Skip to main content
Download PDF

The impact of sex hormones on metabolic syndrome: univariable and multivariable Mendelian randomization studies

Diabetology & Metabolic Syndrome volume 16, Article number: 215 (2024) Cite this article

Abstract

Background

Observational studies have found associations between sex hormones and metabolic syndrome(Mets), but the causal relationships remains unclear. This study utilizes univariable and multivariable Mendelian randomization (MR) to elucidate the associations between sex hormones (including sex hormone-binding globulin(SHBG), estradiol(E2), testosterone(T)) and Mets and its subtypes (including waist circumference(WC), fasting blood glucose(FBG), high blood pressure(HBP), high-density lipoprotein(HDL-C), triglycerides(TG)).

Methods

We utilized summary data from large-scale genome-wide association studies. Univariable Mendelian randomization (UMVMR) analysis was primarily conducted using the inverse variance weighted method (IVW), with secondary analyses employing the weighted median, MR-Egger regression, simple mode method, and weighted mode method. Subsequently, multivariable Mendelian randomization (MVMR) was employed to assess the causal relationships between SHBG, T, E2, and MetS and its components: WC, FPG, HBP, HDL-C, and TG. Sensitivity analyses were conducted to assess result reliability.

Results

Genetically predicted SHBG was significantly negatively associated with MetS (UMVMR: β=-0.72; 95% CI = 0.41 to 0.57; P = 1.28e-17; MVMR: β=-0.60; 95% CI=-0.83 to -0.38; P < 0.001). Positive causal relationships were observed between SHBG and WC(MVMR: β = 0.10; 95% CI = 0.03 to 0.17; P = 0.01) and HDL-C (MVMR: β = 0.41; 95% CI = 0.21 to 0.60; P < 0.001), while negative causal relationships were found between SHBG and HBP (MVMR: β=-0.02; 95% CI=-0.04 to -0.00; P = 0.02), TG (MVMR: β=-0.48; 95% CI=-0.70 to -0.26; P < 0.001). Genetically predicted E2 exhibited a negative association with TG (MVMR: β=-1.49; 95% CI=-2.48 to -0.50; P = 0.003). Genetically predicted T was negatively associated with TG (MVMR: β=-0.36; 95% CI=-0.71 to -0.00; P = 0.049) and WC (MVMR: β=-0.13; 95% CI=-0.24 to -0.02; P = 0.02), with inconsistent sensitivity analyses. Additionally, No other causal associations were found.

Conclusion

Our study indicates that SHBG is a protective factor for MetS, potentially delaying its onset and progression through improvements in HBP and TG. Furthermore, T and E2 may improve TG levels, with T also reducing WC levels. Importantly, our study provides new insights for the prevention and treatment of MetS.

Introduction

Metabolic syndrome (MetS) is a pathological state characterized by metabolic disturbances in substances such as sugar, fat, and protein, including insulin resistance, dyslipidemia, hypertension, and central obesity [1]. With the development of society and the improvement of living standards, the prevalence of MetS is gradually increasing, with approximately 1/4 of the global population estimated to be affected. Due to the lack of specific treatment strategies, the prevalence of MetS continues to rise [2]. Moreover, MetS-related diseases account for two-thirds of non-communicable disease-related deaths [3]. Therefore, effective prevention and treatment of MetS are crucial. Estradiol (E2) is closely related to human reproductive health and participates in a wide range of neurological and physiological functions [4]. Within the physiological range, E2 can enhance insulin sensitivity, while both excessively high and low levels of E2 can lead to insulin resistance, further affecting metabolic function [5]. Testosterone (T) regulates the expression of key regulatory enzymes in the metabolism of glucose and lipids in major insulin-responsive target tissues. Additionally, sex hormone-binding globulin (SHBG) in the circulation affects T levels indirectly by binding to and dissociating from T, thereby indirectly influencing glucose and lipid metabolism [6]. Furthermore, SHBG can directly affect metabolic function through the SHBG receptor-mediated intracellular signaling pathways located on target cell membranes [7]. Peroxisome proliferator-activated receptor-gamma (PPARγ) is a crucial regulator of adipogenesis. Research in molecular medicine has found that PPARγ can influence the activity, functionality, survival rate, longevity, and mitochondrial metabolism of adipose-derived stromal cells (ASCs). SHBG can mimic the effects of PPARγ and is considered a valuable therapeutic agent for regulating mitochondrial activity and improving adipose tissue health [8]. However, current observational studies on the association between T, E2, SHBG, and MetS and its subcomponents yield inconsistent results. A cross-sectional study in Asian male populations found negative correlations between T, SHBG, and MetS, as well as associations between SHBG and HBP, TG, and blood glucose, and between T and WC and TG [9]. Another study did not find an association between SHBG and TG [10]. There are also studies indicating associations between T and MetS and WC, with T replacement therapy improving WC and enhancing quality of life [11, 12]. Conversely, other studies have indicated that lower levels of SHBG are associated with MetS, and higher levels of T increase the risk of metabolic diseases in women but decrease the risk in men [13]. Currently, there is limited research on the relationship between E2 and Mets, and there are also internal connections between E2, T, and SHBG. Observational studies may suffer from confounding biases and small sample sizes, making it difficult to determine causality. Therefore, the specific associations between sex hormones and MetS remain uncertain.

Mendelian randomization (MR) is a method that uses genetic variants as instrumental variables to leverage the natural random allocation of genetic variation, effectively avoiding confounding factors and assessing causal associations from observed data [14, 15]. Randomized controlled trials are often time-consuming and resource-intensive, making MR an alternative method for inferring potential causal relationships between exposure and outcome [16]. Multivariate Mendelian randomization (MVMR) is an emerging MR method that integrates genetic variants of multiple exposures into the same model to minimize confounding factors, allowing for the simultaneous evaluation of the contributions of various exposures to the outcome [17]. To elucidate these underlying relationships, we employed two-sample MR to study the causal relationships between sex hormones (including E2, T, and SHBG) and MetS. Considering potential confounding factors among some sex hormones, we further used MVMR to assess the causal effects of E2, T, and SHBG on MetS. Additionally, we conducted subgroup analyses of MetS to explore the associations between E2, T, SHBG, and FBG, WC, HBP, HDL-C, and TG.

Materials and methods

Data source

Exposure data

Data for T (in males and females), E2 (in males), and SHBG (in males and females) were sourced from summary statistics of a large-scale genome-wide association study (GWAS). This study utilized genotype and phenotype data from the UK Biobank, including 425,097 participants [18]. To ensure the robustness of instrumental variables and exclude weak instruments, stringent thresholds were set. A genetic distance of 10,000 kb, a linkage disequilibrium (LD) parameter (r2) threshold of 0.01, and a P-value threshold of 5 × 10− 8 were applied to ensure the independence of instrumental variables and mitigate the impact of LD on the results. Alignment of allelic effects of single nucleotide polymorphisms (SNPs) and removal of all SNPs with palindromic structures were conducted to obtain significantly associated SNPs. Detailed information is provided in Table 1.

Outcome data

The data for MetS was obtained from the latest data of the Complex Trait Genetics Laboratory (CTG), which includes 461,920 valid subjects of European descent, and identifying the genetic variations of metabolic syndrome through structural equation modeling methods [19]. Genetic variation data for MetS subtypes was extracted from the comprehensive GWAS database. Genetic tools for FBG were extracted from a database containing 58,074 European ancestry participants and adjusted for BMI [20]. Data for WC and HBP were sourced from the MRC-IEU-OpengwasProject database, which includes 462,166 and 463,010 subjects of European descent, respectively. Summary statistics for HDL-C and TG were obtained from the Global Lipids Genetics Consortium (GLGC), which consists of participants from multiple ethnicities, with 96% of them being of European descent [21]. Detailed information is provided in Table 1.

Table 1 Detailed information on the data used in MR
Full size table

Study methods

MR studies need to meet three key assumptions. (1) The Relevance Assumption: This assumption states that the genetic variables used in the analysis should be strongly correlated with the exposure. In this study, significant SNPs were selected by screening p-values for significant correlation, and then the strength of each SNP was assessed by calculating the F-statistic to eliminate the possibility of weak instrumental variables [22].(2) The Independence Assumption: This assumption states that the genetic variants used as instrumental variables for the exposure are independent of confounding factors associated with both the selected exposure and the outcome. In the Mendelian randomization process, due to the lack of individual-level data, it is challenging to use statistical methods to verify this assumption directly. However, MR follows the genetic principle of “random allocation of parental alleles to offspring during gamete formation.” Therefore, confounding factors such as environmental factors and socioeconomic status have minimal impact on the genetic effects.(3) The Exclusion Restriction Assumption: This assumption states that the SNP affects the outcome only through the exposure and not through any other pathways, meaning there is no pleiotropy. In this study, the presence of pleiotropy between SNPs and gout was assessed using the intercept term of the MR-Egger regression.

Univariate Mendelian randomization

In this study, five regression models were utilized to validate the causal relationship between exposure (T, E2, and SHBG) and outcome (MetS) using UMVMR: inverse-variance-weighted (IVW), MR-Egger regression, weighted median estimator (WME), simple mode, and weighted mode.The IVW method aggregates effects from multiple loci through meta-analysis, disregarding the intercept term and utilizing the inverse of the outcome variance as weights for fitting. The IVW analysis results are considered most reliable when there is no pleiotropy present.The WME method calculates the median of the distribution function of SNP effect values after sorting them by weight, providing robust causal inference for significantly outlying SNPs [23]. MR-Egger regression accounts for the presence of the intercept term. The presence of pleiotropy is indicated when the regression intercept term is non-zero and the P-intercept < 0.05 [24]. A significance level of α = 0.05 was employed, where P < 0.05 indicates statistically significant differences.

Multivariable Mendelian randomization

When there may be correlations among multiple exposure factors, using MVMR can help address potential horizontal pleiotropy and provide more accurate results. MVMR allows for the inclusion of multiple instrumental variables without considering their individual association with the exposure of interest. In this study, considering the correlations among E2, T, and SHBG, we used MVMR to combine all instrumental variables for the three exposures to evaluate their independent effects on MetS. Additionally, we employed MVMR to assess the effects among MetS subcomponents.

Sensitivity analysis

Firstly, heterogeneity tests were conducted using the IVW method and MR-Egger method. If the P-value of the test result was > 0.05, it was considered that there was no heterogeneity among the included SNPs, and the impact of heterogeneity on the estimation of causal effects could be ignored. If heterogeneity existed among SNPs, the IVW random effects model was used to estimate the causal effects. Secondly. Secondly, MR-Egger regression and MR-Presso global tests were used to evaluate horizontal pleiotropy of instrumental variables (IVs). The effect estimation of horizontal pleiotropy was represented by the intercept, and a P-value > 0.05 indicated no detection of horizontal pleiotropy [25]. In addition, MR-PRESSO analysis was used to identify and remove significant aberrant SNPs that could introduce horizontal pleiotropy. Finally, a sensitivity analysis was conducted using the leave-one-out method, examining the influence of each SNP on the results. Specifically, for SNPs with P-values < 0.05 in the IVW method, each SNP was individually removed, and the combined effect of the remaining SNPs was calculated to assess whether the MR results were sensitive to that instrumental variable.

Statistical analysis

All statistical analyses were conducted using the “TwoSample MR”, “Mendelian Randomization”, and “MRPRESSO” packages in R version 4.3.3. Summary statistics data related to exposure factors and clinical outcome datasets were harmonized, ensuring that the effects of SNPs on exposure factors and the effects of SNPs on clinical outcome data corresponded to the same alleles. We employed IVW as the primary analysis method and deemed P < 0.05 as indicative of a potential association between sex hormones and MetS.

Results

UMVMR

The relationship between T, E2, SHBG, and MetS predicted by UVMR is depicted in Fig. 1 and summarized in Table 2. After removing SNPs in linkage disequilibrium and those with palindrome sequences, 459, 175, and 13 SNPs were identified in UVMR as instrumental variables for analyzing the causal association between SHBG, T, E2, and the risk of developing MetS, respectively. The F-statistic for each SNP was greater than 10, indicating strong correlation with exposure and no weak instrument bias. IVW method results revealed that genetically predicted SHBG (β=-0.72, 95% CI 0.41 to 0.57, P < 0.001) and T (OR = -0.40, 95% CI 0.54 to 0.84, P < 0.001) significantly decreased the risk of MetS. Cochran’s Q statistic detected significant heterogeneity among SNP effects (all P < 0.001).However, the results of this study remain reliable as the random effects IVW method allows for estimation of causal relationships in the presence of heterogeneity. Nonetheless, sensitivity analysis revealed horizontal pleiotropy between SHBG, T, and MetS (P < 0.05), while no horizontal pleiotropy was found between E2 and MetS (P > 0.05). In the global MR-PRESSO analysis, outliers were identified. After excluding these outliers, no pleiotropy was found between SHBG, T, and MetS, and the overall results remained consistent. For detailed information, please refer to Table 3.

Fig. 1
figure 1

Scatter plots of UVMR (a) for E on MetS; (b) for T on MetS; (c) for SHBG on MetS. T, Testosterone; E2, estradiol; SHBG, sex hormone binding protein; MetS, metabolic syndrom

Full size image
Table 2 UVMR analysis results
Full size table
Table 3 Sensitivity analysis results for UVMR and MVMR
Full size table

MVMR

Based on the SNP selection method mentioned above, a total of 469, 457, 311, 456, 308, and 307 final instrumental variables associated with the three sex hormones and their associations with MetS, WC, FBG, HBP, HDL-C, and TG were selected. The F-statistic for each SNP exceeded 10, indicating that the results were unlikely to be influenced by weak instrument bias.After adjusting for E2 and T, MVMR results showed a causal relationship between SHBG and MetS (β=-0.60, 95% CI -0.83 to -0.38, P < 0.001), while no causal relationship was detected between E2, T, and MetS. The results were consistent across other MR analysis methods. In the subgroup analysis of SHBG, E2, and T with MetS, after adjusting for other variables, according to the IVW results, a causal relationship was observed between T (β = -0.13, 95% CI -0.24 to -0.02, P = 0.02), SHBG (β = 0.10, 95% CI 0.03 to 0.17, P = 0.004), and WC. MR-Egger showed consistent causal directions with IVW. No causal relationships were detected between SHBG, E2, T, and FBG, with consistent causal directions across the three methods. A causal relationship was found between SHBG (β = -0.02, 95% CI -0.04 to -0.00, P = 0.02) and HBP, with consistent causal directions across the three methods. A causal relationship was observed between SHBG (β = 0.41, 95% CI 0.21 to 0.60, P < 0.001) and HDL-C, with consistent causal directions across the three methods. A causal relationship was found between E2 (β = -1.49, 95% CI -2.48 to -0.50, P = 0.003), SHBG (β = -0.48, 95% CI -0.70 to -0.26, P < 0.001), and TG, with MR-Egger showing consistent causal directions with IVW (See Table 4 for details). Cochran’s Q statistic detected significant heterogeneity among SNP effects (all P < 0.001). Since the random-effects IVW allows estimation of causal relationships in the presence of heterogeneity, the results of this study remain reliable. In examining the associations between observational hormones and MetS, FBG, HBP, and TG, the MR-Egger intercept test revealed no significant pleiotropic effects (P > 0.05), indicating that the observed associations were not attributable to pleiotropy. Conversely, for the associations between observational hormones and WC and HDL, the MR-Egger intercept test showed significant directional pleiotropy (P < 0.05). The MR-PRESSO global test identified outliers. However, after these outliers were excluded, no pleiotropy was detected between sex hormones and WC and HDL, and the causal associations between sex hormones and WC and HDL remained consistent. For detailed information, please refer to Table 3.

In summary, after adjusting for E2 and T, SHBG was still causally associated with MetS, consistent with UVMR results, and no pleiotropy was detected. However, MVMR did not detect causal relationships between E2, T, and MetS. Therefore, the results are stable and reliable, suggesting a significant causal association between SHBG and MetS.

Table 4 MVMR analysis results
Full size table

Discussion

The aim of this study is to elucidate the causal relationship between sex hormones and MetS using UMVMR and MVMR methods. Initially, the UMVMR study identified a negative causal relationship between SHBG and MetS, as well as between T and MetS. Further MVMR analysis demonstrated a significant negative correlation between genetically predicted SHBG and MetS, with sensitivity analyses confirming the reliability of these results. Additionally, UMVMR indicated a causal link between T and MetS, but no causal association between E2 and MetS. while no causal relationship was found between E2 and MetS. However, subsequent MVMR analysis did not detect any causal relationship between T, E2, and MetS, likely because SHBG predominantly binds to T, with a binding affinity for T that is three times that of estrogens [26], suggesting that the relationship between T and MetS might occur through SHBG. When SHBG was controlled, no causal relationship between T and MetS was observed. However, Our study also found a causal relationship between genetically predicted SHBG and other MetS components (except for FBG), suggesting SHBG primarily influences MetS through its effects on HBP, TG, WC, and HDL-C. There is extensive evidence that low levels of SHBG are associated with the risk of developing MetS, including obesity, cardiovascular disease, and also with elevated levels of serum inflammatory markers [27].Another prospective study reported a negative correlation between SHBG and HBP, with no apparent gender differences [28], aligning with our findings. Moreover, in examining the associations between E2, T, and subcomponents of MetS, we discovered a causative negative correlation between E2 and TG, supported by observational studies. A cross-sectional study [29] found that E2 is related to TG in women of childbearing age, while another observational study involving 7-year-old preadolescent girls found no association between E2 and metabolic markers like TG [30], possibly due to differences in the age of the study populations. An observational study involving Black, Hispanic, and White male residents of Boston found a significant correlation between E2 and WC, as well as with other obesity indicators such as weight, BMI, and hip circumference. This may be attributed to male E2 being derived from T through aromatase, which is abundant in adipose tissue, suggesting that greater adiposity may facilitate E2 synthesis [31,32,33]. Our study also suggests a potential positive correlation between E2 and WC, though no statistical association was established, possibly due to racial differences.

MVMR analysis also revealed a negative causal relationship between T and TG as well as WC. Observational studies have similarly noted significant improvements in WC following T replacement therapy for MetS [34, 35]. An observational study involving 7,268 healthy male participants demonstrated a negative correlation between T and TG [36], supported by a meta-analysis indicating that T significantly reduces bodily TG levels [37].These findings corroborate our results.T can decrease lipid synthesis and increase fat breakdown, thereby reducing serum TG levels and aiding in the prevention of progression to atherosclerosis [38]. An increase in WC indicates the accumulation of visceral fat, which elevates the risk of cardiovascular disease. It has been reported that low testosterone levels can lead to accumulation of visceral fat [39, 40].

Metabolic Syndrome (MetS) is a severe metabolic disorder significantly affecting patients’ quality of life, with an incidence rate of 37.1% in Asian populations, comparable to that in Europeans [41, 42]. It serves as the pathological basis for cardiovascular events and diabetes, with complex pathophysiological mechanisms. Research [43] indicates that the Mediterranean diet can mitigate cognitive decline in patients with MetS. This beneficial effect is believed to be linked to its role in regulating lipid homeostasis. SHBG can influence MetS through several pathways: Firstly, by reducing hepatic fat content, thereby alleviating MetS. Secondly, inflammation is a crucial component of MetS [44], and inflammatory-driven vascular remodeling can lead to the occurrence of HBP, while SHBG can mitigate MetS and its component HBP by inhibiting inflammatory factors such as TNF-α and IL-6 [27]. Thirdly, previous observational studies and our MR findings suggest that SHBG may mitigate the occurrence of MetS through factors like HBP, TG, WC, and HDL-C. Previous observational studies [45] have suggested an association between SHBG and diabetes. However, our MR study included FBG data, which is susceptible to stress and other factors and is insufficient for diagnosing diabetes, thus the link between SHBG and diabetes requires further verification. Identifying the influencing factors of MetS using reliable methods is crucial for the diagnosis and treatment of the disease. However, as a complex metabolic disorder, there is a lack of evidence on the causal effects of sex hormones on the risk of MetS. Our study addresses this gap. Given the positive causal relationship between SHBG and the development of MetS, clinical guidelines should recommend monitoring SHBG as a primary preventive measure. We propose that more proactive monitoring of SHBG could reduce the incidence of MetS. Likewise, the causal effects of high levels of T and E2 on TG suggest that T and E2 positively influence lipid homeostasis, emphasizing their important role in the progression of hypertriglyceridemia to atherosclerosis.

Conclusion

To our knowledge, this is the first study to establish causal relationships between sex hormones and MetS and its components using an MR framework. Understanding the potential influencing factors of MetS, a complex metabolic disorder, can help with clinical prevention and treatment. Our study provides valuable insights into the etiology and clinical treatment of MetS. Exploring the molecular mechanisms underlying these causal relationships can effectively aid clinicians in developing targeted treatment strategies and personalized management plans for patients.

Strengths and limitations

This study has several advantages: our MR research avoids the inherent weaknesses of observational studies, providing suggestive evidence of a causal protective effect of SHBG on MetS. Additionally, we utilized comprehensive GWAS summary-level data on MetS and its components, thereby possessing high investigative power. Lastly, a series of sensitivity analyses were conducted to enhance the credibility of our results. However, there are some limitations to note: Firstly, our study primarily involved European populations, requiring further research to determine if our results can be generalized to other ethnic groups. Secondly, our data did not stratify by gender, indicating a need for further investigation in the future.

Data availability

No datasets were generated or analysed during the current study.

References

  1. Fahed G, Aoun L, Bou Zerdan M, et al. Metabolic syndrome: updates on pathophysiology and management in 2021[J]. Int J Mol Sci. 2022;23(2):786.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  2. Saklayen MG. The global epidemic of the metabolic Syndrome[J]. Curr Hypertens Rep. 2018;20(2):12.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Zarocostas J. Need to increase focus on non-communicable diseases in global health, says WHO[J]. BMJ. 2010;341:c7065.

    Article  PubMed  Google Scholar 

  4. Taxier LR, Gross KS, Frick KM. Oestradiol as a neuromodulator of learning and memory[J]. Nat Rev Neurosci. 2020;21(10):535–50.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  5. Barros RP, Gustafsson J. Estrogen receptors and the metabolic network[J]. Cell Metab. 2011;14(3):289–99.

    Article  PubMed  CAS  Google Scholar 

  6. Bourebaba N, Ngo T, Śmieszek A, et al. Sex hormone binding globulin as a potential drug candidate for liver-related metabolic disorders treatment[J]. Biomed Pharmacother. 2022;153:113261.

    Article  PubMed  CAS  Google Scholar 

  7. Winters SJ, Gogineni J, Karegar M, et al. Sex hormone-binding globulin gene expression and insulin resistance[J]. J Clin Endocrinol Metab. 2014;99(12):E2780–8.

    Article  PubMed  CAS  Google Scholar 

  8. Marycz K, Wiatrak B, Irwin-Houston JM, et al. Sex hormone binding globulin (SHBG) modulates mitochondrial dynamics in PPARγ-depleted equine adipose derived stromal cells[J]. J Mol Med (Berl). 2024;102(8):1015–36.

    Article  PubMed  CAS  Google Scholar 

  9. Liao CH, Huang CY, Li HY, et al. Testosterone and sex hormone-binding globulin have significant association with metabolic syndrome in Taiwanese men[J]. Aging Male. 2012;15(1):1–6.

    Article  PubMed  Google Scholar 

  10. Sá EQ, Sá FC, Oliveira KC, et al. Association between sex hormone-binding globulin (SHBG) and metabolic syndrome among men[J]. Sao Paulo Med J. 2014;132(2):111–5.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Liu C, Zhao M, Zhao Y, Hu Y. Association between serum total testosterone levels and metabolic syndrome among adult women in the United States, NHANES 2011–2016[J]. Front Endocrinol (Lausanne). 2023;14:1053665.

    Article  PubMed  Google Scholar 

  12. Assyov Y, Gateva A, Karamfilova V, et al. Impact of testosterone treatment on circulating irisin in men with late-onset hypogonadism and metabolic syndrome[J]. Aging Male. 2020;23(5):1381–7.

    Article  PubMed  Google Scholar 

  13. Brand JS, Van Der Tweel I, Grobbee DE, et al. Testosterone, sex hormone-binding globulin and the metabolic syndrome: a systematic review and meta-analysis of observational studies[J]. Int J Epidemiol. 2011;40(1):189–207.

    Article  PubMed  Google Scholar 

  14. Bowden J, Holmes MV. Meta-analysis and mendelian randomization: a review[J]. Res Synth Methods. 2019;10(4):486–96.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Yavorska OO, Burgess S. MendelianRandomization: an R package for performing mendelian randomization analyses using summarized data[J]. Int J Epidemiol. 2017;46(6):1734–9.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Smith GD, Ebrahim S. Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease?[J]. Int J Epidemiol. 2003;32(1):1–22.

    Article  PubMed  Google Scholar 

  17. Sanderson E. Multivariable mendelian randomization and Mediation[J]. Cold Spring Harb Perspect Med. 2021;11(2):a038984.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  18. Ruth KS, Day FR, Tyrrell J, et al. Using human genetics to understand the disease impacts of testosterone in men and women[J]. Nat Med. 2020;26(2):252–8.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  19. Van Walree ES, Jansen IE, Bell NY, et al. Disentangling genetic risks for metabolic Syndrome[J]. Diabetes. 2022;71(11):2447–57.

    Article  PubMed  Google Scholar 

  20. Manning AK, Hivert MF, Scott RA, et al. A genome-wide approach accounting for body mass index identifies genetic variants influencing fasting glycemic traits and insulin resistance[J]. Nat Genet. 2012;44(6):659–69.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Willer CJ, Schmidt EM, Sengupta S, et al. Discovery and refinement of loci associated with lipid levels[J]. Nat Genet. 2013;45(11):1274–83.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  22. Wang B, Liu D, Song M, et al. Immunoglobulin G N-glycan, inflammation and type 2 diabetes in east Asian and European populations: a mendelian randomization study[J]. Mol Med. 2022;28(1):114.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  23. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in mendelian randomization with some Invalid instruments using a weighted median Estimator[J]. Genet Epidemiol. 2016;40(4):304–14.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression[J]. Int J Epidemiol. 2015;44(2):512–25.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Burgess S, Thompson SG. Interpreting findings from mendelian randomization using the MR-Egger method[J]. Eur J Epidemiol. 2017;32(5):377–89.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Abu-Hijleh TM, Gammoh E, Al-Busaidi AS, et al. Common variants in the sex hormone-binding globulin (SHBG) gene influence SHBG levels in women with polycystic ovary Syndrome[J]. Ann Nutr Metab. 2016;68(1):66–74.

    Article  PubMed  CAS  Google Scholar 

  27. Yamazaki H, Kushiyama A, Sakoda H et al. Protective Effect of Sex Hormone-Binding Globulin against Metabolic Syndrome: In Vitro Evidence Showing Anti-Inflammatory and Lipolytic Effects on Adipocytes and Macrophages[J]. Mediators Inflamm, 2018, 2018: 3062319.

  28. Watz MES, Tivesten Å, Ottarsdottir K, et al. Sex hormone-binding globulin levels and development of hypertension in middle-aged men and women[J]. J Hypertens. 2023;41(10):1565–70.

    Article  PubMed  CAS  Google Scholar 

  29. Mezzullo M, Gambineri A, Di Dalmazi G, et al. Steroid reference intervals in women: influence of menopause, age and metabolism[J]. Eur J Endocrinol. 2021;184(3):395–407.

    Article  PubMed  Google Scholar 

  30. Mesa Valencia DC, Mericq V, Corvalán C, Pereira A. Obesity and related metabolic biomarkers and its association with serum levels of estrogen in pre-pubertal Chilean Girls[J]. Endocr Res. 2020;45(2):102–10.

    Article  PubMed  CAS  Google Scholar 

  31. Gates MA, Mekary RA, Chiu GR, et al. Sex steroid hormone levels and body composition in men[J]. J Clin Endocrinol Metab. 2013;98(6):2442–50.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  32. Kaufman JM, Vermeulen A. The decline of androgen levels in elderly men and its clinical and therapeutic implications[J]. Endocr Rev. 2005;26(6):833–76.

    Article  PubMed  CAS  Google Scholar 

  33. Vermeulen A, Kaufman JM, Goemaere S, Van Pottelberg I. Estradiol in elderly men[J]. Aging Male. 2002;5(2):98–102.

    Article  PubMed  CAS  Google Scholar 

  34. Bhattacharya RK, Khera M, Blick G, et al. Effect of 12 months of testosterone replacement therapy on metabolic syndrome components in hypogonadal men: data from the Testim Registry in the US (TRiUS)[J]. BMC Endocr Disord. 2011;11:18.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  35. Kalinchenko SY, Tishova YA, Mskhalaya GJ, et al. Effects of testosterone supplementation on markers of the metabolic syndrome and inflammation in hypogonadal men with the metabolic syndrome: the double-blinded placebo-controlled Moscow study[J]. Clin Endocrinol (Oxf). 2010;73(5):602–12.

    Article  PubMed  CAS  Google Scholar 

  36. Lee JH, Jung HD, Choi JD, et al. Non-linear association between testosterone and LDL concentrations in men[J]. Andrology. 2023;11(6):1107–13.

    Article  PubMed  CAS  Google Scholar 

  37. Kumar S, Khatri M, Memon RA, et al. Effects of testosterone therapy in adult males with hypogonadism and T2DM: a meta-analysis and systematic review[J]. Diabetes Metab Syndr. 2022;16(8):102588.

    Article  PubMed  CAS  Google Scholar 

  38. Raz I, Eldor R, Cernea S, Shafrir E. Diabetes: insulin resistance and derangements in lipid metabolism. Cure through intervention in fat transport and storage[J]. Diabetes Metab Res Rev. 2005;21(1):3–14.

    Article  PubMed  CAS  Google Scholar 

  39. Wang ZY, Li BC, Xing JJ, et al. Associations of waist circumference with sex steroid hormones among 4031 USA children and adolescents[J]. Asian J Androl. 2023;25(4):505–11.

    Article  PubMed  CAS  Google Scholar 

  40. Turan E, Öztekin Ü. Relationship between visceral adiposity index and male infertility[J]. Andrologia. 2020;52(4):e13548.

    Article  PubMed  CAS  Google Scholar 

  41. Čypienė A, Gimžauskaitė S, Rinkūnienė E, et al. The Association between Water Consumption and Hyperuricemia and its relation with early arterial aging in Middle-aged Lithuanian metabolic Patients[J]. Nutrients. 2023;15(3):723.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Ranasinghe P, Mathangasinghe Y, Jayawardena R, et al. Prevalence and trends of metabolic syndrome among adults in the asia-pacific region: a systematic review[J]. BMC Public Health. 2017;17(1):101.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  43. Soldevila-Domenech N, Fagundo B, Cuenca-Royo A, et al. Relationship between sex, APOE genotype, endocannabinoids and cognitive change in older adults with metabolic syndrome during a 3-year Mediterranean diet intervention[J]. Nutr J. 2024;23(1):61.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  44. Guzik TJ, Touyz RM. Oxidative stress, inflammation, and vascular aging in Hypertension[J]. Hypertension. 2017;70(4):660–7.

    Article  PubMed  CAS  Google Scholar 

  45. Simons P, Valkenburg O, Van De Waarenburg MPH, et al. Serum sex hormone-binding globulin is a mediator of the association between intrahepatic lipid content and type 2 diabetes: the Maastricht Study[J]. Diabetologia. 2023;66(1):213–22.

    Article  PubMed  CAS  Google Scholar 

Download references

Acknowledgements

The authors thank CTG-CNCR, GLGC, UKB for providing the data for our research and all the participants and researchers for their participation in this MR study.

Funding

No funds were available to support the study.

Author information

Author notes

  1. Siyuan Liu and Zhuosong Mu contributed equally to this work.

Authors and Affiliations

  1. The Third Clinical Medical College of Zhejiang, University of Traditional Chinese Medicine, Hangzhou, Zhejiang, China

    Siyuan Liu & Xinyi Chen

  2. Jiangnan Hospital Affiliated to Zhejiang, Chinese Medical University (Hangzhou Xiaoshan Hospital of Traditional Chinese Medicine), Hangzhou, Zhejiang, China

    Zhuosong Mu

  3. The Third Affiliated Hospital of Zhejiang, University of Traditional Chinese Medicine, Hangzhou, Zhejiang, China

    Yingying Xu

Authors
  1. Siyuan Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Zhuosong Mu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Xinyi Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yingying Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

LSY, MZS and CXY were involved in the design of this study. LSY and MZS performed data acquisition, analysis, interpretation, and drafted the manuscript. XYY performed data interpretation. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Yingying Xu.

Ethics declarations

Ethical approval

We did not require additional ethical approval because we used publicly available summary-level GWAS data and the included studies were approved by the ethics committee.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, S., Mu, Z., Chen, X. et al. The impact of sex hormones on metabolic syndrome: univariable and multivariable Mendelian randomization studies. Diabetol Metab Syndr 16, 215 (2024). https://doi.org/10.1186/s13098-024-01443-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13098-024-01443-4

Keywords

Diabetology & Metabolic Syndrome

ISSN: 1758-5996

Contact us