High prevalence and distinct patterns of metabolic syndrome in rheumatoid arthritis, psoriatic arthritis, and axial spondyloarthritis: a population-based study
- Open Access
- 01.10.2025
- Observational Research
Erschienen in:
Abstract
Introduction
Inflammatory arthritis (IA), including rheumatoid arthritis (RA), psoriatic arthritis (PsA), and axial spondyloarthritis (axSpA), share common immunogenetic mechanisms, clinical features, and therapeutic options [1]. These conditions also have a potentially bidirectional relationship with metabolic syndrome (MetS); for example, obesity is a shared risk factor, and systemic inflammation in IA may contribute to MetS features like insulin resistance [2, 3]. This association has clinical implications; for instance, metabolic dysfunction-associated steatotic liver disease (MASLD) can restrict the use of liver-toxic disease-modifying antirheumatic drugs (DMARDs) [4], while obesity may affect treatment response through immune modulation and/or pharmacodynamics [5].
MetS affects an estimated quarter of the general population, with prevalence expected to rise in light of the global obesity epidemic [6]. It is characterised by central adiposity, high blood pressure, and abnormalities in serum glucose and lipid levels [7], and often co-exists with obesity [8]. MetS is commonly defined using criteria from the 2001 National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP-III) or the 2009 International Diabetes Federation (IDF) Task Force Joint Interim Statement [7, 9].
Anzeige
The prevalence of MetS is higher among individuals with IA compared to the general population. The prevalence of MetS has been reported as 31% in RA and 46% in PsA, with the prevalence in axSpA estimated to be around 33% [10, 11]. In contrast, the prevalence in the general population is 25.4% [6]. However, these studies are highly heterogeneous in terms of geographic location, IA classification and MetS definition, making comparisons between types of IA challenging [10]. Furthermore, few studies have directly compared MetS prevalence across RA, PsA, and axSpA. Better understanding the epidemiology of MetS across IA is important because it directly influences treatment response, medication choice and the risks of cardiovascular disease (CVD), type 2 diabetes, and all-cause mortality [4, 7, 8, 12, 13]. Understanding the relative risk of MetS between different IAs is essential to providing holistic care to the groups most at risk. We aimed to describe the prevalence of MetS in a large cohort of individuals with RA, PsA, or axSpA.
Methods
We conducted a cross-sectional study using data from the UK Biobank, a cohort study of over 500,000 individuals aged 40 to 69 years recruited between 2006 and 2010. Ethical approval for the study was obtained, and details are available elsewhere [14]. Cross-sectional data from the baseline assessment, performed between 2006 and 2010, were used for the current observational analysis.
Participants with RA, PsA, and axSpA were identified through (1) ICD-10 codes from hospital admission records and/or (2) read codes from primary care records. Individuals with more than one code, for whom the precise diagnosis cannot be ascertained, were excluded. Controls were individuals without a coded diagnosis of RA, PsA or axSpA. Controls with a self-reported diagnosis of RA, PsA or axSpA were also excluded.
MetS was defined using a modified version of the NCEP ATP III criteria [9] (not using fasting blood samples), requiring at least three of the following five criteria: (1) a waist circumference of ≥ 102 cm (40 inches) in men or ≥ 88 cm (35 inches) in women; (2) blood pressure > 130/85 mmHg, a recorded diagnosis of hypertension, or use of antihypertensive medication; (3) a random glucose level ≥ 5.6 mmol/L (≥ 100 mg/dL), a diagnosis of type 2 diabetes, or use of diabetes medications (excluding insulin to avoid misclassification of type 1 diabetes); (4) a random triglyceride level ≥ 1.693 mmol/L (≥ 150 mg/dL); and (5) an HDL cholesterol level < 1.0 mmol/L (< 40 mg/dL) in men or < 1.293 mmol/L (< 50 mg/dL) in women, or use of lipid-lowering therapy.
Anzeige
We used descriptive statistics, presenting means for continuous variables and percentages for categorical variables. Statistical comparison was initially omitted, as the large sample size would likely result in statistically significant differences across all variables, regardless of clinical significance. However, such comparisons were later added at the request of the peer reviewers. For this, continuous variables were compared using ANOVA and categorical using chi-squared test. Prevalence of MetS was compared using a logistic regression model with MetS as the dependent variable and inflammatory arthritis as the independent variable, adjusted for age, sex, CRP and smoking status. Analysis was performed using Stata v15.
Results
The analysis included 498,961 participants, including 3,220 individuals with RA, 804 with PsA and 827 with axSpA. The remaining 494,110 participants without an IA diagnosis served as controls. The RA group were older (mean age 59.8 years), more likely to be female (71.2%) and had a higher mean CRP (6.5 mg/L, SD 9.3) compared to controls and other IAs.
The prevalence of MetS was higher in all IA groups compared to controls (31.8%), with rates of 43.4% in RA, 42.3% in PsA, and 37.1% in axSpA. Detailed results are presented in Table 1. Notably, the proportion of participants meeting the high blood pressure criteria was similar across IA groups at approximately 80%. Dysglycaemia was most prevalent in RA (21.7%), followed by PsA (20.9%). Similarly, an elevated waist circumference was more common in the RA and PsA groups (45.2% and 44.0%, respectively), but the AxSpA group (33.3%) was similar to the controls (33.6%). Derangement of HDLs was most common in RA (44.5%), whereas attainment of the triglyceride criterion was highest in PsA (48.2%). Significant differences were detected between groups across all demographic and metabolic variables.
Logistic regression analysis demonstrated increased odds of MetS in RA (OR 1.15; 95% CI 1.07, 1.24; p < 0.001) and PsA (OR 1.31; 95% CI 1.13, 1.52; p < 0.001) but decreased odds of MetS in axSpA (OR 0.82; 95% CI 0.70, 0.96; p = 0.012) compared to controls, after adjusting for age, sex, CRP and smoking status (Fig. 1).
Table 1
Baseline demographics and prevalence of MetS and its components
Controls | RA | PsA | axSpA | |
|---|---|---|---|---|
n (%) | 494,110 (99.0) | 3,220 (0.6) | 804 (0.2) | 827 (0.2) |
Mean age at recruitment (SD) | 56.5 (8.1) | 59.8 (6.9) | 56.9 (7.4) | 57.6 (7.5) |
Male, n (%) | 225,731 (45.7%) | 927 (28.8%) | 421 (52.4%) | 610 (73.8%) |
CRP (mg/L), mean (SD) | 2.6 (4.3) | 6.5 (9.3) | 5.0 (7.7) | 6.1 (8.2) |
Body mass index (BMI), mean (SD) | 27.4 (4.8) | 28.2 (5.5) | 28.8 (5.3) | 27.4 (4.6) |
Current smoker, n (%) | 51,937 (10.5%) | 390 (12.1%) | 72 (9.0%) | 98 (11.9%) |
Previous smoker, n (%) | 169,676 (34.4%) | 1379 (42.9%) | 331 (41.2%) | 354 (42.9%) |
Never smoker, n (%) | 269,615 (54.7%) | 1421 (44.2%) | 395 (49.2%) | 372 (45.0%) |
Metabolic syndrome, n (%) | 156,648 (31.8%) | 1395 (43.4%) | 340 (42.3%) | 306 (37.1%) |
Fulfilling waist circumference criteria, n (%) | 165,351 (33.6%) | 1447 (45.2%) | 353 (44.0%) | 274 (33.3%) |
Fulfilling high BP criteria, n (%) | 362,086 (73.3%) | 2596 (80.6%) | 639 (79.5%) | 660 (79.8%) |
SBP > 130 mmHg or DBP > 85 mmHg, n (%) | 334,082 (67.6%) | 2280 (70.8%) | 574 (71.4%) | 585 (70.7%) |
On antihypertensives, n (%) | 108,693 (38.7%) | 1206 (68.8%) | 264 (49.3%) | 263 (40.0%) |
Previous diagnosis of HTN, n (%) | 132,783 (27.0%) | 1199 (37.4%) | 308 (38.5%) | 288 (35.0%) |
Fulfilling glucose criteria, n (%) | 74,925 (15.2%) | 697 (21.7%) | 168 (20.9%) | 134 (16.2%) |
Baseline glucose ≥ 5.6 mmol/L, n (%) | 63,587 (15.1%) | 556 (20.5%) | 145 (20.7%) | 110 (15.4%) |
Previous diagnosis of T2DM, n (%) | 25,726 (5.2%) | 311 (9.7%) | 56 (7.0%) | 54 (6.5%) |
On antidiabetic medication, n (%) | 18,523 (3.7%) | 223 (6.9%) | 47 (5.8%) | 37 (4.5%) |
Fulfilling triglycerides criteria, n (%) | 201,429 (41.9%) | 1419 (44.8%) | 379 (48.2%) | 359 (45.2%) |
Fulfilling cholesterol criteria, n (%) | 146,512 (31.3%) | 1385 (44.5%) | 293 (38.9%) | 287 (37.7%) |
HDL cholesterol < 1 mmol/L (m) or < 1.293 mmol/L (w), n (%) | 83,620 (19.8%) | 767 (28.3%) | 181 (25.8%) | 144 (20.2%) |
On cholesterol-lowering medication, n (%) | 84,543 (27.4%) | 896 (34.4%) | 160 (33.7%) | 188 (50.8%) |
Fig. 1
Logistic regression of metabolic syndrome and inflammatory arthritis, adjusted for age, sex, CRP, and smoking history. Definitions: axSpA = axial spondyloarthritis; PsA = psoriatic arthritis; RA = rheumatoid arthritis
Discussion
By directly comparing the prevalence of MetS and its components across RA, PsA, axSpA and controls, this study revealed several notable findings. Firstly, despite MetS being classically associated with PsA [15], RA had a comparable prevalence. Likewise, the odds of comorbid MetS were elevated for both RA and PsA. In contrast, the prevalence of MetS in axSpA was lower than in other IAs, and there were decreased odds of comorbid MetS compared to controls when adjusting for confounders. The prevalence of hypertension and deranged triglycerides and cholesterol was similar in axSpA compared to other IAs, but the prevalence of an elevated waist circumference and dysglycaemia was comparable to controls, resulting in a lower prevalence of MetS overall. Our results also highlight the general prevalence of MetS, with almost a third of our control group meeting the criteria.
Previous studies have reported variable MetS prevalence in IA populations. A large meta-analysis by Hallajzadeh et al. estimated the prevalence of MetS in RA to range from 14.32 to 37.83%, with an overall pooled prevalence of 30.65% [16]. Estimates of the prevalence of MetS in PsA vary, with meta-analyses reporting a range of 29–46% [10, 17]. Data on MetS prevalence in axSpA are limited, with studies estimating it to be between 14% and 34.9% [11, 18, 19]. In a recent cross-sectional study by Guła et al. comparing comorbidities between RA, PsA and axSpA, the axSpA group showed the lowest prevalence of hypertension, obesity, dyslipidaemia and diabetes mellitus, supporting our findings of a lower burden of metabolic comorbidities in these patients compared to other IAs [20]. In the general population, the prevalence of MetS based on the ATP-III definition has been reported as 25.4% globally and 25.3% in Europe [6]. Notably, in our study, the prevalence of MetS and central obesity was similar in RA and PsA.
Although more commonly associated with PsA, MetS is a well-recognised consequence of RA, secondary to chronic inflammation, comorbidities, lifestyle factors and medications [21]. Our findings suggest that MetS prevalence in RA may approximate that of PsA outside of specialist rheumatology services. As disease severity is closely linked to MetS in PsA, these patients may be over-represented in tertiary services compared to other settings [22]. Conversely, the prevalence of MetS may differ with disease duration, particularly as obesity often precedes the diagnosis of PsA [2].
Anzeige
Our estimates for MetS in RA and axSpA are higher than those reported in previous studies. Notably, although the odds ratio for PsA was higher when compared to RA, this is due to a lack of precision and should not be over-interpreted. Comparisons across studies are limited due to variations in geographical location, disease classification criteria and MetS definition. The prevalence of MetS within the same population can vary substantially based on the criteria for MetS used. Furthermore, the performance of specific criteria varies by the ethnicity of the study population [23]. Our study is novel as it employed a consistent definition of MetS within the same population, encompassing participants with different IAs, enabling a direct comparison of the prevalence and odds of MetS by IA type.
Multiple mechanisms may explain the strong association between IA and MetS. Both obesity and MetS are characterised by chronic low-grade inflammation driven by adipokine release from visceral fat [8, 10]. Systemic inflammation, secondary to inflammatory joint disorders, is known to promote insulin resistance, atherosclerosis and endothelial dysfunction [3, 10]. In addition, the pharmacological therapies used for IA can impact MetS. For instance, corticosteroids are associated with weight gain [24], leflunomide is linked to hypertension [25] and Janus kinase inhibitors have been associated with dyslipidaemia [26, 27]. Furthermore, participation in exercise among IA patients is often limited due to both disease-related and psychosocial barriers [28].
MetS in IA is associated with both increased risk of CVD [8, 10, 11] and treatment non-response [13, 22]. Identifying MetS in IA, particularly RA and PsA, is critical to ensure the safe prescribing of DMARDs and provides an opportunity to offer lifestyle interventions or pharmacological treatments (for example, antihypertensives). In particular, weight loss in RA [29] and PsA [22, 30] has shown benefit on disease activity and with the advent of GLP-1 agonists in obesity management, there may be utility for these drugs as an adjunctive therapy in the future [15].
Our study’s strengths include using a large prospective data set of nearly half a million individuals recruited from the general population. Recruiting from the general population, rather than exclusively from specialist centres, enhances the translatability of our results to individuals presenting to primary and secondary care. Importantly, this is one of the first studies directly comparing the prevalence of MetS in RA, PsA and axSpA using the same study population and diagnostic criteria, enabling direct comparisons between these groups.
Anzeige
This study has several limitations. First, the UK Biobank population tends to be healthier and more affluent than the general population. Nonetheless, we observed a high prevalence of MetS within this cohort, suggesting the current literature may be underestimating the scale of this disease in the broader rheumatology and general populations. Second, there is a risk of disease misclassification when using coded or self-reported data, although gender distribution is generally comparable to that reported for these diseases. Although we were able to adjust for several relevant confounders, we were unable to adjust for disease activity and treatment history, which may independently impact the risk of MetS. Lastly, the UK Biobank study did not collect fasting blood samples, which may increase the proportion of participants meeting MetS criteria based on random blood results.
In conclusion, this study provides valuable insights into MetS in IA by directly comparing across RA, PsA, and axSpA. The prevalence and odds of MetS were elevated in those with RA and PsA, but decreased in axSpA. Future research should focus on optimal strategies for managing both MetS and IA as part of a comprehensive treatment approach.
Acknowledgements
This research has been conducted using the UK Biobank Resource under Application Number 72723.
Declarations
Conflict of interest
All authors declare that they have no conflict of interest.
Anzeige
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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 changes were made. 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/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
