喝咖啡、身體質量指數和單核苷酸多態性與骨質疏鬆症的相關性探討

Association of coffee drinking, body mass index, and single nucleotide polymorphism with osteoporosis

吳綺玲(Chi-Ling Wu)

指導教授 ： 廖勇柏

中山醫學大學/健康管理學院/公共衛生學系/博士(2023年)

https://doi.org/10.6834/csmu202300014

摘要

背景
骨質疏鬆症是一種退化性疾病，影響所有種族的女性和男性。骨質疏鬆症的發展部分是由遺傳和生活方式或環境因素之間的相互作用來解釋的。到 2011 年，25.0% 的台灣人患有骨質疏鬆症，比 2001 年的 17.4% 有所增加。到 2050 年可能有 2.12 億人患有骨質疏鬆症，預計亞洲將佔全球所有髖部骨折的約 51.1%。其次，咖啡是世界上越來越普遍且受歡迎的一種生活習慣。另外，隨著人類生活的飲食運動習慣的改變，身體質量指數(BMI)也隨之有所改變。在精準公共衛生的領域中，孟德爾隨機理論在遺傳變異是一個很重要的影響因素。

目的
因此，本論文之目的在於喝咖啡、身體質量指數 (BMI) 與單核苷酸多態性(SNP)來探討與骨質疏鬆症的相關性至為重要。故本篇論文使用台灣人體生物資料庫 (TWB) 的數據研究了喝咖啡、身體質量指數 (BMI)、與單核苷酸多態性來探討與骨質疏鬆症之間的關聯。之前的研究除了從全球的視野或其他種族探討骨質疏疏鬆的相關性，很少屬於台灣地域性的研究探討喝咖啡、身體質量指數 (BMI) 與單核苷酸多態性來探討與骨質疏鬆症的相關性。而本篇論文以台灣地域性的角度去探討以喝咖啡、身體質量指數 (BMI) 與單核苷酸多態性來探討與骨質疏鬆的相關性。綜合本論文的研究價值及建議是對於這些結果提供了對導致台灣骨質疏鬆症的遺傳和生活方式因素的深入了解，並可作為評估該疾病及其相關變量的研究的參考。使更能達對骨質疏鬆症可傷性，易感性及可預測性，更具精準公共衛生的精神所在。
方法
本論文就『研究時間』及『探討的最主要影響因素（ exposure ）』不同，分別為兩個部分：第一部分為喝咖啡與單核苷酸多態性對骨質疏鬆症的相關性探討，第二部分為身體質量指數(BMI)與單核苷酸多態性對骨質疏鬆症的相關性探討。
第一部份，在喝咖啡與單核苷酸多態性對骨質疏鬆的相關性探討研究中，其設計參與者和設置基於人群的橫斷面研究，我們使用了 2016 年至 2019 年間在台灣人體生物資料庫 (TWB)招募的參與者的遺傳、人口統計和生活方式數據。使用多元邏輯回歸份析來確定兩者之間的關係骨質疏鬆症和變異 rs2982573 基因型（TT、TC 和 CC）。
而第二部份，在身體質量指數 (BMI) 與單核苷酸多態性對骨質疏鬆症的相關性探討研究中，我們分析了 在台灣人體生物資料庫 (TWB)招募的參與者10,943 名 30 至 70 歲受試者的數據。 根據髖部的平均 T 評分為 -2.5 及以此來定義骨質疏鬆症。 身體質量指數是按照衛生福利部國民健康署的建議計算的。使用IMPUTE2 (v2.3.1) 程序進行插補。使用多元邏輯回歸進行分析。確定了骨質疏鬆症的優勢比 (OR) 和 95% 置信區間 (CI)。

結果
喝咖啡與單核苷酸多態性對骨質疏鬆症的相關探討中，主要結果是我們確定了台灣 ESR1 rs2982573 個體的基因分型，咖啡攝入量與骨質疏鬆症風險之間的關係。
患有骨質疏鬆症的參與者 (n = 515) 比沒有疾病的參與者年紀更大（平均年齡± SE（年）；61.342 ± 0.361 對 53.068 ±0.130， p<0.001 ）。 rs2982573與骨質疏鬆症之間沒有顯著關聯（OR，0.904；95% CI，0.706-1.157；與 TT 基因型相比，TC+CC 的p=0.422 ）。喝咖啡與較低的骨質疏鬆症風險相關（OR，0.737；95% CI，0.592–0.918； p=0.006） 。在我們的分型分析中，與他們的 TT 相比，喝咖啡的 TC+CC 個體和不喝咖啡的 TC+CC 個體的調整後 OR ( 95% CI ) 分別為 0.635 (0.410–0.985) 和 1.095 (0.809–1.482)基因型對應物。

而在身體質量指數(BMI)與單核苷酸多態性對骨質疏鬆症的相關探討中，主要結果是在多元回歸模型中，　rs2908004 對骨質疏鬆症具有顯著的保護作用（rs2908004 基因分型GA+AA 與 GG：OR，0.650；95% CI = 0.543 至 0.778）。與正常體重相比，體重過輕與更高的骨質疏鬆症風險顯著相關（OR，6.517；95% CI = 4.624 至 9.186），而超重和肥胖具分別有保護作用（OR，0.176；95% CI = 0.140 至 0.221 和 0.057； 95% CI = 0.039 至 0.083。 rs2908004 和 身體質量指數(BMI) 之間存在交互作用（p = 0.0148）。分型分析（使用 rs2908004 基因分型GG/正常體重作為參考組）表明 rs2908004 基因分型GG/體重不足組的 OR 為 7.66（95% CI = 5.153 至 11.394），而在rs2908004 基因分型GA+AA/體重不足組（95% CI=1.509 至 5.974）。 rs2908004 基因分型GG/肥胖組、rs2908004 基因分型GG/超重組、GA+AA/正常體重組、rs2908004 基因分型GA+AA/超重組和rs2908004 基因分型GA+AA/肥胖組的OR顯著降低。

由 身體質量指數(BMI) 和r s2908004 基因型定義的分型的 OR 如表 9 所示。與 GG 基因型相比， GA+AA 個體中骨質疏鬆症的 OR 為0.747（95% CI = 0.608-0.918） 。在體重過輕組中為 0.435 (95% CI = 0.196–0.963)組，超重組為 0.471（95% CI = 0.285-0.780），肥胖組為 0.307（95% CI = 0.115-0.818）。

與女性相比，骨質疏鬆症的校正 OR 為正常體重者 0.108 (95% CI = 0.075-0.156)，體重過輕者 1.630 (95% CI = 0.559-4.753)，在超重時為 0.048 (95% CI = 0.020–0.119) , 和 0.194 (95% CI = 0.063–0.598) 分別在肥胖男性中。分型分析（使用 rs2908004 基因分型GG/正常體重作為參考組）表明 rs2908004 基因分型GG/體重過輕組的 OR 為 7.672 (95% CI = 5.158–11.410)，而在rs2908004 基因分型GA+AA/體重過輕組（95% CI=1.509-5.974）。 rs2908004 基因分型GG/超重組的相應 OR (95% CI) 為 0.197 (0.152–0.256)，rs2908004 基因分型GG/肥胖組為 0.071 (0.047-0.107)，rs2908004 基因分型GA 為 0.747 (0.608-0.919) +AA/正常體重組，rs2908004 基因分型GA+AA/超重組為0.095（0.60-0.150），rs2908004 基因分型GA+AA/肥胖組為0.022（0.009-0.053）(表 10)。。

結論
根據喝咖啡與單核苷酸多態性對骨質疏鬆症的相關性探討的研究，結論是具有 TC+CC 基因型 ESR1 rs2982573 的 TWB 參與者每周至少喝三杯咖啡的人患骨質疏鬆症的可能性較小。而在身體質量指數(BMI)與單核苷酸多態性對骨質疏鬆症的相關性探討的研究，結論是對於 GG 和 GA+AA 基因型，體重不足與骨質疏鬆症風險增加有關，而超重和肥胖與風險較低有關。

關鍵字

喝咖啡 ； 身體質量指數 ； 單核苷酸多態性 ； 骨質疏鬆症

並列摘要

Background
Osteoporosis is a degenerative disease that affects women and men of all races. Osteoporosis is partially explained by the interaction of genes with lifestyle and environmental factors. In 2011, 25.0% of Taiwanese had osteoporosis, an increase from 17.4% in 2001. By 2050, 212 million people could be living with the condition and Asia is expected to account for approximately 51.1% of all hip fractures worldwide. Coffee is an increasingly common and popular lifestyle habit around the world. In addition, as the eating and exercise habits of human beings change, the body mass index (BMI) also changes accordingly. In the field of precision public health, Mendelian randomization theory is important influencing factor in genetic variation.

Objectives
Therefore, drinking coffee, body mass index (BMI), and Single-nucleotide polymorphism (SNP) to explore the correlation with osteoporosis is very important. Therefore, this article uses data of Taiwan Biobank (TWB) to study coffee drinking, body mass index (BMI), and SNP to explore the association with osteoporosis. In this way, in addition to exploring the association of osteoporosis from a global perspective, it is also possible to explore the association of coffee drinking, body mass index (BMI), and SNP with osteoporosis from a regional perspective in Taiwan. To achieve the vulnerability, susceptibility, and predictability of osteoporosis in precision public health.
The paper is divided into two parts, depending on the time of research and the most important influence factor (exposure)：The first part discusses the relationship between drinking coffee and Single-nucleotide polymorphism (SNP) in osteoporosis. In addition, the second part of the study deals with the relationship between BMI and SNP in terms of the relationship with osteoporosis. In this cross-sectional study, participants, and settings based on the population are studied. And, we used individuals recruited from the Taiwan Biobank (TWB) between 2016 and 2019. To explore the correlation between coffee consumption and the Single-nucleotide polymorphism (SNP) associated with osteoporosis, along with participants’ genetic, demographic, and lifestyle data. Using multiple logistic regression analyses, we determined the relationship between rs2982573 (TT, TC, CC) and osteoporosis variants. In a study of the relationship between BMI and SNP with materials and methods for osteoporosis, we analyzed data from 10,943 subjects aged 30 to 70. The definition of osteoporosis was based on the average T score below -2.5 on the hips. The BMI is calculated in accordance with the guidelines of t Health Promotion Administration (MOHW). The imputation was performed using the IMPUTE2 program (v2.3.1). The analysis was done by multiple logistic regressions. Odds ratios (OR) and 95% confidence intervals (CI) for osteoporosis were determined.

Method
This thesis is divided into two parts according to the research time and the most important influencing factors (exposure): the first part discusses the relationship between drinking coffee and SNP in osteoporosis, and the second part discusses the relationship between BMI and SNP on osteoporosis correlation study.
The first part explores the relationship between coffee drinking and SNP in osteoporosis. In this design, participants and the set population-based cross-sectional study, we used genetic, demographic and lifestyle data from participants recruited in the Taiwan Biobank (TWB) between 2016 and 2019. And we used multiple logistic regression analyzes to determine the relationship between osteoporosis and variant genotypes rs2982573 (TT, TC, and CC).In the second part of the study on the relationship between BMI and SNP on osteoporosis, in the materials and methods, we analyzed data from 10,943 subjects aged 30 to 70 years. We defined osteoporosis based on a mean T score of -2.5 and below in the hip. The body mass index was calculated following the guidelines of the Health Promotion Administration. The imputation was performed using the IMPUTE2 (v2.3.1) program. Multiple logistic regression was used for the analysis. The odds ratios (OR) and 95% confidence interval (CI) for osteoporosis were determined.

Result
In the association between coffee drinking and Single-nucleotide polymorphism (SNP) with osteoporosis, the main result was that we identified an association between coffee intake and osteoporosis risk in Taiwanese ESR1 rs2982573 individuals. Individuals with osteoporosis (n = 515) were older than those without the disease (mean age ±SE (year); 61.324±0.361 versus 53.068 ±0.130, p<0.001). There was no significant relationship between rs2982573 and osteoporosis (OR 0.904, (95% CI) 0.706–1.157, p=0.422 between TC + CC and TT genotype). The risk of osteoporosis was reduced by coffee consumption (0.737, 95% confidence interval 0.592–0.918, p=0.006). In our subgroup analyzes, the adjusted ORs (95% CI) were 0.635 (0.410–0.985) in people drinking coffee TC + CC and 1.095 (0.809–1.482) in non-coffee drinking TC+CC individuals, respectively when compared to their TT genotype counterparts.

In the results on the association between BMI and SNP with osteoporosis, the main result is that in the multivariate regression model, the variant rs2908004 was significantly protective against osteoporosis (GA+AA 與 GG：OR，0.650；95% CI = 0.543 - 0.778). Compared to normal weight, underweight was significantly associated with a higher risk of osteoporosis (OR，6.517；95% CI = 4.624 至 9.186), while overweight and obesity were protective respectively (OR，0.176；95% CI = 0.140 至 0.221 和 0.057； 95% CI = 0.039 - 0.083). There was an interaction between rs2908004 and BMI (p =0.0148). Subgroup analyzes (using rs2908004 基因分型GG/normal weight as reference group) indicated ORs of 7.66 (95% CI=5.153 to 11.394) in the rs2908004-GG/underweight group and 3.002 (95% CI=1.509 to 5.974) in the rs2908004-GA+AA/underweight group (95% CI=1.509 to 5.974). The ORs were substantially lower in the rs2908004-GG / obese, rs2908004-GG / overweight, GA + AA / normal weight, rs2908004-GA + AA / overweight and rs2908004-GA+AA/obese groups, respectively.

Conclusion
According to the study exploring the association between coffee consumption and SNP on osteoporosis, it was concluded that participants in TWB with the TC+CC genotype of ESR1 rs2982573 who consumed at least three cups of coffee per week were less likely to have osteoporosis. In contrast, in this study exploring the association between BMI and SNP for osteoporosis, it was concluded that underweight was associated with an increased risk of osteoporosis for both the GG and GA+AA genotypes, while overweight and obesity were associated with a lower risk.

並列關鍵字

coffee drinking ； body mass index ； single nucleotide polymorphism ； osteoporosis

參考文獻

1. Horton, R., Offline: In defence of precision public health. The Lancet, 2018. 392(10157): p. 1504.

2. Liu, J., et al., State of the art in osteoporosis risk assessment and treatment. Journal of endocrinological investigation, 2019. 42(10): p. 1149-1164.

3. Sözen, T., L. Özışık, and N.Ç. Başaran, An overview and management of osteoporosis. European journal of rheumatology, 2017. 4(1): p. 46.

4. Hwang, J.-S., J.-F. Chen, and K.-S. Tsai, Epidemiology of Osteoporosis in Taiwan. Osteoporosis Of The Spine: Asian Perspectives, 2021: p. 1.

5. Chen, F.-P., et al., Secular trends in incidence of osteoporosis in Taiwan: a nationwide population-based study. biomedical journal, 2018. 41(5): p. 314-320.