• Users Online: 32
  • Print this page
  • Email this page


 
 
Table of Contents
ORIGINAL ARTICLE
Year : 2021  |  Volume : 4  |  Issue : 3  |  Page : 75-82

Insulin product decreases risk of varicose vein: Evidence from a Mendelian randomization study


1 Department of Cardiovascular Surgery, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, Guangzhou, China
2 Zhongshan School of Medicine, Sun Yat-Sen University, Guangzhou, China
3 Zhongshan School of Medicine, Sun Yat-Sen University; Department of Vascular Surgery, Sun Yat-Sen First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China

Date of Submission28-Mar-2021
Date of Decision12-May-2021
Date of Acceptance12-May-2021
Date of Web Publication17-Aug-2021

Correspondence Address:
Dr. Chen Yao
No. 58 Zhongshan Second Road, Guangzhou
China
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/2589-9686.323983

Rights and Permissions
  Abstract 


CONTEXT: The association between insulin product treatment and risk of varicose vein (VV) is still unknown.
AIMS: Our study used two-sample Mendelian randomization (MR) to investigate whether treatment of insulin product was causally associated with a lower risk of VV.
SUBJECTS AND METHODS: We searched the summary data from genome-wide association study through MR-Base platform. Data included were from Neale Lab UK-Biobank (UKB)-a-153 (insulin product) and MRC integrative epidemiology unit UKB-b-15592 (VV surgery). Three MR approaches, including inverse variance weighted (IVW) method, MR-Egger, and weighted median method were used to explore the casual effect of insulin product on VV. The exposure in our study was insulin product, and the outcome was VV surgery, both measured by single nucleotide polymorphisms.
RESULTS: Our results showed that insulin product decreased the risk of VV with using IVW method (odds ratio [OR] = 0.73, 95% confidence interval = 0.64–0.84, P < 0.001), which was consistent with the result of MR-Egger and weighted median method. Results of MR-Egger regression showed no evidence for the presence of directional horizontal pleiotropy.
CONCLUSIONS: Our study suggested that insulin product treatment had an inverse association with risk of VV.

Keywords: Mendelian randomization, insulin product, varicose vein


How to cite this article:
Huang K, Shen R, Chen Q, Tian Z, Xia Z, Lin X, Wu G, Chen Z, Yao C. Insulin product decreases risk of varicose vein: Evidence from a Mendelian randomization study. Vasc Invest Ther 2021;4:75-82

How to cite this URL:
Huang K, Shen R, Chen Q, Tian Z, Xia Z, Lin X, Wu G, Chen Z, Yao C. Insulin product decreases risk of varicose vein: Evidence from a Mendelian randomization study. Vasc Invest Ther [serial online] 2021 [cited 2021 Dec 6];4:75-82. Available from: https://www.vitonline.org/text.asp?2021/4/3/75/323983




  Introduction Top


Varicose vein (VV) is one of chronic venous diseases, usually defined as C2 (from C0 to C6) in the Clinical-Etiology-Anatomy-Pathophysiology classification of CVD.[1] About 23% of adults in the United States have VV, which leads to a significant decline in the quality of life.[2],[3] And because of high incidence rate and high treatment cost of related complications such as chronic phlebitis or chronic venous ulcers, VV causes a huge burden of social health resource annually.[4] The pathological causes of VV include venous hypertension, venous valvular insufficiency, inflammation, and blood flow shear stress.[5] Sex, age, body mass index, smoking, low-fiber diet, exercise status, race, pregnancy, and family heredity history are important factors which are associated with VV.[6],[7] Minimally surgery is still the most widely treatment for VV instead of medical therapy nowadays, though which may cause long-term problems of reopening.[8] Research related to molecular therapy or drug treatment for VV should be conducted.

Insulin product therapy is an indispensable way to reduce blood glucose as a treatment for diabetes mellitus (DM).[9] Insulin activation decreases damage of endothelial cells (ECs) and inflammation in blood vessel through PI3K/AKT pathways, which may relieve or prevent the progression of VV.[10] In addition, insulin has function of vasodilation and increasing the blood flow through stimulating the production of nitric oxide (NO) in ECs.[11],[12] However, whether insulin product has an independent and biological effect on VV is still unknown. It is difficult to study insulin alone in randomized controlled trials (RCT) because of ethical reasons and the existence of confounding factors such as DM and the use of other hypoglycemic drugs such as metformin.[13] Mendelian randomization (MR) is a novel method which uses genetic variants as instrument variables (IVs) to investigate the causal association between exposure and outcome.[14] The genetic variants used in MR study, which indicate the effect of exposure, should be causally related to outcome.[15] Moreover, these genetic variants could serve as unconfounded proxies for modifiable risk factors because they were randomly distributed in the population before birth and fixed at conception. MR is similar to RCT but the most difference is that MR can avoid confounding factors and reverse causality in a maximum degree.[16] What's more is that MR is time saving and cost effective and it had been widely used in observational epidemiology of VV recently.[17],[18]

Our study is aimed to investigate the causal association between insulin product treatment and VV with MR method.


  Subjects and Methods Top


Access-related sources and data

Our data were all retrieved from genome-wide association study (GWAS) summary data. GWAS is a summary database which has nondisclosive results from analyzing genetic variants with a phenotype, providing a great method to make inference and explore the causality between environmental exposure and diseases.[19],[20] Through using a platform named MR-Base,[21] we could conveniently use the summary data from GWAS to support MR analysis.[22] We used the data associated with insulin product from Neale Lab consortium in 2017 including 3319 cases and 333840 controls, and the data associated with VV surgery from MRC integrative epidemiology unit consortium in 2018 including 18,818 cases and 444,115 controls rooted Europe [Table 1]. Insulin product (dataset: UK-Biobank [UKB]-a-153) and VV surgery (dataset: UKB-b-15,592) documented in the database were treatment/medication code and operation code, respectively, which meant that studying individuals had received treatment of insulin product or VV surgery. Insulin product documented in the summary data we collected from GWAS for each genetic instrument included the followings: the effect allele, effect allele frequency, beta value, standard error (SE), single nucleotide polymorphisms (SNP), and P value. We also set the parameters of our study as P < 5 × 10−8; linkage disequilibrium r2 <0.1 referring to the previous standard of SNP.[22] The data result is shown in [Table 2]. We used SNPs as IVs associated with insulin product for further MR study.
Table 1: Details of studies and datasets used for analyses

Click here to view
Table 2: Characteristics of the single-nucleotide polymorphism associated with insulin product and varicose vein surgery

Click here to view


Causal effect estimates and sensitivity analysis

The causal effects of insulin product on VV surgery were estimated with the SNP-exposure effects and the SNP-outcome effects obtained from different studies as SNPs were regarded as valid tool variables. There is a range of methods to analyze with GWAS summary data when there are multiple independent instruments for exposure. Moreover, the analysis can be improved by explaining the variance in exposure and evaluating the sensitivity of the estimate to bias arising from violations of the IV assumption (the instrument does not influence the outcome through some pathway other than the exposure) by assuming different patterns of horizontal pleiotropy. We used three MR approaches (inverse variance weighted [IVW] method, MR-Egger, and weighted median method) to explore MR estimates as reference to guideline of use of MR-Base platform.[22] First, we conducted a random effects IVW meta-analysis of Wald ratio for individual SNP used as IV to regress the SNP-insulin product association against the SNP-VV surgery association.[23] Second, weighted median method was conducted to explore the weighted empirical distribution function of ratio estimate of each SNP selected.[24] At last, we performed MR-Egger analysis to carry out a weighted linear regression of SNP-VV risk ratio to SNP-insulin product effect estimates. MR-Egger assumed that horizontal gene polymorphism was independent of SNP exposure effects, which allowed nonzero intercept and unbalanced horizontal gene polymorphism in regression with all SNPs.[25]

Sensitivity analysis

Our MR study was based on the following three assumptions: (a)IVs are strongly associated with insulin product; (b)IVs affect VV surgery only through their effect on insulin product; and (c) IVs have no association with confounders of the association between insulin product and VV surgery. To assess the potential violation of three assumptions above, we conducted MR-Egger sensitivity test to evaluate the directional pleiotropy based on the intercept.[26] At last, a leave-one-out analysis was used to evaluate whether the MR estimates were driven or biased by a single SNP by removing that SNP and using the remaining SNPs as IV to estimate the total effect by IVW method, and a forest plot was also performed to plot the estimate for a particular SNP against its precision.[22],[27] The extent to the individual SNP affecting the causal relevance estimate can be determined through the stability of the effect estimate.


  Results Top


Single nucleotide polymorphisms associated with insulin product

We found eight SNPs (eight SNPs included rs9380190, rs114355928, rs689, rs7744001, rs6679677, rs3129886, rs1064173, rs7903146) which were associated with insulin product at a GWAS threshold of statistical significance (P < 5 × 10−8; linkage disequilibrium r2 < 0.1).

Causal effect from insulin product on varicose vein

Our two-sample MR analysis genetically predicted that insulin product had an inverse association with VV surgery (odds ratio [OR] =0.73, 95% confidence interval [CI] =0.64–0.84), and the P value had statistical significance (P < 0.001). The result of association calculated with IVW method was consistent with weighted median method (OR = 0.68, 95% CI = 0.61–0.76, P < 0.001) and MR-Egger method (OR = 0.65, 95% CI = 0.49–0.86, P = 0.024) [Table 3]. The individual causal effects of SNPs which were selected as IVs (rs9380190, rs114355928, rs689, rs7744001, rs6679677, rs3129886, rs1064173, rs7903146) are shown in [Figure 1]. Among all the SNPs selected, rs7903146 (gene TCF7 L2) showed strongest effect on the association between insulin product and VV surgery. Three MRs regression slopes are shown in [Figure 2].
Table 3: Mendelianrandomizationestimatesoftheassociationsbetweeninsulinproductandriskofvaricosevein

Click here to view
Figure 1: Forest plot of single nucleotide polymorphisms associated with insulin product and their risk of varicose vein. Each black point represents the log odds ratio for varicose vein per standard deviation increase in insulin product using each single nucleotide polymorphisms as instrument variables. Each red point shows the combined causal estimate using all single nucleotide polymorphisms to figure out with each of the three different methods (inverse variance weighted method, weighted median method, and MR-Egger). Horizontal lines denote 95% confidence interval. OR: Odds ratio, SD: Standard deviation, SNP: Single nucleotide polymorphism, IVs: Instrument variables, IVW: Inverse variance weighted, CI: Confidence interval

Click here to view
Figure 2: Scatter plot of single nucleotide polymorphisms associated with insulin product and the risk of varicose vein. The plot shows the effect sizes of the single nucleotide polymorphisms-insulin product associations (X-axis, standard deviation units) and the single nucleotide polymorphisms-varicose vein associations (Y-axis, log odds ratio) with standard error bars. The slopes of the lines correspond to causal estimates using each of the three different methods (inverse variance weighted method, weighted median method, and MR-Egger). OR: Odds ratio; SD: Standard deviation; SNP: Single nucleotide polymorphism, VV: varicose vein

Click here to view


Sensitivity analysis results

In our leave-one-out sensitivity analysis, no single SNP was strongly driving the overall effect of insulin product on VV surgery, as shown in [Figure 3]. There was no evidence for the presence of directional horizontal pleiotropy in the MR-Egger regression analysis (intercept = 5.8e-04, P = 0.39), which was consistent with the hypothesis that genetic pleiotropy was not driving the result [Table 4].
Table 4: Mendelian randomization - Egger pleiotropy test of the associations between insulin product and riskofvaricosevein

Click here to view
Figure 3: Leave-one-out of single nucleotide polymorphisms associated with insulin product and their risk of varicose vein. Each black point represents using the inverse variance weighted method to estimate the causal effect of insulin product on varicose vein excluding the particular variant from the analysis. Red point shows the inverse variance weighted estimate using all single nucleotide polymorphisms. No single nucleotide polymorphisms are strongly driving the overall effect of insulin product on varicose vein. SNP: Single nucleotide polymorphism, IVW: Inverse variance weighted; CI: Confidence interval, VV: Varicose vein

Click here to view



  Discussion Top


VV is the most common illness of vascular vein system, which can progress with developing edema, pigmentation, eczema, and even ulcers.[5] Epidemiological observation study had found that hypoglycemic agents such as metformin used in Type 2 DM may decrease the risk of VV.[13] However, no previous study had shown that insulin product might have treatment effect on VV. In our large MR study, the results showed that insulin product had a decreased association with VV (OR = 0.73, 95% CI = 0.64–0.84, P < 0.001).

Hypertension and inflammatory factors can cause damage to ECs, which are important pathology of VV.[2],[7] The most studied of insulin's effect on the vascular system is that it stimulates the production of NO in the ECs, resulting in vasodilation and increasing blood flow.[28],[29] This mechanism may have a pathological explanation for the relieve of VV as vasodilation may reduce venous hypertension over the vein valve. Insulin can also have anti-inflammatory effect by mediating through the activation of IRS/PI3K/AKT pathway, which inhibits the expression of pro-inflammatory chemokine MCP-1 in macrophages and the adhesion of monocytes.[10],[30] Previous studies had shown that inflammation is a key factor in the development of VV because the recruitment of leukocytes and release of more inflammatory factors and cell adhesion factors can cause damage to the ECs.[31],[32],[33] What's more, hyperglycemia can cause more reactive oxygen species production in ECs, which destroys vascular endothelial system and accelerates ECs senescence or apoptosis.[34],[35] In this way, insulin product may show the role of protecting blood vessels but mostly in DM patients because of the ethical permission for the drug use. What's more, insulin can induce the expression of VEGF in ECs,[36],[37] which promotes the remodeling of microvascular function and helps relieve VV. However, the mechanisms of the effect of insulin product on VV should be studied further.

Our study has several advantages. First, we are the first to study the association between insulin product and VV with MR method. MR method helps us to prevent confounding factors and enables us to use summary-level GWAS results to test the mechanism-which can prevent reverse causation from conventional observational studies. Second, two-sample MR used here is more accurate beyond one-sample MR and has more statistical power.[26] Third, large sample size and robust IVs enable sufficient power to detect robust causal effect estimates with high precision.[38]

There are also limitations in our analysis. First, the main limitation of using GWAS data is that we could not contact the individual patient so that the subgroup analysis cannot be done to confirm the particularity of the association. Second, the data we included were all from European participants and we cannot conclude that our finding is generalized to other populations or races. At last, we were still not able to investigate the mechanism of how insulin affects VV and more work is needed further.


  Conclusions Top


Through eight genetically identified SNPs from GWAS summary data, this large MR study suggests that insulin product decreases the risk of VV. More work is needed to find the causal association between insulin and VV clearly.

Financial support and sponsorship

The project is sponsored by National Natural Science Foundation of China (Project approval number is 81800420).

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Davies AH. The seriousness of chronic venous disease: A review of real-world evidence. Adv Ther 2019;36:5-12.  Back to cited text no. 1
    
2.
Bergan JJ, Schmid-Schönbein GW, Smith PD, Nicolaides AN, Boisseau MR, Eklof B. Chronic venous disease. N Engl J Med 2006;355:488-98.  Back to cited text no. 2
    
3.
Hamdan A. Management of varicose veins and venous insufficiency. JAMA 2012;308:2612-21.  Back to cited text no. 3
    
4.
Shepherd AC, Gohel MS, Lim CS, Davies AH. A study to compare disease-specific quality of life with clinical anatomical and hemodynamic assessments in patients with varicose veins. J Vasc Surg 2011;53:374-82.  Back to cited text no. 4
    
5.
Piazza G. Varicose veins. Circulation 2014;130:582-7.  Back to cited text no. 5
    
6.
Brand FN, Dannenberg AL, Abbott RD, Kannel WB. The epidemiology of varicose veins: The Framingham study. Am J Prev Med 1988;4:96-101.  Back to cited text no. 6
    
7.
Beebe-Dimmer JL, Pfeifer JR, Engle JS, Schottenfeld D. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol 2005;15:175-84.  Back to cited text no. 7
    
8.
Jacobs BN, Andraska EA, Obi AT, Wakefield TW. Pathophysiology of varicose veins. J Vasc Surg Venous Lymphat Disord 2017;5:460-7.  Back to cited text no. 8
    
9.
Danne T, Heinemann L, Bolinder J. New insulins, biosimilars, and insulin therapy. Diabetes Technol Ther 2019;21:S57-78.  Back to cited text no. 9
    
10.
King GL, Park K, Li Q. Selective insulin resistance and the development of cardiovascular diseases in diabetes: The 2015 Edwin Bierman Award Lecture. Diabetes 2016;65:1462-71.  Back to cited text no. 10
    
11.
Potenza MA, Addabbo F, Montagnani M. Vascular actions of insulin with implications for endothelial dysfunction. Am J Physiol Endocrinol Metab 2009;297:E568-77.  Back to cited text no. 11
    
12.
Muniyappa R, Quon MJ. Insulin action and insulin resistance in vascular endothelium. Curr Opin Clin Nutr Metab Care 2007;10:523-30.  Back to cited text no. 12
    
13.
Tseng CH. Metformin reduces risk of varicose veins in patients with type 2 diabetes. Diabetes Metab Res Rev 2020;36:e3206.  Back to cited text no. 13
    
14.
Emdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA 2017;318:1925-6.  Back to cited text no. 14
    
15.
Smith GD, Timpson N, Ebrahim S. Strengthening causal inference in cardiovascular epidemiology through Mendelian randomization. Ann Med 2008;40:524-41.  Back to cited text no. 15
    
16.
Smith GD, Ebrahim S. Mendelian randomization: Prospects, potentials, and limitations. Int J Epidemiol 2004;33:30-42.  Back to cited text no. 16
    
17.
Shadrina AS, Sharapov SZ, Shashkova TI, Tsepilov YA. Varicose veins of lower extremities: Insights from the first large-scale genetic study. PLoS Genet 2019;15:e1008110.  Back to cited text no. 17
    
18.
Fukaya E, Flores AM, Lindholm D, Gustafsson S, Zanetti D, Ingelsson E, et al. Clinical and genetic determinants of varicose veins. Circulation 2018;138:2869-80.  Back to cited text no. 18
    
19.
Welter D, MacArthur J, Morales J, Burdett T, Hall P, Junkins H, et al. The NHGRI GWAS Catalog, a curated resource of SNP-trait associations. Nucleic Acids Res 2014;42:D1001-6.  Back to cited text no. 19
    
20.
Zhu Z, Zheng Z, Zhang F, Wu Y, Trzaskowski M, Maier R, et al. Causal associations between risk factors and common diseases inferred from GWAS summary data. Nat Commun 2018;9:224.  Back to cited text no. 20
    
21.
MR-Base Platform. Available from: http://www.mrbase.org. [Last accessedon 2020 Jul 15].  Back to cited text no. 21
    
22.
Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife 2018;7:e34408.  Back to cited text no. 22
    
23.
Pierce BL, Burgess S. Efficient design for Mendelian randomization studies: Subsample and 2-sample instrumental variable estimators. Am J Epidemiol 2013;178:1177-84.  Back to cited text no. 23
    
24.
Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol 2016;40:304-14.  Back to cited text no. 24
    
25.
Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol 2017;32:377-89.  Back to cited text no. 25
    
26.
Bowden J, Del Greco MF, Minelli C, Davey Smith G, Sheehan NA, Thompson JR. Assessing the suitability of summary data for two-sample Mendelian randomization analyses using MR-Egger regression: The role of the I2 statistic. Int J Epidemiol 2016;45:1961-74.  Back to cited text no. 26
    
27.
Sterne JA, Sutton AJ, Ioannidis JP, Terrin N, Jones DR, Lau J, et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ 2011;343:d4002.  Back to cited text no. 27
    
28.
Steinberg HO, Brechtel G, Johnson A, Fineberg N, Baron AD. Insulin-mediated skeletal muscle vasodilation is nitric oxide dependent. A novel action of insulin to increase nitric oxide release. J Clin Invest 1994;94:1172-9.  Back to cited text no. 28
    
29.
Vincent MA, Dawson D, Clark AD, Lindner JR, Rattigan S, Clark MG, et al. Skeletal muscle microvascular recruitment by physiological hyperinsulinemia precedes increases in total blood flow. Diabetes 2002;51:42-8.  Back to cited text no. 29
    
30.
Mita T, Azuma K, Goto H, Jin WL, Arakawa M, Nomiyama T, et al. IRS-2 deficiency in macrophages promotes their accumulation in the vascular wall. Biochem Biophys Res Commun 2011;415:545-50.  Back to cited text no. 30
    
31.
Castro-Ferreira R, Cardoso R, Leite-Moreira A, Mansilha A. The role of endothelial dysfunction and inflammation in chronic venous disease. Ann Vasc Surg 2018;46:380-93.  Back to cited text no. 31
    
32.
Shadrina A, Voronina E, Smetanina M, Tsepilov Y, Sevost'ianova K, Shevela A, et al. Polymorphisms in inflammation-related genes and the risk of primary varicose veins in ethnic Russians. Immunol Res 2018;66:141-50.  Back to cited text no. 32
    
33.
Lattimer CR, Kalodiki E, Geroulakos G, Hoppensteadt D, Fareed J. Are inflammatory biomarkers increased in varicose vein blood? Clin Appl Thromb Hemost 2016;22:656-64.  Back to cited text no. 33
    
34.
Domingueti CP, Dusse LM, Carvalho Md, de Sousa LP, Gomes KB, Fernandes AP. Diabetes mellitus: The linkage between oxidative stress, inflammation, hypercoagulability and vascular complications. J Diabetes Complications 2016;30:738-45.  Back to cited text no. 34
    
35.
Rizzoni D, Agabiti Rosei E. Small artery remodeling in hypertension and diabetes. Curr Hypertens Rep 2006;8:90-5.  Back to cited text no. 35
    
36.
Jiang ZY, He Z, King BL, Kuroki T, Opland DM, Suzuma K, et al. Characterization of multiple signaling pathways of insulin in the regulation of vascular endothelial growth factor expression in vascular cells and angiogenesis. J Biol Chem 2003;278:31964-71.  Back to cited text no. 36
    
37.
Kuboki K, Jiang ZY, Takahara N, Ha SW, Igarashi M, Yamauchi T, et al. Regulation of endothelial constitutive nitric oxide synthase gene expression in endothelial cells and in vivo: A specific vascular action of insulin. Circulation 2000;101:676-81.  Back to cited text no. 37
    
38.
Brion MJ, Shakhbazov K, Visscher PM. Calculating statistical power in Mendelian randomization studies. Int J Epidemiol 2013;42:1497-501.  Back to cited text no. 38
    


    Figures

  [Figure 1], [Figure 2], [Figure 3]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4]



 

Top
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
Abstract
Introduction
Subjects and Methods
Results
Discussion
Conclusions
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed618    
    Printed14    
    Emailed0    
    PDF Downloaded60    
    Comments [Add]    

Recommend this journal