Background A positive impact on the increased risk for ischemic coronary events in rheumatoid arthritis (RA) could be expected by control of inflammation. Anti-rheumatic drugs might also directly improve risks for ischemic events. In a previous study based on the Swedish Rheumatology Register we reported a relative risk of 0.80 (95% CI 0.52-1.24) for acute coronary syndromes (ACS) comparing early RA patients starting TNF inhibitor therapy (TNFi) to patients with early RA treated otherwise.
Objectives To evaluate the role of TNFi on the risk of ACS in a larger population-based cohort including both incident and prevalent RA cases.
Methods In the Swedish Biologics Register, a national cohort of patients with RA who started their first TNFi between 2001 and 2009 (n=9,869) was identified. For each TNFi exposed patient, three matched referents were randomly selected from the underlying national cohort of all individuals with two or more outpatients visits listing RA (one of which at a dept. of rheumatology/internal medicine) as identified in the National Patient Register. Furthermore, for each TNFi-exposed patient, five matched referents from the general population were identified in the Population Register. Individuals with previous ischemic or congestive heart disease were excluded. Following matching and exclusions, 7,185 patients starting TNFi treatment, 17,702 matched biologics naïve RA referents, and 32,161 matched general population referents remained. Information on any previous diagnoses of hypertension, diabetes mellitus, chronic pulmonary diseases, infections and joint surgery was extracted from the Patient Register. Data on level of education and work disability was extracted from Statistics Sweden. The outcome, first ever ACS, was defined as a discharge diagnosis of myocardial infarction (ICD10: I21) or unstable angina (ICD10: I20.0) during follow-up.
Results We observed 85 ACS (crude incidence 4.8 per 1000 person-yrs) among the TNFi exposed patients, 280 ACS (5.1 per 1000) among the matched RA comparators, and 284 ACS (2.2 per 1000) among the matched general population subjects. Kaplan-Meier curves displayed no differences in event-free survival between RA patients starting TNFi and their matched comparators, but a lower event-free survival in both these cohorts compared with the general population comparator. The age- and sex- adjusted hazard ratios (HR; 95% CI) for TNFi compared with matched, biologics-naïve RA patients was 0.98 (0.78-1.23), and for TNFi vs. general population 2.13 (1.69-2.68). When further adjusted for level of education, work disability, disease duration, cardiovascular risk factors, joint surgery, chronic pulmonary disease and previous infections the RR for TNFi vs. matched biologics-naive RA remained close to 1 (HR 1.12, 95% CI 0.84-1.48).
Conclusions In this nation-wide, population-based, matched cohort study, no association between treatment with TNFi and the risk of ACS in RA was seen. Compared with the general population a doubled risk for ACS was noted among the patients with RA. The lack of risk reduction vs. the general population through TNFi could be due to lack of a true protective effect, residual confounding or channelling, or an insufficient level of RA disease control whether obtained through TNFi or alternative therapies.
Disclosure of Interest None Declared
Statistics from Altmetric.com
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.