腰腿比與糖尿病及代謝特徵因子之相關性探討

Waist-to-thigh Ratio and Its Association with Diabetes and Metabolic Syndrome Components

張芳綸(Fang-Lun Chang)

指導教授 ： 葉志嶸

中山醫學大學/健康管理學院/公共衛生學系碩士班/碩士(2011年)

https://doi.org/10.6834/CSMU.2011.00040

摘要

研究目的：本研究比較大腿圍與其他體位指標，找出與台灣成人代謝危險因子相關的體位測量指標搭配方式；並探討腰腿比指標與糖尿病及代謝特徵因子的關連性，以及腰腿比指標搭配BMI，與身體脂肪、瘦肉分布之相關性。
材料方法：使用台灣營養健康狀況變遷調查（Nutrition and Health Survey in Taiwan, NAHSIT ）2005-2008資料進行分析，以參與體檢及完成問卷男性1269人，女性1303人為第一階段資料分析的研究對象，減維度迴歸分析（Reduced rank regression）找出與代謝危險因子相關的體位指標搭配方式，依接受者特徵操作型曲線（Receiver operation characteristic curve）找出的腰腿比指標於糖尿病之最適切點並搭配身體質量指數(Body mass index)切點進行分組，利用邏輯斯迴歸（Logistic regression model）探討各組與糖尿病之關連。第二階段資料分析採用參與體檢及雙能量骨密度儀（Dual-energy X-ray, DXA）測量且完成問卷男性561人，女性565人為研究對象，以複迴歸（Multiple regression analyses）探討經DXA測量獲得身體軀幹、腿部的脂肪與瘦肉分布情形。
結果：減維度迴歸分析結果顯示於BMI正常(BMI＜24)及過重(BMI≧24)者中，腰圍越粗腿圍越細者糖尿病風險越高，BMI正常者OR值為32.34(95%CI：7.47~139.98)、BMI過重者OR值為5.83（95%CI：3.7~9.19）。男女性腰腿比指標於糖尿病最適切點為1.76，其敏感度與特異度男性為0.78、0.33，女性為0.78、0.27。邏輯斯迴歸顯示男女性腰腿比指標大於1.76會相較於小於1.76者，罹患糖尿病的風險較高，男性BMI正常與過重者，其OR值為3.65（95%CI：1.92~6.92）、5.01（95%CI：2.77~9.05），女性BMI正常與過重者，其OR值為8.01（95%CI：3.59~17.83）、10.13（95%CI：4.85~21.16）。複迴歸分析DXA軀幹與腿部的脂肪、瘦肉結果顯示BMI正常的男性，腰腿比高於1.76者較小於1.76者，腿部的瘦肉顯著較少，其β值為-0.1（P值0.0238）
結論：腰圍越粗且大腿越細者，罹患糖尿病之風險最高。腿部肌肉對男性罹患糖尿病風險之影響較女性明顯。

關鍵字

體位指標 ； 腰腿比 ； 脂肪量 ； 瘦肉量 ； 糖尿病 ； 代謝症候群

並列摘要

Purpose：The purpose of this study is to research for diabetes and metabolic syndrome associated anthropometric in Taiwanese population.
Method：The data was obtained from Nutrition and Health Survey in Taiwan（NAHSIT）2005-2008.We used reduced rank regression to investigate metabolic syndrome associated anthropometric index. Receiver operation characteristic（ROC） analysis was applied to identify the optimal cut-off points of anthropometric index, waist-to-thigh ratio (WTR). Logistic regression was used to analyze type 2 diabetes risk of different WTR and body mass index groups. Body fat mss and lean mass data were obtained by dual-energy X-ray (DXA), multiple regression was used to compare the fat mass and lean mass composition of trunk and leg in different WTR and body mass index groups.
Results：In normal BMI (18.5≦BMI＜24) and overweight (BMI≧24) groups, reduced rank regression show the characteristics of high waist circumference and thigh circumference were significantly associated with type 2 diabetes. The optimal cut-off points of waist-to-thigh ratio for type 2 diabetes was 1.76, with sensitivity 0.78 and 1-specificity 0.33 in male, and sensitivity 0.78 and 1-specificity 0.27 in female. In BMI normal group, WTR was associated with diabetes (male: OR=3.65, 95%CI: 1.92~6.92; female: OR=8.01, 95%CI: 3.59~17.83), and also in overweight group, WTR was associated with diabetes (male: OR=5.01, 95%CI: 2.77~9.05; female: OR=10.13, 95%CI: 4.85~21.16). Multiple regression analyses showed, in male with normal BMI, subjects with WTR>1.76, whose leg lean mass was significantly lower than WTR≦1.76 group (standard β：-0.1；P-value＝0.0238).
Conclusion：Subjects with higher WTR would increase the risk of type 2 diabetes. Male with higher WTR would have less lean mass in leg.

並列關鍵字

Anthropometric index, ； Waist-to-thigh ratio ； fat mass ； lean mass ； diabetes ； metabolic syndrome

參考文獻

2. Knowles, K.M., et al., Waist Circumference, Body Mass Index, and Other Measures of Adiposity in Predicting Cardiovascular Disease Risk Factors among Peruvian Adults. Int J Hypertens, 2011. 2011: p. 931402.

14. 行政院衛生署國民健康局. 修正前後成人代謝症候群之判定標準. 2007; Available from: http://www.bhp.doh.gov.tw.

15. Welborn, T.A., S.S. Dhaliwal, and S.A. Bennett, Waist-hip ratio is the dominant risk factor predicting cardiovascular death in Australia. Med J Aust, 2003. 179(11-12): p. 580-5.

1. Scott M. Grundy, D.B., et al., National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). the National Cholesterol Education Program Coordinating Committee, May 2001.

3. Ford, E.S., Risks for all-cause mortality, cardiovascular disease, and diabetes associated with the metabolic syndrome: a summary of the evidence. Diabetes Care, 2005. 28(7): p. 1769-78.