Skip to main content
Download PDF

The causal relationship between steroid hormones and risk of stroke: evidence from a two-sample Mendelian randomization study

Molecular Brain volume 18, Article number: 6 (2025) Cite this article

Abstract

It is unclear how steroid hormones contribute to stroke, and conducting randomized controlled trials to obtain related evidence is challenging. Therefore, Mendelian randomization (MR) technique was employed in this study to examine this association. Through genome-wide association meta-analysis, the genetic variants of steroid hormones, including testosterone/17β-estradiol (T/E2) ratio, aldosterone, androstenedione, progesterone, and hydroxyprogesterone, were acquired as instrumental variables. Analysis was done on the impact of these steroid hormones on the risk of stroke subtypes. The T/E2 ratio was associated to an elevated risk of small vessel stroke (SVS) according to the inverse variance weighted approach which was the main MR analytic technique (OR, 1.23, 95% CI: 1.05–1.44, p = 0.009). These findings were solid since no heterogeneity nor horizontal pleiotropy were found. The causal association between T/E2 and SVS was also confirmed in the replication study (p = 0.009). Nevertheless, there was no proof that other steroid hormones increased the risk of stroke. According to this study, T/E2 ratio and SVS are causally related. However, strong evidence for the impact of other steroid hormones on stroke subtypes is still lacking. These findings may be beneficial for developing stroke prevention strategies from steroid hormones levels.

Introduction

Steroid hormones are tetracyclic aliphatic hydrocarbon compounds with a cyclopentanopolyhydrophenanthrene ring, including testosterone (T), 17β-estradiol (E2), aldosterone (Aldo), androstenedione (A4), progesterone (P4), and hydroxyprogesterone (17-OHP) [1]. These hormones are mainly biosynthesized from cholesterol in the adrenal cortex and gonads [2]. They play roles in various physiological functions, such as regulating blood pressure, controlling sexual function, and immunoregulation [3,4,5]. There is a complex synthetic network among steroid hormones. When the concentration of one component becomes abnormal due to diseases of the adrenal cortex and gonads, the dynamic balance of steroid hormones in the body may be disrupted, leading to significant changes in their levels. Moreover, steroid hormone secretion disorders can affect other crucial systems, such as the cardiovascular system [6, 7]. For example, primary aldosteronism will lead to water and sodium retention, which will lead to secondary hypertension, and increase the incidence of cardiovascular diseases [6].

Stroke, including ischemic and hemorrhagic types, is a common cardiovascular disease and a major cause of death and disability worldwide, with a continuous upward trend [8, 9]. Recently, hypertension, smoking, alcohol consumption, diabetes, and hyperlipidemia have been identified as stroke risk factors. However, due to the various pathologies of stroke, identifying risk factors is complex [10]. Many observational studies have found that steroid hormone disorders significantly influence the cardiovascular system and contribute to stroke [11, 12]. However, confounders and reverse causality are two main biases that might affect observational research. Randomized controlled trials investigating this association are also difficult to perform due to ethical concerns. Furthermore, most epidemiological data have not examined the connections between stroke subtypes and steroid hormones. Therefore, determining the specific causal impact of different hormones on stroke subtypes can lead to more precise prevention strategies.

In the Mendelian randomization (MR) technique, genetic variants are used as instrumental variables (IVs) to refer to exposure factors, which applies the rule of independent assortment to assess the causal influence of exposure on the outcome. MR generally avoids the effects of confounders and reverse causality [13]. Additionally, single nucleotide polymorphisms (SNPs) can be employed to infer causality in MR research based on data from genome-wide association meta-analysis (GWAMA). Thus, using summary GWAMA data, we performed two-sample MR analyses in this work to investigate the genetic association of steroid hormones and stroke subtypes.

Methods

Study design

In this MR study, genetic variants associated with steroid hormones were used as IVs to examine the causal role of steroid hormones in stroke subtypes. IVs should satisfy the following assumptions: (1) exposure is strongly related to IVs; (2) confounders are not related to IVs; and (3) exposure alone is how IVs influence the result (Fig. 1) [14]. The study process of this MR analysis is presented in Fig. 2. In the original trials, appropriate ethical approval and informed consent from the patients were obtained, and no additional ethical approval was required.

Fig. 1
figure 1

Mendelian randomization model of the present study. The design is based on the assumptions that the instrumental variables (1) are associated with steroid hormones, (2) are not associated with confounders, and (3) influence stroke subtypes only through steroid hormones. SNPs, single nucleotide polymorphisms

Full size image
Fig. 2
figure 2

Workflow of the present Mendelian randomization study

Full size image

Data source

Table S1 displays the participant characteristics from the datasets on steroid hormones and stroke. The participants were matched by ethnicity but not by age or gender. The steroid hormone data were obtained from a recent GWAMA conducted in individuals of European descent [15]. This study produced GWAMA data with adjustments for sex, age, and log-transformed BMI from two separate investigations, LIFE-Adult (1481 cases) [16] and LIFE-Heart (2068 cases) [17], concerning one steroid hormone ratio (T/E2) and the levels of four steroid hormones, including Aldo, A4, P4, and 17-OHP.

Data on 8 stroke subtypes from different GWAMA studies were used as outcome variables. The MEGASTROKE consortium produced the genetic information for any ischemic stroke (AIS, 4,217 cases and 406,111 controls) and its subtypes (large artery stroke (LAS; n = 7,193), cardioembolic stroke (CES; n = 5,386), and small vessel stroke (SVS; n = 4,373)) [18]. Patients with LAS exhibit clinical and brain imaging manifestations of either obvious stenosis or occlusion of the large cerebral artery or cortical branch, possibly caused by atherosclerosis. CES includes patients who may experience arterial occlusion due to an embolus arising in the heart. SVS usually refers to patients with lacunar infarction.

There are two types of hemorrhagic stroke: intracerebral hemorrhage (ICH) and subarachnoid hemorrhage (SAH). The International Stroke Genetics Consortium released European GWAMA consisting 1,545 ICH cases (664 lobar and 881 deep) and 1,481 controls, as the source of the genetic information for ICH [19]. A recent European GWAMA of cerebral aneurysms, which included 5,140 SAH patients caused by cerebral aneurysm rupture and 71,934 controls, provided data on SAH since non-traumatic SAH is predominantly caused by aneurysm rupture [20].

To further verify the reliability of the results, summary statistics for SVS in individuals of European ancestry (6,030 cases, 219,389 controls) were extracted from another GWAMA conducted by Traylor et al. in 2021 for replication analysis [21].

Selection of instrumental variables

Similar to other MR research, the threshold was set at p < 1 × 10− 5 in order to obtain sufficient IVs [22]. This was necessary because there were insufficient SNPs associated with exposure when the threshold was set as 5 × 10− 8. All SNPs with locus-wide significance were then clumped based on linkage disequilibrium threshold (r2 = 0.001) and a distance threshold (10,000 kb) to create index SNPs [23]. Some index SNPs associated with potential risk factors were removed (Table S2) following a search of all index SNPs in PhenoScanner V2, which provides detailed information on the relationship between genotype and phenotype [24]. The initial instrumental SNPs were thus obtained (15 SNPs for T/E2, 16 SNPs for Aldo, 15 SNPs for A4, 28 SNPs for P4, and 10 SNPs for 17-OHP). Next, F-statistics were used to indicate the IVs’ strength [F = (N-K-1) × R2 / K / (1-R2); R2: the percentage of variation accounted for the relationship between the IVs; K: the number of IVs; N: the number of samples]. All remaining SNPs had F-statistics larger than 10 [25]. Finally, SNPs were picked up from and harmonized with outcome data to generate the final instrumental SNPs which were not associated with stroke subtypes and did not contain any mismatched or palindromic SNPs. The characteristics of the final instrumental SNPs related to steroid hormones and stroke subtypes are displayed in Table S3.

Statistical analyses

Three distinct methods were used in the present MR study: inverse-variance weighted (IVW), weighted median, and MR-Egger regression. Among all MR techniques, IVW is the most statistically powerful standard MR algorithm. It provides MR estimate by combining each Wald ratio of instrumental SNPs [26]. If invalid IVs account for less than 50% of the weight, the weighted median can yield accurate estimations [27]. MR-Egger regression is mainly used to detect and explain directional pleiotropy bias [28]. We also displayed the genetic associations of steroid hormones with stroke subtypes using scatterplots of genetic variants.

Furthermore, sensitivity analyses, which included tests for heterogeneity (IVs can affect multiple aspects of exposure factors) and horizontal pleiotropy (IVs can directly affect outcomes without exposure factors), were conducted to identify various violations of assumptions. Cochran’s Q test was used in the IVW approach to quantify heterogeneity. There was heterogeneity in the causal effects across all SNPs when the p value of the Cochran’s Q statistic was less than 0.05 [29]. Horizontal pleiotropy was tested by the intercept test of MR-Egger and the MR-pleiotropy residual sum and outlier (MR-PRESSO) test. First, horizontal pleiotropy was shown by a non-zero MR-Egger regression intercept [28]. Second, the global and outlier test in MR-PRESSO could identify horizontal pleiotropic outliers and adjust pleiotropy by removing outliers [30]. Additionally, leave-one-out analysis was employed to see whether a particular SNP had an impact on the overall MR estimate.

The statistically significant p value criterion was chosen at 0.01 following the Bonferroni correction because this MR study covered five exposures. To determine the suggestive causal association, a p value between the Bonferroni-corrected significance level and the conventional significance threshold (0.05) was employed.

R software (version 4.2.1) and its companion R package, TwoSampleMR (version 0.5.6), were used to perform all of the statistical studies indicated above.

Results

Causal effect of T/E2 ratio on stroke subtypes

According to the IVW technique and the weighted median method, the T/E2 ratio had an obvious causal influence on SVS (OR, 1.23, 95% CI: 1.05–1.44, p = 0.009) and a suggestive causal effect on LAS (OR: 0.82, 95% CI: 0.67–0.99, p = 0.04) (Fig. 3). The estimate was directionally compatible with IVW and weighted median analysis, despite the fact that the MR-Egger study did not yield significant results (Figs. 3 and 4a, and Figure S1). No causal role of the T/E2 ratio was detected in other stroke subtypes. Furthermore, horizontal pleiotropy and heterogeneity were not detected by sensitivity analysis, confirming the MR analysis’s robustness (Table S4). Finally, the IVW point estimate was not dominated by any particular SNP, as the leave-one-out analysis verified. (Fig. 4b and Figure S1). Moreover, there was no gender differences in this association (Table S5).

Fig. 3
figure 3

MR association of genetically determined T/E2 ratio with stroke subtypes. MR, Mendelian randomization; T/E2, testosterone/17β-estradiol; AIS, any ischemic stroke; LAS, large artery stroke; CES, cardioembolic stroke; SVS, small vessel stroke; ICH, intracerebral hemorrhage; SAH, subarachnoid hemorrhage; SNPs, single nucleotide polymorphisms; IVW, inverse variance weighted; OR, odd ratio; CI, confidence interval

Full size image
Fig. 4
figure 4

Scatter plot (a) and leave-one-out test (b) for genetically determined T/E2 ratio and risk of SVS. T/E2, testosterone/17β-estradiol; SVS, small vessel stroke; MR, mendelian randomization; SNP, single nucleotide polymorphism

Full size image

Causal effect of Aldo on stroke subtypes

The IVW results showed that Aldo had no causal influence on any stroke subtype, as seen in Figure S2 (all p > 0.05). The evidence from the weighted median and MR-Egger techniques was comparable with the IVW findings for the association of Aldo with stroke subtypes, with the exception of the weighted median analysis’s suggestive significance for the causal influence of Aldo on LAS (p = 0.02) (Figure S2). Sensitivity analysis showed no heterogeneity, except for the impact of Aldo on AIS (Q, 30.85; p = 0.009). Neither the MR-Egger intercept nor the MR-PRESSO global test indicated directional pleiotropy for Aldo, apart from the relationship between Aldo and AIS (MR-PRESSO global test, p = 0.01) (Figure S2). The MR-PRESSO outlier test identified rs74911566 as an outlier SNP. After correcting for the outlier, the results remained non-significant (Table S6). Finally, the leave-one-out study demonstrated that the risk assessment of genetically predicted Aldo and the risk of stroke subtypes were largely consistent (Figure S3).

Causal effect of A4 on stroke subtypes

A4 showed no significant causal relationship with any stroke subtypes as detected by three MR method (all p > 0.05) (Figure S4). According to Cochran’s Q test, there was no heterogeneity in these results except for SAH (Q, 11.17; p = 0.02) and no pleiotropy was detected. More details are provided in Figure S4. The dependability of the data between A4 and all stroke subtypes was further validated by the leave-one-out analysis. (Figure S5).

Causal effect of P4 on stroke subtypes

A possible correlation between circulating P4 levels and the risk of deep ICH was found by the IVW study (OR: 0.59, 95% CI: 0.35–0.99, p = 0.048). However, there was no proof that P4 and other stroke subtypes were causally related (p > 0.05) (Figure S6). There was no discernible correlation between P4 and other stroke subtypes detected by weighted median or MR-Egger approaches (all p > 0.05) (Figure S6). Some heterogeneity was found between P4 and SVS (Q, 41.02; p = 0.02) as well as P4 and lobar SAH (Q, 27.19; p = 0.01). Though there was no directional pleiotropy between P4 and any of the stroke subtypes according to the MR-Egger intercept test, the MR-PRESSO global test produced a significant result (p = 0.02). One SNP, rs117848367, was identified as an outlier by the MR-PRESSO outlier test in the relationship between P4 and SVS. However, further outlier-corrected results were similar to those before correction (Table S7). Moreover, the causal association between P4 and any stroke subtypes was not demonstrated by a single IV, according to the leave-one-out analysis (Figure S7).

Causal effect of 17-OHP on stroke subtypes

There were no significant results from any MR methods to support the causal impact of 17-OHP on stroke subtypes (all p > 0.05) (Figure S8). According to sensitivity analysis, the results were robust and not affected by heterogeneity or directional pleiotropy (Figure S8). Furthermore, the robustness of the data between 17-OHP and each stroke subtype was further validated using the leave-one-out approach (Figure S9).

Replication analysis

Since SVS was found to have a causal association with the T/E2 ratio, summary statistics for SVS from another GWAMA were used for replication analysis (Figure S10). Although the other two MR analyses did not provide significant causal associations between the T/E2 ratio and SVS in the replication stage, the estimate conducted from the IVW approach agreed with primary findings (OR, 1.23, 95% CI: 1.02–1.47, p = 0.009). No causal relationship was found between the other four hormone exposure factors and SVS. Additionally, sensitivity analysis showed no heterogeneity, directional pleiotropy, or outliers. The causal relationship between each steroid hormone and SVS was not driven by any specific IV, according to the leave-one-out analysis (Figure S11).

Disccusion

After comprehensively evaluating the causal association of different steroid hormones abd various stroke subtypes using data from GWAMA, the current study indicated that an increasing T/E2 ratio had a positive causal correlation with SVS risk but not with other stroke subtypes. This finding was confirmed in the replication stage. Moreover, there was no gender difference in this relationship. The present study suggests that an increase in serum T concentration or a decrease in serum E2 concentration will raise the risk of SVS. However, the study did not support a role for any other steroid hormones in stroke risk.

There is currently no solid evidence about how T affects the risk of stroke. The largest related study included over 80,000 male veterans who had filled a T prescription and over 120,000 who had not [31]. The composite outcome of IS, acute myocardial infarction, or venous thromboembolic disease did not correlate with intramuscular T therapy. However, other research has revealed that fluctuating T levels during a man’s life, with higher levels in young men and lower levels in elderly men, may increase the risk of IS [32, 33]. The effect of E2 levels on stroke is also inconclusive. Alonso de Leciñana et al. studied postmenopausal women in a multicenter, age-matched, case-control study, which indicated that longer-term ovarian E2 exposure might prevent noncardioembolic IS [34]. However, an analysis of data from a postmenopausal women’s nested case-control study from the Nurses’ Health Study by Hu et al. concluded that total or free E2 levels had no role in mediating the risk of IS [35]. Moreover, Abbott et al.‘s follow-up analysis of older men who took part in the Honolulu-Asia Aging analysis revealed that elevated blood E2 levels may be related to an elevated risk of stroke in older men [36].

The conflicting outcomes from these observational studies may be resulted from small sample sizes and the presence of confounding factors that are difficult to eliminate. Additionally, the small sample sizes prevent further analysis of stroke subtypes. Furthermore, it is not obvious if the association between the risk of stroke and circulating levels of T and E2 shown in observational studies is causative or an example of reverse causality. Furthermore, it is not obvious if the association between the risk of stroke and circulating levels of T and E2 shown in observational studies is causal or an example of reverse causality [37,38,39]. To get over the drawbacks of observational research and examine the relationship between the T/E2 ratio and stroke subtypes, we therefore carried out an MR investigation. According to our research, the T/E2 ratio may increase the risk of SVS.

Several potential mechanisms could support our findings. There is robust preclinical evidence of the neuroprotective and anti-inflammatory properties of E2. For example, Ghisletti et al. indicated that E2 could bind to E2 receptor α isoform to inhibit NF-κB signaling pathway and attenuate neuroinflammation. This mechanism involves PI3K activation [40, 41]. However, the role of T in stroke is not clear. Many conflicting findings have been reported in studies evaluating the effects of T on atherosclerosis and lipid metabolism. T has been indicated to increase blood viscosity and platelet activation, promoting thrombosis. Additionally, T could improve renal salt and water retention [42, 43]. Interestingly, by converting to E2 via aromatase in endothelial cells, T may reduce the expression of vascular cell adhesion molecule-1 [44]. This evidence suggests that E2 is more likely to inhibit the occurrence of stroke, whereas the function of T remains unclear. Furthermore, one study has indicated that T could impair microvascular endothelial function [45], and another study showed that compared with large vessels, T was more likely to damage small vessels [46]. This information could explain why an elevated T concentration seems to be a unique risk factor for SVS. Although our research has theoretical support, more studies are needed to reinforce this perspective.

The following limitations should be mentioned in the MR analysis’s findings. First, our data do not support a significant influence of steroid hormones other than the T/E2 ratio on stroke subtype risk. The reason for this phenomenon maybe the sample size of exposure was too small so there was insufficient SNPs to obtain significant MR results. Thus, the sample size of GWAMA about steroid hormones should be increased in subsequent research to collect more reliable IVs for additional validation and more conclusive results. Second, the association between Aldo, A4, and P4 and the risk of stroke subtypes should be assessed with care because potential heterogeneity, pleiotropy, or outliers could be present. Lastly, there was a restriction to extrapolating the discovered causal relationships to other groups with different genetic backgrounds because only data from summary statistics for people with European ancestry were included.

In conclusion, a causal association between the T/E2 ratio and SVS was found by MR analysis. However, strong evidence for the impact of other steroid hormones on stroke subtypes is still lacking. According to our findings, an increase in T or a decrease in E2 during continuous monitoring of steroid hormone levels in the blood may be a biomarker for the occurrence of SVS. Therefore, patients with diseases that may experience this trend, such as men with testicular tumors and menopausal women, should be particularly aware of the possibility of developing SVS and should be closely followed up. Further investigation is warranted to explore the specific mechanisms of T and E2 in the pathogenesis of SVS.

Data availability

All data used in our study are publicly available, which could be available from https://www.mdpi.com/article/10.3390/metabo11110738/s1 (steroid hormones); http://www.megastroke.org/download.html (ischemic stroke); https://cd.hugeamp.org/ downloads.html (intracerebral hemorrhage); https://doiorg.publicaciones.saludcastillayleon.es/10.6084/m9.fifigshare.11303372 (subarachnoid hemorrhage).

Abbreviations

E2:

17β-estradiol

Aldo:

Aldosterone

A4:

Androstenedione

AIS:

Any ischemic stroke

CES:

Cardioembolic stroke

GWAMA:

Genome-wide association meta-analysis

17-OHP:

Hydroxyprogesterone

IVs:

Instrumental variables

ICH:

Intracerebral hemorrhage

IVW:

Inverse-variance weighted

LAS:

Large artery stroke

MR:

Mendelian randomization

MR-PRESSO:

MR-pleiotropy residual sum and outlier

P4:

Progesterone

SNPs:

Single nucleotide polymorphisms

SVS:

Small vessel stroke

SAH:

Subarachnoid hemorrhage

T:

Testosterone

References

  1. Liu K, Fan H, Hu H, Cheng Y, Liu J, You Z. Genetic variation reveals the influence of steroid hormones on the risk of retinal neurodegenerative diseases. Front Endocrinol. 2022;13:1088557. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fendo.2022.1088557.

    Article  Google Scholar 

  2. Holst JP, Soldin OP, Guo T, Soldin SJ. Steroid hormones: relevance and measurement in the clinical laboratory. Clin Lab Med. 2004;24:105–18. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cll.2004.01.004.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Tomaschitz A, Pilz S, Ritz E, Obermayer-Pietsch B, Pieber TR. Aldosterone and arterial hypertension. Nat Rev Endocrinol. 2010;6:83–93. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrendo.2009.263.

    Article  PubMed  CAS  Google Scholar 

  4. Bouman A, Heineman MJ, Faas MM. Sex hormones and the immune response in humans. Hum Reprod. 2005;Update 11:411–23. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/humupd/dmi008.

    Article  CAS  Google Scholar 

  5. Feder HH. Hormones and sexual behavior. Annu Rev Psychol. 1984;35:165–200. https://doiorg.publicaciones.saludcastillayleon.es/10.1146/annurev.ps.35.020184.001121.

    Article  PubMed  CAS  Google Scholar 

  6. Lastra G, Dhuper S, Johnson MS, Sowers JR. Salt, aldosterone, and insulin resistance: impact on the cardiovascular system. Nat Rev Cardiol. 2010;7:577–84. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrcardio.2010.123.

    Article  PubMed  CAS  Google Scholar 

  7. Vitale C, Mendelsohn ME, Rosano GMC. Gender differences in the cardiovascular effect of sex hormones. Nat Rev Cardiol. 2009;6:532–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrcardio.2009.105.

    Article  PubMed  CAS  Google Scholar 

  8. Feigin VL, Nguyen G, Cercy K, Johnson CO, Alam T, Parmar PG, et al. Global, Regional, and Country-Specific Lifetime risks of Stroke, 1990 and 2016. N Engl J Med. 2018;379:2429–37. https://doiorg.publicaciones.saludcastillayleon.es/10.1056/NEJMoa1804492.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Wang J, Li S, Li C, Wu C, Song H, Ma Q, et al. Safety and efficacy of tirofiban in preventing neurological deterioration in acute ischemic stroke (TREND): protocol for an investigator-initiated, multicenter, prospective, randomized, open-label, masked endpoint trial. Brain Circ. 2024;10:168–73. https://doiorg.publicaciones.saludcastillayleon.es/10.4103/bc.bc_93_23.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Boehme AK, Esenwa C, Elkind MSV. Stroke risk factors, Genetics, and Prevention. Circ Res. 2017;120:472–95. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/CIRCRESAHA.116.308398.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  11. Milliez P, Girerd X, Plouin PF, Blacher J, Safar ME, Mourad JJ. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45:1243–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jacc.2005.01.015.

    Article  PubMed  CAS  Google Scholar 

  12. Glisic M, Mujaj B, Rueda-Ochoa OL, Asllanaj E, Laven JSE, Kavousi M, et al. Associations of endogenous estradiol and testosterone levels with plaque composition and risk of stroke in subjects with carotid atherosclerosis. Circ Res. 2018;122:97–105. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/CIRCRESAHA.117.311681.

    Article  PubMed  CAS  Google Scholar 

  13. Smith GD, Ebrahim S. Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol. 2003;32:1–22. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ije.

    Article  PubMed  Google Scholar 

  14. Emdin CA, Khera AV, Kathiresan S, Mendelian Randomization. JAMA. 2017;318:1925–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jama.2017.17219.

    Article  PubMed  Google Scholar 

  15. Pott J, Horn K, Zeidler R, Kirsten H, Ahnert P, Kratzsch J, et al. Sex-specific causal relations between steroid hormones and Obesity-A mendelian randomization study. Metabolites. 2021;11:738. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/metabo11110738.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  16. Loeffler M, Engel C, Ahnert P, Alfermann D, Arelin K, Baber R, et al. The LIFE-Adult-Study: objectives and design of a population-based cohort study with 10,000 deeply phenotyped adults in Germany. BMC Public Health. 2015;15:691. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-015-1983-z.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Scholz M, Henger S, Beutner F, Teren A, Baber R, Willenberg A, et al. Cohort Profile: the Leipzig Research Center for civilization diseases-Heart Study (LIFE-Heart). Int J Epidemiol. 2020;49:1439–h1440. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ije/dyaa075.

    Article  PubMed  Google Scholar 

  18. Malik R, Chauhan G, Traylor M, Sargurupremraj M, Okada Y, Mishra A, et al. Multiancestry genome-wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes. Nat Genet. 2018;50:524–37. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41588-018-0058-3.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  19. Woo D, Falcone GJ, Devan WJ, Brown WM, Biffi A, Howard TD, et al. Meta-analysis of genome-wide association studies identifies 1q22 as a susceptibility locus for intracerebral hemorrhage. Am J Hum Genet. 2014;94:511–21. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.ajhg.2014.02.012.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  20. Bakker MK, van der Spek RAA, van Rheenen W, Morel S, Bourcier R, Hostettler IC, et al. Genome-wide association study of intracranial aneurysms identifies 17 risk loci and genetic overlap with clinical risk factors. Nat Genet. 2020;52:1303–13. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41588-020-00725-7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Traylor M, Persyn E, Tomppo L, Klasson S, Abedi V, Bakker MK, et al. Genetic basis of lacunar stroke: a pooled analysis of individual patient data and genome-wide association studies. Lancet Neurol. 2021;20:351–61. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/S1474-4422(21)00031-4.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  22. Sanna S, van Zuydam NR, Mahajan A, Kurilshikov A, Vich Vila A, Võsa U, et al. Causal relationships among the gut microbiome, short-chain fatty acids and metabolic diseases. Nat Genet. 2019;51:600–5. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41588-019-0350-x.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  23. Noyce AJ, Kia DA, Hemani G, Nicolas A, Price TR, De Pablo-Fernandez E, et al. Estimating the causal influence of body mass index on risk of Parkinson disease: a mendelian randomisation study. PLoS Med. 2017;14:e1002314. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pmed.1002314.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Kamat MA, Blackshaw JA, Young R, Surendran P, Burgess S, Danesh J, et al. PhenoScanner V2: an expanded tool for searching human genotype-phenotype associations. Bioinformatics. 2019;35:4851–3. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/btz469.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  25. Burgess S, Thompson SG. Avoiding bias from weak instruments in mendelian randomization studies. Int J Epidemiol. 2011;40:755–64. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ije/dyr036.

    Article  PubMed  Google Scholar 

  26. Burgess S, Bowden J, Fall T, Ingelsson E, Thompson SG. Sensitivity analyses for robust causal inference from mendelian randomization analyses with multiple genetic variants. Epidemiology. 2017;28:30–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/EDE.0000000000000559.

    Article  PubMed  Google Scholar 

  27. 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. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/gepi.21965.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44:512–25. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ije/dyv080.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Greco MFD, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in mendelian randomisation studies with summary data and a continuous outcome. Stat Med. 2015;34:2926–40. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/sim.6522.

    Article  Google Scholar 

  30. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from mendelian randomization between complex traits and diseases. Nat Genet. 2018;50:693–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41588-018-0099-7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  31. Shores MM, Walsh TJ, Korpak A, Krakauer C, Forsberg CW, Fox AE, et al. Association between Testosterone Treatment and Risk of Incident Cardiovascular events among US male veterans with low testosterone levels and multiple medical comorbidities. J Am Heart Assoc. 2021;10:e020562. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/JAHA.120.020562.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  32. Gibson CL, Attwood L. The impact of gender on stroke pathology and treatment. Neurosci Biobehav Rev. 2016;67:119–24. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.neubiorev.2015.08.020.

    Article  PubMed  Google Scholar 

  33. Herson PS, Palmateer J, Hurn PD. Biological sex and mechanisms of ischemic brain injury. Transl Stroke Res. 2013;4:413–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s12975-012-0238-x.

    Article  PubMed  Google Scholar 

  34. Alonso de Leciñana M, Egido JA, Fernández C, Martínez-Vila E, Santos S, Morales A, et al. Risk of ischemic stroke and lifetime estrogen exposure. Neurology. 2017;68:33–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1212/01.wnl.0000250238.69938.f5.

    Article  CAS  Google Scholar 

  35. Hu J, Lin JH, Jiménez MC, Manson JE, Hankinson SE, Rexrode KM. Plasma estradiol and testosterone levels and ischemic stroke in Postmenopausal Women. Stroke. 2020;51:1297–300. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/STROKEAHA.119.028588.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  36. Abbott RD, Launer LJ, Rodriguez BL, Ross GW, Wilson PWF, Masaki KH, et al. Serum estradiol and risk of stroke in elderly men. Neurology. 2007;68:563–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1212/01.wnl.0000254473.88647.ca.

    Article  PubMed  CAS  Google Scholar 

  37. Belladelli F, Del Giudice F, Kasman A, Salonia A, Eisenberg ML. The association between testosterone, estradiol and their ratio and mortality among US men. Andrologia. 2021;53:e13993. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/and.13993.

    Article  PubMed  CAS  Google Scholar 

  38. Han YY, Forno E, Witchel SF, Manni ML, Acosta-Pérez E, Canino G, et al. Testosterone-to-estradiol ratio and lung function in a prospective study of Puerto Rican youth. Ann Allergy Asthma Immunol. 2021;127:236–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.anai.2021.04.013.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  39. Choi JW, Ryoo IW, Hong JY, Lee KY, Nam HS, Kim WC, et al. Clinical impact of estradiol/testosterone ratio in patients with acute ischemic stroke. BMC Neurol. 2021;21:91. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12883-021-02116-9.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  40. Koellhoffer EC, McCullough LD. The effects of estrogen in ischemic stroke. Transl Stroke Res. 2013;4:390–401. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s12975-012-0230-5.

    Article  PubMed  CAS  Google Scholar 

  41. Ghisletti S, Meda C, Maggi A, Vegeto E. 17beta-estradiol inhibits inflammatory gene expression by controlling NF-kappaB intracellular localization. Mol Cell Biol. 2005;25:2957–68. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/MCB.25.8.2957-2968.2005.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  42. Gagliano-Jucá T, Basaria S. Testosterone replacement therapy and cardiovascular risk. Nat Rev Cardiol. 2019;16:555–74. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41569-019-0211-4.

    Article  PubMed  Google Scholar 

  43. Jones TH, Kelly DM. Randomized controlled trials - mechanistic studies of testosterone and the cardiovascular system. Asian J Androl. 2018;20:120–30. https://doiorg.publicaciones.saludcastillayleon.es/10.4103/aja.aja_6_18.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  44. Mukherjee TK, Dinh H, Chaudhuri G, Nathan L. Testosterone attenuates expression of vascular cell adhesion molecule-1 by conversion to estradiol by aromatase in endothelial cells: implications in atherosclerosis. Proc Natl Acad Sci U S A. 2002;99:4055–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1073/pnas.052703199.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  45. Usselman CW, Yarovinsky TO, Steele FE, Leone CA, Taylor HS, Bender JR, et al. Androgens drive microvascular endothelial dysfunction in women with polycystic ovary syndrome: role of the endothelin B receptor. J Physiol. 2019;597:2853–65. https://doiorg.publicaciones.saludcastillayleon.es/10.1113/JP277756.

    Article  PubMed  CAS  Google Scholar 

  46. Aribas E, Ahmadizar F, Mutlu U, Ikram MK, Bos D, Laven JSE, et al. Sex steroids and markers of micro- and macrovascular damage among women and men from the general population. Eur J Prev Cardiol. 2022;29:1322–30. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/eurjpc/zwaa031.

    Article  PubMed  CAS  Google Scholar 

Download references

Acknowledgements

We acknowledge the investigators and participants of the original GWAS. We are grateful for all GWAS sharing summary data used in this study.

Funding

This study was funded by National Natural Science Foundation of China (82027802, 82102220), Research Funding on Translational Medicine from Beijing Municipal Science and Technology Commission (Z221100007422023), Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support (YGLX202325), Non-profit Central Research Institute Fund of Chinese Academy of Medical (2023-JKCS-09), Beijing Association for Science and Technology Youth Talent Support Program (BYESS2022081), Beijing Municipal Natural Science Foundation (7244510), Science and Technology Innovation Service Capacity Building Project of Beijing Municipal Education Commission (11000023T000002157177), Outstanding Young Talents Program of Capital Medical University (B2305).

Author information

Authors and Affiliations

  1. Department of Neurology, Xuanwu Hospital, Capital Medical University, Beijing, China

    Yang Zhang

  2. Xuanwu Hospital, China-America Institute of Neurology, Capital Medical University, No. 45, Changchun Street, Xicheng District, Beijing, China

    Yang Zhang, Di Wu, Ming Li & Xunming Ji

  3. Beijing Institute of Brain Disorders, Capital Medical University, Beijing, China

    Miaowen Jiang & Xunming Ji

  4. Department of Neurosurgery, Xuanwu Hospital, Capital Medical University, No. 45, Changchun Street, Xicheng District, Beijing, China

    Xunming Ji

Authors
  1. Yang Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Miaowen Jiang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Di Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Ming Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Xunming Ji
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Yang Zhang: Conceptualization (equal); formal analysis (equal); investigation (equal); methodology (equal); software (equal); writing – original draft (equal). Miaowen Jiang: Software (equal); Di Wu: Writing – review and editing (equal); Li Ming: Methodology (equal); writing – review and editing (equal); Xunming Ji: Conceptualization (equal); project administration (lead); writing – review and editing (lead).

Corresponding authors

Correspondence to Ming Li or Xunming Ji.

Ethics declarations

Ethics approval and consent to participate

Only summary data were used in this article. Appropriate ethical approval and patient informed consent were obtained in the original studies.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note

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

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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 changes were made. 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/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, Y., Jiang, M., Wu, D. et al. The causal relationship between steroid hormones and risk of stroke: evidence from a two-sample Mendelian randomization study. Mol Brain 18, 6 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13041-025-01173-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13041-025-01173-2

Keywords

Molecular Brain

ISSN: 1756-6606