Genetic Evidence Supports a Causal Association Between Dietary Factors and Osteoarthritis: A Mendelian Randomization Analysis

doi:10.21203/rs.3.rs-4284464/v1

Download PDF

Research Article

Genetic Evidence Supports a Causal Association Between Dietary Factors and Osteoarthritis: A Mendelian Randomization Analysis

https://doi.org/10.21203/rs.3.rs-4284464/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objective: Observational studies have previously suggested a relationship between diet and osteoarthritis (OA). However, whether there is a causal association between dietary choices and OA, including knee osteoarthritis (KOA), and hip osteoarthritis (HOA), remained uncertain.

Method: We conducted our analysis using summary-level data on dietary factors and OA outcomes (KOA and HOA) obtained from the IEU Open GWAS project. The primary analysis relied on the inverse variance weighted (IVW) method to determine if various dietary factors were causally associated with KOA or HOA.

Results: Our study revealed a genetic association between certain dietary factors and the risk of KOA and HOA. Specifically, alcohol intake (OR 1.26, 95%CI 1.05-1.52, p = 0.014) and coffee consumption (OR 2.05, 95%CI 1.61-2.61, p-value 0.000) were genetically linked to an increased risk of KOA. In contrast, cheese consumption (OR 0.61, 95%CI 0.48-0.76, p = 0.000), cereal intake (OR 0.58, 95%CI 0.37-0.90, p =0.014), and dried fruit consumption (OR 0.44, 95%CI 0.26-0.76, p = 0.003) were identified as protective factors against KOA. Additionally, genetically predicted coffee intake (OR 1.63, 95%CI 1.11-2.61, p = 0.012) and pork consumption (OR 2.18, 95%CI 1.03-4.61, p =0.041) showed positive associations with HOA.

Conclusion: This study provides genetic evidence supporting a causal association between specific dietary choices and the risk of OA. These findings complement earlier retrospective studies and offer valuable insights into potential strategies for preventing OA.

Mendelian randomization

knee osteoarthritis

hip osteoarthritis

dietary intake

GWAS

Osteoarthritis (OA), encompassing knee osteoarthritis (KOA) and hip osteoarthritis (HOA), is a chronic degenerative condition primarily characterized by the deterioration of articular cartilage. The burden of OA is growing due to an aging population, and its prevalence has notably increased over the past decade, particularly for KOA [1, 2]. The primary clinical manifestations are joint pain and deformities. Although oral nonsteroidal anti-inflammatory drugs (NSAIDs) can offer relief from joint pain, late-stage patients often face limited treatment options, with artificial joint replacement being the main alternative [3, 4]. The substantial costs associated with treatment impose a significant economic burden on patients [5]. Various factors contribute to the development and progression of OA. Current research identifies age, gender (especially female), obesity, prior joint injuries, joint deformities, muscle weakness, and high levels of physical activity as known risk factors for OA [6, 7]. A recent study has shown a causal association between serum sex hormones (such as testosterone and dihydrotestosterone) and the development of HOA [8]. Additionally, investigations have highlighted the impact of sleep therapy on OA progression [9]. However, the complete pathogenesis of OA remains unclear. Therefore, a deeper understanding of the causes of OA is of significant clinical importance.

Over the last decade, epidemiological studies have underscored the crucial role of different dietary factors in OA development [10]. Diet can influence the formation of OA through various mechanisms; for example, a high-fat diet may induce obesity, diabetes and other diseases, thereby promoting the onset of OA [11, 12]. Observational studies have suggested that the consumption of fresh vegetables and fruits can delay the progression of OA [13]. However, most existing studies on diet and disease primarily rely on observational data, often collected through questionnaires, which can introduce instability and substantial bias into research outcomes. In the face of the diverse array of dietary choices, establishing a causal relationship between diet and OA is paramount for effective disease treatment and prevention.

Mendelian randomization (MR) studies represent an epidemiological investigative approach wherein genetic variations serve as instrumental variables to explore the links between exposures and outcomes. Since single nucleotide polymorphisms (SNPs) remain constant from birth and are not susceptible to external environmental influences, MR studies offer more robust evidence for inferring causal relationships compared to observational studies [14]. In our pursuit to ascertain whether a causal association exists between OA and dietary factors, we conducted a two-sample MR analysis using publicly available genome-wide association study (GWAS) datasets. Our objective is to provide theoretical insights that may inform future dietary interventions for OA prevention.

Data sources and MR design

A two-sample MR analysis was conducted to investigate the potential causal relationship between OA and 18 different dietary factors. All summary-level data used in this study were sourced from the IEU open GWAS project, with ethical approvals obtained from the original studies. The outcome data for HOA (ebi-a-GCST007091) and KOA (ebi-a-GCST007090) included 393873 and 403124 European individuals, respectively, which were collected through a genome-wide meta-analysis by the UK Biobank and the Arthritis Research UK Osteoarthritis Genetic consortium. The dietary exposure factors comprised various components, including alcoholic intake frequency (ukb-b-5779), processed meat intake (ukb-b-6324), poultry intake (ukb-b-8006), beef intake (ukb-b-2862), non-oily fish intake (ukb-b-17627), oily fish intake (ukb-b-2209), pork intake (ukb-b-5640), lamb intake (ukb-b-14179), bread intake (ukb-b-11348), cheese intake (ukb-b-1489), cooked vegetable intake (ukb-b-8090), tea intake (ukb-b-6066), fresh fruit intake (ukb-b-3881), cereal intake (ukb-b-15926), raw vegetable intake (ukb-b-1996), coffee intake (ukb-b-5237), dried fruit intake (ukb-b-16576), and hot drink temperature (ukb-b-14203). The MR design for this study is displayed in Fig. 1.

Selection of instrumental variables (IVs)

In MR analyses, the selection of appropriate IVs is crucial for examining causal relationships between dietary factors and OA. IVs must meet three fundamental criteria: First, they should exhibit a strong association with the exposure factors. Second, IVs must not be directly related to the outcome of interest. Third, IVs should not be correlated with potential confounding factors. SNPs were extracted from the IEU Open GWAS project, with the following screening criteria [15]: (1) A significant IV association threshold of p < 5 × 10^− 8; (2) Low linkage disequilibrium (r² = 0.001); (3) A maximum physical distance between genes of 10,000 base pairs. In addition, we also applied MR-PRESSO to identify the outliers. Additionally, we employed the MR-PRESSO method to identify and address outliers, utilizing 5,000 distributions in the analysis. To account for potential bias from weak IVs, we calculated F-statistics and R² for each SNP, with F-statistics exceeding 10 indicating a lack of weak IV bias, all of which were included in our MR analyses.

R² = 2*beta2*EAF*(1-EAF)/(2*beta2*EAF*(1-EAF) + se2*2*N*EAF(1-EAF))

F = (N − 2)*R²/(1 − R²)

Statistical analyses

All statistical analyses were conducted using R and the “TwoSampleMR” package. MR analysis is a method that leverages genetic variation as IVs to investigate the impact of exposure factors (food intake) on outcomes (KOA and HOA). After harmonizing the data following IV selection, we employed inverse-variance weighted (IVW) regression as the primary method to estimate the causal effect of food intake on KOA and HOA outcomes. The IVW method accounts for potential heterogeneity by allowing for different mean effects among SNPs. It combines Wald ratio estimates of causal effects obtained from different SNPs, providing a consistent estimate of the causal relationship when each genetic variant adheres to the instrumental variable assumptions. Causal associations were considered significant when p-values were < 0.05. Additionally, we performed supplementary analyses using four other methods, namely, MR-Egger, weighted median, simple mode, and weighted mode.

We assessed IV heterogeneity using Cochran’s Q statistic, with p-values < 0.05 indicating significant heterogeneity. In the MR-PRESSO analysis, we identified outliers (using 5,000 distributions) and removed SNPs contributing disproportionately to heterogeneity. We then conducted MR analysis again, recognizing that heterogeneity and residual confounding could persist due to potential horizontal pleiotropy, even after outlier removal. It is important to note that the presence of heterogeneity does not necessarily invalidate the IVW model.

To assess horizontal pleiotropy in IVs, we employed the MR-Egger regression method. A non-zero intercept distinct from the origin or a p-value > 0.05 can indicate potential pleiotropic effects. A p-value > 0.05 in MR-Egger regression implies the presence of horizontal pleiotropy. In such cases, SNPs (p < 5e-08) in the PhenoScanner database, known to be potential confounders or associated with outcomes, were eliminated.

After sequentially removing each IV using the leave-one-out sensitivity test, IVW MR analysis was performed on the remaining IVs. To determine the existence of abnormal IVs that significantly influence the causal effect estimates, the stability of the effect estimates was assessed.

To investigate the causal relationship between KOA and dietary factors, we examined 18 different exposure factors for analysis. We selected between 6 to 70 SNPs as IVs based on strict criteria. Similarly, we chose between 9 to 82 SNPs as IVs for analyzing the causal relationship between HOA and dietary factors. All SNPs had F-statistics greater than 10, indicating no evidence of weak instrumental bias. More information regarding the SNP associations in this study can be found in the Supplementary Excel file.

We presented the MR estimates from five methods assessing the causality of dietary factors on KOA in Supplementary Table 1. The primary IVW method identified a total of five causal relationships (p < 0.05) in this study, including alcohol intake (OR 1.26, 95% CI 1.05–1.52, p = 0.014), cereal intake (OR 0.58, 95% CI 0.37–0.90, p = 0.014), cheese intake (OR 0.61, 95% CI 0.48–0.76, p = 0.000), coffee intake (OR 2.05, 95% CI 1.61–2.61, p-value 0.000), and dried fruit intake (OR 0.44, 95% CI 0.26–0.76, p = 0.003) (Fig. 2). Other methods yielded similar results to IVW, including cheese intake (Weighted Median: OR 0.610, 95% CI 0.458–0.811, p = 0.001), coffee intake (MR Egger: OR 2.331, 95% CI 1.428–3.666, p = 0.001, Weighted Median: OR 2.102, 95% CI 1.393–2.906, p = 0.000, Weighted Model: OR 2.509, 95% CI 1.750–3.595, p = 0.000), and dried fruit intake (Weighted Median: OR 0.428, 95% CI 0.252–0.727, p = 0.002). This suggests a positive causal association with KOA (OR > 1) for alcohol and coffee intake, while cheese, cereal, and dried fruit intake showed a negative causal association with KOA (OR < 1). No causal association was found between other dietary factors and KOA after the exclusion of outliers and confounding SNPs. Scatter plots are presented in Supplementary Fig. 1.

To assess the causal relationship between dietary factors and HOA, the results are presented in Supplementary Table 2. The primary IVW method identified two causal relationships (p < 0.05) in this study, coffee intake (OR 1.63, 95%CI 1.11–2.61, p = 0.012), and pork intake (OR 2.18, 95%CI 1.03–4.61, p = 0.041) (Fig. 3). Other methods produced results similar to IVW, such as coffee intake (Weighted model: OR 1.553, 95%CI 1.023–2.561, p = 0.047). These findings indicate a positive causal association with HOA (OR > 1) for pork and coffee intake. No causal association was found between other dietary factors and HOA after the exclusion of outliers and confounding SNPs. Scatter plots are also presented in Supplementary Fig. 2.

For sensitivity analysis, we performed Cochran’s Q test, MR-Egger intercept test, and Leave-one-out analysis to validate the reliability of the MR analysis results. Although some heterogeneity persisted after excluding outlier SNPs, we primarily used the random effects model of IVW in this study, and the level of heterogeneity remained acceptable. The MR-Egger intercept test yielded p-values greater than 0.05, indicating no evidence of horizontal pleiotropy (Tables 1 and 2). Leave-one-out analyses did not change the directionality of the results when removing each SNP (Supplementary Figs. 3 and 4).

Table 1

Sensitivity analysis of the Mendelian randomization (MR) analysis results of KOA and dietary factors
Outcome	Exposure	Number of SNPs	MR Egger-intercept	p value	Cochran’s Q	p value
Knee osteoarthritis	Alcohol intake	70	0.014	0.117	193.32	< 0.001
	Beef intake	9	0.01	0.533	4.1	0.874
	Bread intake	24	0.006	0.585	37.20	0.03
	Cereal intake	27	-0.000	0.948	53.75	0.001
	Cheese intake	50	0.001	0.866	72.61	0.016
	Coffee intake	30	-0.003	0.512	31.36	0.348
	Cooked vegetable intake	10	0.006	0.878	8.19	0.515
	Dried fruit intake	31	0.013	0.479	88.68	< 0.001
	Fresh fruit intake	42	0.000	0.917	55.64	0.105
	Lamb intake	26	0.019	0.073	52.44	0.005
	Non-oily fish intake	8	-0.0153	0.39	4.68	0.699
	Oily fish intake	43	0.01	0.325	95.64	< 0.001
	Pork intake	11	0.045	0.205	25.52	0.004
	Poultry intake	6	0.015	0.906	3.468	0.628
	Processed meat intake	20	0.0232	0.215	40.02	0.003
	Raw vegetable intake	16	0.018	0.402	34.80	0.003
	Tea intake	36	-0.0077	0.259	66.97	0.001
	Hot drink temperature	58	0.005	0.588	116.44	< 0.001

Table 2

Sensitivity analysis of the Mendelian randomization (MR) analysis results of HOA and dietary factors
Outcome	Exposure	Number of SNPs	MR Egger-intercept	p Value	Cochran’s Q	p value
Hip osteoarthritis	Alcohol intake	82	0.005	0.221	139.88	< 0.001
	Beef intake	10	0.012	0.652	13.62	0.137
	Bread intake	26	-0.005	0.783	42.28	0.017
	Cereal intake	31	-0.009	0.566	55.81	0.003
	Cheese intake	51	-0.005	0.641	80.49	0.004
	Coffee intake	32	-0.004	0.570	55.4	0.005
	Cooked vegetable intake	11	0.046	0.276	9.95	0.445
	Dried fruit intake	33	0.003	0.859	50.90	0.018
	Fresh fruit intake	44	0.004	0.582	54.88	0.106
	Lamb intake	30	0.01	0.565	35.53	0.046
	Non-oily fish intake	9	0.012	0.628	10.2	0.251
	Oily fish intake	50	-0.004	0.65	73.55	0.013
	Pork intake	12	-0.034	0.234	9.71	0.556
	Poultry intake	6	-0.195	0.266	4.68	0.456
	Processed meat intake	22	-0.02	0.291	23.12	0.186
	Raw vegetable intake	15	-0.027	0.364	37.73	0.001
	Tea intake	35	-0.01	0.099	53.33	0.019
	Hot drink temperature	69	-0.003	0.684	79.10	0.06

In recent years, there has been a steady increase in the incidence of OA, particularly in the cases of KOA, which significantly outnumber those of HOA. KOA typically affects individuals over the age of 50 and can lead to prolonged joint pain, severely impacting their daily lives and mental well-being. For advanced cases of OA, joint replacement surgery becomes necessary to alleviate pain and restore mobility [16, 17]. Epidemiological studies have shown that OA is influenced by various factors, including high BMI and prolonged poor posture. As dietary habits have evolved, dietary factors have also emerged as important contributors to the development and progression of OA [18, 19]. However, no systematic study has been conducted on the causal relationship between different factors and OA. marks the first attempt to employ MR methods to investigate the causal relationships between various dietary factors and both HOA and KOA. We selected 18 different diets from a large database as exposure diets and conducted MR analysis after identifying suitable IVs. Most importantly, this study revealed strong causal relationships (p < 0.05) between KOA and five diets, where "A" represents a risk factor for KOA development. Furthermore, we identified three diets that act as protective factors against KOA. Additionally, we found causal associations between two diets and HOA, both of which are risk factors for HOA development.

The mechanisms through which alcohol causes OA may contribute to osteoarthritis remain somewhat unclear and likely involve several factors. On the one hand, alcohol can trigger the body's immune system, leading to excessive inflammation in the joint space, which in turn damages the joint cartilage surface. There is evidence to suggest that the inflammatory factor IL-6 is associated with the onset of OA [20, 21]. On the other hand, alcohol-induced damage to liver and kidney function can elevate uric acid levels in the bloodstream. The deposition of uric acid crystals in joint fluid can then erode the articular cartilage surface, ultimately causing OA [22, 23]. Additionally, alcohol consumption can lead to increased BMI, which can exacerbate the development of OA. This may also be an indirect factor contributing to alcohol-induced OA [24]. Interestingly, our MR analysis revealed a causal relationship between alcohol consumption and KOA, but not HOA. The consensus regarding alcohol's role in HOA remains somewhat contentious. One study conducted among females found a positive association between alcohol intake and the incidence of HOA [25]. Furthermore, some animal experiments have shown that alcohol consumption can disrupt the articular cartilage structure in rats, leading to the onset of HOA [26]. However, in a cohort study of 840 people followed up for 22 years, the authors found no association between alcohol consumption and HOA [27]. Similarly, a cross-sectional study of 568 women from the U.S. Nurses' Health Study failed to establish a significant association between self-reported hip replacement due to OA and alcohol consumption [28]. Therefore, relying solely on these observational studies makes it challenging to definitively determine the relationship between alcohol consumption and HOA. By employing genetic tools as variables that are not susceptible to external confounding factors, we aim to enhance the reliability of the conclusions drawn from this study.

Coffee consumption has been associated with several diseases, and previous epidemiological studies have found an association between coffee consumption and OA [29]. Several published MR studies have investigated the causal relationship between overall health outcomes and habitual coffee consumption in a cohort of 333,214 individuals of white British descent in the UK Biobank. These studies revealed a causal relationship between coffee consumption and OA [30, 31]. The precise mechanism by which coffee influences the development of OA is not yet fully understood. Some research suggests that caffeine consumption may have a detrimental impact on hyaline cartilage. In studies involving rats, caffeine was found to down-regulate the expression of chondromate-associated proteins like COL2A1 and ACAN. It also reduced articular cartilage matrix synthesis and accelerated articular cartilage degeneration by decreasing proteins associated with the IGF-1 signaling pathway [32, 33]. Furthermore, coffee consumption has been linked to an increased incidence of metabolic-related diseases, such as type 2 diabetes and obesity, which indirectly contribute to the development of OA [34]. In our MR study, coffee consumption exhibited a positive correlation with both KOA and HOA (OR > 1). Thus, we speculate that reducing coffee intake could potentially reduce the risk of developing and progressing OA.

Early observational studies have produced differing perspectives on the relationship between cheese consumption and KOA. Data from the American Organization for Osteoarthritis, based on up to 4 years of follow-up, suggested that high cheese consumption is positively associated with the development of KOA [35]. In contrast, a Dutch study proposed that cheese consumption might have a protective effect on knee joints. This protective effect may be attributed to the presence of trace elements such as calcium and phosphorus in cheese, which could enhance the absorption of other fat-soluble vitamins, thus benefiting the knee joint [36]. Our MR analysis, in alignment with prior research [37] confirmed a negative causal relationship between cheese intake and KOA. We posit that cheese consumption may enhance bone mechanical strength and mitigate joint degeneration in response to long-term weight-bearing, particularly in the knee joint. However, such causal relationship was not observed in HOA, potentially because the hip joint benefits from strong surrounding muscle tissues that share the load, reducing the risk of joint degeneration. Grains are a fundamental component of modern diets, and previous observational studies have suggested that increased grain consumption can significantly reduce the occurrence of KOA and knee pain. This effect is attributed to the dietary fiber content in grains, which can help reduce obesity rates in the population [38, 39]. To the best of our knowledge, this study is the first to employ MR analysis to examine the relationship between cereal consumption and osteoarthritis. Our findings generally align with previous observational studies, showing a negative association between cereal intake and KOA but no significant relationship with HOA. This protective effect may be attributed to weight reduction facilitated by increased cereal consumption. There is currently no epidemiological study confirming whether there is an association between dried fruit consumption as a daily snack and OA. Although dried fruits may have higher sugar content than fresh ones, they are more satiating as solid foods and may help reduce the risk of weight gain [40]. Our MR analysis indicated that dried fruits might reduce the incidence of KOA. However, further observational studies are warranted to validate this conclusion.

In this study, we identified five dietary factors that are causally associated with KOA based on the GWAS database, along with two dietary factors causally associated with HOA. One of the major strengths of this study is its ability to overcome potential confounding factors commonly encountered in observational studies, such as clinical heterogeneity stemming from human factors and variations in sample quality. However, it is noteworthy that we observed a disparity in results between KOA and HOA, with the exception of coffee intake, for which a causal relationship was found in both cases. We are uncertain whether this discrepancy is due to site-specific effects or if the genetic variables selected for HOA are more robust and representative. In the future, it will be necessary to employ advanced techniques such as multivariate MR or higher-level research to elucidate the causal relationship between diet and OA.

Nonetheless, there are limitations to our study. Firstly, the large sample size used in this study primarily comprises individuals of European ancestry, and there may be an overlap of SNPs within the sample, potentially introducing bias into the study results. Negative findings in this study should not be interpreted as a definitive absence of a causal relationship between dietary factors and OA. Future research should involve larger sample sizes and data from diverse racial backgrounds to provide further validation. Secondly, this study included only 18 dietary factors due to the complexity of food choices in modern life. For instance, fish consumption was excluded due to a lack of effective IVs. Finally, this study employed MR on two separate samples, and it is possible that various types of commonly consumed foods in daily life may interact with each other, either in an antagonistic or synergistic manner.

In conclusion, this study marks the first application of MR to investigate the causal relationship between 18 diets and OA. Our findings suggest that five factors may be causally associated with KOA. Among them, two factors (alcohol and coffee) increased the risk of KOA, while the other three exhibited protective effects. In the case of HOA, two diets (coffee and pork) were found to increase the risk of its development. Nevertheless, further research, using additional data and research methodologies, is required to more comprehensively understand the relationship between dietary factors and OA. Most importantly, our study offers valuable insights for clinicians considering dietary interventions and collaborative efforts to prevent OA during diagnosis and treatment.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary material. Further inquiries can be directed to the corresponding authors.

Funding

There was no funding for this work.

Acknowledgments

Special thanks to the IEU open GWAS project developed by The MRC Integrative Epidemiology Unit (IEU) at the University of Bristol. Thank them for extracting relevant GWAS summary-level data from published articles, UK Biobank, and FinnGen biobank.

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interest.

Author Contribution

Conceptualization, Fenyong Chen; data curation and formal analysis, Zhibin Huang and Minjian Lin; writing—original draft, Ying Han; writing—review and editing, Ying Han and Fenyong Chen. All authors read and approved the final manuscript.

S. Glyn-Jones, A.J. Palmer, R. Agricola, A.J. Price, T.L. Vincent, H. Weinans, and A.J. Carr, Osteoarthritis. Lancet (London, England) 386 (2015) 376-87.
R. Barnett, Osteoarthritis. Lancet (London, England) 391 (2018) 1985.
Y. Cho, S. Jeong, H. Kim, D. Kang, J. Lee, S.B. Kang, and J.H. Kim, Disease-modifying therapeutic strategies in osteoarthritis: current status and future directions. Experimental & molecular medicine 53 (2021) 1689-1696.
H. Meng, L. Jiang, Z. Song, and F. Wang, Causal Associations of Circulating Lipids with Osteoarthritis: A Bidirectional Mendelian Randomization Study. Nutrients 14 (2022).
J.H. Salmon, A.C. Rat, H. Achit, W. Ngueyon-Sime, C. Gard, F. Guillemin, D. Jolly, and B. Fautrel, Health resource use and costs of symptomatic knee and/or hip osteoarthritis. Osteoarthritis and cartilage 27 (2019) 1011-1017.
C. Cai, W. Hu, and T. Chu, Interplay Between Iron Overload and Osteoarthritis: Clinical Significance and Cellular Mechanisms. Frontiers in cell and developmental biology 9 (2021) 817104.
C. García-Ibarbia, S. Neila, C. Garcés, M.A. Alonso, M.T. Zarrabeitia, C. Valero, F. Ortiz, and J.A. Riancho, Non-synonymous WNT16 polymorphisms alleles are associated with different osteoarthritis phenotypes. Rheumatology international 37 (2017) 1667-1672.
Y.S. Yan, Z. Qu, D.Q. Yu, W. Wang, S. Yan, and H.F. Huang, Sex Steroids and Osteoarthritis: A Mendelian Randomization Study. Frontiers in endocrinology 12 (2021) 683226.
J. Ni, W. Zhou, H. Cen, G. Chen, J. Huang, K. Yin, and C. Sui, Evidence for causal effects of sleep disturbances on risk for osteoarthritis: a univariable and multivariable Mendelian randomization study. Osteoarthritis and cartilage 30 (2022) 443-450.
Y. Deng, J. Huang, and M.C.S. Wong, Associations between six dietary habits and risk of hepatocellular carcinoma: A Mendelian randomization study. Hepatology communications 6 (2022) 2147-2154.
A. Bortoluzzi, F. Furini, and C.A. Scirè, Osteoarthritis and its management - Epidemiology, nutritional aspects and environmental factors. Autoimmunity reviews 17 (2018) 1097-1104.
S. Sekar, R. Crawford, Y. Xiao, and I. Prasadam, Dietary Fats and Osteoarthritis: Insights, Evidences, and New Horizons. Journal of cellular biochemistry 118 (2017) 453-463.
C. Xu, N.E. Marchand, J.B. Driban, T. McAlindon, C.B. Eaton, and B. Lu, Dietary Patterns and Progression of Knee Osteoarthritis: Data from the Osteoarthritis Initiative. The American journal of clinical nutrition 111 (2020) 667-676.
P. Sekula, M.F. Del Greco, C. Pattaro, and A. Köttgen, Mendelian Randomization as an Approach to Assess Causality Using Observational Data. Journal of the American Society of Nephrology : JASN 27 (2016) 3253-3265.
Z. Dai, W. Xu, R. Ding, X. Peng, X. Shen, J. Song, P. Du, Z. Wang, and Y. Liu, Two-sample Mendelian randomization analysis evaluates causal associations between inflammatory bowel disease and osteoporosis. Frontiers in public health 11 (2023) 1151837.
J. Martel-Pelletier, A.J. Barr, F.M. Cicuttini, P.G. Conaghan, C. Cooper, M.B. Goldring, S.R. Goldring, G. Jones, A.J. Teichtahl, and J.P. Pelletier, Osteoarthritis. Nature reviews. Disease primers 2 (2016) 16072.
B. Abramoff, and F.E. Caldera, Osteoarthritis: Pathology, Diagnosis, and Treatment Options. The Medical clinics of North America 104 (2020) 293-311.
U. Nedunchezhiyan, I. Varughese, A.R. Sun, X. Wu, R. Crawford, and I. Prasadam, Obesity, Inflammation, and Immune System in Osteoarthritis. Frontiers in immunology 13 (2022) 907750.
M.H.J. van den Bosch, Osteoarthritis year in review 2020: biology. Osteoarthritis and cartilage 29 (2021) 143-150.
T. Liu, C. Xu, J.B. Driban, T. McAlindon, C.B. Eaton, and B. Lu, Excessive alcohol consumption and the risk of knee osteoarthritis: a prospective study from the Osteoarthritis Initiative. Osteoarthritis and cartilage 30 (2022) 697-701.
B. Lu, D.H. Solomon, K.H. Costenbader, B.T. Keenan, L.B. Chibnik, and E.W. Karlson, Alcohol consumption and markers of inflammation in women with preclinical rheumatoid arthritis. Arthritis and rheumatism 62 (2010) 3554-9.
J. Hyun, J. Han, C. Lee, M. Yoon, and Y. Jung, Pathophysiological Aspects of Alcohol Metabolism in the Liver. International journal of molecular sciences 22 (2021).
H.K. Choi, K. Atkinson, E.W. Karlson, W. Willett, and G. Curhan, Alcohol intake and risk of incident gout in men: a prospective study. Lancet (London, England) 363 (2004) 1277-81.
K. To, C. Mak, C. Zhang, Y. Zhou, S. Filbay, and W. Khan, The association between alcohol consumption and osteoarthritis: a meta-analysis and meta-regression of observational studies. Rheumatology international 41 (2021) 1577-1591.
N.E. Marchand, Y. Hu, M. Song, B.A. Rosner, E.W. Karlson, C. Ratzlaff, B. Lu, M.H. Liang, and W.C. Willett, Alcohol Consumption and Risk of Total Hip Replacement Due to Hip Osteoarthritis in Women. Arthritis & rheumatology (Hoboken, N.J.) 75 (2023) 1522-1531.
R. Kc, R. Voigt, X. Li, C.B. Forsyth, M.B. Ellman, K.C. Summa, F.W. Turek, A. Keshavarzian, J.S. Kim, and H.J. Im, Induction of Osteoarthritis-like Pathologic Changes by Chronic Alcohol Consumption in an Experimental Mouse Model. Arthritis & rheumatology (Hoboken, N.J.) 67 (2015) 1678-80.
R. Juhakoski, M. Heliövaara, O. Impivaara, H. Kröger, P. Knekt, H. Lauren, and J.P. Arokoski, Risk factors for the development of hip osteoarthritis: a population-based prospective study. Rheumatology (Oxford, England) 48 (2009) 83-7.
E.W. Karlson, L.A. Mandl, G.N. Aweh, O. Sangha, M.H. Liang, and F. Grodstein, Total hip replacement due to osteoarthritis: the importance of age, obesity, and other modifiable risk factors. The American journal of medicine 114 (2003) 93-8.
J.V. Higdon, and B. Frei, Coffee and health: a review of recent human research. Critical reviews in food science and nutrition 46 (2006) 101-23.
K. Nicolopoulos, A. Mulugeta, A. Zhou, and E. Hyppönen, Association between habitual coffee consumption and multiple disease outcomes: A Mendelian randomisation phenome-wide association study in the UK Biobank. Clinical nutrition (Edinburgh, Scotland) 39 (2020) 3467-3476.
Y. Zhang, J. Fan, L. Chen, Y. Xiong, T. Wu, S. Shen, X. Wang, X. Meng, Y. Lu, and X. Lei, Causal Association of Coffee Consumption and Total, Knee, Hip and Self-Reported Osteoarthritis: A Mendelian Randomization Study. Frontiers in endocrinology 12 (2021) 768529.
M. Guillán-Fresco, E. Franco-Trepat, A. Alonso-Pérez, A. Jorge-Mora, M. López-Fagúndez, A. Pazos-Pérez, O. Gualillo, and R. Gómez, Caffeine, a Risk Factor for Osteoarthritis and Longitudinal Bone Growth Inhibition. Journal of clinical medicine 9 (2020).
Y. Tan, K. Lu, J. Li, Q. Ni, Z. Zhao, J. Magdalou, L. Chen, and H. Wang, Prenatal caffeine exprosure increases adult female offspring rat's susceptibility to osteoarthritis via low-functional programming of cartilage IGF-1 with histone acetylation. Toxicology letters 295 (2018) 229-236.
A.T. Nordestgaard, M. Thomsen, and B.G. Nordestgaard, Coffee intake and risk of obesity, metabolic syndrome and type 2 diabetes: a Mendelian randomization study. International journal of epidemiology 44 (2015) 551-65.
B. Lu, J.B. Driban, J. Duryea, T. McAlindon, K.L. Lapane, and C.B. Eaton, Milk consumption and progression of medial tibiofemoral knee osteoarthritis: data from the Osteoarthritis Initiative. Arthritis care & research 66 (2014) 802-9.
K.F.M. Denissen, A. Boonen, J.T.H. Nielen, A.L. Feitsma, E. van den Heuvel, P.J. Emans, C.D.A. Stehouwer, S.J.S. Sep, M. van Dongen, P.C. Dagnelie, and S. Eussen, Consumption of dairy products in relation to the presence of clinical knee osteoarthritis: The Maastricht Study. European journal of nutrition 58 (2019) 2693-2704.
N.R.W. Geiker, C. Mølgaard, S. Iuliano, R. Rizzoli, Y. Manios, L.J.C. van Loon, J.M. Lecerf, G. Moschonis, J.Y. Reginster, I. Givens, and A. Astrup, Impact of whole dairy matrix on musculoskeletal health and aging-current knowledge and research gaps. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 31 (2020) 601-615.
J.P. Zertuche, G. Rabasa, A.H. Lichtenstein, N.R. Matthan, M. Nevitt, J. Torner, C.E. Lewis, Z. Dai, D. Misra, and D. Felson, Alkylresorcinol, a biomarker for whole grain intake, and its association with osteoarthritis: the MOST study. Osteoarthritis and cartilage 30 (2022) 1337-1343.
Z. Dai, J. Niu, Y. Zhang, P. Jacques, and D.T. Felson, Dietary intake of fibre and risk of knee osteoarthritis in two US prospective cohorts. Annals of the rheumatic diseases 76 (2017) 1411-1419.
C. Alasalvar, J.S. Salvadó, and E. Ros, Bioactives and health benefits of nuts and dried fruits. Food chemistry 314 (2020) 126192.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Genetic Evidence Supports a Causal Association Between Dietary Factors and Osteoarthritis: A Mendelian Randomization Analysis

Status:

Version 1

Abstract

Figures

Introduction

Methods

Data sources and MR design

Selection of instrumental variables (IVs)

Statistical analyses

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1