作者：吴晓，杨涛，陈石，徐巍龙

单位：

1.210029江苏省南京市，南京中医药大学附属医院 江苏省中医院呼吸与危重症医学科 2.200040上海市，复旦大学附属华山医院急诊科 3.210029江苏省南京市，南京中医药大学附属医院 江苏省中医院内分泌科 通信作者：徐巍龙，E-mail：xxzjwx2010@163.com

Units：

1.Department of Respiratory and Critical Care Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine/Jiangsu Province Hospital of Chinese Medicine, Nanjing 210029, China 2.Department of Emergency, Huashan Hospital, Fudan University, Shanghai 200040, China 3.Department of Endocrinology, Affiliated Hospital of Nanjing University of Chinese Medicine/Jiangsu Province Hospital ofChinese Medicine, Nanjing 210029, China Corresponding author: XU Weilong, E-mail: xxzjwx2010@163.com

关键词：

糖尿病,2型; 肺炎; 社区获得性肺炎; CURB-65评分; 血镁; 死亡; 预测;

Keywords：

Diabetes mellitus, type 2; Pneumonia; Community-acquired pneumonia; CURB-65 score; Bloodmagnesium; Death; Forecasting

CLC：

DOI：

10.12114/j.issn.1008-5971.2022.00.123

Funds：

江苏省科技厅自然科学基金资助项目（BK20201504）；江苏省中医院院内发展基金项目（Y20030，Y19064）；江苏省高等学校自然科学研究项目（19KJB360010）

摘要：

目的 探讨CURB-65评分联合血镁预测2型糖尿病（T2DM）合并社区获得性肺炎（CAP）患者住院期间死亡的价值。方法 回顾性纳入2016年6月至2020年6月于南京中医药大学附属医院住院治疗的526例T2DM合并CAP患者。根据患者住院期间生存状况将其分为存活组486例和死亡组40例。通过医院电子病历系统提取患者入院时的临床资料（一般资料、入院时生命体征、实验室检查指标、住院期间治疗措施、住院期间生存结局、CURB-65评分），采用多因素Logistic回归分析探讨T2DM合并CAP患者住院期间死亡的影响因素，采用ROC曲线评估CURB-65评分、血镁及二者联合预测T2DM合并CAP患者住院期间死亡的价值。结果 死亡组意识障碍者占比、尿素氮、低镁血症发生率、中性粒细胞与淋巴细胞比值（NLR）、CURB-65评分高于存活组，PaO2低于存活组（P<0.05）。多因素Logistic回归分析结果显示，意识障碍、低镁血症、CURB-65评分升高是T2DM合并CAP患者住院期间死亡的独立危险因素（P<0.05）。ROC曲线分析结果显示，CURB-65评分、血镁及二者联合预测T2DM合并CAP患者住院期间死亡的AUC分别为0.771[95%CI(0.707,0.833)]、0.712[95%CI(0.639,0.785)]、0.892[95%CI(0.846,0.937)]，最佳截断值分别为3.0分、0.7 mmol/L、2.36，灵敏度分别为75.51%、65.00%、80.00%，特异度分别为72.50%、63.99%、78.60%。结论 CURB-65评分和血镁是T2DM合并CAP患者住院期间死亡的影响因素，且二者联合可较好地预测其死亡风险，以快速筛选危重症患者，进而辅助临床决策及个体化治疗，降低患者死亡率，节约医疗资源。

Abstract：

【Abstract】 Objective To investigate the value of CURB-65 score combined with serum magnesium in predictingdeath during hospitalization in patients with type 2 diabetes mellitus (T2DM) and community-acquired pneumonia (CAP) .MethodsA total of 526 patients with T2DM and CAP hospitalized in Affiliated Hospital of Nanjing University of ChineseMedicine from June 2016 to June 2020 were retrospectively included. According to the survival status of the patients duringhospitalization, they were divided into survival group (486 cases) and death group (40 cases) . The clinical data of patients onadmission (general information, vital signs on admission, laboratory indicators, treatment measures during hospitalization, survivaloutcomes during hospitalization, CURB-65 score) were extracted through the electronic medical record system of hospital.Multivariate Logistic regression analysis was used to analyze the influencing factors for death during hospitalization in patientswith T2DM and CAP. ROC curve was used to evaluate the value of CURB-65 score, serum magnesium and their combination inpredicting death during hospitalization in patients with T2DM and CAP. Results The proportion of disturbance of consciousness,blood urea nitrogen, incidence rate of hypomagnesemia, neutrophil to lymphocyte ratio (NLR) , and CURB-65 score in the deathgroup were higher than those in the survival group, and PaO2 was lower than that in the survival group ( P < 0.05) . MultivariateLogistic regression analysis results showed that disturbance of consciousness, hypomagnesemia, and increased CURB-65score were independent risk factors for death during hospitalization in patients with T2DM and CAP (P < 0.05) . ROC curveanalysis results showed that the AUC of CURB-65 score, serum magnesium and their combination in predicting death duringhospitalization in T2DM patients with CAP was 0.771 [95%CI (0.707, 0.833) ] , 0.712 [95%CI (0.639, 0.785) ] , 0.892 [95%CI(0.846, 0.937) ] , the optimal cut-off value was 3.0, 0.7 mmol/L, 2.36, the sensitivities were 75.51%, 65.00%, and 80.00%, andthe specificities were 72.50%, 63.99%, and 78.60%, respectively. Conclusion CURB-65 score and serum magnesium are theinfluencing factors of death during hospitalization in patients with T2DM and CAP. The combined application of the two can betterpredict the risk of death during hospitalization, so as to quickly screen critically ill patients, thereby assist clinical decisionmaking and individualizing treatment, reduce patients' mortality, and save medical resources.

ReferenceList：