Leibovitz Eyal, Khanimov Israel, Wainstein Julio, Boaz Mona
Department of Internal Medicine "A" Yoseftal Hospital, Eilat, Israel.
Department of Internal Medicine "A" Yoseftal Hospital, Eilat, Israel.
Diabetes Metab Syndr. 2019 Jan-Feb;13(1):222-226. doi: 10.1016/j.dsx.2018.07.004. Epub 2018 Jul 17.
To study the association of documented hypoglycemia with length of stay, 30-day mortality, and 1-year mortality, among patients with and without diabetes admitted to internal medicine units.
The electronic medical records of all patients hospitalized in internal medicine departments at E. Wolfson Medical Center, Holon, Israel, between 1/1/2010 and 31/12/2013, were reviewed. Data extracted included all glucose measurements (performed using an institutional blood glucose monitoring system). Patients were considered hypoglycemic if at least one hypoglycemic event was recorded. Regression analysis was used to assess the association between documented hypoglycemia and length of stay, 30-day and one-year mortality. Age, sex, reason for admission, and the Charlson comorbidity index were entered as covariates, and the most conservative model was developed.
The study population included 45,272 patients (mean age 68.9 ± 17.8 years, 49.4% males, 21.0% had diabetes mellitus). The rate of hypoglycemia in the total study population was 7.5% (16.8% among DM patients, 6.0% among patients without diabetes, p < 0.001). Patients with documented hypoglycemia had a longer length of hospital stay (9.3 ± 18.7 vs. 3.1 ± 6.4 days, p < 0.001), as well as higher risk for both 30-day (23.7% vs. 7.0%, p < 0.001) and 1-year mortality (41.6% vs. 15.3%, p < 0.001). Cox regression analysis showed that hypoglycemia significantly increased risk death at one year (HR 2.436, 95% CI 2.298-2.582, p < 0.001) independent of age, sex, the Charlson comorbidity index, DM status and reason for admission.
Documented hypoglycemia is associated with prolonged length of hospital stay and increased risk for both 30-day and 1-year mortality, regardless of diabetes mellitus status.
研究内科病房收治的糖尿病患者和非糖尿病患者中，记录在案的低血糖与住院时间、30天死亡率和1年死亡率之间的关联。
回顾了2010年1月1日至2013年12月31日期间在以色列霍隆市E. 沃尔夫森医疗中心内科住院的所有患者的电子病历。提取的数据包括所有血糖测量值（使用机构血糖监测系统进行）。如果记录到至少一次低血糖事件，则患者被视为低血糖患者。采用回归分析评估记录在案的低血糖与住院时间、30天和1年死亡率之间的关联。将年龄、性别、入院原因和查尔森合并症指数作为协变量纳入分析，并建立最保守的模型。
研究人群包括45272名患者（平均年龄68.9±17.8岁，男性占49.4%，21.0%患有糖尿病）。总研究人群中的低血糖发生率为7.5%（糖尿病患者中为16.8%，非糖尿病患者中为6.0%，p<0.001）。有记录的低血糖患者住院时间更长（9.3±18.7天 vs. 3.1±6.4天，p<0.001），30天死亡率（23.7% vs. 7.0%，p<0.001）和1年死亡率（41.6% vs. 15.3%，p<0.001）的风险也更高。Cox回归分析表明，低血糖显著增加了1年时的死亡风险（风险比2.436，95%置信区间2.298-2.582，p<0.001），且独立于年龄、性别、查尔森合并症指数、糖尿病状态和入院原因。
无论糖尿病状态如何，记录在案的低血糖与住院时间延长以及30天和1年死亡率增加相关。
