摘要:中風在台灣及其他國家仍是造成個案長期失能的主要疾病之ㄧ。目前中風的高發生率及低死亡率更突顯尋求有效復健療法的迫切性。許多著重於提昇中風患者上肢動作之現代復健治療因應而生且廣泛被提倡。機器輔助療法是目前一個極具潛力且以活動為基礎的療法,此療法融入一些有效的中風動作治療策略於系統設計中,如:高度密集訓練、重複練習、任務專一性、回饋提供、及雙側練習等。然而,目前對於此機器輔助療法之療效(如: 動作和日常生活功能)、治療對細胞氧化反應的影響等科學性證據仍非常有限。此外,我們必須尋求適切的治療強度以闡述並探究治療劑量與療效間之關係;而找出顯著的療效預測因子以選擇最適合接受此療法之病人族群亦是重要的議題。因此,本研究將使用隨機控制試驗之研究設計探究當前機器輔助療法運用於中風患者之重要議題。本研究計畫主要包含兩部分:(一)機器復健療效試驗及(二)療效預測與臨床計量特性研究。第一部分之研究旨在探究機器輔助療法運用於中風患者之療效、使用兩組不同治療強度的組別(較高和較低密集訓練)以檢驗治療劑量與成效間關係、及探究不同治療強度對於細胞氧化指標的影響。治療成效評量包含完整的各類功能層面: 動作能力、動作控制策略、基本/延伸性日常生活功能、行動功能、重返社區功能、生活品質、及生物指標(8-OHdG)等。第二部分研究將探討成效之影響因子以界定適當的介入族群、及檢驗成效工具之臨床計量特性。我們將檢驗患者不同之臨床特性是否會影響其治療成效,以定義最可從此療法中獲益的病人族群。此外,我們將檢驗並比較各種成效評量工具的臨床計量特性(如:效度、反應性),並據以選擇可偵測復健治療後臨床重要改變之工具。此四年期的計畫預計募集120位中風患者參與試驗。符合收案標準的中風患者將隨機分配至三種復健治療中的一組: 較高密集性機器輔助療法、較低密集性機器輔助療法、常規復健療法,每位病人皆將接受4週的治療。成效評量工具將包含:傅格梅爾動作量表 (Fugl-Meyer Assessment, FMA)、醫學研究會議之肌力量表 (Medical Research Council scale, MRC)、肌肉功能測量儀 (MYOTON-3)、腕動計(accelerometers)、功能性行走分類量表(Functional Ambulation Categories, FAC), 五公尺步行測試 (Five-meter walk test)、功能性獨立量表 (Functional Independence Measure, FIM)、諾丁漢延伸性日常生活量表 (Nottingham Extended Activities of Daily Living Scale, NEADL)、重返正常生活量表(Reintegration to Normal Living Index, RNL)、中風影響量表 (Stroke Impact Scale, SIS)、運動學分析 (kinematic analysis) 及尿液8-OHdG濃度檢測等。每位患者將於治療前、治療中(治療後2週)、治療後、及治療後3個月各接受一次成效評估。資料分析將使用多變量共變數分析 (MANCOVA) 以檢驗不同治療組別在多個成效指標重複評量之效應,並使用前測分數做為共變數。效應值r 將用以計算三組別在不同成效指標上的療效大小。此外,配對t檢定和單變量共變數分析 (ANCOVA)分別用以檢驗同一組內及不同組間的尿液8-OHdG濃度之變化差異。另外,此研究將使用多重迴歸分析找出各成效指標之顯著預測因子。各種臨床成效評量工具之反應性和效度將分別使用標準化反應平均值(standardized response mean, SRM)和Spearman 相關係數檢定之。完成此計畫後,我們預期將可完整地提供機器輔助療法運用於中風患者之療效、治療劑量與成效間關係、治療對細胞氧化可能影響、影響療效之因子、與成效評量工具之臨床計量特性等實證結果。此研究計畫之結果將促使我們了解此療法提升中風患者功能之原因與機制及密集性訓練對於細胞生理氧化反應之影響。整體而言,此研究計畫將有助於促進我國實證復健之發展,充實中風復健成效之本土資料,並將動作科學研究的知識與技術轉譯至神經復健領域。
Abstract: Stroke remains a leading cause of permanent disability in Taiwan and many other countries. The high incidence of stroke and the decreased mortality from stroke which imply the urgent needs for effective rehabilitation. Various contemporary rehabilitation interventions focused on restoring upper limb motor function have been advocated for stroke rehabilitation. Robot-assisted therapy (RAT), one current prominent activity-based approach, has emerged that incorporates therapeutic elements for success in stroke motor rehabilitation: high-intensity, repetitiveness, task-specificity, feedback, and bilateral training into its design. However, scientific evidence for the effects of the RAT on functional outcomes (e.g., motor and daily functions) and physiological responses (e.g., oxidative responses) in stroke patients remains limited. Furthermore, there is a need to identify the proper level of treatment intensity in order to elucidate the dose-response relations. In addition, it is important to identify the predictors of treatment success in order to inform patient selection. This research project will use randomized controlled trial design to address these issues. This project consists of two parts: Robotic Rehabilitation Trials, and Study of Outcome Predictors and Clinimetric Attributes. In the first part of study, we aim to (1) investigate the treatment effects of RAT in patients with stroke on various outcomes, (2) test the dose-response relations by using two different intensities (higher vs. lower intensity RAT), and (3) investigate the effects of the training intensity on a biomarker of oxidative stress. Treatment outcomes will encompass the spectrum of functioning including motor ability, motor control strategies, basic/extended daily functions, mobility, community reintegration, quality of life, and biomarker (8-OHdG). In the second part of study, the purposes are to define the appropriate populations for RAT and to examine the clinimetric properties of clinical measures relevant for use in robotic rehabilitation research. We will identify the clinical predictors that will potentially influence the functional outcomes after interventions. Finally, we will examine and compare the clinimetric properties (e.g., validity and responsiveness) of the clinical measures of rehabilitation outcome to inform selection of test instruments that may detect clinically meaningful change after rehabilitation therapy. An estimated total of 120 patients with stroke will be recruited in this 4-year project. Eligible participants will be randomized into the higher-intensity RAT, the lower-intensity RAT, or the dose-matched conventional rehabilitation group for a 4-week training period. Outcome measures, including the Fugl-Meyer Assessment (FMA), Medical Research Council scale (MRC), MYOTON-3, accelerometers, Functional Ambulation Category (FAC), Five-meter walk test, Functional Independence Measure (FIM), Nottingham Extended Activities of Daily Living Scale (NEADL), Reintegration to Normal Living Index (RNL), Stroke Impact Scale (SIS), kinematic analysis, and urinary 8-OHdG level (a biomarker of oxidative stress) will be administered to patients before intervention, at 2-week after intervention, immediately after intervention, and at 3-month follow-up. Multivariate analyses of covariance (MANCOVAs) will be used to examine changes in kinematic variables, and clinical measures as a function of treatment while controlling for baseline data. Further, the effect size r will be calculated for each outcome variable to index the magnitude of the performance difference between groups. Paired t test and ANCOVA will be used to examine the changes of 8-OHdG level within a group and to compare the differences of 8-OHdG level between the 3 groups, respectively. In addition, multiple regression models will be used to identify the significant predictors for the outcomes. The responsiveness and validity of the clinical measures relevant for clinimetric scrutiny will be examined by the index of standardized response mean (SRM) and the Spearman correlation coefficients.We anticipate the findings of this research project will shed important light on the evidence of treatment effects of RAT for stroke patients, the dose-response relationships, possible effects of intense rehabilitation therapy on oxidative stress, factors that may influence the outcomes, and clinimetric properties of the outcome measures. The findings will also contribute to our understandings of why the therapy may benefit stroke patients and the physiological oxidative responses in intense training. The overall results will be significant for evidence-based robotic rehabilitation and contribute to the translation of movement science and physiological study into neurorehabilitation.