Skip Navigation


Rheumatology Advance Access originally published online on August 14, 2006
Rheumatology 2007 46(2):350-357; doi:10.1093/rheumatology/kel253
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow CME/CE:
Take the course for this article:
 Rheumatology Fourth Quarter 2006 Quiz
Right arrow All Versions of this Article:
46/2/350    most recent
kel253v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Young, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Young, A.
Related Collections
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Mortality in rheumatoid arthritis. Increased in the early course of disease, in ischaemic heart disease and in pulmonary fibrosis

A. Young1, G. Koduri1, M. Batley1, E. Kulinskaya2, A. Gough1, S. Norton2, J. Dixey1 on behalf of the Early Rheumatoid Arthritis Study (ERAS) group

1ERAS, Department of Rheumatology, City Hospital, St Albans and 2Department of Statistics, University of Hertfordshire, Hertfordshire, UK.

Correspondence to: Dr A. Young, Rheumatology Department, City Hospital, St Albans, Hertfordshire AL3 5PN, UK. E-mail: eras{at}whht.nhs.uk


   Abstract
Top
 Abstract
Introduction
 Patients, materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Objective. To examine the cause of death in a large UK inception cohort of rheumatoid arthritis (RA), and whether this was related to disease duration and severity, treatment effects or extra-articular features and complications of RA.

Methods. Standard clinical, laboratory, radiological and socio-economic measures were recorded at baseline and yearly in an inception cohort started in nine centres in 1986. Date and the cause of death were based on death certificates and the comparisons made with age and sex matched population figures. Risk factors for mortality were identified from baseline measures of disease.

Results. There were 459 deaths (32%) in 1429 patients followed for up to 18 yrs. Standard mortality ratio was 1.27. Survival was significantly lower in the first 7 yrs of RA. Excess mortality was seen in cardiovascular disease (31%), pulmonary fibrosis (4%) and lymphoma (2.3%). Baseline predictors for mortality were men, older age, poor function, lower socio-economic status, extra-articular features, comorbidity, rheumatoid factor, X-ray erosions, high-ESR and low-haemoglobin.

Conclusion. There was a modest increase in mortality in RA, mainly in the first 7 yrs. Deaths from cardiovascular disease and pulmonary fibrosis were higher than expected, but treatment-related deaths were low. Risk factors included less favourable socio-economic status, markers of disease severity and diminished function within the first year.

KEY WORDS: Rheumatoid arthritis, Mortality, Cardiovascular disease, Pulmonary fibrosis, Function


   Introduction
Top
 Abstract
 Introduction
Patients, materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Mortality in rheumatoid arthritis (RA) has been increasingly studied over the last few decades, with most studies showing higher mortality rates than general populations. Most have been mainly hospital based in established RA and although many demonstrated increased standard mortality ratios (SMR), these have ranged widely from 149 to 308, summarized in a recent extensive review [1]. There are fewer reports from inception cohorts, and SMR in these are generally lower (87–184) [2–6]. In fact, recent inception cohorts in northern Europeans did not detect any increase in mortality in early RA [2–4]. Explanations include improved therapies started sooner, but the earlier and milder type of RA recruited to inception cohorts is another possibility [7]. Disease severity has been associated with mortality [8], but the expectation that changes in number and use of disease modifying drugs over the last two decades would be reflected in improved long-term outcomes has not been fully realized [9].

A cross-sectional UK study reported that deaths from cancer, respiratory and cardiovascular causes were all increased in RA [10]. Significant mortality has been reported from non-steroidal anti-inflammatory drugs (NSAIDs) and corticosteroids, but disease-modifying anti-rheumatic drugs (DMARDs)-related deaths are low [11]. Conditions that may contribute to mortality include complications and extra-articular features of RA. However, most recent studies have shown that the increased mortality in RA is mainly from cardiovascular disease (CVD) [1, 12], not only in established disease, but also in some inception cohorts [5,6].

Age and sex have been the most consistent predictors for mortality reported in RA, whereas disease specific features such as severity of RA [8], inflammatory markers [13], functional disability [14] and rheumatoid factor (RF) [5] have variable prognostic importance. This is an important area because of the possibility of identifying and targeting patients at risk in order to reduce mortality in RA. This is especially relevant for ischaemic heart disease (IHD), for which traditional risk factors do not fully explain the increased mortality in RA [1].

The early RA study (ERAS), an inception cohort started in 1986, aimed to measure several dimensions of outcome over time [15]. The size of the cohort and length of follow-up have made it possible to examine whether mortality in RA patients, treated with optimal therapies accepted as part of standard practice for the late 1980s to early 2000s, is increased in an inception cohort design, and whether it is related to severity and duration of disease. Secondary aims were to examine causes of death in detail to establish whether these were attributed to RA, its complications or treatment, and to identify early clinical risk factors.


   Patients, materials and methods
Top
 Abstract
 Introduction
 Patients, materials and methods
Results
 Discussion
 Acknowledgements
 References
 
Since 1986, consecutive patients diagnosed with RA were recruited from nine centres representing all social strata in England. Entry criteria included <2 yrs of symptoms and no prior treatment with DMARDs. Presentation features included onset, family and medical histories. Baseline and yearly assessments included standard clinical, laboratory, functional and socio-economic details, as described previously [16]. These included American College Rheumatology (ACR) criteria [17], swollen and tender joint counts [15], Health Assessment Questionnaire (HAQ) [18], erythrocyte sedimentation rate (ESR), RF and antinuclear antibodies (ANA) titres, body mass index (BMI) (height x height/weight) and hand/feet radiographs [19]. ERAS started before minimum core data sets for disease activity, and response criteria in RA were agreed internationally, but in fact all components of these were used from the start, except for the patients’ and the physicians’ global assessments. We used the original three variable disease activity score (DAS) based on swollen and tender joint counts and ESR [20]. Clinicians recorded extra-articular RA (ExRA) and coexistent medical conditions yearly, causes of death if known and all in-patient episodes [21]. At the start of the study in 1986, the importance of cigarette smoking was not fully appreciated, and was not recorded prospectively. Although added later, it was not always possible to obtain retrospectively in those who had died or moved.

Death certificates
All ERAS patients were tracked by the National Health Service Central Register, which holds computerized records of all patients registered with general practitioners in England and Wales. The date and main cause of death were based on death certificates that allow the recording of three other contributing causes and three comorbid conditions. These were provided and coded by the Office for National Statistics (ONS) using International Classification for Disease (ICD)-10. Only patients who moved from the UK permanently would fail to be recorded under this system. Most died in hospital (n = 328, 71%) or nursing homes (n = 21, 4.5%), 99 (22%) had autopsies and general practitioners signed the remaining death certificates. Five death certificates were not available. Because of reported problems with their reliability as indicators for cause of death [22], certificate details were cross-referenced with the ERAS database, and medical records for pre-morbid conditions and hospital episodes, available in 325 (71%). Discrepancies were low (n = 9) for primary cause of death. There was less agreement for secondary causes (see ‘Results’).

Treatment profiles
All centres followed the framework of UK guidelines for management of RA [23]. Sequential monotherapy was standard practice, ‘step up’ combination therapy reserved for more severe disease. A total of 1191 (84%) patients received at least one DMARD, started at a median 2 months from presentation (68% by 3 months and 87% by 12 months), consistent with the group's early treatment practice of this era. Remaining patients (16%) were managed with NSAIDs and/or low-dose steroids. Drugs used were sulphasalazine (54%), methotrexate (18%), intramuscular gold (13%), D-penicillamine (9%), antimalarials (4%) and various others (2%). Of the DMARD-treated patients, 55% had more than one. Steroids in doses greater than 5 mg daily for at least 12 months were used in 249 (17%).

Statistical analysis
Computer-generated summary statistics have been used to demonstrate baseline clinical features. Age- and sex-specific mortality rates for England and Wales populations in 1986–2002 were obtained from the ONS [24]. Since only 31 patients under 25 (2%) entered the study, we used only patients aged 25 or older in the subsequent analysis. Age of entry was stratified in 10 yearly intervals resulting in seven age categories (the last being 85+). Survival experience for each of the 14 cohorts (age by sex) was tabulated. Poisson distribution was used to obtain SEs. Cumulative survival was then directly standardized to standard European population 25+ yrs old to obtain cumulative survival and its SE. Lower and upper limits for a 95% confidence interval (CI) for survival are based on a normal approximation. Expected survival is based on average survival rates for England and Wales for the last 17 yrs. All cause and cause-specific SMR were calculated using the statistical package STATA-9, where numbers were sufficient and data were available from ONS. In univariate analysis, odds ratios with 95% CIs were calculated to determine baseline clinical risk factors for mortality. For continuous variables, worst quartiles were used as cut off. Multivariate analysis was performed using Cox regression.


   Results
Top
 Abstract
 Introduction
 Patients, materials and methods
 Results
Discussion
 Acknowledgements
 References
 
Demography of ERAS cohort
There were 459 deaths (32%) in 1429 patients recruited from 1986 to 1997, 190 men (41%) and 269 women (59%). Mean age at presentation was 57 in men and 54 in women. Median duration of symptoms to initial visit was 6 months indicating early disease, median follow-up was 9.1 yrs, maximum 19 yrs. Table 1 summarizes baseline features of the whole cohort. These are similar to other early RA studies.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Features at presentation of whole cohort with eventual outcome

 
Survival
All-cause SMR was 1.27 (1.04–1.46). Final results of the analysis of mortality in ERAS are shown in Fig. 1. This shows that the survival of the ERAS population was lower than that expected for the first 7 yrs.


Figure 1
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Graphs show survival in years for men and women. Cumulative survival is in light grey with 95% CIs in dotted lines. Expected survival is in black.

 
Cause of death
Table 2 shows SMR in broad categories of cause of death. IHD was mainly responsible for reduced survival in this cohort. Table 3 shows actual causes of death by age and mean duration of RA. Unexpected findings were early death in women from cardiovascular and respiratory conditions, mainly IHD and pulmonary fibrosis (PF). RA was recorded on death certificates in 82 patients (18%), 17 as contributory and 65 as comorbidity, but never as main cause of death.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Main causes of death (n = 459) as broad categories

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Main causes of death in all patients, and in men and women

 
Cardiovascular disease
IHD was the single most common primary cause of death (n = 109, 24%). SMR was 1.49 (121–177). It was a secondary cause of death in another 13 and a comorbid condition on death certificates in six. Cardiac and non-cardiac vascular disease (aortic aneurysm, thrombo-embolism, stroke and reno-vascular) were primary cause of death in a total of 181 (39%) patients, and recorded as secondary or comorbidity in another 16 (6%) deaths. Mean survival from baseline was 1–2 yrs lower in IHD compared with other major categories (Table 3) and to all other deaths combined (8.7 yrs). Five-year survival was 41% in IHD compared with 69% from other causes. IHD was mainly responsible for the early mortality demonstrated in Fig. 1. IHD was a known comorbid condition at presentation in 18 (16.5%) patients who died of IHD, and in a further 16 (15%) by 3 yrs.

Malignancy (solid tumours 20% and neoplasia of immune system 3.7%)
Cancer mortality was similar to population figures except for non-Hodgkins lymphoma (NHL), a primary cause of death in 11 (2.3%) and a secondary cause in two. Because of the missing smoking data, it was not possible to obtain the adjusted rates for lung cancer.

Deaths related to drug therapy, extra-articular and complications of RA
ExRA was a primary or secondary cause of death in 27 (6%). Complications of RA were contributory in eight.

Pulmonary fibrosis (PF)
This was primary in 18, secondary in five [three CVD and two chronic obstructive pulmonary disease (COPD) deaths] and contributory in three other deaths. In another nine patients who died of other causes, PF was a comorbid condition on the ERAS database but not on death certificates. Of the 26 with PF on death certificates, 17 already had a clinical diagnosis, three at presentation and 10 within 3 yrs of follow-up (mean 30 months). In nine, the diagnosis was only made terminally or at autopsy. Only one patient had scleroderma also. Women died from PF at earlier ages (66 yrs, Table 3) and also had earlier age of onset of RA (58 yrs). A total of 13 (50%) had received steroids, 22 (85%) were on DMARDs, 10 received methotrexate, nine with a diagnosis of PF prior to therapy and none developed pneumonitis.

Vasculitis
RA associated systemic vasculitis (1) and Wegener's Granulomatosis (2) were the secondary causes on three death certificates. A total of 19 deceased patients had major extra-articular conditions diagnosed clinically, but not recorded on death certificates: vasculitis (10 systemic, one ocular/cutaneous), neuromyopathies (6) and Felty's syndrome (2). Although limited by numbers, associations with specific mortality groups (e.g. IHD) were not seen.

Other extra-articular and complications of RA
One patient died from a myopathy. The deceased patients had more extra-articular conditions in the first year than those alive (25 and 15%, Table 1). Except for ExRA described earlier, these were not recorded on death certificates. Sjogren's syndrome was equally present in deceased and alive patients (10%), and present in two lymphoma deaths. RA complications were recorded on 10 certificates, all septicaemia deaths: septic arthritis (3), hip fracture (5) and chronic leg ulcers (2). On the ERAS database, 23 deceased patients had sustained fractures (16 hip, four vertebral and three other), 13 within 12 months of death due to pneumonia (6), stroke (4), CVD (2) and septicaemia (1). Nine were or had been on steroids. There were no immediate post-operative orthopaedic deaths. Of the 20 patients with known cervical spine subluxation, seven died, none with cervical myelopathy.

Drug related deaths
Gastrointestinal (GI) haemorrhage and/or perforations were the primary causes of death in 14 (10 upper, four lower GI) and contributory in three. All except one had been on long-term or recent NSAIDs, and two on steroids. Although not cited in death certificates, eight were considered NSAID related by attending rheumatologists. Upper GI haemorrhage was diagnosed clinically in a further nine patients (from ERAS database) who died of other causes, but not noted on death certificates. We reviewed deceased patients carefully for any other drug attributed deaths, on death certificates and the ERAS database. DMARDs were used in 361 (81%) patients who died, totalling 625 new DMARD starts. Drug-induced blood dyscrasias were not implicated in any deaths. Of the 21 septicaemia deaths, neither DMARDs (n = 17) nor steroids (n = 5) were implicated on death certificates, but steroid toxicity was implicated by ERAS clinicians in six of these patients, one complicated by systemic vasculitis. None of the three with septic arthritis followed intra-articular steroids. The ERAS database recorded one Gold induced and two NSAID nephropathies in renal deaths (eight primary, four secondary), but these were not implicated on death certificates.

Risk factors for mortality
Patients who died had worse function and erosive RA, less favourable socio-economic status and more extra-articular RA or other comorbid conditions at presentation. Table 1 also shows that HAQ, DAS and ESR were worse in these patients, but in fact all baseline and 1-yr clinical and laboratory measures of function and inflammatory activity were worse, and to a similar degree, across all broad categories of death except for solid tumours. Steroids were used for 3–5 yrs in 14% of the living patients, compared with 20–24% in all mortality categories except for the 47% in PF deaths. Time to first DMARD and their use during the first 3–5 yrs was similar between all the mortality categories (data not shown), except for greater use in PF and NHL deaths. The latter all received DMARDs early, seven more than one, because of more severe RA (mean DAS = 4.4, mean HAQ = 1.4). Six had methotrexate and two, azathiaprine.

Univariate analysis
Table 4 shows all baseline variables with predictive value for all-cause mortality. The odds ratios were similar for CVD, stroke and respiratory deaths when analysed separately. No significant differences were seen between early and late CVD deaths. Most variables at 1 yr had improved predictive value, similar to other outcome studies from this group [19, 21].


View this table:
[in this window]
[in a new window]

  		TABLE 4. Risk factors for mortality using features at presentation. Univariate analysis

 
Multivariate analysis
All baseline and 1 yr measures were included in regression models in patients with all these variables (n = 1272). Table 5 shows the most powerful independent predictors for mortality developed from stepwise Cox regression. Controlling for number of DMARDs, steroids and comorbidities in the first year made little difference. For CVD deaths alone, independent predictors were the same, except for the exclusion of haemoglobin. HAQ was included in the model at 1 yr and Fig. 2 shows survival curves by 1st yr HAQ.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Risk factors for mortality using Cox regression

 

Figure 2
View larger version (20K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 2. HAQ at 1st yr has been categorized into quartiles. Patients in the worst quartile have worst survival.

 

   Discussion
Top
 Abstract
 Introduction
 Patients, materials and methods
 Results
 Discussion
Acknowledgements
 References
 
In this UK inception cohort, a modest increase in all-cause mortality was found in RA, with survival noticeably reduced in the first 7 yrs of disease. This was in patients in general hospital settings, with disease onset in the late 1980s/90s, and prior to the biological era. Cause of death was similar to the general population, except for increases in infections, non-Hodgkin's lymphoma, pulmonary fibrosis and CVD, especially IHD in the first 7 yrs. RA-related conditions were primary or secondary causes of death in a total of 35 (8%), mainly PF (n = 27, 6%). Apart from NSAID-related GI deaths (n = 8), we found no definite evidence of drug-related mortality. Deaths attributable to steroids, including septicaemia and osteoporotic hip fractures were uncertain, but not high. Extra-articular features and complications of RA were under reported on death certificates.

Mortality studies differ in mortality rates for both early and established RA, and to the degree to which cardiovascular disease contributes to any excesses [1, 12]. The different study designs (inception, prospective but not early RA or cross sectional studies, clinic-based or population cohorts) and variations in sample sizes, follow-up periods and geographic areas may explain some of these differences and why SMR vary so much. In addition, treatment practices have changed over the periods of study. The more recent studies reflect increased and early referral to secondary care, earlier use of DMARDs, especially methotrexate and greater emphasis on tighter disease control. Increasing awareness of risk management for comorbidities may be another factor [1, 8]. There have been relatively fewer reports from inception cohorts. In the UK, the earliest, started in the 1960s, had good follow-up (15–25 yrs) and reported modest increases in mortality, but in only around 100 patients each [25, 26]. Mortality was increased and mainly due to CVD in studies from Stockport [6], from a community-based study of inflammatory polyarthritis in Norfolk [5], and now in our study of nine centres in central and southern England reported here. In contrast, RA inception cohorts started in the late 1980s in northern Europe reported no increase in mortality [2–4]. The latter two had comparable study designs to ERAS and also followed similar management practice of the 1980/90s, which included the early use of sequential or combined DMARDs. These cohorts were much smaller than ERAS, included only RA with <12 months symptoms and according to HAQ their disease was milder at baseline. These observations would tend to support a design rather than a treatment effect [7], although the latter remains a possibility.

In recent years, considerable attention has been drawn to the increase in cardiac mortality in RA because the inflammatory process itself may play an important role in the development of IHD [27]. Thus the possibility of better disease control could influence this outcome, although the results from observational studies vary [9, 28]. We have shown here that disease activity including ESR was increased during the first year in IHD deaths, but also in all other major categories of death except solid tumours. Inflammatory markers including C-reactive protein (CRP) have been identified as independent risk factors for IHD [29]. Controlling inflammation may well reduce the risk of IHD, despite therapies which themselves may contribute to atherogenesis. Steroid use was not a risk factor for cardiovascular mortality in this study, although it was for all-cause mortality (Table 4), but these patients did have more severe disease when started on treatment. Exposure to anti-tumour necrosis factor-{alpha} (TNF-{alpha}) and COX2 drugs was very low at time of study. There is a complicated relationship between inflammatory activity, disease severity and drug therapies used in RA with death from cardiovascular causes, which is not fully understood [1, 30, 31]. Our findings so far do highlight the increased cardiovascular mortality in both sexes, especially for IHD in the early years of RA, for which one possible explanation is inflammatory activity. The effect of drug therapies, complex and possible beneficial or adverse cardiovascular effects of steroids, DMARDs and NSAIDs and how these interact with genetic factors and classic risk factors like hypertension, smoking and weight are currently under investigation in a subgroup of the ERAS cohort [32].

Other causes of death increased in RA in the literature include infections, respiratory disorders, cancer, lymphomas and stroke [10, 33]. Although we report high-SMR for septicaemia and renal failure, these need careful interpretation as absolute numbers were small and 95% CI wide. Mortality from respiratory causes and especially PF (6% with secondary causes) was higher than other inception cohorts for reasons that are not clear. No associations with drug therapies were found, but disease activity measures like ESR were raised, as seen for all-cause mortality (Table 4). Clinical detection of PF is reported in <5% in RA, although high-resolution computerized tomography (HRCT) improves on this [34]. It was the third most common condition in a retrospective study of extra-articular RA (cumulative incidence 6.8%) [35]. The natural history of PF and RA has not been well-studied, although idiopathic PF is known to have a poor prognosis (median survival 3–5 yrs). Whether PF with RA has a better prognosis is debated, although there are differences in HRCT early on [36]. As PF was often recognized clinically early in disease, it raises the issue of pulmonary screening tests and consideration of more aggressive interventional approaches. RA is associated with a variety of interstitial lung conditions and concern is emerging regarding respiratory complications of DMARDs alone and in combination with anti-TNF-{alpha} therapy [37]. Our findings do not implicate methotrexate in the aetiology of PF, nor that it has an adverse effect on established disease, but clearly it should be used with caution in these clinical situations. Preliminary results from the British Society of Rheumatology (BSR) Biologics Register suggest that increased mortality from PF is probably not explained by anti-TNF-{alpha} therapy [38].

Our study design and small numbers prevents analysis of the relative effect of severity of RA and use of immunosuppressive drugs on the increased risk of developing NHL, although our results support the recent editorial summarizing the evidence for high-cumulative disease activity as a factor in the increase of NHL in RA [39].

DMARD effects are recorded prospectively in ERAS patients, and we have been careful to determine possible associations between DMARDs, comorbidity and mortality. No evidence for such fatalities, either on death certificates or from clinical records was found, similar to a large cross-sectional study, which did, however, report frequent fatal side effects from NSAIDs as well as corticosteroids [11]. In our study, eight patients died from upper gastrointestinal haemorrhages and perforations, all with long-term or recent exposure to NSAIDs. Deaths attributable to steroid use were uncertain and possibly under reported. The figures we report here will be valuable when the long-term safety of biological therapies undergoes evaluation.

Early predictive factors of all-cause mortality were men, elderly, unfavourable socio-economic status, worse functional and disease activity measures, steroid use in the 1st yr, and presence of extra articular features, nodules and coexistent medical conditions. These retained their predictive power, albeit reduced, for CVD, respiratory disorders and stroke. This supports reports that mortality is increased the more aggressive the RA [8]. We have previously reported the value of HAQ [15, 40, 41], especially at 1 yr, ESR and haemoglobin [19, 21] and socio-economic status [16] for functional, radiological, orthopaedic and socio-economic outcomes. HAQ has been found to predict premature mortality in RA [8, 14] and Callahan et al. [42] reported increasing mortality with increasing HAQ. We have previously reported that HAQ initially improves from baseline over the first few years of RA [15, 41], so our finding here that first year HAQ has greater prognostic value than at baseline for both all-cause mortality and CVD is important and could be useful for clinicians. All these measures could be easily collected in ordinary clinics to identify risk groups.

It is increasingly observed from the literature that comorbidity contributes an important role in RA outcomes [1, 12, 43, 44]. Presence of another disease process itself may not increase the severity of the RA nor change RA specific outcomes [43], but RA may worsen or hasten a comorbidity, as Pincus et al. [44] claim for IHD. Socio-economic status and certain patient groups are known to influence access to medical care. Women and the elderly appear less likely to receive potentially life-saving treatment for IHD [45, 46]. Admission rates for IHD in RA were not increased despite increased mortality from this [6]. In our study, of those who died of IHD, the clinical diagnosis had already been made by, or within 3 yrs of presentation in 32% of patients. In PF deaths, 50% had a clinical diagnosis in this time. The numbers of comorbid and extra-articular conditions within the first year of baseline were both predictors of mortality in this report. The results from these and our own study highlight the need by rheumatologists to treat active RA early and effectively [44], and identify those patients who are at additional risk from the coexistent conditions and treat them actively or with preventative measures accordingly. This applies especially to cardiovascular comorbidity [1, 12] and from our study should include pulmonary fibrosis.

Strengths of this study include the length of follow-up (maximum 18, median 9 yrs) in an inception cohort design. This does allow study of a wider spectrum of RA and is more likely to retain patients with definite but milder disease, some of whom achieve prolonged remission. This reflects more accurately the varied severity and prognosis in RA. Retrospective studies may well be biased towards the more severe patients, who would be more likely to be retained in the clinic setting, partly because of comorbidity. The ERAS centres include patients from a wide variation in socio-economic backgrounds in England [16], all being general hospitals rather than tertiary referral centres. These patients are likely to represent those of a typical secondary care setting followed from the late 1980s onwards, and treated with standard therapies of this era. There are problems with mortality studies because of the inherent inaccuracy of death certificates and the under reporting of chronic conditions [22]. This is borne out by our finding that only 18% of the certificates recorded RA, and highlights the need for better training in death certification. However, we had near complete death certification, and were able to confirm the majority of death certificates by cross-reference with the medical case notes and ERAS database. Other possible concerns include left-censorship and case selection, because it is a hospital clinic-based study, although one UK community-based study of inflammatory polyarthritis reported similar findings [5]. Since the patients were not randomized, it has not been possible to account for treatment effects. Except in pulmonary fibrosis and lymphoma, no clear cut differences were observed in number of and time to the first DMARD. From our results it appears unlikely that DMARD or steroid use could explain the worse survival for IHD.

In summary, RA carries a modest but definite increased risk of mortality in the first seven years of disease, and is associated with markers of disease severity and function apparent within the first year of RA. Further consideration needs to be given to tighter disease control in early RA, and to early diagnosis and management of comorbidities, especially IHD and pulmonary fibrosis.


   Acknowledgements
Top
 Abstract
 Introduction
 Patients, materials and methods
 Results
 Discussion
 Acknowledgements
References
 
Dr P. Davies, Clinician, L. Hill, Metrologist—Chelmsford; Dr A. Gough, Dr J. Devlin and Prof. P. Emery, Clinician, L. Waterhouse, Metrologist—Birmingham; Dr P. Prouse, Clinician, C. Boys and A. Kent, Metrologist—Basingstoke; Dr D. James, Clinician, H. Tait, Metrologist—Grimsby; Dr P. Williams, Clinician, D. White, Metrologist—Medway; Dr J. Dixey, Clinician, H. Dart, Metrologist—Oswestry; Dr A. Young, Clinician, A. Seymour, Metrologist—St Albans; Dr J. Winfield, Clinician—Sheffield; Dr N. Cox, Clinician, S. Stafford, Metrologist—Winchester; ERAS Coordinator: Mrs Cathy Mayes, ARC Academic Secretary.

The authors have declared no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Patients, materials and methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Hall FC and Dalbeth N. (2005) Disease modification and cardiovascular risk reduction: two sides of the same coin. Rheumatology 44:1473–82.[Abstract/Free Full Text]
  2. Peltomaa R, Paimela L, Kautiainen H, Leirisalo-Repo M. (2002) Mortality in patients with rheumatoid arthritis treated actively from time of diagnosis. Ann Rheum Dis 61:889–94.[Abstract/Free Full Text]
  3. Lindqvist E and Eberhardt K. (1999) Mortality in rheumatoid arthritis patients with disease in the 1980s. Ann Rheum Dis 58:1–14.[Free Full Text]
  4. Kroot EJA, van Leuuwen MA, van Rijswijk MH, et al. (2000) No increased mortality in patients with rheumatoid arthritis: up to 10 years follow-up from disease onset. Ann Rheum Dis 59:954–8.[Abstract/Free Full Text]
  5. Goodson NJ, Wiles NJ, Lunt M, Barrett EM, Silman A, Symmons DPM. (2002) Mortality in early inflammatory polyarthritis. Arthritis Rheum 46:2010–19.[CrossRef][Web of Science][Medline]
  6. Goodson N, Marks J, Lunt M, Symmons D. (2005) Cardiovascular admissions and mortality in an inception cohort of patients with rheumatoid arthritis with onset in the 1980s and 1990s. Ann Rheum Dis 64:1595–601.[Abstract/Free Full Text]
  7. Ward M. (2001) Recent improvements in survival in patients with rheumatoid arthritis: better outcomes or different study designs? Arthritis Rheum 44:1467–9.[CrossRef][Web of Science][Medline]
  8. Pincus T and Callahan L. (1986) Taking mortality in rheumatoid arthritis seriously – predictive markers, socioeconomic status and comorbidity. J Rheumatol 13:841–5.[Web of Science][Medline]
  9. Galindo-Rodriguez G, Avira Zubrieta J, Russell A, Suarez Almazor M. (1999) Disappointing long term results with disease modifying anti rheumatic drugs. A practice based study. J Rheumatol 16:2337–43.
  10. Symmons DP, Jones MA, Scott DL, Prior P. (1998) Long term mortality outcome in patients with rheumatoid arthritis: early presenters continue to do well. J Rheumatol 25:1072–7.[Web of Science][Medline]
  11. Myllykangas-Luosujarvi R, Aho K, Isomaki H. (1995) Death attributed to antirheumatic medication in a nationwide series of 1666 patients with rheumatoid arthritis who have died. J Rheumatol 22:2200–2.[Web of Science][Medline]
  12. Boers M, Dijkmans B, Gabriel S, Maradit-Kremers H, O'Dell J, Pincus T. (2004) Making an impact on mortality in rheumatoid arthritis. Arthritis Rheum 50:1734–9.[CrossRef][Web of Science][Medline]
  13. Chehata JC, Hassell AB, Clark SA, et al. (2001) Mortality in rheumatoid arthritis: relation to single and composite measures of disease activity. Rheumatol 40:447–52.[Abstract/Free Full Text]
  14. Yelin E, Traupin L, Wong B, Rush S. (2002) The impact of functional status and change in functional status on mortality over 18 years among persons with rheumatoid arthritis. J Rheumatol 29:1851–7.[Abstract/Free Full Text]
  15. Young A, Dixey J, Cox N, et al. (2000) How does functional disability in early Rheumatoid Arthritis affect patients and their lives? Results of 5yr follow up in 781 patients from the Early RA Study (ERAS). Rheumatology 39:603–11.[Abstract/Free Full Text]
  16. Young A, Wilkinson P, Talamo J, et al. (2000) Socio-economic factors in the presentation and outcome of early rheumatoid arthritis. Lessons for the health service? Ann Rheum Dis 59:794–9.[Abstract/Free Full Text]
  17. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries J, Cooper NS. (1988) The ARA 1987 revised criteria for classification of rheumatoid arthritis. Arthritis Rheum 31:315–24.[Web of Science][Medline]
  18. Fries J, Spitz P, Young D. (1982) Dimensions of health outcomes: the health assessment questionnaire, disability and pain scales. J Rheumatol 9:789–93.[Web of Science][Medline]
  19. Dixey J, Solymossy C, Young A. (2004) Is it possible to predict radiological damage in early RA. J Rheumatology 31:Suppl 69, 48–54.[Web of Science]
  20. van der Heijde DMPM, van Hoof MA, van Rile PLCM, et al. (1990) Judging disease activity in clinical practice in rheumatoid arthritis: first step in the development of a disease activity score. Ann Rheum Dis 49:916–20.[Abstract/Free Full Text]
  21. James D, Young A, Dixey J, et al. (2004) Orthopaedic intervention in early rheumatoid arthritis. Occurrence and predictive factors in an inception cohort of 1029 patients followed for 5 years. Rheumatology 43:369–76.[Abstract/Free Full Text]
  22. Smith AE and Hutchins GM. (2001) Problems with proper completion and accuracy of the cause-of-death statement. Arch Intern Med 161:277–84.[Abstract/Free Full Text]
  23. Joint working group of British Society of Rheumatology, & Royal College of Physicians: guidelines and audit measures for the specialist supervision of patients with rheumatoid arthritis. J Roy Coll Phys Lond (1992) 26:76–82.[Web of Science][Medline]
  24. http://www.statistics.gov.uk/STATBASE/Expodata/Spreadsheets/D7131.xls.
  25. Corbett M, Young A, Dalton D, Meadows G, Silman S, Shipley M. (1993) Factors predicting death, survival and functional outcome in a prospective study of early Rheumatoid Disease over fifteen years. Br J Rheumatol 32:717–23.[Abstract/Free Full Text]
  26. Reilly P, Cosh JA, Maddison PJ, Reasker JJ, Silman A. (1990) Mortality and survival in rheumatoid arthritis: a 25-year study of 100 patients. Ann Rheum Dis 49:363–9.[Abstract/Free Full Text]
  27. Manzi S and Wasko MCM. (2000) Inflammation-mediated rheumatic diseases and atherosclerosis. Ann Rheum Dis 59:321–5.[Free Full Text]
  28. Krause D, Schleusser B, Herborn G, Rau R. (2000) Response to methotrexate treatment is associated with reduced mortality in patients with severe Rheumatoid Arthritis. Arthritis Rheum 43:14–21.[CrossRef][Web of Science][Medline]
  29. Ridker P, Cushman M, Stampfer M, et al. (1997) Inflammation, aspirin and the risk of cardiovascular disease in apparently healthy men. N Engl J Med 336:973–9.[Abstract/Free Full Text]
  30. Bacon PA and Townsend JN. (2001) Nails in the coffin: increasing evidence for the role of rheumatic disease in the cardiovascular mortality of rheumatoid arthritis. Arthritis Rheum 44:2707–10.[CrossRef][Web of Science][Medline]
  31. Kitas GD and Erb N. (2003) Tackling ischaemic heart disease in rheumatoid arthritis. Rheumatology 42:607–13.[Free Full Text]
  32. Mattey D, Thompson W, Ollier W, Koduri G, Williams P, Young A. (2004) Association of the HLA-DRB1 shared epitope with mortality in rheumatoid arthritis. Arthritis Rheum 50:Suppl, 1280.
  33. Wolfe F, Freundlich B, Straus WL. (2003) Increase in cardiovascular and cerebrovascular disease prevalence in rheumatoid arthritis. J Rheumatol 30:36–40.[Abstract/Free Full Text]
  34. Kelly CA. (1993) Rheumatoid arthritis, classic lung disease. Bailliere's Clin Rhematology 7:1–16.[CrossRef]
  35. Turesson C, O'Fallon WM, Crowson CS, Gabriel SE, Matteson EL. (2003) Extra-articular disease manifestations in rheumatoid arthritis: incidence trends and risk factors over 46 years. Ann Rheum Dis 62:722–7.[Abstract/Free Full Text]
  36. Rajasekaran BA, Shovlin D, Lord P, Kelly CA. (2001) Interstitial lung disease in patients with rheumatoid arthritis: a comparison with cryptogenic fibrosing alveolitis. Rheumatology 40:1022–5.[Abstract/Free Full Text]
  37. Oster AJK, Chilvers ER, Somerville MF, et al. (2006) Pulmonary complications of infliximab therapy in patients with rheumatoid arthritis. J Rheumatol 33:622–8.[Abstract/Free Full Text]
  38. BSR Report. Anti-TNF{alpha} agents in RA ( June 2005) Biologics Register.
  39. Weyand CM, Goronzy JJ, Kurtin PJ. (2006) Lymphoma in Rheumatoid Arthritis: an immune system set up for failure. Arthritis Rheum 54:685–9.[CrossRef][Web of Science][Medline]
  40. Kobelt G, Jonsson L, Lindgren P, Young A, Eberhardt K. (2002) Modelling the progression of rheumatoid arthritis: a two-country model to estimate costs and consequences of RA. Arthritis Rheum 46:2310–9.[CrossRef][Web of Science][Medline]
  41. Bansback NJ, Young A, Brennan A, Dixey J. (2006) A prognostic model for functional outcome in early rheumatoid arthritis. J Rheumatology 33:1029–38.[Free Full Text]
  42. Callahan LF, Pincus T, Huston JW 3rd, Brooks RH, Nance EP Jr, Kaye JJ. (1997) Measures of activity and damage in rheumatoid arthritis: depiction of changes and prediction of mortality over five years. Arthritis Care Res 10:381–94.[Web of Science][Medline]
  43. Gabriel SE, Crowson CS, O'Fallon WM. (1999) Comorbidity in arthritis. J Rheumatol 26:2475–945.[Web of Science][Medline]
  44. Pincus T, Gibofsky A, Weinblatt ME. (2002) Urgent care and tighter control of rheumatoid arthritis as in Diabetes and hypertension: better treatments but a shortage of rheumatologists. Arthritis Rheum 46:851–4.[CrossRef][Web of Science][Medline]
  45. Clarke KW, Gray D, Keating NA, Hampton JR. (1994) Do women with acute myocardial infarction receive the same treatment as men? Br Med J 309:563–6.[Abstract/Free Full Text]
  46. Huxley R, Barzi F, Woodward M. (2006) Excess risk of fatal coronary heart disease associated with diabetes men and women: meta-analysis of 37 prospective cohort studies. Br Med J 332:73–6.[Abstract/Free Full Text]
Submitted 10 April 2006; revised version accepted 25 May 2006.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Rheumatology (Oxford)Home page
U. Bergstrom, L. T. H. Jacobsson, and C. Turesson
Cardiovascular morbidity and mortality remain similar in two cohorts of patients with long-standing rheumatoid arthritis seen in 1978 and 1995 in Malmo, Sweden
Rheumatology, December 1, 2009; 48(12): 1600 - 1605.
[Abstract] [Full Text] [PDF]

Home page
 The Journal of RheumatologyHome page
T. KURODA, Y. WADA, D. KOBAYASHI, S. MURAKAMI, T. SAKAI, S. HIROSE, N. TANABE, T. SAEKI, M. NAKANO, and I. NARITA
Effective Anti-TNF-{alpha} Therapy Can Induce Rapid Resolution and Sustained Decrease of Gastroduodenal Mucosal Amyloid Deposits in Reactive Amyloidosis Associated with Rheumatoid Arthritis
J Rheumatol, November 1, 2009; 36(11): 2409 - 2415.
[Abstract] [Full Text] [PDF]

Home page
 Rheumatology (Oxford)Home page
C. Meune, E. Touze, L. Trinquart, and Y. Allanore
Trends in cardiovascular mortality in patients with rheumatoid arthritis over 50 years: a systematic review and meta-analysis of cohort studies
Rheumatology, October 1, 2009; 48(10): 1309 - 1313.
[Abstract] [Full Text] [PDF]

Home page
 The Journal of RheumatologyHome page
C. GARCIA-GOMEZ, J. M. NOLLA, J. VALVERDE, J. A. GOMEZ-GERIQUE, M. J. CASTRO, and X. PINTO
Conventional Lipid Profile and Lipoprotein(a) Concentrations in Treated Patients with Rheumatoid Arthritis
J Rheumatol, July 1, 2009; 36(7): 1365 - 1370.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow CME/CE:
Take the course for this article:
 Rheumatology Fourth Quarter 2006 Quiz
Right arrow All Versions of this Article:
46/2/350    most recent
kel253v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Young, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Young, A.
Related Collections
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?