Tumor Necrosis Factor-α Mediates Inflammation-induced Early-Stage Left Ventricular Systolic Dysfunction.

ABSTRACT: Elevated systemic inflammation contributes to pathogenesis of heart failure with preserved ejection fraction (HFpEF), but molecular mechanisms are poorly understood. Although left ventricular (LV) diastolic dysfunction is the main cause of HFpEF, subclinical systolic dysfunction also contributes. We have previously shown that rats with collagen-induced arthritis (CIA) have systemic inflammation, LV diastolic dysfunction, and that increased circulating TNF-α contributes to inflammation-induced HFpEF pathogenesis, but does not mediate LV diastolic dysfunction in CIA rats. Contribution of systemic inflammation to dysfunction of the active process of LV diastolic and systolic function are unknown. In the present study, we used the CIA rat model to investigate the effects of systemic inflammation and TNF-α blockade on systolic function, and mRNA expression of genes involved in active diastolic relaxation and of myosin heavy chain (MyHC) isoforms. Collagen inoculation and TNF-α blockade did not affect LV mRNA expression of genes that mediate active LV diastolic function. Collagen-induced inflammation impaired LV global longitudinal strain ( P = 0.03) and velocity ( P = 0.04). This impairment of systolic function was prevented by TNF-α blockade. Collagen inoculation decreased mRNA expression of α-MyHC ( Myh6, P = 0.03) and increased expression of ß-MyHC ( Myh7, P = 0.0002), a marker, which is upregulated in failing hearts. TNF-α blockade prevented this MyHC isoform-switch. These results show that increased circulating TNF-α changes the relative expression of MyHC isoforms, favoring ß-MyHC, which may underlie changes in contractile function that impair systolic function. Our results indicate that TNF-α initiates early-stage LV systolic, rather than LV diastolic dysfunction.

TNF-α inhibitors reduce inflammation-induced concentric remodelling, but not diastolic dysfunction in collagen-induced arthritis.

OBJECTIVES: To determine biologic disease-modifying anti-rheumatic drug effects on inflammation-induced cardiac geometry and function changes. METHODS: Male and female Sprague-Dawley rats (n=92) were divided into four groups: control group, collagen-induced arthritis (CIA) group, anti-TNF-α group and anti-IL-6 group. Inflammation was induced by injecting bovine type-II collagen emulsified in incomplete Freund's adjuvant at the base of the tail, in all groups except the control group. Following the onset of arthritis, the anti-TNF-α group received etanercept, and the anti-IL-6 group received tocilizumab, for six weeks. Left ventricular (LV) geometry and function were assessed with echocardiography and circulating inflammatory markers were measured with ELISA. RESULTS: Relative wall thickness in the CIA and anti-IL-6 groups were increased compared to controls (p<0.001 and p=0.02, respectively). TNF-α inhibition protected against inflammation-induced LV concentric remodelling, as relative wall thickness in the anti-TNF-α group was similar to controls (p=0.55). Systolic function variables were not different between the groups. In all groups inoculated with collagen, myocardial relaxation (lateral e') were impaired compared to controls (all p<0.001). LV filling pressures (E/e') were increased in the CIA, anti-TNF-α and anti-IL-6 groups compared to controls (p<0.001, p<0.001 and p=0.05, respectively). Independent of concentric remodelling, circulating CRP concentrations were associated with decreased e' and increased E/e', while TNF-α concentrations were associated with E/A. CONCLUSIONS: TNF-α inhibition protected against inflammation-induced adverse cardiac remodelling, but not diastolic dysfunction. IL-6 receptors blocker effects on inflammation-induced cardiac changes were inconclusive. Systemic inflammation likely impacts LV concentric remodelling and diastolic dysfunction through distinct pathways.

Increased protein phosphatase 5 expression in inflammation-induced left ventricular dysfunction in rats.

BACKGROUND: Titin phosphorylation contributes to left ventricular (LV) diastolic dysfunction. The independent effects of inflammation on the molecular pathways that regulate titin phosphorylation are unclear. METHODS: We investigated the effects of collagen-induced inflammation and subsequent tumor necrosis factor-α (TNF-α) inhibition on mRNA expression of genes involved in regulating titin phosphorylation in 70 Sprague-Dawley rats. LV diastolic function was assessed with echocardiography. Circulating inflammatory markers were quantified by enzyme-linked immunosorbent assay and relative LV gene expression was assessed by Taqman® polymerase chain reaction. Differences in normally distributed variables between the groups were determined by two-way analysis of variance (ANOVA), followed by Tukey post-hoc tests. For non-normally distributed variables, group differences were determined by Kruskal-Wallis tests. RESULTS: Collagen inoculation increased LV relative mRNA expression of vascular cell adhesion molecule 1 (VCAM1), pentraxin 3 (PTX3), and inducible nitric oxide synthase (iNOS) compared to controls, indicating local microvascular inflammation. Collagen inoculation decreased soluble guanylate cyclase alpha-2 (sGCα2) and soluble guanylate cyclase beta-2 (sGCß2) expression, suggesting downregulation of nitric oxide-soluble guanylate cyclase-cyclic guanosine monophosphate (NO-sGC-cGMP) signaling. Inhibiting TNF-α prevented collagen-induced changes in VCAM1, iNOS, sGCα2 and sGCß2 expression. Collagen inoculation increased protein phosphatase 5 (PP5) expression. Like LV diastolic dysfunction, increased PP5 expression was not prevented by TNF-α inhibition. CONCLUSION: Inflammation-induced LV diastolic dysfunction may be mediated by a TNF-α-independent increase in PP5 expression and dephosphorylation of the N2-Bus stretch element of titin, rather than by TNF-α-induced downregulation of NO-sGC-cGMP pathway-dependent titin phosphorylation. The steady rise in number of patients with inflammation-induced diastolic dysfunction, coupled with low success rates of current therapies warrants a better understanding of the systemic signals and molecular pathways responsible for decreased titin phosphorylation in development of LV diastolic dysfunction. The therapeutic potential of inhibiting PP5 upregulation in LV diastolic dysfunction requires investigation.

Arterial wave reflection and subclinical atherosclerosis in rheumatoid arthritis.

OBJECTIVES: Atherosclerotic cardiovascular disease risk is increased in rheumatoid arthritis (RA). Wave reflection occurs at arterial branching points, which are particularly prone to atherosclerosis. We explored the relationship of wave reflection with atherosclerosis in RA. METHODS: One hundred and sixty three RA patients (110 white, 31 Asian, 17 black and 5 of mixed ancestry) without cardiovascular disease participated. Arterial stiffness, wave reflection, pressure pulsatility, plaque in the extracranial carotid artery tree and the mean of the left and right common carotid arteries intima-thickness were determined. Associations were identified in multivariable regression models. RESULTS: One SD increase in reflected wave pressure (OR (95% CI) = 2.54 (1.41-4.44), p=0.001), reflection magnitude (OR (95% CI) = 1.84 (1.17-2.89), p=0.008), central pulse pressure (OR (95% CI) = 1.89 (1.12-3.22), p=0.02) and peripheral pulse pressure (OR (95% CI) = 2.09 (1.23-3.57), p=0.007) were associated with plaque. The association of wave reflection with plaque was independent of arterial stiffness and pressure pulsatility, and was present in both hypertensive and normotensive RA patients. In receiver operator characteristic curve analysis, the optimal cutoff value for reflected wave pressure in predicting plaque presence was 25 mmHg with a sensitivity, specificity, positive predictive value and negative predictive value of 45.2%, 89.3%, 78.6% and 66.2%, respectively; a reflected wave pressure of >25 mmHg was associated with plaque in univariate and adjusted analysis (p<0.0001 for both). Arterial function was not independently related to carotid intima-media thickness. CONCLUSIONS: Consideration and therapeutic targeting of wave reflection may improve cardiovascular disease prevention in RA.

Omentin concentrations are independently associated with those of matrix metalloproteinase-3 in patients with mild but not severe rheumatoid arthritis.

Omentin is an adipokine that reportedly protects against cardiometabolic risk. We investigated the relationships between omentin concentrations and subclinical cardiovascular disease in rheumatoid arthritis (RA). Omentin concentrations were measured in 213 (104 black; 109 white) RA patients. Relationships of omentin levels with those of endothelial activation markers, ultrasound determined carotid intima-media thickness and plaque, and matrix metalloproteinase (MMP)-2, -3 and -9 that mediate altered plaque stability, were identified in confounder adjusted multivariate regression models. Omentin concentrations were inversely associated with MMP-3 levels [ß = -364 (0.113), p = 0.002]. This relationship was influenced by population origin, RA activity and the erythrocyte sedimentation rate and joint deformity count (interaction p value = 0.009, 0.04, 0.04 and 0.007, respectively). Accordingly, the omentin-MMP-3 concentration relationship was reproduced in white [ß (SE) = -0.450 (0.153), p = 0.0004)] but not black patients [ß (SE) = -0.099 (0.195), p = 0.6)], in participants with disease remission or mild disease activity [ß (SE) = -0.411 (0.139), p = 0.004] but not with moderate or severe RA activity [ß (SE) = -0.286 (0.202), p = 0.2], and in those with a small [ß (SE) = -0.534 (0.161), p = 0.001] but not large erythrocyte sedimentation rate [-0.212 (0.168), p = 0.2] and without [ß (SE) = -0.554 (0.165), p = 0.0001] but not with large joint deformity counts [-0.110 (0.173), p = 0.5]. Omentin levels were unrelated to endothelial activation and atherosclerosis. Among patients with RA, a lack of plaque stabilizing effects by omentin may contribute to the reported link between severe disease and increased cardiovascular risk. The association between concentrations of omentin and MMP-3 is population specific in RA.

Interleukin-6 Blockers Improve Inflammation-Induced Lipid Metabolism Impairments but Induce Liver Fibrosis in Collagen-Induced Arthritis.

BACKGROUND: Interleukin-6 (IL-6) receptor blockers improve systemic inflammation, however, their inconsistent effects on lipid metabolism and drug-induced liver injuries warrant further investigation. This study aimed to determine the effects of IL-6 receptor blocker therapy on lipid metabolism and liver morphology in collagen-induced arthritis. METHODS: Sixty three Sprague Dawley rats were divided into control (n = 24), inflammation (n = 24), and IL-6 blocker (n = 15) groups. Inflammation was induced in the inflammation and IL-6- blocker groups using Bovine type-II collagen and incomplete Freund's adjuvant. At first signs of arthritis, the IL-6 blocker group received an IL-6 blocker, tocilizumab for six weeks. Serum concentrations of low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and ATP-binding cassette transporter-A1 (ABCA1) were measured. Liver fibrosis was determined by histological stains and liver enzymes were measured using the colorimetric-chemistry analyzer. RESULTS: In the inflammation group, HDL-C and ABCA1 were reduced compared to control (p < 0.0001 and p = 0.04, respectively) and IL-6 blocker (p = 0.0003 and p < 0.0001, respectively) groups. LDL-C was increased in the inflammation compared to control (p = 0.02). Markers of liver fibrosis were increased in the IL-6 blocker group compared to control and inflammation groups (picrosirius red collagen area fraction: p < 0.0001 and p = 0.0008, respectively; Masson's trichrome collagen area fraction: p = 0.0002 and p = 0.01, respectively). Alkaline phosphatase concentrations were increased in the IL-6 blocker group compared to the control (p < 0.0001) and inflammation (p = 0.002) groups. CONCLUSION: IL-6 blockers ameliorated inflammation-induced lipid metabolism impairments, however they induced liver fibrosis. Although IL-6 blockers may reduce inflammation-induced metabolic impairments in chronic inflammatory disorders, routine monitoring of liver function is warranted while on treatment.

The effect of TNF-α inhibitor treatment on microRNAs and endothelial function in collagen induced arthritis.

Chronic inflammation causes dysregulated expression of microRNAs. Aberrant microRNA expression is associated with endothelial dysfunction. In this study we determined whether TNF-α inhibition impacted the expression of miRNA-146a-5p and miRNA-155-5p, and whether changes in the expression of these miRNAs were related to inflammation-induced changes in endothelial function in collagen-induced arthritis (CIA). Sixty-four Sprague-Dawley rats were divided into control (n = 24), CIA (n = 24) and CIA+etanercept (n = 16) groups. CIA and CIA+etanercept groups were immunized with bovine type-II collagen, emulsified in incomplete Freund's adjuvant. Upon signs of arthritis, the CIA+etanercept group received 10mg/kg of etanercept intraperitoneally, every three days. After six weeks of treatment, mesenteric artery vascular reactivity was assessed using wire-myography. Serum concentrations of TNF-α, C-reactive protein, interleukin-6, vascular adhesion molecule-1 (VCAM-1) and pentraxin-3 (PTX-3) were measured by ELISA. Relative expression of circulating miRNA-146a-5p and miRNA-155-5p were determined using RT-qPCR. Compared to controls, circulating miRNA-155-5p, VCAM-1 and PTX-3 concentrations were increased, and vessel relaxation was impaired in the CIA (all p<0.05), but not in the CIA+etanercept (all p<0.05) groups. The CIA group had greater miRNA-146a-5p expression compared to the CIA+etanercept group (p = 0.005). Independent of blood pressure, miRNA-146a-5p expression was associated with increased PTX-3 concentrations (p = 0.03), while miRNA-155-5p expression was associated with impaired vessel relaxation (p = 0.01). In conclusion, blocking circulating TNF-α impacted systemic inflammation-induced increased expression of miRNA-146a-5p and miRNA-155-5p, which were associated with endothelial inflammation and impaired endothelial dependent vasorelaxation, respectively.

Inflammation-induced left ventricular fibrosis is partially mediated by tumor necrosis factor-α.

OBJECTIVE: To determine the mechanisms of inflammation-induced left ventricular (LV) remodeling and effects of blocking circulating tumor necrosis factor alpha (TNF-α) in a model of systemic inflammation. METHODS: Seventy Sprague-Dawley rats were divided into three groups: the control group, the collagen-induced arthritis (CIA) group, and the anti-TNF-α group. Inflammation was induced in the CIA and anti-TNF-α groups. Following the onset of arthritis, the anti-TNF-α group received the TNF-α inhibitor, etanercept, for 6 weeks. LV geometry and function were assessed with echocardiography. Circulating inflammatory markers were measured by ELISA and LV gene expression was assessed by comparative TaqMan® polymerase chain reaction. RESULTS: The LV relative gene expression of pro-fibrotic genes, transforming growth factor ß (TGFß) (p = 0.03), collagen I (Col1) (p < 0.0001), and lysyl oxidase (LOX) (p = 0.002), was increased in the CIA group compared with controls, consistent with increased relative wall thickness (p = 0.0009). Col1 and LOX expression in the anti-TNF-α group were similar to controls (both, p > 0.05) and tended to be lower compared to the CIA group (p = 0.06 and p = 0.08, respectively), and may, in part, contribute to the decreased relative wall thickness in the anti-TNF-α group compared to the CIA group (p = 0.03). In the CIA group, the relative gene expression of matrix metalloproteinase 2 (MMP2) and MMP9 was increased compared to control (p = 0.04) and anti-TNF-α (p < 0.0001) groups, respectively. CONCLUSION: Chronic systemic inflammation induces fibrosis and dysregulated LV extracellular matrix remodeling by increasing local cardiac pro-fibrotic gene expression, which is partially mediated by TNF-α. Inflammation-induced LV diastolic dysfunction is likely independent of myocardial fibrosis.

Early Wave Reflection and Pulse Wave Velocity Are Associated with Diastolic Dysfunction in Rheumatoid Arthritis.

Rheumatoid arthritis (RA) impacts arterial and diastolic function. This study examined whether arterial properties can determine diastolic function in RA. In 173 RA patients, arterial function measures including carotid femoral pulse wave velocity (PWV), central systolic and pulse pressure, pulse pressure amplification, and the magnitude and timing of the forward and reflected waves were measured using applanation tonometry. Diastolic function parameters including the ratio of early-to-late transmitral velocity (E/A) and ratio of E to the mean of the lateral and septal wall myocardial tissue lengthening (e') were measured using echocardiography. The timing of the reflected wave was associated with E/A; PWV was related to E/e'. The timing of the reflected wave, forward wave magnitude, and pulse pressure amplification were associated with impaired relaxation; PWV was related to increased left ventricular (LV) filling pressure. Early wave reflection and PWV are associated with LV-impaired relaxation and increased filling pressure, respectively, in RA.

Disease severity impacts the relationship of apelin with arterial function in patients with rheumatoid arthritis.

Apelin can improve arterial function by enhancing the expression of endothelial nitric oxide synthase but this effect depends markedly on endothelial integrity. We hypothesized that inflammation influences the potential impact of apelin on arterial function in rheumatoid arthritis (RA). We assessed the associations of apelin concentrations with arterial stiffness (pulse wave velocity), wave reflection (augmentation index, reflected wave pressure, and reflection magnitude), and pressure pulsatility (central systolic pressure (CSP), central pulse pressure (CPP), peripheral pulse pressure (PPP), pulse pressure amplification (PPamp), and forward wave pressure (Pf)) among 170 RA patients without cardiovascular disease. In multivariable regression models, apelin concentrations were not independently associated with arterial function measures (p ≥ 0.15) in all patients. Inflammation markers were not consistently associated with apelin levels but joint deformity counts, Disease Activity Score in 28 joints (DAS28), and erythrocyte sedimentation rate (ESR) impacted apelin-pressure pulsatility relations (interaction p ≤ 0.05). In stratified analysis, apelin was associated with CSP (partial r = - 0.33, p = 0.01), CPP (partial r = - 0.26, p = 0.04), PPamp (partial r = 0.27, p = 0.03), and Pf (partial r = - 0.33, p = 0.01) in patients without but not with joint deformities; apelin was related to CSP (partial r = - 0.24, p = 0.05) in those with a DAS28 joint < 2.8 (median value) (partial r = - 0.24, p = 0.05) but not ≥ 2.8, and to CSP (partial r = - 0.36, p = 0.003) in those with an erythrocyte sedimentation rate < 13 mm/h (median value) but not ≥ 13 mm/h. Apelin is associated with reduced pressure pulsatility in RA patients without but not with a high inflammatory burden. A loss of apelin protective effects on arterial function may contribute to the link between RA severity and cardiovascular risk.

Nesfatin-1 and visfatin expression is associated with reduced atherosclerotic disease risk in patients with rheumatoid arthritis.

Nesfatin is an anti-inflammatory molecule that reduces atherosclerotic cardiovascular risk. By contrast, visfatin has pro-inflammatory properties and is pro-atherogenic. We examined the potential impact of nesfatin and visfatin on atherosclerotic disease in 232 (113 black and 119 white) consecutive rheumatoid arthritis (RA) patients from 2 centers. Independent relationships of nesfatin and visfatin concentrations with metabolic risk factors, endothelial activation, carotid atherosclerosis and altered plaque stability were determined in multivariable regression models. Rheumatoid factor (RF) positivity was associated with both nesfatin (ß = 0.650, p < 0.0001) and visfatin levels (ß = 0.157, p = 0.03). Visfatin concentrations were related to increased diastolic blood pressure (ß = 4.536, p = 0.01) and diabetes prevalence (ß = 0.092, p = 0.04). Nesfatin levels were associated with reduced carotid intima-media thickness (ß = -0.017, p = 0.008). Nesfatin (ß = 0.116, p = 0.001) and visfatin concentrations (ß = 0.234, p = 0.001) were related to those of matrix metalloproteinase-2 (MMP-2), a plaque stability mediator. Nesfatin and visfatin concentrations were directly correlated (Spearman's rho = 0.516). The nesfatin-MMP-2 and visfatin-MMP-2 relations were both stronger in RF negative compared to RF positive patients (interaction p = 0.01 and p = 0.04, respectively). Nesfatin is associated with reduced atherosclerosis and increased plaque stability mediator levels in RA. Visfatin is related to adverse cardio-metabolic risk in RA. Increased MMP-2 expression in relation to visfatin may represent a compensatory mechanism aimed at reducing cardiovascular risk in RA.

The Impact of Different Classification Criteria Sets on the Estimated Prevalence and Associated Risk Factors of Diastolic Dysfunction in Rheumatoid Arthritis.

This study compared the estimated prevalence and potential determinants of left ventricular (LV) diastolic dysfunction upon applying different classification criteria in rheumatoid arthritis (RA). LV diastolic function was assessed echocardiographically by pulsed Doppler (E/A), tissue Doppler (E/e', lateral and septal e'), and left atrial volume index in 176 RA patients. Relationships of traditional cardiovascular risk factors and RA characteristics with LV diastolic function and dysfunction according to previous and current criteria were determined in multivariate regression models. Waist-hip ratio was associated with E/A (standardised ß (SE) = -0.28 ± 0.09, p = 0.0002) and lateral e' (standardised ß (SE) = 0.26 ± 0.09, p = 0.01); low diastolic blood pressure was related to E/e' (standardised ß (SE) = -0.16 ± 0.08, p = 0.04). Diastolic dysfunction prevalence differed upon applying previous (59%) compared to current (22%) criteria (p < 0.0001). One SD increase in waist-hip ratio was associated with diastolic dysfunction when applying current criteria (OR = 2.61 (95% CI = 1.51-4.52), p = 0.0006), whereas one SD increase in diastolic blood pressure was inversely related to diastolic dysfunction upon using previous criteria (OR = 0.57 (95% CI = 0.40-0.81), p = 0.002). In conclusion, application of current and previous diastolic dysfunction criteria markedly alters the prevalence and risk factors associated with diastolic dysfunction in RA.

Cardiovascular Risk Factors and Disease Characteristics Are Consistently Associated with Arterial Function in Rheumatoid Arthritis.

OBJECTIVE: Arterial properties influence cardiovascular disease (CVD) risk. We identified potential determinants of arterial function in patients with rheumatoid arthritis (RA). METHODS: Relationships of traditional cardiovascular risk factors and RA characteristics with arterial stiffness (pulse wave velocity), wave reflection (augmentation index, reflected wave pressure, and reflection magnitude), and pressure pulsatility (central systolic and pulse pressure, peripheral pulse pressure, pulse pressure amplification, and forward wave pressure) were identified in multivariable backward regression models among 177 patients without established CVD (118 white, 32 Asian, 22 black, 5 mixed ancestry). RESULTS: Recorded characteristics explained 37% (pulse wave velocity) to 71% (reflected wave pressure) of the variability in arterial function. These factors were particularly associated with wave reflection and pressure pulsatility: RA duration (p = 0.04), rheumatoid factor status (p = 0.01 to 0.03), leukocyte counts (p = 0.02 to 0.05), and total cholesterol (p < 0.01 to 0.03). Body mass index (p < 0.01 to 0.02) and insulin resistance (p < 0.01 to 0.01) were related to reduced wave reflection and peripheral pulse pressure. Exercise (p = 0.02) and alcohol consumption (p < 0.01) were associated with increased pulse pressure amplification and decreased peripheral pulse pressure, respectively. Tumor necrosis factor-α inhibition (p < 0.01) was related to reduced pulse wave velocity, and tetracycline use (p = 0.02) to decreased peripheral pulse pressure. CONCLUSION: Traditional cardiovascular risk factors and disease characteristics are consistently associated with vascular hemodynamic alterations in RA. The relative effect of arterial stiffness, wave reflection, and pressure pulsatility on CVD risk in RA needs further study.

Apelin concentrations are associated with altered atherosclerotic plaque stability mediator levels and atherosclerosis in rheumatoid arthritis.

BACKGROUND AND AIMS: Apelin-APJ signaling reduces cardiovascular disease (CVD) risk. In rheumatoid arthritis (RA), the atherosclerosis burden and plaque vulnerability to rupture are increased. We explored relationships between apelin concentrations and subclinical CVD in RA. METHODS: Apelin levels were measured in 235 (114 black, 121 white) RA patients. Associations between apelin concentrations and ultrasound determined carotid artery intima-media thickness (cIMT) and plaque, and levels of matrix metalloproteinase (MMP)-2 and -9 that mediate plaque stability and vulnerability respectively, were identified in confounder adjusted multivariate regression analysis. RESULTS: In all patients, apelin concentrations were directly associated with those of MMP-2 (ß (SE) = 0.324 (0.112), p = 0.004) and inversely with those of MMP-9 (ß (SE) = -0.239 (0.060), p = 0.000). Apelin concentration-subclinical CVD relations were influenced by population origin, RA disease activity, erythrocyte sedimentation rate (ESR) and interleukin (IL)-6 concentrations (interaction p = 0.001 to 0.04). Accordingly, the apelin-MMP-2 concentration relationship was reproduced in white (ß (SE) = 0.367 (0.146), p = 0.01) but not black RA patients (ß (SE) = 0.197 (0.220), p = 0.4), and only in those without (but not with) large erythrocyte sedimentation rates (ß (SE) = 0.428 (0.143), p = 0.003) or interleukin-6 levels (ß (SE) = 0.485 (0.288), p = 0.04). By contrast, the apelin-MMP-9 concentration relation was reproduced more consistently. Apelin levels were inversely related to cIMT in patients with RA remission or mild (ß (SE) = -0.068 (0.033), p = 0.04) but not moderate or high disease activity (ß (SE) = 0.015 (0.112), p = 0.7). CONCLUSIONS: Apelin concentrations are associated with altered plaque stability mediator levels and atherosclerosis in patients with RA. These relations are partially dependent on population origin and systemic inflammatory status.