Metabolic syndrome and knee osteoarthritis: associations with disease stage and pain

Abstract

Objectives

Metabolic Syndrome (MetS) is associated with OA progression and pain. However, its relationship with outcomes in early-stage disease remains unclear. We assessed associations of MetS with radiographic stage and knee pain in individuals with symptomatic knee OA.

Methods

We analysed cross-sectional baseline data from 251 participants in the Western Ontario Registry for Early Osteoarthritis (WOREO) Knee Study. Associations between MetS and radiographic stage were assessed using modified Poisson regression. Associations with knee pain, measured by the Knee Injury and Osteoarthritis Outcome Score (KOOS) pain scale, were evaluated using multiple linear regression accounting for within-participant correlation. Analyses were stratified by OA stage and adjusted for body mass index (BMI).

Results

MetS was associated with a higher prevalence of late-stage knee OA [adjusted prevalence ratio 1.45 (95% CI 1.16, 1.81)]. After BMI adjustment, hypertension and haemoglobin A1c remained associated with late-stage disease. Participants with MetS reported worse knee pain, with lower KOOS pain scores [β = −14.70 (95% CI −20.92, −8.47)], particularly in early-stage disease [β = −14.15 (95% CI −20.37, −7.93)]. Dyslipidaemia (elevated triglycerides and reduced high-density lipoprotein cholesterol were also associated with worse pain in early-stage disease.

Conclusion

MetS and its components are associated with greater radiographic severity and clinically meaningful worsening of pain, especially in early-stage knee OA, independent of BMI. These findings suggest that early identification and management of metabolic disease may represent a modifiable target to improve OA outcomes.

Key messages.

• Metabolic syndrome (MetS) and its components increase the risk of late-stage knee OA, independent of BMI.

• MetS and its components are associated with worse pain in early-stage knee OA, independent of BMI.

• Interventional studies should assess whether treating MetS slows disease progression in early-stage knee OA.

Introduction

Epidemiological data have confirmed that metabolic syndrome (MetS) is a risk factor for OA [1–6]. The cumulation of evidence has led to the conceptualization that MetS-associated OA (MetS-OA) may be a specific phenotype of OA.

The mechanisms linking MetS and OA are not fully understood. Historically, it was thought that the two shared common but independent risk factors, such as obesity [7]. However, recent evidence suggests that systemic inflammation caused by MetS may accelerate the development and progression of OA [8]. Individual MetS components such as insulin resistance, the presence of oxidized low-density lipoprotein, and hypertension have been linked to mechanisms involved in joint organ damage, such as synovial inflammation, ectopic bone formation, and subchondral ischaemia [7].

Consequently, it has been proposed that treating MetS and its components could improve OA outcomes, such as reducing radiographic joint damage and pain. While it is well established that MetS increases the risk of OA overall, there is limited evidence linking MetS to worse outcomes in early stage disease. Addressing this gap would strengthen the evidence to recommend early screening and treatment of MetS in patients with knee OA [9, 10]. Accordingly, our objective was to investigate the association of MetS with early- and late-stage radiographic damage from knee OA (ES-KOA, LS-KOA), and with OA-related pain.

Methods

Study population

This study was approved by the Western University Research Ethics Board. A total of 251 participants were included in this cross-sectional analysis of baseline data from the Western Ontario Registry for Early Osteoarthritis (WOREO) Knee Study, a single-centre, multi-clinic (St. Joseph’s Health Care London Rheumatology Centre, Fowler Kennedy Sport Medicine Clinic, and Rorabeck Bourne Joint Replacement Clinic) cohort designed to investigate the clinical, biomechanical, and pathophysiological features of early- and late-stage radiographic joint damage. All individuals referred to a rheumatologist or orthopaedic surgeon for assessment of KOA were invited to participate. Eligibility criteria included age >18 years with a diagnosis of KOA based on clinical assessment by a rheumatologist or orthopaedic surgeon and frequent knee symptoms defined as pain, aching or stiffness on most days for the last 4 weeks within the past year. Exclusion criteria were previous total knee arthroplasty, any history of inflammatory arthritis (e.g. RA, SLE, PsA, AS, enteropathic arthritis, vasculitis, SSc, gout, calcium pyrophosphate arthropathy and similar diagnoses), DMARD use, oral corticosteroid use or any knee procedure or corticosteroid injection within 6 months prior to the enrolment assessment. For participants meeting eligibility criteria, demographic data (age, sex, racial background and household income), medications (antihypertensives and antihyperglycaemics), biometrics, radiographic results and patient-reported pain measures were recorded at the enrolment visit. The results of a lipid panel and haemoglobin A1c (HbA1c) performed within 1 year of the enrolment visit were also recorded.

MetS criteria

MetS was defined as meeting three or more of the following five criteria: obesity, defined as a waist circumference ≥102 cm in males and ≥88 cm in females; if abdominal circumference was not recorded, body mass index (BMI) ≥30 kg/m² was used for this criterion (n = 98); elevated triglycerides (TG), defined as ≥1.7 mmol/l; reduced high-density lipoprotein cholesterol (HDL-C), defined as <1.0 mmol/l in males and <1.3 mmol/l in females; hypertension, defined as ≥130 mmHg systolic and/or ≥85 mmHg diastolic or antihypertensive medication (angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, thiazide or thiazide-like diuretics, beta blockers, calcium channel blockers and mineralocorticoid receptor antagonists) in a patient with a history of hypertension; and elevated HbA1c, defined as ≥6.5% or antihyperglycaemic medication (metformin, dipeptidyl peptidase 4 inhibitors, glucagon-like peptide 1 agonists, sodium–glucose cotransporter-2 inhibitors, sulfonylureas, thiazolidinediones and insulin) in a patient with a history of diabetes.

Radiographic damage assessment

Standard fixed-flexion postero-anterior and antero-posterior full-limb standing hip to ankle radiographs were acquired at study enrolment. Radiographic damage grade was assigned to each knee using the Kellgren–Lawrence (KL) grading scale, which classifies radiographic damage severity from grade 0 (none) to 4 (severe) [11]. Radiographs were read and KL grades were assigned by one of three raters. Trainee raters were trained by and calibrated to a rheumatologist (C.T.A.) through formal instruction, followed by radiograph training and calibration image decks. Raters had substantial to excellent intrarater reliability (κ = 0.69–0.95) and substantial interrater reliability [12] (κ = 0.75) for KL grading. Radiographic stage (early/late) was assigned to each patient. ES-KOA was defined as a KL grade ≤2 in both knees and LS-KOA was defined as a KL grade ≥3 in one or both knees. Radiographic data were available for 500 knees. Two knees were excluded due to missing radiographs.

Patient-reported measures

Participants completed the Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaire for each knee separately at study enrolment. KOOS pain data were available for 483 of 500 knees (96%) across the 251 participants. The KOOS is a validated and reliable instrument for assessing individuals with KOA [13, 14]. It includes 42 items in five separately scored subscales. The pain subscale, which includes nine items, was used for this analysis. It is scored from 0 to 100, where lower scores indicate more severe pain. The minimal clinically important difference (MCID) for the KOOS pain subscale for individuals with KOA has been estimated at 10-12 points, representing the smallest change perceived by patients as clinically meaningful [15].

Statistical analyses

Participant characteristics

Descriptive statistics were used to summarize participant characteristics [mean (s.d.)] for continuous variables and proportions for categorical variables].

Radiographic stage analysis

To determine the association between MetS and OA radiographic stage, the primary outcome was LS-KOA vs ES-KOA (reference category). The unit of analysis was the participant. Modified Poisson regression with robust variance estimation was used to estimate prevalence ratios (PRs). Both unadjusted and adjusted models were fitted. Adjusted models included age (continuous), sex, racial background (White vs non-White) and household income (<$25 000, $25 001–59 999, $60 000–89 999 and >$90 000 CAD).

Pain analysis

To examine the association between MetS and knee pain, the outcome was the KOOS pain score (continuous). The unit of analysis was the knee. Multiple linear regression with generalized estimating equations (GEEs) was used to account for within-participant correlation between knees. Models were adjusted for age (continuous), sex, and radiographic stage. An interaction term between MetS and radiographic stage was included to assess effect modification; when a statistically significant effect was detected, analyses were repeated, stratified by radiographic stage.

MetS component and BMI-adjusted analyses

Secondary analyses examined individual MetS components. For components demonstrating significant associations with OA radiographic stage or pain, additional models were fitted with further adjustment for BMI to distinguish metabolic effects from biomechanical loading, as obesity may confound associations between metabolic factors and OA outcomes. BMI was included as a three-level ordinal variable (≤30, ≥30–<35 and >35 kg/m²).

Results are presented as PRs or regression coefficients (β) with 95% CIs. Statistical significance was defined as a two-sided P-value <0.05. All analyses were performed using R version 4.4.0 (packages sandwich and geepack; R Foundation for Statistical Computing, Vienna, Austria) [16, 17].

Results

Participant characteristics

A total of 251 participants (500 knees) were included. We reported baseline demographics and clinical characteristics for the total sample, and by OA radiographic stage (Table 1). There were 103 (41.0%) participants with ES-KOA (defined as a KL grade ≤2 in both knees) and 148 (59.0%) with LS-KOA (defined as a KL grade ≥3 in one or both knees). Of the participants with LS-KOA, 94 (63.5%) had late-stage disease in both knees and 54 (36.5%) had late-stage disease in one knee. For two of the participants with late-stage disease in one knee, radiographic data was not available for the other knee. In total, there were 258 knees with ES-KOA and 242 knees with LS-KOA (Fig. 1).

Table 1.

Baseline and clinical characteristics for total cohort and separated by OA radiographic stage.

Characteristics	Total cohort (N = 251; n = 500 knees)	ES-KOA (N = 103; n = 258 knees)	LS-KOA (N = 148; n = 242 knees)
Age, years, mean (s.d.)	63.26 (10.39)	60.60 (11.29)	65.12 (9.31)
Sex, n (%)
Male	94 (37.45)	32 (31.07)	62 (41.89)
Female	157 (62.55)	71 (68.93)	86 (58.11)
BMI, kg/m2, mean (s.d.)	32.05 (7.46)	29.04 (6.20)	34.15 (7.57)
Race, n (%)
Non-White	18 (7.17)	9 (8.74)	9 (6.08)
White	186 (74.10)	74 (71.84)	112 (75.68)
Household income, n (%)
<$25 000	18 (25.90)	6 (5.83)	12 (8.11)
$25 001–$59 999	65 (25.90)	27 (26.21)	38 (25.68)
$60 000–$89 999	55 (21.91)	26 (25.24)	29 (19.59)
≥$90 000	99 (39.44)	37 (35.92)	62 (41.89)
Participants with MetS, n (%)	110 (43.82)	29 (28.16)	81 (54.36)
KL grade, n (%)
Early stage (KL ≤2)	258 (51.6)	258 (100)	–
Late stage (KL ≥3)	242 (48.4)	–	242 (100)
KOOS pain subscale, mean (s.d.)	62.83 (22.94)	70.30 (22.23)	55.34 (21.17)

Figure 1.

Study population flow diagram. Number of patients and knees included in OA radiographic stage and pain analyses.

Association between MetS and OA radiographic stage

MetS was significantly associated with KOA radiographic stage [PR 1.55 (95% CI 1.26, 1.90)]. The association remained significant after adjusting for all covariates, including age, sex, racial background, and household income [adjusted PR (aPR) 1.45 (95% CI 1.16, 1.81)]. Participants with MetS had a 45% higher prevalence of LS-KOA [aPR 1.45 (95% CI 1.16, 1.81)] compared with participants without MetS (Table 2). We therefore looked for an association between each component of MetS and OA stage in secondary analyses. We found that reduced HDL-C [aPR 1.34 (95% CI 1.04, 1.72)], hypertension [aPR 1.85 (95% CI 1.22, 2.81)] and elevated HbA1c [aPR 1.58 (95% CI 1.27, 1.95)] were associated with increased prevalence of late-stage disease (Table 2). After adjusting for BMI, hypertension [aPR 1.14 (95% CI 1.01, 1.29)] and elevated HbA1c [aPR 1.13 (95% CI 1.03, 1.23)] remained associated with an increased prevalence of late-stage disease (Table 2).

Table 2 Association of MetS and its components with knee OA radiographic stage.

Factor	Unadjusted		Adjusted 1a		Adjusted 2b
	PR (95% CI)	P-value	aPR (95% CI)	P-value	aPR (95% CI)	P-value
MetS	1.55 (1.26, 1.90)	<0.001	1.45 (1.16, 1.81)	0.002	–	–
MetS components
Obesity	1.03 (0.94, 1.12)	0.59	1.08 (0.84, 1.39)	0.54	–	–
Elevated TG	1.10 (1.02, 1.18)	0.02	1.03 (0.95, 1.13)	0.49	–	–
Reduced HDL-C	1.11 (1.02, 1.20)	0.01	1.34 (1.04, 1.72)	0.02	1.08 (0.99, 1.19)	0.57
Hypertension	1.26 (1.14, 1.40)	<0.001	1.85 (1.22, 2.81)	0.004	1.14 (1.01, 1.29)	0.04
Elevated HbA1c	1.22 (1.14, 1.31)	<0.001	1.58 (1.27, 1.95)	<0.001	1.13 (1.03, 1.23)	0.01

Results of modified Poisson regression used to assess the association of MetS and its components with OA radiographic stage (early/late). The analyses were adjusted for age, sex, background and household income level. For the MetS components that demonstrated a significant association with OA radiographic stage after adjusting for covariates, the analyses were further adjusted for BMI as a three-level ordinal variable. Results are reported as PRs and aPRs with 95% CIs (CI). Statistical significance was set at P < 0.05 (two-sided).

^aAdjusted for age, sex, background and household income.

^bFurther adjusted for BMI.

Association between MetS and patient-reported knee pain

We then looked for an association between MetS and worse knee pain. After adjusting for age, sex, and radiographic stage, MetS was significantly associated with worse knee pain. The mean KOOS pain score was much lower (indicating more severe pain) in participants with MetS compared with without MetS [β = −14.70 (95% CI −20.92, −8.47)] (Table 3). We also found that female patients had more severe pain compared with male patients after accounting for MetS and OA stage [β = −4.21 (95% CI −8.38, −0.05)] (Supplementary Table S1).

Table 3 Association of MetS and its components with KOA pain measured by KOOS.

Factors	Unadjusted		Adjusteda		Adjusted 2b
	β (95% CI)	P-value	β (95% CI)	P-value	β (95% CI)	P-value
MetS	−10.61 (−15.0, −6.22)	<0.001	−14.70 (−20.92, −8.47)	<0.001	–	–
MetS components
Obesity	−6.14 (−10.9, −1.43)	0.01	−10.10 (−16.08, −4.13)	0.0009	–	–
Elevated TG	−6.63 (−11.3, −2.01)	0.005	−11.49 (−18.21, −4.77)	0.0008	−8.95 (−15.58, −2.33)	0.008
Reduced HDL-C	−6.22 (−11.1, −1.37)	0.01	−13.11 (−19.48, −6.73)	<0.001	−10.78 (−17.36, −4.20)	0.001
Hypertension	−9.29 (−14.80, −3.75)	0.001	−8.28 (−14.53, −2.03)	0.009	−4.87 (−10.94, 1.21)	0.12
Elevated HbA1c	−9.76 (−15.30, −4.18)	<0.001	−9.52 (−20.12, 1.09)	0.08	–	–

Results of multivariable linear regression with GEEs used to assess the association of MetS and its components with OA pain measured by the KOOS. Analyses were adjusted for age, sex and radiographic stage. For MetS components significantly associated with worse OA pain after adjusting for covariates, analyses were further adjusted for BMI as a three-level ordinal variable. Results are reported as β coefficients with 95% CIs. Statistical significance was set at P < 0.05 (two-sided).

^aAdjusted for age, sex and radiographic stage with interaction term included for radiographic stage.

^bFurther adjusted for BMI.

We tested for effect modification by radiographic stage using an interaction term between MetS and radiographic stage. Since the interaction term was significant (P = 0.006) (Supplementary Table S2), analyses were stratified by OA radiographic stage (Table 4). In ES-KOA, MetS was strongly associated with worse pain [β = −14.15 (95% CI −20.37, −7.93)]. In LS-KOA, the association was smaller and not statistically significant [β = −2.96 (95% CI −8.41, 2.50)].

Table 4 Association of MetS, reduced HDL-C and elevated TG with OA pain stratified by OA radiographic stage.

Factors	Early stage KOA				Late-stage KOA
	Unadjusted		Adjusted		Unadjusted		Adjusted
	β (95% CI)	P-value	β (95% CI)	P-value	β (95% CI)	P-value	β (95% CI)	P-value
MetSa	−14.11 (−20.9, −7.35)	<0.001	−14.15 (−20.37, −7.93)	<0.001	−3.35 (−9.09, 2.39)	0.25	−2.96 (−8.41, 2.50)	0.29
MetS components
Reduced HDL-Cb	−13.10 (−19.30, −6.91)	<0.001	−9.32 (−16.26, −2.38)	0.009	2.07 (−4.62, 8.76)	0.54	3.36 (−2.82, 9.54)	0.29
Elevated TGb	−10.3 (−17.0, −3.63)	0.003	−7.21 (−14.01, −0.42)	0.04	−0.79 (−6.67, 5.09)	0.79	−0.50 (−6.35, 5.36)	0.87

Results of multivariable linear regression with GEEs used to assess the association of MetS, reduced HDL-C and elevated TG with OA pain as measured by the KOOS stratified by KOA radiographic stage (early/late). This analysis was performed due to the significant interaction term between MetS, reduced HDL-C, elevated TG and radiographic stage. Results are reported as β coefficients with 95% CIs. Statistical significance was set at P < 0.05 (two-sided).

^aAdjusted for age, sex and radiographic stage with interaction term included for radiographic stage.

^bFurther adjusted for BMI.

Secondary analyses examined associations between each MetS component and worse knee pain (Table 3). Obesity [β = −10.10 (95% CI −16.08, −4.13)], elevated TG [β = −11.49 (95% CI −18.21, −4.77)], reduced HDL-C [β = −13.11 (95% CI −19.48, −6.73)] and hypertension [β = −8.28 (95% CI −14.53, −2.03)] were associated with worse pain in adjusted models.

For MetS components demonstrating significant associations with worse pain (elevated TG, reduced HDL-C and hypertension), additional models adjusting for BMI were fitted. After BMI adjustment, elevated TG [β = −8.95 (95% CI −15.58, −2.33)] and reduced HDL-C [β = −10.78 (95% CI −17.36, −4.20)] remained significantly associated with worse pain, whereas the association with hypertension was attenuated.

Because interaction terms were significant for elevated TG (P = 0.04) and reduced HDL-C (P = 0.001) with radiographic stage (Supplementary Table S2), these analyses were further stratified by OA radiographic stage. Associations remained significant in early stage disease but not in late-stage disease (Table 4).

Discussion

Although it is well established that MetS increases the risk of OA in general [1–6], evidence linking MetS to worse outcomes in early-stage disease is limited. This is a key knowledge gap. Current clinical management guidelines for OA do not include treat-to-target principles for MetS or its components, even though this is well established in other chronic organ-based diseases such as cardiovascular and renal disease. We found that MetS increases the risk of worse pain and/or radiographic damage in patients with early-stage disease, highlighting an exciting window of opportunity that may exist to modify long-term OA outcomes through treatment of MetS.

We demonstrated that MetS is strongly associated with a high prevalence of LS-KOA and significantly worse KOOS pain scores. Importantly, the magnitude of the association between MetS and KOOS pain exceeded the established MCID for the KOOS pain subscale in KOA (≈10-12 points), indicating that these differences are not only statistically significant but also clinically meaningful. Interestingly, these associations were particularly prominent in early-stage disease. Although we are not able to directly assess causality with this study design, taking these findings in context with existing literature provides strong circumstantial evidence that MetS may contribute to worse pain in early-stage OA, and increased risk of progression to late-stage radiographic joint damage.

There are several possible mechanisms through which MetS may directly or indirectly affect outcomes in joints with OA. Increased body weight, a hallmark of MetS, clearly causes abnormal joint loading that accelerates structural damage and pain [18]. However, the impact of MetS likely extends beyond these biomechanical consequences. MetS drives systemic inflammation, exposing joints to elevated levels of pro-inflammatory cytokines and adipokines, which contribute to synovial damage and heightened pain sensitivity [8]. Additionally, MetS promotes macrophage activation within the synovium, further compromising cartilage integrity [19]. Metabolic dysregulation at the joint level also increases oxidative stress and facilitates the accumulation of advanced glycation end-products, which bind to receptors on chondrocytes and synovial cells, triggering inflammatory cascades and intensifying pain [20]. Furthermore, MetS may heighten pain perception in OA by amplifying central sensitization, ultimately worsening the overall disease burden [21].

Several potential implications arise from linking MetS to worse outcomes in patients with early stage OA. Early in the disease course, more severe pain can hinder patients from participating in physical activity and rehabilitation, which are both important for maintaining muscle strength and mobility [22, 23]. This can result in more severe disability as the disease progresses. Greater disability leads to a more sedentary lifestyle, making the components of MetS even more difficult to control and thus further increasing the risk of developing late-stage disease and cardiometabolic complications, including cardiovascular mortality [24]. Moreover, patients who have progressed to late-stage radiographic disease are at high risk of requiring arthroplasty. Since MetS components are associated with increased rates of postoperative cardiovascular complications [25, 26], venous thromboembolism [27–29], infection [30–32], wound-healing complications [33] and arthroplasty revision [34], comprehensive management of MetS and its components may be an even greater priority in patients with early stage disease. Considering this, our results uncover an urgent need to investigate the effects of treat-to-target strategies for MetS on KOA outcomes in patients with early stage KOA. Indeed, the best window of opportunity to modify KOA outcomes by treating MetS may be early in the disease course, shortly after the onset of frequent knee symptoms and before the development of severe radiographic damage. By identifying early signs of synovial joint disease as an end-organ manifestation of MetS, his proactive (secondary prevention) strategy may herald the potential to improve pain, slow radiographic progression and reduce perioperative risks.

In our analysis of each component of MetS, we found that hypertension, elevated HbA1c and reduced HDL-C were all independently associated with an increased risk of late-stage radiographic KOA. The association remained clinically meaningful for hypertension and elevated HbA1c, even after adjusting for BMI, which we did to account for the possibility that these associations were simply due to increased body weight causing greater knee joint loading. We also found that reduced HDL-C, elevated TG, hypertension, and obesity were all associated with more severe knee pain. In particular, dyslipidaemia (including both reduced HDL-C and elevated TG) remained associated with worse knee pain in early stage disease after adjusting for BMI. While the mechanisms mediating the relationship between metabolic derangement and worse features of OA remain unclear, our findings suggest a role for deranged lipid metabolism, insulin resistance, and hypertension beyond the biomechanical effects of increased loading due to obesity.

Various mechanisms for the effects of each MetS component on OA outcomes have been proposed for several years. Subchondral ischaemia caused by hypertension-induced blood vessel narrowing over time may lead to joint damage and OA progression [35]. Chronic hyperglycaemia has a pro-inflammatory effect on chondrocytes, inducing oxidative stress that can lead to degradation of joint tissue [36]. Low HDL-C has also been linked to oxidative stress [37]. Additionally, there is evidence that higher HDL-C levels in the synovial fluid have a protective effect [38], with one study showing a lower incidence of hand OA in individuals with higher serum HDL-C levels [39], which aligns well with the findings of our study. The pro-inflammatory effects of metabolic derangements may act as noxious stimuli, leading to increased nociceptor activation, which could explain the association we found between MetS and worse knee pain in early stage disease. As the disease progresses and pain typically becomes more severe, the increased pain related to MetS may be less clinically meaningful. Since radiographic joint damage (damage to bone and cartilage) accumulates in late-stage disease and may directly contribute to worse pain experiences, we can speculate that an overriding effect of joint damage on pain may partly explain why we observed stronger associations between MetS and worse pain in early stage disease. Pain experiences in OA are complex, with multiple contributing factors including sex differences and pain sensitization, among others, as evidenced by our finding that female patients report worse knee pain even after accounting for MetS and OA stage.

Our findings complement the results of the Framingham Osteoarthritis Study by Niu et al. [40], which demonstrated that pre-existing MetS and its components are associated with an increased risk of developing radiographic and symptomatic KOA. In their study, these associations were attenuated after adjusting for BMI, consistent with previous research [41]. In our study, after adjusting for BMI, patients with hypertension and elevated HbA1c had a much higher prevalence of LS-KOA. Additionally, patients with reduced HDL-C and elevated TG experienced significantly worse knee pain in early stage disease. Our findings suggest that metabolic derangements contribute to KOA pathophysiology beyond the biomechanical impact of increased body weight, which leads to abnormal joint loading and subsequent joint damage. The association between obesity and OA in non-weight-bearing joints, such as the hands and wrists, further supports the existence of additional metabolic mechanisms due to obesity/MetS [42]. Previous studies have implicated leptin, a hormone produced by adipose tissue, in the development and progression of OA due to its pro-inflammatory properties, among other mechanisms [43]. Although the associations we found between MetS components and both radiographic damage and worse pain remained clinically significant after adjusting for BMI, the effects were reduced, suggesting that MetS induces a combination of biomechanical and metabolic stresses on synovial joints. However, BMI is not an ideal surrogate measure to use when adjusting for biomechanical joint loading, as it may also account for the metabolic effects of obesity, thus underrepresenting the contribution of metabolic derangements to OA pathophysiology [44]. Unfortunately, there is no readily available surrogate for biomechanical loading that can easily be used to account for the effects of increased joint loading in isolation. Thus, BMI continues to be used as a surrogate for lower extremity loading in many studies in lieu of a more direct biomechanical measurement.

Our findings are consistent with prior cross-sectional studies, including the work of Andersson et al. [45], which also demonstrated an association between MetS and radiographic KOA in individuals with earlier symptomatic disease. As with that study, our study design does not permit causal inference between MetS and OA outcomes. Several additional limitations should be considered. Patient-reported pain measures may be subject to recall or reporting bias. Additionally, we did not have detailed information regarding the type or timing of analgesic medication use at the time of pain assessment, which may have influenced patient-reported pain scores and could introduce residual confounding. Although inflammatory rheumatic diseases were excluded, we did not systemically screen for other chronic pain conditions such as fibromyalgia or neuropathic pain syndromes, which may contribute to pain reporting and introduce residual confounding. We also did not include objective measures of physical function, which may provide additional context regarding the clinical impact of MetS on disease severity. Future studies incorporating functional outcomes alongside pain and radiographic measures may further clarify these relationships. Additionally, misclassification of MetS components is possible, as laboratory values were obtained within 1 year of enrolment rather than concurrently for all participants. Finally, because participants were recruited from specialty rheumatology and orthopaedic clinics, referral bias is possible. Individuals with older age and metabolic comorbidities may be more likely to be referred for specialty care, potentially overrepresenting MetS in this cohort and limiting generalizability to younger or community-based populations. These factors should be considered when interpreting our findings.

Despite these limitations, we observed consistent and clinically meaningful associations between MetS and both radiographic severity and pain, even after adjustment for BMI. These findings support the hypothesis that metabolic factors may contribute to worse OA outcomes and highlight the potential value of screening and targeted management of MetS in patients with early stage symptomatic KOA. Prospective and interventional studies are needed to determine whether treating MetS can modify disease progression and symptom burden.

Supplementary data

Supplementary data are available at Rheumatology Advances in Practice online.

Data availability

Some or all datasets generated during and/or analysed during the current study are not publicly available but are available from the corresponding author upon reasonable request.

Authors’ contributions

Jami Kronick (Conceptualization, Methodology, Investigation, Writing—original draft), Trevor B. Birmingham (Methodology, Writing—review & editing), Surim Son: Methodology, Formal analysis, Writing—review & editing), Andrew J. E. Appleton (Conceptualization, Methodology, Writing—review & editing, Supervision), and C. Thomas Appleton (Conceptualization, Methodology, Writing—review & editing, Supervision).

Funding

The WOREO Knee Study is supported by the Academic Medical Organization of Southwestern Ontario INN17-004.

Disclosure statement: The authors have declared no conflicts of interest.