慢性期中風病患大腦皮質活化型態及皮質脊髓徑完整性與其下肢動作功能之關係: 神經塑性研究

Abstract

Study I: 目的：本研究主要目的是探討慢性期中風患者於踝關節進行背屈曲動作時，大腦主要感覺動作皮質區、輔助運動皮質區及扣帶迴動作皮質區之功能性活化強度及側化程度與下肢動作機能損傷及功能性活動能力之相關性。 方法：本研究共徵召15位慢性期單側偏癱中風受試者(6位女性，9 位男性；平均年齡，61.2 ± 7.6歲; 平均發病後時間18.0 ± 15.8個月)及15位同年齡層之健康成年受試者(10位女性，5 位男性；平均年齡，61.6 ± 7.3歲)，以3T磁振造影系統收集功能性磁振造影(functional magnetic resonance imaging, fMRI)，以分析受試者踝關節進行背屈曲動作時(0.17 Hz)之大腦皮質活化型態。所有中風受試者皆接受臨床評估，以傅格-梅爾評估量表下肢動作功能部分(lower extremity motor component of Fugl-Meyer Assessment Scale, FMALE)及三公尺計時起走測試(Timed “Up & Go” Test, TUGT)分別量測下肢動作機能損傷程度及功能性活動能力。大腦皮質功能性活化強度以3對興趣區域- 即兩側之主要感覺動作皮質區(primary sensorimotor cortex, SMC)、輔助運動皮質區(supplementary motor area, SMA)及扣帶迴動作皮質區(cingulate motor area, CMA)之活化t值總和(Σt)代表皮質活化強度絕對指標。每一個ROI之Σt的計算方式為先計算每一對ROI中出現最高之前5%的活化立體像素之平均t值，以此平均t值的50%定義該ROI之活化t閾值，而ROI中超過此t閾值的活化立體像素之t值總和即為Σt。兩側大腦ROI活化之側化程度則以兩側大腦活化之加權側化指數(weighted laterality index, wLI)(Fernandez等，2001)作為皮質活化強度之相對指標。加權側化指數值介於+1到-1，+1代表單側踝關節進行背屈曲動作時，皮質活化完全在對側大腦皮質ROI，而-1則代表完全同側大腦ROI。為瞭解興趣區域之皮質功能性活化型態與臨床評估之相關性，則將兩側SMC, SMA, CMA之Σt值及每對SMC, SMA, CMA之wLI值分別與FMALE及TUGT作凈相關之分析(partial correlation)，並控制年齡及發病後時間的影響。 結果：結果顯示FMALE與健側大腦SMC之Σt值呈現顯著負相關(r= -0.627, p= 0.022)，與wLI則具有正相關之趨勢(r= 0.500, p= 0.082)。此表示患側下肢動作機能損傷程度越高之中風患者，有較高之健側大腦SMC活化強度及較低的SMC活化側化現象。而三公尺計時起走測試之表現則與患側大腦SMC活化之Σt值呈顯著之正相關(r= 0.729, p= 0.005)，並與健側大腦SMC活化之Σt(r= 0.509, p= 0.076)、兩側SMA活化之Σt(r= 0.509, p= 0.076 and r= 0.542, p= 0.056)及患側大腦CMA活化之Σt(r= 0.524, p= 0.066)具有正相關之趨勢，表示功能性活動能力較差者，其兩側之SMC, SMA及患側CMA的活化強度較高。 討論與結論：本研究結果顯示慢性期中風患者下肢不同層面之動作功能表現與大腦兩側SMC、SMA、CMA不同的適應性活化型態相關。患者的患側下肢動作機能損傷程度主要與SMC功能性活化強度及側化程度有關。下肢之功能性活動能力則與多個大腦皮質區(含兩側之SMC, SMA及患側大腦CMA)廣泛的功能性活化有關。因此，這些發現表示中風患者下肢動作機能損傷及功能性活動能力之恢復有賴於不同大腦皮質區的功能性重組。

Study II: 中文摘要 目的：本研究主要為針對慢性期中風病患(1)探討大腦皮質脊髓徑(corticospinal tract, CST)下肢動作纖維之結構完整性及皮質活化型態與下肢動作功能表現的相關性;(2)瞭解大腦CST下肢動作纖維之結構完整性及皮質活化型態對下肢動作功能表現之貢獻。 方法：本研究共徵召18位慢性期單側偏癱之中風受試者(7位女性，11位男性；平均61.0 ± 7.4歲；平均發病後時間16.6 ± 15個月)。所有受試者皆接受臨床評估及3T磁振造影(功能性磁振造影與擴散頻譜造影)掃瞄。臨床評估測試為分別以傅格-梅爾評估量表下肢動作功能部分(Lower extremity motor component of Fugl-Meyer Assessment Scale, FMALE)及三公尺計時起走測試(Timed “Up & Go” Test, TUGT)量測下肢動作機能損傷程度及功能性活動能力。以功能性磁振造影分析患者踝關節進行背屈曲動作時之大腦皮質活化強度，並計算位於三對興趣區域- 即主要感覺動作皮質區(primary sensorimotor cortex, SMC)、輔助運動皮質區(supplementary motor area, SMA)及扣帶迴動作皮質區(cingulate motor area, CMA)的活化t值總和(Σt)以代表皮質活化強度絕對指標，及以兩側大腦興趣區域皮質活化強度的加權側化指數(weighted laterality index, wLI)代表皮質活化強度相對指標。以擴散頻譜造影資料分析大腦CST內囊後肢纖維(posterior limb of internal capsule, PLIC)結構完整性。以患側大腦CST之PLIC的普擴散不等向性分數(generalized fractional anisotropy, GFA)作為代表CST完整性之絕對指標，以及兩側大腦PLIC之GFA相對值(rGFAPLIC) 作為代表CST完整性之相對指標。以單變量線性迴歸(univariate linear regression)分析兩個臨床測試(FMALE及TUGT)與CST結構完整性及大腦功能性皮質活化型態之絕對與相對指標間的相關性，並控制年齡及發病後時間的影響。進一步再進行多變量線性迴歸(multivariate linear regression)以決定CSTLE結構完整性及大腦皮質活化型態對下肢動作機能損傷及功能性活動能力之相對貢獻程度。 結果：單變量線性迴歸結果發現FMALE分數與患側大腦GFAPLIC值(GFAPLIC_AH) (R2= 0.392, p= 0.011)、rGFAPLIC (R2= 0.572, p= 0.001)及健側大腦SMC之Σt值(Σt SMC_UH) (R2= 0.619, p= 0.004)、SMC加權側化指數(wLISMC)(R2= 0.389, p= 0.056)呈中至高度的相關性。TUGT則與健側大腦Σt SMC(Σt SMC_UH)(R2= 0.545, p= 0.034)、患側大腦Σt SMC (Σt SMC_AH)(R2= 0.688, p= 0.004)及患側大腦Σt CMA (Σt CMA_AH)(R2= 0.610, p= 0.014)呈高度相關，但與DSI之量測指標則不具相關性。多變量線性迴歸分析結果顯示，在以影像之絕對指標為預測因子的模型中，Σt SMC_UH(adjusted R2= 0.505, p= 0.004)為預測FMALE之唯一顯著之預測因子；而以影像之相對指標為預測因子的模型中，則發現rGFAPLIC(adjusted R2= 0.486, p= 0.031)為預測FMALE之唯一顯著因子。Σt SMC_AH (adjusted R2= 0.595, p= 0.004)則為預測TUGT之唯一顯著預測因子。 討論與結論:本研究結果顯示兩側大腦CST結構完整性之相對指標及健側大腦主要動作感覺皮質區活化強度之絕對指標為中風慢性期患者下肢動作機能損傷重要的預測因子。但針對功能性活動程度之能力而言，則患側大腦主要動作感覺皮質區之活化強度為最重要的預測因子。因此，建議中風腦傷後不同型態之大腦塑性將影響患者不同面向之臨床動作功能表現。

Study I: Purposes: The aim of this study was to examine how the intensity and lateralization of cortical activations in the primary sensorimotor cortex (SMC), supplementary motor area (SMA), and cingulate motor area (CMA) during ankle movements correlate with motor impairment and functional mobility in patients with chronic stroke. Methods: Functional magnetic resonance imaging (fMRI) data of cortical activations during active ankle dorsiflexion movements (0.17 Hz) were acquired using a 3 Tesla MR scanner from 15 patients (6 females and 9 males; mean age, 61.2 ± 7.6 years) with unilateral hemiplegia following stroke (mean post-onset time, 18.0 ± 15.8 months) and from 15 age-matched healthy subjects (5 males and 10 females; 61.6 ± 7.3 years). The lower extremity motor component of the Fugl-Meyer Assessment (FMALE) and the Timed “Up & Go” Test (TUGT) were used to assess motor impairment of the affected lower extremity and functional mobility of patients, respectively. The activation intensity in three paired regions of interest (ROIs)- the bilateral SMC, SMA, and CMA, were calculated by summing the t values (Σt) of voxels which t values were above the t threshold defined as 50% of the mean of the top 5% maximum t values in the corresponding homologous ROIs. The degree of lateralization of cortical activation in each pair of ROIs was calculated by using a weighted laterality index (wLI) proposed by Fernandez. An wLI value of +1 indicated absolute contralateral cortical activation during ankle movements, whereas an wLI value of -1 indicated absolute ipsilateral cortical activation. The correlations between the Σt values in bilateral SMC, SMA, and CMA, and the wLI values for each pair of SMC, SMA, and CMA with the FMALE and TUGT scores were analyzed using partial correlations, controlling for age and post-onset time. Results: The FMALE scores showed a significant negative correlation with the Σt value of SMC of the unaffected hemisphere (r= -0.627, p= 0.022) and a trend of positive correlation with the wLI of the SMC (r= 0.500, p= 0.082), suggesting that patients with greater affected lower extremity motor impairment presented greater activation intensity in the SMC of unaffected hemisphere and smaller lateralization of SMC activation during affected ankle movements. The TUGT performance showed a significant positive correlation with the Σt value of the SMC of the affected hemisphere (r= 0.729, p= 0.005), and a trend of correlation with the Σt values of the SMC of the unaffected hemisphere (r= 0.491, p= 0.09), with those of SMA of affected and unaffected hemispheres (r= 0.509, p= 0.076 and r= 0.542, p= 0.056, respectively), and with that of the CMA of the affected hemisphere (r= 0.524, p= 0.066). Patients with poorer functional mobility demonstrated greater activation intensity in bilateral SMC, SMA, and CMA of the affected hemisphere during affected ankle movements. Discussion and Conclusions: Results of this study revealed that different aspects of lower extremity motor functions in patients with chronic stroke were correlated with different adaptive cortical activation patterns in bilateral SMC, SMA, and CMA regions in patients with stroke. While patients’ degree of motor impairment of the affected lower extremity was primarily associated with the activation intensity and lateralization of the SMC, their functional mobility was associated with activation intensity in more widespread cortical motor regions, including bilateral SMC and SMA, and CMA of the affected hemisphere. These findings suggest that the recovery of motor impairment and functional mobility of stroke patients may depend upon functional reorganization of different brain regions.

Study II: Purposes: The two purposes of this study were (1) to investigate the associations of the structural integrity of the corticospinal tract lower extremity motor fibers (CSTLE) and cortical activation patterns with the affected lower extremity motor functions in hemiplegic patients following chronic stroke and (2) to determine the relative contributions of the structural integrity of CSTLE and cortical activation patterns to affected lower extremity motor functions in these patients. Methods: Eighteen hemiplegic patients with chronic stroke (7 females and 11 males; mean age, 61.0 ± 7.4 years, mean post-onset time, 16.6 ± 15.0 months) were recruited. All patients underwent clinical measures and MRI scans using a 3 Tesla MR scanner. The lower extremity motor component of the Fugl-Meyer Assessment (FMALE) and the Timed “Up & Go” Test (TUGT) was used to assess motor impairment of affected lower extremity and functional mobility, respectively. Cortical activations during active ankle dorsiflexion movements were measured using an fMRI paradigm. Cortical activation patterns were assessed by calculating absolute summation of t values ( ) above activation threshold in the primary sensorimotor cortex (SMC), supplementary motor area (SMA), and cingulate motor area (CMA) of both hemispheres, as well as the relative activation intensity in these three regions of interest (ROIs) between the two hemispheres, denoted as the weighted laterality index (wLI). The Diffusion spectrum imaging (DSI) was used to assess the structural integrity of the posterior limb of internal capsule (PLIC) segment of the CSTLE. The integrity was indicated by calculating general fractional anisotropy of the PLIC segment of the CSTLE in the affected hemisphere (GFAPLIC_AH) in absolute term and by calculating the relative GFA of the PLIC (rGAFPLIC) between bilateral hemispheres in relative term. The associations of the two clinical measures (FMALE and TUGT) and the absolute and relative DSI (GFAPLIC_AH and rGAFPLIC) and fMRI measures (Σt values of bilateral SMC, SMA, and CMA activations, as well as wLI of SMC, SMA, and CMA activations) were first analyzed using the univariate linear regression analyses, controlling for age and post-onset time. Then, the multivariate linear regression analyses were performed to determine the relative contributions of the structural integrity of CSTLE and cortical activation patterns to FMALE and TUGT. Results: Univariate analyses showed that the FMALE score was moderately to highly correlated with two DSI measures, GFAPLIC_AH (R2= 0.392, p= 0.011) and rGFAPLIC (R2= 0.572, p= 0.001), and with two fMRI measures, Σt of SMC of the unaffected hemisphere (Σt SMC_UH) (R2= 0.619, p= 0.004) and wLISMC (R2= 0.389, p= 0.056). The TUGT performance was only strongly correlated with Σt of SMC of the affected hemisphere (Σt SMC_AH) (R2= 0.688, p= 0.004), Σt SMC_UH (R2= 0.545, p= 0.034), and Σt of CMA of the affected hemisphere (Σt CMA_AH) (R2= 0.610, p= 0.014), but not with DSI measures. Multivariate regression analyses showed that Σt SMC_UH (adjusted R2= 0.505, p= 0.004) was the only significant predictor for FMALE in the model using solely absolute DSI and fMRI measures and rGFAPLIC (adjusted R2= 0.486, p= 0.031) were the only significant predictor for FMALE in the model using solely relative DSI and fMRI measures. The Σt SMC_AH alone significantly predicted TUGT performance (adjusted R2= 0.595, p= 0.004). Discussion and Conclusions: Results of this study suggest that relative CST structural integrity measure and absolute SMC activation intensity measure in the unaffected hemisphere are the most important independent predictors for the lower extremity motor impairment level in patients with chronic patients. On the other hand, only the absolute SMC activation intensity in affected hemisphere was an independent predictor for functional mobility performance of these patients measured by TUGT. These findings suggest that different aspects of brain plasticity following stroke may contribute differentially to different dimensions of clinical motor performance.