Osteoporosis and Stroke: A Bidirectional Mendelian Randomization Study

doi:10.21203/rs.3.rs-4855698/v1

Download PDF

Research Article

Osteoporosis and Stroke: A Bidirectional Mendelian Randomization Study

https://doi.org/10.21203/rs.3.rs-4855698/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objective: Numerous observational studies have identified a link between osteoporosis and stroke. However, the causal genetic relationship between these conditions remains unclear. This study employs a two-sample bidirectional Mendelian randomization (MR) approach to ascertain the causal relationship between osteoporosis and stroke.

Design: We conducted a two-sample Mendelian randomization (MR) study to investigate the potential causal relationship between osteoporosis and stroke, including its subtypes. Genetic data for osteoporosis and stroke, along with their subtypes, were sourced from published genome-wide association studies (GWAS). Single nucleotide polymorphisms (SNPs) demonstrating genome-wide significance (p < 5×10^−8) and independence (r^2 < 0.001) were selected for further analysis, provided they had an F-statistic ≥10. The inverse-variance weighted (IVW) method was employed to evaluate causality, with results reported as odds ratios (ORs). Heterogeneity was assessed using Cochran’s Q test, while pleiotropy was tested using the MR-Egger intercept test. A leave-one-out sensitivity analysis was performed to ensure the robustness of the results.

Results: Employing the IVW method, MR Egger method, and median-weighted method, we found no significant bidirectional causal relationship between osteoporosis and stroke or its subtypes, irrespective of the inclusion of potential pleiotropic SNPs. Sensitivity analyses affirmed the reliability and stability of these findings.

Conclusion: This study indicates that there is no bidirectional causal relationship between osteoporosis and stroke or its subtypes. Nevertheless, these conditions share similar pathophysiological mechanisms and pathways.

osteoporosis

stroke

genome-wide association study

Mendelian randomization

Osteoporosis and stroke are two diseases of significant public health concern worldwide [1, 2]. Each year, approximately 15 million people globally are affected by stroke, which is not only the leading cause of disability among adults but also the fourth leading cause of death [3, 4]. With the aging population, the incidence and associated economic burden of stroke are expected to increase further [5]. Stroke has a high prevalence with notable gender differences; the incidence in men is nearly twice that in women, but women often have worse outcomes [6]. Disabilities caused by stroke may prevent survivors from working normally, leading to severe economic and social problems.[7].

Osteoporosis (OP) is a chronic metabolic bone disease most commonly diagnosed by dual-energy X-ray absorptiometry (DXA), characterized by reduced bone mineral density (BMD) [8, 9], leading to an increased risk of fractures [10]. Osteoporosis is often a major complication following a stroke, with both conditions being highly disabling [11]. Post-stroke osteoporosis typically occurs on the paralyzed side, more commonly affecting the upper limbs rather than the lower limbs. Compared to age-related osteoporosis, metabolic diseases, nutritional issues, and medication-related factors, post-stroke osteoporosis is more prevalent [12]. Osteoporosis can lead to fragile bones, resulting in higher mortality or morbidity and reduced quality of life [13].

Increasing evidence suggests that stroke may negatively impact bone density, increasing the risk of osteoporosis and subsequent major fractures, thereby exacerbating patient morbidity and mortality [14, 15]. This phenomenon may be related to post-stroke inactivity, neurological dysfunction, and endocrine disorders [16]. Additionally, previous studies have shown that bone mineral density(BMD) may be associated with stroke incidence, with patients having lower BMD being at higher risk of stroke [17, 18]. Thus, osteoporosis is believed to play a significant role in the development of stroke. Stroke and osteoporosis share some common risk factors, such as age and lack of physical activity. Inflammatory cytokines are also thought to be related to both conditions [19]. Despite current evidence indicating a complex interplay between stroke and osteoporosis, understanding the relationship between stroke and BMD is crucial for optimizing post-stroke patient management. Moreover, as life expectancy after a stroke increases, bone health issues become increasingly important, especially for individuals at risk of osteoporosis or osteoporotic fractures.

Clarifying the causal relationship between osteoporosis and stroke is essential for prevention and treatment, but it remains uncertain whether such a causal relationship exists. Clinical observational studies often cannot provide a convincing answer regarding the causality between osteoporosis and stroke due to various confounding factors. Mendelian randomization (MR) is a method used to assess the causal relationship between exposure and outcome by using genetic variables (IVs) that are equally, randomly, and independently distributed [20]. The allocation of genotypes is not influenced by age, sex, lifestyle, or environmental factors [21, 22]. Compared to traditional clinical randomized controlled trials, the main advantage of MR is the avoidance of potential confounders [23]. Therefore, we used a two-sample MR design to assess the causal relationship between osteoporosis (measured by BMD) and stroke.

By utilizing summary data from GWASs, we were the first to use the MR method to investigate whether there is a bidirectional causal relationship between stroke and osteoporosis. Based on the two-sample bidirectional MR framework, we explored the potential bidirectional causal effects of osteoporosis on stroke, ischemic stroke (IS), large vessel ischemic stroke (LV-IS), small vessel ischemic stroke (SV-IS), cardioembolic ischemic stroke (CE-IS), and intracerebral hemorrhage (ICH).

Table 1

Data sources used in this study
Exposures or outcome	Sample size (total or cases/controls)	Ancestry	Consortia	PubMed ID or URL of original research	PubMed ID or URL
Total body bone mineral density	56,284	European	GWAS meta-analysis	29304378	https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST005348/
stroke	446,696	European	GWAS meta-analysis	29531354	https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST005838/
Ischemic stroke	440,328	European	GWAS meta-analysis	29531354	https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST006908/
Large vessel ischemic stroke	150,765	European	GWAS meta-analysis	29531354	https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST005840/
Small vessel ischemic stroke	198,048	European	GWAS meta-analysis	29531354	https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST005841/
Cardioembolic ischemic stroke	211,763	European	GWAS meta-analysis	29531354	https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST005842/
Intracerebral hemorrhage	473,513	European	GWAS meta-analysis	34594039	https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90018870/

Study design and data sources

In this two-sample MR study, we used single nucleotide polymorphisms (SNPs) as instrumental variables (IVs) and GWAS data to determine the causal relationship between osteoporosis and stroke. The study design and assumptions for the MR study are outlined in Fig. 1. The genetic data related to osteoporosis and stroke were obtained from published GWAS studies, with detailed data presented in Table 1.

Genetic instrument selection

First, we applied a genome-wide significance criterion with a threshold of P < 5E-8 to mitigate the impact of weak instrument bias, and used a minor allele frequency (MAF) > 0.01 to identify genetic instruments for osteoporosis and stroke. Second, to address significant linkage disequilibrium (LD), we conducted a clumping procedure with an R2 < 0.001 and a window size of 10,000 kb using data from the European ancestry-based 1000 Genomes Project[24]. In cases where SNP pairs exhibited a high LD R2 value, we retained the SNP with the lower P-value. Third, when the targeted SNPs were absent in the outcome genome-wide association study (GWAS), we identified proxy SNPs with high LD levels (R2 > 0.8) relative to the target SNPs. Finally, to generate a harmonized summary set where each SNP in the exposure and outcome corresponded to the same effect allele, we excluded SNPs with discordant alleles and palindromic SNPs[25]. These meticulously selected SNPs served as the final genetic instrumental variables for our subsequent Mendelian Randomization (MR) analysis.

Furthermore, to assess the reliability and validity of each SNP as a genetic instrumental variable (IV) in our Mendelian Randomization (MR) analysis, we calculated the F-statistic for each SNP using the equation F = R² (N − 2) / (1 - R²)[26]. Here, R² represents the proportion of variance in the exposure variable explained by the IV, and N is the sample size of the original genome-wide association study (GWAS). To compute R² for each IV, we used the formula R² = (2 x EAF (1 - EAF) x beta²) / [(2 x EAF (1 - EAF) x beta²) + (2 x EAF (1 - EAF) x N x (SE (beta)²))], where EAF denotes the effect allele frequency, beta represents the estimated genetic effect on the outcome, N is the sample size of the GWAS, and SE is the standard error of the genetic effect[27]. IVs with F-statistics less than 10 were considered unreliable and were excluded from the subsequent MR analysis.

Statistical analysis

We first conducted MR analysis to assess the causal effect of BMD on stroke and then performed the analysis in the opposite direction to determine the causal directionality between the two traits.

For the primary MR analysis, we used the inverse-variance weighted (IVW) technique with random effects, which provides more accurate estimates in the presence of heterogeneity [24]. IVW can yield unbiased estimates as long as all genetic variants are valid instruments [24]. To enhance the reliability of our findings, we also applied two additional MR methods: MR-Egger and median-based estimators, which include the weighted median and weighted mode [25]. The MR-Egger method is less prone to directional pleiotropy and allows for the violation of instrumental variable assumptions by all genetic variants. However, MR-Egger has low statistical power [26]. In contrast, median-based estimators are less sensitive to outliers and can provide valid estimates even if up to 50% of the genetic variants are invalid [27]. We used MR-Egger regression and the IVW method to test for heterogeneity among the selected SNPs and employed Cochrane's Q statistic to assess the impact of heterogeneity. The MR-PRESSO test was used to detect and correct any horizontal pleiotropic outliers, enhancing the accuracy of causal estimates [28]. Additionally, we used MR-Egger regression to test for potential horizontal pleiotropy and conducted leave-one-out sensitivity analysis to evaluate the validity and stability of the MR results.

Then, the SNPs (P < 1×10^− 6) in the GWAS Catalog website (https://www.ebi.ac.uk/gwas) were eliminated due to potential confounders and outcomes, such as coronary artery disease, hypertension, hyperlipemia, wine intake, weight, type 2 diabetes mellitus, and psychiatric diseases (Supplementary Tables 1, 2). We also estimated F-statistics for each SNP, as well as R² for the exposure variance explained by each IV. IVs with F-statistics less than 10 were considered weak instruments and were omitted from MR analysis.

Statistical significance

All statistical analyses were performed using the TwoSampleMR package in R version 4.4.0 (R Foundation for Statistical Computing). To account for multiple testing, we considered associations with a p-value below 0.005 (where P = 0.05/10) as strong evidence of causal association, while associations with a p-value below 0.05 but greater than 0.005 were considered suggestive evidence in MR analysis.

Effect of Bone Mineral Density on Stroke

In our study, we used 82 independent SNPs as instrumental variables (IVs) for bone mineral density (BMD) (Table 2, Fig. 2). The MR analysis results indicated no significant relationship between genetically predicted BMD and overall stroke or its subtypes. The primary IVW results showed no significant association between BMD and stroke [IVW odds ratio (OR): 1.005, 95% confidence interval (CI): 0.963 ~ 1.050, P = 0.81], BMD and IS (OR: 1.031, 95% CI: 0.978 ~ 1.088, P = 0.26), LV-IS (OR: 1.032, 95% CI: 0.927 ~ 1.150, P = 0.57), SV-IS (OR: 0.994, 95% CI: 0.905 ~ 1.092, P = 0.90), CE-IS (OR: 1.032, 95% CI: 0.939 ~ 1.133, P = 0.52), and ICH (OR: 0.922, 95% CI: 0.808 ~ 1.053, P = 0.23). Similar results were observed with MR-Egger regression and median-based estimators (weighted median and weighted mode) (Fig. 2). The horizontal pleiotropy test results indicated no directional pleiotropy (Fig. 2). The weighted median analysis results showed no causal effect of reduced BMD on different stroke subtypes (Fig. 2).

In Cochran’s Q heterogeneity test, heterogeneity was detected in the BMD-overall stroke association (IVW = 101.04, P = 0.05; MR-Egger = 100.35, P = 0.05, Fig. 2, Table 2). Therefore, we conducted an MR-PRESSO test to further examine this relationship. The MR-PRESSO results indicated no significant outliers (Global Test P value = 0.0606). No outliers were identified, so the results for the outlier-corrected MR are set to NA. The leave-one-out sensitivity analysis results showed that no single SNP had a potential impact on the final results. Supplementary Fig. 1 describes the scatter plots, funnel plots, and leave-one-out analysis plots of BMD on overall stroke and stroke subtypes.

To further strengthen our MR assumption, we examined the traits related to our instrumental SNPs. Traits association analysis (see Supplementary Table 1) showed that rs2252865,rs2873195 in TB-BMD, are associated with psychiatric diseases and hyperlipemia, which may have some effect on the risk of overall stroke or its subtypes. Sensitivity analysis by removing the SNPs revealed similar results, though with lower statistical power (see Supplementary Fig. 3).

Effect of Stroke on Bone Mineral Density

We obtained 17, 9, 4, 1, 4, and 2 SNPs from GWAS to serve as IVs for overall stroke, IS, LV-IS, SV-IS, CE-IS, and ICH, respectively (see Fig. 3, Table 3). The MR analysis results indicated no significant relationship between genetically predicted overall stroke and its subtypes with bone mineral density (BMD). The primary IVW results showed no significant association between stroke and BMD (OR: 1.027, 95% CI: 0.947 ~ 1.114, P = 0.52), IS and BMD (OR: 0.953, 95% CI: 0.874 ~ 1.039, P = 0.27), LV-IS and BMD (OR: 0.988, 95% CI: 0.930 ~ 1.049, P = 0.68), SV-IS and reduced (OR: 1.013, 95% CI: 0.915 ~ 1.121, P = 0.81), CE-IS and BMD (OR: 0.969, 95% CI: 0.933 ~ 1.006, P = 0.10), and ICH and BMD (OR: 0.954, 95% CI: 0.847 ~ 1.074, P = 0.43). Similar results were observed with MR-Egger regression and median-based estimators (weighted median and weighted mode) (Fig. 3, Table 2). All Egger regression tests were negative (p > 0.05; Fig. 3), indicating that our MR results were not influenced by horizontal pleiotropy.

In the Cochran’s Q heterogeneity test, heterogeneity was detected in overall stroke, IS, and ICH. For stroke (IVW = 37.16, P = 1.99E-03; MR-Egger = 37.16, P = 1.20E-03), IS (IVW = 17.81, P = 0.02; MR-Egger = 15.95, P = 0.03), and ICH (IVW = 5.89, P = 0.02) (Fig. 3, Table 2). Therefore, we conducted an MR-PRESSO test to further examine these relationships. For overall stroke, SNPs rs3184504 and rs42039 were excluded, and for IS, SNPs rs3184504 and rs4942561 were excluded. After removing the outlier variants, we still found no causal relationship between stroke and reduced BMD (see Fig. 3, Table 2). The leave-one-out sensitivity analysis results showed that no single SNP had a potential impact on the final results. Supplementary Fig. 2 describes the scatter plots, funnel plots, and leave-one-out analysis plots of stroke and its subtypes on BMD.

To further strengthen our MR assumption, we examined the traits related to our instrumental SNPs. The results of the trait association analysis (see Supplementary Table 2) indicated that no SNPs ,which may have some impact on OP.

Table 2

MR Heterogeneity and Pleiotropy Analysis of the Relationship Between Bone Mineral Density and Stroke
Exposure	Outcome	No.of Ivs	Heterogeneity tests		Directional horizontal pleiotropy test
			Methods	Cochran’sQ (P)	MR-Egger intercept (P)
TB-BMD	Stroke	81	MR Egger	100.35(0.05)	-2.44E-03(0.47)
			IVW	101.04(0.05)
TB-BMD	IS	81	MR Egger	96.54(0.09)	-3.02E-03(0.45)
			IVW	97.26(0.09)
TB-BMD	LV-IS	82	MR Egger	94.16(0.13)	-4.78E-03(0.56)
			IVW	94.56(0.14)
TB-BMD	SV-IS	73	MR Egger	84.64(0.13)	-4.55E-03(0.54)
			IVW	85.10(0.14)
TB-BMD	CE-IS	82	MR Egger	100.56(0.06)	-3.29E-03(0.64)
			IVW	100.84(0.07)
TB-BMD	ICH	82	MR Egger	90.56(0.20)	5.99E-03(0.55)
			IVW	90.96(0.21)
Stroke	TB-BMD	17	MR Egger	37.16(1.20E-03)	4.64E-04(0.98)
			IVW	37.16(1.99E-03)
IS	TB-BMD	9	MR Egger	15.95(0.03)	2,16E-02(0.40)
			IVW	17.81(0.02)
LV-IS	TB-BMD	4	MR Egger	2.61(0.27)	4.53E-02(0.26)
			IVW	5.81(0.12)
SV-IS	TB-BMD	1	MR Egger	NA	NA
			IVW	NA
CE-IS	TB-BMD	4	MR Egger	3.07(0.22)	-5.21E-04(0.96)
			IVW	3.07(0.38)
ICH	TB-BMD	2	IVW	5.89(0.02)	NA

To our knowledge, this is the first two-sample MR study to analyze the bidirectional causal relationship between osteoporosis and stroke. In this study, we used bidirectional MR to test the causal relationship between osteoporosis and the risk of stroke, IS, LV-IS, SV-IS, CE-IS, and ICH. Using the largest publicly available GWAS summary data, we found no bidirectional causal relationship between genetically predicted BMD and the increased risk of stroke and its subtypes. Sensitivity analyses supported the reliability of our results.

There is a complex interaction between osteoporosis and stroke, with previous studies indicating that patients with osteoporosis have a higher risk of stroke [29]. The close relationship between osteoporosis and stroke raises the question of whether osteoporosis is a marker rather than a risk factor for stroke. However, our results suggest that osteoporosis does not have a direct causal relationship with stroke but rather serves as a marker. They share common risk factors, pathophysiological pathways, and genetic factors.

Systemic inflammation is a common risk factor for both osteoporosis and arterial diseases. Osteoporosis is associated with systemic inflammation, with elevated levels of inflammatory markers such as C-reactive protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α) [30,31]. This chronic inflammatory state can promote the differentiation and maturation of osteoclasts, leading to accelerated bone resorption and osteoporosis. On the other hand, current evidence suggests that inflammation may play a crucial role in the development of stroke [32,33]. Inflammation is also important in the development and progression of peripheral artery disease . Elevated levels of circulating CRP, IL-6, and TNF-α are associated with a high risk of peripheral artery disease [34,35]. Thus, dysregulation of inflammatory signaling pathways may represent a common mechanism between osteoporosis and stroke.

The connection between osteoporosis and stroke is also linked through vitamin D. Studies have shown that vitamin D receptors are expressed in cardiomyocytes, vascular smooth muscle cells, and endothelial cells [36]. Vitamin D can reduce oxidative stress in endothelial cells and inhibit the formation of atherosclerotic plaques [37,38]. A meta-analysis of limited existing data indicated that moderate to high doses of vitamin D supplementation could reduce cardiovascular risk. Epidemiological studies have suggested that vitamin D deficiency is associated with an increased risk of peripheral artery disease [39]. The link between osteoporosis and stroke may be mediated through the common role of vitamin D.

BMD is a measure of lifelong estrogen exposure. Bone degeneration results from age-related changes in other organs, particularly due to the decline in ovarian function in postmenopausal women. The decrease in estrogen levels after menopause is associated with other diseases, such as cardiovascular disease. BMD can be influenced by estrogen levels and physical activity, particularly in postmenopausal women, and is directly related to arterial calcification [40,41]. The observed age and gender differences in the relationship between BMD and stroke risk may stem from differences in endogenous estrogen levels, which can influence cardiovascular risk [42]. Menopause increases the risk of osteoporosis and cardiovascular diseases, such as atherosclerosis [43,44]. Thus, decreased estrogen levels may simultaneously contribute to both osteoporosis and stroke.

Lack of physical activity is also considered a risk factor for peripheral artery disease [45]. This may be because insufficient physical activity is associated with increased oxidative stress and endothelial dysfunction [46]. At the same time, a lack of physical activity accelerates bone resorption, increasing the risk of osteoporosis [47]. Therefore, it is a common risk factor for both osteoporosis and peripheral artery disease, which could explain the link between the two.

In the opposite direction, the relationship between stroke and osteoporosis is also of interest in the medical field, with various hypotheses proposed to explain their connection. Although we did not find a genetic causal relationship between stroke and osteoporosis, previous observational studies have shown a strong correlation between stroke and osteoporosis [15]. A cohort study found that patients with hemorrhagic stroke had a 2.06 times higher risk of osteoporosis (95% CI 1.83-2.33), and patients with ischemic stroke had a 1.77 times higher risk of osteoporosis (95% CI 1.65-1.89) compared to non-stroke cohorts [48]. Based on our MR study results, which found no genetic causal relationship between stroke and osteoporosis, we believe that the clinical association observed is likely due to secondary effects of stroke rather than an intrinsic direct causal relationship. Many underlying mechanisms lead to bone loss after a stroke, mainly including decreased mobility, vitamin D deficiency, and increased inflammation.

Reduced mobility is one of the main reasons for osteoporosis after a stroke. Evidence over the past few decades suggests that stroke survivors experience a significant decrease in BMD, especially in paralyzed limbs [49,50]. The difference in BMD between paralyzed and non-paralyzed limbs after a stroke can be attributed to various factors, primarily reduced mobility and mechanical loading[51]. Immobilization and muscle weakness promote bone resorption and hinder bone formation, leading to rapid bone loss [52]. Notably, some studies have shown that BMD actually increases on the non-paralyzed side, possibly due to compensatory physical activity on the unaffected side [53]. Yamada et al. reported an increase in lumbar spine BMD in 100 immobilized stroke patients [54]. A meta-analysis showed reduced BMD in the femoral neck of the non-paralyzed side, but no reduction in distal radius BMD of the non-paralyzed side [55]. The independence of upper limb activity might mitigate bone loss through compensatory use of the non-paralyzed side in daily tasks [56,57]. These studies support our findings that there is no direct, intrinsic causal relationship between stroke and osteoporosis. Stroke affects osteoporosis through secondary effects.

Another major factor affecting bone health after a stroke is calcium homeostasis, particularly vitamin D deficiency. Older individuals are less exposed to sunlight, and their skin has a reduced ability to produce vitamin D precursors [58]. Stroke patients are at higher risk because they are more likely to stay indoors, thus receiving less sunlight exposure and having poorer dietary intake [59]. Poole et al. found that most acute stroke patients had vitamin D deficiency throughout the year [60]. This also results from secondary effects post-stroke rather than a direct impact of stroke on osteoporosis.

However, there are some limitations of our study.Selecting SNPs from different large-sample GWAS studies may increase the risk of sample overlap between exposure and outcome variables, potentially biasing the results. Additionally, due to a lack of suitable SNPs for ICH, we were unable to explore the potential causal effect of ICH on osteoporosis. Finally, the MR study population was of European descent. Whether these results can be replicated in Asian populations remains to be explored.

In conclusion, our study found no causal relationship between osteoporosis and stroke in any direction. Based on our findings, although there is no causal relationship, they share similar pathophysiological mechanisms and pathways.

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found at: All GWAS summary statistics can be downloaded from open GWAS for exposures (https://gwas.mrcieu.ac.uk/), GWAS catalog (https://www.ebi.ac.uk/gwas/ and https://data.bris.ac.uk/).

Author contributions

Miao He, Haochuan Yong and Jie Li designed the study. Miao He, Haochuan Yong and Jie Li performed statistical analyses. All authors interpreted the results. Miao He, Haochuan Yong and Jie Li drafted the manuscript. All authors edited the manuscript for intellectual content.All authors read and approved the final manuscript.

Conflict of interest

The authors declare no conflict of interest

Funding

No specific funding was received from any funding bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript.

Acknowledgements

We would like to sincerely thank the original GWASs and the related consortiums for sharing and managing the summary statistics.

Hsu W-L, Chen C-Y, Tsauo J-Y, et al. Balance control in elderly people with osteoporosis. J Formos Med Assoc 2014;113:334–9.PMID: 24650494
Fisher A, Srikusalanukul W, Davis M, et al. Poststroke hip fracture: prevalence, clinical characteristics, mineral-bone metabolism, outcomes, and gaps in prevention. Stroke Res Treat 2013;2013:1-17.PMID: 24187647 PMCID: PMC3800649 DOI: 10.1155/2013/641943
Arnao V, Acciarresi M, Cittadini E, et al. Stroke incidence, prevalence and mortality in women worldwide. Int J Stroke 2016;11:287–301.PMID: 26984193 DOI:10.1177/1747493016632245
Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics-2011 update: a report from the American Heart Association. Circulation 2011;123:e18-209.PMID: 21160056 PMCID: PMC4418670 DOI: 10.1161/CIR.0b013e3182009701
Liu C-T, Wu B-Y, Hu W-L, et al. Gender-based differences in mortality and complementary therapies for patients with stroke in Taiwan. Complement Ther Med 2017;30:113–7. 7.PMID: 28137521 DOI: 10.1016/j.ctim.2016.12.004
Hsieh F-I, Chiou H-Y. Stroke: morbidity, risk factors, and care in Taiwan. J Stroke 2014;16:59–64.PMID: 24949310 PMCID: PMC4060269 DOI: 10.5853/jos.2014.16.2.59
Nicoletti A, Sofia V, Giuffrida S, et al. Prevalence of stroke : a door-to-door survey in rural Bolivia. Stroke 2000;31:882–5.PMID: 10753992 DOI: 10.1161/01.str.31.4.882
Anam AK, Insogna K. Update on osteoporosis screening and management. Med Clin North Am (2021) 105:1117–34. doi: 10.1016/j.mcna.2021.05.016
Cummings SR, Bates D, Black DM. Clinical use of bone densitometry: scientific review. JAMA. 2002; 288:1889–97.https://doi.org/10.1001/jama.288.15.1889 PMID:12377088
Lippuner K. The future of osteoporosis treatment - a research update. Swiss Med Wkly 2012;142:w13624.PMID: 22815185 DOI: 10.4414/smw.2012.13624
Carda S, Cisari C, Invernizzi M, et al. Osteoporosis after stroke: a review of the causes and potential treatments. Cerebrovasc Dis 2009;28:191–200.PMID: 19571530 DOI: 10.1159/000226578
Myint PK, Poole KE, Warburton EA. Hip fractures after stroke and their prevention. QJM. 2007;100:539–545.PMID: 17693418 DOI: 10.1093/qjmed/hcm067
Pasco JA, Henry MJ, Korn S, et al. Morphometric vertebral fractures of the lower thoracic and lumbar spine, physical function and quality of life in men. Osteoporos Int 2009;20:787–92.PMID: 18802658 DOI: 10.1007/s00198-008-0744-6
Borschmann K, Pang MY, Bernhardt J, Iuliano-Burns S. Stepping towards prevention of bone loss after stroke: a systematic review of the skeletal effects of physical activity after stroke. Int J Stroke 2012;7(4):330–5, doi:http://dx.doi.org/10.1111/j.1747- 4949.2011.00645.x.PMID: 21967614 DOI: 10.1111/j.1747-4949.2011.00645.x
Hjelholt TJ, Johnsen SP, Brynningsen PK, Andersen G, Pedersen AB. Impact of stroke history on the risk of recurrent hip fracture or major osteoporotic fractures among patients with incident hip fracture: a nationwide cohort study. J Bone Miner Res 2023;38(2):278–87, doi:http://dx.doi.org/10.1002/jbmr.4760.
Mijajlovic MD, Aleksic V, Stojanovski N, Bornstein NM. Relationship between bone disorders and stroke. Neurol Sci. 2020 Dec;41(12):3579-3587. doi: 10.1007/s10072-020-04748-0. Epub 2020 Oct 2. PMID: 33006058 Review.
Browner WS, Pressman AR, Nevitt MC, Cauley JA, Cummings SR. Association between low bone density and stroke in elderly women. The study of osteoporotic fractures. Stroke 1993;24:940–6.PMID: 8322393 DOI: 10.1161/01.str.24.7.940
Jorgensen L, Engstad T, Jacobsen BK. Bone mineral density in acute stroke patients: low bone mineral density may predict first stroke in women. Stroke 2001;32:47–51.PMID: 11136913 DOI: 10.1161/01.str.32.1.47
Wiseman S, Marlborough F, Doubal F, Webb DJ, Wardlaw J. Blood markers of coagulation, fibrinolysis, endothelial dysfunction and inflammation in lacunar stroke versus non-lacunar stroke and non-stroke: systematic review and meta-analysis. Cerebrovasc Dis 2014;37:64–75.PMID: 24401164 DOI: 10.1159/000356789
Emdin, C. A., Khera, A. V., and Kathiresan, S. (2017). Mendelian Randomization. JAMA 318, 1925–1926. doi: 10.1001/jama.2017.17219
Hartwig, F. P., Borges, M. C., Horta, B. L., Bowden, J., and Davey Smith, G. (2017). Inflammatory biomarkers and risk of schizophrenia: a 2-sample Mendelian randomization study. JAMA Psychiat. 74, 1226–1233. doi: 10.1001/jamapsychiatry.2017.3191
Goto, A., Yamaji, T., SawRAa, N., Momozawa, Y., Kamatani, Y., Kubo, M., et al. (2020). Diabetes and cancer risk: a Mendelian randomization study. Int. J. Cancer 146, 712–719. doi: 10.1002/ijc.32310
Scheltens, P., De Strooper, B., KIVsipelto, M., Holstege, H., Chételat, G., Teunissen, C. E., et al. (2021). Alzheimer’s diseRAe. Lancet Lond. Engl 397, 1577–1590. doi: 10.1016/ S0140-6736(20)32205-4
Hemani G, Zheng J, Elsworth B, WRAe KH, Haberland V, Baird D, et al. The MR-bRAe platform supports systematic causal inference across the human phenome. eLife (2018) 7:e34408. doi: 10.7554/elife.34408
Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and biRA detection through egger regression. Int J Epidemiol (2015) 44:512–25. doi: 10.1093/ije/dyv080
Burgess S, Thompson SG. Interpreting findings from mendelian randomization using the MR-egger method. Eur J Epidemiol (2017) 32:377–89. doi: 10.1007/s10654- 017-0255-x
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. doi: 10.1002/gepi.21965
Verbanck M, Chen C-Y, Neale B, Do R. Detection of widespreRA horizontal pleiotropy in causal relationships inferred from mendelian randomization between complex traits and diseRAes. Nat Genet (2018) 50:693–8. doi: 10.1038/s41588-018-0099-7
Nordström A, Eriksson M, Stegmayr B, Gustafson Y, Nordström P. Low bone mineral density is an independent risk factor for stroke and death. Cerebrovasc Dis 2010;29 (2):130–6, doi:http://dx.doi.org/10.1159/000262308.
de Pablo P, Cooper MS, Buckley CD (2012) Association between bone mineral density and C-reactive protein in a large populationbased sample. Arthritis Rheum 64:2624–2631. https://doi.org/10. 1002/art.34474
Scheidt-Nave C, Bismar H, Leidig-Bruckner G, Woitge H, Seibel MJ, Ziegler R, Pfeilschifter J (2001) Serum interleukin 6 is a major predictor of bone loss in women specifc to the frst decade past menopause. J Clin Endocrinol Metab 86:2032–2042. https:// doi.org/10.1210/jcem.86.5.744516.
Zha L, He L, Liang Y, Qin H, Yu B, Chang L, Xue L (2018) TNF-α contributes to postmenopausal osteoporosis by synergistically promoting RANKL-induced osteoclast formation. Biomedicine & Pharmacotherapy 102:369–374. https://doi.org/10.1016/j. biopha.2018.03.080
Wang QM, Liao JK. ROCKs as immunomodulators of stroke. Expert Opin Ther Targets 2012;16:1013–25.PMID: 22925075 PMCID: PMC3807092 DOI: 10.1517/14728222.2012.715149
Tzoulaki I, Murray GD, Lee AJ, Rumley A, Lowe GD, Fowkes FG (2005) C-reactive protein, interleukin-6, and soluble adhesion molecules as predictors of progressive peripheral atherosclerosis in the general population: Edinburgh Artery Study. Circulation 112:976–983. https://doi.org/10.1161/circulationaha.104.513085
Pande RL, Brown J, Buck S, Redline W, Doyle J, Plutzky J, Creager MA (2015) Association of monocyte tumor necrosis factor expression and serum infammatory biomarkers with walking impairment in peripheral artery disease. J Vasc Surg 61:155–161.https://doi.org/10.1016/j.jvs.2014.06.116
Wang L, Manson JE, Song Y, Sesso HD. Systematic review: Vitamin D and calcium supplementation in prevention of cardiovascular events. Ann Intern Med. 2010;152:315–323.PMID: 20194238 DOI: 10.7326/0003-4819-152-5-201003020-00010
AD (2016) Inhibition of endoplasmic reticulum stress and oxidative stress by vitamin D in endothelial cells. Free Radic Biol Med 99:1–10.https://doi.org/10.1016/j.freeradbiomed.2016.07.020 23.
Oh J, Weng S, Felton SK et al (2009) 1,25(OH)2 vitamin d inhibits foam cell formation and suppresses macrophage cholesterol uptake in patients with type 2 diabetes mellitus. Circulation 120:687–698. https://doi.org/10.1161/circulationaha.109.856070
Melamed ML, Muntner P, Michos ED, Uribarri J, Weber C, Sharma J, Raggi P (2008) Serum 25-hydroxyvitamin D levels and the prevalence of peripheral arterial disease: results from NHANES 2001 to 2004. Arterioscler Thromb Vasc Biol 28:1179– 1185. https://doi.org/10.1161/atvbaha.108.165886
Kang K. Low bone mineral density is associated with intracranial posterior circulation atherosclerosis in women. Bone 2015; 81:669–74.PMID: 26431918 DOI: 10.1016/j.bone.2015.09.016
Tamaki J, Iki M, Hirano Y, Sato Y, Kajita E, Kagamimori S, et al. Low bone mass is associated with carotid atherosclerosis in postmenopausal women: the Japanese Population-based Osteoporosis (JPOS) Cohort Study. Osteoporos Int 2009; 20: 53–60.PMID: 18496639 DOI: 10.1007/s00198-008-0633-z
Ozkaya E, Cakir E, Okuyan E, Cakir C, Ustun G, Kucukozkan T. Comparison of the effects of surgical and natural menopause on carotid intima media thickness, osteoporosis, and homocysteine levels. Menopause 2011;18:73–6.PMID: 20661165 DOI: 10.1097/gme.0b013e3181e5046d
Volpe R, Sotis G, Gavita R, Urbinati S, Valle S, Modena MG. Healthy diet to prevent cardiovascular diseases and osteoporosis: the experience of the ‘ProSa’ project. High Blood Press Cardiovasc Prev 2012;19:65–71.PMID: 22867092 DOI: 10.1007/BF03262456
Zapolski T, Wysokinski A. Safety of pharmacotherapy of osteoporosis in cardiology patients. Cardiol J 2010;17:335–43.PMID: 20690088
Lu Y, Ballew SH, Kwak L, Selvin E, Kalbaugh CA, Schrack JA, Matsushita K, Szklo M (2019) Physical activity and subsequent risk of hospitalization with peripheral artery disease and critical limb ischemia in the ARIC study. J Am Heart Assoc 8:e013534. https://doi.org/10.1161/jaha.119.013534
Sallam N, Laher I (2016) Exercise modulates oxidative stress and infammation in aging and cardiovascular diseases. Oxid Med Cell Longev 2016:7239639. https://doi.org/10.1155/2016/7239639
Pagnotti GM, Styner M, Uzer G, Patel VS, Wright LE, Ness KK, Guise TA, Rubin J, Rubin CT (2019) Combating osteoporosis and obesity with exercise: leveraging cell mechanosensitivity. Nat Rev Endocrinol 15:339–355. https://doi.org/10.1038/ s41574-019-0170-1
Zhang L, Zhang ZH, Wang QR, Su YJ, Lu YY, Zhang CL, Tsai HP, Wu CH. Stroke and osteoporosis: a Taiwan cohort study. Postgrad Med J. 2021 Apr;97(1146):211-216. doi: 10.1136/postgradmedj-2019-136959. Epub 2020 Mar 12.
Sato Y, Maruoka H, Honda Y, Asoh T, Fujimatsu Y, Oizumi K. Development of osteopenia in the hemiplegic finger in patients with stroke. Eur Neurol 1996;36 (5):278–83, doi:http://dx.doi.org/10.1159/000117272.
Poole KE, Reeve J, Warburton EA. Falls, fractures, and osteoporosis after stroke: time to think about protection? Stroke 2002;33(5):1432–6, doi:http://dx.doi.org/ 10.1161/01.str.0000014510.48897.7d.
Cauley JA, Giangregorio L. Physical activity and skeletal health in adults. Lancet Diabetes Endocrinol 2020;8(2):150–62, doi:http://dx.doi.org/10.1016/S2213-8587 (19)30351-1. potential treatments. Cerebrovasc Dis 2009;28(2):191–200, doi:http://dx. doi.org/10.1159/000226578.
J.J. Eng, M.Y. Pang, M.C. Ashe, J. Balance, falls, and bone health: role of exercise in reducing fracture risk after stroke. Rehabil. Res. Dev. 45 (2008) 297–313.PMID: 18566947 DOI: 10.1682/jrrd.2007.01.0014
G.S. Beaupre, H.L. Lew, Bone-density changes after stroke. Am. J. Phys. Med. Rehabil. 85 (2006) 464–472.PMID: 16628156 DOI: 10.1097/01.phm.0000214275.69286.7a
Yamada SM. Changes in bone mineral density in unconscious immobile stroke patients from the acute to chronic phases of brain diseases. Osteoporos Sarcopenia 2022;8 (3):106–11, doi:http://dx.doi.org/10.1016/j.afos.2022.08.001.
Wang J, et al. The effect of stroke on the bone mineral density: A systematic review and meta-analysis. J Nutr Health Aging. 2024. PMID: 38350301
Raghavan P, Santello M, Gordon AM, Krakauer JW. Compensatory motor control after stroke: an alternative joint strategy for object-dependent shaping of hand posture. J Neurophysiol 2010;103(6):3034–43, doi:http://dx.doi.org/10.1152/jn.00936.2009.
Salvietti G, Hussain I, Cioncoloni D, Taddei S, Rossi S, Prattichizzo D. Compensating hand function in chronic stroke patients through the robotic sixth finger. IEEE Trans Neural Syst Rehabil Eng 2017;25(2):142–50, doi:http://dx.doi.org/10.1109/ TNSRE.2016.2529684.
A.R. Webb, C. Pilbeam, N. Hanafin, M.F. Holick, An evaluation of the relative contributions of exposure to sunlight and of diet to the circulating concentrations of 25-hydroxyvitamin D in an elderly nursing home population in Boston. Am. J. Clin. Nutr. 51 (1990) 1075–1081.PMID: 2349922 DOI: 10.1093/ajcn/51.6.1075
J. Marsden, L.M. Gibson, C.E. Lightbody, A.K. Sharma, M. Siddiqi, C. Watkins. Can early onset bone loss be effectively managed in post-stroke patients? An integrative review of the evidence. Age Ageing 37 (2008) 142–150.PMID: 18349011 DOI: 10.1093/ageing/afm198
K.E. Poole, N. Loveridge, P.J. Barker, D.J. Halsall, C. Rose, J. Reeve, E.A. Warburton. Reduced vitamin D in acute stroke. Stroke 37 (2006) 243–245.PMID: 16322500 DOI: 10.1161/01.STR.0000195184.24297.c1

No competing interests reported.

file.docx
Supplementary Figures and Tables

Download PDF

Version 1

posted

You are reading this latest preprint version

Osteoporosis and Stroke: A Bidirectional Mendelian Randomization Study

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

Study design and data sources

Genetic instrument selection

Statistical analysis

Statistical significance

Results

Effect of Bone Mineral Density on Stroke

Effect of Stroke on Bone Mineral Density

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1