Advertisement

Prevalence, risk factors and proteomic bioprofiles associated with heart failure in rheumatoid arthritis: The RA-HF study

Published:November 07, 2020DOI:https://doi.org/10.1016/j.ejim.2020.11.002

      Highlights

      • Rheumatoid arthritis patients have high risk of heart failure. Age, cardiovascular risk factors, and rheumatoid arthritis duration increase the heart failure risk.
      • Patients with rheumatoid arthritis and heart failure largely share similar mechanistic pathways with patients with heart failure without rheumatoid arthritis. These include a higher expression of biomarkers associated with congestion, inflammation, fibrosis and RAAS activation.
      • Treatments that prolong the event-free survival of heart failure patients without rheumatoid arthritis should also be formally tested in patients with concomitant heart failure and rheumatoid arthritis.

      Abstract

      Background

      Rheumatoid arthritis (RA) patients have high risk of heart failure (HF).

      Aims

      Identifying the risk factors and mechanistic pathways associated with HF in patients with RA.

      Methods

      Cohort study enrolling 355 RA patients. HF was defined according to the ESC criteria. 93 circulating protein-biomarkers (91CVDIIOlink®+troponin-T+c-reactive protein) were measured. Regression modeling (multivariate and multivariable) were built and network analyses were performed - based on the identified relevant protein biomarkers.

      Results

      115 (32.4%) patients fulfilled the ESC criteria for HF, but only 24 (6.8%) had a prior HF diagnosis. Patients with HF were older (67 vs. 55yr), had a longer RA duration (10 vs. 14yr), had more frequently diabetes, hypertension, obesity, dyslipidemia, atrial fibrillation, and ischemic arterial disease. Several protein-biomarkers remained independently associated with HF, the top (FDR1%) were adrenomedullin, placenta-growth-factor, TNF-receptor-11A, and angiotensin-converting-enzyme-2. The networks underlying the expression of these biomarkers pointed towards congestion, apoptosis, inflammation, immune system signaling and RAAS activation as central determinants of HF in RA. Similar HF-associated biomarker-pathways were externally found in patients without RA. Having RA plus HF increased the risk of cardiovascular events compared to RA patients without RF; adjusted-HR (95%CI)=2.37 (1.07-5.30), p=0.034

      Conclusion

      Age, cardiovascular risk factors, and RA duration increase the HF odds in patients with RA. Few RA patients had a correct prior HF diagnosis, but the presence of HF increased the patients` risk. RA patients with HF largely share the mechanistic pathways of HF patients without RA. Randomized HF trials should include patients with RA.

      ClinicalTrials.gov ID

      NCT03960515

      Graphical abstract

      Keywords

      Abbreviations:

      RA (rheumatoid arthritis), HF (heart failure), DMARDs (disease modifying antirheumatic drugs), TTE (transthoracic echocardiogram), CRP (c-reactive protein), Hs-TnT (high sensitivity troponin T), NT-pro BNP (N-terminal-pro brain natriuretic peptide), PEA (proximity extension assay), ADM (adrenomedullin), PGF (placenta growth factor), TNFRSF11A (tumor necrosis factor receptor superfamily member 11A), ACE2 (angiotensin-converting enzyme 2), GAL9 (galectin 9), SPON2 (spondin-2), GH (growth hormone), MERTK (tyrosine-protein kinase mer), TNFRSF10A (tumor necrosis factor receptor superfamily member 10A), CEACAM8 (carcinoembryonic antigen-related cell adhesion molecule 8), TF (tissue factor), PTX3 (pentraxin-related protein 3), IL-1ra (Interleukin-1 receptor antagonist protein), PRSS8 (prostasin), AGRP (agouti-related protein), CD4 (T-cell surface glycoprotein CD4)

      1. Introduction

      Rheumatoid Arthritis (RA) is a systemic chronic and progressive inflammatory disease, characterized by persistent synovitis and joint erosion [
      • Gabriel S.E.
      • Crowson C.S.
      • Kremers H.M.
      • Doran M.F.
      • Turesson C.
      • O'Fallon W.M.
      • Matteson E.L.
      Survival in rheumatoid arthritis: a population-based analysis of trends over 40 years.
      ]. It affects around 1% of the general population and mostly young people [
      • Lawrence R.C.
      • Helmick C.G.
      • Arnett F.C.
      • Deyo R.A.
      • Felson D.T.
      • Giannini E.H.
      • Heyse S.P.
      • Hirsch R.
      • Hochberg M.C.
      • Hunder G.G.
      • Liang M.H.
      • Pillemer S.R.
      • Steen V.D.
      • Wolfe F.
      Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States.
      ,
      • Gibofsky A.
      Overview of epidemiology, pathophysiology, and diagnosis of rheumatoid arthritis.
      ]. Management of patients with RA has improved over the last 30 years with the development of disease-modifying antirheumatic drugs (DMARDs) [
      • Ajeganova S.
      • Huizinga T.
      Sustained remission in rheumatoid arthritis: latest evidence and clinical considerations.
      ,
      • Kavanaugh A.
      • Fleischmann R.M.
      • Emery P.
      • Kupper H.
      • Redden L.
      • Guerette B.
      • Santra S.
      • Smolen J.S.
      Clinical, functional and radiographic consequences of achieving stable low disease activity and remission with adalimumab plus methotrexate or methotrexate alone in early rheumatoid arthritis: 26-week results from the randomised, controlled OPTIMA study.
      ]. Notwithstanding, many patients with RA remain undertreated and present a high cardiovascular risk [
      • Radovits B.J.
      • Fransen J.
      • Al Shamma S.
      • Eijsbouts A.M.
      • van Riel P.L.
      • Laan R.F.
      Excess mortality emerges after 10 years in an inception cohort of early rheumatoid arthritis.
      ,
      • Gabriel S.E.
      Cardiovascular morbidity and mortality in rheumatoid arthritis.
      ]. The risk of developing heart failure (HF), in particular, up to 2-fold higher in RA (compared to the general population) and HF is a major determinant of the poor cardiovascular outcomes in this population [
      • Ward M.M.
      Recent improvements in survival in patients with rheumatoid arthritis: better outcomes or different study designs.
      ,
      • Mason J.C.
      • Libby P.
      Cardiovascular disease in patients with chronic inflammation: mechanisms underlying premature cardiovascular events in rheumatologic conditions.
      ,
      • Gossec L.
      • Salejan F.
      • Nataf H.
      • Nguyen M.
      • Gaud-Listrat V.
      • Hudry C.
      • Breuillard P.
      • Dernis E.
      • Boumier P.
      • Durandin-Truffinet M.
      • Fannius J.
      • Fechtenbaum J.
      • Izou-Fouillot M.A.
      • Labatide-Alanore S.
      • Lebrun A.
      • LeDevic P.
      • LeGoux P.
      • Sacchi A.
      • Salliot C.
      • Sparsa L.
      • d'Andre F.L.
      • Dougados M.
      • Network R.R.
      Challenges of cardiovascular risk assessment in the routine rheumatology outpatient setting: an observational study of 110 rheumatoid arthritis patients.
      ,
      • Nicola P.J.
      • Maradit-Kremers H.
      • Roger V.L.
      • Jacobsen S.J.
      • Crowson C.S.
      • Ballman K.V.
      • Gabriel S.E.
      The risk of congestive heart failure in rheumatoid arthritis: a population-based study over 46 years.
      ,
      • Khalid U.
      • Egeberg A.
      • Ahlehoff O.
      • Lane D.
      • Gislason G.H.
      • Lip G.Y.H.
      • Hansen P.R.
      Incident heart failure in patients with rheumatoid arthritis: a nationwide cohort study.
      ]. Importantly, in patients with RA, HF symptoms are often hard to ascertain (due to concomitant osteoarticular disease and functional limitation) and the HF clinical presentation may be atypical [
      • Solomon D.H.
      • Karlson E.W.
      • Rimm E.B.
      • Cannuscio C.C.
      • Mandl L.A.
      • Manson J.E.
      • Stampfer M.J.
      • Curhan G.C.
      Cardiovascular morbidity and mortality in women diagnosed with rheumatoid arthritis.
      ,
      • Peters M.J.
      • van Halm V.P.
      • Voskuyl A.E.
      • Smulders Y.M.
      • Boers M.
      • Lems W.F.
      • Visser M.
      • Stehouwer C.D.
      • Dekker J.M.
      • Nijpels G.
      • Heine R.
      • Dijkmans B.A.
      • Nurmohamed M.T.
      Does rheumatoid arthritis equal diabetes mellitus as an independent risk factor for cardiovascular disease? A prospective study.
      ]. In consequence, these patients tend to be underdiagnosed, managed less aggressively and have poor outcomes [
      • Myasoedova E.
      • Crowson C.S.
      • Kremers H.M.
      • Roger V.L.
      • Fitz-Gibbon P.D.
      • Therneau T.M.
      • Gabriel S.E.
      Lipid paradox in rheumatoid arthritis: the impact of serum lipid measures and systemic inflammation on the risk of cardiovascular disease.
      ].
      Despite the high HF burden among RA patients, little is known about the risk factors, clinical presentation and pathophysiological mechanisms associated with HF in this population [
      • Khalid U.
      • Egeberg A.
      • Ahlehoff O.
      • Lane D.
      • Gislason G.H.
      • Lip G.Y.H.
      • Hansen P.R.
      Incident heart failure in patients with rheumatoid arthritis: a nationwide cohort study.
      ]. HF and RA may share common disease pathways, and one may hypothesize that inflammation and fibrosis could be relevant drivers of HF in RA on top of the conventional cardiovascular risk factors (e.g., age, hypertension, diabetes, obesity, dyslipidemia and smoking), all leading to increased endothelial dysfunction, arterial stiffness and cardiac dysfunction [
      • Mason J.C.
      • Libby P.
      Cardiovascular disease in patients with chronic inflammation: mechanisms underlying premature cardiovascular events in rheumatologic conditions.
      ,
      • Brady S.R.
      • de Courten B.
      • Reid C.M.
      • Cicuttini F.M.
      • de Courten M.P.
      • Liew D.
      The role of traditional cardiovascular risk factors among patients with rheumatoid arthritis.
      ,
      • del Rincon I.
      • Polak J.F.
      • O'Leary D.H.
      • Battafarano D.F.
      • Erikson J.M.
      • Restrepo J.F.
      • Molina E.
      • Escalante A.
      Systemic inflammation and cardiovascular risk factors predict rapid progression of atherosclerosis in rheumatoid arthritis.
      ]. Nonetheless, further research is required to confirm this hypothesis and to better identify the risk factors and the associated disease mechanisms. The combination of clinical variables and biomarkers may improve the insight on the causes and biological pathways of HF in RA; thus, helping to raise the awareness, improving the detection and generating new hypotheses for the management and treatment of HF in RA.
      This study aims to determine the prevalence, clinical risk factors, and proteomic biomarkers with the respective underlying pathways associated with HF in patients with (and without) RA. Also, to estimate the HF associated risk for cardiovascular events.

      2. Methods

      2.1 Study design and patients

      We conducted a prospective single-center study in the Autoimmune Disease Unit of a Portuguese University Hospital (Centro Hospitalar do Porto, Porto, Portugal). This study had ethical approval in the Centro Hospitalar do Porto with the number 2016-023 (020-DEFI/020-CES) and was conducted following the principles of the Declaration of Helsinki. All participants provided written informed consent before enrollment in the study.
      The study was registered in clinicaltrials.gov under the number NCT03960515.
      A restricted database was constructed according to a prespecified case report form systematically filled by MBF. External and independent data cleaning and verification were performed by JPF.
      All patients with RA (2010 ACR/EULAR Classification Criteria [
      • Villeneuve E.
      • Nam J.
      • Emery P.
      2010 ACR-EULAR classification criteria for rheumatoid arthritis.
      ]) aged 18 years or older with follow-up in a specialized RA clinic were considered for inclusion. We excluded patients with an active neoplasm or with the presence of other severe co-morbid conditions with a life expectancy shorter than six months, severe dementia, inability to walk or totally dependent on a third person. The screening period occurred between June 2016 and June 2018. All patients maintained their usual care, according to the decision of their treating physician. Patients the median follow-up time was 1459 days (4.0 years). Cause of death and hospitalization were independently adjudicated. External and independent data cleaning, consolidation, and verification were performed.

      2.2 Patient evaluation, echocardiography and routine laboratory tests

      In the clinical evaluation we collected demographic data, cardiovascular history and risk factors, comorbid conditions, RA history, daily medication intake and a cardiovascular symptom questionnaire. Physical examination included vital signs, pulmonary and cardiac auscultation, and anthropometric measures.
      Transthoracic echocardiogram (TTE) was performed by an experienced echocardiographer and included M-mode, 2D and Doppler measurements acquired according to standard recommendations.
      Blood and urine samples were collected and promptly analyzed in the central laboratory of the Hospital including a complete blood count, c-reactive protein (CRP), glucose, blood electrolytes, lipid profile, creatinine, high sensitivity troponin T (hs-TnT), and N-Terminal-pro Brain Natriuretic Peptide (NT-pro BNP).

      2.3 Plasma proteomic biomarkers

      A large protein biomarker panel with 92 biomarkers from a wide range of pathophysiological domains was measured (Olink® CVDII panel). We selected this panel because it contains known human cardiovascular and inflammatory markers as well as some exploratory human proteins which may have potential as new markers of cardiovascular disease. These proteins are involved in multiple disease processes such as immune/inflammatory response, angiogenesis, coagulation and cell adhesion. An overview of biomarkers, their full names, Uniprot ID and main biological functions are presented in Supplemental Tables 1 and 2. The analytical assay characteristics are described in the Supplemental Addenda. Briefly, the biomarkers were measured using a high-throughput technique using the Olink Proseek® Multiplex CVD II 96 × 96 kit, which measures 92 manually selected cardiovascular-related proteins simultaneously in 1μl plasma samples. The kit uses a proximity extension assay (PEA) technology, where 92 oligonucleotide-labeled antibody probe pairs are allowed to bind to their proper target present in the sample. PEA is a homogeneous assay that uses pairs of antibodies equipped with DNA reporter molecules. When binding to their correct targets, they give rise to new DNA amplicons each ID-barcoding their respective antigens. The amplicons are subsequently quantified using a Fluidigm BioMark™ HD real-time PCR platform. The platform provides relative protein quantification as log2 normalized protein expression (NPX). In addition to the above referenced Olink® biomarkers, hsTnT was measured using the Elecsys Troponin T Gen 5 STAT test by Roche Diagnostics®. CRP was determined using the enzyme-linked immunosorbent assay (Olympus CRPLatex Calibrator Normal Set®). A total of 93 (91CVDII Olink® + hs TnT + CRP) circulating biomarkers were used for the analyses (BNP/NT-pro BNP was excluded as it is already used for HF definition). For the CVDII Olink® panel the mean intra-assay and inter-assay coefficient of variation were between 9.1% and 11.7%. For troponin and CRP the intra-assay and inter-assay coefficient of variation was 10% at a concentration lower than the 99th percentile cut-off.

      2.4 Heart failure prevalence and risk factors

      Two HF definitions were used: HF1, NT-pro BNP >125pg/ml and at least one structural or functional echocardiographic change (i.e., left ventricular hypertrophy, left atrial enlargement or left ventricular systolic or diastolic dysfunction), or use of loop diuretics, or HF history; and HF2, which is HF1 plus signs and symptoms (i.e., rales, peripheral oedema, breathlessness). In patients with RA, the signs and symptoms of HF may be difficult to ascertain due to osteoarticular involvement; thus, we used the HF1 definition for our primary analyses and, unless otherwise specified, the results presented in the manuscript refer to HF1. The HF2 results are shown in the supplement and largely overlap those of HF1 (c.f. results section).

      2.5 External replication/assessment of the findings in patients without rheumatoid arthritis

      To assess if our findings were unique to RA or whether HF in RA shared common pathways with HF in non-RA patients, we externality assessed our findings in two independent case-control cohorts of patients without RA. These cohorts have been previously described and the full results published [
      • Ferreira J.P.
      • Verdonschot J.
      • Collier T.
      • Wang P.
      • Pizard A.
      • Bar C.
      • Bjorkman J.
      • Boccanelli A.
      • Butler J.
      • Clark A.
      • Cleland J.G.
      • Delles C.
      • Diez J.
      • Girerd N.
      • Gonzalez A.
      • Hazebroek M.
      • Huby A.C.
      • Jukema W.
      • Latini R.
      • Leenders J.
      • Levy D.
      • Mebazaa A.
      • Mischak H.
      • Pinet F.
      • Rossignol P.
      • Sattar N.
      • Sever P.
      • Staessen J.A.
      • Thum T.
      • Vodovar N.
      • Zhang Z.Y.
      • Heymans S.
      • Zannad F.
      Proteomic bioprofiles and mechanistic pathways of progression to heart failure.
      ]. In short, a nested-matched case-control design was used with cases (incident HF), and controls (without HF) selected from 3 cohorts (>20 000 individuals). Controls were matched on cohort, follow-up time, age, and sex. Two independent sample sets (Ia with 286 cases and 591 controls and Ib with 276 cases and 280 controls) were used. The analysis presented in this reported are fully adjusted on the matching variables age, sex, and follow-up time plus age, sex, cohort, phase, follow-up time, smoking, diabetes mellitus, history of coronary artery disease, serum creatinine, body mass index, systolic blood pressure, use of antihypertensive medication, and heart rate.

      2.6 Statistical and bioinformatics considerations

      For continuous variables, normality was assessed by Shapiro-Wilks or Kolmogorov-Smirnov tests. If normally distributed, the results were summarized by the mean and standard deviation; otherwise, the median and the interquartile range was used. Student's t-test assessed the differences of the independent variables between the two groups, once normality was demonstrated; otherwise, the Mann-Whitney U was applied. Absolute (n) and relative (%) frequencies were reported for the categorical variables. The chi-square test assessed the association between the independent categorical variables and the outcomes.
      The main aim of this study was to ascertain the association of the multiple circulating proteins with HF and, subsequently, evaluate the relevant underlying mechanistic pathways. Clinical models were built for HF, considering the HF1 and HF2 outcome definitions. These models were used to correct for confounding in all the biomarker analysis. To build these data-driven models we have started with all the variables from the Table 1 with a p-value <0.1, forcing age and sex in the model. We have then applied a forward selection procedure setting a p-value <0.05 for a variable to be retained in the model. The final model was well calibrated and provided the highest area under the curve (AUC). The selected variables in this model were age, sex, diabetes, hypertension, dyspnoea, RA duration, hemoglobin, and renal function (estimated glomerular filtration rate). This pre-specified baseline clinical model presented a good predictive ability (AUC =0.81) and good calibration (goodness-of-fit p-value =0.9). On top of the clinical model, we assessed the prognostic association of each biomarker, correcting for false discoveries (FDR <0.05), as described by Benjamini and Hochberg [
      • Green G.H.
      • Diggle P.J.
      On the operational characteristics of the Benjamini and Hochberg false discovery rate procedure.
      ]. Only the biomarkers that passed this 5% threshold (plus clinical covariate adjustment) were deemed to be associated with HF. Since proteins were measured using NPX values on a log2 scale, the odds ratio for each protein estimates the increase in the odds of HF associated with a doubling in the relative protein concentration. These results are shown in Supplemental Tables 3 and 4.
      Table 1Characteristics of the study population for RA patients with and without HF.
      CharacteristicsNo HFHFp-value
      N.240115-
      Age (years), mean ± SD54.5 ± 12.767.1 ± 9.4<0.001
      Male, n. (%)59 (24.6%)22 (19.1%)0.25
      Comorbidities
      Diabetes mellitus, n. (%)23 (9.6%)26 (22.6%)<0.001
      Dyslipidemia, n. (%)101 (42.1%)70 (60.9%)<0.001
      Hypertension, n. (%)91 (37.9%)77 (67.0%)<0.001
      Myocardial infarction, n. (%)1 (0.4%)5 (4.3%)0.007
      Angina pectoris, n. (%)14 (5.8%)8 (7.0%)0.68
      Cerebrovascular Disease, n. (%)3 (1.3%)8 (7.0%)0.004
      Valvular heart disease, n. (%)4 (1.7%)12 (10.4%)<0.001
      Atrial fibrillation, n. (%)2 (0.8%)11 (9.6%)<0.001
      Prior HF diagnosis, n. (%)024 (15.5%)<0.001
      COPD, n. (%)12 (5.0%)6 (5.2%)0.93
      Anemia, n. (%)26 (10.8%)40 (34.8%)<0.001
      CKD, n. (%)3 (1.3%)9 (7.8%)0.001
      Smoking status
      Never, n. (%)149 (62.1%)80 (69.6%)0.011
      Past, n. (%)48 (20.0%)28 (24.3%)
      Current, n. (%)43 (17.9%)7 (6.1%)
      Symptoms and physical exam
      Dyspnea, n. (%)54 (22.5%)44 (38.3%)0.002
      Fatigue, n. (%)146 (60.8%)75 (65.2%)0.43
      Body mass index (kg/m2), mean ± SD26.5 ± 4.626.7 ± 4.20.62
      Waist circumference (cm), mean ± SD89.9 ± 12.793.7 ± 12.10.008
      SBP (mmHg), mean ± SD130.0 ± 17.3139.0 ± 21.4<0.001
      DBP (mmHg), mean ± SD75.5 ± 10.377.3 ± 9.70.12
      Heart rate (bpm), mean ± SD80.4 ± 13.680.9 ± 16.60.76
      RA history
      RA diagnostic (years), mean ± SD10.2 ± 9.214.4 ± 11.7<0.001
      RF or anti-CCP positive, n. (%)188 (78.3%)89 (77.4%)0.84
      Articular erosions, n. (%)71 (32.9%)52 (50.0%)0.003
      DAS28 VS (ESR), mean ± SD2.7 ± 1.23.0 ± 1.30.12
      DAS28 VS (CRP), mean ± SD2.3 ± 1.12.5 ± 1.20.33
      Disease activity
      Disease activity cutoffs were based on the American College of Rheumatology–recommended disease activity measure DAS28 (ESR or CRP) with remission <2.6, low/minimal ≥2.6 to <3.2, moderate ≥3.2 to ≤5.1, and high/severe >5.1 (Anderson J et al. Rheumatoid Arthritis Disease Activity Measures: American College of Rheumatology Recommendations for Use in Clinical Practice. Arthritis Care & Research. 2012. DOI: 10.1002/acr.21649).
      0.70
      Remission, n (%)115 (54.5%)43 (48.3%)
      Low, n (%)39 (18.5%)16 (18.0%)
      Moderate, n (%)45 (21.3%)24 (27.0%)
      Severe, n (%)12 (5.7%)6 (6.7%)
      Echocardiogram
      LVEF%, mean ± SD61.6 ± 6.259.9 ± 8.70.032
      LVEF <50%, n. (%)6 (2.6%)14 (12.6%)<0.001
      Laboratory tests
      Hb (g/dl), mean ± SD13.4 ± 1.312.5 ± 1.4<0.001
      eGFR (ml/min/1.73m2), mean ± SD92.6 ± 19.279.0 ± 19.2<0.001
      Na+ (mmol/l), mean ± SD142.0 ± 2.2142.2 ± 2.90.45
      K+ (mmol/l), mean ± SD4.4 ± 0.44.5 ± 0.40.084
      NT-pro BNP (pg/ml), median (IQR)56.3 (35.3, 87.8)229.1 (155.2, 544.0)<0.001
      CRP (mg/dl), median (IQR)2.6 (1.1, 6.6)4.2 (1.5, 9.6)0.006
      Medications
      ACEi/ARB, n. (%)72 (30.0%)64 (55.7%)<0.001
      Β-blocker, n. (%)12 (5.0%)28 (24.3%)<0.001
      MRA, n. (%)1 (0.4%)1 (0.9%)0.59
      Loop diuretic, n. (%)0 (0.0%)10 (8.7%)<0.001
      Statins, n. (%)78 (32.5%)56 (48.7%)0.003
      Corticosteroids, n. (%)96 (40.0%)66 (57.4%)0.002
      Corticosteroid dose (mg/day)
      Corticosteroid dose was converted to prednisolone equivalents.
      6.0 ± 2.66.4 ± 4.20.52
      Corticosteroid dose categories0.017
      ≤5 mg/day56 (60.2%)45 (70.3%)
      5-10 mg/day35 (37.6%)13 (20.3%)
      >10 mg/day2 (2.2%)6.9.4%)
      Methotrexate, n. (%)149 (62.1%)67 (58.3%)0.49
      NSAIDs, n. (%)61 (25.4%)24 (20.9%)0.35
      Biological DMARDs, n. (%)45 (18.8%)20 (17.4%)0.76
      Individual DMNARDs0.80
      Anti-TNFα26 (57.8%)12 (60.0%)
      Rituximab7 (15.6%)4 (20.0%)
      Tocilizumab12 (26.7%)4 (20.0%)
      Legend: COPD, chronic obstructive pulmonary disease; Hb, hemoglobin; CKD, chronic kidney disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; CRP, c-reactive protein; RA, rheumatoid arthritis; RF, rheumatoid factor; Anti-CCP, anti-citrulline antibody; DAS28 VS (ESR), Disease Activity Score for Rheumatoid Arthritis with Erythrocyte Sedimentation Rate; LVEF, left ventricular ejection fraction; eGFR, estimated Glomerular Filtration Rate; NT-pro BNP, N-terminal pro B-type Natriuretic Peptide; ACEi/ARB, angiotensin converting enzyme inhibitors or angiotensin-receptor blockers; MRA, mineralocorticoid receptor antagonists; MTX, methotrexate; NSAIDS, nonsteroidal anti-inflammatory drugs; DMARDS, disease-modifying anti-rheumatic drugs.
      low asterisk Disease activity cutoffs were based on the American College of Rheumatology–recommended disease activity measure DAS28 (ESR or CRP) with remission <2.6, low/minimal ≥2.6 to <3.2, moderate ≥3.2 to ≤5.1, and high/severe >5.1 (Anderson J et al. Rheumatoid Arthritis Disease Activity Measures: American College of Rheumatology Recommendations for Use in Clinical Practice. Arthritis Care & Research. 2012. DOI: 10.1002/acr.21649).
      low asterisklow asterisk Corticosteroid dose was converted to prednisolone equivalents.
      Additionally, we also performed a multivariate discriminant analysis using the whole set of measured biomarkers to furtherly explore the association between these circulating proteins with HF and test the internal robustness of our findings. We used partial least squares discriminant analysis (PLS-DA), as a strategy for dimension reduction of a set of independent variables, providing, simultaneously, a mechanism to assess the predictive power, regarding a categorical dependent variable; and orthogonal projection to latent structures (OPLS-DA) for distinguishing the variations in the data that are useful for the prediction of the outcome [
      • Boccard J.
      • Rutledge D.N.
      A consensus orthogonal partial least squares discriminant analysis (OPLS-DA) strategy for multiblock omics data fusion.
      ]. The univariate strategy provides a direct measure of the significance (FDR metric), by testing each one of the circulating biomarkers independently, whereas, the multivariate approach takes all the biomarkers simultaneously and establish a pattern of biomarkers relevance expressed by the variable importance to projection (VIP).
      Therefore, the combination of both relevance metrics (FDR and VIP), allows uncovering a set of robust biomarkers that are important for the characterization of HF. Following the identification of the relevant protein biomarkers, network analysis by consensusPathDB online server was performed, using an induced network approach [
      • Herwig R.
      • Hardt C.
      • Lienhard M.
      • Kamburov A.
      Analyzing and interpreting genome data at the network level with ConsensusPathDB.
      ]. This knowledge-based methodology allows pinpointing the links among the select biomarker proteins based on ontological findings (protein, genetic, biochemical, and gene regulatory interactions). Furthermore, the protein biomarkers were subjected to ClueGO network analysis to extract annotated functional and pathway information and facilitate the biological interpretation.
      Survival analyses were performed using Cox proportional hazards models and the Kaplan-Meier method. All statistical analyses were performed with Stata (version 16; StataCorp 2019, Stata Statistical Software: Release 16; College Station, TX: StataCorp LP), and R (R Core Team 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/).

      3. Results

      3.1 Patients` selection and characteristics

      A total of 408 patients with RA were screened between June 2016 and June 2018. Seven patients had exclusion criteria and 401 patients entered the study. A total of 365 patients underwent transthoracic echocardiogram. Of these, 2 did not perform biomarker assessment and 8 had missing values in main variables for HF definition. The remaining 355 patients were selected for this HF study. Supplemental Fig. 1.
      The baseline characteristics comparing RA patients with and without HF are depicted in Table 1. Of the total 355 RA patients, 115 (32.4%) had HF, of these only 24 (20.9% of the 115 HF cases and 6.8% of the 355-total population) had had a diagnosis of HF prior to the study. Patients with HF were older (67 vs. 55yr), had diabetes, dyslipidemia, hypertension, atherothrombotic disease, valvular heart disease, atrial fibrillation, anemia and CKD more often (p <0.05 for all). The great majority (>85%) of the patients with HF had a LVEF >50%. Patients with HF had longer RA duration (14 vs. 10yr), had more frequently an erosive disease, and used higher doses of corticosteroids (above 10 mg/day of prednisolone equivalent dose). Their NT-pro BNP and CRP levels were significantly higher; their BMI was similar but the waist circumference was larger. Patients with HF were treated with ACEi/ARBs and β-blockers more often, but the overall proportion of their use was low (56% for ACEi/ARBs and 24% for β-blockers). Table 1. The baseline characteristics of the patients with HF2 are similar and are described in Supplemental Table 5.

      3.2 Proteomic biomarkers independently associated with heart failure

      From the 93 measured proteins, 16 were independently associated with HF after clinical adjustment and correction for multiple comparisons. Table 2. Adrenomedullin (ADM), placenta growth factor (PGF), tumor necrosis factor receptor superfamily member 11A (TNFRSF11A), and angiotensin-converting enzyme 2 (ACE2) were associated with HF1 at a 1% FDR. Additionally, galectin-9 (GAL9), spondin-2 (SPON2), growth hormone (GH), tyrosine-protein kinase mer (MerTK), tumor necrosis factor receptor superfamily member 10A (TNFRSF10A), carcinoembryonic antigen-related cell adhesion molecule 8 (CEACAM8), tissue factor (TF), pentraxin-related protein 3 (PTX3), interleukin-1 receptor antagonist protein (IL-1ra), prostasin (PRSS8), agouti-related protein (AGRP), and T-cell surface glycoprotein CD4 (CD4) were associated with HF at a 5% FDR. Table 2. Similar associations were found for HF2 (ADM, ACE2, TNFRSF10A, TNFRSF11A, GAL9, PGF), and also CRP. Supplemental Table 6.
      Table 2Top proteins associated with HF.
      BiomarkerOR (95%CI)p valueFDR q
      HF1
      ADM4.18 (2.13-8.18)<0.0010.001
      PGF5.67 (2.32-13.87)<0.0010.002
      TNFRSF11A3.60 (1.89-6.86)<0.0010.002
      ACE22.43 (1.55-3.80)<0.0010.002
      GAL94.96 (1.99-12.33)<0.0010.011
      SPON211.93 (2.87-49.66)<0.0010.011
      GH1.29 (1.10-1.50)0.0010.017
      MerTK3.06 (1.49-6.31)0.0020.026
      TNFRSF10A3.46 (1.53-7.82)0.0030.026
      CEACAM82.07 (1.28-3.35)0.0030.026
      TF3.55 (1.51-8.37)0.0040.031
      PTX32.13 (1.27-3.60)0.0040.033
      IL-1ra2.02 (1.24-3.29)0.0050.034
      PRSS83.01 (1.38-6.55)0.0060.035
      AGRP2.57 (1.31-5.03)0.0060.035
      CD43.24 (1.40-7.51)0.0060.035
      Legend: ADM: adrenomedullin; PGF: placenta growth factor; TNFRSF11A: tumor necrosis factor receptor superfamily member 11A; ACE2: angiotensin-converting enzyme 2; GAL9: galectin 9; SPON2: spondin-2; GH: growth hormone; MERTK: tyrosine-protein Kinase Mer; TNFRSF10A: tumor necrosis factor receptor superfamily member 10A; CEACAM8: carcinoembryonic antigen-related cell adhesion molecule 8; TF: tissue factor; PTX3: pentraxin-related protein 3; IL-1ra: Interleukin-1 receptor antagonist protein; PRSS8: prostasin; AGRP: agouti-related protein; CD4: T-cell surface glycoprotein CD4.
      HF1 defined by NT-pro BNP >125pg/ml and at least one structural or functional echocardiographic change (i.e., left ventricular hypertrophy, left atrial enlargement or left ventricular systolic or diastolic dysfunction) or use of loop diuretics or HF history.
      OR, odd ratio by doubling in protein concentration (NPX).
      All models adjusted on age, sex, diabetes, hypertension, dyspnoea, RA duration, hemoglobin and renal function (estimated glomerular filtration rate).

      3.3 Modeling regression analysis and prediction profile of proteomic biomarkers for heart failure

      An OPLS-DA model with 1 latent variable and 1 orthogonal component was built using the circulating biomarkers proteins. Supplemental Fig. 2a depicts the scores plot showing the distribution of the samples denoting a small overlap between the two patient groups. This sample distribution is characterized by the variation shown in the loadings plot (Supplemental Fig. 2b), showing clearly that the vast majority of relevant (VIP ≥1) proteins, except for ANGPT1, are over-expressed in the patients with HF. Combining the findings (significant proteins) from both the logistic and multivariate modeling, by depicting the significance score of each biomarker graphically (Fig. 1), we could identify a set of protein biomarkers that had the strongest association with HF. These were ADM, PGF, TNFRSF11A, TNFRSF10A, SPON2, Gal9, TF, PRSS8, MERTK, CD4, and ACE2; that were also subjected to ClueGO and induced network functional analysis to facilitate the biological interpretation and pathway assessment.
      Fig 1
      Fig. 1Selection of relevant biomarkers based on the combination of multivariable logistic regression and multivariable discriminant analysis (significant biomarkers FDR <= 0.05 and VIP >1)
      Legend: ADM, adrenomedullin; PGF, placenta growth factor; TNFRSF11A, tumor necrosis factor receptor superfamily member 11A; TNFRSF10A, tumor necrosis factor receptor superfamily member 10A; SPON2, spondin-2; GAL9, galectin-9; TF, tissue factor; PRSS8, prostasin; MerTK, tyrosine-protein kinase mer; CD4, T-cell surface glycoprotein CD4; ACE2, angiotensin-converting enzyme 2.
      VIP, variable importance to projection; FDR, false-discovery rate.

      3.4 Network analysis

      Network analysis (ClueGO in tandem with consensuspathDB) pointed towards congestion, fibrosis, regulation of blood pressure, inflammation, apoptosis, and immune system signaling as the main mechanisms underlying the expression of these biomarkers (Fig. 2, panels a, b, & c).
      Fig 2
      Fig. 2Network topology, analysis and pathway over-representation of the “top” biomarkers identified from the multivariate modeling and logistic regression. (A) Network topology (ClueGO functional analysis). (B) Induced network analysis using the 11 identified protein biomarkers. (C) Pathway over-representation assessment of the 11 most relevant circulating protein biomarkers
      Fig 2
      Fig. 2Network topology, analysis and pathway over-representation of the “top” biomarkers identified from the multivariate modeling and logistic regression. (A) Network topology (ClueGO functional analysis). (B) Induced network analysis using the 11 identified protein biomarkers. (C) Pathway over-representation assessment of the 11 most relevant circulating protein biomarkers
      Fig 2
      Fig. 2Network topology, analysis and pathway over-representation of the “top” biomarkers identified from the multivariate modeling and logistic regression. (A) Network topology (ClueGO functional analysis). (B) Induced network analysis using the 11 identified protein biomarkers. (C) Pathway over-representation assessment of the 11 most relevant circulating protein biomarkers

      3.5 External replication/association of the biomarker in populations without rheumatoid arthritis

      The top biomarkers found in the RA population with HF were also found in two independent cohorts without RA: ADM, PGF, TNFRSF11A, TNFRSF10A, SPON2, GAL9, ACE2, and CD4 remained strongly associated with HF in these two independent cohorts at risk for HF but without RA. Whereas, PRSS8, MerTK, GH, TF, PTX3, and IL-1ra were not associated with HF in these cohorts without RA. Supplemental Table 7.

      3.6 Outcome associations

      During follow-up, 36 patients had a composite outcome of cardiovascular death or cardiovascular hospitalization. Patients with HF had 24 (20.9%) events and patients without HF had 12 (5.0%) events. The adjusted (age, sex, estimated glomerular filtration rate, diabetes and RA duration) hazard ratio (HR) HF1 was 2.37 (95% confidence interval: 1.07-5.30), p =0.034. Fig. 3.
      Fig 3
      Fig. 3Kaplan-Meier survival curve comparing RA patients with and without HF for the outcomes of cardiovascular death or cardiovascular hospitalization
      Legend: RA, rheumatoid arthritis; HF, heart failure; aHR, adjusted hazard ratio on age, sex, estimated glomerular filtration rate, diabetes and RA duration.
      y-axis, cumulative incidence; x-axis, time in years. P-value for the log-rank test <0.001.
      The summary of the findings is depicted in the Graphical Abstract.

      4. Discussion

      In this cohort of patients with RA, the prevalence of HF was high reaching almost one-third of the patients; but only 7% had had a diagnosis of HF prior to the study. Patients with HF had higher prevalence of traditional cardiovascular risk factors, including diabetes, dyslipidemia and hypertension, and expressed higher concentrations of biomarkers associated with congestion (e.g., adrenomedullin), inflammation (e.g., tumor necrosis factors receptor), fibrosis (e.g., galectin-9) and RAAS activation (e.g., angiotensin-converting enzyme 2). These pathways were also reported to be involved in HF in patients without RA. The presence of HF greatly increased the risk of subsequent cardiovascular events in RA. Collectively, these findings support that HF in RA may be underdiagnosed (and therefore undertreated) and that the mechanisms of the disease share common pathways to those found in HF patients without RA, suggesting that the treatments that improve outcomes in non-RA HF patients may also be effective in RA patients.
      The prevalence of HF in our RA cohort (32.4 %) was considerably higher than in the general population with a similar age (8-10% in people older than 60 years) [
      • Benjamin E.J.
      • Muntner P.
      • Alonso A.
      • Bittencourt M.S.
      • Callaway C.W.
      • Carson A.P.
      • Chamberlain A.M.
      • Chang A.R.
      • Cheng S.
      • Das S.R.
      • Delling F.N.
      • Djousse L.
      • Elkind M.S.V.
      • Ferguson J.F.
      • Fornage M.
      • Jordan L.C.
      • Khan S.S.
      • Kissela B.M.
      • Knutson K.L.
      • Kwan T.W.
      • Lackland D.T.
      • Lewis T.T.
      • Lichtman J.H.
      • Longenecker C.T.
      • Loop M.S.
      • Lutsey P.L.
      • Martin S.S.
      • Matsushita K.
      • Moran A.E.
      • Mussolino M.E.
      • O'Flaherty M.
      • Pandey A.
      • Perak A.M.
      • Rosamond W.D.
      • Roth G.A.
      • Sampson U.K.A.
      • Satou G.M.
      • Schroeder E.B.
      • Shah S.H.
      • Spartano N.L.
      • Stokes A.
      • Tirschwell D.L.
      • Tsao C.W.
      • Turakhia M.P.
      • VanWagner L.B.
      • Wilkins J.T.
      • Wong S.S.
      • Virani S.S.
      Heart Disease and Stroke Statistics-2019 Update: A Report From the American Heart Association.
      ,
      • Timmis A.
      • Townsend N.
      • Gale C.
      • Grobbee R.
      • Maniadakis N.
      • Flather M.
      • Wilkins E.
      • Wright L.
      • Vos R.
      • Bax J.
      • Blum M.
      • Pinto F.
      • Vardas P.
      Group, E. S. C. S. D.
      European Society of Cardiology: cardiovascular disease statistics 2017.
      ]. Notwithstanding, only 6.8% of all RA patients had a previous HF diagnosis, suggesting that an increased HF awareness is needed among the clinicians who treat patients with RA. Results from other cohorts suggest that this is not a pattern unique of our study [
      • Khalid U.
      • Egeberg A.
      • Ahlehoff O.
      • Lane D.
      • Gislason G.H.
      • Lip G.Y.H.
      • Hansen P.R.
      Incident heart failure in patients with rheumatoid arthritis: a nationwide cohort study.
      ]. Several factors could explain the sub-diagnosis of HF in RA. For example, the nonspecific signs and symptoms in the context of the osteoarticular pathology, e.g., tiredness and fatigue may be attributed to RA and not to HF; echocardiograms or natriuretic peptides are seldom performed (and the results of these tests may be unfamiliar to the treating physicians); and low awareness about HF among rheumatologists. Given the high risk of events in patients with concomitant RA and HF, there is an urgent need to improve the screening and treatment of HF in this population [
      • Ponikowski P.
      • Voors A.A.
      • Anker S.D.
      • Bueno H.
      • Cleland J.G.F.
      • Coats A.J.S.
      • Falk V.
      • Gonzalez-Juanatey J.R.
      • Harjola V.-P.
      • Jankowska E.A.
      • Jessup M.
      • Linde C.
      • Nihoyannopoulos P.
      • Parissis J.T.
      • Pieske B.
      • Riley J.P.
      • Rosano G.M.C.
      • Ruilope L.M.
      • Ruschitzka F.
      • Rutten F.H.
      • van der Meer P.
      Authors/Task Force Members; Document Reviewers
      2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC.
      ,
      • Ahmad F.S.
      • Ning H.
      • Rich J.D.
      • Yancy C.W.
      • Lloyd-Jones D.M.
      • Wilkins J.T.
      Hypertension, obesity, diabetes, and heart failure-free survival: the cardiovascular disease lifetime risk pooling project.
      ]. Notwithstanding, the prescription of drugs that improve outcomes in patients with heart failure with reduced ejection fraction, such as ACEi/ARBs and beta-blockers, were prescribed more frequently to patients with HF in our study. This can be explained by the higher prevalence of hypertension, history of myocardial infarction, and atrial fibrillation among HF patients. These co-morbid conditions may be treated with ACEi/ARBs and beta-blockers which, in consequence, were prescribed more frequently in the HF group even in the absence of a formal prior HF diagnosis. This is furtherly confirmed by the low and similar (between HF and no HF groups) prescription of mineralocorticoid receptor antagonists, which are indicated for patients with heart failure with reduced ejection fraction (IA guideline indication) and heart failure with preserved ejection fraction (IIB guideline indication) [
      • Yancy C.W.
      • Jessup M.
      • Bozkurt B.
      • Butler J.
      • Casey Jr., D.E.
      • Colvin M.M.
      • Drazner M.H.
      • Filippatos G.S.
      • Fonarow G.C.
      • Givertz M.M.
      • Hollenberg S.M.
      • Lindenfeld J.
      • Masoudi F.A.
      • McBride P.E.
      • Peterson P.N.
      • Stevenson L.W.
      • Westlake C.
      2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines and the Heart Failure Society of America.
      ,
      • Ponikowski P.
      • Voors A.A.
      • Anker S.D.
      • Bueno H.
      • Cleland J.G.
      • Coats A.J.
      • Falk V.
      • Gonzalez-Juanatey J.R.
      • Harjola V.P.
      • Jankowska E.A.
      • Jessup M.
      • Linde C.
      • Nihoyannopoulos P.
      • Parissis J.T.
      • Pieske B.
      • Riley J.P.
      • Rosano G.M.
      • Ruilope L.M.
      • Ruschitzka F.
      • Rutten F.H.
      • van der Meer P.
      2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC.
      ].
      Similarly to non-RA populations, HF risk in RA also increases with age and with common cardiovascular risk factors; moreover, RA duration likely plays a role in HF development, possibly due to the cumulative exposure to inflammation and pro-inflammatory factors [
      • Braunwald E.
      Heart failure.
      ,
      • Paulus W.J.
      • Tschope C.
      A novel paradigm for heart failure with preserved ejection fraction: comorbidities drive myocardial dysfunction and remodeling through coronary microvascular endothelial inflammation.
      ]. On top of these factors, we found several plasma proteins associated with HF. Most of these proteins share roles in the pathogenesis of both RA and HF [
      • Chosa E.
      • Hamada H.
      • Ohkura T.
      Role of adrenomedullin in patients with rheumatoid arthritis.
      ,
      • Chang Y.
      • Wei W.
      Angiotensin II in inflammation, immunity and rheumatoid arthritis.
      ,
      • Aggarwal S.
      • Dabla P.K.
      • Arora S.
      Prostasin: an epithelial sodium channel regulator.
      ]. For example, adrenomedullin, angiotensin-converting enzyme-2, placenta growth factor, galectin, interleukins, and the TNF superfamily members play an important role in the initiation and progression of chronic arthritis, and the levels of these proteins were found to be higher in the synovial tissues of RA patients compared to patients with osteoarthritis[
      • Chosa E.
      • Hamada H.
      • Ohkura T.
      Role of adrenomedullin in patients with rheumatoid arthritis.
      ,
      • Ribon M.
      • Mussard J.
      • Semerano L.
      • Singer B.B.
      • Decker P.
      Extracellular chromatin triggers release of soluble CEACAM8 upon activation of neutrophils.
      ,
      • Wu Y.
      • Li M.
      • Zeng J.
      • Feng Z.
      • Yang J.
      • Shen B.
      • Zeng Y.
      Differential expression of renin-angiotensin system-related components in patients with rheumatoid arthritis and osteoarthritis.
      ,
      • Michou L.
      • Garnier S.
      • Barbet S.
      • Glikmans E.
      • Ea H.K.
      • Uzan B.
      • Asensio C.
      • Ah Kioon M.D.
      • Lasbleiz S.
      • Bardin T.
      • Cornelis F.
      • Liote F.
      Lack of linkage and association of adrenomedullin and its receptor genes in French Caucasian rheumatoid arthritis trio families.
      ,
      • Cobankara V.
      • Ozturk M.A.
      • Kiraz S.
      • Ertenli I.
      • Haznedaroglu I.C.
      • Pay S.
      • Calguneri M.
      Renin and angiotensin-converting enzyme (ACE) as active components of the local synovial renin-angiotensin system in rheumatoid arthritis.
      ,
      • Bernstein K.E.
      • Khan Z.
      • Giani J.F.
      • Cao D.Y.
      • Bernstein E.A.
      • Shen X.Z.
      Angiotensin-converting enzyme in innate and adaptive immunity.
      ]. In HF, higher levels of these biomarkers have been associated with increased disease severity (regardless of the left ventricular ejection fraction) and worse prognosis (higher mortality and hospital readmission) [
      • Gonzalez-Rey E.
      • Chorny A.
      • O'Valle F.
      • Delgado M.
      Adrenomedullin protects from experimental arthritis by down-regulating inflammation and Th1 response and inducing regulatory T cells.
      ,
      • Voors A.A.
      • Kremer D.
      • Geven C.
      • Ter Maaten J.M.
      • Struck J.
      • Bergmann A.
      • Pickkers P.
      • Metra M.
      • Mebazaa A.
      • Dungen H.D.
      • Butler J.
      Adrenomedullin in heart failure: pathophysiology and therapeutic application.
      ,
      • Kremer D.
      • Ter Maaten J.M.
      • Voors A.A.
      Bio-adrenomedullin as a potential quick, reliable, and objective marker of congestion in heart failure.
      ,
      • Draker N.
      • Torry D.S.
      • Torry R.J.
      Placenta growth factor and sFlt-1 as biomarkers in ischemic heart disease and heart failure: a review.
      ,
      • Hammadah M.
      • Georgiopoulou V.V.
      • Kalogeropoulos A.P.
      • Weber M.
      • Wang X.
      • Samara M.A.
      • Wu Y.
      • Butler J.
      • Tang W.H.
      Elevated soluble fms-like tyrosine kinase-1 and placental-like growth factor levels are associated with development and mortality risk in heart failure.
      ,
      • Molvin J.
      • Jujic A.
      • Bachus E.
      • Gallo W.
      • Tasevska-Dinevska G.
      • Holm H.
      • Melander O.
      • Fedorowski A.
      • Magnusson M.
      Cardiovascular biomarkers predict post-discharge re-hospitalization risk and mortality among Swedish heart failure patients.
      ,
      • Dinarello C.A.
      The role of the interleukin-1-receptor antagonist in blocking inflammation mediated by interleukin-1.
      ,
      • Zhu R.
      • Liu C.
      • Tang H.
      • Zeng Q.
      • Wang X.
      • Zhu Z.
      • Liu Y.
      • Mao X.
      • Zhong Y.
      Serum galectin-9 levels are associated with coronary artery disease in Chinese individuals.
      ]. Other biomarkers, such as MerTK - a major apoptotic cell receptor on macrophages, SPON2 - an extracellular matrix protein, and GH – growth hormone, PTX3 - an acute-phase reactant, CEACAM8 – involved in cell adhesion and migration, PRSS8 - a protein that promotes proteolytic processes in epithelial tissues, and TF - a trigger of coagulation cascade, may also play a role both in RA and HF through several mechanisms linked to inflammation, apoptosis, vascular dysfunction and pro-coagulative states [
      • Paulus W.J.
      • Tschope C.
      A novel paradigm for heart failure with preserved ejection fraction: comorbidities drive myocardial dysfunction and remodeling through coronary microvascular endothelial inflammation.
      ,
      • Aggarwal S.
      • Dabla P.K.
      • Arora S.
      Prostasin: an epithelial sodium channel regulator.
      ,
      • O'Brien M.J.
      • Shu Q.
      • Stinson W.A.
      • Tsou P.S.
      • Ruth J.H.
      • Isozaki T.
      • Campbell P.L.
      • Ohara R.A.
      • Koch A.E.
      • Fox D.A.
      • Amin M.A.
      A unique role for galectin-9 in angiogenesis and inflammatory arthritis.
      ,
      • de Couto G.
      • Jaghatspanyan E.
      • DeBerge M.
      • Liu W.
      • Luther K.
      • Wang Y.
      • Tang J.
      • Thorp E.B.
      • Marban E.
      Mechanism of enhanced MerTK-dependent macrophage efferocytosis by extracellular Vesicles.
      ,
      • Hou X.
      • Chen G.
      • Bracamonte-Baran W.
      • Choi H.S.
      • Diny N.L.
      • Sung J.
      • Hughes D.
      • Won T.
      • Wood M.K.
      • Talor M.V.
      • Hackam D.J.
      • Klingel K.
      • Davogustto G.
      • Taegtmeyer H.
      • Coppens I.
      • Barin J.G.
      • Cihakova D.
      The cardiac microenvironment instructs divergent monocyte fates and functions in myocarditis.
      ].
      Most of the proteins associated with HF in our study were also independently associated with HF in two independent cohorts including patients without RA [
      • Ferreira J.P.
      • Verdonschot J.
      • Collier T.
      • Wang P.
      • Pizard A.
      • Bar C.
      • Bjorkman J.
      • Boccanelli A.
      • Butler J.
      • Clark A.
      • Cleland J.G.
      • Delles C.
      • Diez J.
      • Girerd N.
      • Gonzalez A.
      • Hazebroek M.
      • Huby A.C.
      • Jukema W.
      • Latini R.
      • Leenders J.
      • Levy D.
      • Mebazaa A.
      • Mischak H.
      • Pinet F.
      • Rossignol P.
      • Sattar N.
      • Sever P.
      • Staessen J.A.
      • Thum T.
      • Vodovar N.
      • Zhang Z.Y.
      • Heymans S.
      • Zannad F.
      Proteomic bioprofiles and mechanistic pathways of progression to heart failure.
      ] suggesting that HF in RA shares common pathophysiological mechanisms with HF in non-RA patients. One may postulate that treatments that improve HF in non-RA patients may also be suitable for treating HF in RA, and also that the aggressive reduction of inflammation could improve outcomes in both RA and non-RA HF patients. Mineralocorticoid receptor antagonists (MRAs) may be particularly useful in this context as they lowered inflammatory and pro-fibrotic markers in patients at risk for developing HF and reduced HF hospitalizations in patients with heart failure with preserved ejection fraction (HFpEF) [
      • Yan L.
      • Wei X.
      • Tang Q.Z.
      • Feng J.
      • Zhang Y.
      • Liu C.
      • Bian Z.Y.
      • Zhang L.F.
      • Chen M.
      • Bai X.
      • Wang A.B.
      • Fassett J.
      • Chen Y.
      • He Y.W.
      • Yang Q.
      • Liu P.P.
      • Li H.
      Cardiac-specific mindin overexpression attenuates cardiac hypertrophy via blocking AKT/GSK3beta and TGF-beta1-SMAD signaling.
      ,
      • Pitt B.
      • Pfeffer M.A.
      • Assmann S.F.
      • Boineau R.
      • Anand I.S.
      • Claggett B.
      • Clausell N.
      • Desai A.S.
      • Diaz R.
      • Fleg J.L.
      • Gordeev I.
      • Harty B.
      • Heitner J.F.
      • Kenwood C.T.
      • Lewis E.F.
      • O'Meara E.
      • Probstfield J.L.
      • Shaburishvili T.
      • Shah S.J.
      • Solomon S.D.
      • Sweitzer N.K.
      • Yang S.
      • McKinlay S.M.
      Spironolactone for heart failure with preserved ejection fraction.
      ,
      • Pellicori P.
      • Ferreira J.P.
      • Mariottoni B.
      • Brunner-La Rocca H.P.
      • Ahmed F.Z.
      • Verdonschot J.
      • Collier T.
      • Cuthbert J.J.
      • Petutschnigg J.
      • Mujaj B.
      • Girerd N.
      • Gonzalez A.
      • Clark A.L.
      • Cosmi F.
      • Staessen J.A.
      • Heymans S.
      • Latini R.
      • Rossignol P.
      • Zannad F.
      • Cleland J.G.F.
      Effects of spironolactone on serum markers of fibrosis in people at high risk of developing heart failure: rationale, design and baseline characteristics of a proof-of-concept, randomised, precision-medicine, prevention trial. The Heart OMics in AGEing (HOMAGE) trial.
      ] - the great majority (>85%) of the HF patients in our study had a left ventricular ejection fraction >50%.

      5. Limitations

      Several limitations should be acknowledged in our study. First, this is an observational cohort study; thus, one cannot ascertain causality. Second, this cohort came from a single center and some of the findings may reflect local practice patterns. Third, the replication of the findings in non-RA populations was done in two case-control studies with patients “at-risk” for developing HF (and not with prevalent HF).

      6. Conclusion

      Age, classic cardiovascular risk factors, and RA duration increase the HF risk in patients with RA. These patients have increased congestion, inflammation and RAAS activation, features also found in patients without RA. Few RA patients had a correct prior HF diagnosis, but the presence of HF highly increased the patients` cardiovascular risk of a cardiovascular event. These findings show the clinical relevance of a better screening of HF in RA and underpin the mechanistic pathways in patients with RA and HF. Drugs that improve survival in HF patients without RA, should be tested in patients with concomitant RA and HF.

      Ethics approval and consent to participate

      This study had ethical approval in the Centro Hospitalar do Porto with the number 2016-023 (020-DEFI/020-CES) and was conducted following the principles of the Declaration of Helsinki. All participants provided written informed consent before enrollment in the study.

      Consent for publication

      Not applicable.

      Availability of data and materials

      The data and materials may be available upon reasonable request.

      Authors’ contributions

      M.B.F., drafted the manuscript; J.P.F., performed the statistical analysis and drafted the manuscript; all authors read, provided critical input and approved the final version of the manuscript.

      Funding

      None.

      Declaration of Competing Interest

      None

      Acknowledgment

      JPF, PR, FZ are supported by a public grant overseen by the French National Research Agency (ANR) as part of the second “Investissements d'Avenir” programme (ANR-15-RHU-0004). F.A. Saraiva is supported by Universidade do Porto/FMUP and FSE-Fundo Social Europeu, NORTE 2020-Programa Operacional Regional do Norte, NORTE-08-5369-FSE-000024-Programas Doutorais. ASB is supported by the DOCnet (NORTE-01-0145-FEDER-000003), by Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF).
      Dr. João Pedro Ferreira is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

      Appendix. Supplementary materials

      References

        • Gabriel S.E.
        • Crowson C.S.
        • Kremers H.M.
        • Doran M.F.
        • Turesson C.
        • O'Fallon W.M.
        • Matteson E.L.
        Survival in rheumatoid arthritis: a population-based analysis of trends over 40 years.
        Arthritis Rheum. 2003; 48: 54-58
        • Lawrence R.C.
        • Helmick C.G.
        • Arnett F.C.
        • Deyo R.A.
        • Felson D.T.
        • Giannini E.H.
        • Heyse S.P.
        • Hirsch R.
        • Hochberg M.C.
        • Hunder G.G.
        • Liang M.H.
        • Pillemer S.R.
        • Steen V.D.
        • Wolfe F.
        Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States.
        Arthritis Rheum. 1998; 41: 778-799
        • Gibofsky A.
        Overview of epidemiology, pathophysiology, and diagnosis of rheumatoid arthritis.
        Am J Manage Care. 2012; 18: S295-S302
        • Ajeganova S.
        • Huizinga T.
        Sustained remission in rheumatoid arthritis: latest evidence and clinical considerations.
        Ther Adv Musculoskelet Dis. 2017; 9: 249-262
        • Kavanaugh A.
        • Fleischmann R.M.
        • Emery P.
        • Kupper H.
        • Redden L.
        • Guerette B.
        • Santra S.
        • Smolen J.S.
        Clinical, functional and radiographic consequences of achieving stable low disease activity and remission with adalimumab plus methotrexate or methotrexate alone in early rheumatoid arthritis: 26-week results from the randomised, controlled OPTIMA study.
        Ann Rheum Dis. 2013; 72: 64-71
        • Radovits B.J.
        • Fransen J.
        • Al Shamma S.
        • Eijsbouts A.M.
        • van Riel P.L.
        • Laan R.F.
        Excess mortality emerges after 10 years in an inception cohort of early rheumatoid arthritis.
        Arthritis Care Res. 2010; 62: 362-370
        • Gabriel S.E.
        Cardiovascular morbidity and mortality in rheumatoid arthritis.
        Am J Med. 2008; 121 (10): S9-14
        • Ward M.M.
        Recent improvements in survival in patients with rheumatoid arthritis: better outcomes or different study designs.
        Arthritis Rheum. 2001; 44: 1467-1469
        • Mason J.C.
        • Libby P.
        Cardiovascular disease in patients with chronic inflammation: mechanisms underlying premature cardiovascular events in rheumatologic conditions.
        Eur Heart J. 2015; 36 (482–49c)
        • Gossec L.
        • Salejan F.
        • Nataf H.
        • Nguyen M.
        • Gaud-Listrat V.
        • Hudry C.
        • Breuillard P.
        • Dernis E.
        • Boumier P.
        • Durandin-Truffinet M.
        • Fannius J.
        • Fechtenbaum J.
        • Izou-Fouillot M.A.
        • Labatide-Alanore S.
        • Lebrun A.
        • LeDevic P.
        • LeGoux P.
        • Sacchi A.
        • Salliot C.
        • Sparsa L.
        • d'Andre F.L.
        • Dougados M.
        • Network R.R.
        Challenges of cardiovascular risk assessment in the routine rheumatology outpatient setting: an observational study of 110 rheumatoid arthritis patients.
        Arthritis Care Res. 2013; 65: 712-717
        • Nicola P.J.
        • Maradit-Kremers H.
        • Roger V.L.
        • Jacobsen S.J.
        • Crowson C.S.
        • Ballman K.V.
        • Gabriel S.E.
        The risk of congestive heart failure in rheumatoid arthritis: a population-based study over 46 years.
        Arthritis Rheum. 2005; 52: 412-420
        • Khalid U.
        • Egeberg A.
        • Ahlehoff O.
        • Lane D.
        • Gislason G.H.
        • Lip G.Y.H.
        • Hansen P.R.
        Incident heart failure in patients with rheumatoid arthritis: a nationwide cohort study.
        J Am Heart Assoc. 2018; 7
        • Solomon D.H.
        • Karlson E.W.
        • Rimm E.B.
        • Cannuscio C.C.
        • Mandl L.A.
        • Manson J.E.
        • Stampfer M.J.
        • Curhan G.C.
        Cardiovascular morbidity and mortality in women diagnosed with rheumatoid arthritis.
        Circulation. 2003; 107: 1303-1307
        • Peters M.J.
        • van Halm V.P.
        • Voskuyl A.E.
        • Smulders Y.M.
        • Boers M.
        • Lems W.F.
        • Visser M.
        • Stehouwer C.D.
        • Dekker J.M.
        • Nijpels G.
        • Heine R.
        • Dijkmans B.A.
        • Nurmohamed M.T.
        Does rheumatoid arthritis equal diabetes mellitus as an independent risk factor for cardiovascular disease? A prospective study.
        Arthritis Rheum. 2009; 61: 1571-1579
        • Myasoedova E.
        • Crowson C.S.
        • Kremers H.M.
        • Roger V.L.
        • Fitz-Gibbon P.D.
        • Therneau T.M.
        • Gabriel S.E.
        Lipid paradox in rheumatoid arthritis: the impact of serum lipid measures and systemic inflammation on the risk of cardiovascular disease.
        Ann Rheum Dis. 2011; 70: 482-487
        • Brady S.R.
        • de Courten B.
        • Reid C.M.
        • Cicuttini F.M.
        • de Courten M.P.
        • Liew D.
        The role of traditional cardiovascular risk factors among patients with rheumatoid arthritis.
        J Rheumatol. 2009; 36: 34-40
        • del Rincon I.
        • Polak J.F.
        • O'Leary D.H.
        • Battafarano D.F.
        • Erikson J.M.
        • Restrepo J.F.
        • Molina E.
        • Escalante A.
        Systemic inflammation and cardiovascular risk factors predict rapid progression of atherosclerosis in rheumatoid arthritis.
        Ann Rheum Dis. 2015; 74: 1118-1123
        • Villeneuve E.
        • Nam J.
        • Emery P.
        2010 ACR-EULAR classification criteria for rheumatoid arthritis.
        Rev Bras Reumatol. 2010; 50: 481-483
        • Ferreira J.P.
        • Verdonschot J.
        • Collier T.
        • Wang P.
        • Pizard A.
        • Bar C.
        • Bjorkman J.
        • Boccanelli A.
        • Butler J.
        • Clark A.
        • Cleland J.G.
        • Delles C.
        • Diez J.
        • Girerd N.
        • Gonzalez A.
        • Hazebroek M.
        • Huby A.C.
        • Jukema W.
        • Latini R.
        • Leenders J.
        • Levy D.
        • Mebazaa A.
        • Mischak H.
        • Pinet F.
        • Rossignol P.
        • Sattar N.
        • Sever P.
        • Staessen J.A.
        • Thum T.
        • Vodovar N.
        • Zhang Z.Y.
        • Heymans S.
        • Zannad F.
        Proteomic bioprofiles and mechanistic pathways of progression to heart failure.
        Circ Heart Fail. 2019; 12e005897
        • Green G.H.
        • Diggle P.J.
        On the operational characteristics of the Benjamini and Hochberg false discovery rate procedure.
        Stat Appl Genet Mol Biol. 2007; 6 (Article27)
        • Boccard J.
        • Rutledge D.N.
        A consensus orthogonal partial least squares discriminant analysis (OPLS-DA) strategy for multiblock omics data fusion.
        Anal Chim Acta. 2013; 769: 30-39
        • Herwig R.
        • Hardt C.
        • Lienhard M.
        • Kamburov A.
        Analyzing and interpreting genome data at the network level with ConsensusPathDB.
        Nat Protoc. 2016; 11: 1889-1907
        • Benjamin E.J.
        • Muntner P.
        • Alonso A.
        • Bittencourt M.S.
        • Callaway C.W.
        • Carson A.P.
        • Chamberlain A.M.
        • Chang A.R.
        • Cheng S.
        • Das S.R.
        • Delling F.N.
        • Djousse L.
        • Elkind M.S.V.
        • Ferguson J.F.
        • Fornage M.
        • Jordan L.C.
        • Khan S.S.
        • Kissela B.M.
        • Knutson K.L.
        • Kwan T.W.
        • Lackland D.T.
        • Lewis T.T.
        • Lichtman J.H.
        • Longenecker C.T.
        • Loop M.S.
        • Lutsey P.L.
        • Martin S.S.
        • Matsushita K.
        • Moran A.E.
        • Mussolino M.E.
        • O'Flaherty M.
        • Pandey A.
        • Perak A.M.
        • Rosamond W.D.
        • Roth G.A.
        • Sampson U.K.A.
        • Satou G.M.
        • Schroeder E.B.
        • Shah S.H.
        • Spartano N.L.
        • Stokes A.
        • Tirschwell D.L.
        • Tsao C.W.
        • Turakhia M.P.
        • VanWagner L.B.
        • Wilkins J.T.
        • Wong S.S.
        • Virani S.S.
        Heart Disease and Stroke Statistics-2019 Update: A Report From the American Heart Association.
        Circulation. 2019; 139: e56-e528
        • Timmis A.
        • Townsend N.
        • Gale C.
        • Grobbee R.
        • Maniadakis N.
        • Flather M.
        • Wilkins E.
        • Wright L.
        • Vos R.
        • Bax J.
        • Blum M.
        • Pinto F.
        • Vardas P.
        • Group, E. S. C. S. D.
        European Society of Cardiology: cardiovascular disease statistics 2017.
        Eur Heart J. 2018; 39: 508-579
        • Ponikowski P.
        • Voors A.A.
        • Anker S.D.
        • Bueno H.
        • Cleland J.G.F.
        • Coats A.J.S.
        • Falk V.
        • Gonzalez-Juanatey J.R.
        • Harjola V.-P.
        • Jankowska E.A.
        • Jessup M.
        • Linde C.
        • Nihoyannopoulos P.
        • Parissis J.T.
        • Pieske B.
        • Riley J.P.
        • Rosano G.M.C.
        • Ruilope L.M.
        • Ruschitzka F.
        • Rutten F.H.
        • van der Meer P.
        • Authors/Task Force Members; Document Reviewers
        2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC.
        Eur J Heart Fail. 2016; 18: 891-975
        • Ahmad F.S.
        • Ning H.
        • Rich J.D.
        • Yancy C.W.
        • Lloyd-Jones D.M.
        • Wilkins J.T.
        Hypertension, obesity, diabetes, and heart failure-free survival: the cardiovascular disease lifetime risk pooling project.
        JACC Heart Fail. 2016; 4: 911-919
        • Yancy C.W.
        • Jessup M.
        • Bozkurt B.
        • Butler J.
        • Casey Jr., D.E.
        • Colvin M.M.
        • Drazner M.H.
        • Filippatos G.S.
        • Fonarow G.C.
        • Givertz M.M.
        • Hollenberg S.M.
        • Lindenfeld J.
        • Masoudi F.A.
        • McBride P.E.
        • Peterson P.N.
        • Stevenson L.W.
        • Westlake C.
        2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines and the Heart Failure Society of America.
        J Am Coll Cardiol. 2017; 70: 776-803
        • Ponikowski P.
        • Voors A.A.
        • Anker S.D.
        • Bueno H.
        • Cleland J.G.
        • Coats A.J.
        • Falk V.
        • Gonzalez-Juanatey J.R.
        • Harjola V.P.
        • Jankowska E.A.
        • Jessup M.
        • Linde C.
        • Nihoyannopoulos P.
        • Parissis J.T.
        • Pieske B.
        • Riley J.P.
        • Rosano G.M.
        • Ruilope L.M.
        • Ruschitzka F.
        • Rutten F.H.
        • van der Meer P.
        2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC.
        Eur J Heart Fail. 2016;
        • Braunwald E.
        Heart failure.
        JACC Heart Fail. 2013; 1: 1-20
        • Paulus W.J.
        • Tschope C.
        A novel paradigm for heart failure with preserved ejection fraction: comorbidities drive myocardial dysfunction and remodeling through coronary microvascular endothelial inflammation.
        J Am Coll Cardiol. 2013; 62: 263-271
        • Chosa E.
        • Hamada H.
        • Ohkura T.
        Role of adrenomedullin in patients with rheumatoid arthritis.
        Rheumatoid arthritis - treatment. 2012
        • Chang Y.
        • Wei W.
        Angiotensin II in inflammation, immunity and rheumatoid arthritis.
        Clin Exp Immunol. 2015; 179: 137-145
        • Aggarwal S.
        • Dabla P.K.
        • Arora S.
        Prostasin: an epithelial sodium channel regulator.
        J Biomark. 2013; 2013179864
        • Ribon M.
        • Mussard J.
        • Semerano L.
        • Singer B.B.
        • Decker P.
        Extracellular chromatin triggers release of soluble CEACAM8 upon activation of neutrophils.
        Front Immunol. 2019; 10: 1346
        • Wu Y.
        • Li M.
        • Zeng J.
        • Feng Z.
        • Yang J.
        • Shen B.
        • Zeng Y.
        Differential expression of renin-angiotensin system-related components in patients with rheumatoid arthritis and osteoarthritis.
        Am J Med Sci. 2020; 359: 17-26
        • Michou L.
        • Garnier S.
        • Barbet S.
        • Glikmans E.
        • Ea H.K.
        • Uzan B.
        • Asensio C.
        • Ah Kioon M.D.
        • Lasbleiz S.
        • Bardin T.
        • Cornelis F.
        • Liote F.
        Lack of linkage and association of adrenomedullin and its receptor genes in French Caucasian rheumatoid arthritis trio families.
        Clin Exp Rheumatol. 2008; 26: 1083-1086
        • Cobankara V.
        • Ozturk M.A.
        • Kiraz S.
        • Ertenli I.
        • Haznedaroglu I.C.
        • Pay S.
        • Calguneri M.
        Renin and angiotensin-converting enzyme (ACE) as active components of the local synovial renin-angiotensin system in rheumatoid arthritis.
        Rheumatol Int. 2005; 25: 285-291
        • Bernstein K.E.
        • Khan Z.
        • Giani J.F.
        • Cao D.Y.
        • Bernstein E.A.
        • Shen X.Z.
        Angiotensin-converting enzyme in innate and adaptive immunity.
        Nat Rev Nephrol. 2018; 14: 325-336
        • Gonzalez-Rey E.
        • Chorny A.
        • O'Valle F.
        • Delgado M.
        Adrenomedullin protects from experimental arthritis by down-regulating inflammation and Th1 response and inducing regulatory T cells.
        Am J Pathol. 2007; 170: 263-271
        • Voors A.A.
        • Kremer D.
        • Geven C.
        • Ter Maaten J.M.
        • Struck J.
        • Bergmann A.
        • Pickkers P.
        • Metra M.
        • Mebazaa A.
        • Dungen H.D.
        • Butler J.
        Adrenomedullin in heart failure: pathophysiology and therapeutic application.
        Eur J Heart Fail. 2019; 21: 163-171
        • Kremer D.
        • Ter Maaten J.M.
        • Voors A.A.
        Bio-adrenomedullin as a potential quick, reliable, and objective marker of congestion in heart failure.
        Eur J Heart Fail. 2018; 20: 1363-1365
        • Draker N.
        • Torry D.S.
        • Torry R.J.
        Placenta growth factor and sFlt-1 as biomarkers in ischemic heart disease and heart failure: a review.
        Biomark Med. 2019; 13: 785-799
        • Hammadah M.
        • Georgiopoulou V.V.
        • Kalogeropoulos A.P.
        • Weber M.
        • Wang X.
        • Samara M.A.
        • Wu Y.
        • Butler J.
        • Tang W.H.
        Elevated soluble fms-like tyrosine kinase-1 and placental-like growth factor levels are associated with development and mortality risk in heart failure.
        Circ Heart Fail. 2016; 9e002115
        • Molvin J.
        • Jujic A.
        • Bachus E.
        • Gallo W.
        • Tasevska-Dinevska G.
        • Holm H.
        • Melander O.
        • Fedorowski A.
        • Magnusson M.
        Cardiovascular biomarkers predict post-discharge re-hospitalization risk and mortality among Swedish heart failure patients.
        ESC Heart Fail. 2019; 6: 992-999
        • Dinarello C.A.
        The role of the interleukin-1-receptor antagonist in blocking inflammation mediated by interleukin-1.
        N Engl J Med. 2000; 343: 732-734
        • Zhu R.
        • Liu C.
        • Tang H.
        • Zeng Q.
        • Wang X.
        • Zhu Z.
        • Liu Y.
        • Mao X.
        • Zhong Y.
        Serum galectin-9 levels are associated with coronary artery disease in Chinese individuals.
        Mediat Inflamm. 2015; 2015457167
        • O'Brien M.J.
        • Shu Q.
        • Stinson W.A.
        • Tsou P.S.
        • Ruth J.H.
        • Isozaki T.
        • Campbell P.L.
        • Ohara R.A.
        • Koch A.E.
        • Fox D.A.
        • Amin M.A.
        A unique role for galectin-9 in angiogenesis and inflammatory arthritis.
        Arthritis Res Ther. 2018; 20: 31
        • de Couto G.
        • Jaghatspanyan E.
        • DeBerge M.
        • Liu W.
        • Luther K.
        • Wang Y.
        • Tang J.
        • Thorp E.B.
        • Marban E.
        Mechanism of enhanced MerTK-dependent macrophage efferocytosis by extracellular Vesicles.
        Arterioscler Thromb Vasc Biol. 2019; 39: 2082-2096
        • Hou X.
        • Chen G.
        • Bracamonte-Baran W.
        • Choi H.S.
        • Diny N.L.
        • Sung J.
        • Hughes D.
        • Won T.
        • Wood M.K.
        • Talor M.V.
        • Hackam D.J.
        • Klingel K.
        • Davogustto G.
        • Taegtmeyer H.
        • Coppens I.
        • Barin J.G.
        • Cihakova D.
        The cardiac microenvironment instructs divergent monocyte fates and functions in myocarditis.
        Cell Rep. 2019; 28: 172-189 e7
        • Yan L.
        • Wei X.
        • Tang Q.Z.
        • Feng J.
        • Zhang Y.
        • Liu C.
        • Bian Z.Y.
        • Zhang L.F.
        • Chen M.
        • Bai X.
        • Wang A.B.
        • Fassett J.
        • Chen Y.
        • He Y.W.
        • Yang Q.
        • Liu P.P.
        • Li H.
        Cardiac-specific mindin overexpression attenuates cardiac hypertrophy via blocking AKT/GSK3beta and TGF-beta1-SMAD signaling.
        Cardiovasc Res. 2011; 92: 85-94
        • Pitt B.
        • Pfeffer M.A.
        • Assmann S.F.
        • Boineau R.
        • Anand I.S.
        • Claggett B.
        • Clausell N.
        • Desai A.S.
        • Diaz R.
        • Fleg J.L.
        • Gordeev I.
        • Harty B.
        • Heitner J.F.
        • Kenwood C.T.
        • Lewis E.F.
        • O'Meara E.
        • Probstfield J.L.
        • Shaburishvili T.
        • Shah S.J.
        • Solomon S.D.
        • Sweitzer N.K.
        • Yang S.
        • McKinlay S.M.
        Spironolactone for heart failure with preserved ejection fraction.
        N Engl J Med. 2014; 370: 1383-1392
        • Pellicori P.
        • Ferreira J.P.
        • Mariottoni B.
        • Brunner-La Rocca H.P.
        • Ahmed F.Z.
        • Verdonschot J.
        • Collier T.
        • Cuthbert J.J.
        • Petutschnigg J.
        • Mujaj B.
        • Girerd N.
        • Gonzalez A.
        • Clark A.L.
        • Cosmi F.
        • Staessen J.A.
        • Heymans S.
        • Latini R.
        • Rossignol P.
        • Zannad F.
        • Cleland J.G.F.
        Effects of spironolactone on serum markers of fibrosis in people at high risk of developing heart failure: rationale, design and baseline characteristics of a proof-of-concept, randomised, precision-medicine, prevention trial. The Heart OMics in AGEing (HOMAGE) trial.
        Eur J Heart Fail. 2020; 22: 1711-1723https://doi.org/10.1002/ejhf.1716