心肺停止經急救復甦後心肌功能損傷之臨床評估及抗氧化劑、高二氧化碳再灌流治療之系列研究

Post-resuscitation Myocardial Dysfunction:Clinical Evaluation and Antioxidant & Hypercarbic Reperfusion Therapy

張維典(Wei-Tien Chang)

指導教授 ： 李源德 ； 陳文鍾 ； 鄭劍廷

國立臺灣大學/醫學院/臨床醫學研究所/博士(2007年)

https://doi.org/10.6342/NTU.2007.00640

摘要

猝死及心肺停止為威脅國民健康的重要議題，美國每年約有35至40萬人死於猝死，平均每一分鐘即有一人因猝死而死亡。過去台灣關於猝死的議題並無有系統的研究，一般認為此一問題在心、血管疾病盛行的西方人較易發生，東方社會的發生率應該相當低。然而吾人在台大醫學院內科李源德教授的主持下，針對台北縣金山地區居民的心臟血管疾病作有系統、前瞻性的流行病學調查 ，得到台灣第一個猝死的流病資料，發現台灣猝死的發生率為每年每十萬人口73人 (男性每年每十萬人口108人、女性每年每十萬人口43人)，與同為東方國家的日本相近。雖此結果較之美國的每年每十萬人口100~200人為低，卻不比其他與美國同為高加索人種的某些歐洲國家為低 (如英國的每年每十萬人口40人、芬蘭的每年每十萬人口48人、冰島的每年每十萬人口56人，以及以色列的每年每十萬人口46人等)。另一個由台大醫學院外科林芳郁教授及急診醫學科馬惠明副教授主持，以台北市到院前心肺停止病患為對象所作的研究，亦有類似的發現，其發生率約為每年每十萬人口53人 (男性每年每十萬人口62人、女性每年每十萬人口33人)。雖然此二研究所代表的族群以及收案對象的定義並不盡相同，然而如此相近的結果不僅勾勒出台灣地區猝死或突發性到院前心肺停止的概況，其高發生率亦提醒吾人應就其發生原因及可能的解決之道深入探討，以期在預防醫學以及急診暨急救醫學領域中投注心力，以拯救更多的病患。

到院前心肺停止病人的預後一向悲觀，不論是國內外皆然。過去台灣到院前心肺停止病人急救存活率著實偏低，以台北為例，八十二至八十三年間台北地區的到院前心肺停止病人急救後恢復自發性循環 (return of spontaneous circulation, ROSC) 的比率約 15.8%，而存活率僅 1.4%。近十年來，由於國內緊急醫療體系的建立及日趨完善，到院前心肺停止病患之急救後自體循環恢復比率已有明顯改善，台北地區於八十八年至九十年間到院前心肺停止病患的ROSC rate已可達 28.9%，然而出院存活率仍未盡理想，僅約3.1%。究其原因，有為數不少的病患於ROSC後在醫院內再次惡化而死亡，其中約三分之一乃早期死於心臟及循環衰竭，三分之一因神經學功能衰竭而死於晚期，而另三分之一則是死於感染及其他的併發症。因此，欲有效急救此類病患，除了開始急救時的 “生存之鏈 (chain of survival) ” 須緊密配合之外，對於初步急救成功病患的復甦後照顧，亦為病患能否存活出院的關鍵。此一環節目前已日益獲得重視，且被視為生存之鏈的延伸，因此急救復甦後種種病、生理學變化的研究現正蓬勃進行，期待能對此 “復甦後症候群 (post-resuscitation syndrome)” 有更深的認識，並據以發展有效的治療以改善病患預後。

急救後早期病患往往因心臟及循環衰竭而死亡，對於此，動物實驗指出此常見的復甦後心肌功能損傷實為整體性的心肌缺血暨再灌流傷害 (global myocardial ischemia/reperfusion injury)。對此主題，雖動物研究不少，但臨床上的資料仍相當有限。因此吾人應用心臟超音波技術，針對臨床上急救復甦後病患的心臟功能作一系統性評估，試圖找出影響其功能之相關急救因子，並探究其對病患的預後價值。

自2001年1月至2002年12月為止，吾人以台大醫院急診醫學部的到院前心肺停止成人病患 (大於18嵗) 為對象，經急救復甦恢復自發性心跳及血壓者，於心跳復甦後的第六小時安排心臟超音波的檢查。測量的項目包括：左心室舒張期及收縮期的腔室內徑、心室中隔厚度、以及左心室後壁厚度等。左心室的收縮功能以左心室射出分率 (LVEF) 代表，計算方式乃採用 Teichholz’s method。考慮到每位病患的身高及體重有所差異，以上所有測量值均除以病患的體表面積而以 index value 呈現。至於左心室的舒張功能則以 isovolumic relaxation time (IVRT) 來表達。相關臨床及急救因子資料的收集，包括病患的性別、年齡、過去疾病史、到院前救護資料，以及在急診的各項急救措施與用藥等。復甦後心臟功能與各項因子的關聯性吾人以Mann-Whitney / Wilcoxon test 分析。在單變項分析中達到統計明顯差別者，吾人進一步採 multiple regression model來釐清各因子的獨立性。對於存活預後的分析，吾人以 log-rank test來區分左心室功能對它的影響，並找出其他與病患存活的相關臨床因子。對於所找到者進一步則採用 Cox regression analysis 來分析各變項之獨立性。

在兩年研究期間，總共收集58位病患，其急救後心臟功能在急救後初期最差，之後方才逐漸恢復。影響急救復甦後初期左心室腔室大小與功能的臨床及急救因子列於表三及表四。如表所示，左心室之舒張末期與收縮末期指標 (LVEDDi及LVESDi) 在心因性患者以及過去病史中有缺血性心臟病和陳舊性心肌梗塞者顯著較高，LVESDi 在接受過電擊去顫及epinephrine 劑量超過5 mg者亦較高。至於左心室功能，急救復甦後左心室射出分率 (LVEF) 在具有高血壓、陳舊性心肌梗塞病史病患、急救時間超過20分鐘者、曾接受電擊去顫，以及使用epinephrine 超過5 mg者明顯較差。而左心室舒張功能IVRT則在非心因性病患、初始心律為非心室顫動/頻脈者顯著較長。

在多變項分析中，與 LVEDDi 相關的獨立因子為缺血性心臟病的病史，而與 LVESDi 有關者除了缺血性心臟病的病史外，尚包括急救中使用 epinephrine 的劑量。而與LVEF相關的獨立因子則為陳舊性心肌梗塞病史及急救中使用的 epinephrine 劑量。至於左心室舒張功能，則此次心肺停止的原因 (心因性抑或非心因性) 則為IVRT 的獨立預測因子 (表五)。

以Log-rank test分析復甦後心臟功能對病患預後的預測價值時發現，左心室射出分率大於40%者，其存活出院之機率明顯較小於40%者為佳 (P < 0.05，圖8)，而IVRT小於100 ms者之存活機會優於大於100 ms者更是顯著 (P < 0.001，圖9)。至於其他預測能否存活出院的相關因子包括：病患年齡、初始心律、急救中使用的epinephrine劑量、以及急救復甦之全部時間等。以Cox regression analysis將這些因子列入考慮分析後，IVRT仍是病患能否存活至出院之獨立因子 (hazard ratio 3.3, 95% CI 1.6 to 6.7, P = 0.002)。

至於神經學預後，神經學功能得以恢復者在LVEF大於40% 顯著較佳 (24% vs.0%，P < 0.05)，然而此因子在多變項迴歸分析中則不復存在。預測病患神經學功能恢復的獨立因子包括：初始心律為心室顫動/心室頻脈者、自心肺停止到急救開始時間小於10分鐘者、以及全部急救時間小於20分鐘者。

由以上的研究，吾人可知影響病患急救復甦後心臟功能的因素包括病患潛在的系統、心臟疾病以及諸多急救因子。其中陳舊性心肌梗塞病史固然決定於病患本身而無法改變，然而了解急救中epinephrine的累積劑量會對復甦後心室功能有不良影響，將可提醒吾人於急救中勿過度使用該藥，甚至發展新的藥物以取代之。至於急救後第六小時的心臟功能則提供了臨床醫師一個復甦後早期的預後指標，並可作為發展循環支持療法以改善預後的依據。

基於上述觀察可知病患急救復甦後的心臟功能關係著其預後，若能有效減少心肌細胞缺血暨再灌流傷害的程度，或可改善病患心臟功能及存活機會。因此，吾人從基礎研究著手，嘗試發展潛在可行之治療。因缺血暨再灌流傷害往往造成細胞的過氧化傷害，抗氧化治療為一直接有效的發展方向，因此吾人以flavonoids類抗氧化劑為對象，首先比較不同種類flavonoids間抗氧化能力的差異，並選定其中效果最佳者作深入的探討。在electron spin resonance (ESR) spectrometry實驗中，吾人以五種flavonoids (wogonin、baicalin、baicalein、catechin及procyanidin B2) 對DPPH、superoxide及hydroxyl radical的廓清能力來比較其抗氧化力的高低，結果發現catechin及procyanidin B2 對superoxide有相當好的廓清能力，而baicalein則對hydroxyl radical有最佳的廓清效果。

進一步吾人以一心肌細胞模擬缺血暨再灌流傷害的模型，比較慢性預防性投藥 (chronic treatment) 與急性治療 (I/R treatment及reperfusion treatment) 間效果的差異，並探究影響其療效的原因，以及其他可能並存的保護機轉。在比較慢性與急性治療保護效果的系列實驗中，吾人發現抗氧劑的保護效果除與其自由基廓清能力有關之外，更與治療所涵蓋的時期有密切關聯。大體而言，能充分有效涵蓋缺血及再灌流兩時期的治療，如chronic treatment及I/R treatment，其保護效果較佳，而chronic treatment則因可能事先激活細胞中的抗氧化系統及其他保護機轉，因此保護效果最佳。

然而基於臨床上的實際考量，若病患在缺血性傷害發生前未作預防性治療，則在急性期以增加抗氧化劑量的方式來增進其保護效果，或許為一可行的替代方案。因此吾人選用五種flavonoids中效果最佳的baicalein，於I/R treatment及reperfusion treatment中將劑量從25 μM增加至100 μM，結果發現均呈現一dose-response關係 (圖17-18)，似為臨床上可行的方法。惟治療時機仍是最重要的因素，因為即使於reperfusion treatment中使用100 μM的劑量，其保護效果仍不如chronic treatment中的25 μM。

抗氧化劑的再灌流治療固然可因劑量的增加而彌補其未事先預防的不足，然而若在再灌流期開始之後拖延一段時間方才施予flavonoids抗氧化治療，則其效果將隨時間的延遲而大幅將低。吾人採用上述baicalein 的reperfusion treatment中效果較佳的 100 μM劑量，分別於再灌流開始後0、15、30、60及120分鐘投予，結果發現隨著時間的延遲，baicalein對心肌細胞的保護效果逐步下降。事實上，若baicalein延至再灌流一小時之後才給予，則完全喪失其保護作用 (圖19)。

至於探討baicalein保護機轉的研究中，吾人發現baicalein在心肌細胞內確實能廓清缺血及再灌流過程中所產生的ROS：chronic treatment及I/R treatment二者自缺血時期開始，便能使DCF fluorescence大幅降低，而reperfusion treatment雖對缺血期的ROS沒有影響，但確實可降低再灌流初期的ROS burst (圖24)。至於若將reperfusion treatment中的劑量由25 μM提高至100 μM，則將能使ROS burst進一步地降低。然而，若reperfusion treatment因故延遲，即使只有15分鐘，其亦將錯過再灌流初期的ROS burst (圖25) 而喪失抗氧化保護的效果 (圖19)。

Nitric oxide 在細胞的病、生理學中扮演的重要的角色，許多對缺血暨再灌流傷害中有保護效果的治療 (如preconditioning、hypothermia等) 係透過nitric oxide相關的pathway達到保護的目的。吾人在baicalein的保護機轉研究中，亦使用DAF-2DA作為nitrix oxide的指標，探討nitric oxide在baicalein保護作用中的潛在角色。結果發現chronic treatment及I/R treatment皆可於再灌流的稍後階段 (15分鐘後) 造成nitric oxide 持平的升高。而reperfusion treatment若將劑量由25 μM提高至100 μM，亦可於在再灌流的30-45分鐘後造成nitric oxide的上升。由於上述具有保護效果的治療protocols其nitric oxide profile均與再灌流稍後階段nitric oxide的上升有關，因此baicalein對心肌細胞缺血暨再灌流傷害的保護，至少有部分乃是透過nitric oxide相關的pathway達成。

至於baicalein所造成的nitric oxide上升是否是透過nitric oxide synthase的激活所達成，吾人於baicalein (100 μM) 的reperfusion treatment 中加入nitric oxide synthase的抑制劑L-NAME (200 μM)，來觀察其對nitric oxide profile及保護效果的影響。其結果如圖28所示，加了L-NAME的組別其DAF fluorescence於再灌流的稍後階段的確較未加L-NAME的Baicalein reperfusion treatment組顯著為低 (P < 0.01)，表示baicalein所造成的nitric oxide增加起碼有部分乃是來自於nitric oxide synthase的增加。再者，加了L-NAME (200 μM) 的baicalein reperfusion treatment組其在缺血暨再灌流傷害之後的細胞死亡率 (44.8 ± 1.0%) 亦較未加L-NAME組 (36.4 ± 1.5%) 顯著地高 (P < 0.05；圖29)，此進一步說明了baicalein對心肌細胞缺血暨再灌流傷害的保護作用，起碼有部分是透過nitric oxide synthase的活化，造成再灌流期稍後階段nitric oxide的增加，進一步引起下游的保護機轉而減少心肌細胞的損傷。

細胞凋亡在缺血暨再灌流傷害所造成的細胞死亡中扮演著重要的角色。Baicalein對缺血暨再灌流傷害的保護究竟是否透過減少細胞凋亡來達成，將是一個重要的議題。吾人首先以外源性H2O2過氧化傷害的模型來驗證baicalein的保護效果。其結果如圖20及21可見，baicalein顯著地減少了H2O2所造成的DNA fragmentation，說明baicalein乃是透過減少細胞凋亡保護細胞免受過氧化傷害。同樣地，在心肌細胞的缺血暨再灌流模型中，baicalein亦有效地減少了缺血暨再灌流所造成的DNA laddering，說明了baicalein對心肌細胞的缺血暨再灌流傷害的保護，也是透過減少細胞凋亡達成。

除了藥物方式的治療之外 (如抗氧化劑) ，對於心臟缺血暨再灌流傷害的治療事實上還有許多方式可以考量。在這些方法中，改變再灌流階段氣體的成分 (如二氧化碳濃度，以及隨之而改變的pH值等)，為一方便且實際可行的方式。因此，吾人比較於再灌流階段給予不同二氧化碳濃度的氣體通氣，觀察其對心肌細胞缺血暨再灌流傷害程度的影響。

吾人使用的高二氧化碳灌流氣體為 (hypercarbic reperfusion, 21% O2, 10% CO2, and 69% N2)，低二氧化碳灌流氣體則為 (hypocarbic reperfusion, 21% O2, 1% CO2, and 78% N2)，二者再與標準的I/R control (normocarbic reperfusion, 21% O2, 5% CO2, 74% N2) 相比。其結果發現，低二氧化碳再灌流將造成細胞死亡率大增 (80.4 ± 4.5%，與I/R control的54.8 ± 4.0 % 相較 P < 0.01)，且此細胞死亡之增加與其造成reperfusion ROS burst的增加有關，且此ROS的增加係來自粒腺體的電子傳遞鏈，因為同時給予粒腺體電子傳遞鏈complex III的抑制劑stigmatellin (20 nM)，將可使低二氧化碳再灌流所造成的ROS增加減少，並且減少低二氧化碳再灌流所造成的高細胞死亡率。

另一方面，高二氧化碳再灌流則可使細胞死亡率減少 (26.3 ± 2.8%，與I/R control的54.8 ± 4.0 % 相較 P < 0.001)。其保護的機轉除了與reperfusion ROS burst的減少有關外，亦與其在再灌流期的稍後階段造成nitric oxide的持續增加有關。而此一nitric oxide的增加，類似baicalein於再灌流階段所造成的結果，也和nitric oxide synthase的激活有關，因為nitric oxide synthase抑制劑L-NAME (200 μM)不僅使NO 下降，亦造成hypercarbic reperfusion的保護效果喪失。因此高二氧化碳再灌流對心肌細胞的保護係透過減少再灌流階段的ROS burst，以及nitric oxide synthase媒介的nitric oxide增加來達成。

基於前述臨床的觀察，吾人確知急救復甦後心臟功能對病患預後之重要性，同時找出可能影響此心肌功能損傷的相關因子 (如高劑量的epinephrine等)，對於日後發展替代藥物治療有其潛在價值。另一方面，在基礎層級的研究中，吾人找到一有效的抗氧化治療，具備應用於心肺停止病患急救治療之潛力，同時深入探究了其對心肌細胞缺血暨再灌流傷害的相關保護機轉，發現治療時機對保護機轉及治療效果的關鍵影響。這些研究的發現將可作為轉譯醫學發展的基礎，期待由動物層級的實驗，逐步推展至臨床前期試驗，經由類似第一部分的心臟功能研究，驗證抗氧化治療對復甦後心肌功能損傷及預後之改善，以期進一步進展至大規模的臨床試驗，有朝一日能夠實際應用於臨床病患，以改善病人的預後，達到此系列研究的最終目的。

關鍵字

心肺停止 ； 心肺復甦 ； 急救 ； 心臟衰竭 ； 預後 ； 缺血 ； 再灌流傷害 ； 抗氧化劑 ； 二氧化碳

並列摘要

Ｓudden death has been an issue of significant importance in public health. In the United States, more than 400,000 people die outside the hospitol before they have a chance to reach the hospital and receive medical help. Though generally believed to be low, the incidence of sudden death in Taiwan is comparable to those of the western countries. According to the Chin-Shan Community Cardiovascular Cohort (CCCC) study, a population-based epidemiological study held by Prof. Yuan-Teh Lee in National Taiwan University, the incidence of sudden death in Taiwan is 73/100,000 person-year. The incidence is higher in man (108/100,000 person-year), about 2.5 times that of the woman (43/100,000 person-year). Though lower than that of the United States, it is higher than some of the European countries (e.g. 40/100,000 person-year in England, 46100,000 person-year in Israel, 48/100,000 person-year in Finland, and 56/100,000 person-year in Iceland), and is comparable with that of Japan (68-104/100,000 person-year). Therefore, this issue is worthy of extensive exploration in the oriental societies. Effort should be launched to maximize the primary prevention in the general population and strengthen the emergency cardiac care in the communities.

The prognosis of out-of-hospital cardiac arrest (OHCA) is generally poor. For example, the rate of return of spontaneous circulation (ROSC) in Taipei was 15.8% in 1993, and only 1.4% survived to hospital discharge.With the progress in emergency medical service system for the last 10 years, the ROSC rate in Taipei has improved to 28.9%. However, the survival to discharge rate was still low (3.1%). The discripency between initial ROSC and survival to discharge has been largely attributed to the gradual decline of survival in the post-resuscitation phase. Considering the cause of death in this post-CPR period, one-third die of cardiac or circulatory failure, one-third of neurological failure, and the rest die of other complications such as infection. Circulatory failure usually occurs early, and is responsible for early mortality. In contrast, neurological failure is associated with prolonged ICU stay, and accounts for most of the late mortalities. Both these consistute the main features of the post-resuscitation syndrome.

Myocardial failure usually develops early in the post- resuscitation phase, and has been considered the main cause of early death after initial success in CPR. For the past 10 years, post-resuscitation myocardial dysfunction has attracted keen attention. Resutling from cardiac arrest and CPR, it has been regarded as a specific model of global myocardial ischemia and reperfusion. A lot of studies have been done at the animal level. Experimental studies have implicated a number of factors that might influence its severity, such as the ischemic duration, electrical defibrillation, the number, type, and waveforms of electrical shocks, and the dose of epinephrine. Efforts have also been made to improve such dysfunction via pharmacological or mechanical means. In the real world, however, there are still few studies focusing on characterizing the post-resuscitation myocardial dysfunction in human. Therefore, we launched a prospective observation study using echocardiographic evaluation of the myocardial function in patients suffering from cardiac arrest and CPR, hopefully to identify the factors associated with its severity, as well as the prognostic implications it may have.

Between Jan 2001 and Dec 2002, OHCA patients receiving CPR at the emergency department of National Taiwan University Hospital were prospectively followed up. Those who achieved ROSC were admitted to ICU for post-resuscitation care. Echocardiographic evaluation was performed at the sixth hour post-CPR. The left ventricular (LV) size and systolic and diastolic functions were analyzed in correlation to the patients’ underlying diseases and resuscitation factors. The prognostic value of the LV function was also evaluated.

During the study period, 58 patients were included. The basic characteristics of the patients were listed in Table 2. The associations between clinical factors and post-resuscitation LV dimensions and functions were shown in Table 3. Both LV end-diastolic (LVEDDi) and end-systolic diameter index values (LVESDi) were significantly higher in patients with cardiac etiology and past histories of ischemic heart disease (IHD) and myocardial infarction (MI). The LVESDi was also higher in patients receiving defibrillation and epinephrine ≧ 5 mg. For LV systolic function, poorer LV ejection fraction (LVEF) was associated with hypertension, past history of MI, CPR duration ≧ 20 minutes, defibrillation, and the use of epinephrine ≧ 5 mg. For LV diastolic function, isovolumic relaxation time (IVRT) was significantly longer in patients with non-cardiac etiology and initial rhythm of non-VF/VT. For patients with myocardial ischemia/infarction as the etiology of cardiac arrest, no significant differences were noted in LV dimensions or functions compared to other cardiogenic patients without evident myocardial ischemia.

In multiple regression analysis, past history of IHD was significantly associated with LVEDDi, while epinephrine dose and history of IHD were independent factors for LVESDi (Table 5). For LV systolic function, epinephrine dose and past history of MI were significantly associated with LVEF. For LV diastolic function, the etiology of cardiac arrest was independently associated with IVRT (Table 5).

For prognosis, both LVEF and IVRT were significantly associated with the patients’ survival outcomes (p = 0.03 and 0.0004, respectively). The results were demonstrated by Kaplan-Meier survival curves in Figure 8 and 9. Other factors associated with survival included age, initial cardiac rhythm, epinephrine dose, and total CPR duration. After adjusting for these variables in Cox regression analysis, IVRT ≧ 100 ms still served as an independent predictor for poor survival prognosis (hazard ratio 3.3, 95% CI 1.6 to 6.7, p=0.002).

Neurological recovery was significantly better in patients with LVEF ≧ 40% (24% vs. 0%, p = 0.04), but this significance no longer existed in multiple regression analysis. Factors independently predicting neurological recovery were initial rhythm of VF/VT, duration of no CPR < 10 minutes, and total CPR time < 20 minutes.

According to this study, post-resuscitation LV dysfunction is correlated with a number of clinical factors, among which past history of MI, epinephrine dose, and the etiology of cardiac arrest play independent roles. These results implicate that post-resuscitation myocardial dysfunction may be a combination of the patients’ underlying diseases and the injury resulting from cardiac arrest and CPR. While some of these factors are non-modifiable, certain resuscitation factors can be avoided or substituted by newly developed durgs. For prognosis, both LVEF < 40% and IVRT ≧ 100 ms at the sixth hour post-CPR were associated with poor survival outcomes. This is of prognostic value based on which treatment can be done trying to improve the myocardial function as well as the survival prognosis.

As mentioned above, post-resuscitation myocardial dysfunction is a specific model of global myocardial ischemia/reperfusion (I/R) injury. Given the fact that LV dysfunction in the post-resuscitation phase implicates poor survival outcomes, therapeutic interventions trying to limit or treat such I/R injury become the key to eventually improve myocardial function and prognosis. As I/R is inevitably associated with oxidative stress, and the reactive oxygen species (ROS) produced during I/R can cause damages on lipid, protein and DNA, one reasonable intervention to attenuate I/R injury may be antioxidant therapy. Therefore, we sought to investigate the role of antioxidant therapy in mitigating the myocardial injury resulting from I/R.

Flavonoids are naturally occurring polyphenolic compounds exhibiting potent antioxidant activities. Their roles in combating oxidative injuries such as reperfusion of the ischemic tissue have been investigated in a variety of in vitro and in vivo models. Typically they are pre-administered as preventive measures for evaluating their protection against an up-coming oxidative insult. However, it may not always be feasible to administer flavonoids as a pretreatment in the clinical settings. Therefore, the protective role of flavonoids in an acute treatment model of I/R, i.e. administration of flavonoids concurrently with I/R, or just at the point of reperfusion, is of greater interest.

In this series of study, we first employed electron spin resonance (ESR) spectrometry to evaluate the oxidant scavenging capacities of five selected flavonoids. The oxidants chosen for ESR experiments are superoxide and hydroxyl radical, both being the major ROS produced during the context of I/R. As shown in Fig. 11, addition of the tested flavonoids to xanthine/xanthine oxidase plus DMPO system (25 µM) reduced the DMPO-.OOH signals to different levels compared to that in control. The superoxide scavenging capacity of the five flavonoids was as follows: catechin (90 ± 3%), baicalein (69 ± 5%), procyanidin B2 (68 ± 5%), baicalin (42 ± 4%), and wogonin (8 ± 3%). The hydroxyl radical scavenging effect of the flavonoids was tested using a Fenton reaction system with DMPO as the trapping agent. The addition of flavonoids (200 µM) resulted in differential quenching effects of the DMPO-OH adduct as demonstrated by reduction in the ESR signal intensity (Fig. 12). In general, the scavenging effect of the flavonoids on hydroxyl radicals was weaker than that on superoxide even with higher concentration (200 µM): baicalein (56 ± 4%), catechin (31 ± 2%), procyanidin B2 (28 ± 3%), wogonin (25 ± 5%), and baicalin (15 ± 3%). Only baicalein showed an ESR reduction rate of > 50%.

Then we developed three treatment strategies in an established cardiomyocytes model of I/R to evaluate the antioxidant protection of the flavonoids against I/R injury. The protective effects were compared with the oxidant scavenging capacities as well as among different treatment protocols. As shown in Fig. 13, 14 and 15, the protective effects of flavonoids were generally consistent with the flavonoids’oxidant scavenging activities. Meanwhile, chronic treatment exhibited the best protective effects, while in reperfusion treatment only limited protective effects can be seen. The only exception was catechin, whose antioxidant protection against I/R injury seemed less satisfactory given the outstanding oxidant scavenging activities it demonstrated in ESR esperiments. This can probably be explained by the poor lipid solubility of catechin, which may be an important factor when considering cell membrane access and antioxidant activities within the cell.

In the second part of the study, we chose the flavonoids with the best antioxidant protection against I/R injury – baicalein – for further study and mechanistic exploration. Since our interest has been the role of acute treatment, we first tried to raise the dose of baicalein in acute treatment protocols to see if the protective effect can be enhanced. As shown in Fig. 17 and 18, as the dose of baicalein was raised from 25 μM to 100 μM, the cell death rates after I/R declined in a reverse dose-response manner, both in I/R treatment and reperfusion treatment. This suggests that if chronic prevention cannot be available in face of an acute ischemic insult, raising therapeutic dose for optimizing protective effects may be a practical approach for maximizing antioxidant protection against I/R injury.

Furthermore, as reperfusion ROS burst occurs within minutes of reperfusion, and is associated with tremendous oxidative reperfusion injury thereafter, we employed a delayed reperfusion treatment protocol with different therapeutic time delay in order to justify the critical time window for antioxidant therapy. As shown in Fig. 19, delay of baicalein treatment in the reperfusion phase resulted in attenuation of protection against I/R injury. The protection was completely lost if baicalein was not given until after one hour of reperfusion. This suggests that the therapeutic window for acute antioxidant therapy is within the first 15-30 of reperfusion.

In this cardiomyocyte model of I/R, ROS generation was increased during ischemia, followed by a transient but significant ROS burst in the first minutes of reperfusion (Fig. 23). Given that baicalein exhibits antioxidant activities protecting against oxidant injuries, we sought to compare the ROS scavenging capacities of the different treatment protocols, and their correlations with the protective effects against I/R injury. As shown in Fig. 24, either chronic treatment or I/R treatment (25 μM) significantly abolished the reperfusion ROS burst. Moreover, the DCF fluorescence increase during ischemia was abolished as well. In contrast, in reperfusion treatment the ROS generated during ischemia was not affected at all, and the scavenging of reperfusion ROS burst was less effective. These results were basically consistent with the protective effects seen in the corresponding protocols.

Likewise, when the baicalein dose was raised from 25 to 100 μM in reperfusion treatment, the reduction of reperfusion ROS burst was also augmented (Fig. 25), which is consistent with the enhanced protective effect (Fig. 18). Nevertheless, if the same effective dose (100 μM) of baicalein treatment was delayed by 15 min into reperfusion, this effect on the reperfusion ROS burst was completely lost (Fig. 25), and the protective effect was diminished (Fig. 19). This again supports the notion that the ROS burst in the first min of reperfusion is important in mediating I/R injury. Acute antioxidant therapy effectively covering this critical window of reperfusion ROS burst is essential for oxidant scavenging and protection against I/R injury.

Nitric oxide (NO) is implicated in a number of cardioprotective mechanisms, and has been demonstrated in this cardiomyocyte model mediating cytoprotection in treatment strategies such as preconditioning and hypothermia. To elavuate whether NO, in addition to oxidant scavenging, is involved in baicalein’s protection against I/R injury, we employed DAF-2DA as an indicator of NO and monitored the NO profile of baicalein treatment during I/R. As seen in Fig. 26, in effective treatment strategies such as chronic and I/R treatments (25 μM), there was a sustained increase of NO after the peaking at around 15 min of reperfusion. In contrast, this cannot be seen in less satisfactory treatment protocol such as the reperfusion treatment (25 μM). However, if the baicalein dose was increased from 25 to 100 μM in reperfusion treatment, the NO increased gradually at a later phase of reperfusion (Fig. 27), which may in part accounts for the enhanced protection. Nevertheless, if baicalein treatment was delayed by 15 min at reperfusion, the increase of NO was further delayed and the magnitude was less. These results altogether suggest that NO increase at a later phase of reperfusion may be associated with baicalein’s protection against I/R injury.

As NO synthase has been implicated in a number of protective mechanisms, we further tested if the NO increase seen in baicalein treatment was related to activation of the NO synthase. As shown in Fig. 28, co-administration of NO synthase inhibitor L-NAME (200 μM) with baicalein reperfusion treatment (100 μM) resulted in attenuation of the NO increase. Meanwhile, the protective effects was also partially reversed (Fig. 29). This suggests that baicalein’s protection against I/R injury was, at least in part, attributed to NO synthase-mediated NO increase at the later phase of reperfusion.

As apoptosis plays a significant role in I/R related cell death, we finally evaluated whether baicalein protects by attenuating apoptotic cell death along the course of I/R. As shown in Fig. 20 and 21, baicalein significantly reduced apoptotic cell death induced by exogenous oxidant injury caused by H2O2. Similarly, in endogenous oxidant injury such as that induced by I/R, the DNA fragmentation was also significantly diminished with the treatment of baicalein (Fig. 22), suggesting that baicalein exhibits antiapoptotic protection against I/R injury.

In addition to pharmacological interventions such as antioxidant therapy, there are other methods that can be potentially employed in the compaign of I/R injury. Among these, chemical factors such as CO2 modification, and hence the pH, at the phase of reperfusion may be one that is effective and practically feasible. This is especially true when recent studies suggest that rapid normalization of pH at reperfusion may be associated with increased injury (the pH paradox).

In this series of study, we compared hypercarbic (21% O2, 10% CO2, and 69% N2) and hypocarbic (21% O2, 1% CO2, and 78% N2) reperfusion with normocarbic reperfusion (21% O2, 5% CO2, 74% N2) in I/R control, and see if modification of reperfusion CO2 can have impact on reperfusion injury. As shown in Fig. 30, the cell death seen during the 1 hour ischemia prior to hypocarbic, normocarbic or hypercarbic reperfusion was not significantly different. Following ischemia, percent cell death in hypocarbic reperfusion was significantly higher (80.4 ± 4.5%) compared to normocarbic reperfusion (54.8 ± 4.0 %, P < 0.01). By contrast, hypercarbic reperfusion resulted in significantly lower cell death (26.3 ± 2.8%, P < 0.001 vs. normocarbic reperfusion).

When comparing the ROS profile among the three protocols, we found that hypocarbic reperfusion was associated with increased ROS burst in the first 10 min of reperfusion, in contrast to reduction of the ROS burst by hypercarbic reperfusion (Fig. 31). The increased ROS induced by hypocarbia was mitochondria in origin since addition of mitochondrial complex III ihibitor stigmatellin (20 nM) significantly decreased the ROS (Fig. 32) and cell death (Fig. 33). On the other hand, in addition to attenuation of the reperfusion ROS burst, hypercarbic reperfusion also resulted in a sustained increase of NO at the later phase of reperfusion, a phenomenon similar to those seen in baicalein (Fig. 26, 27) and hypothermia (Fig. 40). This increase was also mediated by NO synthase activation sicne NO synthase inhibitor L-ANME (200 μM) significantly attenuated such NO increase (Fig.38) as well as the protection (Fig. 39).
Therefore, hypercarbic protects cardiomyocytes from I/R injury not only by attenuation of the reperfusion ROS burst, but also activation of NO synthase leading to sustained increase of NO at the later phase of reperfusion.

In conclusion, post-resuscitation myocardial dysfunction is associated with a number of clinical and resuscitation factors, and may impact the patient’s prognosis by influencing hymodynamics and organ perfusion in the early post-resuscitation phase. Identification of these factors may help understanding the pathophysiological mechanism involved in post-resuscitation myocardial dysfunction. It may also help avoiding the potentially harmful resuscitation measures such as unnecessary epinephrine doses. On the other hand, development of therapeutic strategies that can reduce I/R injury and augment myocardial recovery may help improve the post-resuscitation myocardial dysfunction and, possibly, prognosis. Though exciting results have been obtained from the basic research at the cell level, further work is needed while trying to translate these into clinical practice. With the establishment of the animal (rat) model of cardiac arrest and resuscitation, application of hypercarbic reperfusion or baicalein treatment for assessing the protective effects in the in-vivo model may offer valuable data and serve as a solid base when further translating these results to clinical use.

並列關鍵字

Cardiac arrest ； cardiopulmonary resuscitation ； resuscitation ； myocardial dysfunction ； prognosis ； ischemia ； reperfusion injury ； reactive oxygen species ； antioxidant ； carbon dioxide

參考文獻

Afanas'ev, I. B., Dorozhko, A. I., Brodskii, A. V., Kostyuk, V. A., and Potapovitch, A. I. Chelating and free radical scavenging mechanisms of inhibitory action of rutin and quercetin in lipid peroxidation. Biochem Pharmacol 1989;38:1763-1769.

Aufderheide, T. P., and Lurie, K. G. Death by hyperventilation: a common and life-threatening problem during cardiopulmonary resuscitation. Crit Care Med 2004;32:S345-351.

Becker, L. B. New concepts in reactive oxygen species and cardiovascular reperfusion physiology. Cardiovasc Res 2004;61:461-470.

Becker, L. B., vanden Hoek, T. L., Shao, Z. H., Li, C. Q., and Schumacker, P. T. Generation of superoxide in cardiomyocytes during ischemia before reperfusion. Am J Physiol 1999;277:H2240-2246.

Berg, R. A., Otto, C. W., Kern, K. B., Sanders, A. B., Hilwig, R. W., Hansen, K. K., and Ewy, G. A. High-dose epinephrine results in greater early mortality after resuscitation from prolonged cardiac arrest in pigs: a prospective, randomized study. Crit Care Med 1994;22:282-290.