Running head: Incidence and prevalence of MACE in RA, PsA and AxSpA

Title: Incidence and prevalence of major adverse cardiovascular events in rheumatoid

arthritis, psoriatic arthritis, and axial spondyloarthritis

1Kim Lauper, MD, Delphine S. Courvoisier, PhD, Paola Chevallier, MD, Axel Finckh, MD, MSc,

Cem Gabay, MD, Division of Rheumatology, Geneva University Hospitals, Geneva,

Switzerland.

1Division of Rheumatology, Geneva University Hospitals, Geneva, Switzerland

Keywords: Rheumatoid arthritis; psoriatic arthritis; axial spondyloarthritis; cardiovascular

risk

Word count: 3245

Funding information: This study was supported by an unrestricted grant by Abbvie (Abott

Rheumatology Grant 2011). Swiss Clinical Quality Management in Rheumatic Diseases

(SCQM) database is sponsored by public and industrial support

(http://scqm.ch/en/sponsoren/).

Correspondence:

Cem Gabay, MD

Division of Rheumatology, Geneva University Hospitals, 26 Avenue Beau-Séjour, 1206

Geneva, Switzerland. cem.gabay@hcuge.ch

Disclosure statement: CG has received research fees from Roche, Pfizer, AB2 Bio, and

consultant and speaker's fee from Roche, Pfizer, Abbvie, MSD, BMS, Sanofi-Regeneron,

Novartis, AB2 Bio. DC received consulting fees from BMS, Pfizer, and Janssen. AF has

received consultancies or honoraria from Abbvie, AB2 Bio, BMS, MSD, Pfizer, Roche, and

UCB. KL and PC have nothing to disclose.

Funding sources had no role in the design of this study, its execution, analyses,

interpretation of the data, or decision to submit results.

2

Abstract

Objective

The primary objective of this study was to compare the risk of major cardiovascular events

(MACE) in a large observational cohort of rheumatoid arthritis (RA), psoriatic arthritis (PsA)

and axial spondyloarthritis (AxSpA) patients

Methods

We conducted a mixed retrospective and prospective cohort study using data from patients

with RA, PsA and axSpA included in the Swiss Clinical Quality Management registry. The

primary outcome of interest was a composite of myocardial infarction, transient or

permanent cerebrovascular event, or CV associated death.

Results

5,315 patients were eligible for the analysis of incidence with a total follow-up time of

37,495 patient years for RA, 19,837 for AxSpA, and 9,171 for PsA. The unadjusted incidence

rate of MACE per 1000 person-years was 2.67 for RA, 1.41 for AxSpA, and 1.42 for PsA.

Compared to RA patients, unadjusted incidence rate ratio (IRR) were 0.53 for AxSpA (95%

confidence interval (CI) 0.34-0.80, p=0.003) and 0.53 for PsA (95% CI 0.30-0.95, p=0.03).

After adjustment with traditional CV risk factors, age at disease onset, gender, and disease

duration, the difference was not statistically significant between RA and AxSpA (adjusted

IRR: 0.93, 95% CI: 0.51-1.69, p=0.80) nor between RA and PsA (adjusted IRR: 0.56, 95% CI:

0.27-1.14, p = 0.11). We found a similar result with the analysis of prevalence.

Conclusion

3

There was no significant difference in the incidence and prevalence of MACE between RA

and AxSpa or PsA, suggesting that inflammation drives the increased risk of CV disease

rather than a particular disease.

4

Significance and Innovations

Incidence and prevalence of major cardiovascular adverse events were not

statistically significantly different between rheumatoid arthritis, psoriatic arthritis and

axial spondyloarthritis.

We should improve the management of cardiovascular risk in all these inflammatory

rheumatic diseases.

5

Introduction

Rheumatoid arthritis (RA) is a chronic inflammatory disease, leading to joint deformity,

disability and an excess mortality, largely attributable to cardiovascular and cerebrovascular

disease (1,2). This increased cardiovascular (CV) risk is not explained solely by an increase in

traditional CV risk factors, thus suggesting that RA is an independent risk factor (3–5).

Similarly, psoriatic arthritis (PsA) and ankylosing spondylitis (AS), seem to confer also an

increased risk of cardiovascular diseases (CVD) (6–8). The increase in CV risk has been

evaluated in a recent meta-analysis on PsA to be 43% for cardiovascular diseases and 22%

for cerebrovascular diseases, independently from traditional cardiovascular risk factors (9).

In AS, the relative risk of acute coronary syndrome compared to the general population

varies from 1.4 (95% IC 1.2-1.4) to 3.0 (95% IC 1.3-6.7), depending on the study (10). Schier

et al. estimated the relative risk of myocardial infarction in RA compared to the general

population at 1.52 (95% IC 1.37-1.69) in their meta-analysis (11).

Inflammation has been shown to play an important role in the pathogenesis of

atherosclerosis, and systemic markers of inflammation predict cardiovascular disease in the

general population (12). RA and atherosclerosis share similar processes in their

pathogenesis, with implication of immune cells, adhesion molecules, endothelin and pro-

inflammatory cytokines, suggesting that the systemic inflammation present in RA accelerates

atherogenesis (13,14). Disease modifying antirheumatic drugs (DMARDs) such as

methotrexate (MTX) or tumor necrosis factor (TNF) inhibitors decrease the risk of CV

morbidity and mortality in RA and psoriasis, supporting the role of systemic inflammation in

atherogenesis (15–18).

6

It is not known if this increased CV risk is similar between various rheumatic diseases. In

particular, the prevalence and incidence of major adverse cardiovascular events (MACE)

between RA, PsA and axial spondyloarthritis (AxSpA) have not been directly compared in a

large population cohort so far. The purpose of this study was to compare the prevalence and

incidence of MACE between these inflammatory diseases after adjusting for traditional CV

risk factors in a Swiss cohort of patients. The secondary objective was to find predictors of

the incidence of MACE in the RA population, since this was the largest group of patients with

available data.

Patients and Methods

Study design

We conducted a mixed retrospective and prospective cohort study using data from patients

with RA, PsA and axSpA included in the Swiss Clinical Quality Management (SCQM) registry.

The SCQM registry is a longitudinal cohort of patients, which has been described in detail

elsewhere (19). Ethical approval for the SCQM and related studies was obtained from the

Swiss Academy of Medical Sciences review board, and all the patients have provided written

consent prior to enrolment. At inclusion, demographic and disease characteristics,

concomitant treatments, and co-morbidities are collected by the treating rheumatologists

and patients fill-out self-administered questionnaires such as health assessment

questionnaire disability index (HAQ-DI). Follow-up assessments are performed at regular

intervals, approximately one to four times a year (disease activity, antirheumatic treatments,

side effects, reasons of discontinuation, co-morbidities, etc.). The Swiss Society of

Rheumatology recommends the inclusion of all the patients treated with biologic DMARDs

7

(bDMARDs). Patients in SCQM come from diverse clinical settings, with more than 50% from

private practice, 30 % from non-academic centres and 20 % from academic centres.

For this study, we added a specific self-reported questionnaire asking the patients for CV

events and associated risk factors, encompassing the period before the inclusion in the

register, to the existing longitudinal data collection. When available we also included in the

analysis the data regarding CV risks factors that were captured prospectively in the SCQM

registry by the treating rheumatologists. For patients lost of follow-up (no follow-up visit ≥

18 months) as well as in cases with incomplete or inconsistent responses, we asked the

treating rheumatologists for further information or confirmation of patient-reported data. In

patients lost of follow-up, we enquired if the patients had died and if so, whether the cause

of death was related to a CV event. Because this procedure required de-anonymization of

part of the data in order to contact the treating rheumatologist, we obtained a separate

approval for this study by the Ethics Committee of the Geneva University Hospitals.

For the analysis of incidence rates, inclusion criteria were a diagnostic of RA, PsA, or axSpA

established by the treating rheumatologist, age ≥ 18 years at disease onset, and at least at

one visit after disease onset and between March 2012 (start of collection of patient

cardiovascular risk questionnaire) and March 2015 (end of the study and locking of the

database). Exclusion criteria were a history of non-lethal MACE at time of disease onset and

the non-completion of the cardiovascular questionnaire.

For the analysis of prevalence, inclusion criteria were a diagnostic of RA, PsA, or axSpA

established by the treating rheumatologist and age ≥ 18 years at disease onset and at least

one answer to the questionnaire between March 2012 and March 2015.

8

For the secondary objective of analyzing factors associated with the incidence of

cardiovascular diseases in RA, we restricted our analysis to the subgroup of RA patients who

experienced a MACE after their inclusion in the SCQM registry and were free of MACE before

inclusion in SCQM.

Outcome of interest

MACE was defined as myocardial infarction, transient or permanent cerebrovascular event,

or CV-associated death. For the calculation of an incidence rate of MACE, we recorded the

year of occurrence from time of disease onset until the last visit before the end of study. The

exposure time was defined as the time in years between onset of the rheumatic disease until

either the first MACE or the end of follow-up, whichever occurred first. For patients who

died from a CV event, we recorded the year at the last visit. For the prevalence, we

considered all visits in the study period with complete information regarding non-lethal

myocardial infarction and non-lethal stroke, to derive presence of a history of non-lethal

MACE.

Exposure of interest and covariates

Of primary interest was the type of inflammatory rheumatic disease: RA, PsA, or axSpA. In

addition to the year of disease onset, sex and age at disease onset, and disease duration, we

assessed the following other traditional CV risk factors: known family history of early MACE,

as defined by a MACE occurring before the age of 50 in a first-degree relative, ever-

hypertension, ever-diabetes, ever-hyperlipidemia (hypercholesterolemia was not specifically

captured in the questionnaire), and ever-smoking. All available information as provided by

the patient until end of follow-up was considered and the value set to „yes“ in case the

patient indicated presence of such a condition at least once.

9

Other variables of interest

For the RA patients, we assessed the medication history with respect to bDMARDs and use

of MTX, glucocorticoids (GC), and nonsteroidal anti-inflammatory drugs (NSAIDs). We also

assessed seropositivity for rheumatoid factor and/or anti-citrullinated protein antibodies,

Health Assessment Questionnaire Disability Index (HAQ-DI) and Disease Activity Score

(DAS)28-CRP at the time of inclusion in the SCQM registry.

Statistical analyses

We compared covariates between diseases using the Kruskal-Wallis test for continuous or

discrete variables and the Fisher’s exact test for categorical variables. To estimate and

compare incidence rates of first MACE between diseases, and for the analysis of predictors

of MACE in RA, we applied Poisson regression with offsets corresponding to the patients’

time under observation for the first MACE. For the analysis of prevalence, we applied logistic

regression to estimate and compare the odds for a history of non-lethal MACE between

diseases. Due to the relatively small number of MACE, we restricted the number of variables

to include in the multivariable models (20). We included all cardiovascular risk factors and

disease duration to ensure the best possible adjustment for confounding. However, other

characteristics were only added to the multiple adjusted model if their p-values were <0.1 in

the univariable model (21). A significance level of 0.05 was used and no correction for

multiple testing was applied, as all the analysis performed were pre-specified. For missing

values of covariates, we performed a multiple imputation analysis using chained equations.

All analyses were performed using R v.3.3.2 and the mice package.

Results

10

5,315 patients were eligible for the analysis of incidence (Figure 1) and contributed a total

follow-up time of 37’495 patient years for RA, 19,837 for AxSpA, and 9,171 for PsA.

Rheumatic diseases started between 1950 and 2014 and differed significantly with respect to

assessed CV risk factors and disease duration (Table 1). A total of 100 RA, 28 AxSpA, and 13

PsA patients experienced at least one incident MACE since disease onset. The unadjusted

incidence rate of MACE per 1000 person-years was 2.67 for RA, 1.41 for AxSpA, and 1.42 for

PsA (p = 0.005). Compared to RA patients, unadjusted incidence rate ratio (IRR) were 0.53 for

AxSpA (p=0.003) and 0.53 for PsA (p=0.03). IRR adjusted for gender were 0.34 for AxSpA

(p<0.001) and 0.37 for PsA (p=0.001), whereas IRR adjusted for age were 1.16 for AxSpA

(p=0.52) and 0.75 for PsA (p=0.34). The age-adjusted IRR strongly swayed the IRR toward 1

(the null), suggesting that age is an important confounder when comparing incidence rates

across rheumatic diseases.

When we further adjusted for traditional CV risk factors, age at disease onset, gender, and

disease duration, the difference between RA and AxSpA was no longer statistically significant

(adjusted IRR: 0.93, 95% confidence interval (CI): 0.51-1.69, p=0.80), nor between RA and

PsA (adjusted IRR: 0.56, 95% CI: 0.27-1.14, p=0.11). Older age at disease onset, male gender,

known familial history, ever-hypertension, ever-hyperlipidemia, and longer disease duration

were significantly positively associated with the risk of incident MACE (Table 2).

In the analysis of prevalent MACE, we included 3123 RA, 1,468 AxSpA and 792 PsA. The

unadjusted prevalence of non-lethal MACE was 4.89% for RA, 2.24% for AxSpA and 2.90% for

PsA (p < 0.001). After adjusting for the traditional CV risk factors, age at disease onset,

gender, and disease duration, the difference was no longer significant between RA and

AxSpA (prevalence ratio (PR) 0.99, 95% CI 0.60-1.58) and between RA and PsA (PR 0.68, 95%

11

CI 0.39-1.10). As for the incident MACE, the most important confounder was age, and

adjusting only for this variable removed the statistical significant differences in prevalence of

MACE between diseases (Table 3).

For the predictors of the incidence of MACE in RA, 3019 patients were eligible for analysis

with 49 events. In univariable analysis, age, male gender, ever-smoking, ever-hypertension,

ever-hyperlipidemia and HAQ-DI were statistically significantly associated with the

occurrence of MACE, whereas seropositivity and use of NSAIDs were negatively associated

(Table 4). When adjusting for age and gender only, seropositivity and NSAIDs were no longer

associated with the occurrence of MACE. In multivariable analysis, age, male gender, ever-

smoking, ever-hyperlipidemia and disease duration were significantly associated with the

occurrence of MACE. Ever-hypertension, ever-diabetes, and HAQ-DI at inclusion were not

associated with the occurrence of MACE.

Discussion

In this study, we did not find an independent association between the type of rheumatic

disease and the incidence and prevalence of MACE. The apparent increase in MACE in RA as

compared to PsA and AxSpA was largely attributable to difference in population

characteristics, in particular owing to older age. In the RA population, age, gender, ever-

smoking, ever-hyperlipidemia and disease duration were independent predictors of the

incidence of MACE.

A study from Jamnitski et al. directly compared the incidence of MACE between RA and PsA

patients. The study was limited to 489 PsA and 353 RA patients aged between 50 and 75

years, who answered a mailed questionnaire (22). There was no difference in the odds of

MACE between RA and PsA after stratification by age and sex. In a meta-regression with

12

adjustment for age and sex only, there was also no difference between RA and PsA in the

risk of myocardial infarction (11). We only found one cross-sectional study reporting

significantly lower odds of CVD in patients with PsA (n=1,147) compared to RA patients

(n=2,152), after adjustments for demographic variables, traditional CV risk factors and

educational level [OR 0.46 (0.23–0.92), p=0.028) (23). In a recent study linking national and

population-based registers, the age and sex-adjusted relative risks for acute coronary

syndrome compared with the general population was 1.3 (95% CI 1.0 to 1.7) for AS and 1.7

(1.4 to 2.0) for RA. For stroke, the relative risks were 1.5 (1.1 to 2.0) in AS and 1.5 (1.2 to 1.8)

in RA, compared with the general population. The important overlap between the relative

risks suggests that the CV risk is comparable in RA and AS (24). We found one study that

directly compared AS and RA, with a lower hazard ratio for myocardial infarction of 0.8 (95%

IC 0.7-0.9) in AS as compared to RA, but no significant difference in the hazard of stroke (10).

Inflammation is likely to play an important role in the pathogenesis of atherosclerosis, which

may explain a similar risk of MACE in RA, PsA and AxSpA. Endothelial activation, infiltration

of inflammatory cells in the blood vessel walls, secretion of chemokines and cytokines and

formation of a plaque by macrophages turned into foam cells are present in atherosclerosis

(13,25). The pathogenesis of RA is very similar to that of atherosclerosis with endothelial cell

activation, recruitment of inflammatory cells, and cytokines release (26). In particular, TNF

may accelerate atherosclerosis in RA (27) and TNF is also implicated in the pathogenesis of

AxSpA and PsA (28–30). In fact, myocardial dysfunction and early signs of atherosclerosis

have been shown in patients with AxSpA and AS (31,32). In addition, the IL-17/IL23 axis that

exerts an important role in the pathogenesis of AxSpA and PsA (33,34), has also been shown

to contribute to endothelial dysfunction and arterial hypertension (35). Recently the

CANTOS trial has shown that anti-inflammatory therapy targeting IL-1β may lower the risk of

13

a new CV event in patients with a history of myocardial infarction and blood level of C-

reactive protein ≥ 2 mg/L despite the use of aggressive secondary prevention strategies (36).

RA disease duration has already been associated to an increase in CV risk. However, this

association is not linear, with an increased risk already present in the first years of the

disease, which remains unchanged thereafter (37–39). The use of NSAIDs was not a

predictor of MACE in our population. Similarly, a recent Taiwanese nation-wide study did not

find an increase in coronary disease in patients taking etoricoxib or celecoxib after

adjustment for gender, age, comorbidities, hypertension, hyperlipidemia, and DMARDs (40).

Finding similar results in this study is reassuring, but a confounding by indication cannot be

excluded, for example if patients less at risk for MACE are taking more NSAIDs.

Our study has several limitations. We may have a selection bias in our registry, which can

have led to either an increased or a decreased estimate in the incidence and prevalence of

MACE. Indeed, patients included in the SCQM are more likely to be treated with bDMARDs

and thus, comprise a group of more severe patients. However, these patients may also have

benefited from better treatments and follow-up, and may also be more aware of their health

and prone to follow health advices due to more regular disease assessments (41,42). We can

also not exclude a survival bias, with patients with less severe rheumatologic or CV

conditions included in our register.

The sample of patients for PsA is small and thus the absence of statistically significant

difference with RA in the incidence and prevalence of MACE may be due to limited statistical

power.

We obtained information on MACE using self-reporting by patients. Questionnaires are often

used in epidemiological studies, in the absence of available registries. Validation studies

14

estimates that the sensitivity of self reported myocardial infarction and stroke is moderate

to high for both conditions (43,44), To minimize this potential reporting bias, we ascertained

the results with a physician report.

To investigate the effect of rheumatic diseases on MACE, we chose to exclude patients with

a history of non-lethal MACE before the onset of RA, PsA and AxSpa. RA patients are older,

and thus at higher risk for MACE. Maybe for this reason, slightly more patients have been

excluded for previous MACE in the RA population (1.7%) than in AxSpA (0.3%) or PsA (1.3%).

By excluding patients with a previous history MACE, we may have differentially selected

patients at lower risk, thus decreasing the difference between RA and the two other

rheumatic diseases. Of note, according to one study, the risk of ischemic heart disease does

not seem to be increased in RA before the onset of the rheumatic disease, suggesting that

MACE occurring before the onset of RA is independent of the rheumatic disease (38).

However, literature is conflicting on this point with another study finding a higher risk of

coronary heart disease preceding the ACR criteria based diagnosis of RA (45).

We could not compare the incidence of MACE of our study population with the general

Swiss population, as the data are not reported in the same categories (46,47).

We assessed the effect of concomitant medications such as NSAIDS, biological DMARDs and

corticosteroids only in the prospective RA cohort, because treatments prescribed before the

inclusion in the register were not accurately reported. We also did not adjust for body mass

index (BMI), HAQ, or disease activity in patients who were not included in the prospective

cohort for the same reason.

15

Patients were assessed during 10 years on average. In RA, the risk of MACE increased rapidly

in the first 4 years of the disease, but we cannot exclude that a longer latency is necessary to

see a difference between the diseases (48).

Finally, a substantial proportion of MACE occurred before the patient’s inclusion in the

register, which did not allow their inclusion in the analysis of predictors of the incidence of

MACE in the RA cohort. Limited statistical power could explain why we did not see an effect

of disease activity and NSAID use on MACE, unlike other studies (18,49,50). Accordingly, we

could also not use all the pre-specified variables in the multivariable analysis, to avoid

overadjustment.

The strengths of our study are that the sample is relatively large, with a long duration of

follow-up. The data were collected prospectively and included a wide range of variables,

including parameters of function and disease activity in the prospective RA cohort.

In conclusion, the incidence and prevalence in MACE in RA was not statistically significantly

different to those of AxSpA and PsA after adjustment for confounding factors. However,

considering the small sample of patients with PsA, it is difficult to reach a definite conclusion

due to limited statistical power. Our findings suggest that inflammation drives the increased

risk of CV disease rather than a particular disease. In addition, traditional CV risk factors,

such as smoking, hypertension and hyperlipidemia were significantly associated with MACE,

stressing the importance of the detection and management of traditional CV risk factors in

patients with inflammatory arthritis. Accordingly, the recent recommendations of the EULAR

insist on the necessity of clinicians to be aware of the higher risk of CV events in RA, and

probably in PsA and AxSpA, and suggest that the rheumatologist should ensure that CV risk

management is performed. One of the main recommendations is that an adequate control

16

of disease activity is necessary to lower the CV risk. Thus, additional studies are needed to

explore MACE in AxSpA and PsA to better understand the effects of DMARDs and

management of CV risk factors on the occurrence of MACE.

17

Acknowledgments

The authors thank all the rheumatologists and patients who participated in SCQM.

Contributing institutions are listed on www.scqm.ch/institutions. We also thank Myriam Riek

from the SCQM foundation who provided the dataset and contributed to the statistical

analysis.

18

References

1. Aviña-Zubieta JA, Choi HK, Sadatsafavi M, Etminan M, Esdaile JM, Lacaille D. Risk of

cardiovascular mortality in patients with rheumatoid arthritis: a meta-analysis of

observational studies. Arthritis Rheum 2008;59:1690–7.

2. Dadoun S, Zeboulon-Ktorza N, Combescure C, Elhai M, Rozenberg S, Gossec L, et al.

Mortality in rheumatoid arthritis over the last fifty years: Systematic review and meta-

analysis. Jt Bone Spine 2013;80:29–33.

3. Lindhardsen J, Ahlehoff O, Gislason GH, Madsen OR, Olesen JB, Torp-Pedersen C, et al. The

risk of myocardial infarction in rheumatoid arthritis and diabetes mellitus: a Danish

nationwide cohort study. Ann Rheum Dis 2011;70:929–34.

4. Rincón ID del, Williams K, Stern MP, Freeman GL, Escalante A. High incidence of

cardiovascular events in a rheumatoid arthritis cohort not explained by traditional cardiac

risk factors. Arthritis Rheum 2001;44:2737–45.

5. Gonzalez A, Maradit Kremers H, Crowson CS, Ballman K V, Roger VL, Jacobsen SJ, et al. Do

cardiovascular risk factors confer the same risk for cardiovascular outcomes in rheumatoid

arthritis patients as in non-rheumatoid arthritis patients? Ann Rheum Dis 2008;67:64–9.

6. Jamnitski A, Symmons D, Peters MJL, Sattar N, McInnes I, McIinnes I, et al. Cardiovascular

comorbidities in patients with psoriatic arthritis: a systematic review. Ann Rheum Dis

2013;72:211–6.

7. Mathieu S, Pereira B, Soubrier M. Cardiovascular events in ankylosing spondylitis: An

updated meta-analysis. Semin Arthritis Rheum 2015;44:551–555.

8. Haroon NN, Paterson JM, Li P, Inman RD, Haroon N. Patients With Ankylosing Spondylitis

19

Have Increased Cardiovascular and Cerebrovascular Mortality: A Population-Based Study.

Ann Intern Med 2015;163:409–16.

9. Polachek A, Touma Z, Anderson M, Eder L. Risk of Cardiovascular Morbidity in Patients

With Psoriatic Arthritis: A Meta-Analysis of Observational Studies. Arthritis Care Res

(Hoboken) 2017;69:67–74.

10. Eriksson JK, Jacobsson L, Bengtsson K, Askling J. Is ankylosing spondylitis a risk factor for

cardiovascular disease, and how do these risks compare with those in rheumatoid arthritis?

Ann Rheum Dis 2017;76:364–370.

11. Schieir O, Tosevski C, Glazier RH, Hogg-Johnson S, Badley EM. Incident myocardial

infarction associated with major types of arthritis in the general population: a systematic

review and meta-analysis. Ann Rheum Dis 2017:annrheumdis-2016-210275.

12. Ridker PM, Buring JE, Cook NR, Rifai N. C-reactive protein, the metabolic syndrome, and

risk of incident cardiovascular events: an 8-year follow-up of 14 719 initially healthy

American women. Circulation 2003;107:391–7.

13. Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation

2002;105:1135–43.

14. Sattar N, McCarey DW, Capell H, McInnes IB. Explaining how “high-grade” systemic

inflammation accelerates vascular risk in rheumatoid arthritis. Circulation 2003;108:2957–

63.

15. Westlake SL, Colebatch AN, Baird J, Kiely P, Quinn M, Choy E, et al. The effect of

methotrexate on cardiovascular disease in patients with rheumatoid arthritis: a systematic

literature review. Rheumatology (Oxford) 2010;49:295–307.

20

16. Micha R, Imamura F, Wyler von Ballmoos M, Solomon DH, Hernán MA, Ridker PM, et al.

Systematic review and meta-analysis of methotrexate use and risk of cardiovascular disease.

Am J Cardiol 2011;108:1362–70.

17. Barnabe C, Martin B-J, Ghali W a. Systematic review and meta-analysis: anti-tumor

necrosis factor α therapy and cardiovascular events in rheumatoid arthritis. Arthritis Care

Res (Hoboken) 2011;63:522–9.

18. Roubille C, Richer V, Starnino T, McCourt C, McFarlane A, Fleming P, et al. The effects of

tumour necrosis factor inhibitors, methotrexate, non-steroidal anti-inflammatory drugs and

corticosteroids on cardiovascular events in rheumatoid arthritis, psoriasis and psoriatic

arthritis: a systematic review and meta-analysis. Ann Rheum Dis 2015;74:480–489.

19. Ciubotariu E, Gabay C, Finckh A, Physicians of the Swiss Clinical Quality Management

Program for Rheumatoid Arthritis. Joint damage progression in patients with rheumatoid

arthritis in clinical remission: do biologics perform better than synthetic antirheumatic

drugs? J Rheumatol 2014;41:1576–82.

20. Courvoisier DS, Combescure C, Agoritsas T, Gayet-Ageron A, Perneger T V. Performance

of logistic regression modeling: Beyond the number of events per variable, the role of data

structure. J Clin Epidemiol 2011;64:993–1000.

21. Bursac Z, Gauss CH, Williams DK, Hosmer DW. Purposeful selection of variables in logistic

regression. Source Code Biol Med 2008;3:17. Available at:

http://scfbm.biomedcentral.com/articles/10.1186/1751-0473-3-17.

22. Jamnitski A, Visman IM, Peters MJL, Boers M, Dijkmans B a C, Nurmohamed MT.

Prevalence of cardiovascular diseases in psoriatic arthritis resembles that of rheumatoid

21

arthritis. Ann Rheum Dis 2011;70:875–6.

23. Fernández-Gutiérrez B, Perrotti PP, Gisbert JP, Domènech E, Fernández-Nebro A, Cañete

JD, et al. Cardiovascular disease in immune-mediated inflammatory diseases: A cross-

sectional analysis of 6 cohorts. Medicine (Baltimore) 2017;96:e7308.

24. Eriksson JK, Jacobsson L, Bengtsson K, Askling J. Is ankylosing spondylitis a risk factor for

cardiovascular disease, and how do these risks compare with those in rheumatoid arthritis?

Ann Rheum Dis 2017;76:364–370.

25. Libby P. Inflammation in atherosclerosis. Nature 2002;420:868–74.

26. Skeoch S, Bruce IN. Atherosclerosis in rheumatoid arthritis: is it all about inflammation?

Nat Rev Rheumatol 2015;11:390–400.

27. Rho YH, Chung CP, Oeser A, Solus J, Asanuma Y, Sokka T, et al. Inflammatory mediators

and premature coronary atherosclerosis in rheumatoid arthritis. Arthritis Rheum

2009;61:1580–5.

28. Sieper J, Poddubnyy D. Axial spondyloarthritis. Lancet (London, England) 2017;6736:1–

12.

29. Fitzgerald O, Winchester R. Psoriatic arthritis: from pathogenesis to therapy. Arthritis Res

Ther 2009;11:214.

30. Mease PJ. Tumour necrosis factor (TNF) in psoriatic arthritis: pathophysiology and

treatment with TNF inhibitors. Ann Rheum Dis 2002;61:298–304.

31. Chen Y, Chung H-Y, Zhao C-T, Wong A, Zhen Z, Tsang HH-L, et al. Left ventricular

myocardial dysfunction and premature atherosclerosis in patients with axial

22

spondyloarthritis. Rheumatology (Oxford) 2015;54:292–301.

32. Caliskan M, Erdogan D, Gullu H, Yilmaz S, Gursoy Y, Yildirir A, et al. Impaired coronary

microvascular and left ventricular diastolic functions in patients with ankylosing spondylitis.

Atherosclerosis 2008;196:306–12.

33. Suzuki E, Mellins ED, Gershwin ME, Nestle FO, Adamopoulos IE. The IL-23/IL-17 axis in

psoriatic arthritis. Autoimmun Rev 2014;13:496–502.

34. Layh-Schmitt G, Colbert RA. The interleukin-23/interleukin-17 axis in spondyloarthritis.

Curr Opin Rheumatol 2008;20:392–7.

35. Karbach S, Croxford AL, Oelze M, Schüler R, Minwegen D, Wegner J, et al. Interleukin 17

drives vascular inflammation, endothelial dysfunction, and arterial hypertension in psoriasis-

like skin disease. Arterioscler Thromb Vasc Biol 2014;34:2658–68.

36. Ridker PM, Everett BM, Thuren T, MacFadyen JG, Chang WH, Ballantyne C, et al.

Antiinflammatory Therapy with Canakinumab for Atherosclerotic Disease. N Engl J Med

2017;377:1119–1131.

37. Kremers HM, Crowson CS, Therneau TM, Roger VL, Gabriel SE. High ten-year risk of

cardiovascular disease in newly diagnosed rheumatoid arthritis patients: a population-based

cohort study. Arthritis Rheum 2008;58:2268–74.

38. Holmqvist ME, Wedrén S, Jacobsson LTH, Klareskog L, Nyberg F, Rantapää-Dahlqvist S, et

al. No increased occurrence of ischemic heart disease prior to the onset of rheumatoid

arthritis: results from two Swedish population-based rheumatoid arthritis cohorts. Arthritis

Rheum 2009;60:2861–9.

39. Arts EE a, Fransen J, Broeder A a den, Popa CD, Riel PLCM van. The effect of disease

23

duration and disease activity on the risk of cardiovascular disease in rheumatoid arthritis

patients. Ann Rheum Dis 2015;74:998–1003.

40. Hung Y, Lin L, Chen C, Chiou J, Wang Y-H, Wang PY, et al. The effect of anti-rheumatic

medications for coronary artery diseases risk in patients with rheumatoid arthritis might be

changed over time: A nationwide population-based cohort study. Wallace GR, ed. PLoS One

2017;12:e0179081.

41. Zink A, Askling J, Dixon WG, Klareskog L, Silman AJ, Symmons DPM. European biologicals

registers: methodology, selected results and perspectives. Ann Rheum Dis 2009;68:1240–

1246.

42. Curtis JR, Chen L, Bharat A, Delzell E, Greenberg JD, Harrold L, et al. Linkage of a de-

identified United States rheumatoid arthritis registry with administrative data to facilitate

comparative effectiveness research. Arthritis Care Res (Hoboken) 2014;66:1790–8.

43. Machón M, Arriola L, Larrañaga N, Amiano P, Moreno-Iribas C, Agudo A, et al. Validity of

self-reported prevalent cases of stroke and acute myocardial infarction in the Spanish cohort

of the EPIC study. J Epidemiol Community Health 2013;67:71–75.

44. Eliassen B-M, Melhus M, Tell GS, Borch KB, Braaten T, Broderstad AR, et al. Validity of

self-reported myocardial infarction and stroke in regions with Sami and Norwegian

populations: the SAMINOR 1 Survey and the CVDNOR project. BMJ Open 2016;6:e012717.

45. Maradit-Kremers H, Crowson CS, Nicola PJ, Ballman K V., Roger VL, Jacobsen SJ, et al.

Increased unrecognized coronary heart disease and sudden deaths in rheumatoid arthritis: A

population-based cohort study. Arthritis Rheum 2005;52:402–411.

46. Obsan. Infarctus aigu du myocarde par classe d’âge et sexe.

24

http://www.obsan.admin.ch/data/xls/011_ag_f.xls 2017.

47. Obsan. Attaque cérébrale par classe d’âge et sexe.

http://www.obsan.admin.ch/data/xls/013_ag_f.xls 2017.

48. Holmqvist ME, Wedrén S, Jacobsson LTH, Klareskog L, Nyberg F, Rantapää-Dahlqvist S, et

al. Rapid increase in myocardial infarction risk following diagnosis of rheumatoid arthritis

amongst patients diagnosed between 1995 and 2006. J Intern Med 2010;268:578–585.

49. Solomon DH, Reed GW, Kremer JM, Curtis JR, Farkouh ME, Harrold LR, et al. Disease

activity in rheumatoid arthritis and the risk of cardiovascular events. Arthritis Rheumatol

(Hoboken, NJ) 2015;67:1449–55.

50. Arts EE, Fransen J, Broeder AA Den, Riel PLCM van, Popa CD. Low disease activity

(DAS28≤3.2) reduces the risk of first cardiovascular event in rheumatoid arthritis: a time-

dependent Cox regression analysis in a large cohort study. Ann Rheum Dis 2017;76:1693–

1699.

25

Table 1. Baseline demographic and disease characteristics

RA (n=3,070) PsA (n=782) AxSpa (n=1,463)

Age [yrs] 47 (37-56) 39 (30-49) 29 (23-39)

Male 768 (25%) 418 (53%) 804 (55%)

Known familial history of MACE 357 (13%) 116 (17%) 174 (14%)

Ever-hypertension 1,141 (39%) 277 (37%) 346 (26%)

Ever-diabetes 214 (7%) 50 (7%) 63 (4%)

Ever-hyperlipidemia 664 (25%) 164 (25%) 208 (18%)

Ever-smoking 1,701 (56%) 451 (58%) 892 (61%)

Disease duration at inclusion [yrs] 4.1 (1.5-10.1) 6.1 (2.2-12.2) 7.4 (2.8-15.2)

Disease duration at MACE or end

of follow-up [yrs]10.0 (5.0-17.0) 11.0 (6.0-19.0) 10.0 (5.0-16.0)

Follow-up time since inclusion

[yrs]4.0 (2.0-9.0) 3.0 (1.0-5.0) 3.0 (1.0-6.0)

Seropositivity 2,290 (79%) - -

26

Reported values are median and interquartile range for continuous variables and n (%) for categorical variables.

For categorical characteristics with missing data, the percentage may differ slightly from the number of cases divided by the total number of patients with the disease.

AxSpA: Axial spondyloarthritis; MACE: Major adverse cardiovascular events; PsA: Psoriatic arthritis; RA: Rheumatoid arthritis.

27

Table 2. Unadjusted and adjusted incidence rate ratio for MACE

Events/py Rate per

1000 py

IRR 95%CI P-value

Unadjusted -- -- --

RA 100/37495 2.63 -- -- --

AxSpA 28/19837 1.41 0.53 0.34-0.80 0.003

PsA 13/9171 1.42 0.53 0.30-0.95 0.03

Adjusted

RA -- -- --

AxSpA 0.93 0.51-1.69 0.8

PsA 0.56 0.27-1.14 0.11

Age [yrs] 1.07 1.05-1.09 <0.001

Gender (male) 2.09 1.37-3.19 <0.001

Known familial history of

MACE

1.70 1.05-2.75 0.03

Ever-smoking 1.46 0.95-2.24 0.09

Hypertension 1.56 1.01-2.41 0.04

Diabetes 0.63 0.33-1.19 0.16

Hyperlipidemia 4.74 3.10-7.24 <0.001

Disease duration [yrs] 1.04 1.01-1.07 0.002

AxSpA: Axial spondyloarthritis; CI: Confidence interval; IRR:

Incidence rate ratio; MACE: Major adverse cardiovascular

events; PsA: Psoriatic arthritis; RA: Rheumatoid arthritis.

28

Table 3. Unadjusted and adjusted prevalence ratio for MACE

Events/

patients

Rate (%) PR 95%CI P-value

Unadjusted

RA 153/3123 4.89 -- -- --

AxSpA 33/1468 2.25 0.44 0.30-0.64 <0.001

PsA 23/792 2.90 0.51 0.36-0.89 0.02

Adjusted

RA -- -- --

AxSpA 0.99 0.60-1.58 0.97

PsA 0.68 0.39-1.10 0.14

Age [yrs] 1.07 1.06-1.09 <0.001

Gender (male) 1.76 1.28-2.44 <0.001

Known familial history

of MACE

1.67 1.14-2.40 0.007

Ever-smoker 1.64 1.20-2.35 0.004

Hypertension 1.38 0.96-1.85 0.053

Diabetes 0.86 0.53-1.31 0.54

29

Hyperlipidemia 3.57 2.67-5.06 <0.001

Disease duration [yrs] 1.08 1.06-1.10 <0.001

AxSpA: Axial spondyloarthritis; CI: Confidence interval; MACE: Major adverse cardiovascular events; PsA: Psoriatic arthritis; PR: Prevalence ratio; RA: Rheumatoid arthritis.

30

Table 4. Predictor of MACE in RA – Univariable and multivariable analysis

Univariable analysis Age and gender

adjusteda

Multivariable analysis

IRR 95% CI p IRR 95% CI p IRR 95% CI p

Age [yrs]) 1.04 1.01-1.06 0.003 1.03 1.01-1.05 0.007 1.07 1.03-1.12 0.001

Gender (male) 2.42 1.36-4.29 0.003 2.17 1.21-3.9 0.009 2.41 1.17-4.96 0.02

Known family

history of early

MACE

1.48 0.69-3.19 0.31 1.76 0.82-3.82 0.15 2.19 0.89-5.06 0.09

Ever-smoking 2.17 1.18-3.98 0.01 1.79 0.96-3.34 0.07 3.22 1.39-7.45 0.006

Hypertension 2.23 1.23-4.03 0.008 2.03 1.11-3.69 0.02 1.21 0.58-2.5 0.61

Diabetes 1.99 0.84-4.67 0.12 1.56 0.66-3.69 0.32 0.68 0.23-2.01 0.49

Hyperlipidemia 4.57 2.53-8.27 < 0.001 4.48 2.48-8.11 <0.001 3.87 1.90-7.92 <0.001

Disease

duration [yrs]

1.02 0.99- 1.04 0.17 1.07 1.04-1.10 <0.001 1.09 1.04-1.14 <0.001

Seropositivity 0.47 0.24-0.91 0.02 0.53 0.27-1.04 0.07 0.64 0.29-1.42 0.27

HAQ-DI at

inclusion1

1.48 1.007-2.17 0.04 1.68 1.13-2.50 0.01 1.27 0.77-2.11 0.34

DAS28-CRP

at inclusion

Low disease 0.80 0.23-2.72 0.72 0.83 0.24-2.84 0.77

31

activity (DAS28-

CRP ≥2.6-<3.2)

Moderate

disease activity

(DAS28-CRP

≥3.2–≤5.1)

1.13 0.48-2.64 0.78 1.22 0.52-2.86 0.65

High disease

activity (DAS28-

CRP >5.1)

0.90 0.34-2.36 0.83 0.93 0.35-2.45 0.89

Use of bDMARDs 1.09 0.51-2.32 0.83 1.34 0.62-2.88 0.46

Use of GC 0.75 0.43-1.32 0.32 0.69 0.39-1.21 0.19

Use of MTX 0.84 0.42-1.68 0.62 0.74 0.37-1.48 0.39

Use of NSAIDs 0.55 0.31-0.97 0.04 0.57 0.33-1.00 0.0503 1.02 0.51-2.02 0.95

a. age-adjusted for gender; gender-adjusted for age;

bDMARDs: Biologic disease modifying antirheumatic drugs; CI: Confidence interval; GC:

glucocorticoid; HAQ-DI: Health Assessment Quality Disability Index; IRR: Incidence rate

ratio; MACE: Major adverse cardiovascular events; MTX: methotrexate; NSAIDs:

Nonsteroidal anti-inflammatory drugs; RA: Rheumatoid arthritis.

32

Figure legend:

Fig. 1. Flowchart for the analysis of the incidence of major cardiovascular events.

33