Hirata Yu, Tomioka Hiromi, Sekiya Reina, Yamashita Shyuji, Kaneda Toshihiko, Kida Yoko, Nishio Chihiro, Kaneko Masahiro, Fujii Hiroshi, Nakamura Takehiro
Department of Diabetes and Endocrinology, Kobe City Medical Center West Hospital, Japan.
Intern Med. 2013;52(21):2431-8. doi: 10.2169/internalmedicine.52.9594.
Information available on the clinical features and outcomes of pneumonia in diabetic patients is limited. There are no data on the association between glycemic control during hospitalization and mortality in this population. The objective of this study is to examine whether the presence of hyperglycemia on admission and during hospitalization is associated with mortality in diabetic patients admitted to the hospital for pneumonia.
This study is a retrospective observational cohort study of diabetic adults hospitalized for the first time for pneumonia between 2005 and 2011 in a 358-bed community hospital. Univariate and multivariate analyses were performed for 30-day all-cause hospital mortality adjusted for sex, age, type of pneumonia (community-acquired pneumonia or nursing and health care-associated pneumonia), severity of pneumonia according to the A-DROP score and various comorbidities in consideration of the serum glucose and hemoglobin A1c levels on admission and the mean plasma glucose level during hospitalization.
Of the 1,499 pneumonia patients evaluated, 185 (12.3%) (mean age 75 years) had diabetes mellitus. Fourteen (7.6%) of the 185 diabetic patients died within 30 days after admission. According to the univariate analysis, 30-day mortality was significantly associated with the A-DROP score (p<0.0001), the admission glucose level (p=0.01) and the mean plasma glucose level during hospitalization (p<0.0001). Even after adjusting for factors related to the severity of pneumonia, the mean plasma glucose level during hospitalization remained significantly associated with 30-day mortality (p=0.004).
Hyperglycemia determined according to the mean plasma glucose level during hospitalization is independently associated with 30-day all-cause hospital mortality in diabetic patients admitted for pneumonia.
关于糖尿病患者肺炎临床特征和转归的现有信息有限。尚无该人群住院期间血糖控制与死亡率之间关联的数据。本研究的目的是探讨入院时及住院期间高血糖的存在是否与因肺炎入院的糖尿病患者的死亡率相关。
本研究是一项回顾性观察性队列研究，研究对象为2005年至2011年期间在一家拥有358张床位的社区医院首次因肺炎住院的成年糖尿病患者。对30天全因住院死亡率进行单因素和多因素分析，校正了性别、年龄、肺炎类型（社区获得性肺炎或护理及医疗保健相关性肺炎）、根据A-DROP评分的肺炎严重程度以及各种合并症，并考虑了入院时的血糖和糖化血红蛋白水平以及住院期间的平均血浆葡萄糖水平。
在评估的1499例肺炎患者中，185例（12.3%）（平均年龄75岁）患有糖尿病。185例糖尿病患者中有14例（7.6%）在入院后30天内死亡。根据单因素分析，30天死亡率与A-DROP评分（p<0.0001）、入院血糖水平（p=0.01）和住院期间的平均血浆葡萄糖水平（p<0.0001）显著相关。即使校正了与肺炎严重程度相关的因素后，住院期间的平均血浆葡萄糖水平仍与30天死亡率显著相关（p=0.004）。
根据住院期间平均血浆葡萄糖水平确定的高血糖与因肺炎入院的糖尿病患者30天全因住院死亡率独立相关。
