動脈瘤破裂蜘蛛膜下腔出血後腦脊髓液生化指標預測慢性水腦症與預後

Abstract

研究背景與目的 自發性蜘蛛膜下腔出血的主要原因為動脈瘤破裂導致血液流入蜘蛛膜下腔。而腦部皮質血管就浸泡在腦脊髓液中，正常腦脊髓液循環會經過第四腦室的三個小孔流入蜘蛛膜下腔；一旦大量蜘蛛膜下腔出血，腦脊髓液的通道雖在結構上並未阻塞，但在在功能上會被區隔成兩區，新鮮的腦脊髓液由腦室中的脈絡叢製造，但蜘蛛膜下腔因含有大量血塊及其代謝產物，會變得相當濃稠，因此大量蜘蛛膜下腔出血後可能會使腦室與蜘蛛膜下腔產生隔間(compartmentalization)的現象。一旦如同我們的假說有隔間現象產生，蜘蛛膜下腔的血塊代謝與發炎反應必定遠大於腦室。根據我們的假說，在大量蜘蛛膜下腔出血後產生分隔現象，乳酸將會在蜘蛛膜下腔堆積，尤其是在水腦症產生的狀況下。因此第一個研究的目的即是觀測動脈瘤破裂併發modified Fisher Grade III 與 IV 蜘蛛膜下出血的病人，其腦室與腰椎腦脊髓液的生化差異，並分析乳酸與水腦症的相關性。 在蜘蛛膜下腔出血後，紅血球會被分解而放出hemoglobin，接下來進一步被代謝成heme與鐵。游離的鐵離子對於腦細胞有毒，需與蛋白質結合成ferritin。在這個過程中，一個可被誘發的酵素稱為heme oxygenase (HO) 1，是hemoglobin代謝過程速率決定步驟。 先前許多報告顯示heme代謝的許多產物與頭部外傷或蜘蛛膜下出血的預後相關， 因此第二個研究目的在探討紅血球代謝過程的各項產物與預後的關係。 研究方法: 我們第一步先分析蜘蛛膜下腔出血後，腦室與腰椎腦脊髓液中各項生化指標的差異；因此我們同時收集出血後第七天腦室與腰椎腦脊髓液，進一步利用多變項分析探討蜘蛛膜下腔出血後造成慢性水腦症的相關因子。 第二個研究則接續於第一個研究的結果，於第七天藉由腰椎穿刺取得腦脊髓液，測量了紅血球代謝過程的各種酵素與產物，包括HO1，oxyhemoglobin，ferritin與 bilirubin；並研究這些生化指標與病人出血後三個月的神經功能的關係。 研究結果 第一個研究顯示，腦室內與髓內腦脊髓液，其中蛋白質、鐵蛋白、紅球素與乳酸在脊髓內腦脊髓液接顯著高於腦室內(p<0.001 in each factor)。所有單變項顯著因子進行多變項分析，結果顯示有腦室內出血(IVH)(odds ratio [OR]: 18.0; 95% confidence interval [95%CI]: 1.0-325.5; p=0.05] 與脊髓內腦脊髓液乳酸濃度(OR: 4.8; 95%CI: 1.1-20.4; p=0.036)顯著與引流管依賴水腦症SDHC相關。再進一步，藉由GMA模式，脊椎內腦脊髓液乳酸lactate ≥5.5 mmol/L時會增加32倍產生引流管依賴水腦症SDHC 的機會(95%CI: 3.8-270.8; p=0.0015)。 第二個研究，有關腦脊髓液中的數據，單變量分析顯示，造成病人結果不好的因子包括HO1，氧合血紅蛋白，膽紅素，鐵蛋白，和乳酸，較高的白細胞計數與較低的淋巴細胞比例。在調整了年齡，WFNS臨床參數（≥或<3），急性腦積水和腦室內出血後，較高的HO1與不好的結果相關(odds ratio=0.920; 0.850–0.995, p=0.038)。此外，藉由GAM所取得的臨界點，H0 1>81.2對於病患有不利的結果（64.7％的敏感性，特異性為100％，100％，陽性預測值為80.0％，陰性預測值82.4％）。 結果與展望 蜘蛛膜下腔出血後，一旦產生水腦症，腰椎腦脊髓液與腦室腦脊髓液會有隔間(compartalization)的情況，且在此狀況下，乳酸會快速堆積，尤其在水腦症患者，此種情形必定會加劇，第二個研究進一步證實在出血後血塊會開始代謝，過程中HO1的濃度與預後有明顯關聯。 既然出血後腰椎腦脊髓液與腦室腦脊髓液會有隔間(compartalization)的情況，那麼如何能將這些物質加快代謝，就有可能改善病人癒後。參考文獻與我們的假設，腰椎引流雖然無法完全將蜘蛛膜下腔的血塊洗出，但相較於腦室外引流，腰椎引流應更有機會排出溶解的血塊，甚至可以減少造成血管痙攣的物質。另外針對延遲性缺血性神經病變的探討，我們發展一個新的觀測方法，可活體觀測蜘蛛膜下腔出血後老鼠的微血流，並同時監測腦局部血流與氧分壓，我們發現蜘蛛膜下腔出血後微血流有明顯痙攣，且伴隨局部血流與氧分壓降低；這樣的結果與最近的一些研究相當吻合，大血管的痙攣的治療並不符合預期，也許微血流與微血栓對於延遲性缺血性腦病變有更大的影響。

Background and objective Spontaneous subarachnoid hemorrhage (SAH) is most frequently caused by aneurysmal rupture. The cortical vessels were immersed in cerebrospinal fluid. In the normal cerebrospinal fluid circulation, CSF flow through three holes of the fourth ventricle into subarachnoid space; once a large number of subarachnoid hemorrhage, as shown, cerebral spinal fluid, although anatomically not blocked, but the functionally may be separated, fresh cerebrospinal fluid in the ventricles can’t flush into subarachnoid space. Thus, subarachnoid clots and its metabolites, will become high viscosity. Our hypothesis is that compartmentalization happened if hydrocephalus presented. Thus, blood clots and inflammation metabolites in subarachnoid space certainly are much higher than the ventricle. According to our hypothesis, after subarachnoid hemorrhage, lactic acid will accumulate in the subarachnoid space, especially under conditions resulting in hydrocephalus. Therefore, the aim of the first study was to analyze its biochemical differences between ventricle and lumbar cerebrospinal fluid, in patients with modified Fisher's Grade III and IV subarachnoid hemorrhage, and further analyze the relationship between lactate and shunt dependent hydrocephalus. After subarachnoid hemorrhage, red blood cells will be broken down and released hemoglobin, followed by further metabolized to heme and iron. Free iron ions are toxic to brain cells, and protein should be combined into ferritin. In this process, which may be induced by an enzyme known as heme oxygenase (HO) 1, is the rate determining step of metabolism of hemoglobin. Many reports about head trauma or SAH previously showed that heme metabolism is related to outcome.. Materials and Methods Our first step is to confirm that after subarachnoid hemorrhage, the difference between ventricle and lumbar cerebrospinal fluid. So, we collected both ventricle and lumbar CSF on the seventh day after SAH. We further analysis of lumbar cerebrospinal fluid lactate and correlate with the hydrocephalus relevance. The Second study enrolled ptients with Fisher’s grade III aneurysmal SAH receiving early obliteration. The levels of heme oxygenase 1 (HO1), oxyhemoglobin, ferritin, and bilirubin in intra-thecal CSF were measured on the seventh day post-hemorrhage. The associations of functional outcome with clinical and CSF parameters were analyzed. Results The first study showed Intra-thecal CSF had significantly higher levels of total protein, ferritin, hemoglobin, and lactate but lower glucose level than intra-ventricular CSF (all p<0.0001). By multivariate analysis of clinical and CSF parameters, elevated intra-thecal CSF lactate (p=0.036) and the presence of intra-ventricular hemorrhage (p=0.05) were independent factors associated with SDHC. Moreover, intra-thecal lactate >5.5 μM effectively predicted the occurrence of SDHC (odds ratio: 32, 95% confidence interval: 3.8-270.8; p=0.0015). The second study showed age >60 years, admission World Federation of Neurosurgeons Score ≥3, and the presence of acute hydrocephalus were independent factors associated with an unfavorable outcome. After adjusting for clinical parameters, a higher level of HO1 appeared to be the most significant CSF parameter related to an unfavorable outcome among all tested CSF molecules (odds ratio=0.934, 95% CI: 0.883–0.989; p=0.018). Further analysis using a generalized additive model identified a cut-off value of HO1 > 81.2 uM to predict patients with an unfavorable outcome (82.4% accuracy). Conclusion and Prospect We suggested that there was compartalization between ventricle and subarachnoid space after massive SAH and intra-thecal lactate level is a useful predictive parameter for long-term SDHC in patients with aneurysmal SAH patients. Further, the second study found that the concentration of HO1 significantly associated with prognosis. Since SAH may cause compartalization of the ventricle and subarachnoid space, then how can we accelerate metabolism of the blood clot may possible improve patient prognosis. Therefore, we propose that continuous lumbar drainage may improve the outcome, clinical trials have been carried out. In order to verified the pathophysiology of DIND after SAH, we develop a new, in vivo observation of microcirculation after subarachnoid hemorrhage in rat. We can monitor cerebral blood flow and oxygen partial pressure at the same time. We found significant spasm of microcirculation, accompanied with decreased regional blood flow and oxygen partial pressure. This result is quite consistent with some recent studies, the treatment of spasm of the great vessels do not meet expectations, perhaps microcirculaiotn and microthrombus have a greater impact for the delayed ischemic brain lesions.