Article Text
PDF
Abstract
Background Acute coronary syndrome (ACS) and other cardiovascular diseases are the main drivers of the increased morbidity and preterm mortality in rheumatoid arthritis (RA). ACS in RA has been linked to inflammation and RA severity. During recent years and with new therapeutic options and treat-to-target strategies, increasing efforts have been made to reach RA remission as soon as possible after diagnosis, and the average level of RA disease activity has declined. Whether this has resulted in declining excess risks for RA comorbidities remains unclear.
Methods We performed a nationwide population-based cohort study of patients with new-onset RA from 1997 to 2014, and matched general population comparators. In the Swedish healthcare system, all residents have equal access to healthcare services. Healthcare is monitored using high-quality population-based registers that can be linked together. 15 744 patients with new-onset RA, identified from the Swedish Rheumatology Quality Register, and 70 899 general population comparator subjects were included.
Results Seven hundred and seventy two patients with RA developed an ACS during 103 835 person-years of follow-up (crude incidence, 7.4 per 1000), corresponding to an overall HR versus the general population of 1.41 (95% CI 1.29 to 1.54). Whereas the ACS incidence declined over calendar time in both the RA and the general population cohort, the excess and the relative risks of ACS remained the same.
Conclusions Despite improved disease control in new-onset RA, the elevated risk of ACS in RA remains a concern.
- Rheumatoid Arthritis
- cardiovascular
- cohort
- risk
- Sweden
Statistics from Altmetric.com
Introduction
Cardiovascular disease (CVD) is the main driver of the excess morbidity and preterm mortality in patients with rheumatoid arthritis (RA).1–3 We and others have previously demonstrated an increased risk of acute coronary syndrome (ACS) in patients with RA compared with the general population, already within a few years of the RA diagnosis.4 The mechanisms behind this risk increase are not completely elucidated, although the risk of ACS in RA has repeatedly been linked to RA disease activity and severity.5–8 Because of this, and in addition to close monitoring of traditional cardiovascular risk factors, clinical cardioprotective recommendations in RA include active treatment aiming for control of the RA inflammatory activity.9
During the last two decades, more intense treatment strategies and new therapeutic options have been introduced in the management of new-onset RA, and the average level of RA disease control has improved. Taken together, ‘modern rheumatology’ might thus have impacted the risk of RA-associated comorbidities, including the elevated risk of CVD in RA.10–13 Few studies have, however, addressed the risk of ACS in RA over calendar time periods and disease duration with focus on patients diagnosed with RA during the most recent decade. The aim of this study was therefore to investigate whether improved management in terms of treatment options, treatment algorithms and cardiovascular vigilance in new-onset RA collectively have led to a reduction in the excess risk of ACS.
Methods
Design
We performed a nationwide population-based cohort study of patients with newly diagnosed RA (defined as RA diagnosis within 12 months of patient-reported symptom onset, to be able to assess the temporal development of ACS in relation to disease onset, not just diagnosis) with matched general population comparator subjects, based on prospectively recorded register data.
Setting
The publicly funded Swedish healthcare system enables access to all healthcare services, including specialised care for chronic diseases such as RA, for all residents. Patients with RA are diagnosed and cared for by rheumatologists or internists, typically in hospital outpatient settings. Prescribed drugs are subsidised and, beyond an upper annual limit of SEK2200 (approximately US$260), provided free of charge. All residents are assigned a unique personal identity number that can be used for linkage of different data resources, including several national health registers of high quality.14
Study population
The new-onset RA cohort
The Swedish Rheumatology Quality (SRQ) Register was initiated in 1995 and includes individuals aged 16 years or older fulfilling the 1987 American College of Rheumatology criteria for RA.15 The current coverage (on a national basis) is estimated to above 80% of all new-onset RA.16 The register collects information on age, sex, rheumatoid factor (RF) status, date of first symptom of RA, date of inclusion into the register and the personal identification number. The register also contains information on drug treatment and disease activity, for example, number of swollen and tender joints, erythrocyte sedimentation rate, serum concentration of C reactive protein, patient’s and physician’s assessment of global disease activity, and Disease Activity Score 28-joint count (DAS28), from the inclusion visit and onwards. For this study, we identified all individuals diagnosed with RA between 1 January 1997 and 31 December 2014 who were included in the register within 12 months of first symptoms of RA (n=16 214). Validations against other data sources suggest that less than 0.8% of all new-onset RA in SRQ represents prevalent RA misclassified as new-onset (D di Giuseppe, personal communication, SRQ 2016). For this study, the date of inclusion into the register (typically the date of RA diagnosis) was used as index date.
General population comparators cohort
For each unique patient with RA, we randomly selected up to five individuals from the Swedish Population Register (which includes all Swedish residents), matched on sex, year of birth and residential area (n=72 939). Each subject was assigned the same index date as the corresponding patient with RA.
Data sources used for follow-up and to detect ACS
Using the personal identification number, we linked the cohort of patients with RA and the matched general population comparator cohort with the following data sources, for which data were available through 31 December 2014: The National Patient Register, the Population Register, the Cause of Death register, and the Integrated Database for Labor and Education. The National Patient Register contains information on inpatient care since 1964, with nationwide full coverage since 1987.17 The register lists date of admission, date of discharge and the discharge diagnosis (primary and secondary diagnoses) as set by the discharging physician and classified according to the calendar year-specific version of the International Classification of Diseases (ICD, since 1997: ICD-10). The Population Register includes information on deaths, emigration and immigration for the entire Swedish population. The Cause of Death register holds information on cause of death coded according to ICD. The Integrated Database for Labor and Education holds information on educational level. Through these linkages, we identified all hospitalisations and non-primary care outpatient visits, before or after index date, and all deaths and emigrations during follow-up. The nationwide and near complete coverage of the National Patient Register ensured very low (but formally not assessable) missingness.
Follow-up and occurrence of ACS
ACS was defined as hospitalisation listing a primary ICD-10 diagnosis of I21 (acute myocardial infarction) or I20.0 (unstable angina pectoris), or an acute myocardial infarction listed as the underlying cause of death. This definition has a positive predictive value of 95%.18 In both cohorts, all individuals with a diagnosis of ACS prior to the start of follow-up were excluded. The RA cohort and the comparison cohort were followed from the index date until the first ACS, death, emigration or 31 December 2014, whichever came first.
Statistical analyses
Descriptive baseline data were summarised and presented as proportions, means and medians as appropriate. The incidence of ACS was described and assessed by dividing the number of ACS (overall and in subgroups based on sex, age and calendar year of diagnosis) with the corresponding person-years of follow-up. The excess incidence of ACS was defined as the difference between the incidence in the RA cohort and the corresponding incidence in the general population cohort. The HR (used as measure of relative risk) of ACS in RA compared with the general population was calculated using Cox’ regression models adjusted for residential area, sex, year of diagnosis, age at diagnosis and educational level. Analyses were stratified by RF status, sex, age at index date, calendar period of index date, time since start of follow-up and DAS28 (≤3.2 and >3.2) at RA diagnosis. All analyses were carried out with SAS V.9.4 software package. This study was approved by the Stockholm Ethics Review Board.
Results
After excluding individuals with a history of ACS at start of follow-up (470 (2.9%) patients with RA and 2040 (2.8%) general population comparator subjects), 15 744 patients with RA and 70 899 comparator subjects remained for analysis. Sixty-nine per cent of all individuals were women and the mean age at index date was 57 years (table 1).
Demographics at year of diagnosis among Swedish patients with rheumatoid arthritis (RA) overall and by calendar period of RA diagnosis, and in matched general population comparator subjects
On average, patients with RA had a somewhat lower level of education than the general population comparators (25% vs 29% with >12 years of education). When our study period was split into four groups as defined by calendar period of RA diagnosis, the gender distribution was stable, but the distributions of age (increasingly higher), RF status (increasingly more seronegative RA) and educational level (increasingly higher) varied over time (table 1).
Clinical RA characteristics
During the study period, the duration of RA symptoms at index date (diagnosis) decreased somewhat from 1997 to 2002 and from 2011 to 2014 (table 2). The disease activity at index date and at the return visit at 3–6 months also declined modestly. The use of any disease-modifying antirheumatic drug within the first year after RA diagnosis was high; the proportion prescribed glucocorticoids, methotrexate or biological drugs during the first year after RA diagnosis increased during the study period.
Clinical characteristics of Swedish patients with new-onset rheumatoid arthritis (RA) at inclusion in the Swedish Rheumatology Quality register, including data on treatments initiated within 1 year of inclusion, by calendar period of RA diagnosis
Incidence of ACS during follow-up
During 103 835 person-years of follow-up (median/interquartile follow-up=5.7/7.1 years) in the RA cohort, 772 individuals developed a first-ever ACS. In the comparator cohort (median/interquartile follow-up=5.7/7.3), 2418 individuals developed a first-ever ACS during 466 930 person-years of follow-up (table 3). In both the RA and the general population comparator cohort, the incidence of ACS was higher among men, increased with age and decreased markedly over successive calendar periods of start of follow-up (figure 1). During follow-up, 1685 (10.7%) of the patients with RA and 7336 (10.4%) of the general population subjects died. Seventy seven (0.5%) of patients with RA and 824 (1.2%) of the general population emigrated from Sweden during follow-up and where therefore censored at these time-points.
Number of events (N), person-years at risk (PY) and incidence rates (N/1000 PY) of acute coronary syndrome in new-onset rheumatoid arthritis (RA) identified between 1997 and 2014 and in an individually matched general population comparator
Panels refer to the number of events/1000 person-years by calendar period and time since diagnosis in RA and general population.
Excess incidence of ACS by calendar period and time since RA diagnosis
Overall, the excess incidence of ACS increased from around 1/1000 person-years during the first year after RA diagnosis to between 2/1000 and 3/1000 person-years during the next 10 years. There was no evidence of any clear secular trend in the point estimates for excess risk, but the excess risk remained similar in all calendar periods (around 1/1000 person-years during the first year to between 2/1000 and 3/1000 person-years during the next 10 years).
Relative risk of ACS by calendar period and time since RA diagnosis
Overall, RA was associated with approximately 40% higher risk of ACS, HR 1.41 (95% CI 1.29 to 1.54, table 3). We noted statistically significantly increased risks, and similar relative risks, in all subsets defined by age and sex, but also that the excess risk was confined to patients with seropositive RA and to patients with DAS28 above 3.2 at diagnosis (table 3).
Further, within each calendar period of RA diagnosis, the overall relative risks remained similar: 1997–2002, HR=1.41 (95% CI 1.24 to 1.60); 2003–2006, HR=1.47 (95% CI 1.24 to 1.74); 2007–2010, HR=1.38 (95% CI 1.13 to 1.68); and 2011–2014, HR=1.19 (95% CI 0.85 to 1.67) (p=0.9). When the calendar years of RA diagnosis and RA disease duration were cross-tabulated, we noted similar HRs per follow-up interval for each calendar period of RA diagnosis (table 4).
HR and 95% CI adjusted for sex, residential area, year of diagnosis age and educational level
The same pattern of lack of distinct calendar trend emerged when the same analyses were performed separately for seropositive RA and seronegative RA (for which the relative risks were typically not increased), and by DAS28 at diagnosis (online supplementary tables 1 and 2).
Discussion
In this nationwide population-based study, we observed a close to 40% decline in the incidence of ACS in the general Swedish population in 1997–2012. In our inception cohort of patients with RA, the level of decline in incidence was similar, but the overall risk of ACS was approximately 40% higher than in the general population. The increase was restricted to patients with DAS28 above 3.2 at RA diagnosis and to patients who were RF-positive. Because the decline in incidence was equally pronounced in the RA cohort as in the general population, there was no evidence of any decline in the excess risk for patients diagnosed with RA in more recent years.
Studies evaluating time trends of mortality and morbidity in CVD in RA based on data from 1950 to 2000 have shown conflicting results. A declining trend of standardised mortality ratio (SMR) from myocardial infarction was observed in an earlier study in patients with debut of RA symptoms before 1970 compared with RA diagnosed from 1980 through 1997.19 A previous Swedish study observed a decrease in the overall mortality (incidence), but similar standardised ratios (relative risks) of morbidity and mortality from CVD comparing patients with established RA assessed in 1978 with patients assessed in 1995.20 This finding is in line with a meta-analysis that presented unchanged excess risks of death (here: SMRs, relative risks) from CVD in RA studies between approximately 1945 and 1995.3 Published studies evaluating time trends in CVD incidence in RA cohorts comprising patients with disease debut after 2000 are scarce. Our present study thus contributes new knowledge regarding contemporary patients, but otherwise corroborates the results from RA cohorts from previous decades and treatment paradigms.
On the one hand, our results suggest that patients with RA may have benefited from the CVD prevention and risk factors intervention that have been implemented in society at large, at least to the same extent as the rest of the population. On the other hand, our results suggest that despite improved RA disease control and increasing recognition of cardiovascular (CV) risks in RA over time, the gap in risk between patients with RA and the general population remains, at least among patients with seropositive RA. Potential explanations include that the increasing RA treatment intensity and efficiency have not affected the excess risk of CV comorbidity, or that such true gains have been offset by risks inherent with these treatment strategies.
A distinct decrease in incidence and mortality from coronary artery disease in the general population has been observed in developed countries since half a century. In part this has been attributed to population-level changes in CV risk factors (eg, smoking, dyslipidaemia, hypertension, inactivity), in part to improved medical treatment.21 It is probable that the positive changes in the general population have also affected individuals with present or forthcoming RA disease. However, as several CV risk factors are also associated with the development of RA and with active RA disease (eg, smoking, dyslipidaemia, physical inactivity, obesity), seropositive RA in particular, it could be argued that the improvement in risk factors might have been less pronounced in the subpopulation that later will develop RA. Under such a scenario, an excess risk of CVD would be observed already prior to RA symptom onset, which was not the case in one of our earlier studies but has been reported from elsewhere (and in our current study, the proportion with a history of ACS at entry was not higher in the RA cohort).22 23 Even if traditional CV risk factors seem to have less impact on the risk of CVD in RA than in the general population, some level of risk reduction due to improved CV risk factor profile is to be expected,24–26 although it is not possible to tell whether the remaining gap in morbidity from ACS is attributable to factors related to the RA disease and its treatment rather than to any increased background risk in the subpopulation with RA.
Our observation of a rapid increase in the risk of ACS after RA diagnosis is in line with our previous results from this cohort and the results from other studies.4 5 20 Patients with recent RA onset have typically experienced a period with active RA disease before the effect of treatment, in the present study with a mean of DAS28 4.6–5.3 at diagnosis, and disease activity at RA diagnosis was clearly linked to excess ACS risk. This inflammatory activity could theoretically have negative effects on plaque stability as well as the formation of thrombus.27 28 Use of prescribed or over-the-counter COX inhibitors or oral glucocorticoids during the first months of RA disease duration might be more frequent than during periods with well-controlled disease activity. It is thus possible that these factors may contribute to the increased morbidity during the first year. The average RA disease activity in our study was well controlled 3–6 months after diagnosis, with the lowest disease activity estimate in the most recent calendar period. The use of methotrexate in early RA increased, and the use of biological treatment doubled during the study period, which, if anything, is likely to exert beneficial rather than detrimental effects on CV risks.29 Interestingly, however, there was also an increasing use of glucocorticoids, for which adverse CV effects are documented.30 The main strength of this study is the large, nationwide, population-based RA inception cohort, with the possibility to add prospectively collected and linked data on comorbidity and covariates from mandatory public registers. In terms of patients with RA, we believe that the present study has a good generalisability as our cohort represents patients from daily routine clinical care rather than patients recruited from, for example, tertiary referral centres or into a research cohort. Similarly, the publicly funded healthcare in Sweden might limit the impact of socioeconomic factors. Also, the Swedish RA treatment guidelines are in good agreement with most other RA guidelines, at least in the western world. In terms of ACS, the trend towards declining incidences in the general population is observed in many countries. Therefore we believe that our results may also be applicable to other settings other than Sweden. The risk of misclassification of the outcome is low; a previous validation found a positive predictive value of 95% for this definition of ACS in the context of early RA.18 Misclassification of prevalent as new-onset RA in SRQ is, as indicated, very low, and thus unlikely to have influenced our results. Although the coverage of the SRQ is high, some degree of selection is possible as individuals with an expected short survival due to age or concomitant morbidity might stand a lower chance of being included in any longitudinal clinical monitoring system. Such selection would, however, not explain the increases in overall HRs in our study, and can only explain the absence of time trends in HRs if the coverage of patients with poor prognosis (and elevated CV risks) has increased over time. Irrespective, our results suggest that RA is still associated with an increased risk of new-onset coronary artery disease. Similarly, despite an increasing proportion of seronegative RA and an increasing proportion of patients with DAS28 <3.2 at entry (both of which would have inflated any declining calendar trend), we did not observe any marked reduction in relative risk during the more recent calendar periods of RA diagnosis. Although this is a large cohort, the numbers of events in some strata were low. We did not have data on traditional risk factors for patients or referents, which is a limitation, although our primary aim was to assess rather than to attribute clinical excess risks. The frequencies of unidentified but true ACS (eg, silent myocardial infarction) in the RA and comparator cohorts are unknown, but based on previous studies unlikely to be higher among the general population comparator subjects.22
In conclusion, in spite of a decline in the absolute risks, the excess and the relative risks of ACS have not declined over calendar time, such that patients with early RA are at an approximately 40% increased risk of ACS compared with the general population. Whether RA disease-related or treatment-related or not, coronary artery disease in RA thus remains a concern and calls for continuous vigilance and implementation of preventive measures early in the RA disease course.
References
Footnotes
Contributors MH had full access to all of the data used for analyses in this study and takes full responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: JA, MH, LL. Acquisition of data: MH, JA. Statistical analysis: MH, LL. Analysis and interpretation of data: MH, LL, JA. Drafting of manuscript: MH, LL. Critical revision of manuscript and final approval given: MH, LL, JA. Obtained funding: JA. Study supervision: JA.
Funding The Swedish Research Council, the Swedish Foundation for Strategic Research, Stockholm County Council (ALF), Heart Lung Foundation and Karolinska Institutet (Strategic Research Area Epidemiology). Funders had no impact on the design or interpretation of the study or its results.
Competing interests None declared.
Ethics approval Ethics committee in Stockholm, Sweden.
Provenance and peer review Not commissioned; externally peer reviewed.