Original Article

Split Viewer

J Rheum Dis 2020; 27(2): 88-95

Published online April 1, 2020

© Korean College of Rheumatology

The Uric Acid and Gout have No Direct Causality With Osteoarthritis: A Mendelian Randomization Study

Young Ho Lee, M.D., Ph.D., Gwan Gyu Song, M.D., Ph.D.

Department of Rheumatology, Korea University College of Medicine, Seoul, Korea

Correspondence to : Young Ho Lee http://orcid.org/0000-0003-4213-1909
Department of Rheumatology, Korea University Anam Hospital, Korea University College of Medicine, 73 Inchon-ro, Seongbuk-gu, Seoul 02841, Korea.
E-mail:lyhcgh@korea.ac.kr

Received: August 20, 2019; Revised: October 24, 2019; Accepted: November 15, 2019

This is an Open Access article, which permits unrestricted non-commerical use, distribution, and reproduction in any medium, provided the original work is properly cited.

Objective. To examine whether uric acid level or gout is causally associated with the risk of osteoarthritis. Methods. We performed a two-sample Mendelian randomization (MR) analysis using inverse-variance weighted (IVW), MR-Egger regression, and weighted median methods. We used the publicly available summary statistics datasets of uric acid level or gout genome-wide association studies (GWASs) as the exposure, and a GWAS in 3,498 patients with osteoarthritis in the arcOGEN study and 11,009 controls of European ancestry as the outcome. Results. Six single nucleotide polymorphisms (SNPs) from the GWAS data on uric acid level and 12 SNPs from the GWAS data on gout were selected as instrumental variables (IVs). The IVW analysis did not support a causal association between uric acid level or gout and risk of osteoarthritis (beta=–0.026, standard error [SE]=0.096, p=0.789; beta=–0.018, SE=0.025, p=0.482). MR-Egger regression revealed no causal association between uric acid level or gout and risk of osteoarthritis (beta=0.028, SE=0.142, p=0.852; beta=–0.056, SE=0.090, p=0.548). Similarly, no evidence of a casual association was provided by the weighted median approach (beta=0.004, SE=0.064, p=0.946; beta=–0.005, SE=0.025, p=0.843). Conclusion. The results of MR analysis demonstrates that uric acid level and gout may be not causally associated with the increased risk of osteoarthritis. Considering MR study is not susceptible to bias from unmeasured confounders or reverse causation, the epidemiological evidence for an association between uric acid level or gout and a higher risk of osteoarthritis may be due to residual confounding or reverse causation rather than direct causality.

Keywords Uric acid, Gout, Osteoarthritis, Mendelian randomization analysis

Osteoarthritis is the most common disease of joints, characterized by a progressive degeneration of articular cartilage, joint pain, and immobility [1]. It is a major cause of pain and disability in elderly people, and its prevalence continues to increase worldwide [2]. The health burden of osteoarthritis is growing with increase in the average longevity worldwide. Therefore, a comprehensive understanding of the risk factors associated with the occurrence of osteoarthritis is needed. Hyperuricemia is a condition in which the serum urate level exceeds the limit of urate solubility (6.8 mg/dL), which reflects supersaturation of the extracellular fluid with urate, and predisposes the body to gout [3]. Gout is an inflammatory disorder characterized by hyperuricemia and effects of urate crystal deposition, including episodic gout flares, gouty arthropathy, tophi, and urolithiasis [3]. Hyperuricemia leads to the deposition of monosodium urate (MSU) crystals in tissues, and increase in the levels of hyperuricemia is correlated with increasing incidences of gouty arthritis [4].

Epidemiological studies have demonstrated that higher serum uric acid levels and gout are associated with an increased risk of osteoarthritis, suggesting uric acid and gout as possible risk factors for osteoarthritis [5,6]. The relationship among hyperuricemia, gout, and osteoarthritis may be explained by the finding that MSU crystal deposition on/in cartilage may create local mechanical and/or inflammatory damage, promoting osteoarthritis development, and that uric acid transporters are present on human articular chondrocytes, suggesting that these cells may internalize uric acid with potential pro-oxidant effects [7]. However, the shared risk factors due to obesity and aging represent a common feature among hyperuricemia, gout and osteoarthritis [8]. Hyperuricemia and gout may predispose joints to osteoarthritis, and osteoarthritis-caused cartilage damage may promote MSU deposition; further these conditions may trigger or result in shared inflammatory cascades. Observational studies are prone to biases, such as reverse causation interpretations and residual confounding, preventing a clear understanding of the effects of uric acid level and gout on osteoarthritis.

Mendelian randomization (MR) is a technique that uses genetic variants as instrumental variables (IVs) to assess whether an observational association between a risk factor and an outcome is consistent with a causal effect [9]. However, to date, MR has not been used to explore the causal effects of uric acid level and gout on the risk of osteoarthritis. The aim of this study was to examine, using a two-sample MR analysis, whether uric acid level and gout are causally associated with the risk of osteoarthritis.

Data sources and selection of genetic variants

We searched NHGRI-EBI GWAS (https://www.ebi.ac.uk/gwas), a comprehensive catalog of reported associations from published GWAS studies. We used publicly available summary statistics datasets from uric acid level or gout GWASs. For the exposure dataset, we used publicly available summary statistics datasets from a meta-analysis of GWASs on uric acid levels from 14 studies with a total 28,141 participants of European descent [10], and three GWASs on gout (European: n=968 cases and 15,506 controls [11]; Japanese: n=945 cases and 1,213 controls, followed by 1,048 cases and 1,334 controls [12]), and replicated the results using a dataset from a GWAS with 1,396 cases and 1,268 controls from Caucasian and New Zealand Polynesian populations [13]. We used summary statistics from a GWAS based on 3,498 patients with osteoarthritis of the knee and hip in the arcOGEN study and 11,009 controls from the UK or European descent, as an outcome [14]. A two-sample MR study uses genetic variants associated with uric acid level or gout as IVs to improve the inference. We obtained summary statistics (beta coefficients and standard errors [SE]) for single-nucleotide polymorphisms (SNPs) associated with uric acid level and gout as IVs from uric acid level and gout GWASs, and utilized the summary data of SNPs from osteoarthritis GWAS as an outcome. The ‘top hits’ from a GWAS can be used to define genetic instruments for exposures in MR analyses. The selection criteria for IVs were as follows; 1) Association p-values of SNPs with exposure should be less than a threshold of p<2.04e-05 in GWAS dataset of exposure. 2) The SNPs from exposure dataset also should be present in GWAS dataset of outcome.

Statistical analysis for Mendelian randomization

MR analysis requires genetic variants to be related to, but not potential confounders of, an exposure [15]. First, we assessed the independent association of SNPs with uric acid level or gout. Second, we examined the association between each SNP and the risk of osteoarthritis. Third, we combined these findings to estimate the uncompounded causal association between uric acid level or gout and osteoarthritis risk using MR analysis. We performed two-sample MR, which is a method used to estimate the causal effect of an exposure (uric acid level and gout) on outcomes (osteoarthritis) using summary statistics from different GWASs [16], to assess the causal relationships between uric acid level or gout and osteoarthritis risk.

The IVW method uses a meta-analysis approach to combine the Wald ratio estimates of the causal effect obtained from different SNPs, and provides a consistent estimate of the causal effect of the exposure on the outcome when each of the genetic variants satisfies the assumptions of an instrumental variable [17]. Although the inclusion of multiple variants in an MR analysis results in increased statistical power, it has the potential to include pleiotropic genetic variants that are not valid IVs [16]. To explore and adjust for pleiotropy (association of genetic variants with more than one variable), the weighted median and MR-Egger regression methods were used. MR-Egger regression analysis tests and accounts for the presence of unbalanced pleiotropy by introducing a parameter for this bias, through incorporation of summary data estimates of causal effects from multiple individual variants, making it robust to invalid instruments [18]. MR-Egger method performs a weighted linear regression of the gene-outcome coefficients on the gene-exposure coefficients [18]. The slope of this regression represents the causal effect estimate, and the intercept can be interpreted as an estimate of the average horizontal pleiotropic effect across the genetic variants [19]. The weighted median estimator provides a consistent estimate of the causal effect, even when up to 50% of the information contributing to the analysis comes from genetic variants that are invalid IVs [20]. The weighted median estimator has the advantage of retaining greater precision in the estimates than those from the MR-Egger analysis [20]. Tests were considered statistically significant at p<0.05. All MR analyses were performed on MR Base platform [21].

Heterogeneity and sensitivity test

We assessed heterogeneities among SNPs using Cochran’s Q-statistics and funnel plots [22]. We also performed a “leave one out” analysis to investigate the possibility that the causal association was driven by a single SNP.

Studies included in the meta-analysis

1) Instrumental variables for Mendelian randomization

We selected six SNPs as IVs from the GWAS data on uric acid level and 12 SNPs from the GWAS data on gout as IVs to improve the inference (Table 1, Figure 1). Two of the IVs (rs2231142, rs734553) were associated with both uric acid level and gout (Table 1). The selected IVs were associated with uric acid level and gout at genome-wide significance, except for rs10791821, 11733284, and rs734553 in gout (Table 1).

Table 1 . Instrumental SNPs from uric acid level (A) or gout (B) and osteoarthritis GWASs

A. Uric acid levels

Instrumental SNPEffect alleleGeneExposure (uric acid)Outcome (osteoarthritis)


BetaSEp-valueCase (n)Control (n)BetaSEp-value
rs12356193ASLC16A90.0800.0141.00×10−83,49811,0090.0100.0290.657
rs17300741ASLC22A110.0600.0087.00×10−143,49811,009−0.0200.0220.259
rs2231142TABCG20.1400.0221.00×10−103,49811,009−0.0940.0340.006
rs734553TSLC2A90.4000.0131.00×10−1923,49811,0090.0100.0260.679
rs742132ASCGN0.0500.0099.00×10−93,49811,0090.0300.0210.169
rs780094TGCKR0.0500.0081.00×10−93,49811,009−0.0300.0200.190

B. Gout

Instrumental SNPEffect alleleGeneExposure (gout)Outcome (osteoarthritis)


BetaSEp-valueCase (n)Control (n)BetaSEp-value

rs1014290TSLC2A90.4510.0427.00×10−263,49811,0090.0410.0240.075
rs10791821GMAP3K110.4510.0841.00×10−73,49811,0090.0000.0260.973
rs1165176GSLC17A10.3510.0571.00×10−93,49811,0090.0100.0210.717
rs11733284ANIPAL10.2150.0459.00×10−73,49811,009−0.0410.0240.099
rs11758351GHIST1H2BF0.3360.0582.00×10−83,49811,0090.0490.0310.088
rs1260326TGCKR0.3070.0432.00×10−123,49811,009−0.0410.0240.089
rs2231142TABCG20.5130.0753.00×10−123,49811,009−0.0940.0340.006
rs2728125CABCG20.7130.0467.00×10−543,49811,009−0.0730.0360.043
rs3114020CABCG20.6370.0519.00×10−353,49811,009−0.0300.0210.193
rs4073582GCNIH-20.5070.0866.00×10−93,49811,0090.0200.0240.259
rs4766566TCUX20.4120.0464.00×10−203,49811,0090.0300.0260.201
rs734553TSLC2A90.3290.0652.00×10−73,49811,0090.0100.0260.652

SNP: single nucleotide polymorphism, GWAS: genome-wide association study, Beta: beta coefficient, SE: standard error, SLC16A9: Solute Carrier Family 16 Member 9, SLC22A11: Solute Carrier Family 22 Member 11, ABCG2: ATP Binding Cassette Subfamily G Member 2 (Junior Blood Group), SLC2A9: Solute Carrier Family 2 Member 9, SCGN: Secretagogin, EF-Hand Calcium Binding Protein, GCKR: Glucokinase Regulator, MAP3K11: Mitogen-Activated Protein Kinase Kinase Kinase 11, SLC17A1: Solute Carrier Family 17 Member 1, NIPAL1: NIPA-Like Domain Containing 1, HIST1H2BF: Histone Cluster 1 H2B Family Member F, CNIH-2: Cornichon Family AMPA Receptor Auxiliary Protein 2, CUX2: Cut-Like Homeobox 2.


Fig. 1. Forrest plot of the causal effects of uric acid level (A) or gout (B)-associated SNPs on osteoarthritis. SNP: single nucleotide polymorphism, MR: Mendelian randomization, IVW: inverse-variance weighted, Na: not available.

2) Mendelian randomization results

The IVW method yielded no evidence to support a causal association between uric acid level or gout and the risk of osteoarthritis (beta=−0.026, SE=0.096, p=0.789; beta=−0.018, SE=0.025, p=0.482) (Table 2, Figures 1, 2). In the MR-Egger test, the intercept represents the average pleiotropic effect across the genetic variants, i.e., the average direct effect of a variant on the outcome. If the intercept differs from zero, it serves as an evidence of directional pleiotropy. MR-Egger regression revealed that directional pleiotropy was unlikely to be biasing uric acid level (intercept=−0.013, p=0.610) or uric acid results (intercept=0.018, p=0.668), while yielding no causal association between uric acid level or gout and the risk of osteoarthritis (beta=0.028, SE=0.142, p=0.852; beta=−0.056, SE=0.090, p=0.548) (Table 2, Figures 1, 2). In addition, the weighted median approach provided no evidence of a causal association between uric acid level or gout and the risk of osteoarthritis (beta=0.004, SE=0.064, p=0.946; beta=−0.005, SE=0.025, p=0.843) (Table 2, Figures 1, 2). We performed subgroup analysis using exposure dataset of European. However, the MR result from European did not change significantly the overall MR result.

Table 2 . The MR estimates of the causal effect of uric acid level (A) and gout (B) on osteoarthritis risk, derived using different methods

A. Uric acid level

MR methodNumber of SNPBetaSEAssociation p-valueCochran Q statisticHeterogeneity p-value
Inverse variance weighted6−0.0260.0960.78913.020.232
MR-Egger60.0280.1420.85212.090.166
Weighted median60.0040.0640.946nana

B. Gout

MR methodNumber of SNPBetaSEAssociation p-valueCochran Q statisticHeterogeneity p-value

Inverse variance weighted12−0.0180.0250.48226.530.005
MR-Egger12−0.0560.0900.54826.030.004
Weighted median12−0.0050.0250.834nana

MR: Mendelian randomization, SNP: single nucleotide polymorphism, Beta: beta coefficient, SE: standard error, na: not available.


Fig. 2. Scatter plots of genetic associations of uric acid level (A) or gout (B) against the genetic associations of osteoarthritis. The slopes of each line represent the causal association for each method. Blue line represents the IVW estimate, green line represents the weighted median estimate, and dark blue line represents the MR-Egger estimate. IVW: inverse-variance weighted, SNP: single nucleotide polymorphism, MR: Mendelian randomization, Na: not available.

3) Heterogeneity and sensitivity test

Cochran’s Q-test indicated no evidence of heterogeneity between IV estimates based on the individual variants of uric acid level, but not gout (Table 2). The funnel test yielded symmetry, indicating no evidence of heterogeneity in the MR analyses of uric acid level and gout (Figure 3). Results from the “leave one out” analysis demonstrated that no single SNP was driving the IVW point estimate.

Fig. 3. Funnel plot to assess the heterogeneity of the causal effects of uric acid level (A) or gout (B)-associated SNPs on osteoarthritis. Blue line represents the IVW estimate, and dark blue line represents the MR-Egger estimate. SNP: single nucleotide polymorphism, IVW: inverse-variance weighted, MR: Mendelian randomization, SE: standard error, β: beta coefficient.

In conclusion, the results of our MR analysis demonstrates that uric acid level and gout may be not causally associated with the increased risk of osteoarthritis. Considering MR study is not susceptible to bias from unmeasured confounders or reverse causation, the epidemiological evidence for an association between uric acid level or gout and a higher risk of osteoarthritis may be due to residual confounding or reverse causation rather than direct causality. Well-designed epidemiological and MR studies using more variants that explain a greater proportion of uric acid level or gout are warranted to confirm or rule out its causal relationship with osteoarthritis.

No potential conflict of interest relevant to this article was reported.

Y.H.L. was involved in conception and design of study, acquisition of data, analysis and interpretation of data, drafting the manuscript, and revising the manuscript. G.G.S. was involved in conception and design of study, analysis and interpretation of data, and drafting the manuscript.

  1. Dieppe PA, Lohmander LS. Pathogenesis and management of pain in osteoarthritis. Lancet 2005;365:965-73.
    Pubmed CrossRef
  2. Litwic A, Edwards MH, Dennison EM, Cooper C. Epidemiology and burden of osteoarthritis. Br Med Bull 2013;105:185-99.
    Pubmed KoreaMed CrossRef
  3. Wortmann RL. Gout and hyperuricemia. Curr Opin Rheumatol 2002;14:281-6.
    Pubmed CrossRef
  4. Terkeltaub RA. Clinical practice. Gout. N Engl J Med 2003;349:1647-55.
    Pubmed CrossRef
  5. Ding X, Zeng C, Wei J, Li H, Yang T, Zhang Y, et al. The associations of serum uric acid level and hyperuricemia with knee osteoarthritis. Rheumatol Int 2016;36:567-73.
    Pubmed CrossRef
  6. Howard RG, Samuels J, Gyftopoulos S, Krasnokutsky S, Leung J, Swearingen CJ, et al. Presence of gout is associated with increased prevalence and severity of knee osteoarthritis among older men: results of a pilot study. J Clin Rheumatol 2015;21:63-71.
    Pubmed KoreaMed CrossRef
  7. Dalbeth N, Aati O, Kalluru R, Gamble GD, Horne A, Doyle AJ, et al. Relationship between structural joint damage and urate deposition in gout: a plain radiography and dual-energy CT study. Ann Rheum Dis 2015;74:1030-6.
    Pubmed CrossRef
  8. Yokose C, Chen M, Berhanu A, Pillinger MH, Krasnokutsky S. Gout and osteoarthritis: associations, pathophysiology, and therapeutic implications. Curr Rheumatol Rep 2016;18:65.
    Pubmed CrossRef
  9. Burgess S, Daniel RM, Butterworth AS, Thompson SG; EPIC-InterAct Consortium. Network Mendelian randomization: using genetic variants as instrumental variables to investigate mediation in causal pathways. Int J Epidemiol 2015;44:484-95.
    Pubmed KoreaMed CrossRef
  10. Kolz M, Johnson T, Sanna S, Teumer A, Vitart V, Perola M, et al. Meta-analysis of 28,141 individuals identifies common variants within five new loci that influence uric acid concentrations. PLoS Genet 2009;5:e1000504.
    Pubmed KoreaMed CrossRef
  11. Sulem P, Gudbjartsson DF, Walters GB, Helgadottir HT, Helgason A, Gudjonsson SA, et al. Identification of low-frequency variants associated with gout and serum uric acid levels. Nat Genet 2011;43:1127-30.
    Pubmed CrossRef
  12. Matsuo H, Yamamoto K, Nakaoka H, Nakayama A, Sakiyama M, Chiba T, et al. Genome-wide association study of clinically defined gout identifies multiple risk loci and its association with clinical subtypes. Ann Rheum Dis 2016;75:652-9.
    Pubmed KoreaMed CrossRef
  13. Nakayama A, Nakaoka H, Yamamoto K, Sakiyama M, Shaukat A, Toyoda Y, et al. GWAS of clinically defined gout and subtypes identifies multiple susceptibility loci that include urate transporter genes. Ann Rheum Dis 2017;76:869-77.
    Pubmed KoreaMed CrossRef
  14. Zeggini E, Panoutsopoulou K, Southam L, Rayner NW, Day-Williams AG, Lopes MC, et al; arcOGEN Consortium, arcOGEN Collaborators. Identification of new susceptibility loci for osteoarthritis (arcOGEN): a genome-wide association study. Lancet 2012;380:815-23.
    Pubmed KoreaMed CrossRef
  15. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013;37:658-65.
    Pubmed KoreaMed CrossRef
  16. Hartwig FP, Davies NM, Hemani G, Davey Smith G. Two-sample Mendelian randomization: avoiding the downsides of a powerful, widely applicable but potentially fallible technique. Int J Epidemiol 2016;45:1717-26.
    Pubmed KoreaMed CrossRef
  17. Pierce BL, Burgess S. Efficient design for Mendelian randomization studies: subsample and 2-sample instrumental variable estimators. Am J Epidemiol 2013;178:1177-84.
    Pubmed KoreaMed CrossRef
  18. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015;44:512-25.
    Pubmed KoreaMed CrossRef
  19. Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol 2017;32:377-89.
    Pubmed KoreaMed CrossRef
  20. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol 2016;40:304-14.
    Pubmed KoreaMed CrossRef
  21. Hemani G, Zheng J, Wade KH, Laurin C, Elsworth B, Burgess S, et al. MR-Base: a platform for systematic causal inference across the phenome using billions of genetic associations. bioRxiv 2016;16:78972.
    CrossRef
  22. Egger M, Smith GD, Phillips AN. Meta-analysis: principles and procedures. BMJ 1997;315:1533-7.
    Pubmed KoreaMed CrossRef
  23. Neogi T, Krasnokutsky S, Pillinger MH. Urate and osteoarthritis: Evidence for a reciprocal relationship. Joint Bone Spine 2019;86:576-82.
    Pubmed CrossRef
  24. Hill HA, Kleinbaum DG. Bias in observational studies. Wiley StatsRef . DOI: 10.1002/9781118445112.stat05111.
    CrossRef
  25. Lee YH, Bae SC, Song GG. Hepatitis B virus (HBV) reactivation in rheumatic patients with hepatitis core antigen (HBV occult carriers) undergoing anti-tumor necrosis factor therapy. Clin Exp Rheumatol 2013;31:118-21.
    Pubmed
  26. Smith GD, Ebrahim S. Mendelian randomization: genetic variants as instruments for strengthening causal inference in observational studies. In: Weinstein M, Vaupel JW, Wachter KW, eds. Biosocial surveys. Washington D.C., National Academies Press (US), 2008.
  27. VanderWeele TJ, Tchetgen Tchetgen EJ, Cornelis M, Kraft P. Methodological challenges in Mendelian randomization. Epidemiology 2014;25:427-35.
    Pubmed KoreaMed CrossRef
  28. Thompson JR, Minelli C, Bowden J, Del Greco FM, Gill D, Jones EM, et al. Mendelian randomization incorporating uncertainty about pleiotropy. Stat Med 2017;36:4627-45.
    Pubmed CrossRef
  29. Paaby AB, Rockman MV. The many faces of pleiotropy. Trends Genet 2013;29:66-73.
    Pubmed KoreaMed CrossRef
  30. Smith GD, Ebrahim S. Mendelian randomization: prospects, potentials, and limitations. Int J Epidemiol 2004;33:30-42.
    Pubmed CrossRef
  31. Swerdlow DI, Kuchenbaecker KB, Shah S, Sofat R, Holmes MV, White J, et al. Selecting instruments for Mendelian randomization in the wake of genome-wide association studies. Int J Epidemiol 2016;45:1600-16.
    Pubmed KoreaMed CrossRef
  32. Burgess S, Dudbridge F, Thompson SG. Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods. Stat Med 2016;35:1880-906.
    Pubmed KoreaMed CrossRef

Article

Original Article

J Rheum Dis 2020; 27(2): 88-95

Published online April 1, 2020 https://doi.org/10.4078/jrd.2020.27.2.88

Copyright © Korean College of Rheumatology.

The Uric Acid and Gout have No Direct Causality With Osteoarthritis: A Mendelian Randomization Study

Young Ho Lee, M.D., Ph.D., Gwan Gyu Song, M.D., Ph.D.

Department of Rheumatology, Korea University College of Medicine, Seoul, Korea

Correspondence to:Young Ho Lee http://orcid.org/0000-0003-4213-1909
Department of Rheumatology, Korea University Anam Hospital, Korea University College of Medicine, 73 Inchon-ro, Seongbuk-gu, Seoul 02841, Korea.
E-mail:lyhcgh@korea.ac.kr

Received: August 20, 2019; Revised: October 24, 2019; Accepted: November 15, 2019

This is an Open Access article, which permits unrestricted non-commerical use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Objective. To examine whether uric acid level or gout is causally associated with the risk of osteoarthritis. Methods. We performed a two-sample Mendelian randomization (MR) analysis using inverse-variance weighted (IVW), MR-Egger regression, and weighted median methods. We used the publicly available summary statistics datasets of uric acid level or gout genome-wide association studies (GWASs) as the exposure, and a GWAS in 3,498 patients with osteoarthritis in the arcOGEN study and 11,009 controls of European ancestry as the outcome. Results. Six single nucleotide polymorphisms (SNPs) from the GWAS data on uric acid level and 12 SNPs from the GWAS data on gout were selected as instrumental variables (IVs). The IVW analysis did not support a causal association between uric acid level or gout and risk of osteoarthritis (beta=–0.026, standard error [SE]=0.096, p=0.789; beta=–0.018, SE=0.025, p=0.482). MR-Egger regression revealed no causal association between uric acid level or gout and risk of osteoarthritis (beta=0.028, SE=0.142, p=0.852; beta=–0.056, SE=0.090, p=0.548). Similarly, no evidence of a casual association was provided by the weighted median approach (beta=0.004, SE=0.064, p=0.946; beta=–0.005, SE=0.025, p=0.843). Conclusion. The results of MR analysis demonstrates that uric acid level and gout may be not causally associated with the increased risk of osteoarthritis. Considering MR study is not susceptible to bias from unmeasured confounders or reverse causation, the epidemiological evidence for an association between uric acid level or gout and a higher risk of osteoarthritis may be due to residual confounding or reverse causation rather than direct causality.

Keywords: Uric acid, Gout, Osteoarthritis, Mendelian randomization analysis

INTRODUCTION

Osteoarthritis is the most common disease of joints, characterized by a progressive degeneration of articular cartilage, joint pain, and immobility [1]. It is a major cause of pain and disability in elderly people, and its prevalence continues to increase worldwide [2]. The health burden of osteoarthritis is growing with increase in the average longevity worldwide. Therefore, a comprehensive understanding of the risk factors associated with the occurrence of osteoarthritis is needed. Hyperuricemia is a condition in which the serum urate level exceeds the limit of urate solubility (6.8 mg/dL), which reflects supersaturation of the extracellular fluid with urate, and predisposes the body to gout [3]. Gout is an inflammatory disorder characterized by hyperuricemia and effects of urate crystal deposition, including episodic gout flares, gouty arthropathy, tophi, and urolithiasis [3]. Hyperuricemia leads to the deposition of monosodium urate (MSU) crystals in tissues, and increase in the levels of hyperuricemia is correlated with increasing incidences of gouty arthritis [4].

Epidemiological studies have demonstrated that higher serum uric acid levels and gout are associated with an increased risk of osteoarthritis, suggesting uric acid and gout as possible risk factors for osteoarthritis [5,6]. The relationship among hyperuricemia, gout, and osteoarthritis may be explained by the finding that MSU crystal deposition on/in cartilage may create local mechanical and/or inflammatory damage, promoting osteoarthritis development, and that uric acid transporters are present on human articular chondrocytes, suggesting that these cells may internalize uric acid with potential pro-oxidant effects [7]. However, the shared risk factors due to obesity and aging represent a common feature among hyperuricemia, gout and osteoarthritis [8]. Hyperuricemia and gout may predispose joints to osteoarthritis, and osteoarthritis-caused cartilage damage may promote MSU deposition; further these conditions may trigger or result in shared inflammatory cascades. Observational studies are prone to biases, such as reverse causation interpretations and residual confounding, preventing a clear understanding of the effects of uric acid level and gout on osteoarthritis.

Mendelian randomization (MR) is a technique that uses genetic variants as instrumental variables (IVs) to assess whether an observational association between a risk factor and an outcome is consistent with a causal effect [9]. However, to date, MR has not been used to explore the causal effects of uric acid level and gout on the risk of osteoarthritis. The aim of this study was to examine, using a two-sample MR analysis, whether uric acid level and gout are causally associated with the risk of osteoarthritis.

MATERIALS AND METHODS

Data sources and selection of genetic variants

We searched NHGRI-EBI GWAS (https://www.ebi.ac.uk/gwas), a comprehensive catalog of reported associations from published GWAS studies. We used publicly available summary statistics datasets from uric acid level or gout GWASs. For the exposure dataset, we used publicly available summary statistics datasets from a meta-analysis of GWASs on uric acid levels from 14 studies with a total 28,141 participants of European descent [10], and three GWASs on gout (European: n=968 cases and 15,506 controls [11]; Japanese: n=945 cases and 1,213 controls, followed by 1,048 cases and 1,334 controls [12]), and replicated the results using a dataset from a GWAS with 1,396 cases and 1,268 controls from Caucasian and New Zealand Polynesian populations [13]. We used summary statistics from a GWAS based on 3,498 patients with osteoarthritis of the knee and hip in the arcOGEN study and 11,009 controls from the UK or European descent, as an outcome [14]. A two-sample MR study uses genetic variants associated with uric acid level or gout as IVs to improve the inference. We obtained summary statistics (beta coefficients and standard errors [SE]) for single-nucleotide polymorphisms (SNPs) associated with uric acid level and gout as IVs from uric acid level and gout GWASs, and utilized the summary data of SNPs from osteoarthritis GWAS as an outcome. The ‘top hits’ from a GWAS can be used to define genetic instruments for exposures in MR analyses. The selection criteria for IVs were as follows; 1) Association p-values of SNPs with exposure should be less than a threshold of p<2.04e-05 in GWAS dataset of exposure. 2) The SNPs from exposure dataset also should be present in GWAS dataset of outcome.

Statistical analysis for Mendelian randomization

MR analysis requires genetic variants to be related to, but not potential confounders of, an exposure [15]. First, we assessed the independent association of SNPs with uric acid level or gout. Second, we examined the association between each SNP and the risk of osteoarthritis. Third, we combined these findings to estimate the uncompounded causal association between uric acid level or gout and osteoarthritis risk using MR analysis. We performed two-sample MR, which is a method used to estimate the causal effect of an exposure (uric acid level and gout) on outcomes (osteoarthritis) using summary statistics from different GWASs [16], to assess the causal relationships between uric acid level or gout and osteoarthritis risk.

The IVW method uses a meta-analysis approach to combine the Wald ratio estimates of the causal effect obtained from different SNPs, and provides a consistent estimate of the causal effect of the exposure on the outcome when each of the genetic variants satisfies the assumptions of an instrumental variable [17]. Although the inclusion of multiple variants in an MR analysis results in increased statistical power, it has the potential to include pleiotropic genetic variants that are not valid IVs [16]. To explore and adjust for pleiotropy (association of genetic variants with more than one variable), the weighted median and MR-Egger regression methods were used. MR-Egger regression analysis tests and accounts for the presence of unbalanced pleiotropy by introducing a parameter for this bias, through incorporation of summary data estimates of causal effects from multiple individual variants, making it robust to invalid instruments [18]. MR-Egger method performs a weighted linear regression of the gene-outcome coefficients on the gene-exposure coefficients [18]. The slope of this regression represents the causal effect estimate, and the intercept can be interpreted as an estimate of the average horizontal pleiotropic effect across the genetic variants [19]. The weighted median estimator provides a consistent estimate of the causal effect, even when up to 50% of the information contributing to the analysis comes from genetic variants that are invalid IVs [20]. The weighted median estimator has the advantage of retaining greater precision in the estimates than those from the MR-Egger analysis [20]. Tests were considered statistically significant at p<0.05. All MR analyses were performed on MR Base platform [21].

Heterogeneity and sensitivity test

We assessed heterogeneities among SNPs using Cochran’s Q-statistics and funnel plots [22]. We also performed a “leave one out” analysis to investigate the possibility that the causal association was driven by a single SNP.

RESULTS

Studies included in the meta-analysis

1) Instrumental variables for Mendelian randomization

We selected six SNPs as IVs from the GWAS data on uric acid level and 12 SNPs from the GWAS data on gout as IVs to improve the inference (Table 1, Figure 1). Two of the IVs (rs2231142, rs734553) were associated with both uric acid level and gout (Table 1). The selected IVs were associated with uric acid level and gout at genome-wide significance, except for rs10791821, 11733284, and rs734553 in gout (Table 1).

Table 1 . Instrumental SNPs from uric acid level (A) or gout (B) and osteoarthritis GWASs.

A. Uric acid levels

Instrumental SNPEffect alleleGeneExposure (uric acid)Outcome (osteoarthritis)


BetaSEp-valueCase (n)Control (n)BetaSEp-value
rs12356193ASLC16A90.0800.0141.00×10−83,49811,0090.0100.0290.657
rs17300741ASLC22A110.0600.0087.00×10−143,49811,009−0.0200.0220.259
rs2231142TABCG20.1400.0221.00×10−103,49811,009−0.0940.0340.006
rs734553TSLC2A90.4000.0131.00×10−1923,49811,0090.0100.0260.679
rs742132ASCGN0.0500.0099.00×10−93,49811,0090.0300.0210.169
rs780094TGCKR0.0500.0081.00×10−93,49811,009−0.0300.0200.190

B. Gout

Instrumental SNPEffect alleleGeneExposure (gout)Outcome (osteoarthritis)


BetaSEp-valueCase (n)Control (n)BetaSEp-value

rs1014290TSLC2A90.4510.0427.00×10−263,49811,0090.0410.0240.075
rs10791821GMAP3K110.4510.0841.00×10−73,49811,0090.0000.0260.973
rs1165176GSLC17A10.3510.0571.00×10−93,49811,0090.0100.0210.717
rs11733284ANIPAL10.2150.0459.00×10−73,49811,009−0.0410.0240.099
rs11758351GHIST1H2BF0.3360.0582.00×10−83,49811,0090.0490.0310.088
rs1260326TGCKR0.3070.0432.00×10−123,49811,009−0.0410.0240.089
rs2231142TABCG20.5130.0753.00×10−123,49811,009−0.0940.0340.006
rs2728125CABCG20.7130.0467.00×10−543,49811,009−0.0730.0360.043
rs3114020CABCG20.6370.0519.00×10−353,49811,009−0.0300.0210.193
rs4073582GCNIH-20.5070.0866.00×10−93,49811,0090.0200.0240.259
rs4766566TCUX20.4120.0464.00×10−203,49811,0090.0300.0260.201
rs734553TSLC2A90.3290.0652.00×10−73,49811,0090.0100.0260.652

SNP: single nucleotide polymorphism, GWAS: genome-wide association study, Beta: beta coefficient, SE: standard error, SLC16A9: Solute Carrier Family 16 Member 9, SLC22A11: Solute Carrier Family 22 Member 11, ABCG2: ATP Binding Cassette Subfamily G Member 2 (Junior Blood Group), SLC2A9: Solute Carrier Family 2 Member 9, SCGN: Secretagogin, EF-Hand Calcium Binding Protein, GCKR: Glucokinase Regulator, MAP3K11: Mitogen-Activated Protein Kinase Kinase Kinase 11, SLC17A1: Solute Carrier Family 17 Member 1, NIPAL1: NIPA-Like Domain Containing 1, HIST1H2BF: Histone Cluster 1 H2B Family Member F, CNIH-2: Cornichon Family AMPA Receptor Auxiliary Protein 2, CUX2: Cut-Like Homeobox 2..


Figure 1. Forrest plot of the causal effects of uric acid level (A) or gout (B)-associated SNPs on osteoarthritis. SNP: single nucleotide polymorphism, MR: Mendelian randomization, IVW: inverse-variance weighted, Na: not available.

2) Mendelian randomization results

The IVW method yielded no evidence to support a causal association between uric acid level or gout and the risk of osteoarthritis (beta=−0.026, SE=0.096, p=0.789; beta=−0.018, SE=0.025, p=0.482) (Table 2, Figures 1, 2). In the MR-Egger test, the intercept represents the average pleiotropic effect across the genetic variants, i.e., the average direct effect of a variant on the outcome. If the intercept differs from zero, it serves as an evidence of directional pleiotropy. MR-Egger regression revealed that directional pleiotropy was unlikely to be biasing uric acid level (intercept=−0.013, p=0.610) or uric acid results (intercept=0.018, p=0.668), while yielding no causal association between uric acid level or gout and the risk of osteoarthritis (beta=0.028, SE=0.142, p=0.852; beta=−0.056, SE=0.090, p=0.548) (Table 2, Figures 1, 2). In addition, the weighted median approach provided no evidence of a causal association between uric acid level or gout and the risk of osteoarthritis (beta=0.004, SE=0.064, p=0.946; beta=−0.005, SE=0.025, p=0.843) (Table 2, Figures 1, 2). We performed subgroup analysis using exposure dataset of European. However, the MR result from European did not change significantly the overall MR result.

Table 2 . The MR estimates of the causal effect of uric acid level (A) and gout (B) on osteoarthritis risk, derived using different methods.

A. Uric acid level

MR methodNumber of SNPBetaSEAssociation p-valueCochran Q statisticHeterogeneity p-value
Inverse variance weighted6−0.0260.0960.78913.020.232
MR-Egger60.0280.1420.85212.090.166
Weighted median60.0040.0640.946nana

B. Gout

MR methodNumber of SNPBetaSEAssociation p-valueCochran Q statisticHeterogeneity p-value

Inverse variance weighted12−0.0180.0250.48226.530.005
MR-Egger12−0.0560.0900.54826.030.004
Weighted median12−0.0050.0250.834nana

MR: Mendelian randomization, SNP: single nucleotide polymorphism, Beta: beta coefficient, SE: standard error, na: not available..


Figure 2. Scatter plots of genetic associations of uric acid level (A) or gout (B) against the genetic associations of osteoarthritis. The slopes of each line represent the causal association for each method. Blue line represents the IVW estimate, green line represents the weighted median estimate, and dark blue line represents the MR-Egger estimate. IVW: inverse-variance weighted, SNP: single nucleotide polymorphism, MR: Mendelian randomization, Na: not available.

3) Heterogeneity and sensitivity test

Cochran’s Q-test indicated no evidence of heterogeneity between IV estimates based on the individual variants of uric acid level, but not gout (Table 2). The funnel test yielded symmetry, indicating no evidence of heterogeneity in the MR analyses of uric acid level and gout (Figure 3). Results from the “leave one out” analysis demonstrated that no single SNP was driving the IVW point estimate.

Figure 3. Funnel plot to assess the heterogeneity of the causal effects of uric acid level (A) or gout (B)-associated SNPs on osteoarthritis. Blue line represents the IVW estimate, and dark blue line represents the MR-Egger estimate. SNP: single nucleotide polymorphism, IVW: inverse-variance weighted, MR: Mendelian randomization, SE: standard error, β: beta coefficient.

CONCLUSION

In conclusion, the results of our MR analysis demonstrates that uric acid level and gout may be not causally associated with the increased risk of osteoarthritis. Considering MR study is not susceptible to bias from unmeasured confounders or reverse causation, the epidemiological evidence for an association between uric acid level or gout and a higher risk of osteoarthritis may be due to residual confounding or reverse causation rather than direct causality. Well-designed epidemiological and MR studies using more variants that explain a greater proportion of uric acid level or gout are warranted to confirm or rule out its causal relationship with osteoarthritis.

CONFLICT OF INTEREST

No potential conflict of interest relevant to this article was reported.

AUTHOR CONTRIBUTIONS

Y.H.L. was involved in conception and design of study, acquisition of data, analysis and interpretation of data, drafting the manuscript, and revising the manuscript. G.G.S. was involved in conception and design of study, analysis and interpretation of data, and drafting the manuscript.

Fig 1.

Figure 1.Forrest plot of the causal effects of uric acid level (A) or gout (B)-associated SNPs on osteoarthritis. SNP: single nucleotide polymorphism, MR: Mendelian randomization, IVW: inverse-variance weighted, Na: not available.
Journal of Rheumatic Diseases 2020; 27: 88-95https://doi.org/10.4078/jrd.2020.27.2.88

Fig 2.

Figure 2.Scatter plots of genetic associations of uric acid level (A) or gout (B) against the genetic associations of osteoarthritis. The slopes of each line represent the causal association for each method. Blue line represents the IVW estimate, green line represents the weighted median estimate, and dark blue line represents the MR-Egger estimate. IVW: inverse-variance weighted, SNP: single nucleotide polymorphism, MR: Mendelian randomization, Na: not available.
Journal of Rheumatic Diseases 2020; 27: 88-95https://doi.org/10.4078/jrd.2020.27.2.88

Fig 3.

Figure 3.Funnel plot to assess the heterogeneity of the causal effects of uric acid level (A) or gout (B)-associated SNPs on osteoarthritis. Blue line represents the IVW estimate, and dark blue line represents the MR-Egger estimate. SNP: single nucleotide polymorphism, IVW: inverse-variance weighted, MR: Mendelian randomization, SE: standard error, β: beta coefficient.
Journal of Rheumatic Diseases 2020; 27: 88-95https://doi.org/10.4078/jrd.2020.27.2.88

Table 1 . Instrumental SNPs from uric acid level (A) or gout (B) and osteoarthritis GWASs.

A. Uric acid levels

Instrumental SNPEffect alleleGeneExposure (uric acid)Outcome (osteoarthritis)


BetaSEp-valueCase (n)Control (n)BetaSEp-value
rs12356193ASLC16A90.0800.0141.00×10−83,49811,0090.0100.0290.657
rs17300741ASLC22A110.0600.0087.00×10−143,49811,009−0.0200.0220.259
rs2231142TABCG20.1400.0221.00×10−103,49811,009−0.0940.0340.006
rs734553TSLC2A90.4000.0131.00×10−1923,49811,0090.0100.0260.679
rs742132ASCGN0.0500.0099.00×10−93,49811,0090.0300.0210.169
rs780094TGCKR0.0500.0081.00×10−93,49811,009−0.0300.0200.190

B. Gout

Instrumental SNPEffect alleleGeneExposure (gout)Outcome (osteoarthritis)


BetaSEp-valueCase (n)Control (n)BetaSEp-value

rs1014290TSLC2A90.4510.0427.00×10−263,49811,0090.0410.0240.075
rs10791821GMAP3K110.4510.0841.00×10−73,49811,0090.0000.0260.973
rs1165176GSLC17A10.3510.0571.00×10−93,49811,0090.0100.0210.717
rs11733284ANIPAL10.2150.0459.00×10−73,49811,009−0.0410.0240.099
rs11758351GHIST1H2BF0.3360.0582.00×10−83,49811,0090.0490.0310.088
rs1260326TGCKR0.3070.0432.00×10−123,49811,009−0.0410.0240.089
rs2231142TABCG20.5130.0753.00×10−123,49811,009−0.0940.0340.006
rs2728125CABCG20.7130.0467.00×10−543,49811,009−0.0730.0360.043
rs3114020CABCG20.6370.0519.00×10−353,49811,009−0.0300.0210.193
rs4073582GCNIH-20.5070.0866.00×10−93,49811,0090.0200.0240.259
rs4766566TCUX20.4120.0464.00×10−203,49811,0090.0300.0260.201
rs734553TSLC2A90.3290.0652.00×10−73,49811,0090.0100.0260.652

SNP: single nucleotide polymorphism, GWAS: genome-wide association study, Beta: beta coefficient, SE: standard error, SLC16A9: Solute Carrier Family 16 Member 9, SLC22A11: Solute Carrier Family 22 Member 11, ABCG2: ATP Binding Cassette Subfamily G Member 2 (Junior Blood Group), SLC2A9: Solute Carrier Family 2 Member 9, SCGN: Secretagogin, EF-Hand Calcium Binding Protein, GCKR: Glucokinase Regulator, MAP3K11: Mitogen-Activated Protein Kinase Kinase Kinase 11, SLC17A1: Solute Carrier Family 17 Member 1, NIPAL1: NIPA-Like Domain Containing 1, HIST1H2BF: Histone Cluster 1 H2B Family Member F, CNIH-2: Cornichon Family AMPA Receptor Auxiliary Protein 2, CUX2: Cut-Like Homeobox 2..


Table 2 . The MR estimates of the causal effect of uric acid level (A) and gout (B) on osteoarthritis risk, derived using different methods.

A. Uric acid level

MR methodNumber of SNPBetaSEAssociation p-valueCochran Q statisticHeterogeneity p-value
Inverse variance weighted6−0.0260.0960.78913.020.232
MR-Egger60.0280.1420.85212.090.166
Weighted median60.0040.0640.946nana

B. Gout

MR methodNumber of SNPBetaSEAssociation p-valueCochran Q statisticHeterogeneity p-value

Inverse variance weighted12−0.0180.0250.48226.530.005
MR-Egger12−0.0560.0900.54826.030.004
Weighted median12−0.0050.0250.834nana

MR: Mendelian randomization, SNP: single nucleotide polymorphism, Beta: beta coefficient, SE: standard error, na: not available..


References

  1. Dieppe PA, Lohmander LS. Pathogenesis and management of pain in osteoarthritis. Lancet 2005;365:965-73.
    Pubmed CrossRef
  2. Litwic A, Edwards MH, Dennison EM, Cooper C. Epidemiology and burden of osteoarthritis. Br Med Bull 2013;105:185-99.
    Pubmed KoreaMed CrossRef
  3. Wortmann RL. Gout and hyperuricemia. Curr Opin Rheumatol 2002;14:281-6.
    Pubmed CrossRef
  4. Terkeltaub RA. Clinical practice. Gout. N Engl J Med 2003;349:1647-55.
    Pubmed CrossRef
  5. Ding X, Zeng C, Wei J, Li H, Yang T, Zhang Y, et al. The associations of serum uric acid level and hyperuricemia with knee osteoarthritis. Rheumatol Int 2016;36:567-73.
    Pubmed CrossRef
  6. Howard RG, Samuels J, Gyftopoulos S, Krasnokutsky S, Leung J, Swearingen CJ, et al. Presence of gout is associated with increased prevalence and severity of knee osteoarthritis among older men: results of a pilot study. J Clin Rheumatol 2015;21:63-71.
    Pubmed KoreaMed CrossRef
  7. Dalbeth N, Aati O, Kalluru R, Gamble GD, Horne A, Doyle AJ, et al. Relationship between structural joint damage and urate deposition in gout: a plain radiography and dual-energy CT study. Ann Rheum Dis 2015;74:1030-6.
    Pubmed CrossRef
  8. Yokose C, Chen M, Berhanu A, Pillinger MH, Krasnokutsky S. Gout and osteoarthritis: associations, pathophysiology, and therapeutic implications. Curr Rheumatol Rep 2016;18:65.
    Pubmed CrossRef
  9. Burgess S, Daniel RM, Butterworth AS, Thompson SG; EPIC-InterAct Consortium. Network Mendelian randomization: using genetic variants as instrumental variables to investigate mediation in causal pathways. Int J Epidemiol 2015;44:484-95.
    Pubmed KoreaMed CrossRef
  10. Kolz M, Johnson T, Sanna S, Teumer A, Vitart V, Perola M, et al. Meta-analysis of 28,141 individuals identifies common variants within five new loci that influence uric acid concentrations. PLoS Genet 2009;5:e1000504.
    Pubmed KoreaMed CrossRef
  11. Sulem P, Gudbjartsson DF, Walters GB, Helgadottir HT, Helgason A, Gudjonsson SA, et al. Identification of low-frequency variants associated with gout and serum uric acid levels. Nat Genet 2011;43:1127-30.
    Pubmed CrossRef
  12. Matsuo H, Yamamoto K, Nakaoka H, Nakayama A, Sakiyama M, Chiba T, et al. Genome-wide association study of clinically defined gout identifies multiple risk loci and its association with clinical subtypes. Ann Rheum Dis 2016;75:652-9.
    Pubmed KoreaMed CrossRef
  13. Nakayama A, Nakaoka H, Yamamoto K, Sakiyama M, Shaukat A, Toyoda Y, et al. GWAS of clinically defined gout and subtypes identifies multiple susceptibility loci that include urate transporter genes. Ann Rheum Dis 2017;76:869-77.
    Pubmed KoreaMed CrossRef
  14. Zeggini E, Panoutsopoulou K, Southam L, Rayner NW, Day-Williams AG, Lopes MC, et al; arcOGEN Consortium, arcOGEN Collaborators. Identification of new susceptibility loci for osteoarthritis (arcOGEN): a genome-wide association study. Lancet 2012;380:815-23.
    Pubmed KoreaMed CrossRef
  15. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013;37:658-65.
    Pubmed KoreaMed CrossRef
  16. Hartwig FP, Davies NM, Hemani G, Davey Smith G. Two-sample Mendelian randomization: avoiding the downsides of a powerful, widely applicable but potentially fallible technique. Int J Epidemiol 2016;45:1717-26.
    Pubmed KoreaMed CrossRef
  17. Pierce BL, Burgess S. Efficient design for Mendelian randomization studies: subsample and 2-sample instrumental variable estimators. Am J Epidemiol 2013;178:1177-84.
    Pubmed KoreaMed CrossRef
  18. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015;44:512-25.
    Pubmed KoreaMed CrossRef
  19. Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol 2017;32:377-89.
    Pubmed KoreaMed CrossRef
  20. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol 2016;40:304-14.
    Pubmed KoreaMed CrossRef
  21. Hemani G, Zheng J, Wade KH, Laurin C, Elsworth B, Burgess S, et al. MR-Base: a platform for systematic causal inference across the phenome using billions of genetic associations. bioRxiv 2016;16:78972.
    CrossRef
  22. Egger M, Smith GD, Phillips AN. Meta-analysis: principles and procedures. BMJ 1997;315:1533-7.
    Pubmed KoreaMed CrossRef
  23. Neogi T, Krasnokutsky S, Pillinger MH. Urate and osteoarthritis: Evidence for a reciprocal relationship. Joint Bone Spine 2019;86:576-82.
    Pubmed CrossRef
  24. Hill HA, Kleinbaum DG. Bias in observational studies. Wiley StatsRef . DOI: 10.1002/9781118445112.stat05111.
    CrossRef
  25. Lee YH, Bae SC, Song GG. Hepatitis B virus (HBV) reactivation in rheumatic patients with hepatitis core antigen (HBV occult carriers) undergoing anti-tumor necrosis factor therapy. Clin Exp Rheumatol 2013;31:118-21.
    Pubmed
  26. Smith GD, Ebrahim S. Mendelian randomization: genetic variants as instruments for strengthening causal inference in observational studies. In: Weinstein M, Vaupel JW, Wachter KW, eds. Biosocial surveys. Washington D.C., National Academies Press (US), 2008.
  27. VanderWeele TJ, Tchetgen Tchetgen EJ, Cornelis M, Kraft P. Methodological challenges in Mendelian randomization. Epidemiology 2014;25:427-35.
    Pubmed KoreaMed CrossRef
  28. Thompson JR, Minelli C, Bowden J, Del Greco FM, Gill D, Jones EM, et al. Mendelian randomization incorporating uncertainty about pleiotropy. Stat Med 2017;36:4627-45.
    Pubmed CrossRef
  29. Paaby AB, Rockman MV. The many faces of pleiotropy. Trends Genet 2013;29:66-73.
    Pubmed KoreaMed CrossRef
  30. Smith GD, Ebrahim S. Mendelian randomization: prospects, potentials, and limitations. Int J Epidemiol 2004;33:30-42.
    Pubmed CrossRef
  31. Swerdlow DI, Kuchenbaecker KB, Shah S, Sofat R, Holmes MV, White J, et al. Selecting instruments for Mendelian randomization in the wake of genome-wide association studies. Int J Epidemiol 2016;45:1600-16.
    Pubmed KoreaMed CrossRef
  32. Burgess S, Dudbridge F, Thompson SG. Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods. Stat Med 2016;35:1880-906.
    Pubmed KoreaMed CrossRef
JRD
Jan 01, 2025 Vol.32 No.1, pp. 1~7
COVER PICTURE
Cumulative growth of rheumatology members and specialists (1980~2024). Cumulative distribution of the number of the (A) Korean College of Rheumatology members and (B) rheumatology specialists. (J Rheum Dis 2025;32:63-65)

Stats or Metrics

Share this article on

  • line

Related articles in JRD

Journal of Rheumatic Diseases

pISSN 2093-940X
eISSN 2233-4718
qr-code Download