背景：阿茲海默氏症目前的主要治療方式為經由乙醯膽鹼水解酶抑制劑，來重建中樞之乙醯膽鹼系統。然而，不是所有的患者都對乙醯膽鹼水解酶抑制劑有良好的反應。本研究希望藉由患者神經心理測驗的表現，來預測乙醯膽鹼水解酶抑制劑在阿茲海默氏症的療效。
方法：本研究共收案六十位阿茲海默氏症患者。受試者完成三項神經心理測驗：路徑描繪測試、Stroop叫色測試及工作記憶測試。在接受乙醯膽鹼水解酶抑制劑治療六個月後，患者將依據簡短版心智測驗（MMSE）分數的減少是否小於0，而被分為有效組 (N=35)及無效組 (N=12)。
結果：在Stroop叫色測試當中，無效組在色、字不一致的情況下，反應時間明顯較有效組為長 (p=.010)。在路徑描繪測試及工作記憶測試的表現，則無法區分出有效組及無效組。
結論：本實驗之結果發現，在六個月的追蹤時間內，Stroop叫色測試可以預測阿茲海默氏症患者對乙醯膽鹼水解酶抑制劑的療效。我們認為，Stroop叫色測試具有預測力的原因是因為此測驗可特別探測到前回扣帶和背外側前額葉間的聯繫，或是前額葉之留存功能。這是第一個利用Stroop叫色測試來預測乙醯膽鹼水解酶抑制劑的療效的實驗。Stroop叫色測試為一臨床可行的工具，希望本篇拋磚引玉，未來能有更多這方面的相關探索。
Background: Current major treatment of Alzheimer's disease is restoring the central cholinergic system by acetylcholinesterase inhibitors (AChEIs). However, not all patients respond to AChEIs well. The goal of this study was to predict the effectiveness of AChEIs in Alzheimer's disease by baseline neuro-psychologic tasks.
Methods: In total 60 patients with Alzheimer's disease were included. They completed three neuro-psychologic tasks: color trail making task (cTMT), Stroop task and working memory task. After 6-months AChEI therapy, the participants were separated into responders (N=35) and non-responders (N=12), by the criterion of decrement of less than 0 point in Mini-Mental State Examination (MMSE) score compare to baseline.
Results: The non-responders showed longer reaction time in the incongruent condition in the Stroop task than the responders (p=.010). Meanwhile, the performance of the cTMT and the working memory task cannot distinguish responders and non-responders.
Conclusions: Our data suggest that the Stroop task is able to predict AChEIs effectiveness in Alzheimer's disease during a 6-months follow up. We argue that the predictive power of the Stroop task may be from a specific probe on the connection between anterior cingulate gyrus and dorsolateral prefrontal cortex, or the general reserved function of the prefrontal cortex. This is the first study to use Stroop task to predict the effectiveness of AChEIs in Alzheimer's disease. Stroop task is a clinical practical tool and deserves further exploration on the robustness of our findings.
