臺灣社區老年人之平均紅血球體積與認知功能障礙之關聯：六年世代研究

Abstract

研究目的：本研究目的為探討臺灣社區老年人之平均紅血球體積與認知功能隨時間變化之關聯。 研究背景：隨著人口快速老化，老年人的健康問題變得日益嚴重，認知功能障礙被列為公共衛生的優先項目。較大的紅細胞更為脆弱並容易破裂，難以運輸氧氣和營養物質到大腦中，從而導致個體產生認知功能障礙。但過去僅有少部分研究在探討平均紅血球體積和認知功能障礙的關聯性，且研究結果並不一致。因此本研究目的為探討臺灣社區老年人之平均紅血球體積與認知功能在六年間之關聯。 研究方法：本研究為追蹤6年之世代研究，於基線時（2011至2013年）招納了605位年齡≥65歲的受試者，排除患有阿茲海默氏症或正在服用治療阿茲海默氏症的藥物、有中風病史、腦部腫瘤≥3公分者、曾失去意識≥30分鐘者、蒙特利爾認知評估臺灣版總分（Montreal Cognitive Assessment- Taiwanese version , MoCA-T）≤21者、平均紅血球體積有缺失值者、兩次測量的蒙特利爾認知評估臺灣版總分分數相差≥12分，最終一個納入522（86.3%）位長者。本研究的依變項為在基線、第2年追蹤、第4年追蹤及第6年追蹤時進行評估的受試者之認知功能狀態，包括整體認知功能和四個認知功能領域（邏輯記憶功能，注意力，語言流暢度及執行功能）。本研究的自變項為在基線測量的平均紅血球體積。將受試者依平均紅血球體積在基線時之數值排序平分為三等分，三組中數值最高的一組（T3）定義為“高”平均紅血球體積，平均紅血球體積數值較低的兩組合併起來（T1+T2）定義為“低”平均紅血球體積，使用廣義線性混合模型（Generalized linear mixed model, GLMM）探討平均紅血球體積與認知功能於六年間之關聯，模型中調整性別、年齡、教育年數、血小板數值、紅細胞體積分佈寬度的變異係數、葉酸、維生素B12以及追蹤年數，並且納入年齡與上述調整變項之交互作用項，將上述迴歸分析中達統計顯著的關係進一步依據性別、年齡、葉酸濃度、紅血球分佈寬度變異係數、維生素B12濃度及血小板數值進行分層分析。 研究結果：在平均紅血球體積較高（T3,平均紅血球體積≥92.9 fL）的受試者中,年齡每增加一歲,蒙特利爾認知評估臺灣版（MoCA-T）的分數會減少0.063（平均紅血球體積×年齡：β= -0.063, 95% CI = -0.122至-0.003, p值=0.04），即隨著年齡增加基線平均紅血球體積較大對於整體認知功能具有負面影響且平均紅血球體積與年齡存在交互作用。以整個族群的平均年齡（73歲）來看,高平均紅血球體積（T3）與較差的整體認知功能（以蒙特利爾認知評估臺灣版評估）有關（平均紅血球體積與平均紅血球體積×年齡β之總和：-0.11=4.49-73×0.063），而且平均紅血球體積與年齡之間存在交互作用（Pinteraction=0.04）。依據重要因子進行分層後，在某些次族群中可以觀察到顯著的關聯性。在年齡≥75歲的受試者中，高平均紅血球體積組別（≥92.9 fL）比起低平均紅血球體積的組別（<92.8 fL）其整體認知功能表現較差（平均紅血球體積：β=-0.96，95% CI：-1.61至-0.30）；在葉酸濃度較低的受試者（≤15.1 pg/ml）中，以整個族群的平均年齡（73歲）來看，高平均紅血球體積（T3）與較差的整體認知功能（以蒙特利爾認知評估臺灣版評估）有關（平均紅血球體積與平均紅血球體積×年齡β之總和：-0.2=6.37-73×0.09）；在維生素B12濃度較低的受試者（≤825 pg/ml）中，以整個族群的平均年齡（73歲）來看，高平均紅血球體積（T3）與較差的整體認知功能（以蒙特利爾認知評估臺灣版評估）也有關（平均紅血球體積與平均紅血球體積×年齡β之總和：-0.11=7.19-73×0.1）。 結論：以整個族群的平均年齡73歲來看，基線平均紅血球體積較大對於整體認知功能具有負面影響，且平均紅血球體積與年齡存在交互作用。分層分析發現上述關係只出現在葉酸濃度較低的受試者（≤15.1 pg/ml）及維生素B12濃度較低的受試者（≤825 pg/ml）中。希望未來能有更多大型及長期之世代追蹤的研究，來探討平均紅血球體積對於認知功能的影響及因果關係。 關鍵字：老年人、平均紅血球體積、認知功能、認知功能障礙、蒙特利爾認知評估

Purpose: The aim of this study was to explore the association between mean corpuscular volume (MCV) and cognitive impairment in the community-dwelling older adults. Background: As population aging rapidly, cognitive impairment becomes an important health concern. Mean corpuscular volume (MCV) is the average volume of red blood cells (RBC). Larger MCV is fragile and has difficulty to carry oxygen and nutrients to brain thus leading to cognitive impairment. Few studies have relating MCV levels to cognitive impairment and their findings were inconsistent. This study aimed to explore this association in community-dwelling older adults. Methods: This study was a 6-year cohort study, a total of 605 elderly participants (age≥ 65) at baseline (2011-2013), recruited from the annual elderly health checkup program at the National Taiwan University Hospital from 2011 to 2013. Participants who meet the following criteria were excluded: took medications for Alzheimer’s disease, had the history of stroke, brain tumor (≥3cm), loss of consciousness≥30 min, baseline Montreal Cognitive Assessment–Taiwanese version (MoCA-T) score<22 (highly suspect with dementia), missing data of MCV, and the change of MoCA-T score over 2 years≥12. A total of 522 elderly (age≥ 65) were enrolled in the study. Global and domain-specific (logical memory, attention, verbal fluency, and executive function) cognition were assessed for each participant at baseline, 2-year (2013-2015), 4-year (2015-2017), and 6-year (2017-2019) follow-ups. MCV was assessed at baseline (2011-2013), high MCV indicates the highest tertile (T3); Low MCV indicates the lower tertiles (T1+T2). The generalized linear mixed models (GLMM) were used to estimate the association between MCV and cognitive impairment adjusting for important covariates, which included age, sex, years of education, platelet level, red cell volume distribution width - coefficient of variation (RDW-CV), serum folate level, vitamin B12 level and years of follow up. Results: With the population average age of 73 years old, high baseline MCV (T3, MCV≥92.9 fL) was associated with a poor performance of global cognition (sum of β of MCV and β of MCV×age: 4.49-73×0.063 = -0.11) and there existed an interaction between the MCV and age (p-value=0.04). After stratified analyses, the highest tertile MCV was significantly interact with age (＜75 years and ≥75 years) on global cognition (MoCA-T). In older adults aged ≥75 years old, increased baseline MCV was associated with poor performance of global cognition (β = -0.96, 95% CI = -1.61 to -0.30). In older adults with lower folate (≤15.1 pg/ml), of which the average age of the population was 73 years old, high MCV (T3: ≥92.9 fL) was associated with poor global cognition (sum of β of MCV and β of MCV×age: 6.37-73×0.09 = -0.2). In older adults with lower vitamin B12 (≤825 pg/ml), of which the average age of the population was 73 years old, high MCV (≥92.9 fL) was associated with poor global cognition (sum of β of MCV and β of MCV×age: 7.19-73×0.1=-0.11). No significant findings were observed in cognitive domains or other strata. Conclusion: High baseline MCV was associated with poor global cognitive function and there existed a MCV and age interaction. Taken together, MCV level may be used as an early marker for screening dementia in the community-dwelling older adults. Key words: elderly, mean corpuscular volume, cognition, cognitive impairment, Montreal Cognitive Assessment–Taiwanese version