Rheumatoid arthritis (RA) is one of the most prevalent inflammatory diseases in the general population. While the most important pathogenic lesion is inflammatory synovitis, extra-articular features are equally important, considering the fact that inflammation has a systemic magnitude [1]. Mortality is higher among RA patients compared to the general population and the majority of premature deaths are attributable to cardiovascular (CV) disease [2]. CV morbidity is also enhanced, with an increased prevalence of all stages of atherogenesis—from endothelial dysfunction to increased thickness and plaque in carotid arteries, and ultimately to fatal and nonfatal myocardial infarction and stroke [3, 4, 5, 6, 7, 8]. Classic risk factors for CV diseases in the general population—such as hypertension, dyslipidemia, diabetes, and smoking—also appear to be important in RA, but only account in part for the excess CV risk [9]. Chronic inflammation plays an important role in the pathogenesis of both atherosclerosis and RA and accentuates established CV risk factors [2].

Carotid intima-media thickness (CIMT) of the common carotid artery is a useful noninvasive anatomical structural measure of subclinical CV disease and the most popular technique used to study early structural changes in the arterial wall [10]. Increased CIMT has been reported in RA patients compared to controls [11, 12]. Large artery stiffness is known to increase in patients with atherosclerosis and is both a surrogate marker and an independent risk factor for atherosclerosis [13]. In chronic inflammatory diseases, elevated proinflammatory cytokines may contribute to the increased risk for CV disease by promoting vascular changes and increased arterial stiffness. Arterial stiffness can be assessed noninvasively by measurement of pulse wave velocity (PWV), a simple and reproducible method [14].

Osteoprotegerin (OPG) is a member of the tumor necrosis factor (TNF) family and is known to inhibit osteoclastogenesis [15]. Clinical studies have shown an association between OPG and vascular atherosclerotic diseases. OPG was also independently associated with coronary artery calcification in RA patients with long-standing disease [16]. To our knowledge, there is no available study reporting on the relationship between OPG and PWV in RA patients without CV disease or risk factors. Therefore, the aim of this study was to evaluate OPG levels in RA patients compared to healthy controls and to assess their correlation with arterial stiffness measured by carotid femoral PWV, CIMT, and other laboratory and disease activity indices.

Materials and methods

Patients

For this cross-sectional study, 68 consecutive patients (age 48.4 ± 8.1 years; 56 women and 12 men) who fulfilled the 2010 American College of Rheumatology (ACR)/European League against Rheumatism (EULAR) classification criteria [17] for RA were recruited. All patients were taking conventional synthetic disease-modifying antirheumatic drugs (DMARD) including methotrexate, sulfasalazine, leflunomide, and hydroxychloroquine either as monotherapy or in combination, and 11 patients also were treated with biologics. Patients with known prior CV disease (coronary angina, myocardial infarction, cerebral ischemic stroke, or peripheral vascular disease), hypertension, diabetes mellitus, or hyperlipidemia were excluded from the study. As normal controls, 48 age- and gender-matched healthy volunteers (age 46.6 ± 6.2 years; 37 women and 11 men), who had no rheumatic or CV disease, were included. Written informed consent was obtained from all participants prior to the study. The study protocol was approved by the local Ethics Committee of our institution and performed in accordance with the principles stated in the Declaration of Helsinki.

Clinical assessment

Demographics of all participants were recorded. Disease activity was assessed by the 28-joint disease activity score (DAS28) in RA patients [18]. Pain intensity was evaluated using a 100-mm horizontal visual analog scale (VAS). Erythrocyte sedimentation rate (ESR) was recorded. Based on DAS28, the patients were subdivided into three subgroups: mild (DAS28 ≤ 3.2), moderate (3.2 < DAS28 ≤  5.1), and severe (DAS28 >  5.1).

A single experienced cardiologist, who was blind to patient clinical data, performed vascular assessments for both the patients and controls. Reported blood pressures were all obtained in the morning after an overnight fast and at the same visit as patients’ vascular assessment. Resting blood pressures were measured in the seated position using an automated sphygmomanometer.

Laboratory analysis

Overnight fasting blood samples were obtained from all subjects for determination of glucose, total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) levels. Glucose and lipid profile were determined by standard methods.

Anti-cyclic citrullinated peptide (anti-CCP) antibody was measured by enzyme-linked immunosorbent assay (ELISA), using a commercial kit according to the manufacturer’s instructions. A result was considered positive for anti-CCP antibodies if the titer was above 15 IU/ml. Serum levels of rheumatoid factor (RF)-IgM were measured by the nephelometric method, and a result was considered positive for RF when its concentration was above 20 IU/ml. Serum high-sensitivity C-reactive protein (hsCRP) level was measured with an immunoturbidimetric method using an Abbott auto-analyzer (Architect C1600; Abbott, USA). The normal hsCRP range was defined as ≤ 0.5 mg/dl.

Determination of serum osteoprotegerin

Blood samples from patients and controls were collected after an overnight fast. Serum was obtained after centrifugation and stored at − 70 ℃ until analysis. The concentration of OPG was measured using a commercially available human OPG ELISA kit (eBioscience, Vienna, Austria) according to the instructions provided by the manufacturer. Absorbance was measured at a wavelength of 450 nm using an ELISA reader. The levels of OPG were recorded in pg/ml. The sensitivity of the OPG assay was 2.5 pg/ml.

Assessment of carotid intima-media thickness

Ultrasound imaging examination of the carotid arteries was performed using a high-resolution ultrasonography scanner (VingMed Vivid 3; GE Medical Systems, Horten, Norway) with a 7-MHz linear array transducer. Measurements were performed on the right and left carotid arteries with the subject in the supine position. The region 1 cm proximal to the carotid bifurcation was identified, and the CIMT of the far wall was evaluated as the distance between the lumen–intima interface and the media–adventitia interface. The CIMT measurement was obtained from four contiguous sites at 1-mm intervals on each carotid artery, and the mean value of all eight measurements (in millimeters) was calculated and used for analysis.

Assessment of pulse wave velocity

Carotid femoral PWV was measured as an index of arterial stiffness. Assessments were performed after the subjects had rested for 15 min in the supine position in a quiet room. The carotid femoral PWV was calculated from the measurements of pulse transmission time and the distance between the two recording sites by a validated noninvasive device (SphygmoCor, AtCor Medical, Sydney, Australia). The distances from the sternal notch to the right carotid artery and from the sternal notch to the right femoral artery were measured. The distance traveled by the pulse wave over the surface of the body, as determined with a tape measure, was divided by the transit time; the carotid femoral PWV was then calculated automatically in meters/second (m/s).

Statistical analysis

All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS), version 13.0, for Windows (SPSS Inc., Chicago, IL, USA). Continuous variables are presented as the mean ± standard deviation or median (95 % confidence interval, CI). The normality of the distribution for all variables was assessed by the Kolmogorov–Smirnov test. Intergroup comparisons were made using Student’s t-test for normally distributed variables and the Mann–Whitney U test for nonparametric variables. Analyses of evaluated parameters in subgroups with RA were performed using a one-way analysis of variance and the Kruskal–Wallis test. To assess the correlations between variables, Spearman’s rank or Pearson’s correlation analysis were used according to data distribution. Values of p < 0.05 were considered statistically significant.

Results

The demographic and clinical characteristics of patients and controls are shown in Tab. 1. Patients and controls did not significantly differ in terms of age or gender (48.4 ± 8.1 vs. 46.6 ± 6.2 years, p = 0.183; female/male 56/12 vs. 37/11, p = 0.642, respectively), but the body mass index (BMI) was significantly higher in RA patients than controls (p = 0.01). The mean disease duration in RA patients was 5.6 ± 3.9 years.

Tab. 1 Demographic and clinical characteristics of patients with rheumatoid arthritis (RA) and control subjects (mean ± standard deviation)
Full size table

The comparison of laboratory parameters, PWV, and CIMT between patients and controls is shown in Tab. 2. TG and LDL-C levels did not differ between groups (p > 0.05), whereas TC, HDL-C, blood glucose, and systolic and diastolic blood pressures were significantly higher in RA patients compared to controls (all p < 0.05). Patients with RA demonstrated significantly higher levels of hsCRP and ESR compared to controls (1.2 ± 1.4 vs. 0.2 ± 0.1 mg/dl, p < 0.001; 28.9 ± 17.9 vs. 8.7 ± 4.6 mm/h, p < 0.001, respectively). The mean serum OPG level was significantly higher in RA patients than in controls (116.9 ± 92.7 vs. 57.1 ± 12.2 pg/ml, p < 0.001). The mean PWV and CIMT values were also significantly increased in RA patients compared to controls (8.2 ± 1.9 vs. 6.2 ± 1.2 m/s, p < 0.001; 0.80 ± 0.12 vs. 0.57 ± 0.07 mm, p < 0.001, respectively).

Tab. 2 Cardiovascular parameters and laboratory findings of patients with rheumatoid arthritis (RA) and controls (mean ± standard deviation)
Full size table

Although the patients with severe disease activity had the highest level of OPG, there was no statistically significant difference in mean serum OPG levels among subgroups with low, mild, and severe disease activity (98.9 ± 81.1, 118.6 ± 85.7, and 148.2 ± 126.6 pg/ml, respectively; p = 0.296); nor did the mean CIMT and PWV values demonstrate significant differences among RA subgroups (CIMT: 0.80 ± 0.14, 0.81 ± 0.11, and 0.78 ± 0.13 mm; p = 0.814; PWV: 8.6 ± 2.3, 7.9 ± 1.5, and 7.9 ± 2 m/s; p = 0.390, respectively).

In RA patients, serum OPG levels were significantly correlated with PWV and CIMT (r = 0.273, p = 0.02; r = 0.335, p = 0.005, respectively; Fig. 1 a, b). Serum OPG levels also demonstrated significant correlations with RF and anti-CCP antibody, but not with DAS28, hsCRP, or ESR (p = 0.007, p = 0.007, p = 0.202, p = 0.179, p = 0.237, respectively; Tab. 3). In addition, CIMT was significantly correlated with hsCRP, PWV, and patient age, but not with ESR or DAS28 (r = 0.263, p = 0.031; r = 0.460, p < 0.001; r = 0.405, p = 0.001; r = 0.192, p = 0.118; r = 0.007, p = 0.954, respectively). While PWV showed no significant correlation with hsCRP, ESR, DAS28, or disease duration (all p > 0.05), it was significantly correlated with patient age (r = 0.336, p = 0.005). Serum lipid profiles were not correlated with any other studied parameters (all p > 0.05; data not shown).

Fig. 1
figure 1

The correlations of osteoprotegerin (OPG) with (a) carotid femoral pulse wave velocity (PWV) and (b) carotid intima-media thickness (CIMT) in patients with rheumatoid arthritis

Full size image
Tab. 3 The correlations of osteoprotegerin with laboratory findings and disease activity indices
Full size table

Discussion

There is increasing evidence from controlled clinical trials that patients with RA present more extensive atherosclerosis and coronary calcification than those without [19]. A prospective Dutch study has also demonstrated that the magnitude of CV risk in RA may equal the CV risk in patients with type 2 diabetes [20]. In this study, the CV risk in RA patients with no traditional risk factors was investigated with regard to the association of serum osteoprotegerin level with PWV and CIMT.

CIMT is an important marker for early, preclinical atherosclerosis and a predictor of future CV events [21]. RA is considered an independent risk factor for intima-media thickening of the common carotid and femoral arteries [19, 22]. Additionally, in previously reported studies in RA patients, CIMT demonstrated significant correlations with traditional CV risk factors [23, 24], disease severity [11], and CRP [25] and ESR values [26]. CIMT ≥ 0.60 mm is a common marker of atherosclerosis and the presence of carotid plaques is the best indication of advanced atherosclerosis [10]. In accordance with previous studies, the mean CIMT in this study was higher in RA patients compared to controls (0.80 vs. 0.57 mm). In addition, CIMT demonstrated statistically significant correlations with hsCRP and age in patients with RA, but no correlation with ESR, DAS28, or lipid profile.

Arterial stiffness is an important arterial phenotype and an independent predictor of CV disease [27]. In this study, arterial stiffness was evaluated by carotid femoral PWV, the established gold-standard measure of aortic stiffness, and a significantly greater PWV was observed in patients with RA as compared to controls. Similar results also have been reported in other studies [28, 29, 30]. In contrast, a study by Fan et al. [13] found no significant difference in terms of brachial ankle PWV among groups including patients with RA, inflammatory bowel disease, and healthy subjects. Increased arterial stiffness has also been linked with disease duration [31, 32], disease activity scores [33], patient age, and CRP values [30] in RA patients. However, the latter studies included patients with traditional CV risk factors. In the present study, PWV was only significantly correlated with patient age, and showed no significant correlation with hsCRP, ESR, or DAS28. Similar to our results, PWV was not significantly correlated with disease duration, biochemical parameters of inflammation (ESR, CRP), or DAS28 in a study by Pieringer et al. [29]. As in our study design, RA patients with traditional CV risk factors were excluded from the Pieringer study to eliminate possible confounders. We additionally found significant correlations between PWV and CIMT in patients with RA.

Osteoprotegerin is currently defined as an important biomarker in CV disease. By acting as a soluble decoy receptor competing for receptor activator of nuclear factor κB ligand (RANKL), OPG prevents receptor activator of nuclear factor κB (RANK)–RANKL interactions, and thus osteoclast differentiation and bone resorption. Mechanisms supporting a pro-atherosclerotic role for OPG include its in vitro ability to enhance expression of endothelial cell adhesion molecules and subsequent infiltration of leukocytes and monocytic cells. Furthermore, OPG might contribute to endothelial dysfunction by blocking RANKL signaling, which is able to activate protective intracellular endothelial pathways such as the nitric oxide synthase pathway. OPG also influences the production of other agents important in plaque stability, such as matrix metalloproteinases [34]. A recent systematic update including 14 studies with clearly defined cohorts qualified for review concluded that OPG concentration is associated with the presence and severity of stable coronary artery disease, acute coronary syndrome, and cerebrovascular disease [35].

RA patients with no CV risk factors had higher OPG levels and increased CIMT and PWV values

In the present study, serum OPG concentration was evaluated along with CIMT and carotid femoral PWV in RA patients. Consistent with previous studies, serum OPG levels were significantly higher in RA patients compared to controls. Furthermore, statistically significant, albeit not strong, correlations of serum OPG with PWV, CIMT, and age were also demonstrated. In fact, RA patients with CV disease or risk factors were excluded from this study. In addition, our patients had short disease duration (mean 5.6 years), moderate disease activity (3.73), and relatively low CRP values (1.2 mg/dl). It is known that preclinical atherosclerosis is associated with CV risk factors, disease duration, disease severity, and level of CRP [1]. These factors may explain the weak correlation between serum OPG levels and cardiac parameters. Nevertheless, to our knowledge, this study is the first to report the correlation of OPG with carotid femoral PWV in RA patients. We also observed significant correlations of OPG with RF and anti-CCP antibodies. However, no correlations were identified between serum OPG and disease activity indices such as ESR, hsCRP, and DAS28. All of the patients were on drug treatment, including conventional synthetic or biologic DMARDs. These treatment regimens could have influenced the association of OPG levels with inflammatory markers. A small number of previous studies have evaluated OPG as an indicator of CV disease in RA patients. In a study by Asanuma et al. [16], similar to our results, higher OPG concentrations were observed in patients with early and long-standing RA as compared to control subjects. This study also detected significant correlations of OPG with ESR, hsCRP, DAS28, and disease duration. However, contrary to our study design, the Asanuma study included patients and controls with CV risk factors such as hypertension, smoking, diabetes, and family history of coronary artery disease. Furthermore, the patients had older age, longer disease duration, and higher CRP values than those in our study. A later study also found that serum OPG level was significantly higher in RA patients with carotid plaques than in those without plaques, and independently associated with carotid plaques [36]. In a recent uncontrolled study by Dessein et al. [37] that included 30 patients with severe RA, OPG concentrations were significantly associated with both endothelial activation and carotid atherosclerosis, and this finding was consistent before and after infliximab infusion. This study also included patients with traditional CV risk factors. In addition, similar to our results, there were no associations of OPG with ESR, DAS28, or CRP in their study.

Limitations

This study has some limitations. First, the study design was cross-sectional, and a longitudinal study would have been more conclusive. Second, the number of patients with low and severe disease activity was relatively small, and the mean disease duration was short. Therefore, the results may not be generalizable to all RA patients. Third, it was not possible to evaluate patients before and after treatment with DMARDs and biologics, which might have reflected the significance of effective anti-inflammatory therapy in risk management strategies for CV disease in patients with RA.

Conclusion

This study demonstrated that RA patients with no CV risk factors had higher levels of OPG and increased CIMT and PWV values compared to controls, and detected significant correlations of OPG with CIMT and PWV in this population. However, further follow-up studies involving large samples are required to clarify the diagnostic, predictive, and prognostic value of OPG for the risk management of CV disease in RA patients.