You are here

How does established rheumatoid arthritis develop, and are there possibilities for prevention?

Best Practice & Research Clinical Rheumatology

Abstract

Established rheumatoid arthritis (RA) is a chronic state with more or less joint damage and inflammation, which persists after a phase of early arthritis. Autoimmunity is the main determinant of persistence. Although the autoimmune response is already fully developed in the phase of early arthritis, targeted treatment within the first months produces better results than delayed treatment. Prevention of established RA currently depends on the success of remission-targeted treatment of early disease. Early recognition is aided by the new criteria for RA. Further improvement may be possible by even earlier recognition and treatment in the at-risk phase. This requires the improvement of prediction models and strategies, and more intervention studies. Such interventions should also be directed at modifiable risk factors such as smoking and obesity. The incidence of RA has declined for decades in parallel with the decrease of smoking rates; however, a recent increase has occurred that is associated with obesity.

Keywords: Preclinical, Early and established rheumatoid arthritis, Undifferentiated arthritis, Prediction, Risk factors, Prevention.

Introduction

The concept of “established rheumatoid arthritis” (RA) appears to be clear for the clinician. The picture arises of a patient with a “longstanding disease” that has caused a certain amount of joint and comorbid damage, and it remains in a fixed state with more or less active disease. The counterpart is the concept of “early (rheumatoid) arthritis,” a more fluid state of recent synovitis where everything is still possible, including spontaneous or induced complete remission. Although the contrasting states are clear, the transition between them is gradual and less well defined. It is reasonable to expect that causative factors for RA also influence the course of the disease, in this case the progression from early to established RA. For example, anti-citrullinated protein antibodies (ACPAs) are associated with both the risk of developing RA and the risk of a severe, unremitting course of RA.

In this chapter, we review risk factors for the development of early RA and for the transition to established RA. The concept of undifferentiated arthritis (UA) as a separate entity in a continuum from health to RA is undergoing changes due to new definitions. Finally, we focus on efforts to prevent RA from occurring (primary prevention) or from progressing from UA to RA (secondary prevention).

Apart from the uncertainty over the transitions between the different phases of RA, there is also considerable uncertainty over the question of whether RA is a modern or an ancient disease. The name RA first appears in the medical literature in 1876[1], and the first unequivocal description of RA dates from 1631 [2]. There is a scarcity of descriptions of the disease in Europe between 1700 and 1900 [3]. This, combined with the fact that the evidence of erosions compatible with RA has been found in ancient skeletons in North America, but not in Europe or the Middle East [4], has led to the suggestion that RA may be a communicable disease brought to the Old World after contact with the New World [1]. A good candidate factor for such an effect may be tobacco smoking, a habit imported from the New World that increased tremendously in the late 19th century followed by a decrease in the second half of the 20th century, roughly in parallel with changes in the incidence of RA.

Risk factors for RA development

The risk of developing RA is determined by genetic susceptibility combined with environmental factors [6] and [5]. Certain environmental factors operate already early in life, and they may help to lay the foundation for autoimmunity. In a large part of those later developing seropositive RA, there is a phase of autoimmunity and subclinical inflammation, during which another transient cause of inflammation such as an infection is thought to trigger the onset of clinically apparent disease [7].

In the following, we present a short overview of genetic and environmental risk factors for RA, with a focus on recent publications. Due to the preclinical phase that many later patients go through, biomarkers of autoimmunity and inflammation can also be used as risk factors or predictors of disease. Recently, several prediction models have been constructed using information from various cohorts of persons at risk of RA.

Genetic risk factors

Approximately 65% of RA risk has been shown to be heritable, and >100 risk loci are now known. Most of these confer a low risk, and together they explain approximately 16% of total susceptibility [8]. It has become clear that ACPA-negative and ACPA-positive disease have a genetically different background [5] and [9]. The major histocompatibility complex (MHC) class II, DR beta 1 (human leukocyte antigen (HLA)-DRB1) alleles play a central role in the genetic risk of “seropositive” (ACPA and/or rheumatoid factor (RF)-positive) RA, mainly in patients who are ACPA positive [5]. Multiple alleles from this complex are associated with RA, which all share a region of similarity termed the shared epitope (SE). Besides these, several non-HLA genes have been identified. Most of the evidence comes from genome-wide association studies (GWAS) [10]. Until now, most GWAS investigating RA have been performed in seropositive individuals with a European background [10]. Recently, a review was published of specific genetic risk in Asian populations [11]. Although most single-nucleotide polymorphisms (SNPs) have the same effect sizes for developing RA in European and Asian people, some differences are found, mainly for PADI4 and PTPN22, which are more strongly associated with RA in Asian populations. Furthermore, the genetic risk in certain high-risk populations of North American Natives has been described, showing that most of the risk is conferred by a high prevalence of the SE in this population [12]. Evidence is lacking for many other populations. However, it seems that common SNPs found in ACPA-positive individuals with a European background also make individuals with a different ethnicity more susceptible to developing RA [9]. This was also shown, to a lesser extent, for ACPA-negative patients.

A disadvantage of the GWAS method is that the implicated SNPs are not necessarily causally linked to the development of RA itself. Moreover, until now, they cannot be used for individual prediction because of their low effect sizes. Most have odds ratios (ORs) for developing ACPA-positive RA of 1.1–1.2, with a few exceptions having an individual OR of around 2.0 (e.g., locus 1p13 on the PTPN22 gene, and 6p21 on the HLA*04 genes) [9].

Genetic risk scores

Given the many involved genes with small effect sizes, genetic risk scores (GRS) have been developed to help individual prediction of RA by adding up multiple validated genetic risk loci. In the next step, these can be combined with environmental factors in prediction models. GRS for RA usually take both the number of alleles an individual possesses and the effect size of the alleles into account. Published GRS prediction models for RA, some including environmental factors, are presented in Table 1. In the case of multiple publications from one cohort, only the last publication is shown.

Table 1

Prediction models for development of rheumatoid arthritis using genetic, clinical, and behavioral (smoking) data.

 

Reference Cohort; variables Numbers Results
van der Helm

2010 [15]
Early arthritis cohort, the Netherlands

Genetic loci: HLADRB1 SE alleles, 11 SNPs

Clinical parameter: smoking
570 UA Model with genetic loci combined: AUC 0.54 (CI: 0.48–0.59)

Genetic loci and clinical parameter: AUC 0.89 (CI: 0.86–0.95)
Kurreeman

2011 [9]
EHR cohort, USA

Genetic loci: 1 HLA allele, 29 SNPs
1552 ACPA + RA

1504 controls
European ancestry: AUC 0.71 (CI: 0.68–0.73)

African ancestry: AUC 0.63 (CI: 0.56–0.70)

East Asian ancestry: AUC 0.74 (CI: 0.59–0.89)

Hispanic ancestry: AUC 0.66 (CI: 0.56–0.76)
Scott

2013 [16]
WTCCC and UKRAGG, UK

Genetic loci: 25 HLA alleles, 31 SNPs

Clinical parameter: smoking
WTCCC/UKRAGG:

1516/2623RA

1647/1500 controls
HLA-SNP model WTCCC: AUC 0.80 (CI: 0.78–0.81)

UKRAGG: AUC 0.76 (CI 0.72–0.79)

HLA-SNP-smoking model WTCCC: AUC 0.84 (CI: 0.81–0.87)

UKRAGG: AUC 0.86 (CI: 0.80–0.91)
Yarwood

2015 [17]
Immunochip Consortium

Validation in CORRONA

Genetic loci: 45 SNPs, imputed amino acids at HLA-DRB1 [10], [70], and [73] and HLA-DPB1 (position 9) HLA-B (position 9)

Clinical parameters: gender, smoking
Immunochip/

CORRONA:

11,366/2206 RA

15,489/1863 controls
Genetic loci combined – Immunochip:

OR 2.0 (CI: 2.0–2.0), AUC 0.74, sens 35%, spec 91%

Genetic loci combined – CORRONA:

OR 2.0 (CI: 1.9–2.1), AUC 0.72, sens 30%, spec 92%

Addition of smoking improved the AUC to 0.80, without improving sens and spec
Sparks

2015 [13]
NHS, USA (only females)

Validation in EIRA, Sweden

Genetic loci: 8 HLA alleles, 31 SNPs

Clinical parameters: family history, epidemiologic factors, HLA-smoking interaction
NHS/EIRA:

381/1752 RA

410/1361 controls
Genetic loci combined – NHS: AUC 0.62 (CI: 0.58–0.67)

Genetic loci and clinical parameters: AUC 0.74 (CI: 0.70–0.78)

Genetic loci combined – EIRA: AUC 0.58 (CI: 0.55–0.60)

Genetic loci and clinical parameters: AUC 0.69 (CI: 0.67–0.72)

ACPA = anti-citrullinated protein antibodies, AUC = area under the receiver operating characteristic curve, CI = confidence interval (excluding 0.50 means statistically significant predictive value), CORRONA = Consortium of Rheumatology Researchers of North America registry, EHR = Electronic Health Records, EIRA = Epidemiologic Investigation of Rheumatoid Arthritis, HLA = human leukocyte antigen, NHS = Nurses׳ Health study, RA = rheumatoid arthritis, SE = shared epitope, sens = sensitivity, SNPs = single-nucleotide polymorphisms, spec = specificity, UA = undifferentiated arthritis, UK = United Kingdom, UKRAGG = RA Genetics Group Consortium UK, USA = United States of America, WTCCC = Wellcome Trust Case–Control Consortium.

In summary, these studies show ORs of different models of around 2.0, and a wide variation of area under the curves (AUC) from a low value of 0.54 to a high value of 0.89 (with the highest values also including clinical parameters). A relatively high specificity for identifying individuals at risk (75–90%) is unfortunately accompanied by a very low sensitivity (30–45%). Therefore, apart from the disadvantage of its high cost, genetic risk prediction is thus still not precise enough to be used in current clinical practice, even though more and more genetic loci are being discovered. However, a recent study shows that a GRS plus environmental factors in family members of RA patients provides enough discrimination to enable the selection of high-risk subjects for intervention studies [13]. To support future research, Nagai et al. have made the open access database “RAvariome,” which was developed to list all RA-associated genetic variants and to check reproducibility over different ethnicities [14]. Their website (http://hinv.jp/hinv/rav/) also provides a “genetic risk predictor,” which gives the lifetime risk on developing RA per individual. Unfortunately, as with the different GRS, the timing of RA development cannot be predicted by using this database.

Environmental and behavioral factors

New risk factors for RA are being found, and systematic reviews have reevaluated established or controversial risk factors. The present situation is summarized in Table 2[6] and [5].

Table 2

Environmental risk factors for development of rheumatoid arthritis.

 

Validated risk factor Comment
Traditional risk factors
Family history 65% of RA risk is thought to be heritable
Female gender Females have 2–4 times higher risk
Aging Onset usually around sixth decade of life
Smoking One of the main risk factors, dose-dependent risk effect
Lower education level Possibly linked to lifestyle or certain occupations
Silica exposure Industrial exposure: mining, construction, agriculture, electronics
Pregnancy Increased risk in the year after childbirth
High birth weight >4 kg
Fish oil, olive oil Protective effect; believed to have anti-inflammatory properties
Comorbid conditions Diabetes mellitus type 1 and 2, inflammatory lung diseases, dyslipidemia. Schizophrenia (protective)
New risk factors or new information on known risk factors
Suggested risk factor Comment
Sugar-sweetened soda May induce obesity, insulin resistance, and inflammation
Obesity Conflicts about whether it increases risk of both seronegative and seropositive RA
Physical activity Associated with less and milder RA
Infections Frequent infections may predispose, although some contradict this finding, no specific pathogens causally linked to RA
Sleep disorders The non-apnea types show higher RA rates later on
Autoimmune thyroid disease Subsequent RA seems more frequent
Tetanus vaccination One study reported an increased risk of <1 year after vaccination
Recent reviews
Alcohol consumption Protective effect, mainly for seropositive RA
Fish consumption No significant relationship with RA development
Coffee consumption Coffee consumption gives a higher risk of seropositive RA
Reproductive/hormonal factors Controversy continues
Use of oral contraceptives No significant relationship with RA development
Geographic area RA is more prevalent in Northern countries as compared to countries near the equator
Inconclusive/conflicting results
Age at menarche and menstrual cycles, parity and age at first childbirth, breastfeeding, oral contraceptives, postmenopausal hormone use
Periodontitis
Previous blood transfusion
Consumption of coffee and tea, red meat
Ultraviolet B exposure and vitamin D levels, antioxidant, and trace element intake, exposure to toxic elements, and air pollution
Silicone implants

One controversial factor was alcohol consumption, which was shown earlier to be protective, even in small quantities [6]. Two reviews [19] and [18] confirmed this protective effect, although the effect size was small (summary ORs of 0.78 and 0.86, respectively), and one only found the effect in individuals later developing ACPA-positive RA. A nonlinear relationship was found in the dose–response meta-analysis. Lu et al. confirmed the finding that the association between alcohol and less development of RA was stronger in seropositive women [20]. Second, fish consumption (number of servings per week) was addressed in a systematic review [21]. This dose–response meta-analysis showed an inverse association between fish consumption of one to three servings per week versus never consumption and the risk of RA, with a relative risk (RR) of 0.76 (CI: 0.57–1.01) (not statistically significant). Third, the meta-analysis of the consumption of coffee and tea showed that only the use of coffee was related to RA development. The RR of total coffee intake was 1.3 for developing seropositive RA [22]. Fourth, much controversy exists about reproductive factors and sex hormone levels in both women and men in relation to RA. This holds true for menstrual cycle, parity, pregnancy, age at menopause, hormone use, and male testosterone levels. More recently published articles still show varying results, as also reflected in a recent review [23]. A publication that was not included in this review reported that pregnancy complications, namely preeclampsia, and poor self-rated health during pregnancy were related to a higher risk of later RA [24]. Baydoun et al. investigated reproductive history and postmenopausal RA, but only found menopausal age below 40 years to confer the risk of RA after menopause [25]. Moreover, no significant relationship could be found between the use of oral contraceptives and the development of RA in two reviews incorporating a total of 28 studies [26]. Two other publications produced conflicting results of testosterone levels in men. One did not show a difference between testosterone levels in pre-RA cases versus controls [27], and the other found lower testosterone levels before the diagnosis of RF-negative RA [28]. Finally, a recent article publishes information about geographic area and RA incidence, and prevalence and mortality rates [29]. Although the focus was more on the burden of disease, the authors do present data showing that RA is more prevalent in Northern countries as compared with countries near the equator.

More focus has been directed lately toward different dietary components and the risk of RA development. Already in 2004, a review suggested the possible role of diet, but it could not quantify the risk [30]. Recent publications have focused more on different types of diet. No significant relations could be found for a Mediterranean-type diet [31], a carbohydrate-restricted diet [31], and sodium intake (which only led to a significantly increased risk when combined with smoking) [32]. Interestingly, sugar-sweetened soda consumption of ≥1 serving/day (compared with <1 serving/month) was significantly related to the development of both seropositive and late-onset-seropositive RA (age after 55 years) in women with hazard ratios (HRs) of 1.63 and 2.62, respectively (corrected for other lifestyle components) [33]. The amount of added sugar in these drinks may contribute to the pathogenesis of RA by inducing obesity, insulin resistance, and inflammation. In light of the recent rise in obesity prevalence and RA incidence (see subsequently), this might be an important point of interest, and it suggests a possibility to intervene in the at-risk subjects.

Most environmental risk factors seem to be more related to seropositive than to seronegative RA. However, obesity was shown to be related mainly to seronegative RA in most publications [5], [35], and [34], with only one report also showing a higher risk of ACPA-positive RA in women [36]. All underline the importance of obesity as a risk factor. As obesity may be in part related to little exercise, it was hypothesized that regular exercise protects against RA. This was confirmed by two studies, which showed regular physical activity indeed leads to less RA, and, if it did occur, patients presented with milder disease [38] and [37].

Besides obesity, several other comorbidities have since long been linked to the development of RA, such as diabetes mellitus and schizophrenia. Recently, two other diseases have been investigated. Sleep disorders (without sleep apnea) had an HR of 1.45, [39] and autoimmune thyroid disease was seen more frequently in RA cases than in controls (together with more thyroxin substitution before RA development) [40].

The exact mechanism as to how systemic autoimmunity advances into local inflammation in the joints still needs to be further investigated [7]. It is thought that infections may trigger the onset of clinically apparent disease. Some recent publications have focused on the presence of infections before RA onset and specific pathogens. Prior infection-related medical visits and bacterial colonization are shown to predispose the development of RA, mostly in the year preceding diagnosis [42] and [41]. However, another study found a decreased risk of gastrointestinal and urinary tract infections and no relation for other infections [43]. So far, no specific pathogen could be quantitatively linked to RA development [6]. Regarding the related subject of vaccination, only one out of many studies reported an increased risk of <1 year after tetanus vaccination [44].

Finally, de Roos et al. investigated living in the proximity to traffic, ambient air pollution, and community noise. They found a higher risk of RA when living within 50 m from the highway (OR 1.37), but they could not relate this to ambient air pollution or noise [45]. In this study, it is good to note that it was not possible to correct for confounding factors such as low social economic status, nonwhite race, and smoking. Therefore, the results may be biased. Besides, another study could also not find a relationship between air pollution and the development of RA [46].

A distinction was made between traditional risk factors, meaning generally accepted risk factors before at least 4 years ago (most already presented in previous edition of Best Practice & Research), and the ones receiving more attention over the past years and generating new insights.

Gene–environment interactions and environmental factors influencing each other

A strong interaction exists between smoking and genetic background (namely HLA-DRB1 alleles) [10]. Besides, smoking interacts with autoantibody-positive status, gender (higher influence in males), and consumption of dietary sodium [5] and [32] to a lesser extent. Furthermore, adding positive family history of RA to genetic risk models increases the predictive capacity. However, in general, the gene–environment interactions add too little information to the models to be of clinical use [13].

Autoimmunity and biomarkers

Approximately two-thirds of RA patients test positive for RF and/or ACPA at diagnosis, underlying their importance in this disease. Other antibodies preceding and predicting a diagnosis of RA, independent of RF and ACPA status, are anti-carbamylated protein antibodies and anti-peptidyl arginine deiminase type 4 antibodies [6]. The discovery of new related autoantibody systems may in the future give more insight into the pathogenesis of RA.

Other blood-based biomarkers such as acute phase reactants or cytokines were not found to have predictive capacity for RA [6].

Clinical prediction models

Quantifying progression to RA with genetic modeling alone is not ready for clinical use, as we have shown earlier. Several studies have taken a different approach by using a combination of clinical characteristics, symptoms, and sometimes imaging findings. The resulting prediction rules are summarized in Table 3. Validation is still needed for all models. With this restriction, they can be useful to inform persons with musculoskeletal symptoms about their risk of arthritis/RA, especially in the presence of RA-related antibodies.

Table 3

Clinical prediction models for development of rheumatoid arthritis.

 

First author and year (ref) Cohort; variables Numbers Results
van de Stadt

2013 [47]
Seropositive arthralgia patients

Prediction rule variables: alcohol nonuse, family history, several symptoms, autoantibody status
Arthralgia: 374 (131 developed arthritis) Prediction rule: AUC 0.82 (CI: 0.75–0.89)

Intermediate vs. low risk group: HR 4.52 (CI: 2.42–8.77)

High vs. low risk group: HR 14.86 (CI: 8.40–28)
de Hair

2013 [48]
Seropositive arthralgia patients

Predictive variables: smoking, BMI
Arthralgia: 55 (15 developed arthritis) Smoking (ever vs. never) and risk of RA: HR 9.6 (CI: 1.3–73)

Obesity (BMI ≥25 vs. <25) and risk of RA: HR 5.6 (CI: 1.3–25)
Lahiri

2014 [49]
European Prospective Investigation of Cancer, UK

Prediction rule variables: alcohol use, smoking, occupation, BMI, diabetes mellitus, parity
Total participants: 25,455

(184 developed IP, 138 developed RA)
Pack-years smoking in men and risk of IP: HR 1.21 (CI: 1.08–1.37)

Seropositive in men and risk of IP: HR 1.24 (CI: 1.10–1.41)

Having DM (I or II) and risk of IP: HR 2.54 (CI: 1.26–5.09)

Alcohol and risk of IP (per unit/day): HR 0.36 (CI: 0.15–0.89)

Overweight and risk of seronegative IP: HR 2.75 (CI: 1.39–5.46)

Parity ≥2 and risk of IP: HR 2.81 (CI: 1.37–5.76)

Breastfeeding and risk of IP: HR 0.66 (CI: 0.46–0.94)
Rakieh

2014 [50]
ACPA-positive arthralgia patients

Prediction rule variables: several symptoms, high-positive ACPA, positive ultrasound power

Doppler signal
Arthralgia: 100 (50 developed RA) Power Doppler model: Harrell׳s C 0.67 (CI: 0.59–0.74)

Progression to IA:

Low risk (0 points) 0%

Moderate risk (1–2 points) 31%

High risk (≥3 points) 62%

ACPA = anti-citrullinated protein antibody, AUC = area under the receiver operating characteristic curve, BMI = body mass index, CI = confidence interval, DM = diabetes mellitus, EIRA = Epidemiological Investigation of RA, HR = hazard ratio, IA = inflammatory arthritis, IP = inflammatory polyarthritis, NHS = Nurses׳ Health Study, OR = odds ratio, RA = rheumatoid arthritis, VAS = visual analog scale. Reproduced from Turk et al., 2014 [7].

Changing incidence rates and mode of presentation of RA

In 1979, it was hypothesized that RA as a disease entity would disappear eventually [51]. Currently, more evidence exists of a pattern of rises and falls over the decades. Over the first half of the 20th century, no data are available. Alamanos et al. summarized studies on incidence and prevalence rates of RA (according to the 1987 American College of Rheumatology (ACR) criteria) from the second half of the 20th century [52]. Two out of the three studies, which evaluated time trends of RA occurrence, reported a declining RA incidence of 15% and 47% in 1 and 4 decade(s), respectively (1980–1990 in MN, USA, and 1955–1994 in Finland). In Greece, the incidence remained stable between 1987 and 1995. Studies in Japan and of North American Natives in the USA have also noted a declining incidence of RA [54] and [53]. The decline in incidence combined with a shift toward higher age at the onset of disease has been attributed to a so-called birth cohort effect [55]. This is a term used in social science to describe characteristics of an area of study over time among individuals who are defined by certain early life influences. Following generations will benefit or be harmed by these influences of their ancestors, in this case leading to a decline in RA incidence. However, which specific risk factors would be implicated in the decline of the incidence has not been specified.

More recent studies in Denmark [56] and Minnesota [57] suggest that the incidence may be rising again, with annual increases of RA incidence of 6% (1995–2001) and 2.5% (1995–2007), respectively, remarkably only in women. However, in Finland, a further annual decline of 2% was seen for RF-positive RA over the period of 1980–2000 [55]. It was speculated that a combination of environmental changes leading to either increased risk or loss of protection plays a role in the increasing RA incidence found in the above-mentioned countries. As alluded to in the earlier text, obesity seems to be an important emerging risk factor of RA development. Crowson et al. linked the recent increase of obesity in the population to the higher incidence of RA [58]. It was calculated that an increase in obesity could explain 52% of the increase in the incidence of RA among women in the period 1995–2007. Furthermore, other factors may play a role, for example, lower doses of hormones in the oral contraceptives over the years, slower decline of smoking rates in women compared with men, and more vitamin D deficiency [57].

Another important note about changes in incidence rates over time is that the timing of the measurement and used RA criteria can vary between studies, and it also depends on the duration of the study period, mode of presentation, awareness of the disease by general practitioners, and the delay of referral after symptom onset. In the following, we describe two of these factors in more detail. First, the new ACR/European League Against Rheumatism (EULAR) 2010 criteria for RA (see subsequent discussion) are more sensitive than the earlier criteria, which will probably lead to earlier detection (and treatment) and thereby affect the measurement of incidence rates in the coming years [59]. Second, within Europe, the variation in the delay of first assessment of RA patients is substantial, with a median range of 16–38 weeks per center and a difference at its highest of 34% in seeing patients within 12 weeks of symptom onset [60]. This could partly explain differences of changes in incidence rates across European countries, and even less is known about such a variation outside Europe.

In conclusion, relevant trends are a steady decrease of worldwide RA incidence during the period 1955–1995, followed by a recent increase in at least Denmark and the USA, probably explained in part by changing environmental factors. Furthermore, factors such as differences in the use of RA criteria and differences in the awareness of RA across countries can affect the incidence rates over time.

UA, past and present

The term “UA” suggests that the condition in the patient concerned is in a stage of transition from an unspecified type of arthritis toward either RA, another arthritis-associated diagnosis, or spontaneous remission. The incidence of UA ranges from 41 (in Finland) to 149 (in Sweden) per 100,000 adults, and 13–54% of these patients will eventually develop RA, according to the 1987 ACR criteria [61]. In the past, the transition from UA to RA was equivalent to fulfilling the 1987 ACR criteria for RA [62] after a phase with arthritis in which these criteria were not yet fulfilled. In practice, this mainly applied to the progression from oligoarthritis to polyarthritis and/or the development of erosive disease, as other elements of the criteria set such as RF or nodules do not often appear in early arthritis, if not present at the first presentation [63]. Therefore, the transition from UA to RA could be viewed as the development of a more severe arthritis in inadequately controlled early RA, which made this an outcome of interest. The main predictor of the transition was the ACPA status of the patient [64].

The 2010 ACR/EULAR criteria for RA aim to increase sensitivity in early disease [65], which is mainly achieved by a focus on small joint involvement and serology. Thus, a patient with one swollen finger joint of 6 weeks duration and a high-titer ACPA will already classify as RA. The consequence is that the subgroup of UA in early arthritis patients is strongly reduced, and it is now composed mainly of seronegative (oligo-) arthritis. On average, these “2010 UA” patients will have a milder and more heterogeneous disease than “1987 UA” patients [66]. Although both the 1987 and 2010 criteria for RA are classification and not diagnostic criteria, the 2010 criteria were specifically developed for use in early disease, and they reflect the trend among clinicians to diagnose RA earlier and even in the presence of only a few involved joints.

Just as was the case with 1987 UA patients, a part of 2010 UA patients will remit and a part will go on to have a severe disease course. In a recent study of three early arthritis cohorts, the Leiden prediction rule (developed to predict 1987 RA in 1987 UA patients) and the ACPA status failed to predict the development of 2010 RA in 2010 UA patients [67]. New biomarkers are needed that can help to detect the 2010 UA patients at high risk of disease progression, so that they may be considered for more aggressive therapy than the remaining UA patients, for whom symptomatic treatment may be sufficient. An example is anti-CarP antibodies, which were shown to predict radiographic damage in early ACPA-negative RA patients [68]. Next to blood-based biomarkers, imaging modalities such as ultrasound or magnetic resonance imaging (MRI) may prove to be useful in this respect [70] and [69].

When does early RA become established RA?

This question gives rise to the suggestion that there is a difference between the pathology at the beginning of the disease and what is found later on, and that this distinction has clinical significance. In fact, this is closely related to the concept of a therapeutic “window of opportunity,” which states that treatment initiated at an early stage of the disease is more successful than when it is started later on. “Early” would mean that there is joint inflammation of recent onset, which may at this stage still resolve without further consequences or at least decrease to a barely detectable minimum, if treated adequately. “Established” on the other hand would mean the inflammation is there to stay, more or less pronounced, whatever intervention is applied. Moreover, the concept of “established” RA will generally include damage to the joints, and diverse comorbidities with their complications such as osteoporosis or cardiovascular disease, which may arise as a consequence of the ongoing inflammation.

To begin with, the pathology of RA does not suddenly start around the onset of clinical arthritis. RA-specific systemic autoimmunity as well as nonspecific subclinical inflammation occurs in concert on average 5 years before the onset of symptoms [72] and [71]. During the period of presymptomatic autoimmunity, there is a maturation of the immune response to citrullinated and carbamylated antigens, which is consistent with an increasing break of tolerance [73]. Thus, the number and levels of different ACPA specificities increase toward the onset of arthritis; however, there is no further increase once clinical arthritis has begun [74]. Accordingly, the number and type of ACPA specificities do not differ largely between early and late disease [75]. Anti-immunoglobulin G (IgG) antibodies or RF arise later and less frequently than ACPA, and they may continue to increase in prevalence after the onset of arthritis [74] and [76].

The synovial infiltrate of knee joints of RA patients that had not been clinically swollen before, nevertheless, showed chronic inflammation [77]. In animal models of RA, inflammation in joint pathological specimens precedes clinically detectable inflammation. Persons at an increased risk of RA have increased numbers of T-cells in their knee synovium even if they did not yet have knee symptoms [78], again suggesting that the transition to chronic inflammation takes place before the onset of clinically apparent arthritis. Once the symptoms begin, a higher number of recognized ACPA specificities are associated with a higher rate of transition to clinical arthritis [73]. This means that once a person notices the first symptoms of RA, the pathological immune response has matured to a large extent, but not completely.

Although the immunological driving processes of RA do not seem to differ between early and late RA, it is well known that better clinical results can be obtained by treating RA patients early and aggressively [79]. A recent analysis concluded that this window of opportunity starts to close 4 months after the onset of symptoms [80]. This implies that it is still possible during that period to interrupt certain processes perpetuating the chronicity of inflammation. One of these could be the total burden of inflammation, which builds up in the early clinical phase. It is conceivable that once a critical mass of inflammatory tissue has been reached, it is no longer possible to control it effectively. This theory is difficult to test, as there is no technique available at present, which can reliably test the total load of inflammatory tissue in a person.

Primary prevention of RA

The different stages of RA development offer opportunities for preventive interventions, varying from (primary) prevention of the development of arthritis in the at-risk phase to (secondary) prevention of progression from UA to RA or from early to established RA.

The list of risk factors for RA (Table 2) shows that there are several opportunities for lifestyle changes to help prevent RA. Smoking is the strongest environmental risk factor for RA, in particular for ACPA-positive RA, and it has been calculated in Denmark and Sweden that population-wide cessation of smoking would result in more than one-third less cases of ACPA-positive RA [82] and [81]. Other potentially modifiable factors include dietary changes, weight reduction, and dental care to reduce periodontitis. These are currently being addressed in the PRE-RA Family Study Boston, which is exploring the attitudes of family members of RA patients toward a lifestyle intervention based on a genetic plus environmental risk assessment [83]. Participants are randomized to receive feedback and education concerning their personalized RA risk based on demographics, RA-associated behaviors, genetics, and biomarkers or to receive standard RA information. Four behavioral RA risk factors are included in the risk estimate: smoking, excess body weight, poor oral health, and low fish intake. The trial outcomes will be changes in willingness to alter behaviors. As we learn more about these relations, such information programs can be refined. At present, the most important advice is for family members of ACPA-positive RA patients, to refrain from smoking [82].

The concept of primary prevention of RA with drugs has become possible through the recognition of a prolonged at-risk phase with variable symptoms and/or autoimmunity before the outbreak of clinical RA. The first clinical trial was a post hoc analysis of the effect of vitamin E in a study designed to prevent coronary heart disease in the general population [84]. Although the trial was negative for the prevention of both heart disease and RA, there was a trend toward protection against RF-positive RA. The next study was a trial of two intramuscular injections of 100 mg dexamethasone or placebo in ACPA- and/or RF-positive arthralgia patients [85]. Furthermore, this trial did not affect the onset of arthritis, although autoantibody levels were suppressed for 6 months. Meanwhile, trials of rituximab (Prevention of RA by B cell-directed therapy (PRAIRI) trial, NTR1969; www.trialregister.nl), of abatacept (Arthritis Prevention In the Pre-clinical Phase of Rheumatoid Arthritis (APIPPRA) trial; www.isrctn.com/ISRCTN46017566), and of atorvastatin (STAtins in the Prevention of RA (STAPRA) trial; NTR5265; www.trialregister.nl) in the same patient category are ongoing.

Some clinicians confronted with seropositive arthralgia patients will try antimalarial treatment. Apart from being a relatively nontoxic RA remedy, the rationale for this treatment comes from the experience with antimalarials in the treatment of palindromic rheumatism, a rather ill-defined syndrome of intermittently occurring peripheral arthritis. A subgroup of those patients is RF or ACPA positive with a tendency to develop RA [86], and this tendency was found to be markedly reduced in a retrospective survey in those taking antimalarials [87]. Another retrospective study reported a marked reduction in the frequency and duration of attacks in palindromic rheumatism patients taking chloroquin [88].

In conclusion, no intervention has yet showed an effect in a randomized controlled trial in the primary prevention of RA. The scarcity of data gives rise to the suggestion that it is not easy to perform clinical trials in the at-risk phase of RA, and that positive outcomes are not readily obtained. A major ethical issue with intervening pharmacologically in this phase is that persons are exposed to potentially toxic drugs, whereas a part of the study subjects will never develop RA.

Secondary prevention of RA

One of the explicit goals of the 2010 ACR/EULAR criteria for RA was to facilitate the performance of trials in early RA [65], in order to make even better use of the window of opportunity in early disease. The underlying idea was that it would be easier to design a trial for patients who were classified as RA instead of as UA. Nevertheless, already before the publication of the 2010 criteria, a number of trials had been conducted with the intention to prevent the progression of early disease, mostly not classifying as RA according to the 1987 ACR criteria [62]. Part of the outcome measures of these trials was a reduction of the transition of UA to RA, which means that a successful outcome could be regarded as secondary 1987 ACR criteria prevention of RA.

The results of the PROMPT study of methotrexate to prevent progression of UA to RA (1987 criteria) and its long-term follow-up showed less progression to RA, but only in ACPA-positive patients and only as long as the treatment was continued [89]. Other trials in early oligoarthritis or UA have noted some transient benefit from treatment with intramuscular (STIVEA trial) or intraarticular corticosteroids compared with placebo or nonsteroidal anti-inflammatory drugs [91] and [90]. However, the Stop Arthritis Very Early (SAVE) trial observed that the development of 1987 RA was not delayed by intramuscular glucocorticoid treatment in oligoarthritis patients [92].

Biologics have also been tested for this indication. Three months of infliximab did not prevent progression to 1987 RA after 1 year [93]. Six months of abatacept slightly reduced the progression of UA to 1987 RA from 67% to 46% [94]. Abatacept treatment also had an impact on radiographic and MRI inhibition, which was maintained for 6 months after treatment stopped. The STREAM study, a trial of aggressive treatment including adalimumab aimed at remission versus usual care in oligoarthritis patients, did not show a better outcome for aggressive treatment, although there was a trend toward less radiographic damage in the aggressively treated group [95]. In a larger two-step study aiming at early remission of early oligoarthritis or RA (IMPROVED study), similar rates of remission were achieved after 6 months of 61%. Of those not in remission at 6 months, more patients achieved remission at 12 months with adalimumab than with conventional disease-modifying antirheumatic drug (DMARD) combination therapy [96].

In conclusion, intervening in the early phase of clinical arthritis with minimal joint involvement leads to similar remission rates as are found in early RA, and there is not much evidence supporting the halting of progression from UA to RA. This suggests that it is not easy to further enhance the benefit of treating RA patients early, by treating patients with fewer involved joints even earlier in the disease course.

The broader question to what extent the transition to established RA can be prevented in patients with early RA is answered by the relative but not yet absolute success of early targeted treatment during the window of opportunity. Secondary prevention in this case could be defined as the goal of achieving and maintaining remission by early and aggressive treatment followed by minimization of therapy [97]. Spontaneous remission occurs frequently in early arthritis, especially in seronegative arthritis, and only rarely in established RA (Fig. 1) [61]. Patients who achieve early remission can sometimes maintain their remission for prolonged periods after stopping medication [98]. For patients with established RA in remission, it is less often possible to maintain a drug-free remission [100] and [99]. Taken together, it appears that DMARD-free remission can occur (13–50%), and it is not so rare as previously thought (4–6%). At any rate, there is hope that by achieving early remission with aggressive therapy, the disease can be controlled with less total medication in the long run than with milder treatment regimens.

[locator:gr1] (link type not set)

Fig. 1

Remission in different stages of rheumatoid arthritis.

 

Summary

The increasingly successful management of RA now leads to more patients achieving early and sustained remission, and this will lead to less patients progressing to the classical state of established RA. A next goal in the management of RA can be the improved recognition and intervention in the early or even at-risk phase of RA.

Prediction depends on the knowledge of risk factors. Recent advances in the risk factor assessment of RA include alcohol consumption as a confirmed protective factor, whereas fish consumption could not be confirmed as a protective factor. New risk factors are coffee consumption, sugar consumption, sleep disorders, and thyroid disease, whereas exercise and recent infections have been put forward as protective factors. Increasingly, risk factors are being combined to establish prediction rules. Those containing genetic risk plus environmental factors are not yet ready for general use. However, new prediction rules for arthralgia subjects using clinical characteristics, serology, and sometimes imaging are quite simple to perform, and they can be used to inform patients of their risk of RA.

Interestingly, RA incidence seems to have been declining since 1955, when formal measurements started, at least until the end of the last century. However, recent reports suggest that the incidence is on the rise again, mainly in seronegative females, and that this can be ascribed largely to the recent increase in obesity. When comparing trends in different countries, it becomes necessary to take into account the large variation between countries in the public and physician awareness of the need to identify RA early.

The problem of assessing UA has been reduced considerably by the introduction of the 2010 RA criteria. Many former UA patients can now be classified as RA, leaving a smaller group of UA patients with more heterogeneous and milder disease. Treating UA patients early gives results similar to early treatment of RA. In line with the concept of an early “window of opportunity,” a few studies have attempted to treat patients at an even earlier stage, before clinical arthritis becomes apparent. These primary prevention studies with pharmacological interventions have not yet produced positive results. Although these efforts are continued, the identification of modifiable risk factors for RA such as smoking, obesity, and lack of exercise should incite physicians to promote healthy behavior in persons at risk of RA.

Practice points
  • Worldwide RA incidence showed a steady decrease during the period 1955–1995, followed by a recent increase in at least Denmark and the USA.
  • New possible risk factors for the development of RA are nonalcohol use, coffee consumption, sugar-sweetened soda intake, obesity, physical inactivity, sleep disorders, and thyroid disease.
  • Possible options for the primary prevention of RA include dietary changes, weight reduction, and dental care. No drug intervention has proven to be effective in the prevention of RA.
  • With the advent of the 2010 ACR/EULAR criteria, the subgroup of UA in early arthritis patients is strongly reduced, and it contains mainly seronegative (oligo-) arthritis patients with a mild disease course.
  • Secondary prevention of RA is becoming less of an issue due to the high sensitivity of the 2010 ACR/EULAR criteria in early disease, and the tendency to treat early arthritis rapidly.
Research agenda
  • Improve prediction models of RA by integrating personal characteristics, symptoms, and genetic information with new biomarkers.
  • Establish simple prediction aids for different situations, for example, in the general practitioner (GP) office, the rheumatology clinic, or the general public.
  • Controlled intervention studies in persons at risk of RA in different stages.
  • Improved identification of UA with poor prognosis.

References

  • [1] C.L. Short. The antiquity of rheumatoid arthritis. Arthritis and Rheumatism. 1974;17(3):193-205 Crossref
  • [2] J. Dequeker. Siebrandus Sixtius: evidence of rheumatoid arthritis of the robust reaction type in a seventeenth century Dutch priest. Annals of the Rheumatic Diseases. 1992;51(4):561-562 Crossref
  • [3] G.O. Storey, M. Comer, D.L. Scott. Chronic arthritis before 1876: early British cases suggesting rheumatoid arthritis. Annals of the Rheumatic Diseases. 1994;53(9):557-560 Crossref
  • [4] B.M. Rothschild, K.R. Turner, M.A. DeLuca. Symmetrical erosive peripheral polyarthritis in the Late Archaic period of Alabama. Science. 1988;241(4872):1498-1501
  • *[5] I.C. Scott, S. Steer, C.M. Lewis, A.P. Cope. Precipitating and perpetuating factors of rheumatoid arthritis immunopathology: linking the triad of genetic predisposition, environmental risk factors and autoimmunity to disease pathogenesis. Best Practice and Research Clinical Rheumatology. 2011;25(4):447-468
  • *[6] S.A. Turk, M.H. van Beers-Tas, D. van Schaardenburg. Prediction of future rheumatoid arthritis. Rheumatic Disease Clinics of North America. 2014;40(4):753-770 Crossref
  • *[7] W.P. Arend, G.S. Firestein. Pre-rheumatoid arthritis: predisposition and transition to clinical synovitis. Nature Reviews Rheumatology. 2012;8(10):573-586 Crossref
  • [8] A. Yarwood, T.W. Huizinga, J. Worthington. The genetics of rheumatoid arthritis: risk and protection in different stages of the evolution of RA. Rheumatology. 2014; [Epub 2014 Sep 18]
  • [9] F. Kurreeman, K. Liao, L. Chibnik, et al. Genetic basis of autoantibody positive and negative rheumatoid arthritis risk in a multi-ethnic cohort derived from electronic health records. The American Journal of Human Genetics. 2011;88(1):57-69 Crossref
  • [10] J.A. Sparks, K.H. Costenbader. Genetics, environment, and gene-environment interactions in the development of systemic rheumatic diseases. Rheumatic Diseases Clinics of North America. 2014;40(4):637-657 Crossref
  • [11] K. Yamamoto, Y. Okada, A. Suzuki, Y. Kochi. Genetics of rheumatoid arthritis in Asia-present and future. Nature Reviews Rheumatology. 2015; [Epub 2015 Feb 10]
  • [12] H.S. El-Gabalawy, D.B. Robinson, N.A. Daha, et al. Non-HLA genes modulate the risk of rheumatoid arthritis associated with HLA-DRB1 in a susceptible North American native population. Genes and Immunity. 2011;12(7):568-574 Crossref
  • *[13] J.A. Sparks, C.Y. Chen, X. Jiang, et al. Improved performance of epidemiologic and genetic risk models for rheumatoid arthritis serologic phenotypes using family history. Annals of the Rheumatic Diseases. 2015;74(8):1522-1529 Crossref
  • [14] Y. Nagai, T. Imanishi. RAvariome: a genetic risk variants database for rheumatoid arthritis based on assessment of reproducibility between or within human populations. Database (Oxford). 2013; 2013:bat073
  • [15] A.H. van der Helm-van Mil, R.E. Toes, T.W. Huizinga. Genetic variants in the prediction of rheumatoid arthritis. Annals of the Rheumatic Diseases. 2010;69(9):1694-1696 Crossref
  • [16] I.C. Scott, S.D. Seegobin, S. Steer, et al. Predicting the risk of rheumatoid arthritis and its age of onset through modelling genetic risk variants with smoking. PLOS Genetics. 2013;9(9) [Epub 2013 Sep 19]
  • [17] A. Yarwood, B. Han, S. Raychaudhuri, et al. A weighted genetic risk score using all known susceptibility variants to estimate rheumatoid arthritis risk. Annals of the Rheumatic Diseases. 2015;74(1):170-176 Crossref
  • [18] Z. Jin, C. Xiang, Q. Cai, et al. Alcohol consumption as a preventive factor for developing rheumatoid arthritis: a dose-response meta-analysis of prospective studies. Annals of the Rheumatic Diseases. 2014;73(11):1962-1967 Crossref
  • [19] I.C. Scott, R. Tan, D. Stahl, et al. The protective effect of alcohol on developing rheumatoid arthritis: a systematic review and meta-analysis. Rheumatology. 2013;52(5):856-867 Crossref
  • [20] B. Lu, D.H. Solomon, K.H. Costenbader, E.W. Karlson. Alcohol consumption and risk of incident rheumatoid arthritis in women: a prospective study. Arthritis and Rheumatology. 2014;66(8):1998-2005 Crossref
  • [21] D. Di Giuseppe, A. Crippa, N. Orsini, A. Wolk. Fish consumption and risk of rheumatoid arthritis: a dose-response meta-analysis. Arthritis Research and Therapy. 2014;16(5):446 Crossref
  • [22] Y.H. Lee, S.C. Bae, G.G. Song. Coffee or tea consumption and the risk of rheumatoid arthritis: a meta-analysis. Clinical Rheumatology. 2014;33(11):1575-1583 Crossref
  • [23] W. Marder, E.C. Somers. Is pregnancy a risk factor for rheumatic autoimmune diseases?. Current Opinion In Rheumatology. 2014;26(3):321-328 Crossref
  • [24] K.T. Jorgensen, M.C. Harpsoe, S. Jacobsen, et al. Increased risk of rheumatoid arthritis in women with pregnancy complications and poor self-rated health: a study within the Danish national birth cohort. Rheumatology. 2014;53(8):1513-1519 Crossref
  • [25] H.A. Beydoun, R. el-Amin, M. McNeal, et al. Reproductive history and postmenopausal rheumatoid arthritis among women 60 years or older: third National Health and Nutrition Examination Survey. Menopause: the journal of the North American Menopause Society. 2013;20(9):930-935 Crossref
  • [26] Q. Chen, Z. Jin, C. Xiang, et al. Absence of protective effect of oral contraceptive use on the development of rheumatoid arthritis: a meta-analysis of observational studies. International Journal of Rheumatic Diseases. 2014;17(7):725-737 Crossref
  • [27] A.T. Masi, A.A. Rehman, R.T. Chatterton, et al. Controlled cohort study of serum gonadal and adrenocortical steroid levels in males prior to onset of rheumatoid arthritis (pre-RA): a comparison to pre-RA females and sex differences among the study groups. International Journal of Rheumatology. 2013;2013:284145
  • [28] M. Pikwer, A. Giwercman, U. Bergstrom. Association between testosterone levels and risk of future rheumatoid arthritis in men: a population-based case control study. Annals of the Rheumatic Diseases. 2014;73(3):573-579 Crossref
  • [29] M. Cross, E. Smith, D. Hoy, et al. The global burden of rheumatoid arthritis: estimates from the global burden of disease 2010 study. Annals of the Rheumatic Diseases. 2014;73(7):1316-1322 Crossref
  • [30] D.J. Pattison, R.A. Harrison, D.P. Symmons. The role of diet in susceptibility to rheumatoid arthritis: a systematic review. The Journal of Rheumatology. 2004;31(7):1310-1319
  • [31] B. Sundstrom, I. Johansson, S. Rantapaa-Dahlqvist. Diet and alcohol as risk factors for rheumatoid arthritis: a nested case-control study. Rheumatology International. 2015;35(3):533-539 Crossref
  • [32] B. Sundstrom, I. Johansson, S. Rantapaa-Dahlqvist. Interaction between dietary sodium and smoking increases the risk for rheumatoid arthritis: results from a nested case-control study. Rheumatology. 2015;54(3):487-493 Crossref
  • [33] Y. Hu, K.H. Costenbader, X. Gao, et al. Sugar-sweetened soda consumption and risk of developing rheumatoid arthritis in women. The American Journal of Clinical Nutrition. 2014;100(3):959-967 Crossref
  • [34] A. Finckh, C. Turesson. The impact of obesity on the development and progression of rheumatoid arthritis. Annals of the Rheumatic Diseases. 2014;73(11):1911-1913 Crossref
  • [35] A. Wesley, C. Bengtsson, A.C. Elkan, et al. Association between body mass index and anti-citrullinated protein antibody-positive and anti-citrullinated protein antibody-negative rheumatoid arthritis: results from a population-based case-control study. Arthritis Care and Research. 2013;65(1):107-112 Crossref
  • [36] B. Lu, L.T. Hiraki, J.A. Sparks, et al. Being overweight or obese and risk of developing rheumatoid arthritis among women: a prospective cohort study. Annals of the Rheumatic Diseases. 2014;73(11):1914-1922 Crossref
  • [37] D. Di Giuseppe, M. Bottai, J. Askling, A. Wolk. Physical activity and risk of rheumatoid arthritis in women: a population-based prospective study. Arthritis Research and Therapy. 2015;17(1):40 Crossref
  • [38] M.E. Sandberg, S. Wedren, L. Klareskog, et al. Patients with regular physical activity before onset of rheumatoid arthritis present with milder disease. Annals of the Rheumatic Diseases. 2014;73(8):1541-1544 Crossref
  • [39] Y.H. Hsiao, Y.T. Chen, C.M. Tseng, et al. Sleep disorders and increased risk of autoimmune diseases in individuals without sleep apnea. Sleep. 2015;38(4):581-586
  • [40] X.F. Pan, J.Q. Gu, Z.Y. Shan. Increased risk of thyroid autoimmunity in rheumatoid arthritis: a systematic review and meta-analysis. Endocrine. 2015;50(1):79-86
  • [41] M.I. Arleevskaya, A.G. Gabdoulkhakova, Y.V. Filina, et al. A transient peak of infections during onset of rheumatoid arthritis: a 10-year prospective cohort study. British Medical Journal Open. 2014;4(8) [Epub 2014 Sep 3]
  • [42] M.A. Rogers, D.A. Levine, N. Blumberg, et al. Antigenic challenge in the etiology of autoimmune disease in women. Journal of Autoimmunity. 2012;38(2–3):J97-J102 Crossref
  • [43] M.E. Sandberg, C. Bengtsson, L. Klareskog, et al. Recent infections are associated with decreased risk of rheumatoid arthritis: a population-based case-control study. Annals of the Rheumatic Diseases. 2015;74(5):904-907 Crossref
  • [44] P. Ray, S. Black, H. Shinefield, et al. Risk of rheumatoid arthritis following vaccination with tetanus, influenza and hepatitis B vaccines among persons 15-59 years of age. Vaccine. 2011;29(38):6592-6597 Crossref
  • [45] A.J. De Roos, M. Koehoorn, L. Tamburic, et al. Proximity to traffic, ambient air pollution, and community noise in relation to incident rheumatoid arthritis. Environmental Health Perspectives. 2014;122(10):1075-1080 Crossref
  • [46] J.E. Hart, H. Kallberg, F. Laden, et al. Ambient air pollution exposures and risk of rheumatoid arthritis: results from the Swedish EIRA case-control study. Annals of the Rheumatic Diseases. 2013;72(6):888-894 Crossref
  • *[47] L.A. van de Stadt, B.I. Witte, W.H. Bos, D. van Schaardenburg. A prediction rule for the development of arthritis in seropositive arthralgia patients. Annals of the Rheumatic Diseases. 2013;72(12):1920-1926 Crossref
  • [48] M.J. de Hair, R.B. Landewe, M.G. van de Sande, et al. Smoking and overweight determine the likelihood of developing rheumatoid arthritis. Annals of the Rheumatic Diseases. 2013;72(10):1654-1658 Crossref
  • *[49] M. Lahiri, R.N. Luben, C. Morgan, et al. Using lifestyle factors to identify individuals at higher risk of inflammatory polyarthritis (results from the European prospective investigation of Cancer-Norfolk and the Norfolk Arthritis Register–the EPIC-2-NOAR Study). Annals of the Rheumatic Diseases. 2014;73(1):219-226 Crossref
  • [50] C. Rakieh, J.L. Nam, L. Hunt, et al. Predicting the development of clinical arthritis in anti-CCP positive individuals with non-specific musculoskeletal symptoms: a prospective observational cohort study. Annals of the Rheumatic Diseases. 2015;74(9):1659-1666 Crossref
  • [51] W.W. Buchanan, R.M. Murdoch. Hypothesis: that rheumatoid arthritis will disappear. The Journal of Rheumatology. 1979;6(3):324-329
  • [52] Y. Alamanos, P.V. Voulgari, A.A. Drosos. Incidence and prevalence of rheumatoid arthritis, based on the 1987 American College of Rheumatology criteria: a systematic review. Seminars in Arthritis and Rheumatism. 2006;36(3):182-188 Crossref
  • [53] L.T. Jacobsson, R.L. Hanson, W.C. Knowler, et al. Decreasing incidence and prevalence of rheumatoid arthritis in Pima Indians over a twenty-five-year period. Arthritis and Rheumatism. 1994;37(8):1158-1165 Crossref
  • [54] K. Shichikawa, K. Inoue, S. Hirota, et al. Changes in the incidence and prevalence of rheumatoid arthritis in Kamitonda, Wakayama, Japan, 1965-1996. Annals of the Rheumatic Diseases. 1999;58(12):751-756 Crossref
  • [55] O. Kaipiainen-Seppanen, H. Kautiainen. Declining trend in the incidence of rheumatoid factor-positive rheumatoid arthritis in Finland 1980-2000. The Journal of Rheumatology. 2006;33(11):2132-2138
  • [56] J.K. Pedersen, A.J. Svendsen, K. Horslev-Petersen. Incidence of rheumatoid arthritis in the Southern part of Denmark from 1995 to 2001. The Open Rheumatology Journal. 2007;1:18-23
  • [57] E. Myasoedova, C.S. Crowson, H.M. Kremers, et al. Is the incidence of rheumatoid arthritis rising? Results from Olmsted County, Minnesota, 1955-2007. Arthritis and Rheumatism. 2010;62(6):1576-1582 Crossref
  • [58] C.S. Crowson, E.L. Matteson, J.M. Davis 3rd, S.E. Gabriel. Contribution of obesity to the rise in incidence of rheumatoid arthritis. Arthritis Care and Research (Hoboken). 2013;65(1):71-77 Crossref
  • [59] J.H. Humphreys, D.P. Symmons. Postpublication validation of the 2010 American College of Rheumatology/European league against rheumatism classification criteria for rheumatoid arthritis: where do we stand?. Current Opinion In Rheumatology. 2013;25(2):157-163 Crossref
  • [60] K. Raza, R. Stack, K. Kumar, et al. Delays in assessment of patients with rheumatoid arthritis: variations across Europe. Annals of the Rheumatic Diseases. 2011;70(10):1822-1825 Crossref
  • [61] J.M. Hazes, J.J. Luime. The epidemiology of early inflammatory arthritis. Nature Reviews Rheumatology. 2011;7(7):381-390 Crossref
  • [62] F.C. Arnett, S.M. Edworthy, D.A. Bloch, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis and Rheumatism. 1988;31(3):315-324 Crossref
  • [63] J. Ursum, W.H. Bos, R.J. van de Stadt, et al. Different properties of ACPA and IgM-RF derived from a large dataset: further evidence of two distinct autoantibody systems. Arthritis Research and Therapy. 2009;11(3):R75 Crossref
  • [64] A.H. van der Helm-van Mil, S. le Cessie, H. van Dongen, et al. A prediction rule for disease outcome in patients with recent-onset undifferentiated arthritis: how to guide individual treatment decisions. Arthritis and Rheumatism. 2007;56(2):433-440 Crossref
  • [65] D. Aletaha, T. Neogi, A.J. Silman, et al. 2010 rheumatoid arthritis classification criteria: an American College of Rheumatology/European league against rheumatism collaborative initiative. Annals of the Rheumatic Diseases. 2010;69(9):1580-1588 Crossref
  • *[66] A. Krabben, T.W. Huizinga, A.H. van der Helm-van Mil. Undifferentiated arthritis characteristics and outcomes when applying the 2010 and 1987 criteria for rheumatoid arthritis. Annals of the Rheumatic Diseases. 2012;71(2):238-241 Crossref
  • [67] A. Krabben, A. Abhishek, K. Britsemmer, et al. Risk of rheumatoid arthritis development in patients with unclassified arthritis according to the 2010 ACR/EULAR criteria for rheumatoid arthritis. Rheumatology. 2013;52(7):1265-1270 Crossref
  • [68] J. Shi, R. Knevel, P. Suwannalai, et al. Autoantibodies recognizing carbamylated proteins are present in sera of patients with rheumatoid arthritis and predict joint damage. Proceedings of the National Academy of Sciences of the United States of America. 2011;108(42):17372-17377 Crossref
  • [69] A. Duer-Jensen, K. Horslev-Petersen, M.L. Hetland, et al. Bone edema on magnetic resonance imaging is an independent predictor of rheumatoid arthritis development in patients with early undifferentiated arthritis. Arthritis and Rheumatism. 2011;63(8):2192-2202 Crossref
  • [70] A. Filer, P. de Pablo, G. Allen, et al. Utility of ultrasound joint counts in the prediction of rheumatoid arthritis in patients with very early synovitis. Annals of the Rheumatic Diseases. 2011;70(3):500-507 Crossref
  • [71] M.M. Nielen, D. van Schaardenburg, H.W. Reesink, et al. Simultaneous development of acute phase response and autoantibodies in preclinical rheumatoid arthritis. Annals of the Rheumatic Diseases. 2006;65(4):535-537 Crossref
  • [72] M.M. Nielen, D. van Schaardenburg, H.W. Reesink, et al. Specific autoantibodies precede the symptoms of rheumatoid arthritis: a study of serial measurements in blood donors. Arthritis and Rheumatism. 2004;50(2):380-386 Crossref
  • *[73] L.A. van de Stadt, A.R. van der Horst, M.H. de Koning, et al. The extent of the anti-citrullinated protein antibody repertoire is associated with arthritis development in patients with seropositive arthralgia. Annals of the Rheumatic Diseases. 2011;70(1):128-133 Crossref
  • [74] J. Ursum, W.H. Bos, N. van Dillen, et al. Levels of anti-citrullinated protein antibodies and IgM rheumatoid factor are not associated with outcome in early arthritis patients: a cohort study. Arthritis Research and Therapy. 2010;12(1):R8 Crossref
  • [75] D. van der Woude, S.W. Syversen, E.I. van der Voort, et al. The ACPA isotype profile reflects long-term radiographic progression in rheumatoid arthritis. Annals of the Rheumatic Diseases. 2010;69(6):1110-1116 Crossref
  • [76] L.A. van de Stadt, H. de Vrieze, N.I. Derksen, et al. Antibodies to IgG4 hinge can be found in rheumatoid arthritis patients during all stages of disease and may exacerbate chronic antibody-mediated inflammation. Arthritis and Rheumatology. 2014;66(5):1133-1140 Crossref
  • [77] M.C. Kraan, H. Versendaal, M. Jonker, et al. Asymptomatic synovitis precedes clinically manifest arthritis. Arthritis and Rheumatism. 1998;41(8):1481-1488 Crossref
  • [78] M.J. de Hair, M.G. van de Sande, T.H. Ramwadhdoebe, et al. Features of the synovium of individuals at risk of developing rheumatoid arthritis: implications for understanding preclinical rheumatoid arthritis. Arthritis and Rheumatology. 2014;66(3):513-522 Crossref
  • [79] A. Finckh, M.H. Liang, C.M. van Herckenrode, P. de Pablo. Long-term impact of early treatment on radiographic progression in rheumatoid arthritis: a meta-analysis. Arthritis and Rheumatism. 2006;55(6):864-872 Crossref
  • *[80] J.A. van Nies, R. Tsonaka, C. Gaujoux-Viala, et al. Evaluating relationships between symptom duration and persistence of rheumatoid arthritis: does a window of opportunity exist? Results on the Leiden Early Arthritis Clinic and ESPOIR cohorts. Annals of the Rheumatic Diseases. 2015;74(5):806-812 Crossref
  • [81] H. Kallberg, B. Ding, L. Padyukov, et al. Smoking is a major preventable risk factor for rheumatoid arthritis: estimations of risks after various exposures to cigarette smoke. Annals of the Rheumatic Diseases. 2011;70(3):508-511 Crossref
  • [82] M. Pedersen, S. Jacobsen, P. Garred, et al. Strong combined gene-environment effects in anti-cyclic citrullinated peptide-positive rheumatoid arthritis: a nationwide case-control study in Denmark. Arthritis and Rheumatism. 2007;56(5):1446-1453 Crossref
  • [83] J.A. Sparks, M.D. Iversen, R. Miller Kroouze, et al. Personalized risk estimator for rheumatoid arthritis (PRE-RA) family study: rationale and design for a randomized controlled trial evaluating rheumatoid arthritis risk education to first-degree relatives. Contemporary Clinical Trials. 2014;39(1):145-157 Crossref
  • [84] E.W. Karlson, N.A. Shadick, N.R. Cook, et al. Vitamin E in the primary prevention of rheumatoid arthritis: the Women׳s Health Study. Arthritis and Rheumatism. 2008;59(11):1589-1595 Crossref
  • [85] W.H. Bos, B.A. Dijkmans, M. Boers, et al. Effect of dexamethasone on autoantibody levels and arthritis development in patients with arthralgia: a randomised trial. Annals of the Rheumatic Diseases. 2010;69(3):571-574 Crossref
  • [86] L. Gonzalez-Lopez, J.I. Gamez-Nava, G.S. Jhangri, et al. Prognostic factors for the development of rheumatoid arthritis and other connective tissue diseases in patients with palindromic rheumatism. The Journal of Rheumatology. 1999;26(3):540-545
  • [87] L. Gonzalez-Lopez, J.I. Gamez-Nava, G. Jhangri, et al. Decreased progression to rheumatoid arthritis or other connective tissue diseases in patients with palindromic rheumatism treated with antimalarials. The Journal of Rheumatology. 2000;27(1):41-46
  • [88] W. Youssef, A. Yan, A.S. Russell. Palindromic rheumatism: a response to chloroquine. The Journal of Rheumatology. 1991;18(1):35-37
  • [89] H. van Dongen, J. van Aken, L.R. Lard, et al. Efficacy of methotrexate treatment in patients with probable rheumatoid arthritis: a double-blind, randomized, placebo-controlled trial. Arthritis and Rheumatism. 2007;56(5):1424-1432 Crossref
  • [90] H. Marzo-Ortega, M.J. Green, A.M. Keenan, et al. A randomized controlled trial of early intervention with intraarticular corticosteroids followed by sulfasalazine versus conservative treatment in early oligoarthritis. Arthritis and Rheumatism. 2007;57(1):154-160 Crossref
  • [91] S.M. Verstappen, M.J. McCoy, C. Roberts, et al. Beneficial effects of a 3-week course of intramuscular glucocorticoid injections in patients with very early inflammatory polyarthritis: results of the STIVEA trial. Annals of the Rheumatic Diseases. 2010;69(3):503-509 Crossref
  • [92] K.P. Machold, R. Landewe, J.S. Smolen, et al. The Stop Arthritis Very Early (SAVE) trial, an international multicentre, randomised, double-blind, placebo-controlled trial on glucocorticoids in very early arthritis. Annals of the Rheumatic Diseases. 2010;69(3):495-502 Crossref
  • [93] B. Saleem, S. Mackie, M. Quinn, et al. Does the use of tumour necrosis factor antagonist therapy in poor prognosis, undifferentiated arthritis prevent progression to rheumatoid arthritis?. Annals of the Rheumatic Diseases. 2008;67(8):1178-1180 Crossref
  • [94] P. Emery, P. Durez, M. Dougados, et al. Impact of T-cell costimulation modulation in patients with undifferentiated inflammatory arthritis or very early rheumatoid arthritis: a clinical and imaging study of abatacept (the ADJUST trial). Annals of the Rheumatic Diseases. 2010;69(3):510-516 Crossref
  • [95] I.C. van Eijk, M.M. Nielen, I. van der Horst-Bruinsma, et al. Aggressive therapy in patients with early arthritis results in similar outcome compared with conventional care: the STREAM randomized trial. Rheumatology. 2012;51(4):686-694 Crossref
  • *[96] L. Heimans, K.V. Wevers-de Boer, K. Visser, et al. A two-step treatment strategy trial in patients with early arthritis aimed at achieving remission: the IMPROVED study. Annals of the Rheumatic Diseases. 2014;73(7):1356-1361 Crossref
  • [97] J.S. Smolen, R. Landewe, F.C. Breedveld, et al. EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update. Annals of the Rheumatic Diseases. 2014;73(3):492-509 Crossref
  • [98] N. van Herwaarden, A.A. den Broeder, W. Jacobs, et al. Down-titration and discontinuation strategies of tumor necrosis factor-blocking agents for rheumatoid arthritis in patients with low disease activity. The Cochrane Database of Systematic Reviews. 2014;9 CD010455
  • [99] N. van Herwaarden, A. van der Maas, M.J. Minten, et al. Disease activity guided dose reduction and withdrawal of adalimumab or etanercept compared with usual care in rheumatoid arthritis: open label, randomised controlled, non-inferiority trial. British Medical Journal. 2015;350 h1389
  • [100] B. Fautrel, T. Pham, T. Alfaiate, et al. Step-down strategy of spacing TNF-blocker injections for established rheumatoid arthritis in remission: results of the multicentre non-inferiority randomised open-label controlled trial (STRASS: spacing of TNF-blocker injections in Rheumatoid Arthritis Study). Annals of the Rheumatic Diseases. 2015; [Epub 2015 Jun 23]

Footnotes

a Amsterdam Rheumatology and Immunology Center, Reade, Doctor Jan van Breemenstraat 2, 1056 AB Amsterdam, The Netherlands

b Amsterdam Rheumatology and Immunology Center, Reade and Academic Medical Center, Amsterdam, The Netherlands

Corresponding author. Department of Rheumatology, Jan van Breemen Research, Institute/Reade, PO Box 58271, 1040 HG Amsterdam, The Netherlands. Tel.: +31 205896222; fax: +31 20 6833498.

1 M.H. van Beers-Tas and S.A. Turk contributed equally to this work.

2 Tel.: +31 20 5896222; fax: +31 20 6833498.

3 Tel.: +31 20 5896263; fax: +31 206834464.

Stay informed

Subscribe to our E-Alert to keep up to date with the new items in the Resource Centre.

Clinical Rheumatology

Journal cover clinical Rheumatoloy - Elsevier Resource Centre

Spread the word

If you like this resource centre, please share it.