基于孟德尔随机化探索臀围与2型糖尿病的因果关系

李文超; 李洪凯; 刘新辉; 司书成; 于媛媛; 李云霞; 袁同慧; 侯蕾; 刘璐; 周宇畅; 薛付忠; 刘言训

doi:10.16462/j.cnki.zhjbkz.2020.01.003

基于孟德尔随机化探索臀围与2型糖尿病的因果关系

doi: 10.16462/j.cnki.zhjbkz.2020.01.003

250012 济南, 山东大学公共卫生学院生物统计系/山东大学健康医疗大数据研究院

基金项目: 国家自然科学基金(81773547)

详细信息

通讯作者:
刘言训, E-mail:liu-yx@sdu.edu.cn

中图分类号: R587.1
计量
- 文章访问数: 523
- HTML全文浏览量: 429
- PDF下载量: 119
- 被引次数: 0
出版历程
- 收稿日期: 2019-07-22
- 修回日期: 2019-10-25
- 刊出日期: 2020-01-10

Exploring the causal relationship between hip circumference and type 2 diabetes based on mendelian randomization

Department of Biostatistics, School of Public Health, Shandong University and Healthcare Big Data Institute of Shandong University, Jinan 250012, China

Funds: National Natural Science Foundation of China(81773547)

More Information

Corresponding author: LIU Yan-xun, E-mail:liu-yx@sdu.edu.cn

摘要

摘要: 目的采用孟德尔随机化分析方法, 探讨臀围与2型糖尿病的因果关联。方法将分别来自人体测量特征遗传研究(genetic investigation of anthropometric traits, GIANT)和糖尿病遗传学验证和荟萃分析(diabetes genetics replication and Meta-analysis, DIAGRAM)数据库的臀围和2型糖尿病遗传位点数据, 据单核苷酸多态性(single nucleotide polymorphism, SNP)编号匹配合并, 以与臀围密切相关的SNP为工具变量, 运用逆方差加权法、MR-Egger回归模型和加权中位数法分析臀围对2型糖尿病的因果效应。结果不分性别队列、女性队列和男性队列中分别匹配到52、9和15个SNP。异质性检验结果提示SNP间同质。三个队列的OR值及其95%的置信区间分别为1.065(1.030~1.100)、1.103(1.057~1.150)和1.583(1.273~1.968), 均有统计学意义(均有P < 0.05)。敏感性分析结果显示三个队列逐一移除SNP, 其因果效应不受影响, 结果稳健。结论人体臀围与2型糖尿病之间均存在因果关联, 臀围可能为2型糖尿病的危险因素。
- 孟德尔随机化 /
- 臀围 /
- 2型糖尿病 /
- 因果推断
Abstract: Objective To investigate the causal association between hip circumference(HC) and type 2 diabetes mellitus(T2 DM) based on Mendelian randomization. Methods The genetic variants data of the HC and T2 DM from the Genetic Investigation of Anthropometric Traits(GIANT) and DIAbetes Genetics Replication And Meta-analysis(DIAGRAM) database were matched according to the single nucleotide polymorphism(SNP) rsID. Genetic loci strongly related to the HC were used as instrumental variables; and the inverse-variance weighting, MR-Egger regression model and weighting median method were carried out to analyze the causal effect of HC on T2 DM. Results Fifty-two, nine and fifteen SNPs were matched in the total cohort, female cohort and male cohort, respectively. Heterogeneity test suggested the SNPs were homogeneous. We found HC to be positively associated with T2 DM risk(OR=1.065, 95% CI: 1.030-1.100, OR=1.103, 95% CI: 1.057-1.150 and OR=1.583, 95% CI: 1.273-1.968, respectively) in above three cohorts, respectively. Sensitivity analysis showed the results were robust. Conclusions There is a relationship between HC and T2 DM of people, and HC may be the risk factor of T2 DM.
- Mendelian randomization /
- Hip circumference /
- Type 2 diabetes mellitus /
- Causal inference

HTML全文

图 1 工具变量的有向无环示意图

Figure 1. Directed acyclic graph of instrumental variable

下载: 全尺寸图片幻灯片

图 2 在各队列中HC与T2DM间OR(95% CI)值的森林图

Figure 2. Forest plots of the value of OR(95% CI) of HC and T2DM in different cohort

下载: 全尺寸图片幻灯片

表 1 各SNP对HC和T2DM的效应值

Table 1. SNPs and magnitude effect on HC and T2DM

SNP	有效等位基因	其他等位基因	有效等位基因频率	HC			T2DM
SNP	有效等位基因	其他等位基因	有效等位基因频率	β	s_x	P值	β	s_x	P值
rs10132280	C	A	0.666 7	0.022 0	0.003 8	4.8×10^-9	0.029 6	0.019 8	0.120 0
rs10182181	A	G	0.500 0	-0.023 0	0.003 5	6.4×10^-11	0.019 8	0.017 6	0.340 0
rs10929925	C	A	0.650 0	0.019 0	0.003 6	4.5×10^-8	0.048 8	0.019 4	0.016 0
rs10938397	G	A	0.433 3	0.030 0	0.003 7	9.3×10^-17	0.029 6	0.019 8	0.110 0
rs11057405	A	G	0.091 7	-0.040 0	0.006 3	2.1×10^-10	0.095 3	0.046 5	0.055 0
rs11165623	A	G	0.483 3	0.022 0	0.003 5	3.0×10^-10	0.039 2	0.017 3	0.048 0
rs11672660	T	C	0.175 0	-0.028 0	0.004 5	7.1×10^-10	0.019 8	0.027 4	0.390 0
rs12086130	C	T	0.900 0	-0.038 0	0.006 3	2.6×10^-9	0.095 3	0.032 5	0.005 0
rs1294409	T	C	0.616 7	-0.020 0	0.003 7	3.0×10^-8	<0.000 1	0.017 8	0.980 0
rs13098327	A	G	0.183 3	0.027 0	0.004 4	1.2×10^-9	0.019 8	0.020 0	0.400 0
rs1351394	C	T	0.516 7	-0.023 0	0.003 5	6.4×10^-11	0.029 6	0.019 8	0.120 0
rs13695	C	T	0.725 0	-0.024 0	0.004 4	4.3×10^-8	0.010 0	0.025 3	0.800 0
rs143384	G	A	0.400 0	0.026 0	0.003 8	3.9×10^-12	0.019 8	0.022 4	0.230 0
rs1516725	C	T	0.908 3	0.031 0	0.005 4	1.0×10^-8	0.067 7	0.028 6	0.018 0
rs1548457	C	T	0.375 0	0.025 0	0.004 5	1.7×10^-8	0.019 8	0.020 0	0.340 0
rs1561277	A	C	0.225 0	-0.024 0	0.004 1	6.0×10^-9	0.019 8	0.020 0	0.360 0
rs16894959	C	T	0.100 0	0.037 0	0.004 9	9.6×10^-14	0.010 0	0.025 3	0.740 0
rs16905212	C	T	0.325 0	-0.021 0	0.003 8	1.9×10^-8	0.010 0	0.020 2	0.740 0
rs17024393	C	T	0.041 7	0.063 0	0.010 0	2.8×10^-10	0.157 0	0.056 4	0.006 0
rs17066856	T	C	0.866 7	0.036 0	0.006 3	6.3×10^-9	0.067 7	0.033 4	0.041 0
rs1808579	C	T	0.525 0	0.023 0	0.003 5	2.1×10^-10	0.029 6	0.019 8	0.130 0
rs2112347	T	G	0.625 0	0.025 0	0.003 6	9.7×10^-12	0.039 2	0.017 1	0.025 0
rs2206277	T	C	0.091 7	0.039 0	0.004 6	3.1×10^-17	0.039 2	0.024 5	0.097 0
rs2293576	G	A	0.633 3	0.023 0	0.003 8	2.1×10^-9	0.010 0	0.017 6	0.520 0
rs2301573	C	T	0.075 0	0.035 0	0.006 2	2.1×10^-8	<0.000 1	0.030 6	0.970 0
rs2820443	T	C	0.700 0	-0.034 0	0.003 9	4.9×10^-18	0.058 3	0.019 3	0.004 0
rs3087591	G	A	0.375 0	0.021 0	0.003 8	2.5×10^-8	0.010 0	0.020 2	0.490 0
rs355838	T	G	0.366 7	-0.022 0	0.003 6	1.6×10^-9	0.039 2	0.017 3	0.029 0
rs3800229	T	G	0.691 7	0.021 0	0.003 8	3.2×10^-8	0.019 8	0.020 0	0.390 0
rs3888190	A	C	0.358 3	0.035 0	0.003 6	9.4×10^-22	0.010 0	0.017 8	0.770 0
rs4132228	C	T	0.766 7	-0.021 0	0.003 9	3.1×10^-8	0.067 7	0.019 1	0.001 0
rs4883723	A	G	0.125 0	0.032 0	0.005 1	5.2×10^-10	0.067 7	0.028 6	0.015 0
rs4889606	G	A	0.358 3	-0.021 0	0.003 6	8.8×10^-9	0.010 0	0.015 2	0.590 0
rs543874	G	A	0.266 7	0.045 0	0.004 5	1.8×10^-23	0.039 2	0.022 2	0.093 0
rs6163	C	A	0.608 3	-0.021 0	0.003 7	1.6×10^-8	0.019 8	0.017 6	0.400 0
rs6265	C	T	0.825 0	0.034 0	0.004 5	1.5×10^-14	0.019 8	0.020 0	0.420 0
rs6569648	T	C	0.758 3	-0.029 0	0.004 2	6.5×10^-12	0.067 7	0.019 1	0.001 0
rs663129	A	G	0.283 3	0.050 0	0.004 2	1.4×10^-32	0.067 7	0.021 6	0.002 0
rs6678622	C	T	0.583 3	-0.022 0	0.003 7	1.1×10^-9	<0.000 1	0.020 4	0.990 0
rs6755502	C	T	0.875 0	0.054 0	0.004 7	2.4×10^-30	0.019 8	0.022 4	0.320 0
rs7138803	A	G	0.441 7	0.029 0	0.003 7	1.9×10^-15	0.039 2	0.017 3	0.037 0
rs7144011	T	G	0.275 0	0.030 0	0.004 2	8.6×10^-13	0.067 7	0.019 1	0.001 0
rs7183263	G	T	0.475 0	0.019 0	0.003 6	4.8×10^-8	0.019 8	0.020 0	0.380 0
rs7531118	C	T	0.608 3	0.024 0	0.003 7	4.0×10^-11	0.010 0	0.017 6	0.450 0
rs7632381	T	C	0.516 7	-0.036 0	0.003 5	5.3×10^-24	0.039 2	0.017 1	0.015 0
rs7903146	T	C	0.250 0	-0.026 0	0.003 9	2.1×10^-11	0.336 5	0.020 0	5.5×10^-65
rs798528	C	A	0.300 0	-0.021 0	0.003 8	1.7×10^-8	<0.000 1	0.020 4	0.990 0
rs806794	G	A	0.275 0	-0.032 0	0.004 0	2.8×10^-16	0.010 0	0.020 2	0.740 0
rs879620	C	T	0.408 3	-0.029 0	0.004 7	8.9×10^-10	0.010 0	0.020 2	0.740 0
rs887912	T	C	0.316 7	0.022 0	0.003 9	1.6×10^-8	0.029 6	0.019 8	0.160 0
rs951252	G	A	0.525 0	-0.028 0	0.003 5	4.5×10^-15	0.010 0	0.020 2	0.580 0
rs9939973	A	G	0.475 0	0.070 0	0.003 6	2.3×10^-86	0.095 3	0.016 2	1.2×10^-8

下载: 导出CSV

表 2 leave-one-out法敏感性分析结果

Table 2. Result of "leave one out" method

除去某SNP	OR(95% CI)值	除去某SNP	OR(95% CI)值
rs6755502	1.655 (1.304~2.100)	rs10784502	1.444(1.163~1.791)
rs9846396	1.627(1.314~2.016)	rs3118910	1.190(1.077~1.314)
rs13130484	1.627(1.320~2.005)	rs17109256	1.542(1.233~1.927)
rs951252	1.597(1.284~1.986)	rs4788102	1.629(1.323~2.005)
rs806794	1.581(1.273~1.964)	rs12936587	1.619(1.224~2.141)
rs2744937	1.583(1.275~1.966)	rs6567160	1.470(1.130~1.912)
rs2206277	1.570(1.262~1.954)	rs143384	1.622(1.313~2.003)
rs6569648	1.545(1.232~1.937)

下载: 导出CSV

参考文献(24)

[1]	Seuring T, Archangelidi O, Suhrcke M. The economic costs of type 2 diabetes: a global systematic review[J]. PharmacoEconomics, 2015, 33(8): 811-831. DOI: 10.1007/s40273-015-0268-9.
[2]	Global report on diabetes[R]. World Health Organization, 2016. https://www.who.int/diabetes/publications/grd-2016/en/index1.html.
[3]	Cameron AJ, Magliano DJ, Söderberg S. A systematic review of the impact of including both waist and hip circumference in risk models for cardiovascular diseases, diabetes and mortality[J]. Obes Rev, 2013, 14(1): 86-94. DOI: 10.1111/j.1467-789X.2012.01051.x.
[4]	江流芳, 周蓦, 佟伟军.臀围与2型糖尿病发病的关联: Meta分析[J].上海医药, 2014, 37(1): 60-64. Jiang LF, Zhou M, Tong WJ. Correlation between hip circumference and type 2 diabetes: a meta-analysis[J]. Shanghai Med J, 2014, 37(1): 60-64.
[5]	李杰, 钱玲玲, 汤在祥, 等. 2型糖尿病患者臀围与代谢综合征危险因素的相关性分析[J].中国糖尿病杂志, 2016, 24(8): 673-677. DOI: 10.3969/j.issn.1006-6187.2016.08.001. Li J, Qian LL, Tang ZX, et al. Association between hip circumference and metabolic syndrome in Chinese patients with type 2 diabetes[J]. Chin J Diabetes, 2016, 24(8): 673-677. DOI: 10.3969/j.issn.1006-6187.2016.08.001.
[6]	刘洋, 刘永跃, 王健, 等.蒙古族居民臀围与代谢综合征关系[J].中国公共卫生, 2012, 28(1): 25-27. DOI: 10.11847/zgggws2012-28-01-10. Liu Y, Liu YY, Wang J, et al. Association of hip circumference with metabolic syndrome in Chinese Mongolian population[J]. Chin J Publ Heal, 2012, 28(1): 25-27. DOI: 10.11847/zgggws2012-28-01-10.
[7]	Janghorbani M, Amini M. Associations of hip circumference and height with incidence of type 2 diabetes: the Isfahan diabetes prevention study[J]. Acta Diabetol, 2012, 49(1): 107-114. DOI: 10.1007/s00592-011-0351-4.
[8]	Katz EG, Stevens J, Truesdale KP, et al. Hip circumference and incident metabolic risk factors in Chinese men and women: the People's Republic of China study[J]. Metab Syndr Relat Disord, 2011, 9(1): 55-62. DOI: 10.1089/met.2010.0045.
[9]	Lawlor DA, Harbord RM, Sterne JA, et al. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology[J]. Stat Med, 2008, 27(8): 1133-1163. DOI: 10.1002/sim.3034.
[10]	Locke AE, Kahali B, Berndt SI, et al. Genetic studies of body mass index yield new insights for obesity biology[J]. Nature, 2015, 518(7538): 197-206. DOI: 10.1038/nature14177.
[11]	Morris AP, Voight BF, Teslovich TM, et al. Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes[J]. Nat Genet, 2012, 44(9): 981-990. DOI: 10.1038/ng.2383.
[12]	Noyce AJ, Kia DA, Hemani G, et al. Estimating the causal influence of body mass index on risk of parkinson disease: a mendelian randomisation study[J]. PLoS Med, 2017, 14(6): e1002314. DOI: 10.1371/journal.pmed.1002314.
[13]	Bowden J, Davey SG, Burqess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression[J]. Int J Epidemiol, 2015, 44(2): 512-525. DOI: 10.1093/ije/dyv080.
[14]	Zhu Z, Zhang F, Hu H, et al. Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets[J]. Nat Genet, 2016, 48(5): 481-487. DOI: 10.1038/ng.3538.
[15]	White J, Swerdlow DI, Preiss D, et al. Association of lipid fractions with risks for coronary artery disease and diabetes[J]. JAMA Cardiol, 2016, 1(6): 692-699. DOI: 10.1001/jamacardio.2016.1884.
[16]	Bowden J, Davey Smith G, Haycock PC, et al. Consistent estimation in mendelian randomization with some invalid instruments using a weighted median estimator[J]. Genet Epidemiol, 2016, 40(4): 304-314. DOI: 10.1002/gepi.21965.
[17]	Vazquez G, Duval S, Jacobs DR, et al. Comparison of body mass index, waist circumference, and waist/hip ratio in predicting incident diabetes: a meta-analysis[J]. Epidemiol Rev, 2007, 29(1): 115-128. DOI: 10.1093/epirev/mxm008.
[18]	王燕, 阿布都克尤木·赛买提, 张文杰.臀围在筛查新疆喀什贫困地区维吾尔族妇女肥胖中的应用研究[J].第三军医大学学报, 2017, 39(3): 296-300. DOI: 10.16016/j.1000-5404.201611167. Wang Y, Abudukeyoumu SMT, Zhang WJ. Application of hip circumference in screening Uyghur obese women in poor areas of Kashi, Xinjiang[J]. J Third Military Medical Univ, 2017, 39(3): 296-300. DOI: 10.16016/j.1000-5404.201611167.
[19]	周帆, 王先令, 窦京涛.从2017年版ADA指南看2型糖尿病患者肥胖管理新策略[J].药物评价, 2018, 15(3): 10-13, 28. Zhou F, Wang XL, Dou JT. A new strategy for obesity management in type 2 diabetes mellitus from the 2017 ADA standards of med ical care in diabetes[J]. Drug Evaluation, 2018, 15(3): 10-13, 28.
[20]	苏清清, 皮红英, 闻智, 等. 2型糖尿病患者体质量控制及自我管理行为现状调查[J].解放军医学院学报, 2019, 40(3): 207-210, 251. DOI: 10.3969/j.issn.2095-5227.2019.03.002. Su QQ, Pi HY, Wen Z, et al. Current status of weight control and self-management behaviors in patients with type 2 diabetes mellitus[J]. Academic J Chin PLA Medical School, 2019, 40(3): 207-210, 251. DOI: 10.3969/j.issn.2095-5227.2019.03.002.
[21]	Boo S, Yoon YJ, Oh H. Evaluating the prevalence, awareness, and control of hypertension, diabetes, and dyslipidemia in Korea using the NHIS-NSC database: A cross-sectional analysis[J]. Medicine, 2018, 97(51): 13713. DOI: 10.1097/MD.0000000000013713.
[22]	Kolahdooz F, Nader F, Daemi M, et al. Prevalence of known risk factors for type 2 diabetes mellitus in multiethnic urban youth in Edmonton: findings from the why act now Project[J]. Can J Diabetes, 2019, 43(3): 207-214. DOI: 10.1016/j.jcjd.2018.10.002.
[23]	Smokovski I, Milenkovic T, Cho NH. First stratified diabetes prevalence data for republic of macedonia derived from the national ehealth system[J]. Diabetes Res Clin Pract, 2018, 143: 179-183. DOI: 10.1016/j.diabres.2018.07.015.
[24]	Alkandari A, Longenecker JC, Barengo NC, et al. The prevalence of pre-diabetes and diabetes in the Kuwaiti adult population in 2014[J]. Diabetes Res Clin Pract, 2018, 144: 213-223. DOI: 10.1016/j.diabres.2018.08.016.