血清总胆红素与糖尿病肾病因果关系分析

夏淼; 王敏珍; 赵增毅; 邢玉微; 赵维丽; 杨洪燕; 刘赞朝

doi:10.16462/j.cnki.zhjbkz.2023.05.010

血清总胆红素与糖尿病肾病因果关系分析

doi: 10.16462/j.cnki.zhjbkz.2023.05.010

1.
730000 兰州，兰州大学公共卫生学院流行病与卫生统计学研究所
2.
050051 石家庄，河北省糖尿病基础医学研究重点实验室，河北省石家庄市第二医院内分泌科

基金项目:

河北省卫生厅科研基金项目重点科技研究计划 20190156

详细信息

通讯作者:
王敏珍，E-mail: wangmzhen@lzu.edu.cn

刘赞朝, E-mail: liuzanchao2007@163.com

中图分类号: R181.1
计量
- 文章访问数: 210
- HTML全文浏览量: 151
- PDF下载量: 31
- 被引次数: 0
出版历程
- 收稿日期: 2022-02-07
- 修回日期: 2022-06-10
- 刊出日期: 2023-05-10

Casual analysis between serum total bilirubin and diabetic nephropathy

1.
Department of Epidemiology and Statistics, School of Public Health, Lanzhou University, Lanzhou 730000, China
2.
Hebei Province Key Laboratory of Basic Medicine for Diabetes, Department of Endocrinology, The Second Hospital of Shijiazhuang City, Shijiazhuang 050051, China

Funds:

Key Scientific Research Program of Hebei Health Department Scientific Research Fund Project 20190156

More Information

Corresponding author: WANG Minzhen，E-mail: wangmzhen@lzu.edu.cn; LIU Zanchao, E-mail: liuzanchao2007@163.com

摘要

摘要: 目的采用断点回归设计(regression discontinuity design, RDD)探讨血清总胆红素(total bilirubin，TBIL)与糖尿病肾病(diabetic nephropathy，DN)的因果关系。方法以石家庄市第二医院就诊2型糖尿病(type 2 diabetes mellitus, T2DM)患者为研究对象，分析其DN患病现状，应用logistic回归分析模型探讨DN影响因素，建立风险预测模型，得出DN个体患病概率值，以TBIL为分组变量，DN风险概率为结局变量，进行RDD分析。结果 4 173例糖尿病患者中，DN 507例，检出率12.15%。在DN风险预测模型中，预测因素为高脂血症、吸烟、饮酒、饮食控制、运动情况、降糖药物使用、糖尿病病程、尿素氮、SBP、TBIL。DN风险预测模型受试者工作特征(receiver operating characteristic，ROC)曲线下面积(area under curve, AUC)为0.758(95% CI：0.736~0.779)。TBIL是DN的保护因素(OR=0.967，95% CI: 0.948~0.987)；当TBIL=8 μmol/L时，DN风险下降5.02%(P=0.004)。结论 TBIL是DN的保护因素；当TBIL=8 μmol/L时，患DN风险降低。
- 血清总胆红素 /
- 糖尿病肾病 /
- 断点回归设计 /
- 风险预测
Abstract: Objective To explore the causal relationship between serum total bilirubin(TBIL) and diabetic nephropathy (DN) using a regression discontinuity design (RDD) approach. Methods We analyzed data from 4 173 participants with type 2 diabetes mellitus (T2DM) at the Second Hospital of Shijiazhuang City. The prevalence of DN was assessed, and significant influencing factors were selected through logistic regression analysis. A risk prediction model was established, and individual DN risk was calculated. RDD was performed using TBIL as the grouping variable and individual DN risk as the outcome variable. Results Of the 4 173 patients with T2DM, 507 had DN (12.15%). The DN risk prediction model identified hyperlipidemia, smoking, drinking, diet control, physical exercise, hypoglycemic drugs, diabetes course, blood urea nitrogen, systolic blood pressure and TBIL as predictive factors. The area under the curve (AUC) for the model was 0.758(95% CI: 0.736-0.779). TBIL was found to be an protective factor for DN(OR=0.967, 95% CI: 0.948-0.987); at TBIL=8 μmol/L, the risk of DN significantly decreased by 5.02%(P=0.004). Conclusions TBIL is a protective factor for DN, and when TBIL=8 μmol/L, the risk of developing DN sharply declines.
- Total bilirubin /
- Diabetic nephropathy /
- Regression discontinuity design /
- Risk prediction

HTML全文

图 1 logistic回归分析模型计算的患病概率对DN的预测效果ROC曲线图

Figure 1. ROC curve of the predictive effect of disease probability on DN calculated by the logistic regression analysis model

下载: 全尺寸图片幻灯片

图 2 RDD中T2DM患者罹患DN的概率在TBIL=8 μmol/L时的变化

Figure 2. Changes in the probability of DN in T2DM before and after breakpoints in the RDD (TBIL=8 μmol/L)

下载: 全尺寸图片幻灯片

图 3 RDD中TBIL在8 μmol/L前后的密度分布曲线

Figure 3. Density distribution curve of TBIL before and after 8 μmol/L in RDD

下载: 全尺寸图片幻灯片

表 1 研究对象人群基本特征[n(%)]

Table 1. Basic characteristics of the research subjects [n(%)]

变量	合计(N=4 173)	非DN(n=3 666)	DN(n=507)	P值^a
年龄(x±s, 岁)	61.79±12.27	61.70±12.40	62.48±11.24	0.179
性别				0.010
男	2 060(49.36)	1 773(48.36)	287(56.61)
女	2 113(50.64)	1 893(51.64)	220(43.39)
糖尿病病程[M(IQR), 年]	6.00(10.00)	6.00(9.00)	10.00(11.00)	0.010
高血压	1 834(43.95)	1 557(42.47)	277(54.64)	0.010
高脂血症	1 083(25.95)	859(23.43)	224(44.18)	0.010
吸烟	991(23.75)	811(22.12)	180(35.50)	0.010
饮酒	989(23.70)	818(22.31)	171(33.73)	0.010
饮食控制				0.010
不控制	2 620(62.78)	2 369(64.62)	251(49.51)
一般控制	1 415(33.91)	1196(32.62)	219(43.20)
严格控制	138(3.31)	101(2.76)	37(7.30)
运动情况				0.040
不运动	1 242(29.76)	1 091(29.76)	151(29.78)
偶尔运动	1 055(25.28)	973(26.54)	82(16.17)
经常运动	1 876(44.96)	1602(43.70)	274(54.04)
降糖药物使用				0.010
未使用	1 102(26.41)	1 052(28.70)	50(9.86)
单用口服药	1 901(45.55)	1 686(45.99)	215(42.41)
单用胰岛素	479(11.48)	379(10.34)	100(19.72)
口服药联合胰岛素	691(16.56)	549(14.98)	142(28.01)
钙拮抗剂	600(14.38)	498(13.58)	102(20.12)	0.010
ACEI类	254(6.09)	202(5.51)	52(10.26)	0.010
ARB类	294(7.05)	252(6.87)	42(8.28)	0.245
受体阻滞剂	293(7.02)	246(6.71)	47(9.27)	0.034
硝酸酯类	228(5.46)	190(5.18)	38(7.50)	0.032
他汀类	907(21.73)	806(21.99)	101(19.92)	0.291
阿司匹林	943(22.60)	800(21.82)	143(28.20)	0.010
BMI(x±s, kg/m²)	25.82±3.67	25.81±3.67	25.47±3.06	0.018
HbA1c(x±s, g/L)	8.17±1.76	8.16±1.76	8.26±1.79	0.227
FPG(x±s, mmol/L)	9.25±3.45	9.24±3.44	9.34±3.49	0.528
SBP(x±s, mmHg)	131.93±16.01	131.44±15.68	135.44±17.83	0.010
DBP(x±s, mmHg)	80.33±9.86	80.20±9.80	81.25±10.21	0.026
TC(x±s, mmol/L)	4.82±1.20	4.81±1.21	4.85±1.09	0.451
TG[M(IQR), mmol/L]	1.57(0.85)	1.55(0.85)	1.68(0.97)	0.195
HDL-C(x±s, mmol/L)	1.22±0.31	1.22±0.30	1.24±0.36	0.237
LDL-C(x±s, mmol/L)	2.72±0.82	2.72±0.83	2.76±0.79	0.261
BUN(x±s, mmol/L)	5.68±2.71	5.56±2.37	6.58±4.31	0.010
UA(x±s, μmol/L)	280.54±85.86	279.46±85.52	288.38±87.96	0.028
ALT(x±s, U/L)	23.43±16.01	23.61±16.33	22.11±13.48	0.023
AST(x±s, U/L)	22.26±12.99	22.38±13.56	21.31±7.71	0.010
TBIL(x±s, μmol/L)	14.44±5.19	14.60±5.17	13.27±5.23	0.010
注：^a表示t检验、秩和检验或$\chi$ ²检验; 血管紧张素转化酶抑制剂(angiotensin converting enzyme inhibitors，ACEI)；血管紧张素Ⅱ受体拮抗剂(angiotensin Ⅱ receptor antagonist，ARB)；血尿素氮(blood urea nitrogen，BUN)；尿酸(acid uric，UA)；谷丙转氨酶(alanine aminotransferase，ALT)；谷草转氨酶(aspartate transaminase，AST)。

下载: 导出CSV

表 2 DN多因素logistic回归分析模型分析

Table 2. Multivariate logistic regression analysis model analysis of DN

变量	β值	s_x	Wald $\chi$²值	OR值(95% CI)	P值
高脂血症	0.827	0.104	62.751	2.286(1.863~2.805)	< 0.010
吸烟	0.353	0.138	6.545	1.424(1.086~1.867)	0.011
饮酒	0.325	0.139	5.437	1.383(1.053~1.817)	0.020
饮食控制^a
一般	0.340	0.112	9.319	1.406(1.130~1.749)	< 0.010
严格控制	0.746	0.220	11.464	2.109(1.369~3.249)	< 0.010
运动情况^b
偶尔运动	-0.467	0.155	9.142	0.627(0.463~0.848)	< 0.010
经常运动	-0.001	0.125	0.000	0.999(0.782~1.278)	0.997
降糖药物使用^c
单用口服药	0.707	0.171	17.102	2.027(1.450~2.833)	< 0.010
单用胰岛素	1.110	0.199	31.050	3.033(2.053~4.481)	< 0.010
口服药联合胰岛素	1.039	0.189	30.223	2.827(1.952~4.094)	< 0.010
糖尿病病程	0.036	0.007	29.789	1.037(1.023~1.050)	< 0.010
BUN	0.082	0.015	29.496	1.085(1.054~1.117)	< 0.010
SBP(mmHg)	0.010	0.003	11.401	1.010(1.004~1.017)	< 0.010
TBIL(μmol/L)	-0.034	0.010	10.422	0.967(0.948~0.987)	< 0.010
注：^a以饮食“不控制”为参照；^b以运动情况为“不运动”为参照；^c以降糖药情况为“未使用”为参照。

下载: 导出CSV

表 3 RDD中相关协变量的连续性检验

Table 3. Continuity test of correlation covariables in RDD

变量	β值	s_x	Wald值	95% CI	P值
高脂血症	-0.035	0.052	-0.67	-0.137~0.067	0.505
饮食	-0.079	0.071	-1.10	-0.219~0.061	0.270
运动	-0.137	0.109	-1.25	-0.350~0.077	0.210
糖尿病病程	-1.825	0.999	-1.83	-3.783~0.133	0.068
BUN	-0.357	0.422	-0.85	-1.184~0.470	0.397
SBP	-0.540	1.964	-0.27	-4.390~3.311	0.784

下载: 导出CSV

表 4 RDD稳健性检验

Table 4. Robustness test of RDD

类别	β值	s_x	Wald值	95% CI	P值
核函数类型
三角核	-5.021	1.726	-2.91	-8.404~-1.639	0.004
矩形核	-5.035	1.436	-3.51	-7.850~-2.220	0.001
带宽(μmol/L)
4.11	-5.021	1.726	-2.91	-8.404~-1.639	0.004
2.06	-5.403	2.206	-2.45	-9.726~-1.080	0.014
8.22	-4.907	1.460	-3.36	-7.767~-2.046	0.001
控制变量	-2.250	0.850	-2.65	-3.914~-0.584	0.008
注：控制变量包括高脂血症、饮食、运动、糖尿病病程、BUN、SBP。

下载: 导出CSV

参考文献(32)

[1]	Chan JCN, Malik V, Jia WP, et al. Diabetes in Asia: epidemiology, risk factors, and pathophysiology [J]. JAMA, 2009, 301(20): 2129-2140. DOI: 10.1001/jama.2009.726.
[2]	Intemational DF. IDF diabetes atlas-10th edition. [EB/OL]. (2021-12-06) [2022-02-07]. https://blog.csdn.net/wanjiac/article/details/106085105.
[3]	Zhang XX, Kong J, Yun K. Prevalence of diabetic nephropathy among patients with type 2 diabetes mellitus in China: a meta analysis of observational studies [J]. J Diabetes Res, 2020, 2020: 2315607. DOI: 10.1155/2020/2315607.
[4]	Gerstein HC, Mann JF, Yi Q, et al. Albuminuria and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic individuals [J]. JAMA, 2001, 286(4): 421-426. DOI: 10.1001/jama.286.4.421.
[5]	Han SS, Na KY, Chae DW, et al. High serum bilirubin is associated with the reduced risk of diabetes mellitus and diabetic nephropathy [J]. Tohoku J Exp Med, 2010, 221(2): 133-140. DOI: 10.1620/tjem.221.133.
[6]	Targher G, Zoppini G, Cesare Guidi G, et al. Relationship between serum bilirubin and kidney function in non-diabetic and diabetic individuals [J]. Kidney Int, 2009, 75(8): 863. DOI: 10.1038/ki.2008.677.
[7]	Targher G, Bosworth C, Kendrick J, et al. Relationship of serum. bilirubin concentrations to kidney function and albuminuria in the United States adult population. Findings from the national health and nutrition examination survey 2001-2006 [J]. Clin Chem Lab Med, 2009, 47(9): 1055-1062. DOI: 10.1515/CCLM.2009.244.
[8]	Guo ZY, Liu L, Yu FF, et al. The causal association between body mass index and type 2 diabetes mellitus-evidence based on regression discontinuity design [J]. Diabetes Metab Res Rev, 2021, 37(8): e3455. DOI: 10.1002/dmrr.3455.
[9]	Song D, Hurley JC, Lia M. Estimated treatment effects of tight glycaemic targets in mild gestational diabetes mellitus: a multiple cut-off regression discontinuity study design [J]. Int J Environ Res Public Health, 2020, 17(21): 7725. DOI: 10.330/ijer-ph17217725.
[10]	中华医学会糖尿病学分会. 中国2型糖尿病防治指南(2020年版) [J]. 中华糖尿病杂志, 2021, 13(4): 315-409. DOI: 10.3760/cma.j.cn115791-2021022100095.Chinese Diabetes Society. Guideline for the prevention and treatment of type 2 diabetes mellitus in China(2020 edition) [J]. Chin J Diabetes Mellit, 2021, 13(4): 315-409. DOI: 10.3760/cma.j.cn115791-2021022100095.Chinesecma.j.cn115791-2021022100095.
[11]	中华医学会糖尿病学分会微血管并发症学组. 中国糖尿病肾脏病防治指南(2021年版) [J]. 国际内分泌代谢杂志, 2021, 41(4): 388-410. DOI: 10.3760/cma.j.cn121383-20210825-08064.Microvascular Complications Group of Chinese Diabetes Society. Clinical guideline for the prevention and treat-ment of diabetie kidney disease in China(2021 edition) [J]. Int J Endocrinol Metab, 2021, 41(4): 388-410. DOI: 10.3760/cma.j.cn121383-20210825-08064.
[12]	刘力生. 中国高血压防治指南2010 [J]. 中华高血压杂志, 2011, 19(8): 701-743. DOI: 10.16439/j.cnki.1673-7245.2011.08.009.Liu LS. 2010 Chinese guidelines for the management of hypetension [J]. Chin J Hypertens, 2011, 19(8): 701-743. DOI: 10.16439/j.cnki.1673-7245.2011.08.009.
[13]	诸骏仁, 高润霖, 赵水平, 等. 中国成人血脂异常防治指南(2016年修订版) [J]. 中华全科医师杂志, 2017, 16(1): 15-35. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGXH201610002.htm Zhu JR, Gao RL, Zhao SP, et al. Guidelines for prevention and treatment of dyslipidenia in Chinese adults (rvised 2016) [J]. Chin J Gen Pract, 2017, 16(1): 15-35. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGXH201610002.htm
[14]	Chen Q. Advanced econometrics and stata applications (2nd Edi- tion) [M]. Beijing: Higher Education Press, 2014: 10.
[15]	柯箫韵, 翟海龙, 黄飞. 黄石社区2型糖尿病肾病患病率调查及其相关因素研究[J]. 中国全科医学, 2013, 16(12): 1373-1375. DOI: 10.3969/j.issn.1007-9572.2013.04.091.Ke XY, Zhai HL, Huang F. A epidemiological survey on the incidence of type 2 diabetic nephropathy and its related risk factors in Huangshi community [J]. Chin Gen Prac, 2013, 16(12): 1373-1375. DOI: 10.3969/j.issn.1007-9572.2013.04.091.
[16]	郁雨婷, 赵琦, 王娜, 等. 上海市某社区成人2型糖尿病及其合并肾功能不全流行现况及相关因素[J]. 中华疾病控制杂志, 2019, 23(8): 951-955, 976. DOI: 10.16462/j.cnki.zhjbkz.2019.08.013.Yu YT, Zhao Q, Wang N, et al. Epidemiology investigation of type 2 diabetes mellitus and its associated renal insfciency in a community based population in Shanghai [J]. Chin J Dis Control Prev, 2019, 23(8): 951-955, 976. DOI: 10.16462/j.cnki.zhjbkz.2019.08.013.
[17]	胡颖辉, 陈宏, 杨锐, 等. 广东省超重肥胖2型糖尿病患者糖尿病肾病的患病率及危险因素分析[J]. 中国现代医学杂志, 2016, 26(23): 89-94. DOI: 10.3969/j.isn.1005-8982.2016.23.018. Hu YH, Chen H, Yang R, et al. Prevalence and risk factors for diabetic nephropathy in overweight or obese patients with type 2 diabetes mellitus in Guangdong Province [J]. Chin J Mod Med, 2016, 26(23): 89-94. DOI: 10.3969/j.issn.1005-8982.2016.23.018.
[18]	Juutilainen A, Kastarinen H, Antikainen R, et al. Comparison of the MDRD study and the CKD-EPI study equations in evaluating trends of etimated kidney function at population level: findings from the national FINRISK study [J]. Nephrol Dial Transplant, 2012, 27(8): 3210-3217. DOI: 10.1093/ndt/gfs047.
[19]	刘锐, 杨智鹏, 陈磊, 等. 社区2型糖尿病患者并发糖尿病肾病现况及危险因素[J]. 中国老年学杂志, 2019, 39(19): 4679- 4682. DOI: 10.3969/j.issn.1005-9202.2019.19.016. LiuR, Yang ZP, Chen L, et al. Status and risk factors of diabetie nephropathy in conmunity patients with type 2 diabetes mellitus [J]. Chinese Joumal of Gerontology, 2019, 39(19): 4679-4682. DOI: 10.3969/j.issn.1005-9202.2019.19.016.
[20]	Toya K, Babazono T, Hanai K, et al. Association of serum bilirubin levels with development and progression of albuminuria, and decline in estimated glomerular filtration rate in patients with type 2 diabetes mellitus [J]. J Diabetes Investig, 2014, 5(2): 228-235. DOI10.1111/idi. 12134. doi: 10.1111/idi.12134
[21]	Rodella L, Lamon BD, Rezzani R, et al. Carbon monoxide and biliverdin prevent endothelial cell sloughing in rats with type Ⅰ dia betes [J]. Free Radic Biol Med, 2006, 40(12): 2198-2205. DOI: 10.1016/j.freeradbiomed.2006.02.018.
[22]	Boon AC, Bulmer AC, Coombes JS, et al. Circulating bilirubir and defense against kidney disease and cardiovascular mortality: mechanisms contributing to protection in clinical investigations [J]. Am J Physiol Renal Physiol, 2014, 307(2): F123-F136. DOI: 10.1152/aiprenal.00039.2014.
[23]	Hull TD, Agarwal A. Bilirubin: a potential biomarker and therapeutic target for diabetic nephropathy [J]. Diabetes, 2014, 63(8): 2613-2616. DOI: 10.2337/db14-0691.
[24]	Oda E, Aoyagi R, Aizawa Y. Hypobilirubinemia might be a possible risk factor of end-stage kidney disease independently of esti mated glomerular filtration rate [J]. Kidney Blood Press Res, 2012, 36(1): 47-54. DOI: 10.1159/000339027.
[25]	Wu Y, Zhang J, Wang J, et al. The association of serum bilirubin on kidney clinicopathologic features and renal outcome in patients with diabetic nephropathy: a biopsy-based study [J]. Endocr Pract, 2019, 25(6): 554-561. DOI: 10.4158/EP-2018-0560.
[26]	Sakoh T, Nakayama M, Tanaka S, et al. Association of serum total bilirubin with renal outcome in Japanese patients with stages 3-5 chronic kidney disease [J]. Metabolism, 2015, 64(9): 1096-1102. DOI: 10.1016/j.imetabol.2015.06.006.
[27]	Liu M, Li Y, Li J, et al. Elevated serum total bilirubin levels are negatively associated with major diabetic complications among Chinese senile diabetic patients [J]. J Diabetes Complicat, 2017, 31(1): 213-217. DOI: 10.1016/jjdiacomp.2016.08.023.
[28]	Wang J, Li Y, Han X, et al. Association between serum bilirubin levels and decline in estimated glomerular filtration rate among patients with type 2 diabetes [J]. J Diabetes Complicat, 2016, 30(7): 1255-1260. DOI: 10.1016/ijdiacomp.2016.05.013.
[29]	Liu M, Li J, Lyu X, et al. Bilirubin and its changes were negatively associated with diabetic kidney disease incidence and progression: a five year's cohort study based on 5 323 Chinese male diabetic patients [J]. J Diabetes Complicat, 2018, 32(11): 1012-1017. DOI: 10.1016/j.jdiacomp.2018.08.006.
[30]	Okada H, Fukui M, Tanaka M, et al. Low serum bilirubin concentration is a novel risk factor for the development of albuminuria in patients with type 2 diabetes [J]. Metabolism, 2014, 63(3): 409-414. DOI: 10.1016/j.imetabol.2013.11.011.
[31]	Chan WK, Tsai SS, Li YR, et al. Association between serum bilirubin levels and progression of albuminuria in Taiwanese with type 2 diabetes mellitus [J]. Biomed J, 2021, 44(2): 201-208. DOI: 10.1016/j.bi.2019.12.004.
[32]	Tanaka M, FukuiM, Okada H, et al. Low serum bilirubin concentration is a predictor of chronic kidney disease [J]. Atherosclerosis, 2014, 234(2): 421-425. DOI: 10.1016/j.atherosclerosis.2014.03.015.