發生嚴重低血糖於急診就診之第二型糖尿病患死亡之相關因素分析 -- 五年回溯型世代研究

Analysis of the cause of death among the severe hypoglycemia type 2 diabetes with emergency aids, a five years retrospective cohort study

蔡麗卿(Leih-Ching Tsai)

指導教授 ： 張浤榮 副教授

中山醫學大學/醫學院/醫學研究所/碩士(2012年)

https://doi.org/10.6834/CSMU.2012.00060

摘要

研究背景及目的 糖尿病的盛行率日益增加，依台灣健保資料推算，糖尿病門診人數自2004年佔總人口數之4.8％並逐年攀升(2007年5.4%,2010年7.1%)。糖尿病的盛行率也由1993- 1996年的5.33%，2002年6.5%，上升到2009年9.2%。台灣地區主要死因，2008至2010年糖尿病為第五位，2011年則躍居第四位。可見糖尿病已成為危害國人健康的主要慢性病之一，而其急、慢性併發症中以低血糖為最常見的急性併發症，會造成醫療經濟負擔及病患的心理壓力，而重複的低血糖發作會造成病患生活品質及情緒上不良的影響。過去的研究較少探討嚴重低血糖病人死亡的預測因子。本研究目的在於分析中部地區某區域教學醫院第二型糖尿病病患，發生嚴重低血糖急診室就診後至死亡期間其體重的變化，與相關併發症及其他預測死亡的相關因素。

方法 本研究為病例回溯型世代研究，收案對象為中部某區域教學醫院之第二型糖尿病病患，於2007年1月至2012年1月間，因為低血糖血糖值小於70mg/dL或者需要他人協助，而到本院急診室就診之共同照護網之第二型糖尿病病患，以ICD-9-CM codes 249.8x, 250.8x, 251.0, 251.1, 251.2等代碼做病例篩選出483人(排除不符合者62人，失聯者57人，資料不齊全者174人) 共有190人納入研究，用以比較存活與死亡組以下變項的差異。例如: 糖尿病罹病時間，胰島素治療之時程，糖尿病共照網收案時程，DKA次數，HHNS次數，急診低血糖就診次數，急診觀察時間，總入院次數，昏迷指數等。用Cox regression 找出與死亡相關的危險因子，以p-value＜0.05表示有顯著性差異，再以Kaplan-Meier methods 來畫存活分析圖。

結果 2007年1月至2012年1月期間，死亡者42人，存活者共148人，體重下降大於3%者比體重改變±2.9%者風險 (Hazard ratio, HR) 上升2.88倍 (95% CI 1.38-6.02, p=0.005)達到統計上之意義。體重上升3%者風險為體重改變±2.9%者的0.4倍(95% CI 0.11-1.51, p= 0.175)。BMI小於18.5者比BMI 18.5-23.9者HR上升5.1倍(95% CI 1.56-16.72，p=0.007)。BMI大於25者為BMI 18.5-23.9之間者HR上升0.76倍 (95% CI 0.37-1.56, p=0.456)未達統計上之意義。血紅素低貧血者，其死亡者HR為存活者的3.44倍(95% CI 1.55-7.63, p= 0.002)。淋巴球總數小於1500 cell/mm3 病人死亡的HR上升2.87倍(95% CI為1.38-5.98, p=0.005)。有心血管疾病 (CVD) 的病人的風險為沒有CVD者的2.19倍(95% CI 1.10-4.36, p=0.026)。總入院次數高者死亡的風險上升1.24倍(95% CI 1.08-1.43, p=0.003)，皆達到統計上之意義。

結論 本研究中發現因嚴重低血糖被送到急診室就醫之第二型糖尿病病患，其預後除了心血管疾病之外，還要將糖尿病人的營養狀況及抵抗力等條件也列入考量。而急診室的報告中的血色素低有貧血，總淋巴球數低(TLC)，病人有低的身體質量比值(BMI)，隨後的追蹤有體重下降3% 以上加上總入院次數高者其死亡率也較高，這些變項都是有用且便宜又方便的臨床指標，可以用來有效的預測病患的死亡機率是否上升，以提早提出因應之對策。

關鍵字

第二型糖尿病 ； 低血糖 ； 死亡預測

並列摘要

BACKGROUND Global prevalence of diabetes is increasing. The prevalence of diabetes was estimated to be 2.8% in 2000 and 4.4% in 2030. The total number of diabetes is projected to rise from 171 million in 2000 to 366 million in 2030. According to the database of Taiwan National Health Insurance (NHI), the number of diabetes outpatient clinic increased from 4.8%, 5.4%, 7.1% of 2004, 2007, 2010 respectively. Trends in diabetes prevalence from 1993 to 2009 in Taiwan increased from 5.33% to 9.2%. Diabetes is the 5th leading cause of death in Taiwan from 2008 to 2010, then the 4th in 2011. Hypoglycemia is a severe adverse reaction and causes frequent acute complication of diabetes in emergency services. It creates economic impact on medical care. Recurrent hypoglycemia will cause a negative effect on life quality and build emotional distress. There is little known about the hypoglycemic type 2 diabetes whose following after their presenting to the emergency department (ED) and the predicting of death. The aim of this study was to determine whether ED hypoglycemic episodes were associated with higher mortality, body weight change and co-morbidity in this group of patients, and analyzed the cause of death.

METHODS A retrospective observational cohort studies. During January 2007 to January 2012, eligible patients were derived from a teaching hospital in rural area of mid-Taiwan. When patients had hypoglycemia-related ED visit, they were enrolled via ED diagnosis ICD-9-CM codes (249.8x, 250.8x, 251.0, 251.1, 251.2) and the diabetes medical improvement program. Hypoglycemia was defined as blood glucose below 70mg/dl, or need for third party assistance and attendance at hospital. Initially, 483 patients were selected, and 293 patients were excluded (not meeting inclusion criteria n=52, lost to follow-up n=57 or without complete data n=174) with 190 patients recruited. Differences between subjects who died and survived were analyzed by t test and chi-square test. Cox regression was used to establish the risk factors related to death.

RESULTS In those 5 years retrospective analysis, total 190 patients (452 episodes/ person-years) who presented to ED due to hypoglycemia, revealed 42 (22.1%) died, 148 (77.9%) alive. Related to the moderate body weight change (MBWC) ±2.9%, Cox regression pointed out patients with body weight gain (BWG) over 3% had a 2.87-fold increased risk of death (CI 1.38-6.00, p=0. 005). Related to body mass index (BMI) 18.5-24.9, BMI less than 18.5 had a 5.59-fold increased death risk (CI 1.70-18.38, p=0.005). For patients with anemia, male hemoglobin (Hb) <13g/dL, female Hb <12g/dL, had a 3.04-fold increased risk (CI 1.36-6.77, p=0.007). Subjects in the total lymphocyte count <1500 cell/mm3 group, had a 3.05-fold increased risk (CI 1.44-6.46, p=0.004). Participants with cardiovascular disease, had a 2.03-fold increased risk (CI 1.02-4.05, p=0.045). Patients with higher total admission rate, had a 1.22-fold increased risk (CI 1.05-1.40, p=0.008).

CONCLUSIONS Patients who presented to the ED with severe hypoglycemia, subsequent greater body weight loss over 3%, BMI less than 18.5, anemia, TLC less than 1500 cells/mm3, established CVD, and the higher total admission rate had increased mortality.

並列關鍵字

type 2 diabetes mellitus ； hypoglycemia ； mortality prediction

參考文獻

老年人糖尿病之盛行現況 (2012-6-03上網)

行政院衛生署，台灣地區老人營養健康狀況調查1999~2000：

complications in type II diabetes (VA CSDM).Diabetes

Akizawa T, et al (2011). Management of anemia in chronic kidney disease patients: baseline findings from Chronic Kidney. Disease Japan Cohort Study.

Clin Exp Nephrol. 2011 Apr;15(2):248-57 Epub Jan 15.