- Top of page
- Conflict of Interest
Rheumatoid arthritis (RA) is associated with enhanced atherosclerosis and impaired endothelial function early after the onset of the disease and cardiovascular (CV) disease represents one of the leading causes of morbidity and mortality. It is well known that disease modifying antirheumatic drugs (DMARDs) are able to improve the course of the disease and the quality of life of these patients, but little is known about the effects of DMARDs on CV risk and endothelial dysfunction. Our goal was to examine the effects of long-term therapy with DMARDs on endothelial function and disease activity in early RA (ERA). Twenty-five ERA patients (mean age 52 ± 14.6 years, disease duration 6.24 ± 4.10 months) without evidence of CV involvement were evaluated for disease activity score (DAS-28), 2D-echo derived coronary flow reserve (CFR), common carotid intima-media thickness (IMT) and plasma asymmetric dimethylarginine (ADMA) levels at baseline and after 18 months of treatment with DMARDs (10 patients with methotrexate and 10 with adalimumab). DMARDs significantly reduced DAS-28 (6.0 ± 0.8 vs. 2.0 ± 0.7; P < 0.0001) and improved CFR (2.4 ± 0.2 vs. 2.7 ± 0.5; P < 0.01). Common carotid IMT and plasma ADMA levels did not show significant changes. The present study shows that DMARDs, beyond the well known antiphlogistic effects, are able to improve coronary microcirculation without a direct effect on IMT and ADMA, clinical markers of atherosclerosis. Treatment strategies in ERA patients with high inflammatory activity must be monitored to identify beneficial effects on preclinical markers of vascular function.
- Top of page
- Conflict of Interest
Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease associated to high grade inflammatory status with a multiorgan involvement. It affects about 1% of the general adult population and is characterized by a reduced life expectancy  mainly due to cardiovascular (CV) disease. The excess of CV morbidity and mortality is detectable early after the onset of the disease and it could not be fully explained by traditional risk factors [2–4].
Chronic inflammation [5,6], immune dysregulation  and neoangiogenesis  have been widely recognized as important players both in the pathogenesis of RA and in the development of atherosclerosis. Moreover, standard therapy (i.e., corticosteroids and cyclo-oxygenase inhibitors) may accelerate atherosclerosis [9,10] while, recent evidences showed a positive effect on CV outcomes of methotrexate and anti-tumor necrosis factor alpha (TNFα) [11,12].
Wislowska et al.  found that RA patients with unknown CV involvement had a high prevalence of silent coronary artery disease (CAD), and a number of other authors reported that CAD occurs rapidly in RA patients and at a younger age than in the general population [14,15]. The importance of recognising and treating patients in the early stages of RA is therefore due to the risk that active disease may lead to progressive joint and cardiovascular damage [15,16].
In particular, we could define early rheumatoid arthritis (ERA) as disease duration ≤12 months in absence of any antirheumatic therapy (biological or nonbiological disease-modifying antirheumatic drugs (DMARDs) or steroids) .
Transthoracic Doppler-derived coronary flow reserve (CFR) has been used to identify patients with known or suspected CAD , and its prognostic value has also been confirmed in various CV settings . Furthermore, the measurement of carotid artery intima-media thickness (IMT) is clinically useful in identifying early atherosclerosis and closely correlates with CAD .
Recently, plasma asymmetric dimethylarginine (ADMA) levels have been associated with CV risk [21,22], and increased plasma ADMA has been observed in patients with diseases associated with enhanced atherosclerosis, such as hypercholesterolemia , hypertriglyceridemia , peripheral arterial disease , hypertension , type 2 diabetes mellitus , acute coronary syndromes [28,29], and end-stage renal failure . It has also been hypothesised that ADMA is causally involved in the pathophysiology of atherosclerosis and its complications.
ADMA is released into plasma and inhibits nitric oxide (NO) production by NO synthase, thus causing endothelial dysfunction, which is closely associated with the development of atherosclerosis. It has been recognized as a major endogenous inhibitor of all three isoforms of NO synthase , and it is known that high plasma ADMA levels can impair NO generation and promote the progression of atherosclerosis . Recent studies postulated that ADMA is not only a marker but also a potent mediator of endothelial dysfunction and atherosclerosis  which have been recently demonstrated in young patients with long-term RA and low disease activity . Moreover, ADMA pathways have been recognized to play critical roles during infection and inflammation .
Chronic inflammation, accelerated atherosclerosis and functional abnormalities of the endothelium suggest a subclinical CV involvement beginning rapidly soon after the onset of the disease and progressing with disease duration. In a previous study, we detected an impaired CFR associated with increased plasma ADMA levels in 25 ERA patients without evidence of CV involvement .
It is well known that early DMARDs improve the clinical course of the disease and retard the progression of structural damage . However, little is known about the effects of DMARDs on CV involvement and endothelial dysfunction in ERA patients.
The aim of the present study was to evaluate the effects of DMARDs on endothelial function and coronary microcirculation in ERA patients.
- Top of page
- Conflict of Interest
Descriptive statistics of our study population as well as haemodynamic data at baseline and after follow-up are summarized in Table 1.
Table 1. Characteristics of the study population
| ||ERA patients baseline (n = 25)||Posttreatment (n = 20)||P value|
|Sex, no. (%) female||21 (84%)||18 (90%)||–|
|Age (years)||52.0 ± 14.6||55.1 ± 14.9||–|
|BMI (kg/m2)||21.9 ± 2.7||23.1 ± 3.7||NS|
|Systolic blood pressure (mmHg)||136.4 ± 22.1||127.1 ± 16.02||NS|
|Diastolic blood pressure (mmHg)||80.8 ± 9.4||74.3 ± 6.75||0.025|
|Heart rate (beats/min)||75.9 ± 12.8||75.36 ± 10.7||NS|
|Disease duration (months)||6.24 ± 4.10||23.85 ± 3.76||–|
|Rheumatoid Factor (mg/dL)||97.11 ± 143.72||79.36 ± 75.43||NS|
|Anti-CCP (UI)||81.40 ± 123.84||64.75 ± 75.45||NS|
|DAS-28||5.86 ± 0.64||2.01 ± 0.74||0.0001|
|Total cholesterol (mg/dL)||180.6 ± 20.8||175.2 ± 18.3||NS|
|HDL cholesterol (mg/dL)||59.5 ± 10.5||56.6 ± 12.1||NS|
|LDL cholesterol (mg/dL)||101.0 ± 14.5||98.1 ± 11.6||NS|
|Triglycerides (mg/dL)||120.3 ± 15.4||102.7 ± 74.1||NS|
|Creatinine (mg/dL)||0.85 ± 0.15||0.88 ± 0.23||NS|
|GFR (mL/min/1.73 sqm)||90.5 ± 8.3||92.1 ± 9.4||NS|
|Homocysteinemia (μmoles/L)||9.6 ± 2.5||8.9 ± 1.8||NS|
|Glycaemia (mg/dL)||85.0 ± 15.0||92.3 ± 12.1||–|
|CRP (mg/dL)||14.04 ± 15.27||2.24 ± 3.74||0.0001|
|ESR (mm/h)||35.2 ± 22.5||20.85 ± 9.54||NS|
|EF (%)||62.4 ± 8.4||63.3 ± 5.3||NS|
|IMT (mm)||0.68 ± 0.10||0.66 ± 0.15||NS|
|CFR||2.4 ± 0.2||2.7 ± 0.5||0.01|
|ADMA (μmoles/L)||0.65 ± 0.07||0.7 ± 0.17||NS|
| Methotrexate—no. (%)||–||10 (50%)|| |
| Anti TNF-α agents—no. (%)||–||10 (50%)|| |
Only 20 patients were available for analysis because of 5 patients dropped out: 3 patients developed side-effects early after the beginning of treatment (1 patient showed cutaneous rash after adalimumab administration and 2 developed gastrointestinal intolerance with methotrexate) and 2 patients have been lost at the follow-up.
Noteworthy, since baseline evaluation, ERA patients had impaired CFR (3.5 ± 0.8 vs. 2.4 ± 0.2, P < 0.001), higher plasma ADMA levels (0.57 ± 0.07 vs. 0.65 ± 0.07, P < 0.01) and common carotid IMT (0.56 ± 0.11 vs. 0.68 ± 0.10, P < 0.01) compared with 25 healthy controls matched for factors affecting endothelial function, such as cholesterol, blood pressure, smoking habits and renal function (data not shown).
Both at baseline and after treatment all patients had normal ECG track, heart rate and blood pressure. Echocardiographic and Doppler parameters, lipid profile and renal function were in normal ranges.
At baseline all patients had high disease activity. Long term DMARDs treatment induced a reduction of inflammatory status: CRP lowered from 14.04 ± 15.27 to 2.24 ± 3.74 mg/dL (P= 0.0001) (Table 1 and Figure 3). DAS-28 significantly improved from 5.86 ± 0.64 to 2.01 ± 0.74 (P < 0.0001) (Figures 1 and 3) during treatment. No significant differences were observed between the 2 groups treated with methotrexate and adalimumab concerning the effect on disease activity and inflammatory markers (Table 2).
Table 2. Comparison between methotrexate and adalimumab
| ||Methotrexate (n = 10)||Adalimumab (n = 10)||P value|
|CFR||2.68 ± 0.5||2.71 ± 0.4||NS|
|IMT (mm)||0.66 ± 0.12||0.67 ± 0.09||NS|
|ADMA (μmoles/L)||0.68 ± 0.17||0.7 ± 0.17||NS|
|ESR (mm/h)||20.9 ± 6.57||20.8 ± 6.88||NS|
|CRP (mg/dL)||2.31 ± 2.3||2.43 ± 2.1||NS|
|DAS-28||2.02 ± 0.37||1.99 ± 0.7||NS|
CFR improved during DMARDs treatment from 2.4 ± 0.2 to 2.7 ± 0.5 (P < 0.01) while common carotid IMT (Figure 3) and plasma ADMA levels did not show significant changes after therapy (Figure 3). When separately considering the two arms of treatment, we did not observe any significant difference in CFR, common carotid IMT and plasma ADMA levels between patients receiving methotrexate and those on adalimumab (Table 2).
None of the correlation evaluated between CFR, IMT and ADMA and markers of inflammation or disease activity resulted statistically significant (Figures 4, 5, 6).
Figure 5. Dispersion plot of variables, the straight line portray the interpolation while the dotted line represents 95% confidence limits of the regression predicted value; Spearman correlation coefficients and 95% confidence intervals are reported on the table below. CFR, coronary flow reserve; IMT, intima-media thickness; ADMA, asymmetric dimethylarginine; DAS-28, disease activity score; bas, baseline; post, posttreatment.
Download figure to PowerPoint
Figure 6. Dispersion plot of variables, the straight line portray the interpolation while the dotted line represents 95% confidence limits of the regression predicted value; Spearman correlation coefficients and 95% confidence intervals are reported on the table below. ESR, erythrocyte sedimentation rate; DAS-28, disease activity score; bas, baseline; post, posttreatment.
Download figure to PowerPoint
An example of CFR Doppler signal is presented in Figure 2.
Figure 2. Examples of CFR values obtained at baseline (left panels: CFR = 2.58) and at the end of the follow-up (right panels: CFR = 2.97) in a ERA patient. * indicates the diastolic velocity in Doppler signal at baseline (upper panels) and during hyperaemia (lower panels).
Download figure to PowerPoint
The interobserver variability for CFR assessment resulted < 7%.
- Top of page
- Conflict of Interest
Our results demonstrated the efficacy of long-term DMARDs treatment, both methotrexate and adalimumab, not only in reducing disease activity, but also in improving coronary microcirculation function. It is well known that atherogenesis is a long-term process that develops throughout lifespan in all individuals, but it is enhanced in disease conditions characterized by systemic inflammation. Recently, del Rincon et al.  demonstrated that established RA patients with longer disease duration had more atherosclerosis as compared to healthy controls. In addition to disease duration, also high grade systemic inflammation  and exposure to certain drugs  have been considered responsible for accelerated atherosclerosis in RA. Accordingly to these data, since baseline evaluation, we observed a high incidence of endothelial dysfunction associated with impaired coronary microcirculation in our ERA patients with high disease activity early after the onset of the disease. However, common carotid IMT was still in normal ranges, even if increased compared with healthy controls. So, we hypothesize that CFR evaluation is able to detected an early stage of atherosclerotic process, when peripheral anatomic vascular changes were not yet occurred.
Infact, carotid IMT is a structural marker, which reflects a chronic vascular thickening and atherosclerosis, while CFR reflects coronary microvascular function which seems to precede atherosclerotic structural damages. So, early detection of preclinical coronary atherosclerosis in ERA patients is of great interest to improve both quality of life and survival in this population.
Although early treatment is crucial to achieve optimal results concerning disease progression and joint structural damage , little is known on vascular effects of traditional DMARDs and anti-TNFα drugs.
Previous studies [52–55] reported controversial data on the effects of antirheumatic drugs on endothelial function. In particular, Hurlimann et al.  for the first time demonstrated that short time anti-TNFα treatment is able not only to ameliorate the disease process but also substantially to improve vascular function in 11 RA patients. Moreover, Bilsborough et al.  observed that the addition of anti-TNFα agents to traditional antirheumatic drugs reduced inflammatory symptoms and improved endothelial dysfunction in patients with severe RA.
On the contrary, Hansel et al. (33) observed no beneficial effect on endothelial function for specific blockade of TNFα compared to traditional DMARDs in eight patients with stable RA and low disease activity.
Our results showed that the administration of both methotrexate and anti-TNFα drugs early after the onset of RA, not only reduced disease activity and chronic inflammatory status, as detected by the finding of significant decrease in DAS28 and CRP, but also retarded and even reversed the progression of endothelial dysfunction and atherosclerosis as showed by the increase in CFR. Interestingly, we did not observe any difference between adalimumab and methotrexate regarding the effects on disease activity and coronary microcirculation.
The absence of any changes in plasma ADMA levels despite improvement of CFR and reduction of disease activity could suggest that antirheumatic drugs may affect vascular function throughout pathophysiological patways different from NO cycle.
In accordance to this hypothesis, Rueda-Clausen et al.  had demonstrated that the severity of coronary artery disease in dyslipidemic patients is associated with inflammation process but not endothelial dysfunction evaluated by flow mediated dilation.
As pointed out by Avouac et al.  the effects of any powerful DMARDs on CV risk seem to be, at least in part, associated with the concomitant suppression of systemic inflammation which is known as one of the leading causes of enhanced atherosclerosis in RA.
Moreover, early data suggest that modification of the levels of inflammatory indices or improvement of the noninvasive markers of atherosclerosis, by therapeutic interventions may have a significant impact on the CV risk .
We could suppose that the improvement in the inflammatory status in our patients could be, at least partially, involved in the improvement of CFR after antirheumatic treatment and the lack of correlation between CRP levels and CFR could be due to the small number of patients enrolled.
Noteworthy, CRP not only reflects the effect of local vascular inflammation, but it may have a direct proinflammatory effect on endothelium related to enhancement of the proinflammatory inducible NO synthase and inhibition of the atheroprotective isoform of NO synthase . This would be of interest in relation to the association between endothelial dysfunction and atherosclerosis: on the basis of previous studies  increased circulating CRP may cause endothelial dysfunction. Moreover, in literature, it has been documented that values in the very low (<0.5 mg/L) and very high range (>10 mg/L) of CRP levels correspond to very low and very high risk for adverse CV events .
Recent studies evaluated the endothelial effects of the new drugs used in RA therapy such as anti-tumor necrosis factor alpha (TNFα) agents: Hurlimann et al.  have found that TNFα antagonism not only reduces RA activity but also improves endothelial function.
In conclusion, our study demonstrated that an early identification of patients with recent onset of RA and the early treatment of the disease with traditional or biological DMARDs are crucial not only to achieve disease activity and inflammatory status control but also to improve coronary microvascular dysfunction and CV outcomes.
Since the impact on CV risk resulting from changes in inflammatory mediators and noninvasive markers of subclinical atherosclerosis remain to be investigated, large-scale clinical trials would be necessary to better define the effects of DMARDs on long-term prognosis and prevention of CV complications in ERA patients.