胰島素控制腦損傷患者血糖之統合分析(實證醫學)

Meta-analysis of insulin for glycemic control in brain injury (Evidence-Based Medicine)

蕭如君(Ju-Chun Hsiao)

指導教授 ： 蔡崇弘

中山醫學大學/醫學院/醫學研究所/碩士(2011年)

https://doi.org/10.6834/CSMU.2011.00097

摘要

研究背景:人體遭遇危急狀態時會產生自發性的壓力性高血糖表現。目前的研究已經證實高血糖與增加死亡率及導致預後變差有密切相關性，也可能影響神經功能復原情形。在2001年及2006年由Van den Berghe等人所領導的大型隨機對照試驗中，證實嚴格的使用胰島素來控制危急病人血糖於≦110mg/dl對降低死亡率及減低罹病率是有好處的。但在這些試驗中鮮少有人特別去分析急性腦損傷(acute brain injury)的病人，考量人體腦部的最主要能量來源就是葡萄糖，一旦發生中風或嚴重腦損傷的情形，葡萄糖在血腦障壁及神經細胞膜的運輸機制就會改變，導致無法供應正常的細胞代謝。在急性中風病人身上有40~70%會出現高血糖的表現，但目前對急性腦損傷病人的血糖控制值仍沒有定論。
研究目的:探討以胰島素控制急性腦損傷患者血糖值控制範圍之相關研究。
研究方法: 本研究利用統合分析的方法，於Pubmed,Medline, Cochrane等資料庫中搜尋大量相關隨機對照試驗。設定納入的標準為:急性腦損傷以胰島素控制血糖於正常值的相關隨機對照試驗，研究設計為:以胰島素嚴格控制患者血糖值於 ≦120mg/dl至少24小時的實驗組與對照組(安慰劑或傳統治療)兩組之間在死亡率、感染率、神經復原情形，以及發生低血糖事件的比率。
研究結果:共納入14個隨機對照試驗，整合同質性的研究數據，運用RevMan軟體得出統計的結果，整合發現實驗組設定的血糖值控制在≦130 mg/dl，與對照組相比OR:1.01(95%CI 0.84~1.21)，在死亡率來說是沒有差別的。實驗組設定血糖值控制在≦120 mg /dl，與對照組相比OR:0.95 (95%CI 0.74 ~1.22)，在死亡率也同樣沒有差別。感染率的分析顯示血糖值控制在≦120mg/dl相較於對照組OR:0.81 (95% CI 0.55~1.19)，就感染率而言也沒有差別。神經復原療效評估項目則分別比較GOS及 mRS分數的變化，發現兩組也沒有差別。而發生低血糖事件的比率，嚴格的以胰島素來控制血糖組發生低血糖的風險(血糖值≦80 mg/dl)是傳統治療組的7.76倍(95% CI 5.14~ 11.70)。
結論與建議:本研究發現在急性腦損傷病人，積極控制血糖/以胰島素來介入治療(血糖值控制在≦120 mg /dl)，以目前的研究證據得知其臨床效益目前仍無法證實，反而相對提高發生低血糖的風險。目前對腦損傷患者血糖值，依據美國中風協會建議控制在≦140 mg/dl應是最適當的選擇。

關鍵字

實證醫學 ； 統合分析 ； 中風 ； 腦損傷 ； 高血糖 ； 胰島素

並列摘要

Background: The human body will produce spontaneous stress hyperglycemia while encountering the critical state, even in nondiabetes patients also will present such condition. Present research has already proved hyperglycemia has been close linked with increasing mortality and causing the prognosis worsen, may influence the recovery of neurologic functional outcome,too. The large randomized control trials led by Van den Berghe,et al. in 2001 and 2006, have proved that intensive insulin therapy to control blood glucose level in normal range(≦110 mg/ dl) of patients in intensive care units had benefit in reduced the mortality and lower morbidity. Human brain is dependent on a steady supply of glucose to maintain normal energy metabolism.Transport of glucose at the blood– brain barrier(B.B.B.) and the neuronal plasma membrane may become inadequate to satisfy brain cellular metabolism after acute brain injury.There is 40~70% will present stress hyperglycemia after acute brain injury.But still does not have a conclusion about controlling patient's blood glucose level after acute brain injury at present.
Objective : The subject of the relevant research of intensive insulin therapy to control the blood glucose level after acute brain injury .
Methods and Materials : This study use the method of meta-analysis to search a large amount of relevantly clinical trials from Pubmed, Medline, Cochrane databases.This study set up the included criteria is relevantly random controlled trials to investigate intensive insulin therapy to control patient’s blood glucose level in normal range after acute brain injury . Study design was : intensive insulin therapy to matain patient’s blood glucose level on≦120mg/dl for above 24 hours and the control group (placebo or conventional insulin therapy) to compare the mortality、 infection rate、functional outcome and the rate of the hypoglycemia incident between two groups.
Results : After screening the relevantly random controlled trials in high quality after critical appraisal studies, include 14 random controlled trials finally.Combined the homogeneity study datas, use the RevMan software to obtain the results.Through counting the data analysis, combine result data this study find experimental group that sets up maintain blood glucose level ≦130 mg/dl, compared with control group OR: 1.01(95% CI 0.84~1.21) had no difference in mortality. And experimental group that sets up maintain blood glucose level ≦120 mg/dl, compared with control group OR: 0.95 (95%CI 0.74 ~1.22) had no difference in mortality,too. In the infection rate this study find that sets up maintain blood glucose level ≦120 mg/dl, compared with control group OR: 0.81(95% CI 0.55~1.19) had no difference in mortality,too. Assess the functional outcomes compare GOS and mRS scores this study find that no difference between two groups. But the rate of hypoglycemia incident (blood glucose level≦ 80 mg/dl), in strict insulin therapy group had 7.76 times (95% CI 5.141.70) compared with control group .
Conclusion and Suggestion : In this study conclusion for acute brain injury, control blood glucose level strictly / intervention with insulin (maintain blood glucose in ≦120 mg /dl),the clinical benefit is still unable to prove at present with the present research evidence, and increased the risk of the hypoglycemia incidence relatively. According to American Stroke Association,it should be the most appropriate to control blood glucose level ≦140 mg/dl in brain injury patient at present.

並列關鍵字

evidence-based medicine ； meta-analysis ； stroke ； brain injury ； hyperglycemia ； insulin

參考文獻

35. 盧美秀. 實證醫學、實證護理與實證管理. 新台北護理期刊. 2001;3:1-3.

49. 宋惠娟、張淑敏. 臨床決策： 實證實務的步驟. 慈濟護理雜誌. 2006;5:73-80.

52. 陳杰峰、邱文達、曾珮娟. IOM五大核心能力∼基於實證醫學的專業執行. 醫療品質雜誌. 2007;1:32-40.

57. 中央健康保險局, 財團法人國家衛生研究院, 財團法人醫院評鑑暨醫療品質策進會. 臨床診療指引發展手冊. 2004.

the medical curriculum. BMJ. 2008;337:a1253.