，P=0.001；χ²=12.56，P=0.006）。肺炎严重指数（PSI）、CURB-65评分在低风险和高风险组AKI发生比例分别为48.0%（61/127）比52.0%（66/127）、29.9%（38/127）比70.1%（89/127），随着肾功能损伤加重，4组患者需要进行无创机械通气、有创机械通气、血管活性药物及连续性肾脏替代治疗（continuous renal replacement therapy，CRRT）的比例也随之增高。4组患者住院天数分别为（12.2±2.4）d、（17.1±2.4）d、（19.8±3.0）d、（29.8±4.7）d，死亡比例分别为0.4%（1/281）、1.6%（1/61）、6.4%（3/47）、47.4%（9/19），4组间比较差异均有统计学意义。相关性分析显示，AKI分级与住院天数（r=0.747）、死亡比例（r=0.992）呈正相关。结论　老年社区获得性肺炎AKI发生率高，且与年龄呈正相关，重症肺炎及合并高血压、糖尿病导致AKI发生率增高，住院天数延长，病死率升高。

关键词: 急性肾损伤, RIFLE分级, 老年肺炎, 社区获得性肺炎, 预后