腦部病灶負荷對體顯性腦動脈血管病變合併皮質下腦梗塞及腦白質病變患者認知功能與步態表現的影響

Abstract

背景：腦部小血管疾病是導致中風、認知退化與步態障礙的主要原因之一。腦部小血管疾病病理變化會造成腦部病灶負荷增加，包括白質病變、腔隙性腦梗塞與腦部微出血，並可以用磁振造影加以量化。過往針對散發性腦部小血管疾病的研究顯示，主要腦部病灶指標──包括白質高訊號、腦部微出血與腔隙性腦梗塞─與認知功能受損（特別是執行功能）具有一致的關聯，且病灶位置與整體並灶負荷同樣重要。這些腦部病灶亦與步態障礙相關，例如較慢的步行速度、較短的步幅、增加的步態變異度，以及改變的雙重任務成本。體顯性腦動脈血管病變合併皮質下腦梗塞及腦白質病變（CADASIL）是由NOTCH3基因突變所致的最常見單基因型的腦部小血管疾病。然而，針對NOTCH3 R544C突變型 CADASIL，探討不同疾病嚴重度分期之認知功能與步態差異，以及腦部病灶負荷與認知功能與步態表現之相關性研究仍相當有限。 目的：本研究旨在（1）比較NOTCH3 R544C突變型CADASIL患者於三個NOTCH3-SVD分期階段之 CADASIL 患者在認知功能，以及單一與雙重任務步態表現；（2）探討定量腦部病灶指標（白質高訊號體積、腦部微出血數量、腔隙性腦梗塞數量）與 CADA-MRIT 分項分數（周邊、深部與表淺白質高訊號、腦部微出血與腔隙性腦梗塞分數）與認知功能及單一與雙重任務步態表現之關聯性。 方法：本橫斷式研究將自「臺灣遺傳性與非遺傳性腦部小血管病變研究計畫」（TAG-SVD）收案57名經基因檢測確診為NOTCH3 R544C突變型CADASIL的受試者（男性26名、女性31名；平均年齡64.4 ± 7.0歲）。所有受試者將接受腦部磁振造影以量化腦部病灶負荷，並依NOTCH3-SVD分期系統進行疾病嚴重度分期。步態使用 GAITRite 系統進行單一任務與認知雙重任務（連續減 7）步態評估。以標準化認知測驗評估整體認知功能、執行功能、注意力及處理速度。依資料分佈情形，認知功能與步態雙重任務成本以單因子變異數分析或Kruskal–Wallis 檢定進行組間比較。單一任務與雙重任務步態表現則以雙因子（3 組 × 2 條件）重複測量變異數分析進行檢定。此外，以多元線性迴歸分析探討腦部病灶負荷與認知及單一與雙重任務步態參數之獨立關聯。 結果：隨著 NOTCH3-SVD 分期增加，CADASIL患者的認知功能與單一任務及認知雙重任務步態表現皆顯著變差（p < 0.05）。依病灶類型進行之迴歸分析顯示，各類病灶呈現不同的功能關聯。就白質高訊號而言，較高的深部白質高訊號分數與較長的站立時間（β = 0.426，p = 0.027）、較長的雙腳支撐時間（β = 0.465，p = 0.013）、較大的站立期比例（β = 0.440，p = 0.026）與雙腳支撐期比例（β = 0.443，p = 0.023），以及較短的擺動期比例（β = −0.437，p = 0.027）顯著相關；在認知雙重任務條件下，較高的深部白質高訊號分數亦與較短的步長（β = −0.427，p = 0.033）、較大的站立期比例（β = 0.436，p = 0.042）、較大的雙腳支撐期比例（β = 0.242，p = 0.047），以及較短的擺動期比例（β = −0.439，p = 0.041）顯著相關。就腦部微出血而言，無論以腦部微出血數量或CADA-MRIT腦部微出血分數評估，較高的腦部微出血負荷皆與較差的執行功能表現（β = −0.512 與 −0.591，皆 p < 0.001）、較慢的單一任務步行速度（β = −0.390，p = 0.031；β = −0.379，p = 0.024），以及較大的擺動時間變異度（β = 0.408，p = 0.027；β = 0.425，p = 0.034）顯著相關。至於腔隙性腦梗塞，特別是以CADA-MRIT腔隙性腦梗塞分數評估時，呈現最一致的關聯結果；較高的腔隙性腦梗塞分數與較差的執行功能表現（β = -0.309，p = 0.038），以及多項步態參數之雙重任務成本增加顯著相關，包括步態週期時間（β = 0.400，p = 0.035）、站立時間（β = 0.388，p = 0.042）、步長變異度（β = 0.454，p = 0.015）與雙支撐時間變異度（β = 0.439，p = 0.018）。 結論：在NOTCH3 R544C突變型CADASIL患者中，隨著NOTCH3-SVD分期增加，認知功能以及單一任務與認知雙重任務步態表現皆呈現下降趨勢，且腦部病灶負荷與認知功能及步態表現顯著相關，並展現不同病灶類型具有其特定的功能關聯性。尤為重要的是，CADA-MRIT 各次項目分數相較於單純病灶數量或整體病灶體積，更能有效捕捉具臨床意義的病灶—功能關聯，突顯整合腦部影像、認知評估與步態分析於疾病監測上的臨床價值。

Background: Cerebral small vessel disease (SVD) is a major cause of stroke, cognitive decline, and gait impairment in older adults. Pathological changes of SVD lead to cerebral lesion burden, such as change in brain white matter, lacunar infarcts, and cerebral microbleeds (CMBs), quantified by using magnetic resonance imaging (MRI). Evidence from sporadic SVD indicates that major cerebral lesion markers—including white matter hyperintensity (WMHs), CMBs, and lacunes—are consistently associated with cognitive impairment, particularly in executive function, with lesion location appearing as relevant as overall lesion burden. These lesion markers are also linked to gait impairment, such that greater WMH, lacune, and CMB burdens are associated with slower gait velocity, shorter stride length, increased gait variability, and altered dual-task costs. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), caused by mutations in the NOTCH3 gene, is the most common monogenic form of SVD. However, in CADASIL caused by the NOTCH3 R544C mutation, studies examining cognitive functions and gait differences across disease severity stages and the associations between lesion burdens with cognitive functions and gait performance remain limited. Purpose: This study aimed to (1) compare cognitive functions and single- and cognitive dual-task gait performance across three NOTCH3-SVD staging groups in individuals with CADASIL carrying the NOTCH3 R544C mutation; and (2) investigate the associations of quantitative cerebral lesion markers (WMH volume, CMB number, and lacune number) and Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy - MRI Inventory Tool (CADA-MRIT) subitem scores (periventricular WMH, deep WMH, and superficial WMH, CMB, and lacune score) with cognitive functions and single- and dual-task gait performance. Methods: This cross-sectional study included 57 participants (26 males and 31 females; mean age = 64.4 ± 7.0 years) with genetically confirmed CADASIL due to the NOTCH3 R544C mutation from the Taiwan Associated Genetic and Nongenetic Small Vessel Disease (TAG-SVD) cohort. All participants underwent brain MRI for quantification of cerebral lesion burdens and were classified into disease severity stages using the NOTCH3-SVD staging system. Gait performance was assessed using the GAITRite system during single-task and cognitive dual-task (serial-7 subtraction) walking conditions. Standardized cognitive tests were administered to evaluate global cognition, executive function, attention, and processing speed. Group differences in cognitive functions and dual-task costs (DTCs) of gait performance were examined using one-way ANOVA or Kruskal–Wallis tests, as appropriate based on data distribution. Group differences in single- and dual-task gait performance were examined using a two-way (3 groups x 2 conditions) repeated-measures ANOVA. Multiple linear regression analyses were conducted to explore the independent associations of cerebral lesion burdens with cognitive functions and single- and dual-task gait parameters. Results: More advanced NOTCH3-SVD staging was associated with poorer cognitive function and single- and dual-task gait performance (p < 0.05). Lesion-specific regression analyses revealed distinct functional associations. For WMHs, a higher deep WMH score was associated with longer stance time (β = 0.426, p = 0.027), longer double support time (β = 0.465, p = 0.013), greater stance phase duration (β = 0.440, p = 0.026), and greater double support phase duration (β = 0.443, p = 0.023), accompanied by a reduced swing phase duration (β = −0.437, p = 0.027). Under the dual-task condition, higher deep WMH scores were associated with shorter step length (β = −0.427, p = 0.033), greater stance phase duration (β = 0.436, p = 0.042), greater double support phase duration (β = 0.242, p = 0.047), and reduced swing phase duration (β = −0.439, p = 0.041). For CMBs, both the CMB number and the CADA-CMB score showed consistent associations with poorer functional performance. Specifically, a higher CMB burden was associated with poorer executive performance (β = −0.512 and −0.591, both p < 0.001), lower single-task gait velocity (β = −0.390, p = 0.031; β = −0.379, p = 0.024), and greater swing time variability (β = 0.408, p = 0.027; β = 0.425, p = 0.034). For lacunes, particularly when assessed using the CADA-MRIT lacune score, the most consistent associations were observed. A higher lacune score was associated with poorer executive performance (β = −0.309, p = 0.038), greater DTC of cycle time (β = 0.400, p = 0.035), stance time (β = 0.388, p = 0.042), step length variability (β = 0.454, p = 0.015), and double support time variability (β = 0.439, p = 0.018). Conclusion: In people with CADASIL carrying the NOTCH3 R544C mutation, worsening cognitive function and single- and dual-task gait performance were observed across increasing NOTCH3-SVD stages were observed across increasing NOTCH3-SVD stages, and cerebral lesion burden was associated with cognitive function and gait performance, with lesion-specific functional correlates. Importantly, itemized CADA-MRIT scores provided advantages in capturing clinically relevant lesion–function relationships beyond global lesion volume or simple lesion counts, highlighting the value of integrated MRI, cognitive, and gait assessments for disease monitoring in this population.