[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to main content
Download PDF

Dyslipidemia induced inflammation mediated the association between obesity and Osteoarthritis: a population-based study

BMC Public Health volume 24, Article number: 3155 (2024) Cite this article

Abstract

Background

This study aims to evaluate the mediation effect of dyslipidemia induced inflammation on the causal associations between obesity and Osteoarthritis (OA).

Methods

This cross-sectional study used data from the National Health and Nutrition Examination Survey (1999–2010). The association between general and abdominal obesity (exposure), OA (outcome) and mediators (total cholesterol, high-density lipoprotein, and C-reactive protein) was assessed using multivariate linear and logistic regression models and mediation analysis.

Results

A total of 23,308 participants were enrolled in this study, and 2,180 were diagnosed with OA. Participants with obesity were more likely to have OA (general obesity: OR = 2.508, 95%CI: 1.602, 4.197, P < 0.001; abdominal obesity: OR = 3.814, 95%CI: 3.242, 4.509, P < 0.001) than those without the obesity. High quantile of total cholesterol (OR:1.399; 95%CI:1.235, 1.257; P < 0.001), high-density lipoprotein (OR:1.644; 95%CI:1.443, 1.874; P < 0.001) and C-reactive protein (OR:1.952; 95%CI:1.707, 2.237; P < 0.001) increased the risk of OA when compared to lowest quartile. In the linear regression, the betas varied from 0.668 (95%CI: 0.635, 0.741; P < 0.001) to 0.693 (95%CI: 0.674, 0.712; P < 0.001), suggesting that individual with obesity had higher C-reactive protein levels. Additionally, total cholesterol and high-density lipoprotein were associated with C-reactive protein. Mediation analyses showed that the causal association of obesity with OA risk was mediated by high-density lipoprotein and C-reactive protein, with the mediation proportion ranging from 17.216 to 45.058%. Moreover, high-density lipoprotein to C-reactive protein path acting as serial mediators in the associations between obesity and OA (general obesity: β = 0.012; 95%CI: 0.009–0.014; abdominal obesity: β = 0.011; 95%CI: 0.008–0.014).

Conclusion

The association between obesity and OA is partially mediated by systemic inflammation caused by dyslipidemia. Our study suggested anti-lipid therapy may be positive for obese individuals with OA.

Peer Review reports

Introduction

Osteoarthritis (OA), a degenerative disease, predominantly affects the joints, causing pain, limited mobility, and a substantial financial burden on society [1]. This is especially the case among the millions of adults in the United States who are afflicted by the obesity epidemic [2]. Obesity is well known to be a significant risk factor for the development and onset of OA, as it not only increases adipokine production and joint loading but also raises cholesterol and related inflammatory indicators through adipose tissue infiltration [3, 4].

Numerous investigations demonstrate that obesity primarily increases the amounts of fatty acids delivered to the liver, leading to increased levels of dyslipidemia [5, 6], including heightened levels of total cholesterol (TCHO), marginally decreased levels of high-density lipoprotein cholesterol (HDL), and marginally increased levels of low-density lipoprotein (LDL) [7]. Despite the evidence from intervention trials is sparse, MRI imaging study suggested that dyslipidemia plays a role in the development of cartilage degradation [8]. Notably, Rotterdam study found that using lipid-lowering medicines was associated with a lower incidence and development of knee OA [9]. Additionally, recent literature suggests a causal link between these two disorders: mice on a high-cholesterol diet exhibit more severe cartilage degradation and synovial thickening, possibly due to the release of inflammatory chemicals like TNF-α and IL-1β [10]. Although research has enhanced our understanding of how cholesterol deposition in organs leads to systemic inflammatory indicators like C-reactive protein (CRP) via a variety of interrelated pathways [11, 12]. there is still no clinical data to support this notion.

Mediation analyses are a useful tool for estimating the potential causal pathway by which an intervention or exposure could produce an outcome [13]. The importance of mediation analyses in observational research for public health policies has been recognized in a recent JAMA article [14]. The aim of this study is to explore the link between obesity, dyslipidemia, systemic inflammation, and OA risk. Furthermore, we will examine whether inflammation caused by dyslipidemia could mediate the causal relationship between obesity and OA through serial mediation analysis. We hypothesize that dyslipidemia associated with obesity may exacerbate inflammation and result in the onset of OA.

Methods

Study population and data source

This population-based survey employed a cross-sectional design based on the National Health and Nutrition Examination Survey (NHANES). NHANES is a national research designed to assess the health and nutrition status of children and adults in the United States. The URL https://www.cdc.gov/nchs/nhanes/ provides public access to detailed NHANES protocols and data. In our study, the 1999–2010 continuous cycles of NHANES were selected as they include a comprehensive periodontal examination, measurements of body mass index (BMI) and waist circumference, as well as serum concentrations of cholesterol and CRP. Likewise, we can also collect data on socio-demographic factors, health-risk behaviors, physical examination findings, medical history, and serum biomarker parameters.

Ethics approval and consent to participate

The NHANES data is publicly available. The protocols of the NHANES were approved by the institutional review board of the National Center for Health Statistics, Centers for Disease Control. Written informed consent was obtained from each participant prior to participation in this study.

Exclusion criteria

This research involved participants who reported having OA (n = 29,157). The following exclusion criteria were used in the final sample analysis: (1) being under 21 years of age (n = 582), (2) pregnancy (n = 1,214), (3) incomplete BMI information (n = 2,048), (4) missing data on waist circumference (n = 914), and (5) absence of CRP data (n = 1,091). Finally, 23,308 in 6 cycles of NHANES were included in the study (Fig. 1).

Fig. 1
figure 1

Flow chart of participants selection, NHANES. Abbreviations: NHANES, National Health and Nutrition Examination Survey; OA, Osteoarthritis; BMI, body mass index; CRP, C-reactive protein

Full size image

Outcome identification

Study suggested that questionnaires have the greater potential to detect OA in the community with an 81% agreement between self-report and clinical diagnosis of OA [15]. In the NHANES, OA was diagnosed by professionals and the information was collected using a questionnaire survey. Briefly, participants were asked two arthritis-related questions. First, they were asked: “Has a doctor or other health professional ever told you that you have arthritis?” Participants who answered affirmatively were subsequently prompted with a follow-up inquiry: “Which type of arthritis was it?” Those who gave a positive answer about OA to this question will be considered individuals with OA.

Obesity measurements and exposure

General obesity (defined by BMI) and abdominal obesity (measured by waist circumference) were identified as exposure factors. Following the World Health Organization or related guidelines, general obesity was defined as BMI values ≥ 30 kg/m², and abdominal obesity was defined as waist circumference ≥ 110 cm for males and ≥ 105 cm for females [16]. Heights, weights, and waist circumference were measured in adults by trained health technicians in the Mobile Examination Center. BMI (kg/m²) was computed as weight in kilograms divided by height in meters squared. Waist circumference was measured at the iliac crest with a tape, to the nearest millimeter.

Laboratory analysis and mediators

Dyslipidemia biomarkers and systemic inflammation were set as mediation variables. A complete blood count was obtained from blood specimens collected from eligible participants. The blood specimens for biological markers test were stored under appropriate temperature conditions until they were shipped to a contract laboratory for testing. These biological marker information were obtained from NHANES and included triglyceride (mmol/L), LDL (mmol/L), TCHO (mmol/L), HDL (mmol/L), and CRP (mg/dL).

Covariates

To control potential confounding effect of related risk factors that may contribute to OA, these basic socio-demographic characteristics, including gender, age, race/ethnicity, marital status, education background and medical insurance, were initially considered [17]. Other related covariates comprise: smoking status (never, ever) [18], drinking status (never, ever) [19], hypertension (yes, no) [20], and diabetes (yes, no) [21], and physical activity (vigorous, moderate, low or never) [22]. All covariates listed above can be determined directly based on related questionnaires or related medical history.

Statistical analysis

Data were analyzed using R (version 4.3.1) and SPSS (version 26.0). QQ plots were used to evaluate normality assumptions, and continuous variables were given as mean ± SD or median (interquartile range, IQR) based on data distribution. Binary and multi-categorical variables were reported as percentages. Simple demographic differences between individuals with and without OA were assessed using t-tests or Mann-Whitney U test (for continuous data), and chi-square testing (for categorical data). In our data analysis, triglyceride and LDL were excluded for high levels of missingness (51.647% and 54.058%).

Given the CRP data is obviously skewed to the right and unevenly distributed, we must first log-transform its values before performing statistical analysis. Logistic regression analysis was utilized to assess the correlations between exposure (BMI and waist circumference), mediators (TCHO, HDL, and CRP), and OA risk. Univariate and multivariate logistic and linear regression models were then developed for investigating the association between exposure (BMI and waist circumference) and mediators (TCHO, HDL, and CRP). Continuous variables of TCHO, HDL were categorized into four groups (Q1, Q2, Q3, and Q4) based on quartile; BMI was divided into four groups according to the cut-off points: 18.5, 25, and 30 kg/m2. Similarly, WC was divided into four categories: 90/80, 100/90, and 110/105 cm in male/female. All variables were enrolled in multivariable regression model after excluding multicollinearity. Overall, the odds ratio (OR) and 95% confidence interval (CI), as well as β coefficient, was calculated. Additionally, we assumed that the covariates missing at random. Sensitive analysis was conducted in regression analysis under a multivariable imputation procedure (MICE) with 5 datasets [23]. All analyses were two-tailed, and a p-value of < 0.05 indicated significance.

In mediation analyses, we made the assumption that the exposure preceded the mediators1 (M1, defined as TCHO and HDL), M1 preceded the mediators2 (M2, defined as CRP), and M2 preceded the outcome (Fig. 2). The possible serial mediation effects were estimated using serial mediation analyses (PROCESS for SPSS, version 4.1) with 5000 bootstraps, and if 0 was not contained in the 95% CI, the mediation effect was judged statistically significant [24, 25].

Fig. 2
figure 2

Mediation analysis model. Four prespecified routes were used as follow: Route 1 (direct effect: c’): Exposure (obesity)→Outcome (Osteoarthritis); Route 2 (indirect effect: a1*b1): Exposure (obesity)→Mediator 1 (dyslipidemia)→Outcome (Osteoarthritis); Route 3 (indirect effect: a2*b2): Exposure (obesity)→Mediator 2 (inflammation)→Outcome (Osteoarthritis); Route 4 (serial indirect effect: a1*d1*b2): Exposure (obesity)→Mediator 1 (dyslipidemia) →Mediator 2 (inflammation)→Outcome (Osteoarthritis); Proportion of mediation = (|a*b|/|a*b|+|c’|)*100%

Full size image

Results

Baseline characteristics

Overall, 23,308 participants were included in this study, and 2,180 of these were diagnosed with OA with 63.303% being female, 60.046% aged over 60 years and 71.697% being non-Hispanic white. Compared to those were free of OA, participants with OA had a lower rate of drinking history (65.717% vs. 71.794%), higher rate of smoking history (53.557% vs. 47.428%), and higher prevalence of hypertension (69.083% vs. 39.145%) and diabetes (16.100% vs. 9.424%). Participants with an OA diagnosis had a greater BMI (29.794 ± 6.537 vs. 28.380 ± 6.256) than those without OA. Comparably, survey participants without OA had lower WC (97.217 ± 02.364 vs. 102.117 ± 15.101) than those with OA. Moreover, participants with OA also exhibited higher values of TCHO, HDL, and CRP compared to those without OA (all P < 0.001) (Table 1).

Table 1 Basic characteristics of include participants, NHANES 1999–2010
Full size table

Logistic regression analyses

Table 2 showed the associations between obesity, serum biomarkers and OA risk. In the unadjusted model, BMI (OR: 1.033; 95%CI:1.026, 1.040; P < 0.001), WC (OR: 1.019; 95%CI:1.017, 1.022; P < 0.001) and CRP (OR: 1.103; 95%CI:1.058, 1.147; P < 0.001) was associated with OA. Participants with obesity (BMI: OR = 2.508; 95%CI: 1.602 to 4.197, P < 0.001; WC: OR = 3.814; 95%CI: 3.242 to 4.509, P < 0.001) had a higher risk of developing OA. Similarly, after adjust other covariates, the positive associations can be found in the model 2 and model 3. In the unadjusted model, the relationship between TCHO (OR:1.117; 95%CI:1.075, 1.161; P < 0.001), HDL (OR:1.477; 95%CI:1.336, 1.633; P < 0.001) and OA was significant positive. The highest quantile of TCHO (OR:1.399; 95%CI:1.235, 1.538; P < 0.001) and HDL (OR:1.644; 95%CI:1.443, 1.874; P < 0.001) increased the risk of OA compared to Q1. However, after adjusting for covariates, those positive correlation became insignificant. Further, each 1-unit increase in log-CRP was associated with an increased OA risk (OR:1.103; 95%CI: 1.058, 1.147; P < 0.001). The highest quantile of CRP (OR:1.952; 95%CI:1.707, 2.237; P < 0.001) increased the risk of OA compared to Q1. These were consistent with the quantile analysis in model 2 and model 3 (all P < 0.001). The sensitivity analyses based on multiple imputation was consistent with results mentioned above (Table S1).

Table 2 Associations of BMI, waist circumference, biological markers with OA risk
Full size table

Linear regression analyses

In Table 3, linear regression analyses presented that obesity, assessed by BMI (β:0.688 95%CI:0.635, 0.741; P < 0.001) and WC (β:0.693; 95%CI:0.674, 0.712; P < 0.001), was associated with Log-CRP compared to underweight. These findings were confirmed after adjusted multiple covariates (all P < 0.001). We observed positive association of TCHO (β: 0.046, 95%CI: 0.039, 0.052, P < 0.001) with log-CRP value, and negative association between HDL (β: -0.231, 95%CI: -0.248, -0.214; P < 0.001) and log-CRP. With increasing quantiles of TCHO, log-CRP was increased (β: 0.134, 95%CI: 0.115, 0.154; P < 0.001). And with increasing quantiles of HDL, log-CRP was decreased (β: -0.265, 95%CI: -0.285, -0.245; P < 0.001). These were consistent with the outcome in Model 2 and Model 3 (all P < 0.001). The results generated by the sensitivity analysis were consistent with these models (Table S2).

Table 3 Associations between BMI, waist circumference, TCHO, HDL, and log-CRP
Full size table

Mediation analyses

As presented in Figs. 3 and 4 and Table S3, the direct effect of general obesity and abdominal obesity were both significant (All P < 0.001). Association between general obesity and OA was mediated by TCHO (indirect effect = 0.005, 95%CI: 0.002, 0.009), HDL (indirect effect =-0.110, 95%CI: -0.131, -0.089), and CRP (indirect effect = 0.155, 95%CI: 0.119, 0.191) in the unadjusted models. Mediation proportion of TCHO and HDL was 2.577% and 27.778%, separately. Mediation proportion of CRP ranging from 36.865 to 45.058%, respectively. Apparently, the association between abdominal obesity and OA was mediated by HDL (indirect effect =-0.115, 95%CI: -0.135, -0.094) and CRP (indirect effect = 0.151, 95%CI: 0.116, 0.187). Mediation proportion of HDL was 17.216%. Mediated proportion of CRP ranging from 21.449 to 22.712%, respectively. Furthermore, HDL-CRP path act as a serial mediator in the association between obesity and OA (general obesity: β = 0.012, 95%CI: 0.009, 0.014; abdominal obesity: β = 0.011, 95%CI: 0.008, 0.014), as well as TCHO-CRP path (general obesity: β = 0.001, 95%CI: 0.000, 0.001; abdominal obesity: β = 0.000, 95%CI: 0.000–0.000). Similarly, same results can be found with additional adjustment for other potential covariates in both model 2 and model 3 (Table S4-5).

Fig. 3
figure 3

Mediation effect of dyslipidemia induced inflammation on the associations of general obesity with Osteoarthritis. (A) all routes were significant suggesting TCHO, CRP and TCHO-CRP path are effective mediators to the causation between whole obesity and Osteoarthritis. (B) all routes were significant suggesting HDL, CRP and HDL-CRP path are effective mediators to the causation between whole obesity and Osteoarthritis

Full size image
Fig. 4
figure 4

Mediation effect of dyslipidemia induced inflammation on the associations of abdominal obesity with Osteoarthritis. (A) all routes except route 1 was significant suggesting CRP and TCHO-CRP path are effective mediators to the causation between abdominal obesity and osteoarthritis. (B) all routes were significant suggesting HDL, CRP and HDL-CRP path are effective mediators to the causation between abdominal obesity and osteoarthritis

Full size image

Discussion

To the best of our knowledge, this is the first clinical investigation to show how dyslipidemia and systemic inflammation, which are mediators of obesity and OA, are related. Our research specifically showed that the HDL-CRP axis has strong serial mediation effects on the causative relationship between obesity and OA, indicating that inflammation and dyslipidemia together may be important factors in the pathological development of OA.

The origins and causes of inflammation in OA may be classified as mechanical damage and metabolic stress, which is mainly produced from adipose tissue [26]. Therefore, it is reasonable to suggest that obesity, accompanied by increased adiposity, contribute to the occurrence of OA not only the mechanism of increasing joint stress but also through obesity-related metabolic inflammation [27]. Utilizing data collected from the NHANES, we found that the association between increased adiposity and the risk of OA among American adults is particularly mediated by dyslipidemia and elevated inflammation. Ali et al. noted that chondrocytes may internalize lipids through hedgehog signaling, leading to the accumulation of cholesterol within chondrocytes [28], and these mechanisms may inhibit extracellular matrix synthesis and increase the expression of MMP13 and Adamts4 [29]. Additionally, Choi et al. [30] pointed out that the accumulation of cholesterol directly activates the cholesterol receptor RORα, which subsequently upregulates the expression of matrix-degrading enzymes in chondrocytes. However, this change in lipid accumulation in the cartilage inducing cellular toxicity can be partially impeded by lipophilic statins through the downregulation of pivotal cartilage-degrading enzymes [31].

The impact of body fat distribution on health varies. For example, excessive visceral fat is closely associated with insulin resistance and metabolic disorders, while subcutaneous fat may have a lesser impact [32, 33]. Herein, we measured BMI for overall obesity and waist circumference for abdominal obesity. The complexity of the inflammatory system and its involvement in multivariable feedback mechanisms are well-known. Dyslipidemia-induced ectopic lipid deposition within chondrocytes has been suggested to alter the action of pro-inflammatory cytokines. CRP is a phase-protein that can be produced by adipocytes and regulated by proinflammatory cytokines [34]. Cohort study found significant correlations between CRP and the incidence and symptoms of OA, which was consistent with our finding [35]. Based on the results of regression analyses, we further explore the underlying path from obesity to OA. We observed significant mediated effects of total cholesterol on the associations of obesity with OA risk, with proportion of mediation ranging from 0.654 to 27.778%. More importantly, serial mediation analyses showed the associations of general obesity and abdominal obesity with OA were mediated by TCHO to CRP and HDL to CRP path. Inspired by our findings, there must be alternative biomechanical path that contributed to the relationship between obesity and OA. Overall, our study not only clinically validated the important role of dyslipidemia in the risk of OA, but also provided the finding that a serial mediation effect of dyslipidemia-inflammation path between obesity and OA. These results contribute to a better understanding of the complex relationship between obesity, dyslipidemia, inflammation and OA.

There were several strengths in our study. First, our data analysis based on a large national survey, highlighting the broad applicability of our findings. Second, two different obesity assessments and dyslipidemia markers were adopted in each survey which could also show that the association remains significant irrespectively. What is more, by applying a variety of methods to explore the relationships between obesity and OA in a relatively large population, we were able to enhance the reliability of our study findings. There were some limitations in this study that we need to acknowledge. First, given the existence of variances in the body fat distribution across racial groups, it is plausible that the findings may have been impacted by this factor. Second, although studies indicated that TCHO and HDL were more reliability of plasma level of lipid biomarkers than triglycerides and LDL [36], due to a relatively high proportion of missing values, we did not apply them in the study. This may hamper a more comprehensive understanding of the relationship between OA and dyslipidemia-induced systemic inflammation. Subsequently, obtaining clinically or radiologically diagnosed data for a large-scale population study is a formidable challenge. Although the pooled sensitivity and specificity of self-reported OA compared to clinically confirmed OA are acceptable, as demonstrated by Peeters et al. [37]. In this NHANES study, the information regarding the assessment of OA obtained through self-reported questionnaires may indeed constrain the credibility of the results. Most crucially, our research on the causal inference between lipid-induced systemic inflammation and OA is grounded in a retrospective cross-sectional study. Given that other inflammatory indicators (such as white blood cell count or erythrocyte sedimentation rate) were not incorporated into the mediation model, this might lead to an overestimation of the mediation proportion of CRP. Fifth, there exists a disparity in OA grades among different levels of obesity. It is probable that a more severe OA is accompanied by a higher CRP [38]. Nevertheless, we were unable to conduct these specific investigations with our present data source. Further studies on this subject matter are requisite. Considering the limitations presented, these findings should be interpreted with caution.

Conclusion

In conclusion, our study suggested that dyslipidemia with systemic inflammation might be a critical mediator in the causative relationship of obesity with OA, as evidenced by indirect effect of TCHO to CRP and HDL to CRP path. Meaningfully, adequate lipid-lowering strategy may be positive for obese individuals with OA.

Data availability

Publicly available datasets were analyzed in this study. This data can be found here: https://www.cdc.gov/nchs/nhanes.

References

  1. Peat G, Thomas MJ. Osteoarthritis year in review 2020: epidemiology & therapy. OSTEOARTHR Cartil. 2021;29(2):180–9.

    Article  CAS  Google Scholar 

  2. Zhao G, Zhu S, Zhang F, Zhang X, Zhang X, Li T, Li D, Zhu W. Global Burden of osteoarthritis associated with high body mass index in 204 countries and territories, 1990–2019: findings from the Global Burden of Disease Study 2019. ENDOCRINE 2023, 79(1):60–71.

  3. Sampath S, Venkatesan V, Ghosh S, Kotikalapudi N. Obesity, metabolic syndrome, and Osteoarthritis-An updated review. CURR OBES REP. 2023;12(3):308–31.

    Article  PubMed  Google Scholar 

  4. Zhu J, Ruan G, Cen H, Meng T, Zheng S, Wang Y, Li B, Zhu Z, Han W, Winzenberg T, et al. Association of serum levels of inflammatory markers and adipokines with joint symptoms and structures in participants with knee osteoarthritis. Rheumatology. 2022;61(3):1044–52.

    Article  CAS  PubMed  Google Scholar 

  5. Jung UJ, Choi MS. Obesity and its metabolic complications: the role of adipokines and the relationship between obesity, inflammation, insulin resistance, dyslipidemia and nonalcoholic fatty liver disease. INT J MOL SCI. 2014;15(4):6184–223.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Mallick R, Basak S, Das RK, Banerjee A, Paul S, Pathak S, Duttaroy AK. Fatty acids and their proteins in adipose tissue inflammation. CELL BIOCHEM BIOPHYS; 2023.

  7. Vekic J, Zeljkovic A, Stefanovic A, Jelic-Ivanovic Z, Spasojevic-Kalimanovska V. Obesity and dyslipidemia. METABOLISM. 2019;92:71–81.

    Article  CAS  PubMed  Google Scholar 

  8. Pan F, Tian J, Mattap SM, Cicuttini F, Jones G. Association between metabolic syndrome and knee structural change on MRI. Rheumatology. 2020;59(1):185–93.

    PubMed  Google Scholar 

  9. Clockaerts S, Van Osch GJ, Bastiaansen-Jenniskens YM, Verhaar JA, Van Glabbeek F, Van Meurs JB, Kerkhof HJ, Hofman A, Stricker BH, Bierma-Zeinstra SM. Statin use is associated with reduced incidence and progression of knee osteoarthritis in the Rotterdam study. ANN RHEUM DIS. 2012;71(5):642–7.

    Article  CAS  PubMed  Google Scholar 

  10. Cao C, Shi Y, Zhang X, Li Q, Zhang J, Zhao F, Meng Q, Dai W, Liu Z, Yan W, et al. Cholesterol-induced LRP3 downregulation promotes cartilage degeneration in osteoarthritis by targeting Syndecan-4. NAT COMMUN. 2022;13(1):7139.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. O’Hagan R, Berg AR, Hong CG, Parel PM, Mehta NN, Teague HL. Systemic consequences of abnormal cholesterol handling: interdependent pathways of inflammation and dyslipidemia. FRONT IMMUNOL. 2022;13:972140.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Tall AR, Yvan-Charvet L. Cholesterol, inflammation and innate immunity. NAT REV IMMUNOL. 2015;15(2):104–16.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Lee H, Herbert RD, McAuley JH. Mediation analysis. JAMA-J AM MED ASSOC. 2019;321(7):697–8.

    Article  Google Scholar 

  14. Lee H, Cashin AG, Lamb SE, Hopewell S, Vansteelandt S, VanderWeele TJ, MacKinnon DP, Mansell G, Collins GS, Golub RM, et al. A Guideline for reporting mediation analyses of randomized trials and observational studies: the AGReMA Statement. JAMA-J AM MED ASSOC. 2021;326(11):1045–56.

    Article  Google Scholar 

  15. March LM, Schwarz JM, Carfrae BH, Bagge E. Clinical validation of self-reported osteoarthritis. OSTEOARTHR Cartil. 1998;6(2):87–93.

    Article  CAS  Google Scholar 

  16. Ross R, Neeland IJ, Yamashita S, Shai I, Seidell J, Magni P, Santos RD, Arsenault B, Cuevas A, Hu FB, et al. Waist circumference as a vital sign in clinical practice: a Consensus Statement from the IAS and ICCR Working Group on visceral obesity. NAT REV ENDOCRINOL. 2020;16(3):177–89.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Nikolajsen H, Larsen CM, Holsgaard-Larsen A, Juul-Kristensen B, Hestbaek L. Physical impairments among adults in Denmark: a register-based study. BMC Public Health. 2022;22(1):2416.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Ni J, Wang P, Yin KJ, Huang JX, Tian T, Cen H, Sui C, Xu Z, Pan HF. Does smoking protect against developing osteoarthritis? Evidence from a genetically informed perspective. SEMIN ARTHRITIS RHEU. 2022;55:152013.

    Article  CAS  Google Scholar 

  19. He K, Huang H. The significant role of Alcohol in the relationship between C-Reactive protein and self-reported osteoarthritis. J NUTR. 2024;154(2):600–9.

    Article  CAS  PubMed  Google Scholar 

  20. Yang ZJ, Liu Y, Liu YL, Qi B, Yuan X, Shi WX, Miao L. Osteoarthritis and hypertension: observational and mendelian randomization analyses. ARTHRITIS RES THER. 2024;26(1):88.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Eitner A, Culvenor AG, Wirth W, Schaible HG, Eckstein F. Impact of diabetes Mellitus on knee Osteoarthritis Pain and Physical and Mental Status: data from the Osteoarthritis Initiative. ARTHRIT CARE RES. 2021;73(4):540–8.

    Article  Google Scholar 

  22. Lopes DG, Costa D, Cruz EB, Mendonça N, Henriques AR, Branco J, Canhão H, Rodrigues AM. Association of physical activity with physical function and quality of life in people with hip and knee osteoarthritis: longitudinal analysis of a population-based cohort. ARTHRITIS RES THER. 2023;25(1):14.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Sterne JA, White IR, Carlin JB, Spratt M, Royston P, Kenward MG, Wood AM, Carpenter JR. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. BMJ-BRIT MED J. 2009;338:b2393.

    Article  Google Scholar 

  24. Livingston MD, Haardörfer R. A Gentle introduction to mediation and moderation. J PEDIATR-US. 2019;214:246–8.

    Article  Google Scholar 

  25. Hayes AF. Introduction to Mediation, Moderation, and conditional process analysis: a regression-based Approach: introduction to Mediation, Moderation, and conditional process analysis. A Regression-Based Approach; 2013.

  26. Berenbaum F, Wallace IJ, Lieberman DE, Felson DT. Modern-day environmental factors in the pathogenesis of osteoarthritis. NAT REV RHEUMATOL. 2018;14(11):674–81.

    Article  PubMed  Google Scholar 

  27. Gløersen M, Steen PP, Neogi T, Jafarzadeh SR, Vistnes M, Thudium CS, Bay-Jensen AC, Sexton J, Kvien TK, Hammer HB, et al. Associations of Body Mass Index with Pain and the Mediating role of inflammatory biomarkers in people with Hand Osteoarthritis. ARTHRITIS RHEUMATOL. 2022;74(5):810–7.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Ali SA, Al-Jazrawe M, Ma H, Whetstone H, Poon R, Farr S, Naples M, Adeli K, Alman BA. Regulation of cholesterol homeostasis by hedgehog signaling in osteoarthritic cartilage. ARTHRITIS RHEUMATOL. 2016;68(1):127–37.

    Article  CAS  PubMed  Google Scholar 

  29. Su YP, Chen CN, Huang KC, Chang HI, Lee KC, Lo CM, Chang SF. Leptin induces MMP1/13 and ADAMTS 4 expressions through bone morphogenetic protein-2 autocrine effect in human chondrocytes. J CELL BIOCHEM. 2018;119(4):3716–24.

    Article  CAS  PubMed  Google Scholar 

  30. Choi WS, Lee G, Song WH, Koh JT, Yang J, Kwak JS, Kim HE, Kim SK, Son YO, Nam H, et al. The CH25H-CYP7B1-RORα axis of cholesterol metabolism regulates osteoarthritis. Nature. 2019;566(7743):254–8.

    Article  CAS  PubMed  Google Scholar 

  31. Barter MJ, Hui W, Lakey RL, Catterall JB, Cawston TE, Young DA. Lipophilic statins prevent matrix metalloproteinase-mediated cartilage collagen breakdown by inhibiting protein geranylgeranylation. ANN RHEUM DIS. 2010;69(12):2189–98.

    Article  CAS  PubMed  Google Scholar 

  32. Shi J, Bao G, Hong J, Wang S, Chen Y, Zhao S, Gao A, Zhang R, Hu J, Yang W, et al. Deciphering CT texture features of human visceral fat to evaluate metabolic disorders and surgery-induced weight loss effects. EBIOMEDICINE. 2021;69:103471.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Lin Y, Zhong X, Lu D, Yao W, Zhou J, Wu R, Feng F. Association of visceral and subcutaneous fat with bone mineral density in US adults: a cross-sectional study. SCI REP-UK. 2023;13(1):10682.

    Article  CAS  Google Scholar 

  34. Eklund CM. Proinflammatory cytokines in CRP baseline regulation. ADV CLIN CHEM. 2009;48:111–36.

    Article  CAS  PubMed  Google Scholar 

  35. Stannus OP, Jones G, Blizzard L, Cicuttini FM, Ding C. Associations between serum levels of inflammatory markers and change in knee pain over 5 years in older adults: a prospective cohort study. ANN RHEUM DIS. 2013;72(4):535–40.

    Article  CAS  PubMed  Google Scholar 

  36. Lee SA, Wen W, Xiang YB, Fazio S, Linton MF, Cai Q, Liu D, Zheng W, Shu XO. Stability and reliability of plasma level of lipid biomarkers and their correlation with dietary fat intake. DIS MARKERS. 2008;24(2):73–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Peeters GM, Alshurafa M, Schaap L, de Vet HC. Diagnostic accuracy of self-reported arthritis in the general adult population is acceptable. J CLIN EPIDEMIOL. 2015;68(4):452–9.

    Article  PubMed  Google Scholar 

  38. Arendt-Nielsen L, Eskehave TN, Egsgaard LL, Petersen KK, Graven-Nielsen T, Hoeck HC, Simonsen O, Siebuhr AS, Karsdal M, Bay-Jensen AC. Association between experimental pain biomarkers and serologic markers in patients with different degrees of painful knee osteoarthritis. ARTHRITIS RHEUMATOL. 2014;66(12):3317–26.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The authors thank the data collection team and NHANES administration and staff for the data and reports made available through the NHANES website that allowed us to generate this paper and also Thank all the participants who have contributed to this study.

Funding

This study was supported by the National Natural Science Foundation of China (grant number:82274553). The funder has no role in the conceptualization, design, data collection, analysis, decision to publish, or preparation of the manuscript.

Author information

Author notes

  1. Yuxin Zheng and Yuelong Cao contributed equally to this work.

Authors and Affiliations

  1. Shi’s Center of Orthopedics and Traumatology, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, Institute of Traumatology & Orthopedics, Shanghai Academy of Traditional Chinese Medicine, No. 528, Zhangheng Road, Pudong New Area, Shanghai, 201203, China

    Laijun Yan, Haiya Ge, Qinguang Xu, Ding Jiang, Anping Shen, Muyun Yang, Yuxin Zheng & Yuelong Cao

Authors
  1. Laijun Yan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Haiya Ge
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Qinguang Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ding Jiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Anping Shen
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Muyun Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Yuxin Zheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Yuelong Cao
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

LJY: Conception and design of the study, analysis and interpretation of the data, methodology, drafting of the article. HYG: Conception and design of the study, analysis and interpretation of the data, graphics processing. QGX, DJ, APS, MYY: revising the article critically for important intellectual content. YLC and YXZ: Statistical expertise, obtaining of funding, administrative, critical revision of the article for important intellectual content, final approval of the article. YLC and YXZ contribute equally to this work. All the authors read and approved the final manuscript.

Corresponding authors

Correspondence to Yuxin Zheng or Yuelong Cao.

Ethics declarations

Ethics approval and consent to participate

The NHANES data is publicly available. The protocols of the NHANES were approved by the institutional review board of the National Center for Health Statistics, Centers for Disease Control. Written informed consent was obtained from each participant prior to participation in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yan, L., Ge, H., Xu, Q. et al. Dyslipidemia induced inflammation mediated the association between obesity and Osteoarthritis: a population-based study. BMC Public Health 24, 3155 (2024). https://doi.org/10.1186/s12889-024-20616-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12889-024-20616-4

Keywords

BMC Public Health

ISSN: 1471-2458

Contact us