Causal Effect of the 25-Hydroxyvitamin D Concentration on Cerebral Small Vessel Disease: A Mendelian Randomization Study
Abstract
BACKGROUND:
Previous observational studies reported that a lower serum 25-hydroxyvitamin D [25(OH)D] concentration is associated with a higher burden of cerebral small vessel disease (cSVD). The causality of this association is uncertain, but it would be clinically important, given that 25(OH)D can be a target for intervention. We tried to examine the causal effect of 25(OH)D concentration on cSVD-related phenotypes using a Mendelian randomization approach.
METHODS:
Genetic instruments for each serum 25(OH)D concentration and cSVD-related phenotypes (lacunar stroke, white matter hyperintensity, cerebral microbleeds, and perivascular spaces) were derived from large-scale genome-wide association studies. We performed 2-sample Mendelian randomization analyses with multiple post hoc sensitivity analyses. A bidirectional Mendelian randomization approach was also used to explore the possibility of reverse causation.
RESULTS:
We failed to find any significant causal effect of 25(OH)D concentration on cSVD-related phenotypes (odds ratio [95% CI], 1.00 [0.87–1.16], 1.01 [0.96–1.07], 1.06 [0.85–1.33], 1.00 [0.97–1.03], 1.02 [0.99–1.04], 1.01 [0.99–1.04] for lacunar stroke, white matter hyperintensity, cerebral microbleeds, and white matter, basal ganglia, hippocampal perivascular spaces, respectively). These results were reproduced in the sensitivity analyses accounting for genetic pleiotropy. Conversely, when we examined the effects of cSVD phenotypes on 25(OH)D concentration, cerebral microbleeds were negatively associated with 25(OH)D concentration (0.94 [0.92–0.96]).
CONCLUSIONS:
Given the adequate statistical power (>0.8) of the analyses, our findings suggest that the previously reported association between 25(OH)D concentration and cSVD phenotypes might not be causal and partly attributed to reverse causation.
Graphical Abstract

Cerebral small vessel disease (cSVD) is a common disease entity, a well-known indicator of vascular brain health, and results in neurological disorders such as stroke or dementia.1–3 There is some evidence supporting that optimal blood pressure control, adequate diet, and physical exercise might prevent the progression of cSVD; yet, these interventions have been not adequately proven to be effective in preventing cSVD in large randomized clinical trials, although some trial evidence supports a benefit of antihypertensive treatment to slow down the progression of cSVD.3–5
Several observational studies have reported a negative association between serum 25-hydroxyvitamin D [25(OH)D] concentration and cSVD.6–8 Lowering blood pressure through the renin-angiotensin-aldosterone system, attenuating inflammation, and having a neuroprotective effect have been suggested as potential mechanisms.9,10 However, this association was not reproduced in other studies,11,12 which led to doubt as to whether it is truly causal.
Mendelian randomization is a method that uses single-nucleotide polymorphisms (SNPs) as a genetic instrumental variable for a given exposure to seek evidence for a causal relation with a given outcome. This method is based on the fact that genetic alleles are assorted randomly during meiosis, making the analysis less prone to confounding and reverse causation than observational studies.13 In other words, by using MR, one can simulate a randomized controlled trial. Two-sample MR studies derive their estimates from summary statistics of genome-wide association studies (GWAS) for the exposure and the outcome, respectively.14 Using recent large GWAS of 25(OH)D concentration and cSVD phenotypes, the present study aimed to elucidate whether 25(OH)D concentration is causally associated with 4 cSVD phenotypes, which the GWAS summary statistics were available, lacunar stroke, white matter hyperintensity (WMH) volume, cerebral microbleeds (CMBs), and perivascular spaces (PVSs).
METHODS
Data Availability and Ethics Statement
All data used for this study is publicly available from the respective GWAS. Ethical approval was not required for this study as we used anonymized summary-level data that were available to the public.
Data
We used summary statistics of the most recent GWASs of both 25(OH)D concentration and cSVD phenotypes: lacunar stroke, WMH volume, CMBs, and PVSs. The exposure dataset was composed of the summary statistics of the most recent and largest GWAS of the 25(OH)D concentration that was derived from an analysis of the UK Biobank population (N=417 580).15 The outcome dataset was composed of the most recent GWASs of each cSVD phenotype: lacunar stroke (6030 cases and 248 929 controls), WMH volume (N=48 454), CMBs (N=25 862), and PVSs (white matter [WM] PVSs, N=38 598; basal ganglia (BG) PVSs, N=38 903; and hippocampal PVSs, N=38 871).16–19 25(OH)D concentration was mostly measured at the initial assessment visit of the UK Biobank, and rank-based inverse-normal transformation was applied. Information on the month of assessment was included as a covariate, taking into account the well-known seasonal variation associated with 25(OH)D concentration,20 in addition to information on supplement intake. Lacunar stroke was defined according to the TOAST (Trial of ORG 10172 in Acute Stroke Treatment) criteria21 or as a subcortical infarct with a diameter ≤15 mm on diffusion-weighted imaging for acute infarcts and on fluid-attenuated inversion recovery or T1 sequences for nonacute infarcts.16 WMH volume was measured from T1, T2, and fluid-attenuated inversion recovery sequences (with proton density in some cases) usually using a fully automated software, and was transformed with rank-based inverse-normal transformation.17 CMBs were defined as small, hypointense lesions on susceptibility-weighted imaging or T2*-weighted gradient echo sequences.18,22 Among participants who had CMBs, 61.3% were classified as having lobar CMB, when they did not have any CMBs in the deep or infratentorial region, while the remaining were classified as mixed CMBs, defined as deep or infratentorial CMBs with a possible combination of CMBs in the lobar region.18 PVSs were quantified in the cerebral WM, BG, and hippocampus using visual semiquantitative rating scales or automated computational methods. These were identified as round, ovoid, or linear shape fluid-filled spaces, mimicking the signal of cerebrospinal fluid, usually with a maximum diameter of <3 mm. They exhibit no hyperintense rim on T2-weighted or fluid-attenuated inversion recovery sequences, and are located in areas supplied by perforating arteries.19,23 Due to the use of different scales across studies, the PVS burden was dichotomized using a cutoff point closest to the top quartile of the semiquantitative scale distribution within each cohort. The selected GWASs included participants of predominantly European ancestry. We used the summary statistics of GWAS conducted in exclusively European ancestry participants except for CMBs, for which available GWAS summary statistics included a small proportion of participants of other ancestries (5.8%).
Bidirectional Mendelian Randomization Analysis
SNPs that met the genome-wide significance level (P<5×10−8) were initially selected as genetic instruments. If no SNPs met the genome-wide significance level, a lower threshold of P<1×10−5 was used. Among the selected SNPs, linkage disequilibrium (LD) clumping with the threshold of R2<0.01 was performed to exclude the correlated SNPs. Heterogeneity among the SNPs was examined using the I2 statistic of the inverse-variance weighted method.
The primary 2-sample MR analysis was performed using the inverse-variance weighted method along with multiple sensitivity analyses with the weighted median, mode-based, MR-Egger, MR-Lasso, and MR-PRESSO (Mendelian Randomization Pleiotropy Residual Sum and Outlier) methods to rule out pleiotropy.24–29 Genetic pleiotropy was assessed using the intercept of the MR-Egger analysis. As the exposure GWAS provided the 25(OH)D concentration unit as rank-based inverse-normal transformed values, the effect size of this study was presented as a change in odds of each cSVD phenotype, the presence versus absence of lacunar stroke, CMBs, and PVSs, and the 1 SD of WMH volume, per 1 SD of the 25(OH)D concentration.
Additionally, a reverse MR analysis was performed using each cSVD phenotype as an exposure and the 25(OH)D concentration as an outcome to explore reverse causation between cSVD phenotypes and the 25(OH)D concentration. The selection of the genetic instruments and other analytic methods for this reverse MR analysis were identical to those of the primary analysis described earlier. For CMBs, analysis according to the subtypes of lobar CMBs and mixed CMBs was also performed. The statistical power of each MR analysis was calculated by the method established by Brion et al (https://shiny.cnsgenomics.com/mRnd/).30
Cross-Trait LD Score Regression Analysis
Genetic correlations between the 25(OH)D concentration and cSVD phenotypes were examined using cross-trait LD score regression analysis. This analysis determines the estimate of the genetic correlation between 2 phenotypes using genetic covariance and the LD score from the GWAS summary statistics of each.31,32 The information of the LD structure for the European ancestry was derived from the 1000 genome project.33
Statistical significance was considered at a 2-sided P<0.05. Analyses were performed using the Mendelian Randomization (version 0.5.0) and MR-PRESSO package for bidirectional MR analysis, and ieugwasr (version 0.1.5) for LD clumping and the LDSC package (version 1.0.1) for cross-trait LD score regression analysis.
RESULTS
GWASs used for this study are summarized in the Table. After data harmonization, the number of independent SNPs selected as instruments for 25(OH)D concentration was 94 for the analysis of association with lacunar stroke, 83 with WMH volume, 95 with CMBs, and 106 for PVSs. The primary 2-sample MR analysis showed no significant association between genetically determined 25(OH)D concentration and cSVD phenotypes; the odds ratios (95% CI) were 1.00 (0.87–1.16) for lacunar stroke, 1.01 (0.96–1.07) for WMH volume, 1.06 (0.85–1.33) for CMBs, 1.00 (0.97–1.03) for WM PVSs, 1.02 (0.99–1.04) for BG PVSs, and 1.01 (0.99–1.04) for hippocampal PVSs. These results were reproduced in the sensitivity analyses (Figures 1A through 1F and 2A through 2F). Moderate heterogeneity was suspected regarding the association between 25(OH)D concentration and WMH volume (I2=42.7; Table S1). However, the MR-PRESSO analysis, which provides results after excluding outlier SNPs, supported no association between the 25(OH)D concentration and WMH volume (Figure 1). Besides, there was only little heterogeneity regarding the associations between 25(OH)D levels and each of lacunar stroke, CMBs, WM PVSs, BG PVSs, and hippocampal PVSs (I2=0.0, 11.6, 18.5, 0.0, and 0.0, respectively). The MR-Egger intercept, which is an indicator of genetic pleiotropy, was small and statistically insignificant for all 6 cSVD phenotypes (Table S1).
Variable | Source GWAS | Ethnicity | N | Adjusted variable |
---|---|---|---|---|
Exposure | ||||
25-hydroxyvitamin D concentration15 | Nat Commun. 2020. Apr 2;11(1):1647. | European | 417 580 | Age, sex, assessment month, assessment center, supplement-intake information, genotyping batch, and PCs |
Outcome | ||||
Lacunar stroke16 | Lancet Neurol. 2021 May;20(5):351–361. | European | 6030 cases and 248 929 controls | PCs |
White matter hyperintensity volume17 | Nat Commun. 2020. Dec 8;11(1):6285. | European | 48 454 | Age, sex, PCs, and intracranial volume |
Cerebral microbleeds18 | Neurology. 2020;95:e3331–e3343. | Multiethnic (mainly European, 94.2%) | 3556 cases and 22 306 controls | Age, sex, PCs, family, and relations (if applicable) |
White matter perivascular spaces19 | Nat Med. 2023 Apr;29(4):950–962. | European | 9324 cases and 29 274 controls | Age, sex, intracranial volume, PCs, and study site |
Basal ganglia perivascular spaces19 | Nat Med. 2023 Apr;29(4):950–962. | European | 8913 cases and 29 990 controls | Age, sex, intracranial volume, PCs, and study site |
Hippocampal perivascular spaces19 | Nat Med. 2023 Apr;29(4):950–962. | European | 9223 cases and 29 648 controls | Age, sex, intracranial volume, PCs, and study site |
GWAS indicates genome-wide association study; and PCs, principal components.


In reverse MR analyses, 7 SNPs were selected as the genetic instruments for lacunar stroke, 21 SNPs for WMH volume, and 17 for WM PVSs. However, for CMBs, 11 SNPs met the criteria for the P value of genome-wide association significance and only 1 SNP remained after LD clumping. Thus, we used the post-clumped single SNP by the Wald method34 as a primary analysis for CMBs. Also, there was no SNP that met the criteria for the P value of genome-wide association significance for BG PVSs and hippocampal PVSs, so we used a lenient threshold of P<1×10−5 for these analyses. The primary analysis using the inverse-variance weighted method showed no significant association of genetic liability to lacunar stroke, PVSs, or genetically determined WMH volume with 25(OH)D concentration (Figures 3A, 3B, and 3D through 3F). In the sensitivity analysis, a significant negative association between WMH volume and 25(OH)D concentration was indicated by the weighted median, mode-based analysis, and MR-Lasso analysis, while a negative association between lacunar stroke and the 25(OH)D concentration was only suggestive based on the mode-based and MR-Egger analyses (Figures 3A, 3B, 4A, and 4B). However, there was no association between PVS phenotypes and 25(OH)D concentration (Figures 3D through 3F and 4D through 4F). Furthermore, substantial heterogeneity was suspected regarding these phenotypes (I2=72.4, 88.1, 37.4, 43.2, and 64.4 for the WMH volume, lacunar stroke, WM PVSs, BG PVSs, and hippocampal PVSs, respectively). The MR-PRESSO analysis showed no significant association after excluding the outlier SNPs, and the MR-Egger intercepts were not statistically significant (Figures 3A, 3B, and 3D through 3F; Table S1).


In contrast, the reverse MR analysis for CMBs based on the analysis using a single, post LD-clumped SNP showed a significant negative association between CMBs and the 25(OH)D concentration (odds ratio [95% CI], 0.94 [0.92–0.96]; Figure 3C).
We found no association between the 25(OH)D concentration and CMB subtypes, with the exception of a positive association identified in the MR-Lasso analysis for the reverse MR analysis between lobar CMBs and 25(OH)D concentration (odds ratio [95% CI], 1.01 [1.00–1.02]). It should, however, be noted that a lenient threshold of P<1×10−5 was applied for the reverse MR analysis, given that no SNP met the initial P value threshold, resulting in decreased statistical power (as detailed in Table S2).
The statistical power was ≥80% in most of the bidirectional MR analysis regarding associations between 25(OH)D concentration and WMH volume or CMBs, mode-based, and MR-Egger analysis regarding the bidirectional association between the 25(OH)D concentration and lacunar stroke. For PVSs, some of the sensitivity analysis such as weighted median method regarding the association between 25(OH)D concentration and hippocampal PVS and both mode-based and MR-Egger methods for the association between WM PVS and 25(OH)D concentration in the reverse MR analysis held ≥80% statistical power, while only mode-based analysis achieved this for BG PVS (Table S3).
The cross-trait LD score regression analysis did not show any genetic correlation between 25(OH)D concentration and lacunar stroke, WMH volume, or PVSs (Table S4). The genetic correlation between 25(OH)D concentration and CMBs could not be estimated, probably because of the lack of statistical power due to the small sample size of the source GWAS for CMBs.
DISCUSSION
Our results suggest that a low serum 25(OH)D concentration is not a direct cause of cSVD, which contradicts findings of previous observational studies.6–8 This result was supported by the absence of genetic correlation between 25(OH)D concentration and cSVD phenotypes in our study. However, the results of the reverse MR analysis suggested that cSVD phenotypes, especially CMBs, were negatively associated with serum 25(OH)D concentration; thus, cSVD burden reduced the serum 25(OH)D concentration.
There were a series of observational studies about the associations between serum 25(OH)D concentration and cSVD phenotypes. Two small studies of single hospital-based adult outpatients undergoing magnetic resonance imaging (N=56) and community-dwelling elderly people receiving home care (N=318) showed that a lower 25(OH)D concentration was associated with a higher burden of cSVD phenotypes, such as a larger WMH volume or lacunar stroke.7,8 This negative association was reproduced by another study of hospitalized patients with stroke (N=759): a 25(OH)D deficiency was associated with severe WMH, lacunes, and deep CMBs in a dose-dependent manner.6 However, the weakness of the aforementioned studies, including the small sample size and cross-sectional study design, might have limited generalizability of the causality of relationship. Furthermore, a well-designed, prospective, population-based cohort (N=888) could not find an association between the baseline 25(OH)D concentration and WMH grade cross-sectionally or WMH progression during an ≈10-year follow-up.11
MR analysis is an analogue of a randomized controlled trial, and it can control for unmeasured confounding and reverse causation, which is commonly found in observational studies.13,34 It would be infeasible and somewhat unethical to randomize vitamin D deficiency, but MR analysis made it possible to seek evidence for a causal effect of a low 25(OH)D concentration on cSVD phenotypes.34 Additionally, MR analyses estimate a lifelong effect of an exposure, which would be impossible in a randomized clinical trial. Thus, MR analysis would be a good choice for answering such a question of whether a vitamin D supplement has a beneficial effect on preventing cSVD. Moreover, we were able to secure a statistical power >90% in most of the analyses for the WMH volume and CMBs, and a power >80% in some analyses for lacunar stroke and PVSs. We were confident that we could confirm the associations between the 25(OH)D concentration and cSVD phenotypes in this study (Table S3).
Our results contradict those of a recent MR study based on the patients with ischemic stroke or myocardial infarction (N=441) and controls (N=5331) from Hong Kong.35 The study showed that vitamin D is causally protective against recurrent ischemic vascular events. However, stroke is a heterogeneous disease entity, and no protection against lacunar stroke does not mean no protection against large artery atherosclerosis or cardioembolism.21,36 Recently, a 2-sample MR study using publicly available summary statistics from published GWASs reported that lower 25(OH)D concentrations were associated with an increased risk of intracerebral hemorrhage (ICH), which is another phenotype of cSVD.37 The difference in results between this study and ours could be attributed to the difference in pathogenesis between ICH and CMB. In this recent ICH study, 25(OH)D concentration was negatively associated with nonlobar (deep) ICH, but not with lobar ICH. However, in our study, most patients had lobar CMBs, and the data for those with strictly nonlobar CMBs were unavailable.
It is quite interesting that reverse causation between cSVD phenotypes, especially CMBs, and the 25(OH)D concentration was observed in our study. people with a higher cSVD burden, who are prone to stroke or vascular cognitive impairment, are susceptible to poor nutritional status and frailty, which leads to lack of outdoor activity and a lower 25(OH)D concentration eventually.38,39 This association may be prominent with CMBs, as CMBs in our study were mainly lobar or mixed types suggestive of cerebral amyloid angiopathy, and Alzheimer disease shares a pathophysiology with cerebral amyloid angiopathy.40,41 The Rotterdam study, a prospective population-based cohort study, found that a lower 25(OH)D concentration was associated with prevalent stroke, not with incident stroke, which implies that a lower 25(OH)D is a consequence rather than a cause of stroke.42 Reverse causation is an important bias that should be considered in observational studies to investigate causal inference.43 Our results suggest that the negative association between the 25(OH)D concentration and cSVD burden might have been driven by reverse causation. Yet, people with CMBs may be at risk of 25(OH)D deficiency and might be suggested as a target population for screening serum 25(OH)D levels and providing supplementation if needed. This finding might add value to this field, as there is not enough evidence for screening for vitamin D deficiency in asymptomatic adults.44
Our study has several limitations. First, because the GWASs used in our study were mostly based on the patients with European ancestry, generalizability of our findings to other ethnicities might be limited. Second, we were unable to account for the nonlinear association in the analysis, which needs a single dataset with all the exposure and outcome variables along with information of the genetic instruments.45,46 However, a recent study that performed an nonlinear MR analysis using the UK Biobank database showed there was no association between the 25(OH)D concentration and WMH volume.47 Third, the subjects of each GWAS study were derived from either population-based cohorts or hospital-based cohorts and this difference in the source population may be a potential source of bias. Forth, it should be noticed that there are other cSVD phenotypes, such as cortical atrophy, superficial hemosiderosis, or microinfarcts, which were not analyzed in this study as GWAS results are not available yet.3 Lastly, although we performed various sensitivity analyses, possibilities of uncontrolled pleiotropies or heterogeneities cannot be excluded.
CONCLUSIONS
This 2-sample MR analysis using recent large GWASs indicates that 25(OH)D concentration does not have a causal effect on cSVD phenotypes. The association shown in previous observational studies might have been derived from unmeasured confounding and reverse causation, especially for CMBs.
ARTICLE INFORMATION
Supplemental Material
Tables S1–S4
STROBE-MR checklist
Acknowledgments
None.
Footnote
Nonstandard Abbreviations and Acronyms
- 25(OH)D
- 25-hydroxyvitamin D
- BG
- basal ganglia
- CMB
- cerebral microbleed
- cSVD
- cerebral small vessel disease
- GWAS
- genome-wide association study
- ICH
- intracerebral hemorrhage
- LD
- linkage disequilibrium
- PVS
- perivascular space
- SNP
- single-nucleotide polymorphism
- WM
- white matter
- WMH
- white matter hyperintensity
Supplemental Material
File (str_stroke-2023-042980_supp1.pdf)
- Download
- 468.31 KB
REFERENCES
1.
Cuadrado-Godia E, Dwivedi P, Sharma S, Santiago AO, Gonzalez JR, Balcells M, Laird J, Turk M, Suri HS, Nicolaides A, et al. Cerebral small vessel disease: a review focusing on pathophysiology, biomarkers, and machine learning strategies. J Stroke. 2018;20:302–320. doi: 10.5853/jos.2017.02922
2.
Gorelick PB, Furie KL, Iadecola C, Smith EE, Waddy SP, Lloyd-Jones DM, Bae HJ, Bauman MA, Dichgans M, Duncan PW, et al; American Heart Association/American Stroke Association. Defining optimal brain health in adults: a presidential advisory from the American Heart Association/American Stroke Association. Stroke. 2017;48:e284–e303. doi: 10.1161/STR.0000000000000148
3.
Wardlaw JM, Smith C, Dichgans M. Small vessel disease: mechanisms and clinical implications. Lancet Neurol. 2019;18:684–696. doi: 10.1016/S1474-4422(19)30079-1
4.
Smith EE, Saposnik G, Biessels GJ, Doubal FN, Fornage M, Gorelick PB, Greenberg SM, Higashida RT, Kasner SE, Seshadri S; American Heart Association Stroke Council; Council on Cardiovascular Radiology and Intervention; Council on Functional Genomics and Translational Biology; and Council on Hypertension. Prevention of stroke in patients with silent cerebrovascular disease: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2017;48:e44–e71. doi: 10.1161/STR.0000000000000116
5.
Wardlaw JM, Debette S, Jokinen H, De Leeuw F-E, Pantoni L, Chabriat H, Staals J, Doubal F, Rudilosso S, Eppinger S, et al. ESO Guideline on covert cerebral small vessel disease. Eur Stroke J. 2021;6:CXI–CLXII. doi: 10.1177/23969873211012132
6.
Chung PW, Park KY, Kim JM, Shin DW, Park MS, Chung YJ, Ha SY, Ahn SW, Shin HW, Kim YB, et al. 25-hydroxyvitamin D status is associated with chronic cerebral small vessel disease. Stroke. 2015;46:248–251. doi: 10.1161/STROKEAHA.114.007706
7.
Prager JM, Thomas C, Ankenbrandt WJ, Meyer JR, Gao Y, Ragin A, Sidharthan S, Hutten R, Wu YG. Association of white matter hyperintensities with low serum 25-hydroxyvitamin D levels. Am J Neuroradiol. 2014;35:1145–1149. doi: 10.3174/ajnr.A3840
8.
Buell JS, Dawson-Hughes B, Scott TM, Weiner DE, Dallal GE, Qui WQ, Bergethon P, Rosenberg IH, Folstein MF, Patz S, et al. 25-Hydroxyvitamin D, dementia, and cerebrovascular pathology in elders receiving home services. Neurology. 2010;74:18–26. doi: 10.1212/WNL.0b013e3181beecb7
9.
Buell JS, Dawson-Hughes B. Vitamin D and neurocognitive dysfunction: preventing “D”ecline? Mol Aspects Med. 2008;29:415–422. doi: 10.1016/j.mam.2008.05.001
10.
Danik JS, Manson JAE. Vitamin D and cardiovascular disease. Curr Treat Options Cardiovasc Med. 2012;14:414–424. doi: 10.1007/s11936-012-0183-8
11.
Michos ED, Carson KA, Schneider ALC, Lutsey PL, Xing L, Sharrett AR, Alonso A, Coker LH, Gross M, Post W, et al. Vitamin D and subclinical cerebrovascular disease: the atherosclerosis risk in communities brain magnetic resonance imaging study. JAMA Neurol. 2014;71:863–871. doi: 10.1001/jamaneurol.2014.755
12.
Littlejohns TJ, Kos K, Henley WE, Lang IA, Annweiler C, Beauchet O, Chaves PHM, Kestenbaum BR, Kuller LH, Langa KM, et al. Vitamin D and risk of neuroimaging abnormalities. PLoS One. 2016;11:e0154896–e0154812. doi: 10.1371/journal.pone.0154896
13.
Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ. 2018;362:k601. doi: 10.1136/bmj.k601
14.
Lawlor DA. Commentary: two-sample Mendelian randomization: opportunities and challenges. Int J Epidemiol. 2016;45:908–915. doi: 10.1093/ije/dyw127
15.
Revez JA, Lin T, Qiao Z, Xue A, Holtz Y, Zhu Z, Zeng J, Wang H, Sidorenko J, Kemper KE, et al. Genome-wide association study identifies 143 loci associated with 25 hydroxyvitamin D concentration. Nat Commun. 2020;11:1647. doi: 10.1038/s41467-020-15421-7
16.
Traylor M, Persyn E, Tomppo L, Klasson S, Abedi V, Bakker MK, Torres N, Li L, Bell S, Rutten-Jacobs L, et al; Helsinki Stroke, Study Dutch Parelsnoer Institute-Cerebrovascular Accident (CVA) Study Group. Genetic basis of lacunar stroke: a pooled analysis of individual patient data and genome-wide association studies. Lancet Neurol. 2021;20:351–361. doi: 10.1016/S1474-4422(21)00031-4
17.
Sargurupremraj M, Suzuki H, Jian X, Sarnowski C, Evans TE, Bis JC, Eiriksdottir G, Sakaue S, Terzikhan N, Habes M, et al; International Network against Thrombosis (INVENT) Consortium. Cerebral small vessel disease genomics and its implications across the lifespan. Nat Commun. 2020;11:6285. doi: 10.1038/s41467-020-19111-2
18.
Knol MJ, Lu D, Traylor M, Adams HHH, Romero JRJ, Smith AV, Fornage M, Hofer E, Liu J, Hostettler IC, et al; Alzheimer's Disease Neuroimaging Initiative. Association of common genetic variants with brain microbleeds: a genome-wide association study. Neurology. 2020;95:e3331–e3343. doi: 10.1212/WNL.0000000000010852
19.
Duperron M-G, Knol MJ, Le Grand Q, Evans TE, Mishra A, Tsuchida A, Roshchupkin G, Konuma T, Trégouët D-A, Romero JR, et al; CHARGE consortium. Genomics of perivascular space burden unravels early mechanisms of cerebral small vessel disease. Nat Med. 2023;29:950–962. doi: 10.1038/s41591-023-02268-w
20.
Levis S, Gomez A, Jimenez C, Veras L, Ma F, Lai S, Hollis B, Roos BA. Vitamin D deficiency and seasonal variation in an adult south Florida population. J Clin Endocrinol Metab. 2005;90:1557–1562. doi: 10.1210/jc.2004-0746
21.
Adams HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE. Classification of subtype of acute ischemic stroke Definitions for use in a multicenter clinical trial TOAST Trial of Org 10172 in Acute Stroke Treatment. Stroke. 1993;24:35–41. doi: 10.1161/01.str.24.1.35
22.
Greenberg SM, Vernooij MW, Cordonnier C, Salman RA, Edin F, Warach S, Lenore J, Van Buchem MA, Breteler MMB. Cerebral microbleeds: a field guide to their detection and interpretation. Lancet Neurol. 2009;8:165–174. doi: 10.1016/S1474-4422(09)70013-4
23.
Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R, Lindley RI, O’Brien JT, Barkhof F, Benavente OR, et al; STandards for ReportIng Vascular changes on nEuroimaging (STRIVE v1). Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol. 2013;12:822–838. doi: 10.1016/S1474-4422(13)70124-8
24.
Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37:658–665. doi: 10.1002/gepi.21758
25.
Bowden J, Smith GD, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44:512–525. doi: 10.1093/ije/dyv080
26.
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;40:304–314. doi: 10.1002/gepi.21965
27.
Hartwig FP, Smith GD, Bowden J. Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption. Int J Epidemiol. 2017;46:1985–1998. doi: 10.1093/ije/dyx102
28.
Rees JMB, Wood AM, Dudbridge F, Burgess S. Robust methods in Mendelian randomization via penalization of heterogeneous causal estimates. PLoS One. 2019;14:e0222362–e0222324. doi: 10.1371/journal.pone.0222362
29.
Verbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50:693–698. doi: 10.1038/s41588-018-0099-7
30.
Brion MJA, Shakhbazov K, Visscher PM. Calculating statistical power in Mendelian randomization studies. Int J Epidemiol. 2013;42:1497–1501. doi: 10.1093/ije/dyt179
31.
Bulik-Sullivan B, Loh PR, Finucane HK, Ripke S, Yang J, Patterson N, Daly MJ, Price AL, Neale BM, Corvin A, et al. LD score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet. 2015;47:291–295. doi: 10.1038/ng.3211
32.
Bulik-Sullivan B, Finucane HK, Anttila V, Gusev A, Day FR, Loh PR, Duncan L, Perry JRB, Patterson N, Robinson EB, et al; ReproGen Consortium. An atlas of genetic correlations across human diseases and traits. Nat Genet. 2015;47:1236–1241. doi: 10.1038/ng.3406
33.
Auton A, Abecasis GR, Altshuler DM, Durbin RM, Bentley DR, Chakravarti A, Clark AG, Donnelly P, Eichler EE, Flicek P, et al. A global reference for human genetic variation. Nature. 2015;526:68–74. doi: 10.1038/nature15393
34.
Lawlor DA, Harbord RM, Sterne JAC, Timpson N, Davey Smith G. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Stat Med. 2008;27:1133–1163. doi: 10.1002/sim.3034
35.
Chan YH, Mary Schooling C, Zhao J, Au Yeung SL, Hai JJ, Neil Thomas G, Cheng KK, Jiang CQ, Wong YK, Au KW, et al. Mendelian randomization focused analysis of vitamin D on the secondary prevention of ischemic stroke. Stroke. 2021;52:3926–3937. doi: 10.1161/STROKEAHA.120.032634
36.
Ko Y, Lee S, Chung J-W, Han M-K, Park J-M, Kang K, Park TH, Park S-S, Cho Y-J, Hong K-S, et al. MRI-based algorithm for acute ischemic stroke subtype classification. J Stroke. 2014;16:161–172. doi: 10.5853/jos.2014.16.3.161
37.
Szejko N, Acosta JN, Both CP, Leasure A, Matouk C, Sansing L, Gill TM, Hongyu Z, Sheth K, Falcone GJ. Genetically-proxied levels of vitamin D and risk of intracerebral hemorrhage. J Am Heart Assoc. 2022;11:e024141. doi: 10.1161/JAHA.121.024141
38.
Verhaar BJH, de Leeuw FA, Doorduijn AS, Fieldhouse JLP, van de Rest O, Teunissen CE, van Berckel BNM, Barkhof F, Visser M, de van der Schueren MAE, et al. Nutritional status and structural brain changes in Alzheimer’s disease: the NUDAD project. Alzheimers Dement (Amst). 2020;12:e12063. doi: 10.1002/dad2.12063
39.
Siejka TP, Srikanth VK, Hubbard RE, Moran C, Beare R, Wood A, Phan T, Callisaya ML. Frailty and cerebral small vessel disease: a cross-sectional analysis of the tasmanian study of cognition and gait (TASCOG). J Gerontol A Biol Sci Med Sci. 2018;73:255–260. doi: 10.1093/gerona/glx145
40.
Charidimou A, Boulouis G, Frosch MP, Baron JC, Pasi M, Albucher JF, Banerjee G, Barbato C, Bonneville F, Brandner S, et al. The Boston criteria version 20 for cerebral amyloid angiopathy: a multicentre, retrospective, MRI–neuropathology diagnostic accuracy study. Lancet Neurol. 2022;21:714–725. doi: 10.1016/S1474-4422(22)00208-3
41.
Greenberg SM, Bacskai BJ, Hernandez-Guillamon M, Pruzin J, Sperling R, van Veluw SJ. Cerebral amyloid angiopathy and Alzheimer disease — one peptide, two pathways. Nat Rev Neurol. 2020;16:30–42. doi: 10.1038/s41582-019-0281-2
42.
Berghout BP, Fani L, Heshmatollah A, Koudstaal PJ, Ikram MA, Zillikens MC, Ikram MK. Vitamin D status and risk of stroke: the Rotterdam study. Stroke. 2019;50:2293–2298. doi: 10.1161/STROKEAHA.119.025449
43.
Sattar N, Preiss D. Reverse causality in cardiovascular epidemiological research: more common than imagined? Circulation. 2017;135:2369–2372. doi: 10.1161/CIRCULATIONAHA.117.028307
44.
Krist AH, Davidson KW, Mangione CM, Cabana M, Caughey AB, Davis EM, Donahue KE, Doubeni CA, Epling JW, Kubik M, et al. Screening for vitamin D deficiency in adults: US preventive services task force recommendation statement. JAMA - J Am Med Assoc. 2021;325:1436–1442. doi: 10.1001/jama.2021.3069
45.
Arvanitis M, Qi G, Bhatt DL, Post WS, Chatterjee N, Battle A, McEvoy JW. Linear and nonlinear Mendelian randomization analyses of the association between diastolic blood pressure and cardiovascular events: the J-curve revisited. Circulation. 2021;143:895–906. doi: 10.1161/CIRCULATIONAHA.120.049819
46.
Burgess S, Davies NM, Thompson SG; EPIC-InterAct Consortium. Instrumental variable analysis with a nonlinear exposure-outcome relationship. Epidemiology. 2014;25:877–885. doi: 10.1097/EDE.0000000000000161
47.
Navale SS, Mulugeta A, Zhou A, Llewellyn DJ, Hyppönen E. Vitamin D and brain health: an observational and Mendelian randomization study. Am J Clin Nutr. 2022;116:531–540. doi: 10.1093/ajcn/nqac107
Information & Authors
Information
Published In
Copyright
© 2023 The Authors. Stroke is published on behalf of the American Heart Association, Inc., by Wolters Kluwer Health, Inc. This is an open access article under the terms of the Creative Commons Attribution Non-Commercial-NoDerivs License, which permits use, distribution, and reproduction in any medium, provided that the original work is properly cited, the use is noncommercial, and no modifications or adaptations are made.
Versions
You are viewing the most recent version of this article.
History
Received: 19 February 2023
Revision received: 31 May 2023
Accepted: 13 June 2023
Published online: 19 July 2023
Published in print: September 2023
Keywords
Subjects
Authors
Disclosures
Disclosures None.
Sources of Funding
This work was supported by the Seoul National University Bundang Hospital Research Fund (grant number no: 13-2021-0014) and partly supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT; no: 2023R1A2C2002925) and Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT; no: 2020-0-00121, Development of data improvement and dataset correction technology based on data quality assessment, “기여울”).
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
- Causal effect of serum matrix metalloproteinase levels on venous thromboembolism: a Mendelian randomization study, Epidemiology and Health, 46, (e2024046), (2024).https://doi.org/10.4178/epih.e2024046
- Causal effect of serum 25 hydroxyvitamin D concentration on cardioembolic stroke: Evidence from two-sample Mendelian randomization, Nutrition, Metabolism and Cardiovascular Diseases, 34, 5, (1305-1313), (2024).https://doi.org/10.1016/j.numecd.2024.02.013
- Causal effect of air pollution on the risk of brain health and potential mediation by gut microbiota, Ecotoxicology and Environmental Safety, 285, (117080), (2024).https://doi.org/10.1016/j.ecoenv.2024.117080
- Matrix Metalloproteinase and Aortic Aneurysm: A Two-sample Mendelian Randomization Study, Annals of Vascular Surgery, 105, (227-235), (2024).https://doi.org/10.1016/j.avsg.2024.02.011
- An inverse causal relationship between serum 25‐hydroxyvitamin D levels and pulmonary hypertension: A two‐sample Mendelian randomization study, Pulmonary Circulation, 14, 1, (2024).https://doi.org/10.1002/pul2.12350
Loading...
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
Purchase this article to access the full text.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.