Genetically Determined Uric Acid and the Risk of Cardiovascular and Neurovascular Diseases: A Mendelian Randomization Study of Outcomes Investigated in Randomized Trials

Anthoula Efstathiadou; Dipender Gill; Frances McGrane; Terence Quinn; Jesse Dawson

doi:10.1161/JAHA.119.012738

. 2019 Aug 23;8(17):e012738. doi: 10.1161/JAHA.119.012738

Genetically Determined Uric Acid and the Risk of Cardiovascular and Neurovascular Diseases: A Mendelian Randomization Study of Outcomes Investigated in Randomized Trials

Anthoula Efstathiadou ^1,^†,^✉, Dipender Gill ^1,^†, Frances McGrane ², Terence Quinn ^2,^‡, Jesse Dawson ^2,^‡

PMCID: PMC6755826 PMID: 31438759

Abstract

Background

Higher serum uric acid levels are associated with cardiovascular and neurovascular disease, but whether these relationships are causal is not known. We applied Mendelian randomization approaches to assess the association between genetically determined uric acid levels and outcomes under study in large clinical trials.

Methods and Results

We used 28 genetic variants related to serum uric acid as instruments to perform a range of 2‐sample Mendelian randomization methods. Our analysis had statistical power to detect clinically relevant effects of genetically determined serum uric acid levels on the considered clinical outcomes; cognitive function, Alzheimer disease, coronary heart disease, myocardial infarction, systolic blood pressure, and stroke. There was some suggestive evidence for an association between higher genetically determined serum uric acid and cognitive function. There was also some suggestive evidence of a relationship between coronary heart disease, systolic blood pressure, and the serum uric acid genetic instruments, but likely related to genetic pleiotropy. Overall, there was no consistent evidence of a clinically relevant effect of genetically determined serum uric acid on any of the considered outcomes.

Conclusions

This Mendelian randomization study does not support a clinically relevant causal effect of genetically determined serum urate on a range of cardiovascular and neurovascular outcomes. The weak association of genetically determined serum urate with coronary heart disease and systolic blood pressure may be because of pleiotropic effects. If urate lowering drugs such as allopurinol are found to affect these outcomes in clinical trials, then the effects may be mediated through urate independent mechanisms.

Keywords: Mendelian randomization, neurovascular disease, uric acid

Subject Categories: Genetic, Association Studies; Hypertension; Ischemic Stroke; Acute Coronary Syndromes; Coronary Artery Disease

Clinical Perspective

What Is New?

It is not known whether there is an association between serum uric acid and various cardiovascular and neurovascular outcomes.
Mendelian randomization is a statistical approach that uses genetic variants such as single‐nucleotide polymorphisms as genetic instruments to make causal inferences on the nature of an exposure‐outcome relationship.
This study used the Mendelian randomization technique to investigate the effect of genetically determined uric acid levels on cardiovascular and neurovascular outcomes under investigation in clinical trials.

What Are the Clinical Implications?

We did not find evidence of any clinically relevant causal association between uric acid and coronary heart disease, myocardial infarction, systolic blood pressure, cognitive function, Alzheimer disease or any type of ischemic stroke.
The protective role of urate lowering drugs on cardiovascular disease suggested by many observational studies and clinical trials might be attributable to urate independent mechanisms.

Observational studies have shown a relationship between higher serum uric acid levels and increased risk of stroke, coronary heart disease (CHD), total cardiovascular events and incident hypertension.1, 2, 3, 4 There may also be a complicated relationship with dementia—hyperuricemia may accelerate cerebrovascular disease resulting in vascular dementia,5 but may be neuroprotective in Alzheimer or Parkinson dementia, because of its antioxidant properties.6 However, causality has not been confirmed and there is considerable potential for confounding. The Mendelian randomization (MR) approach can overcome some of the limitations of observational studies such as confounding or reverse causation to make causal inferences.7 MR uses genetic variants randomly allocated at conception that are not associated with environmental confounders, such as single‐nucleotide polymorphisms (SNPs), as instruments to study the effect of varying an exposure on risk of a particular outcome.7 SNPs associated with uric acid can therefore be used as proxies for phenotypic serum uric acid levels to produce valid causal estimates when the underlying requisite assumptions of MR are met.8

Horizontal pleiotropy represents a violation of the requisite MR assumptions and occurs when the instruments are associated directly with the outcome independently of the exposure.9 MR analyses studying the effects of uric acid on cardiovascular disease and hypertension have previously been performed, with some supporting a causal association.10, 11, 12 In addition, a recent MR study did not find an association between serum uric acid levels and Alzheimer disease risk, but other dementia types or cognitive function were not investigated.13 Similarly, to our knowledge there is no published MR investigation of the role of uric acid on ischemic stroke and its subtypes.

Clinical trials of allopurinol, the most widely used uric acid lowering drug, are underway in the setting of cardiovascular disease and are exploring effects on blood pressure, cardiovascular event rate in patients with CHD (ALL‐HEART)14 and on cognitive and cardiovascular outcomes in patients with stroke (XILO‐FIST [Xanthine oxidase inhibition for improvement of long‐term outcomes following ischaemic stroke and transient ischaemic attack]).15 Numerous other small studies are in progress exploring surrogate outcomes. Allopurinol lowers serum uric acid in a dose dependent fashion, but also reduces the formation of reactive oxygen species and some of its vascular effects may be independent of uric acid reduction.16 These clinical trials will help inform on whether allopurinol improves outcomes, although interpretation of the mechanism of any improvement will be limited.

In this study, we performed MR analyses investigating the relationship between genetically determined serum uric acid levels and the range of cardiovascular and associated outcomes under study in trials (cognitive performance, Alzheimer disease, CHD, myocardial infarction [MI], systolic blood pressure [SBP], ischemic stroke and its subtypes [large‐artery atherosclerotic stroke {LAS}, cardioembolic stroke, and stroke caused by small‐vessel disease {SVS}]). Methods for dealing with potential violations of the modelling assumptions were incorporated, including a range of statistical sensitivity analyses performed to investigate the robustness of the findings.

Methods

All data used in this study come from GWAS (genome‐wide association study) meta‐analyses for which ethical approval and patient consent were previously obtained. The data used are available upon reasonable request from the corresponding author.

Genetic Association Estimates

We used 28 SNPs associated with serum uric acid concentration at genome‐wide significance (P<5×10⁻⁸) as genetic instruments. These were identified from a GWAS meta‐analysis of 110 347 participants of European ancestry.17 All 28 SNPs were replicated, and overall explained 5.8% of the variability in uric acid levels. In the largest study of the GWAS meta‐analysis, the mean uric acid concentration was 6.0 mg/dL and the standard deviation (SD) was 1.5 mg/dL. The strength of each instrument was evaluated using F‐statistic values, which were in turn calculated based on R², the proportion of phenotypic variance explained by each SNP.18, 19

Genetic association estimates of SNP‐outcome relationships were extracted from different data sets. For cognitive ability, SNPs were derived from a GWAS meta‐analysis performed by the UK Biobank and Cognition Genomics Consortium (COGENT) on 257 841 participants of European origin.20 Cognitive function was measured using a test of verbal‐numerical reasoning in the UK Biobank, and neuropsychological assessments in COGENT study, further details for which are available in the original reporting studies.21, 22 Cognitive performance was standardized (mean=0 and SD=1). Genetic association estimates for Alzheimer disease were extracted from the discovery stage of the International Genomics of Alzheimer's Project (IGAP), a meta‐analysis of 4 GWASs with a total of 17 008 cases and 37 154 controls, all of European ancestry.23 The Coronary Artery Disease Genome Wide Replication and Meta‐Analysis (CARDIoGRAM) plus The Coronary Artery Disease (C4D) (CARDIoGRAMplusC4D) 1000 Genomes‐based GWAS meta‐analysis was used to derive the association estimates for CHD and MI.24 There were 60 801 CHD cases and 123 504 controls, with the majority of participants (77%) being of European ancestry. The CHD definition included MI (≈70% of the total number of cases), acute coronary syndrome and angina pectoris. For SBP, data on 473 891 European‐ancestry participants from the UK Biobank were used, with SBP estimates obtained from automated readings at baseline assessment, and having a mean value of 137.82 and an SD of 19.3 in the considered population.25 Finally, for ischemic stroke and its subtypes LAS, cardioembolic stroke, and SVS, genetic association estimates were derived from a multi‐ancestry GWAS performed by the MEGASTROKE Consortium, on 60 341 cases of any type of ischemic stroke (6688 of LAS; 9006 of cardioembolic stroke; 11 710 of SVS) and 454 450 controls.26 Table 1 contains the information for all the data sets used in this study.

Table 1.

Characteristics of All Outcome Data Sets

Outcome	Consortium	Definition	Total Population	Ethnicity	Cases (%)	Controls	References
Cognitive performance	UK Biobank and COGENT	Cognitive performance was assessed using a test of verbal‐numerical reasoning in UK Biobank and various neuropsychological tests in COGENT study	257 841	European	···	···	Lee et al (2018)20
Alzheimer disease	IGAP(Stage 1)	WHO definition; the most common form of dementia. Dementia is a syndrome in which there is deterioration in memory, thinking, behavior and the ability to perform everyday activities	54 162	European	17 008 (31.4%)	37 154	Lambert et al (2013)23
Coronary heart disease	CARDIoGRAMplusC4D	Coronary stenosis >30%, documented angina, documented myocardial infarction	184 305	77% European	60 801 (33.0%)	23 504	Nikpay et al (2015)24
Myocardial infarction	CARDIoGRAMplusC4D	Documented myocardial infarction; myocardial cell necrosis attributable to major and constant ischemia	54 162	77% European	42 560 (78.6%)	11 602	Nikpay et al (2015)24
Systolic blood pressure	UK Biobank	Electronically measured systolic blood pressure at baseline visit	473 891	White British	···	···	UK Biobank (Neale's lab) (2018)25
Ischemic stroke—any type	MEGASTROKE	WHO definition; rapidly developing clinical signs of focal (or global) disturbance of cerebral function, with symptoms lasting ≥24 h or leading to death, with no apparent cause other than of vascular origin of a neurological deficit persisting >24 h related to a vascular insult to the brain	514 791	Mixed	60 341 (11.7%)	454 450	Malik et al (2018)26
Ischemic stroke—cardioembolic stroke	MEGASTROKE	A type of ischemic stroke attributable to emboli coming from heart	514 791	Mixed	9006 (1.7%)	505 785	Malik et al (2018)26
Ischemic stroke—large‐artery atherosclerotic stroke	MEGASTROKE	A type of ischemic stroke attributable to atherosclerosis of major intracranial arteries	514 791	Mixed	6688 (1.3%)	508 103	Malik et al (2018)26
Ischemic stroke—small‐vessel disease stroke	MEGASTROKE	A type of ischemic stroke attributable to pathological processes of small arteries, arterioles, venules, and capillaries of the brain	514 791	Mixed	11 710 (2.3%)	503 081	Malik et al (2018)26

Open in a new tab

COGENT indicates Cognition Genomics Consortium; IGAP, International Genomics of Alzheimer Project; WHO, World Health Organization.

Mendelian Randomization Power Calculation

Information on the available sample size and the percentage of phenotypic variance explained by the instruments was used to perform power calculations for conventional MR analyses using the mRnd power calculator (available at http://cnsgenomics.com/shiny/mRnd/).27 The smallest effect of uric acid (on all outcomes separately) required to achieve at least 80% statistical power given the sample size and percentage of phenotypic variance explained was calculated for the main analysis.

Mendelian Randomization Analyses and Investigation of Pleiotropy

For inverse‐variance weighted (IVW) ratio method MR, we generated MR estimates for each SNP using the Wald estimator (ratio of SNP‐outcome estimate over SNP‐exposure estimate), with standard errors calculated using the Delta method to account for possible measurement error in both the exposure and outcome association estimates.28, 29 In the main analysis, the final effect estimate for each outcome was derived by pooling all MR estimates using the fixed‐effects IVW method.30 As this method assumes the absence of any horizontal pleiotropic effects of the genetic instruments on the considered outcome through pathways independent of the exposure (which in this case is serum uric acid) and is vulnerable to bias if this assumption does not hold, further MR methods that have less stringent assumptions on horizontal pleiotropy were therefore performed in sensitivity analyses. We estimated the intercept and slope of MR‐Egger regression, which represent the average horizontal pleiotropy and a pleiotropy‐adjusted MR estimate, respectively.31 An intercept value for this regression that does not include the null (P<0.05) was considered indicative of horizontal pleiotropy.31 MR‐Egger makes the assumption that the strength of the instruments is not correlated to any pleiotropic effect that they have.31 Additionally, the weighted median estimator was performed.32 This is the weighted median effect of all the MR estimates produced by individual instruments, with weights equal to the inverse of the standard error.32 It is valid when more than half the information from the analysis comes from valid instruments.32 Finally, the Mendelian Randomization Pleiotropy Residual Sum and Outlier (MR‐PRESSO) method was performed, which excludes outliers determined by the square of residual errors from the SNP‐outcome against SNP‐exposure regression to calculate an outlier‐free effect estimate.33

Increased body mass index (BMI) has been causally associated with higher levels of serum uric acid in a recent MR study by Palmer et al.11 To reduce any possible genetic confounding related to associations of the uric acid instruments with BMI, we adjusted for this using conventional regression‐based multivariable MR.34 Multivariable MR is a linear regression‐based method with >1 explanatory variable.34 The genetic association estimates for BMI were derived from a GWAS meta‐analysis of the Genetic Investigation of Anthropometric Traits (GIANT) Consortium and UK Biobank, including ≈700 000 European‐ancestry participants.35

An additional sensitivity analysis to deal with any potential horizontal pleiotropy was performed. We examined whether the 28 SNPs were associated with any traits other than uric acid levels, using the PhenoScanner online data set of publicly available GWAS results (http://www.phenoscanner.medschl.cam.ac.uk/phenoscanner, accessed February 1, 2019), identifying associations that were genome‐wide significant (P<5×10⁻⁸), and also considering proxy SNPs (linkage disequilibrium r ²>0.8).36 The whole analysis was then repeated using only the SNPs that were exclusively associated with uric acid and/or gout. For the considered 9 outcomes, we accounted for multiple testing in the main analysis using a Bonferroni correction.

We used R software (version 3.5.1) and the MendelianRandomization37 and MRPRESSO33 software packages to perform analyses.

Results

All 28 SNPs had F‐statistic values >10, suggesting that they were unlikely to introduce marked weak instrument bias into the MR analyses.18 Table S1 contains the individual association estimates for uric acid and F‐statistic values of each instrument SNP. Investigating potentially pleiotropic associations identified that only 7 out of the 28 instrument SNPs were exclusively associated with either uric acid levels and/or gout (Table S2). Power calculations for the conventional IVW MR analyses using the 28 SNPs indicated >80% statistical power to detect an odds ratio (OR) <0.97 or >1.04 per 1 mg/dL increase in uric acid for the urate‐CHD/MI relationships and a beta estimate smaller or >0.096 per 1 mg/dL increase in uric acid for the urate‐SBP relationship. Similar power calculations were found for the MR analysis using the 7 SNPs specific to urate/gout. Tables S3 and S4 contain the power calculations for all outcomes for the 28 SNPs and 7 SNPs analyses, respectively. All results in this study are reported per 1 mg/dL increase in genetically determined uric acid.

Cognition and Alzheimer Disease

The results for uric acid and cognition are shown in Table 2. All analyses found an inverse relationship between serum uric acid level and cognitive performance. However, only the effect estimates from the weighted median method had a significant effect after adjusting for multiple testing using the Bonferroni correction (P<0.005) (weighted median estimate for the 7 only‐urate associated SNPs: β −0.03; 95% CI −0.05–−0.01; P=3.77×10⁻⁰⁴).38 The intercept of the MR‐Egger regression did not suggest the presence of directional pleiotropy (P=0.72). Results were consistent between all 28 and 7 SNPs.

Table 2.

Results From All MR Analyses for the Association of Uric Acid With Cognitive Performance and Alzheimer Disease

Uric Acid on Cognition	β (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	−0.02 (−0.04–0.01)	0.16
Weighted median	−0.03 (−0.05–−0.01)	1.00×10⁻⁰³
MR‐Egger	−0.02 (−0.06–0.01)	0.22
MR‐PRESSO (outliers corrected; 3 outliers)	−0.02 (−0.04–−0.01)	0.02
MR‐PRESSO (raw)	−0.02 (−0.04–0.01)	0.17
MVMR adjusting for BMI	−0.02 (−0.04–0.01)	0.19
Analysis with 7 SNPs^a
Fixed‐effects IVW	−0.03 (−0.06–0.00)	0.08
Weighted median	−0.03 (−0.05–−0.01)	3.77×10⁻⁰⁴
MR‐Egger	−0.03 (−0.07–0.02)	0.27
MR‐PRESSO (raw; 0 outliers)	−0.03 (−0.07–0.00)	0.09

Uric Acid on Alzheimer	OR (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	1.03 (0.96–1.10)	0.47
Weighted median	1.04 (0.95–1.14)	0.42
MR‐Egger	1.08 (0.98–1.20)	0.12
MR‐PRESSO (raw; 0 outliers)	1.03 (0.96–1.10)	0.49
MVMR adjusting for BMI	1.03 (0.96–1.11)	0.35
Analysis with 7 SNPs^a
Fixed‐effects IVW	1.05 (0.93–1.17)	0.48
Weighted median	1.05 (0.95–1.15)	0.31
MR‐Egger	1.06 (0.90–1.25)	0.48
MR‐PRESSO (raw; 0 outliers)	1.04 (0.93–1.17)	0.49

Open in a new tab

BMI indicates body mass index; IVW, inverse variance weighted; MR‐PRESSO, Mendelian Randomization Pleiotropy Residual Sum and Outlier; MR, Mendelian randomization; MVMR, multivariable Mendelian randomization analysis; SNP, single‐nucleotide polymorphisms.

The 7 non‐pleiotropic SNPs that are associated with only uric acid or/and gout.

Regarding Alzheimer disease, there was no demonstrable relationship with uric acid in any analysis (IVW method for the 7 SNPs analysis: OR 1.05; 95% CI 0.93–1.17; P=0.48). In addition, no horizontal pleiotropy was suggested by the MR‐Egger test (P=0.13). Detailed results can be found in Table 2.

Coronary Heart Disease and Myocardial Infarction

There was an apparent effect of uric acid on CHD in outlier‐corrected MR‐PRESSO (OR 1.07; 95% CI 1.03–1.12; P=2.99×10⁻³), but not in any other method or in analyses restricted to the 7 SNPs specific for urate (Table 3). The MR‐Egger test did not provide evidence of horizontal pleiotropy (P=0.06). Similar results were found for MI (Table 3).

Table 3.

Results From All MR Analyses for the Association of Uric Acid With Coronary Heart Disease, Myocardial Infarction, and Systolic Blood Pressure

Uric Acid on CHD	OR (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	1.08 (1.02–1.14)	0.01
Weighted median	1.05 (0.99–1.11)	0.10
MR‐Egger	1.02 (0.94–1.12)	0.63
MR‐PRESSO (outlier corrected; 1 outlier)	1.07 (1.03–1.12)	2.99×10⁻⁰³
MR‐PRESSO (raw)	1.08 (1.02–1.16)	0.02
MVMR adjusting for BMI	1.07 (1.01–1.14)	0.04
Analysis with 7 SNPs^a
Fixed‐effects IVW	1.04 (0.97–1.11)	0.29
Weighted median	1.03 (0.97–1.09)	0.29
MR‐Egger	1.03 (0.95–1.11)	0.58
MR‐PRESSO (raw; 0 outliers)	1.04 (0.97–1.11)	0.29

Uric Acid on MI	OR (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	1.09 (1.02–1.16)	0.02
Weighted median	1.06 (0.99–1.12)	0.08
MR‐Egger	1.01 (0.92–1.12)	0.80
MR‐PRESSO (outlier corrected; 1 outlier)	1.08 (1.03–1.14)	4.41×10⁻⁰³
MR‐PRESSO (raw)	1.10 (1.01–1.18)	0.03
MVMR adjusting for BMI	1.08 (1.00–1.17)	0.05
Analysis with 7 SNPs^a
Fixed‐effects IVW	1.04 (0.97–1.12)	0.28
Weighted median	1.04 (0.98–1.11)	0.22
MR‐Egger	1.03 (0.93–1.14)	0.54
MR‐PRESSO (raw; 0 outliers)	1.04 (0.97–1.12)	0.28

Uric Acid on SBP	β (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	0.47 (−0.02 to 0.95)	0.07
Weighted median	0.23 (−0.07 to 0.53)	0.13
MR‐Egger	−0.25 (−0.97 to 0.47)	0.49
MR‐PRESSO (outlier corrected; 4 outliers)	0.58 (0.19–0.97)	0.01
MR‐PRESSO (raw)	0.57 (−0.02 to 1.16)	0.07
MVMR adjusting for BMI	0.56 (−0.05 to 1.17)	0.08
Analysis with 7 SNPs^a
Fixed‐effects IVW	0.34 (−0.18 to 0.86)	0.25
Weighted median	0.21 (−0.09 to 0.51)	0.15
MR‐Egger	−0.12 (−0.53 to 0.30)	0.59
MR‐PRESSO (outliers corrected; 1 outlier)	0.31 (−0.19 to 0.80)	0.28
MR‐PRESSO (raw)	0.36 (−0.19 to 0.90)	0.25

Open in a new tab

BMI indicates body mass index; CHD, coronary heart disease; IVW, inverse variance weighted; MR‐PRESSO, Mendelian Randomization Pleiotropy Residual Sum and Outlier; MI, myocardial infarction; MR, Mendelian randomization; MVMR, multivariable Mendelian randomization analysis; SBP, systolic blood pressure.

The 7 non‐pleiotropic SNPs that are associated with only uric acid or/and gout.

Systolic Blood Pressure

In the main analysis using the 28 SNPs, 1 mg/dL increase in uric acid was associated with 0.47 mm Hg increase in SBP (IVW method: 95% CI −0.02–0.95; P=0.07). There was a suggestive association when performing the outlier corrected MR‐PRESSO method with all 28 SNPs, but this was not statistically significant after Bonferroni correction (MR‐PRESSO method; β 0.58; 95% CI 0.19–0.97; P=0.01) (Table 3). In addition, the MR‐Egger test suggested the presence of horizontal pleiotropy when considering all 28 SNPs (P=2×10⁻³).

Stroke

The effect estimates for ischemic stroke and its subtypes were consistent throughout all analysis methods and were not suggestive of an association between uric acid and ischemic stroke or its subtypes (Table 4). The OR for the effect of genetically determined uric acid on any type of ischemic stroke when using the IVW method in the 28 SNPs analysis was 1.00 (95% CI 0.94–1.06; P=0.99) and was consistent after adjusting for BMI in multivariable Mendelian randomization (OR 0.99; 95% CI 0.92–1.07; P=0.89) or after using the 7 SNPs associated with only uric acid/gout (IVW: OR 0.96; 95% CI 0.91–1.02; P=0.20). The MR‐Egger intercepts of all stroke analyses were found to be close to 0, indicating the absence of horizontal pleiotropy.

Table 4.

Results From All MR Analyses for the Association of Uric Acid With Ischemic Stroke and its Subtypes

Uric Acid on Ischemic Stroke	OR (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	1.00 (0.94–1.06)	0.99
Weighted median	0.98 (0.93–1.03)	0.42
MR‐Egger	0.95 (0.85–1.05)	0.31
MR‐PRESSO (outlier corrected; 1 outlier)	1.00 (0.95–1.05)	0.87
MR‐PRESSO (raw)	1.00 (0.94–1.08)	0.86
MVMR adjusting for BMI	0.99 (0.92–1.07)	0.89
Analysis with 7 SNPs^a
Fixed‐effects IVW	0.96 (0.91–1.02)	0.20
Weighted median	0.97 (0.92–1.03)	0.33
MR‐Egger	0.99 (0.93–1.06)	0.82
MR‐PRESSO (raw; 0 outliers)	0.96 (0.91–1.02)	0.20

Uric Acid on CES	OR (95% CI)	P Value
Analysis with 28 SNP
Fixed‐effects IVW	0.97 (0.89–1.05)	0.44
Weighted median	0.95 (0.86–1.05)	0.32
MR‐Egger	0.92 (0.82–1.04)	0.20
MR‐PRESSO (raw; 0 outliers)	0.97 (0.89–1.06)	0.49
MVMR adjusting for BMI	0.97 (0.89–1.06)	0.50
Analysis with 7 SNPs^a
Fixed‐effects IVW	0.93 (0.87–1.00)	0.08
Weighted median	0.95 (0.85–1.06)	0.33
MR‐Egger	0.97 (0.85–1.10)	0.60
MR‐PRESSO (raw; 0 outliers)	0.93 (0.87–1.00)	0.08

LAS	OR (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	1.01 (0.88–1.15)	0.94
Weighted median	0.94 (0.83–1.07)	0.36
MR‐Egger	0.89 (0.74–1.08)	0.24
MR‐PRESSO (outliers corrected; 1 outlier)	1.00 (0.88–1.13)	0.98
MR‐PRESSO (raw)	1.01 (0.88–1.16)	0.87
MVMR adjusting for BMI	1.00 (0.86–1.15)	0.98
Analysis with 7 SNPs^a
Fixed‐effects IVW	0.92 (0.84–1.01)	0.13
Weighted median	0.93 (0.81–1.06)	0.27
MR‐Egger	0.93 (0.79–1.10)	0.41
MR‐PRESSO (raw; 0 outliers)	0.92 (0.84–1.01)	0.13

SVS	OR (95% CI)	P Value
Analysis with 28 SNPs
Fixed‐effects IVW	1.04 (0.92–1.16)	0.55
Weighted median	1.05 (0.94–1.17)	0.41
MR‐Egger	1.08 (0.90–1.28)	0.42
MR‐PRESSO (outliers corrected; 1 outlier)	1.03 (0.93–1.13)	0.62
MR‐PRESSO (raw)	1.04 (0.92–1.17)	0.54
MVMR adjusting for BMI	1.02 (0.90–1.15)	0.79
Analysis with 7 SNPs^a
Fixed‐effects IVW	1.02 (0.93–1.12)	0.71
Weighted median	1.04 (0.92–1.17)	0.56
MR‐Egger	1.09 (0.94–1.26)	0.27
MR‐PRESSO (raw; 0 outliers)	1.02 (0.93–1.12)	0.72

Open in a new tab

BMI indicates body mass index; CES, cardioembolic stroke; IVW, inverse variance weighted; LAS, large‐artery atherosclerotic stroke; MR, Mendelian randomization; MR‐PRESSO, Mendelian Randomization Pleiotropy Residual Sum and Outlier; MVMR, multivariable Mendelian randomization analysis; SVS, small‐vessel stroke.

The 7 non‐pleiotropic SNPs that are associated with only uric acid or/and gout.

The association estimates of the SNPs with cognitive performance, Alzheimer disease, CHD, MI, SBP, and ischemic stroke and its subtypes are presented in Tables S5 through S13, respectively. Figures S1 through S9 are Forest plots representing the individual SNP MR estimates of the 28 SNPs for all outcomes, respectively. Funnel and Radial plots to visualize the presence of heterogeneity for every outcome are provided in Figures S10 through S18. Forest plots containing the association estimates found using each method for the uric acid‐cognition and uric acid‐Alzheimer disease analyses are provided in Figures S19 and S20, respectively. Similarly, Figure S21 is a Forest plot of the uric acid‐CHD and uric acid‐MI relationships, with Figures S22 and S23 the same for uric acid‐SBP and uric acid‐ischemic stroke and its subtypes, respectively.

Discussion

Our study did not provide consistent MR evidence to support a causal effect of genetically determined serum uric acid levels on cognitive function, Alzheimer disease, CHD, MI, or ischemic stroke, including its subtypes (cardioembolic stroke, SVS, and LAS), despite the associations observed in many observational studies.1, 2, 3, 4, 5 Although there was some evidence for an association with cognitive function, coronary outcomes and SBP, the findings did not survive correction for multiple testing and sensitivity analyses investigating bias related to pleiotropy. Although the analysis with 7 instrument SNPs had lower statistical power than that with 28 SNPs, it is important to appreciate that all the IVW MR analyses had sufficient power to detect clinically meaningful associations. These results will therefore help interpret results from ongoing clinical trials and inform the direction of future study.

Our results were suggestive of an inverse association between uric acid and cognitive performance, as the effect estimates of this relationship were consistent throughout all analysis methods and approached statistical significance, even after accounting for multiple testing. Whether this effect is of clinical importance, however, needs to be clarified and more research is needed. We also need to acknowledge that this study does not exclude the presence of smaller effects, or effects of allopurinol lowering drugs distinct from the effect of general urate lowering. Any relationship between uric acid levels and cognitive function or dementia is likely to be complex given the heterogeneous nature of dementias, the antioxidant effects of uric acid and the detrimental effects of chronic hyperuricemia on the vasculature. Observational studies suggest the relationship may differ by dementia subtype. In a recent systematic review and meta‐analysis, serum uric acid levels were lower in people with Alzheimer disease and dementia associated with Parkinson disease but there was no association with vascular dementia.39 Experimental studies suggest a neuroprotective effect of uric acid which may be beneficial in some neurodegenerative disorders,6 but also that chronic hyperuricemia could induce cognitive impairment via vascular damage.5 Our data support this. Studies assessing the relationship between instrumental variables for uric acid and vascular dementia would be helpful.

Considering the other outcomes, while for CHD and SBP there was some evidence of an association, the importance and clinical relevance of this is debatable. Our analysis for SBP suggested that 1 mg/dL increase in genetically determined uric acid would increase blood pressure by 0.34 mm Hg (IVW for the 7 SNPs analysis: 95% CI −0.18–0.86; P=0.25). However, other MR studies do support a causal role for hyperuricemia in hypertension10, 12 (although there are discordant reports11, 40) and clinical trials suggest a larger effect. A clinical trial of uric acid lowering therapies in adolescents showed an ≈10 mm Hg fall in SBP with both allopurinol and probenecid.41, 42 If a fall of this magnitude was present in older adults, it would be associated with a 41% reduction in stroke events and a 22% reduction in CHD events.41, 42 Whether these changes exist in older adults at risk of cardiovascular disease now needs to be established in clinical trials. The presence of a small effect in an MR study in the predominantly “healthy” UK Biobank population does not exclude a similar larger and important effect of uric acid reduction in hyperuricemia adults at increased risk. Further clinical trials are needed to assess this. The fact that both a xanthine oxidase inhibitor and a uricosuric drug lowered blood pressure in adolescents raises the possibility that this effect is mediated by uric acid itself and not by xanthine oxidase inhibition or a reduction in oxidative stress. However, there are alternative explanations. The effects of xanthine oxidase inhibitors may be uric acid independent; high doses of allopurinol, but not probenecid, improved measures of endothelial function in adults with heart failure and it is of interest that trials of uric acid reduction have been neutral in this setting.43, 44 Further, probenecid may exert its effects via changes in renal function and MR studies have suggested that activity of uric acid transporters rather than the level of uric acid itself is related to renal function.45 Whether the observed effects of probenecid on renal function are because of this mechanism or a uric acid effect is unclear. Further clinical study should aim to establish this.

MR studies have also shown an association between uric acid and CHD10, 12 but again there are conflicting reports.40 Hypertension is the biggest risk factor for stroke and if hyperuricemia causes this, a downstream effect on the risk of stroke would be expected. Stroke is a heterogeneous condition and the presence of an association with CHD makes it attractive to hypothesize there would be an association with LAS, which overlap in its etiology.26 However, we saw no MR evidence of this.

The findings of this study will help plan and interpret results of clinical trials of uric acid reduction. Both allopurinol and probenecid have been shown to reduce blood pressure in adolescents by between 6 and 10 mm Hg systolic.41, 42 Allopurinol has also been shown to reduce left ventricular hypertrophy and carotid intima media thickness.43, 46, 47 Left ventricular hypertrophy is a key risk factor for stroke and the lack of association with stroke (and LAS in particular in our study) raises the possibility that these effects are mediated by xanthine oxidase and not uric acid. Our data do not exclude potentially beneficial effects of allopurinol that are mediated via uric acid independent mechanism. Indeed, if clinical trials of allopurinol yield benefit on outcomes such as stroke, we could infer these effects are uric acid independent. Furthermore, patients with stroke have a higher risk of cognitive impairment and cardiac disease so a potentially beneficial effect of uric acid reduction in this population cannot be excluded. Overall however, we feel our data suggest trials of uric acid reduction in the setting of cardiovascular disease should focus on cardiac end points or blood pressure reduction. It is likely that if clinically important effects are seen that they will at least in part be because of urate independent mechanisms. Further, the potential for genetic information to identify people with particularly harmful forms of hyperuricemia, or who are most likely to respond to uric acid reduction, should be considered. This could be done by analysis of genetic data in ongoing trials.

Strengths of our study include the large sample size for each MR analysis with good statistical power. Both the 28 and 7 SNPs analyses had sufficient power to detect clinically relevant effect estimates (Tables S3 and S4). The availability of large‐scale GWASs for a range of relevant neurological and vascular outcomes further allowed investigation across the relevant outcome phenotypes in a manner more efficient that afforded by clinical trial,48 and also overcoming the issues of confounding and reverse causation bias that can limit traditional observational research.49 Furthermore, robust methods were applied to deal with possible violations of the requisite assumptions of MR, including horizontal pleiotropy.50

Limitations of this work include the fact that the sample size for subtypes of stroke was smaller than for other analyses, having potential implications for fine‐mapping the effects of uric acid on stroke subtypes that vary in their underlying etiology and pathophysiology.26 Furthermore, it is important to appreciate that MR measures the cumulative effect of lifelong exposure to genetic variants related to serum uric acid levels, and that this is not the same as studying the effect of a discrete clinical intervention in adult life.51 These MR results should therefore not be extrapolated to assume the effect of clinical intervention on uric acid levels, particularly as the MR approach may be subject to some residual bias related to pleiotropy despite the pleiotropy robust approaches. Furthermore, therapies for uric acid reduction may be having effects on neurological and vascular disease partly unrelated to their effects on uric acid.43, 44, 45, 46, 47 Finally, because of the fact that there is no available power calculation technique for MR‐Egger, we were not able to provide statistical power calculations for this analysis. Therefore, our analysis could contain false negative results on the MR‐Egger test for directional pleiotropy.

Conclusions

In conclusion, our study did not provide consistent evidence to support that genetically determined serum uric acid has a clinically relevant causal effect on risk of the considered cardiovascular and neurovascular outcomes. If there is an effect of urate lowering drugs on these mechanisms, it may be mediated by urate independent mechanisms.

Sources of Funding

Dr Gill is funded by the Wellcome Trust 4i programme at Imperial College London. Dr Quinn is supported by a Joint Chief Scientist Office and Stroke Association Senior Clinical Lectureship. The MEGASTROKE project received funding from sources specified at http://www.megastroke.org/acknowledgments.html.

Disclosures

Prof Dawson and Dr Quinn hold research funding to explore the effect of xanthine oxidase inhibition on blood pressure, cardiovascular and cognitive outcomes after stroke. The remaining authors have no disclosures to report.

Supporting information

Table S1. All 28 SNPs for Serum Urate From Köttgen et al Study (1)

Table S2. The 7 SNPs That are Only Associated With Serum Urate or/and Gout (1) After Excluding the Pleiotropic SNPs Using Phenoscanner (4)

Table S3. Power Calculations for All Analyses Using the 28 SNPs

Table S4. Power Calculations for All Analyses Using the 7 SNPs

Table S5. The Association Estimates of the 28 SNPs for Urate (1) With Cognitive Performance (7)

Table S6. The Association Estimates of the 28 SNPs for Urate (1) with Alzheimer's Disease (9)

Table S7. The Association Estimates of the 28 SNPs for Urate (1) With Coronary Heart Disease (CHD) (6)

Table S8. The Association Estimates of the 28 SNPs for Urate (1) With Myocardial Infarction (MI) (6)

Table S9. The Association Estimates of the 28 SNPs for Urate (1) With Systolic Blood Pressure (SBP) (8)

Table S10. The Association Estimates of the 28 SNPs for Urate (1) With Any Ischemic Stroke (IS) (10)

Table S11. The Association Estimates of the 28 SNPs for Urate (1) With Cardio‐Embolic Stroke (CES) (10)

Table S12. The Association Estimates of the 28 SNPs for Urate (1) With Large‐Artery Atherosclerotic Stroke (LAS) (10)

Table S13. The Association Estimates of the 28 SNPs for Urate (1) With Small‐Vessel Stroke (SVS) (10)

Figure S1. Forest plot of the 28 MR estimates of the urate‐cognitive performance relationship.

Figure S2. Forest plot of the 28 MR estimates of the urate‐Alzheimer's disease relationship.

Figure S3. Forest plot of the 28 MR estimates of the urate‐coronary heart disease (CHD) relationship.

Figure S4. Forest plot of the 28 MR estimates of the urate‐myocardial infarction (MI) relationship.

Figure S5. Forest plot of the 28 MR estimates of the urate‐systolic blood pressure (SBP) relationship.

Figure S6. Forest plot of the 28 MR estimates of the urate‐any ischemic stroke relationship.

Figure S7. Forest plot of the 28 MR estimates of the urate‐cardioembolic stroke ischemic stroke (CES) relationship.

Figure S8. Forest plot of the 28 MR estimates of the urate‐large‐artery atherosclerotic ischemic stroke (LAS) relationship.

Figure S9. Forest plot of the 28 MR estimates of the urate‐small‐artery stroke ischemic stroke (SAS) relationship.

Figure S10. Funnel plot (A) and radial plot (B) for the urate‐cognitive performance relationship.

Figure S11. Funnel plot (A) and radial plot (B) for the urate‐Alzheimer's disease relationship.

Figure S12. Funnel plot (A) and radial plot (B) for the urate‐coronary heart disease (CHD) relationship.

Figure S13. Funnel plot (A) and radial plot (B) for the urate‐myocardial infarction (MI) relationship.

Figure S14. Funnel plot (A) and radial plot (B) for the urate‐ systolic blood pressure (SBP) relationship.

Figure S15. Funnel plot (A) and radial plot (B) for the urate‐any ischemic stroke relationship.

Figure S16. Funnel plot (A) and radial plot (B) for the urate‐ cardioembolic ischemic stroke (CES) relationship.

Figure S17. Funnel plot (A) and radial plot (B) for the urate‐ large‐artery atherosclerotic ischemic stroke (LAS) relationship.

Figure S18. Funnel plot (A) and radial plot (B) for the urate‐ small‐artery ischemic stroke (SVS) relationship.

Figure S19. Forest plot for the association of uric acid with cognitive performance.

Figure S20. Forest plot for the association of uric acid with Alzheimer disease.

Figure S21. Forest plots for the association of uric acid with coronary heart disease (CHD), myocardial infarction (MI).

Figure S22. Forest plot for the association of uric acid with systolic blood pressure (SBP).

Figure S23. Forest plots for the association of uric acid with ischemic stroke (any type) and its subtypes (cardioembolic stroke [CES], stroke caused by small‐vessel disease (small‐vessel stroke [SVS] and large‐artery atherosclerotic stroke [LAS]).

Click here for additional data file.^{(2.7MB, pdf)}

Acknowledgments

The authors offer their thanks and appreciation to the UK Biobank, COGENT consortium, IGAP, CARDIoGRAMplusC4D Consortium, GIANT consortium, MEGASTROKE consortium, for making the data used in this study publicly available. Details of all MEGASTROKE authors are available at http://www.megastroke.org/authors.html.

(J Am Heart Assoc. 2019;8:e012738 DOI: 10.1161/JAHA.119.012738.)

References

1. Storhaug HM, Norvik JV, Toft I, Eriksen BO, Løchen M‐L, Zykova S, Solbu M, White S, Chadban S, Jenssen T. Uric acid is a risk factor for ischemic stroke and all‐cause mortality in the general population: a gender specific analysis from The Tromsø Study. BMC Cardiovasc Disord. 2013;13:115. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Muiesan ML, Agabiti‐Rosei C, Paini A, Salvetti M. Uric acid and cardiovascular disease: an update. Eur Cardiol. 2016;11:54–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Chang C‐C, Wu C‐H, Liu L‐K, Chou R‐H, Kuo C‐S, Huang P‐H, Chen LK, Lin SJ. Association between serum uric acid and cardiovascular risk in nonhypertensive and nondiabetic individuals: the Taiwan I‐Lan Longitudinal Aging Study. Sci Rep. 2018;8:5234. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Jossa F, Farinaro E, Panico S, Krogh V, Celentano E, Galasso R, Mancini M, Trevisan M. Serum uric acid and hypertension: the Olivetti heart study. J Hum Hypertens. 1994;8:677–681. [PubMed] [Google Scholar]
5. Vannorsdall TD, Jinnah HA, Gordon B, Kraut M, Schretlen DJ. Cerebral ischemia mediates the effect of serum uric acid on cognitive function. Stroke. 2008;39:3418–3420. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Bakshi R, Zhang H, Logan R, Joshi I, Xu Y, Chen X, Schwarzschild M. Neuroprotective effects of urate are mediated by augmenting astrocytic glutathione synthesis and release. Neurobiol Dis. 2015;82:574–579. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. 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] [PubMed] [Google Scholar]
8. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23:R89–R98. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Hemani G, Bowden J, Davey Smith G. Evaluating the potential role of pleiotropy in Mendelian randomization studies. Hum Mol Genet. 2018;27:R195–R208. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Kleber ME, Delgado G, Grammer TB, Silbernagel G, Huang J, Kramer BK, Ritz E, März W. Uric acid and cardiovascular events: a Mendelian randomization study. J Am Soc Nephrol. 2015;26:2831–2838. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Palmer TM, Nordestgaard BG, Benn M, Tybjaerg‐Hansen A, Davey Smith G, Lawlor DA, Timpson NJ. Association of plasma uric acid with ischaemic heart disease and blood pressure: Mendelian randomisation analysis of two large cohorts. BMJ. 2013;347:f4262. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Li X, Meng X, Spiliopoulou A, Timofeeva M, Wei W‐Q, Gifford A, Shen X, He Y, Varley T, McKeigue P, Tzoulaki I, Wright AF, Joshi P, Denny JC, Campbell H, Theodoratou E. MR‐PheWAS: exploring the causal effect of SUA level on multiple disease outcomes by using genetic instruments in UK Biobank. Ann Rheum Dis. 2018;77:1039–1047. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Williams DM, Hägg S, Pedersen NL. Circulating antioxidants and Alzheimer disease prevention: a Mendelian randomization study. Am J Clin Nutr. 2018;109:90–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Mackenzie IS, Ford I, Walker A, Hawkey C, Begg A, Avery A, Taggar J, Wei L, Struthers AD, MacDonald TM; ALL‐HEART study group . Multicentre, prospective, randomised, open‐label, blinded end point trial of the efficacy of allopurinol therapy in improving cardiovascular outcomes in patients with ischaemic heart disease: protocol of the ALL‐HEART study. BMJ Open. 2016;6:e013774. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Dawson J, Broomfield N, Dani K, Dickie DA, Doney A, Forbes K, Houston G, Kean S, Lees K, McConnachie A, Muir KW, Quinn T, Struthers A, Walters M. Xanthine oxidase inhibition for the improvement of long‐term outcomes following ischaemic stroke and transient ischaemic attack (XILO‐FIST)—protocol for a randomised double blind placebo‐controlled clinical trial. Eur Stroke J. 2018;3:281–290. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. George J, Carr E, Davies J, Belch JJF, Struthers A. High‐dose allopurinol improves endothelial function by profoundly reducing vascular oxidative stress and not by lowering uric acid. Circulation. 2006;114:2508–2516. [DOI] [PubMed] [Google Scholar]
17. Köttgen A, Albrecht E, Teumer A, Vitart V, Krumsiek J, Hundertmark C, Pistis G, Ruggiero D, O'Seaghdha CM, Haller T, Yang Q, Tanaka T, Johnson AD, Kutalik Z, Smith AV, Shi J, Struchalin M, Middelberg RP, Brown MJ, Gaffo AL, Pirastu N, Li G, Hayward C, Zemunik T, Huffman J, Yengo L, Zhao JH, Demirkan A, Feitosa MF, Liu X, Malerba G, Lopez LM, van der Harst P, Li X, Kleber ME, Hicks AA, Nolte IM, Johansson A, Murgia F, Wild SH, Bakker SJ, Peden JF, Dehghan A, Steri M, Tenesa A, Lagou V, Salo P, Mangino M, Rose LM, Lehtimäki T, Woodward OM, Okada Y, Tin A, Müller C, Oldmeadow C, Putku M, Czamara D, Kraft P, Frogheri L, Thun GA, Grotevendt A, Gislason GK, Harris TB, Launer LJ, McArdle P, Shuldiner AR, Boerwinkle E, Coresh J, Schmidt H, Schallert M, Martin NG, Montgomery GW, Kubo M, Nakamura Y, Tanaka T, Munroe PB, Samani NJ, Jacobs DR Jr, Liu K, D'Adamo P, Ulivi S, Rotter JI, Psaty BM, Vollenweider P, Waeber G, Campbell S, Devuyst O, Navarro P, Kolcic I, Hastie N, Balkau B, Froguel P, Esko T, Salumets A, Khaw KT, Langenberg C, Wareham NJ, Isaacs A, Kraja A, Zhang Q, Wild PS, Scott RJ, Holliday EG, Org E, Viigimaa M, Bandinelli S, Metter JE, Lupo A, Trabetti E, Sorice R, Döring A, Lattka E, Strauch K, Theis F, Waldenberger M, Wichmann HE, Davies G, Gow AJ, Bruinenberg M; LifeLines Cohort Study , Stolk RP, Kooner JS, Zhang W, Winkelmann BR, Boehm BO, Lucae S, Penninx BW, Smit JH, Curhan G, Mudgal P, Plenge RM, Portas L, Persico I, Kirin M, Wilson JF, Mateo Leach I, van Gilst WH, Goel A, Ongen H, Hofman A, Rivadeneira F, Uitterlinden AG, Imboden M, von Eckardstein A, Cucca F, Nagaraja R, Piras MG, Nauck M, Schurmann C, Budde K, Ernst F, Farrington SM, Theodoratou E, Prokopenko I, Stumvoll M, Jula A, Perola M, Salomaa V, Shin SY, Spector TD, Sala C, Ridker PM, Kähönen M, Viikari J, Hengstenberg C, Nelson CP; CARDIoGRAM Consortium; DIAGRAM Consortium; ICBP Consortium; MAGIC Consortium , Meschia JF, Nalls MA, Sharma P, Singleton AB, Kamatani N, Zeller T, Burnier M, Attia J, Laan M, Klopp N, Hillege HL, Kloiber S, Choi H, Pirastu M, Tore S, Probst‐Hensch NM, Völzke H, Gudnason V, Parsa A, Schmidt R, Whitfield JB, Fornage M, Gasparini P, Siscovick DS, Polašek O, Campbell H, Rudan I, Bouatia‐Naji N, Metspalu A, Loos RJ, van Duijn CM, Borecki IB, Ferrucci L, Gambaro G, Deary IJ, Wolffenbuttel BH, Chambers JC, März W, Pramstaller PP, Snieder H, Gyllensten U, Wright AF, Navis G, Watkins H, Witteman JC, Sanna S, Schipf S, Dunlop MG, Tönjes A, Ripatti S, Soranzo N, Toniolo D, Chasman DI, Raitakari O, Kao WH, Ciullo M, Fox CS, Caulfield M, Bochud M, Gieger C. Genome‐wide association analyses identify 18 new loci associated with serum urate concentrations. Nat Genet. 2013;45:145–154. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Palmer TM, Lawlor DA, Harbord RM, Sheehan NA, Tobias JH, Timpson NJ, Davey Smith G, Sterne J. Using multiple genetic variants as instrumental variables for modifiable risk factors. Stat Methods Med Res. 2012;21:223–242. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM, Koseki M, Pirruccello JP, Ripatti S, Chasman DI, Willer CJ, Johansen CT, Fouchier SW, Isaacs A, Peloso GM, Barbalic M, Ricketts SL, Bis JC, Aulchenko YS, Thorleifsson G, Feitosa MF, Chambers J, Orho‐Melander M, Melander O, Johnson T, Li X, Guo X, Li M, Shin Cho Y, Jin Go M, Jin Kim Y, Lee JY, Park T, Kim K, Sim X, Twee‐Hee Ong R, Croteau‐Chonka DC, Lange LA, Smith JD, Song K, Hua Zhao J, Yuan X, Luan J, Lamina C, Ziegler A, Zhang W, Zee RY, Wright AF, Witteman JC, Wilson JF, Willemsen G, Wichmann HE, Whitfield JB, Waterworth DM, Wareham NJ, Waeber G, Vollenweider P, Voight BF, Vitart V, Uitterlinden AG, Uda M, Tuomilehto J, Thompson JR, Tanaka T, Surakka I, Stringham HM, Spector TD, Soranzo N, Smit JH, Sinisalo J, Silander K, Sijbrands EJ, Scuteri A, Scott J, Schlessinger D, Sanna S, Salomaa V, Saharinen J, Sabatti C, Ruokonen A, Rudan I, Rose LM, Roberts R, Rieder M, Psaty BM, Pramstaller PP, Pichler I, Perola M, Penninx BW, Pedersen NL, Pattaro C, Parker AN, Pare G, Oostra BA, O'Donnell CJ, Nieminen MS, Nickerson DA, Montgomery GW, Meitinger T, McPherson R, McCarthy MI, McArdle W, Masson D, Martin NG, Marroni F, Mangino M, Magnusson PK, Lucas G, Luben R, Loos RJ, Lokki ML, Lettre G, Langenberg C, Launer LJ, Lakatta EG, Laaksonen R, Kyvik KO, Kronenberg F, König IR, Khaw KT, Kaprio J, Kaplan LM, Johansson A, Jarvelin MR, Janssens AC, Ingelsson E, Igl W, Kees Hovingh G, Hottenga JJ, Hofman A, Hicks AA, Hengstenberg C, Heid IM, Hayward C, Havulinna AS, Hastie ND, Harris TB, Haritunians T, Hall AS, Gyllensten U, Guiducci C, Groop LC, Gonzalez E, Gieger C, Freimer NB, Ferrucci L, Erdmann J, Elliott P, Ejebe KG, Döring A, Dominiczak AF, Demissie S, Deloukas P, de Geus EJ, de Faire U, Crawford G, Collins FS, Chen YD, Caulfield MJ, Campbell H, Burtt NP, Bonnycastle LL, Boomsma DI, Boekholdt SM, Bergman RN, Barroso I, Bandinelli S, Ballantyne CM, Assimes TL, Quertermous T, Altshuler D, Seielstad M, Wong TY, Tai ES, Feranil AB, Kuzawa CW, Adair LS, Taylor HA Jr, Borecki IB, Gabriel SB, Wilson JG, Holm H, Thorsteinsdottir U, Gudnason V, Krauss RM, Mohlke KL, Ordovas JM, Munroe PB, Kooner JS, Tall AR, Hegele RA, Kastelein JJ, Schadt EE, Rotter JI, Boerwinkle E, Strachan DP, Mooser V, Stefansson K, Reilly MP, Samani NJ, Schunkert H, Cupples LA, Sandhu MS, Ridker PM, Rader DJ, van Duijn CM, Peltonen L, Abecasis GR, Boehnke M, Kathiresan S. Biological, clinical and population relevance of 95 loci for blood lipids. Nature. 2010;466:707–713. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Lee JJ, Wedow R, Okbay A, Kong E, Maghzian O, Zacher M, Nguyen‐Viet TA, Bowers P, Sidorenko J, Karlsson Linnér R, Fontana MA, Kundu T, Lee C, Li H, Li R, Royer R, Timshel PN, Walters RK, Willoughby EA, Yengo L; 23andMe Research Team; COGENT (Cognitive Genomics Consortium); Social Science Genetic Association Consortium , Alver M, Bao Y, Clark DW, Day FR, Furlotte NA, Joshi PK, Kemper KE, Kleinman A, Langenberg C, Mägi R, Trampush JW, Verma SS, Wu Y, Lam M, Zhao JH, Zheng Z, Boardman JD, Campbell H, Freese J, Harris KM, Hayward C, Herd P, Kumari M, Lencz T, Luan J, Malhotra AK, Metspalu A, Milani L, Ong KK, Perry JRB, Porteous DJ, Ritchie MD, Smart MC, Smith BH, Tung JY, Wareham NJ, Wilson JF, Beauchamp JP, Conley DC, Esko T, Lehrer SF, Magnusson PKE, Oskarsson S, Pers TH, Robinson MR, Thom K, Watson C, Chabris CF, Meyer MN, Laibson DI, Yang J, Johannesson M, Koellinger PD, Turley P, Visscher PM, Benjamin DJ, Cesarini D. Gene discovery and polygenic prediction from a genome‐wide association study of educational attainment in 1.1 million individuals. Nat Genet. 2018;50:1112–1121. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Trampush JW, Yang MLZ, Yu J, Knowles E, Davies G, Liewald DC, Starr JM, Djurovic S, Melle I, Sundet K, Christoforou A, Reinvang I, DeRosse P, Lundervold AJ, Steen VM, Espeseth T, Räikkönen K, Widen E, Palotie A, Eriksson JG, Giegling I, Konte B, Roussos P, Giakoumaki S, Burdick KE, Payton A, Ollier W, Horan M, Chiba‐Falek O, Attix DK, Need AC, Cirulli ET, Voineskos AN, Stefanis NC, Avramopoulos D, Hatzimanolis A, Arking DE, Smyrnis N, Bilder RM, Freimer NA, Cannon TD, London E, Poldrack RA, Sabb FW, Congdon E, Conley ED, Scult MA, Dickinson D, Straub RE, Donohoe G, Morris D, Corvin A, Gill M, Hariri AR, Weinberger DR, Pendleton N, Bitsios P, Rujescu D, Lahti J, Le Hellard S, Keller MC, Andreassen OA, Deary IJ, Glahn DC, Malhotra AK, Lencz T. GWAS meta‐analysis reveals novel loci and genetic correlates for general cognitive function: a report from the COGENT consortium. Mol Psychiatry. 2017;22:336–345. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Lyall DM, Cullen B, Allerhand M, Smith DJ, Mackay D, Evans J, Anderson J, Fawns‐Ritchie C, McIntosh AM, Deary IJ, Pell JP. Cognitive test scores in UK Biobank: data reduction in 480,416 participants and longitudinal stability in 20,346 participants. PLoS One. 2016;11:e0154222. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Lambert J‐C, Ibrahim‐Verbaas CA, Harold D, Naj AC, Sims R, Bellenguez C, DeStafano AL, Bis JC, Beecham GW, Grenier‐Boley B, Russo G, Thorton‐Wells TA, Jones N, Smith AV, Chouraki V, Thomas C, Ikram MA, Zelenika D, Vardarajan BN, Kamatani Y, Lin CF, Gerrish A, Schmidt H, Kunkle B, Dunstan ML, Ruiz A, Bihoreau MT, Choi SH, Reitz C, Pasquier F, Cruchaga C, Craig D, Amin N, Berr C, Lopez OL, De Jager PL, Deramecourt V, Johnston JA, Evans D, Lovestone S, Letenneur L, Morón FJ, Rubinsztein DC, Eiriksdottir G, Sleegers K, Goate AM, Fiévet N, Huentelman MW, Gill M, Brown K, Kamboh MI, Keller L, Barberger‐Gateau P, McGuiness B, Larson EB, Green R, Myers AJ, Dufouil C, Todd S, Wallon D, Love S, Rogaeva E, Gallacher J, St George‐Hyslop P, Clarimon J, Lleo A, Bayer A, Tsuang DW, Yu L, Tsolaki M, Bossù P, Spalletta G, Proitsi P, Collinge J, Sorbi S, Sanchez‐Garcia F, Fox NC, Hardy J, Deniz Naranjo MC, Bosco P, Clarke R, Brayne C, Galimberti D, Mancuso M, Matthews F; European Alzheimer's Disease Initiative (EADI); Genetic and Environmental Risk in Alzheimer's Disease; Alzheimer's Disease Genetic Consortium; Cohorts for Heart and Aging Research in Genomic Epidemiology , Moebus S, Mecocci P, Del Zompo M, Maier W, Hampel H, Pilotto A, Bullido M, Panza F, Caffarra P, Nacmias B, Gilbert JR, Mayhaus M, Lannefelt L, Hakonarson H, Pichler S, Carrasquillo MM, Ingelsson M, Beekly D, Alvarez V, Zou F, Valladares O, Younkin SG, Coto E, Hamilton‐Nelson KL, Gu W, Razquin C, Pastor P, Mateo I, Owen MJ, Faber KM, Jonsson PV, Combarros O, O'Donovan MC, Cantwell LB, Soininen H, Blacker D, Mead S, Mosley TH Jr, Bennett DA, Harris TB, Fratiglioni L, Holmes C, de Bruijn RF, Passmore P, Montine TJ, Bettens K, Rotter JI, Brice A, Morgan K, Foroud TM, Kukull WA, Hannequin D, Powell JF, Nalls MA, Ritchie K, Lunetta KL, Kauwe JS, Boerwinkle E, Riemenschneider M, Boada M, Hiltuenen M, Martin ER, Schmidt R, Rujescu D, Wang LS, Dartigues JF, Mayeux R, Tzourio C, Hofman A, Nöthen MM, Graff C, Psaty BM, Jones L, Haines JL, Holmans PA, Lathrop M, Pericak‐Vance MA, Launer LJ, Farrer LA, van Duijn CM, Van Broeckhoven C, Moskvina V, Seshadri S, Williams J, Schellenberg GD, Amouyel P. Meta‐analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease. Nat Genet. 2013;45:1452–1458. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Nikpay M, Goel A, Won H‐H, Hall LM, Willenborg C, Kanoni S, Saleheen D, Kyriakou T, Nelson CP, Hopewell JC, Webb TR, Zeng L, Dehghan A, Alver M, Armasu SM, Auro K, Bjonnes A, Chasman DI, Chen S, Ford I, Franceschini N, Gieger C, Grace C, Gustafsson S, Huang J, Hwang SJ, Kim YK, Kleber ME, Lau KW, Lu X, Lu Y, Lyytikäinen LP, Mihailov E, Morrison AC, Pervjakova N, Qu L, Rose LM, Salfati E, Saxena R, Scholz M, Smith AV, Tikkanen E, Uitterlinden A, Yang X, Zhang W, Zhao W, de Andrade M, de Vries PS, van Zuydam NR, Anand SS, Bertram L, Beutner F, Dedoussis G, Frossard P, Gauguier D, Goodall AH, Gottesman O, Haber M, Han BG, Huang J, Jalilzadeh S, Kessler T, König IR, Lannfelt L, Lieb W, Lind L, Lindgren CM, Lokki ML, Magnusson PK, Mallick NH, Mehra N, Meitinger T, Memon FU, Morris AP, Nieminen MS, Pedersen NL, Peters A, Rallidis LS, Rasheed A, Samuel M, Shah SH, Sinisalo J, Stirrups KE, Trompet S, Wang L, Zaman KS, Ardissino D, Boerwinkle E, Borecki IB, Bottinger EP, Buring JE, Chambers JC, Collins R, Cupples LA, Danesh J, Demuth I, Elosua R, Epstein SE, Esko T, Feitosa MF, Franco OH, Franzosi MG, Granger CB, Gu D, Gudnason V, Hall AS, Hamsten A, Harris TB, Hazen SL, Hengstenberg C, Hofman A, Ingelsson E, Iribarren C, Jukema JW, Karhunen PJ, Kim BJ, Kooner JS, Kullo IJ, Lehtimäki T, Loos RJF, Melander O, Metspalu A, März W, Palmer CN, Perola M, Quertermous T, Rader DJ, Ridker PM, Ripatti S, Roberts R, Salomaa V, Sanghera DK, Schwartz SM, Seedorf U, Stewart AF, Stott DJ, Thiery J, Zalloua PA, O'Donnell CJ, Reilly MP, Assimes TL, Thompson JR, Erdmann J, Clarke R, Watkins H, Kathiresan S, McPherson R, Deloukas P, Schunkert H, Samani NJ, Farrall M. A comprehensive 1000 Genomes–based genome‐wide association meta‐analysis of coronary artery disease. Nat Genet. 2015;47:1121–1130. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. UK Biobank—Neale lab[Internet]. 2018. Available at: http://www.nealelab.is/uk-biobank/. Accessed December 14, 2018.
26. Malik R, Chauhan G, Traylor M, Sargurupremraj M, Okada Y, Mishra A, Rutten‐Jacobs L, Giese AK, van der Laan SW, Gretarsdottir S, Anderson CD, Chong M, Adams HHH, Ago T, Almgren P, Amouyel P, Ay H, Bartz TM, Benavente OR, Bevan S, Boncoraglio GB, Brown RD Jr, Butterworth AS, Carrera C, Carty CL, Chasman DI, Chen WM, Cole JW, Correa A, Cotlarciuc I, Cruchaga C, Danesh J, de Bakker PIW, DeStefano AL, den Hoed M, Duan Q, Engelter ST, Falcone GJ, Gottesman RF, Grewal RP, Gudnason V, Gustafsson S, Haessler J, Harris TB, Hassan A, Havulinna AS, Heckbert SR, Holliday EG, Howard G, Hsu FC, Hyacinth HI, Ikram MA, Ingelsson E, Irvin MR, Jian X, Jiménez‐Conde J, Johnson JA, Jukema JW, Kanai M, Keene KL, Kissela BM, Kleindorfer DO, Kooperberg C, Kubo M, Lange LA, Langefeld CD, Langenberg C, Launer LJ, Lee JM, Lemmens R, Leys D, Lewis CM, Lin WY, Lindgren AG, Lorentzen E, Magnusson PK, Maguire J, Manichaikul A, McArdle PF, Meschia JF, Mitchell BD, Mosley TH, Nalls MA, Ninomiya T, O'Donnell MJ, Psaty BM, Pulit SL, Rannikmäe K, Reiner AP, Rexrode KM, Rice K, Rich SS, Ridker PM, Rost NS, Rothwell PM, Rotter JI, Rundek T, Sacco RL, Sakaue S, Sale MM, Salomaa V, Sapkota BR, Schmidt R, Schmidt CO, Schminke U, Sharma P, Slowik A, Sudlow CLM, Tanislav C, Tatlisumak T, Taylor KD, Thijs VNS, Thorleifsson G, Thorsteinsdottir U, Tiedt S, Trompet S, Tzourio C, van Duijn CM, Walters M, Wareham NJ, Wassertheil‐Smoller S, Wilson JG, Wiggins KL, Yang Q, Yusuf S; AFGen Consortium; Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium; International Genomics of Blood Pressure (iGEN‐BP) Consortium; INVENT Consortium; STARNET , Bis JC, Pastinen T, Ruusalepp A, Schadt EE, Koplev S, Björkegren JLM, Codoni V, Civelek M, Smith NL, Trégouët DA, Christophersen IE, Roselli C, Lubitz SA, Ellinor PT, Tai ES, Kooner JS, Kato N, He J, van der Harst P, Elliott P, Chambers JC, Takeuchi F, Johnson AD; BioBank Japan Cooperative Hospital Group; COMPASS Consortium; EPIC‐CVD Consortium; EPIC‐InterAct Consortium; International Stroke Genetics Consortium (ISGC); METASTROKE Consortium; Neurology Working Group of the CHARGE Consortium; NINDS Stroke Genetics Network (SiGN); UK Young Lacunar DNA Study; MEGASTROKE Consortium , Sanghera DK, Melander O, Jern C, Strbian D, Fernandez‐Cadenas I, Longstreth WT Jr, Rolfs A, Hata J, Woo D, Rosand J, Pare G, Hopewell JC, Saleheen D, Stefansson K, Worrall BB, Kittner SJ, Seshadri S, Fornage M, Markus HS, Howson JMM, Kamatani Y, Debette S, Dichgans M. Multiancestry genome‐wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes. Nat Genet. 2018;50:524–537.29531354 [Google Scholar]
27. Brion M‐JA, Shakhbazov K, Visscher PM. Calculating statistical power in Mendelian randomization studies. Int J Epidemiol. 2013;42:1497–1501. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Didelez V, Meng S, Sheehan NA. Assumptions of IV methods for observational epidemiology. Stat Sci. 2010;25:22–40. [Google Scholar]
29. Thompson J, Minelli C. & Del Greco M.F. Mendelian Randomization using Public Data from Genetic Consortia. The International Journal of Biostatistics. 2016;12(2). Retrieved 26 Jul. 2019. DOI: 10.1515/ijb-2015-0074. [DOI] [PubMed] [Google Scholar]
30. Burgess S, Bowden J. Integrating summarized data from multiple genetic variants in Mendelian randomization: bias and coverage properties of inverse‐variance weighted methods. 2015; arXiv:1512.04486.
31. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44:512–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. 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] [PMC free article] [PubMed] [Google Scholar]
33. 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;50:693–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Burgess S, Thompson SG. Multivariable Mendelian randomization: the use of pleiotropic genetic variants to estimate causal effects. Am J Epidemiol. 2015;181:251–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Yengo L, Sidorenko J, Kemper KE, Zheng Z, Wood AR, Weedon MN, Frayling TM, Hirschhorn J, Yang J, Visscher PM; GIANT Consortium . Meta‐analysis of genome‐wide association studies for height and body mass index in ∼700000 individuals of European ancestry. Hum Mol Genet. 2018;27:3641–3649. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Staley JR, Blackshaw J, Kamat MA, Ellis S, Surendran P, Sun BB, Paul DS, Freitag D, Burgess S, Danesh J, Young R, Butterworth AS. PhenoScanner: a database of human genotype–phenotype associations. Bioinformatics. 2016;32:3207–3209. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Yavorska OO, Burgess S. MendelianRandomization: an R package for performing Mendelian randomization analyses using summarized data. Int J Epidemiol. 2017;46:1734–1739. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Qu H‐Q, Tien M, Polychronakos C. Statistical significance in genetic association studies. Clin Invest Med. 2010;33:E266–E270. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Khan AA, Quinn TJ, Hewitt J, Fan Y, Dawson J. Serum uric acid level and association with cognitive impairment and dementia: systematic review and meta‐analysis. Age (Dordr). 2016;38:16. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Li X, Meng X, Timofeeva M, Tzoulaki I, Tsilidis KK, Ioannidis PA, Campbell H, Theodoratou E. Serum uric acid levels and multiple health outcomes: umbrella review of evidence from observational studies, randomised controlled trials, and Mendelian randomisation studies. BMJ. 2017;357:j2376. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Soletsky B, Feig DI. Uric acid reduction rectifies prehypertension in obese adolescents. Hypertension. 2012;60:1148–1156. [DOI] [PubMed] [Google Scholar]
42. Feig DI, Soletsky B, Johnson RJ. Effect of allopurinol on blood pressure of adolescents with newly diagnosed essential hypertension: a randomized trial. JAMA. 2008;300:924–932. [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Kao MP, Ang DS, Gandy SJ, Nadir MA, Houston JG, Lang CC, Struthers AD. Allopurinol benefits left ventricular mass and endothelial dysfunction in chronic kidney disease. J Am Soc Nephrol. 2011;22:1382–1389. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Noman A, Ang DS, Ogston S, Lang CC, Struthers AD. Effect of high‐dose allopurinol on exercise in patients with chronic stable angina: a randomised, placebo controlled crossover trial. Lancet. 2010;375:2161–2167. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Hughes K, Flynn T, de Zoysa J, Dalbeth N, Merriman TR. Mendelian randomization analysis associates increased serum urate, due to genetic variation in uric acid transporters, with improved renal function. Kidney Int. 2014;85:344–351. [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Liu P, Wang H, Zhang F, Chen Y, Wang D, Wang Y. The effects of allopurinol on the carotid intima‐media thickness in patients with type 2 diabetes and asymptomatic hyperuricemia: a three‐year randomized parallel‐controlled study. Intern Med. 2015;54:2129–2137. [DOI] [PubMed] [Google Scholar]
47. Szwejkowski BR, Gandy SJ, Rekhraj S, Houston JG, Lang CC, Morris AD, George J, Struthers AD. Allopurinol reduces left ventricular mass in patients with type 2 diabetes and left ventricular hypertrophy. J Am Coll Cardiol. 2013;62:2284–2293. [DOI] [PubMed] [Google Scholar]
48. Frieden TR. Evidence for health decision making—beyond randomized, controlled trials. N Engl J Med. 2017;377:465–475. [DOI] [PubMed] [Google Scholar]
49. Davey Smith G, Ebrahim S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease?*. Int J Epidemiol. 2003;32:1–22. [DOI] [PubMed] [Google Scholar]
50. 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. [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Burgess S, Butterworth A, Malarstig A, Thompson SG. Use of Mendelian randomisation to assess potential benefit of clinical intervention. BMJ. 2012;345:e7325. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials