You are here

  • Home
  • Archive
  • Volume 9, Issue 5
  • Levels of Lipoprotein (a) in patients with coronary artery disease with and without inflammatory rheumatic disease: a cross-sectional study

Article Text

Article menu

Download PDFPDF

XML
Cardiovascular medicine
Research
Levels of Lipoprotein (a) in patients with coronary artery disease with and without inflammatory rheumatic disease: a cross-sectional study
  1. Sverre Holm1,2,
  2. Ingvild Oma3,4,
  3. Tor-Arne Hagve4,5,
  4. Kjell Saatvedt6,
  5. Frank Brosstad1,4,
  6. Knut Mikkelsen2,
  7. Hans Rydningen7,
  8. Ivar Risnes7,
  9. Sven Martin Almdahl8,
  10. Thor Ueland1,4,
  11. Pål Aukrust1,4,9,
  12. Bente Halvorsen1,4,
  13. Ivana Hollan2,3,10,11
  1. 1 Research Institute of Internal Medicine, Oslo University Hospital, Oslo, Norway
  2. 2 Rheumatology Department, Hospital for Rheumatic Diseases, Lillehammer, Norway
  3. 3 Department of Pathology, Sykehuset Innlandet HF Divisjon Lillehammer, Lillehammer, Norway
  4. 4 Faculty of Medicine, Institute of Clinical Medicine, University of Oslo, Oslo, Norway
  5. 5 Division of Diagnostics and Technology, Akershus University Hospital, Lorenskog, Norway
  6. 6 Department of Thoracic Surgery, Oslo University Hospital Rikshospitalet, Oslo, Norway
  7. 7 Department of Cardiac Surgery, Feiring Heart Clinic, Feiring, Norway
  8. 8 Department of Cardiothoracic and Vascular Surgery, University Hospital of North Norway, Tromso, Norway
  9. 9 Section of Clinical Immunology and Infectious Diseases, Oslo University Hospital Rikshospitalet, Oslo, Norway
  10. 10 Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts, USA
  11. 11 Harvard Medical School, Harvard, Boston, USA
  1. Correspondence to Dr. Sverre Holm; sverre.holm{at}rr-research.no

Abstract

Objectives Patients with various inflammatory rheumatic diseases (IRDs) have increased risk of atherothrombotic disease. Lipoprotein (a) (Lp(a)) is a risk factor for atherosclerosis but its role in IRD with accompanying coronary artery disease (CAD) is still unclear. We aimed to examine if serum Lp(a) levels differed between CAD patients with and without accompanying IRD.

Design A cross-sectional observational, patient-based cohort study.

Setting Referred centre for coronary artery bypass grafting in the South Eastern part of Norway.

Participants 67 CAD patients with IRD (CAD/IRD) and 52 CAD patients without IRD (CAD/non-IRD). All patients were Caucasians, aged >18 years, without any clinically significant infection or malignancy.

Methods Lp(a) levels in serum were analysed by particle enhanced immunoturbidimetric assay, and Lp(a) levels were related to clinical and biochemical characteristics of the patient population.

Results We found no differences in serum levels of Lp(a) between CAD patients with and without IRD. In general, we found that Lp(a) correlated poorly with clinical and biochemical parameters including C reactive protein with the same pattern in the CAD/non-IRD and CAD/IRD groups.

Conclusions Our data do not support a link between inflammation and Lp(a) levels in CAD and in general Lp(a) levels were not correlated with other risk factors for cardiovascular disease.

  • coronary heart disease
  • rheumatology
  • atherosclerosis

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjopen-2019-030651

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • The study investigates a unique and well characterised group of patients.

    • We present measurement of Lipoprotein (a) in two distinct coronary artery disease populations, that is, patients with and without accompanying inflammatory rheumatic disease.

    • The main limitation of the study is the relatively small sample size.

    • Another limitation is that the design of the study does not allow for confirming of any cause–effect relationships.

    Introduction

    Patients with inflammatory rheumatic diseases (IRDs) have increased cardiovascular risk, primarily due to accelerated atherothrombosis.1–4 The reasons are not fully understood, but seem to involve chronic inflammation as a common mediator of both IRD and atherosclerosis.5 6 Thus, persistent chronic inflammation in IRD patients has been shown to contribute both to accelerated atherosclerosis7 and cardiovascular events.8 Besides chronic inflammation, traditional cardiovascular risk factors and genetic components are also implicated in the increased risk of cardiovascular disease (CVD) observed in patients with chronic IRD.9

    Lipoprotein (a)10 (Lp(a)) is composed of a low-density lipoprotein (LDL)-like particle bound with a cringle-structured glycoprotein, named apolipoprotein (apo) (a).11 12 Epidemiological studies have suggested that elevated levels of Lp(a) is a significant risk factor for atherosclerotic disorders including myocardial infarction (MI) and ischaemic stroke.12 13 In those with established coronary artery disease (CAD), high Lp(a) levels are associated with increased cardiovascular risk.10 12 Moreover, Lp(a) seems to be a marker of atherosclerotic disorders and to exert both proatherogenic and prothrombotic effects, some of which are primarily related to the LDL component whereas others are apo(a)-dependent.12 14

    Interestingly, besides genetic factors, also inflammation may influence Lp(a) levels. In support of this notion, interleukin 6 (IL-6) inhibition has been shown to down-regulate Lp(a) levels in patients with IRD.15 Moreover, we have previously demonstrated that methotrexate and tumour necrosis factor inhibition down-regulate Lp(a) in patients with rheumatoid arthritis (RA).16 In contrast, the majority of lipid-lowering strategies, except for lipid apheresis, have little influence on Lp(a) levels.17 18

    Lp(a) levels have been reported to be elevated in IRDs including RA and systemic lupus erythematosus (SLE).19 However, it is still unclear if the increased levels of Lp(a) in IRDs is modified by accompanying CAD. Our aim in this study was to investigate if Lp(a) levels differed between CAD patients with and without accompanying IRD. Our secondary aim was to examine if Lp(a) levels were associated with relevant clinical and laboratory variables in these patients.

    Materials and methods

    Study population

    From the Norwegian Feiring Heart Biopsy Study, described elsewhere,20 21 we examined CAD patients with IRD (CAD/IRD group; n=67) and CAD patients without IRD (CAD/non-IRD group; n=52). Briefly, all patients were Caucasians, aged >18 years, without any clinically significant infection or malignancy. The CAD groups were recruited among patients referred to coronary artery bypass grafting (CABG) due to CAD. The IRD diagnoses were confirmed according to accepted diagnostic criteria.21 The patients were consecutively enrolled in the study. The CAD/IRD and CAD/non-IRD groups were matched for age and sex at group level.

    The CAD/IRD group consisted of patients with RA (n=24), polymyalgia rheumatica (n=15), psoriatic arthritis (n=10), ankylosing spondylitis (n=6), giant cell arteritis (n=6), SLE (n=3), primary Sjögren’s syndrome (n=1), reactive arthritis (n=1) and undifferentiated connective tissue disease (n=1).

    Data collection

    The CAD groups were examined by interview, physical examination, self-reported questionnaires and blood tests within 2 days before CABG. A positive family history of CAD was defined as CAD in first-degree relatives at age <65 years and hypercholesterolaemia as a total serum cholesterol level >5.5 mmol/L registered in medical records, or use of lipid-lowering drugs.20

    Blood sampling protocol

    Venous blood samples were collected after a minimum of 4 hours fasting. The samples were collected in sterile containers with EDTA (plasma) or without additives (serum). The EDTA tubes were immediately immersed in melting ice, and centrifuged within 30 min (2000 g for 10 min). The serum tubes were centrifuged after coagulation in room temperature (<2 hours). Plasma and serum were stored at −75 °C in multiple aliquots, and sent to the respective laboratory on dry ice. Samples were not thawed prior to the laboratory analyses, and they were analysed in batches, in random order and in a blinded manner.

    Lp(a) assay

    Lp(a) levels in serum were analysed by particle enhanced immunoturbidimetric assay, LPALX Tina-quant Lp(a) (Latex) (Roche, Rotkreuz, Switzerland). The precipitate was determined turbidimetrically at 552 nm, using Cobasc 501 system from Roche.

    Miscellaneous

    Plasma concentrations of plasminogen activator inhibitor were analysed by enzyme immunoassay provided from R&D Systems (Minneapolis, Minnesota, USA). Serum levels of C-reactive protein (CRP) were determined using a particle-enhanced, high-sensitive immunoturbidimetric assay (Tina-Quant CRP Gen.3, Roche). Routine test standards of the hospital laboratory were used to analyse erythrocyte sedimentation rate, leucocytes, neutrophils, triglycerides, total cholesterol, high-density lipoprotein (HDL) cholesterol and LDL cholesterol and uric acid.

    Statistical analyses

    Distributions of categorical variables were compared using χ2 test. Comparisons of continuous variables in two groups were performed using Student’s t-test (normally distributed variables) or Mann-Whitney U-test (non-normally distributed variables). The p values (two-sided) were considered significant when<0.05.

    Patient and public involvement

    The outcomes measures were selected based on the current level of evidence, which aims to contribute to the provision of the missing pieces. The plan for the study was developed in accordance with suggestions given by representatives and members of patient organisations, in particular the Norwegian Rheumatism association during frequent meetings/lectures about CVD in IRD. The patients wished to clarify reasons for the increased CV risk in IRD, especially with the aim to identify factors that could be possible to modify, and therefore ameliorate CVD morbidity and mortality in these diseases. The patients themselves were not involved in recruitment and conduction of the study. As for other substudies from this biobank, also this study will be disseminated to IRD patients through lectures for patient organisations and/or through popular scientific information in magazines published by these organisations.

    Results

    Characteristics of the cohort

    Characteristics of the study population are shown in table 1. Except for the use of immunosuppressive drugs, the two CAD groups were matched for most of the parameters. However, the CAD/IRD group had more often a history of previous acute coronary syndrome (ACS) and higher CRP levels (table 1).

    View this table:
    Table 1

    Characteristics of the study population

    Serum LP(a) levels in CAD/IRD and CAD/non-IRD patients

    As illustrated in figure 1, serum levels of Lp(a) were similar, with no significant differences between the CAD/IRD and CAD/non-IRD groups. As ACS and CRP was not associated with Lp(a) (see below), no adjustments were made when comparing levels between CAD/IRD and CAD/non-IRD groups.

    Figure 1
    Figure 1

    Lp(a) levels in CAD patients with (n=67) and without IRD (n=52). In addition to individual values, the figure shows median and 25th and 75th percentile levels. The comparison between the groups was performed by the Mann-Whitney U-test. CAD, coronary artery disease; IRD, inflammatory rheumatic disease; Lp(a), lipoprotein (a).

    Associations between Lp(a) and relevant clinical and demographic variables

    In general, Lp(a) correlated poorly with clinical and biochemical parameters in the study population as a whole, with the same pattern in CAD/IRD and CAD-non-IRD patients. This included correlations with age, gender and body mass index, medical history, biochemical parameters and medications, including CRP and a history of ACS that differed between the two CAD groups (table 2). However, Lp(a) correlated significantly with LDL-cholesterol in the study group as a whole, mainly reflecting a positive correlation in the CAD/non-IRD group (table 2). Moreover, whereas HDL cholesterol correlated positively with Lp(a) in the CAD/non-IRD group, Lp(a) was inversely correlated with HDL cholesterol in the CAD/IRD group (table 2).

    View this table:
    Table 2

    Correlations between serum levels of Lp(a) and different clinical and biochemical characteristics of the study population

    Discussion

    In the present study we found no differences in serum levels of Lp(a) between CAD patients with and without IRD. Moreover, we found that Lp(a) correlated poorly with clinical and biochemical parameters including CRP with the same pattern in the CAD/non-IRD and CAD/IRD groups. These data do not support the notion that Lp(a) levels are driven by inflammation, as reflected by CRP levels, in these populations.

    Several studies have shown that elevated serum Lp(a) levels are associated with a higher risk for developing CAD, and in those with established CAD, high levels of Lp(a) are associated with increased risk for cardiovascular events.10 12 13 There are also some studies showing increased serum levels of Lp(a) in IRD patients, potentially associated with increased risk for developing atherosclerotic disease.19 These latter studies may indicate an interaction between Lp(a) and inflammation. However, herein we found no differences in Lp(a) levels between CAD/non-IRD and CAD/IRD patients, although the latter group had significantly higher CRP levels. Indeed, we found no correlation between Lp(a) and CRP with the same pattern in the two CAD groups. Moreover, whereas therapeutic intervention targeting IL-6 by the IL-6 receptor inhibitor tocilizumab has been shown to down-regulate Lp(a) levels in patients with IRD,22–24 we recently observed no effect of tocilizumab treatment on Lp(a) levels in patients with Non-ST-elevation MI.25 Thus, at present the link between inflammation and Lp(a) levels is questionable, and our findings herein do not support the ‘inflammatory-driven’ Lp(a) hypothesis.

    There are some reports of a positive correlation between Lp(a) and HDL cholesterol,26 and in the present study this was also seen in the CAD/non-IRD group. In contrast, an opposite association was seen in the CAD/IRD, with a significant inverse correlation between Lp(a) and HDL cholesterol. At present we have no explanation for these different associations between HDL cholesterol and Lp(a) in the two CAD groups, and it could be by chance owing to a large number of correlation analyses. However, future studies exploring this pattern in larger study populations could potentially contribute to our understanding of the complex regulation of Lp(a) in CVD.

    Although some correlation with LDL and HDL cholesterol, Lp(a) levels did not correlated with other risk factor for CAD. More recent studies have also noted that Lp(a) levels correlated poorly with other risk factors for CVD.27 28 This does, however, not exclude that Lp(a) could contribute to development and progression of atherosclerotic disorders. Thus, Lp(a) has been suggested to promote atherogenesis and plaque instability by inhibiting angiogenesis, enhancing endothelial cell activation and thrombogenesis through inhibiting fibrinolysis and tissue factor inhibitor pathway and inducing platelet activation.10 12 14 Lp(a) has also been shown to contribute to foam cell formation, and through binding to oxidised phospholipids it may cause plaque instability.12 If these mechanisms are operating in the same degree in CAD patients with and without accompanying IRD has to this end not been studied.

    This study has some potential limitations. First, due to a relatively small sample size, the apparent lack of differences and associations might be caused by type-II error and our data should be interpreted with some cautions. Second, the design of the study does not allow for confirming of any cause-effect relationships.

    In conclusion, we found no differences in Lp(a) levels between CAD/IRD and CAD/non-IRD patients, and Lp(a) levels correlated poorly with other risk factors for CAD progression including CRP. Our data do not support a link between inflammation and Lp(a) levels in CVD.

    Acknowledgments

    We will thank the patient advisers from Norwegian Rheumatism association as well as Norwegian Women’s Public Health Association and People with Heart and Lung Diseases for their support.

    References

    1. 1.↵
      2. Mackey RH ,
      3. Kuller LH ,
      4. Moreland LW
      . Cardiovascular disease risk in patients with rheumatic diseases. Clin Geriatr Med 2017;33:105–17.doi:10.1016/j.cger.2016.08.008
      OpenUrl
    2. 2.↵
      2. Mellana WM ,
      3. Aronow WS ,
      4. Palaniswamy C , et al
      . Rheumatoid arthritis: cardiovascular manifestations, pathogenesis, and therapy. Curr Pharm Des 2012;18:1450–6.doi:10.2174/138161212799504795
      OpenUrlCrossRefPubMed
    3. 3.↵
      2. Nurmohamed MT ,
      3. Heslinga M ,
      4. Kitas GD
      . Cardiovascular comorbidity in rheumatic diseases. Nat Rev Rheumatol 2015;11:693–704.doi:10.1038/nrrheum.2015.112
      OpenUrlCrossRefPubMed
    4. 4.↵
      2. Skeoch S ,
      3. Bruce IN
      . Atherosclerosis in rheumatoid arthritis: is it all about inflammation? Nat Rev Rheumatol 2015;11:390–400.doi:10.1038/nrrheum.2015.40
      OpenUrlCrossRefPubMed
    5. 5.↵
      2. Breland UM ,
      3. Hollan I ,
      4. Saatvedt K , et al
      . Inflammatory markers in patients with coronary artery disease with and without inflammatory rheumatic disease. Rheumatology 2010;49:1118–27.doi:10.1093/rheumatology/keq005
      OpenUrlCrossRefPubMed
    6. 6.↵
      2. Hartman J ,
      3. Frishman WH
      . Inflammation and atherosclerosis: a review of the role of interleukin-6 in the development of atherosclerosis and the potential for targeted drug therapy. Cardiol Rev 2014;22:147–51.doi:10.1097/CRD.0000000000000021
      OpenUrlCrossRefPubMed
    7. 7.↵
      2. Gonzalez-Gay MA ,
      3. Gonzalez-Juanatey C ,
      4. Piñeiro A , et al
      . High-grade C-reactive protein elevation correlates with accelerated atherogenesis in patients with rheumatoid arthritis. J Rheumatol 2005;32:1219–23.
      OpenUrlAbstract/FREE Full Text
    8. 8.↵
      2. Gonzalez-Gay MA ,
      3. Gonzalez-Juanatey C ,
      4. Lopez-Diaz MJ , et al
      . HLA-DRB1 and persistent chronic inflammation contribute to cardiovascular events and cardiovascular mortality in patients with rheumatoid arthritis. Arthritis Rheum 2007;57:125–32.doi:10.1002/art.22482
      OpenUrlCrossRefPubMedWeb of Science
    9. 9.↵
      2. López-Mejías R ,
      3. Castañeda S ,
      4. González-Juanatey C , et al
      . Cardiovascular risk assessment in patients with rheumatoid arthritis: The relevance of clinical, genetic and serological markers. Autoimmun Rev 2016;15:1013–30.doi:10.1016/j.autrev.2016.07.026
      OpenUrl
    10. 10.↵
      2. Niccoli G ,
      3. Cin D ,
      4. Scalone G , et al
      . Lipoprotein (a) is related to coronary atherosclerotic burden and a vulnerable plaque phenotype in angiographically obstructive coronary artery disease. Atherosclerosis 2016;246:214–20.doi:10.1016/j.atherosclerosis.2016.01.020
      OpenUrl
    11. 11.↵
      2. Manocha A ,
      3. Srivastava LM
      . Lipoprotein (a): a Unique Independent Risk Factor for Coronary Artery Disease. Indian J Clin Biochem 2016;31:13–20.doi:10.1007/s12291-015-0483-3
      OpenUrl
    12. 12.↵
      2. Nordestgaard BG ,
      3. Langsted A
      . Lipoprotein (a) as a cause of cardiovascular disease: insights from epidemiology, genetics, and biology. J Lipid Res 2016;57:1953–75.doi:10.1194/jlr.R071233
      OpenUrlAbstract/FREE Full Text
    13. 13.↵
      2. Gurdasani D ,
      3. Sjouke B ,
      4. Tsimikas S , et al
      . Lipoprotein(a) and risk of coronary, cerebrovascular, and peripheral artery disease: the EPIC-Norfolk prospective population study. Arterioscler Thromb Vasc Biol 2012;32:3058–65.doi:10.1161/ATVBAHA.112.255521
      OpenUrlAbstract/FREE Full Text
    14. 14.↵
      2. Boffa MB ,
      3. Koschinsky ML
      . Lipoprotein (a): truly a direct prothrombotic factor in cardiovascular disease? J Lipid Res 2016;57:745–57.doi:10.1194/jlr.R060582
      OpenUrlAbstract/FREE Full Text
    15. 15.↵
      2. Schultz O ,
      3. Oberhauser F ,
      4. Saech J , et al
      . Effects of inhibition of interleukin-6 signalling on insulin sensitivity and lipoprotein (a) levels in human subjects with rheumatoid diseases. PLoS One 2010;5:e14328.doi:10.1371/journal.pone.0014328
    16. 16.↵
      2. Hjeltnes G ,
      3. Hollan I ,
      4. Førre O , et al
      . Serum levels of lipoprotein(a) and E-selectin are reduced in rheumatoid arthritis patients treated with methotrexate or methotrexate in combination with TNF-α-inhibitor. Clin Exp Rheumatol 2013;31:415–21.
      OpenUrlPubMed
    17. 17.↵
      2. Boffa MB
      . Emerging Therapeutic Options for Lowering of Lipoprotein(a): Implications for Prevention of Cardiovascular Disease. Curr Atheroscler Rep 2016;18:69.doi:10.1007/s11883-016-0622-1
      OpenUrl
    18. 18.↵
      2. van Capelleveen JC ,
      3. van der Valk FM ,
      4. Stroes ES
      . Current therapies for lowering lipoprotein (a). J Lipid Res 2016;57:1612–8.doi:10.1194/jlr.R053066
      OpenUrlAbstract/FREE Full Text
    19. 19.↵
      2. Missala I ,
      3. Kassner U ,
      4. Steinhagen-Thiessen E
      . A Systematic Literature Review of the Association of Lipoprotein(a) and Autoimmune Diseases and Atherosclerosis. Int J Rheumatol 2012;2012:1–10.doi:10.1155/2012/480784
      OpenUrl
    20. 20.↵
      2. Hollan I ,
      3. Bottazzi B ,
      4. Cuccovillo I , et al
      . Increased levels of serum pentraxin 3, a novel cardiovascular biomarker, in patients with inflammatory rheumatic disease. Arthritis Care Res 2010;62:378–85.doi:10.1002/acr.20094
      OpenUrl
    21. 21.↵
      2. Hollan I ,
      3. Scott H ,
      4. Saatvedt K , et al
      . Inflammatory rheumatic disease and smoking are predictors of aortic inflammation: a controlled study of biopsy specimens obtained at coronary artery surgery. Arthritis Rheum 2007;56:2072–9.doi:10.1002/art.22690
      OpenUrlCrossRefPubMedWeb of Science
    22. 22.↵
      2. Gabay C ,
      3. McInnes IB ,
      4. Kavanaugh A , et al
      . Comparison of lipid and lipid-associated cardiovascular risk marker changes after treatment with tocilizumab or adalimumab in patients with rheumatoid arthritis. Ann Rheum Dis 2016;75:1806–12.doi:10.1136/annrheumdis-2015-207872
      OpenUrlAbstract/FREE Full Text
    23. 23.↵
      2. Müller N ,
      3. Schulte DM ,
      4. Türk K , et al
      . IL-6 blockade by monoclonal antibodies inhibits apolipoprotein (a) expression and lipoprotein (a) synthesis in humans. J Lipid Res 2015;56:1034–42.doi:10.1194/jlr.P052209
      OpenUrlAbstract/FREE Full Text
    24. 24.↵
      2. García-Gómez C ,
      3. Martín-Martínez MA ,
      4. Castañeda S , et al
      . Lipoprotein(a) concentrations in rheumatoid arthritis on biologic therapy: Results from the CARdiovascular in rheuMAtology study project. J Clin Lipidol 2017;11:749–56.doi:10.1016/j.jacl.2017.02.018
      OpenUrl
    25. 25.↵
      2. Ueland T , et al
      . Serum lipoprotein(a) is not modified by interleukin-6 receptor antagonism or associated with inflammation in non-ST-elevation myocardial infarction. Int J Cardiol 2018.
    26. 26.↵
      2. Sharma S ,
      3. Merchant J ,
      4. Fleming SE
      . Lp(a)-cholesterol is associated with HDL-cholesterol in overweight and obese African American children and is not an independent risk factor for CVD. Cardiovasc Diabetol 2012;11:10.doi:10.1186/1475-2840-11-10
      OpenUrlCrossRefPubMed
    27. 27.↵
      2. Afshar M ,
      3. Pilote L ,
      4. Dufresne L , et al
      . Lipoprotein(a) Interactions With Low-Density Lipoprotein Cholesterol and Other Cardiovascular Risk Factors in Premature Acute Coronary Syndrome (ACS). J Am Heart Assoc 2016;5.doi:10.1161/JAHA.115.003012
    28. 28.↵
      2. Puri R ,
      3. Ballantyne CM ,
      4. Hoogeveen RC , et al
      . Lipoprotein(a) and coronary atheroma progression rates during long-term high-intensity statin therapy: Insights from SATURN. Atherosclerosis 2017;263:137–44.doi:10.1016/j.atherosclerosis.2017.06.026
      OpenUrl

    Footnotes

    • Contributors SH, IO, BH and IH conceived and design research; T-AH and FB performed biochemical analyses; IO, KS, KM, HR, IR, SMA and IH recruited patients and collected samples; SH, IO, TU and IH performed statistical analyses; SH, PA and IH wrote the paper; SH, PA, BH and IH had the responsibility for the final content. All authors read, critically revised and approved the final manuscript.

    • Funding This work was supported by grants from the South-Eastern Norway Regional Authority. We will also thank the Norwegian Rheumatism association as well as Norwegian Women’s Public Health Association and People with Heart and Lung Diseases for financial support.

    • Competing interests None declared.

    • Ethics approval The Regional Ethics Committee for Medical Research approved the study protocol.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement All data presented in this study is available from the corresponding author upon reasonable request.

    • Patient consent for publication Not required.