Objective To compare the prevalence of metabolic syndrome (MetS) and the levels of related biomarkers in patients with psoriatic arthritis (PsA) and psoriasis without arthritis (PsC).
Methods This study compared patients with PsA and patients with PsC. The presence of MetS was determined. Serum levels of insulin, adiponectin and leptin were measured. The homeostasis model assessment for insulin resistance (HOMA-IR) was calculated. HOMA-IR, adiponectin and leptin were log-transformed. Continuous variables were compared using the t test and the χ2 test was used for discrete variables. Multivariate regression models were used to investigate the association of MetS and adiponectin with PsA compared to PsC after adjusting for potential confounding variables.
Results 203 PsA and 155 PsC patients were analysed. The prevalence of MetS was higher in PsA patients compared to those with PsC. However, this did not reach statistical significance (36.5% vs 27.1%, p=0.056). The levels of adipokines were significantly higher in PsA compared to PsC: adiponectin (8.8±5.2 vs 7.4±4.5 log (µg/ml), p=0.009) and leptin in women (3.1±0.8 vs 2.8±0.8, log (ng/ml), p=0.04). HOMA-IR was also higher in PsA (0.97±0.63 vs 0.68±0.81, p<0.001). No difference was observed in leptin levels in men. In multivariate regression analysis, PsA (p=0.04) and the psoriasis area and severity index score (p=0.02) were associated with MetS. Adiponectin was significantly associated with PsA (p=0.005), the use of anti-tumour necrosis factor α therapy (p=0.03) and active joint count (p=0.001).
Conclusions MetS and related adipokines correlated with an increased burden of skin and joint inflammation.
- Psoriatic Arthritis
Statistics from Altmetric.com
Psoriatic arthritis (PsA) is an inflammatory arthritis that affects 7–42% of patients with psoriasis.1 The chronic inflammatory nature of psoriasis and PsA predisposes patients to cardiovascular diseases and metabolic syndrome (MetS).2 MetS is a cluster of metabolically related cardiovascular risk factors that has become a major public health challenge worldwide.3 MetS predicts all-cause and cardiovascular mortality, diabetes mellitus and stroke,4 and is associated with systemic inflammation and pro-inflammatory cytokines.5
The prevalence of cardiovascular risk factors is increased in patients with psoriasis compared to the general population.6–8 Central obesity, a major component of MetS, is strongly associated with psoriasis and psoriasis severity.9–11 People with psoriasis tend to have atherogenic lipid profiles with high triglycerides and decreased high-density lipoprotein (HDL) cholesterol levels.12–14 The prevalence of cardiovascular risk factors and MetS is also increased in PsA patients compared to the general population.15–17 However, only a few studies have directly compared the burden of cardiovascular risk factors in patients with PsA and those with cutaneous psoriasis without arthritis (PsC). Our group has recently found that the burden of subclinical atherosclerosis, as measured by the carotid total plaque area, is higher in PsC patients compared to those with PsA.18
Adipose tissue produces a large number of metabolically active molecules that share functional and structural characteristics with cytokines and are therefore termed adipokines. Adiponectin is a regulator of insulin sensitivity and metabolism that has anti-inflammatory properties. It correlates negatively with obesity, insulin resistance and MetS.19 Leptin is another adipocyte-specific protein that effects many physiological and metabolic pathways, including the regulation of body weight, appetite, energy expenditure and immunity.20 Serum leptin levels reflect body fat mass. Leptin promotes the synthesis of type I pro-inflammatory cytokines and suppresses type II cytokines.21 Lower levels of adiponectin and higher levels of leptin were reported in patients with psoriasis compared to healthy controls.22 ,23 Leptin also correlated with the severity of psoriasis while adiponectin was negatively associated with the severity of psoriasis.24 To the best of our knowledge, these adipokines have not been evaluated in PsA.
In the present study we aimed to investigate several aspects of MetS by comparing its prevalence, the level of adipokines and measures of insulin resistance in patients with PsA and those with PsC. A secondary aim of the study was to assess the correlation between adipokines and clinical measures of joint and skin inflammation and damage. We hypothesised that as the burden of inflammation of both skin and joints is higher in patients with PsA, biomarkers that correlate with obesity and insulin resistance will be higher in patients with PsA compared to those with PsC.
Patients and setting
In this cross-sectional study patients with PsA were compared to patients with PsC. Consecutive adult PsA patients who satisfied the classification of psoriatic arthritis criteria25 were recruited from the University of Toronto psoriatic arthritis cohort that was established in 1978 as part of an ongoing prospective study. The patients are followed according to a standardised protocol every 6–12 months.26 The patients in the PsA clinic represent a wide spectrum of the disease that is related to the broad referral base of the clinic. The clinic serves as a primary, secondary and tertiary referral centre for PsA patients from the greater Toronto area and southern Ontario. Furthermore, patients are followed regularly, irrespective of their disease activity. Many of the patients were referred from dermatology clinics for the assessment of PsA, thus minimising ascertainment bias between the PsA and the psoriasis groups, because most of the psoriasis patients were also recruited from dermatology clinics. The reference group of PsC patients was recruited from the University of Toronto psoriasis cohort. The cohort was described in detail previously.27 Briefly, all potential study subjects have a diagnosis of psoriasis confirmed by a dermatologist and have been assessed by a rheumatologist to exclude a diagnosis of PsA. The cohort was established with the aim of studying risk factors for the development of PsA. Patients are recruited mainly from dermatology clinics and phototherapy centres. All participants are followed according to the same protocol as in the PsA cohort, and are assessed annually for symptoms or signs of arthritis. If inflammatory arthritis is diagnosed, the subject is considered to have developed the outcome of interest and is censored. This process ensures that all of the psoriasis cohort's patients are free of arthritis.
The study was approved by the University Health Network Research Ethics Board and all patients gave their informed consent.
Cardiovascular risk factor assessment
MetS was defined as the presence of at least three of the following: (1) increased triglyceride levels of 1.7 mmol/l (150 mg/dl) or greater or specific treatment for this lipid abnormality; (2) reduced HDL-cholesterol of less than 1.0 mmol/l (40 mg/dl) in men or less than 1.3 mmol/l (50 mg/dl) in women or specific treatment for this lipid abnormality; (3) elevated blood pressure: systolic blood pressure of 130 mm Hg or greater or diastolic blood pressure of 85 mm Hg or greater or treatment of previously diagnosed hypertension; (4) raised fasting plasma glucose of 5.6 mmol/l (100 mg/dl) or greater or previously diagnosed type 2 diabetes; (5) central obesity defined by a waist circumference of 88 cm or greater for women or 102 cm or greater for men.28 Height and weight were measured and body mass index (BMI) was calculated. Smoking status and alcohol consumption were recorded. Blood samples were collected after 12 h of overnight fasting and analysed for glucose, lipid profile, high sensitivity C-reactive protein (hsCRP), adiponectin, leptin and insulin levels. Serum insulin was measured using an ELISA kit (Invitrogen, Camarillo, California, USA). The homeostasis model assessment for insulin resistance (HOMA-IR) was calculated as follows: fasting insulin (mU/l) × fasting glucose (mmol/l)/22.5. Adiponectin, leptin and hsCRP were measured by ELISA using commercially available kits (R&D Systems, Minneapolis, Minnesota, USA and Invitrogen).
Assessment of disease activity and damage
Disease-related information included: age at diagnosis of psoriasis and PsA, current use of non-steroidal anti-inflammatory drugs (NSAID), disease-modifying antirheumatic drugs (DMARD) and anti-tumour necrosis factor alpha (TNFα) agents. Tender, swollen and damaged joint counts were assessed on physical examination. Current psoriasis activity was determined by the psoriasis area and severity index (PASI).
Continuous data were described by means and SD and categorical variables as frequencies and percentages. Comparisons between the two categories were made using two-tailed t tests for continuous variables and by the χ2 test for categorical variables. As the distribution of leptin, adiponectin and HOMA-IR was skewed to the left, these variables were log transformed. A logistic regression model was fit with MetS as the binary outcome and with disease status as a covariate, adjusting for age, sex, ethnicity, smoking, alcohol consumption, use of lipid-lowering agents and BMI. The effect of each factor was measured through OR that were estimated along with their respective 95% CI and reported with p values. Non-contributory variables were dropped from the models through backwards elimination and the final models were reported including only significant variables. Pearson and Spearman correlation coefficients were used when appropriate to study the relationship between clinical variables and the biomarkers. A multivariate linear regression model was used to estimate the correlation between adiponectin and disease status (PsA vs PsC) and the following disease-related variables: active and damaged joint count, PASI, anti-TNFα and DMARD therapy. Adjustment was made for the following variables: age, sex, ethnicity, smoking, alcohol consumption, use of lipid- lowering agents and BMI. The effect of each factor was expressed as a regression coefficient (β) along with their respective 95% CI and reported with p values. The effect of a covariate was considered statistically significant if the p value from the two-sided Wald test was less than 0.05. Model fit for linear regression models was assessed with a coefficient of variation (R2). The statistical computation was performed using SAS 9.2.
A total of 203 PsA and 155 PsC patients participated in the study. There were no significant differences between PsA and PsC patients with respect to their age, gender or ethnicity; however, PsA patients had a longer duration of psoriasis (table 1). PsC patients exhibited more active psoriasis, reflected by their higher PASI scores and were less likely to use NSAID, DMARD or anti-TNFα agents than PsA patients. Furthermore, the frequency of obesity (BMI > 30), hypertension and impaired glucose tolerance was higher in PsA patients. No difference was observed in the levels of the various lipid particles between the two groups.
MetS in PsA compared to PsC patients
A higher prevalence of MetS in PsA patients compared to PsC was observed but was not statistically significant (74 (36.5%) vs 42 (27.1%), p=0.056). In the multivariate regression analysis the following variables were associated with MetS: group (PsA vs PsC, OR 1.7, 95% CI 1 to 2.7), PASI (OR 1.6, 95% CI 1.1 to 1.9) and age (OR 1.6, 95% CI 1.3 to 2). The results are presented in table 2.
Adipokines and insulin resistance in PsA compared to PsC
Serum levels of adiponectin were significantly higher in patients with PsA compared to those with PsC (2.01±0.58 vs 1.83±0.60 log of μg/ml, p=0.005, figure 1A). The level of insulin resistance, as measured by HOMA-IR, was higher in PsA patients compared to those with PsC (0.97±0.63 vs 0.68±0.81, p=0.0003, figure 1B). Due to the significantly different cut-off levels of leptin among healthy men and women, comparison of leptin levels between the two groups was stratified by sex. Leptin levels were higher in women with PsA compared to those with PsC (3.09±0.82 vs 2.81±0.82 log of ng/ml, p=0.04, figure 1C). No difference was found in leptin levels between the two groups among men (figure 1D).
The correlation between disease-related factors and biomarkers of MetS was investigated using Pearson and Spearman correlation coefficients. The results are presented in table 3.
As expected from previous studies,19 adiponectin positively correlated with HDL and negatively correlated with triglycerides, HOMA-IR and BMI, while leptin positively correlated with HOMA-IR and BMI. HOMA-IR also positively correlated with BMI and triglycerides and negatively correlated with HDL.
With respect to disease-related variables, adiponectin showed a positive weak correlation with active and damaged joint counts (r=0.12, p=0.02; r=0.13, p=0.02, respectively) and a negative weak correlation with PASI (r=−0.11, p=0.03). No correlation was observed between hsCRP and adiponectin. Leptin showed a borderline weak correlation with the active joint count (r=0.1, p=0.05) and moderately correlated with hsCRP (r=0.36, p<0.001). HOMA-IR correlated with hsCRP (r=0.31, p<0.001).
The association between adiponectin and disease-related variables by multivariate regression analysis is presented in table 4. The following variables were associated with adiponectin in the multivariate regression analysis: PsA group (β=0.09, p=0.005), age (β=0.09, p<0.001), female gender (β=0.45, p<0.001), race (β=0.17, p=0.02), anti-TNFα treatment (β=−0.14, p=0.03), active joint count (β=0.14, p=0.001), BMI (β=−0.03, p<0.001) and lipid-lowering agents use (β=−0.26, p<0.001). A trend for an association was found between the damaged joint count and adiponectin (β=0.05, p=0.06).
The association between HOMA-IR and disease-related variables was further analysed using a multivariate linear regression model (R2=0.4). Anti-TNFα treatment (β=0.23, p=0.003), male gender (β=0.14, p=0.02), BMI (β=0.06, p<0.001) and the use of lipid-lowering agents (β=0.37, p<0.001) were associated with HOMA-IR in the multivariate regression analysis, while the association with the PsA group was of borderline significance (β=−0.07, p=0.05).
In this study we found that the occurrence of MetS as well as adipose tissue biomarkers and insulin resistance was associated with PsA compared to PsC. These biomarkers also correlated with the burden of disease, particularly with joint inflammation and damage.
Obesity, which has become a worldwide epidemic, is a major component of MetS and directly contributes to the development of other components including insulin resistance, hypertension and dyslipidaemia. There is a strong link between obesity and psoriatic disease. Obesity predicts the development of psoriasis and was also suggested to be associated with PsA.10 ,11 ,29 ,30 Obese patients tend to have more severe psoriasis and are less likely to respond to therapy.31–34 In the present study we found that patients with PsA are more likely to be obese and have MetS. The occurrence of MetS was also associated with more severe psoriasis. Furthermore, leptin levels, which reflect body fat mass,20 were elevated in women with PsA, and the extent of insulin resistance was higher among PsA patients compared to those with PsC, reflecting the increased extent of metabolic derangement among the former group. These findings are in accordance with a study by Bhole et al35 that reported a higher prevalence of obesity among PsA patients compared to PsC patients. Another study from our group has found that the prevalence of hypertension, obesity, hyperlipidaemia and type 2 diabetes mellitus were 1.5–2.6-fold higher in PsA than in PsC.36 The ‘chicken and egg’ dilemma arises as it is still unclear which comes first, PsA or obesity. In other words, whether obesity directly increases the risk of developing of PsA among patients with psoriasis or whether arthritis leads to lower levels of physical activity causing weight gain. Recent results from a prospective cohort study among women support the paradigm that obesity is a risk factor for the development of PsA among patients with psoriasis.37 Additional prospective studies are needed to investigate this link.
The relation between adiponectin and inflammation is intriguing as this adipokine in general has anti-inflammatory properties that improve metabolic functions.38 Adiponectin treatment improves macrovascular endothelial inflammation in adiponectin knockout mice39 and inhibits TNFα-induced responses through the nuclear factor κ B pathway.40 In contrast to its beneficial effect on metabolic functions, elevated adiponectin levels were found in several inflammatory conditions including rheumatoid arthritis (RA) and systemic lupus erythematosus.41 ,42 Recent studies have suggested that adiponectin has pro-inflammatory properties in the joints, leading to increased joint damage in RA patients.43 ,44 Obese individuals are expected to have low levels of adiponectin. However, in the present study, adiponectin levels were higher in PsA patients compared to those with PsC despite the higher prevalence of obesity in the former group. Similar to findings in RA patients, adiponectin correlated with a higher burden of joint inflammation and damage.
It is still unclear what leads to the induction of adiponectin production in inflammatory conditions, but it is clear that its effect may exceed the inhibitory effect of TNFα, and that the negative feedback loop between these two cytokines is dysregulated. It might be expected that TNFα blockade would increase adiponectin levels; however, prospective studies in RA and PsA patients did not find any effect of this treatment on serum adiponectin levels.45 In the present study anti-TNFα treatment negatively correlated with adiponectin in the multivariate regression analysis. However, as this study is cross-sectional we cannot determine whether this treatment affects adiponctin levels in patients with PsA or PsC. Furthermore, the synthesis of adiponectin may be compartment-dependent and its local effect and regulation may differ completely between the joint, the skin and the blood vessel. The notion of a compartment-specific adiponectin effect is supported by the negative correlation between adiponectin levels and skin inflammation such as in psoriasis or scleroderma23 ,46 in contrast to its positive correlation with joint inflammation as in the present study and in RA.
Some of the patients included in this study had carotid artery ultrasound and plaque area measured.18 An analysis of those who participated in both studies revealed a correlation between plaque area and HOMA-IR (data not shown).
Our study had several limitations. Its cross-sectional nature precludes a firm conclusion regarding any causal relation between adipokines and joint damage. Second, the study lacked a healthy control group, although previous studies found that obesity and related metabolic abnormalities are increased in patients with both PsA and PsC compared to healthy individuals.7 ,8 ,17 Finally, although our study population was well phenotyped and we were able to adjust for differences in important variables between the groups through the regression analysis, we are aware that there may be other confounders that were not considered and may explain the differences in MetS and adipokine levels between the two groups. Despite these limitations, the strengths of our study include its relatively large sample size and the accurate phenotyping of the study population. Furthermore, to our knowledge, this is the first study to compare the levels of several metabolic biomarkers between PsA and PsC patients.
In summary, we have found that the presence of MetS and related adipokines correlated with an increased burden of skin and joint inflammation. These findings highlight the important association between obesity and arthritis among psoriasis patients. Additional prospective studies are needed to investigate this relationship further.
The authors would like to acknowledge Sutha Shanmugarajah and Daniel Pereira for their coordination of this study.
Handling editor Tore K Kvien
Contributors All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. All authors were likewise involved in the study conception and design, acquisition of data as well as analysis and interpretation of data. Dr Gladman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Funding The Psoriatic Arthritis Program is funded in part by The Arthritis Society, Canadian Institutes of Health Research, and the Krembil Foundation.
Competing interests None.
Ethics approval The study was approved by the University Health Network Research Ethics Board.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement All authors were involved in the study conception and design, acquisition of data, analysis and interpretation of data.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.