Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Differences in referral path, clinical and radiographic outcomes between seronegative and seropositive rheumatoid arthritis Mexican Mestizo patients: A cohort study

  • Guillermo Arturo Guaracha-Basáñez ,

    Contributed equally to this work with: Guillermo Arturo Guaracha-Basáñez, Irazú Contreras-Yáñez, Ana Belén Ortiz-Haro, Virginia Pascual-Ramos

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – review & editing

    Affiliation Department of Immunology and Rheumatology, Instituto Nacional de Ciencias Médicas y Nutrición Salvador-Zubirán (INCMyN-SZ), Mexico City, Mexico

    ⨯
  • Irazú Contreras-Yáñez ,

    Contributed equally to this work with: Guillermo Arturo Guaracha-Basáñez, Irazú Contreras-Yáñez, Ana Belén Ortiz-Haro, Virginia Pascual-Ramos

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Software, Supervision, Validation, Writing – review & editing

    Affiliation Department of Immunology and Rheumatology, Instituto Nacional de Ciencias Médicas y Nutrición Salvador-Zubirán (INCMyN-SZ), Mexico City, Mexico

    ⨯
  • Ana Belén Ortiz-Haro ,

    Contributed equally to this work with: Guillermo Arturo Guaracha-Basáñez, Irazú Contreras-Yáñez, Ana Belén Ortiz-Haro, Virginia Pascual-Ramos

    Roles Conceptualization, Investigation, Validation, Writing – review & editing

    Affiliation Department of Immunology and Rheumatology, Instituto Nacional de Ciencias Médicas y Nutrición Salvador-Zubirán (INCMyN-SZ), Mexico City, Mexico

    ⨯
  • Virginia Pascual-Ramos

    Contributed equally to this work with: Guillermo Arturo Guaracha-Basáñez, Irazú Contreras-Yáñez, Ana Belén Ortiz-Haro, Virginia Pascual-Ramos

    Roles Conceptualization, Formal analysis, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    virtichu@gmail.com

    Affiliation Department of Immunology and Rheumatology, Instituto Nacional de Ciencias Médicas y Nutrición Salvador-Zubirán (INCMyN-SZ), Mexico City, Mexico

    ⨯

Abstract

Background

The study compared the referral path, the first two-year clinical outcomes, and the first five-year radiographic outcomes between seronegative patients (SNPs) from a recent-onset rheumatoid arthritis dynamic cohort initiated in 2004 and seropositive patients (SPPs). Predictors of incidental erosive disease were investigated.

Patients and methods

Up to March 2023, one independent observer reviewed the charts from 188 patients with at least two years of clinical assessments and up to five years of annual radiographic assessments. SNPs were defined when baseline RF and ACPA serum titers were within local normal ranges. The erosive disease was defined on hand and/or foot radiographs when at least one unequivocal cortical bone defect was detected. The incidental erosive disease was defined in baseline erosive disease-free patients who developed erosions at follow-ups. Multivariate Cox regression analyses identified hazard ratios (95% confidence interval) for factors to predict incidental erosive disease.

Results

There were 17 (9%) SNPs, and they had a shorter time from symptoms onset to first physician evaluation, visited a lower number of physicians, and received less intensive treatment at referral and during the first years of follow-up than SPPs. Also, they had fewer 0–66 swollen joints and were less frequently persistent on therapy. The erosive disease was detected only in SPPs, and its frequency increased from 10.1% at baseline to 36.1% at the five-year radiographic assessment. There were 53 (31.4%) patients with incidental erosive disease, and differences between SPPs and SNPs were statistically significant at the feet location. Incidental erosive disease was predicted by baseline ACPA, ESR, substantial morning stiffness, and cumulative CRP.

Conclusions

SNPs showed mild differences in their referral path and clinical outcomes compared to SPPs. However, erosive disease was detected only in SPPs and was predicted by baseline and cumulative clinical and serologic variables.

Citation: Guaracha-Basáñez GA, Contreras-Yáñez I, Ortiz-Haro AB, Pascual-Ramos V (2024) Differences in referral path, clinical and radiographic outcomes between seronegative and seropositive rheumatoid arthritis Mexican Mestizo patients: A cohort study. PLoS ONE 19(6): e0304953. https://doi.org/10.1371/journal.pone.0304953

Editor: Jan René Nkeck, Faculty of Medicine and Biomedical Sciences, the University of Yaoundé I, Yaoundé, Cameroon, CAMEROON

Received: January 5, 2024; Accepted: May 21, 2024; Published: June 6, 2024

Copyright: © 2024 Guaracha-Basáñez et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data that support the findings of this study are not openly available due to reasons of sensitivity and privacy. However, our complete data are available from the corresponding author upon reasonable request. Data requests can also be placed with the chair of the Research Ethics Committee. Dr. Sergio C. Hernández Jiménez is currently the chair of the Research Ethics Committee (e-mail: sergio.hernandezj@incmnsz.mx). The details that ensure long-term data storage and availability include (but are not limited to): duplicate charts (paper and electronic), database updated on regular bases, regular electronic data backup and staff backup for the person in charge of the data availability.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Rheumatoid arthritis (RA) is increasingly recognized as a highly heterogeneous syndrome whose expression encompasses various clinical phenotypes, responses to treatment, and outcomes [1, 2]. The recognition of RA-associated autoantibodies, with the rheumatoid factor (RF) and antibodies against citrullinated proteins (ACPA) being the most acknowledged ones, supports the auto-immune nature of the disease [3]. RA patients who are positive for serum RF and/or ACPA are defined as "seropositive" patients (SPPs) and are considered to display a different etiology, disease nature, and course compared to the so-called "seronegative" patients (SNPs) who had been described as a not well-characterized group [3].

A recent review of seronegative RA highlights a high variability in the frequency of seronegative disease within general RA cohorts [1, 4]. At the same time, an increased incidence is observed due to changes in demographics and environmental risk factors [1, 5, 6]. Also, the review addresses relevant differences in the susceptibility loci [1], risk factors [1, 5–8], pathogenic pathways [1], and the benefits attributed to specific drug combinations [9, 10] between the two subsets of RA. In addition, it emphasizes a shorter and more abrupt pre-clinical history of SNPs compared to that of SPPs [1, 11, 12], a precocious and common involvement of tendons [1, 13], and better clinical, serological, and radiographic outcomes [1, 14, 15], but persistent pain [1, 16]. However, the current knowledge regarding the peculiarities of seronegative RA has been hampered by diagnostic difficulties, treatment differences in patients’ management, and conceived based on clinical evidence obtained from primarily Caucasian populations, which limits the comprehensiveness of the topic. It is worth mentioning that the seropositive subgroup highly represents RA patients. In this subgroup, distinctive characteristics had been described among RA patients from the Latin-American region (LATAM) compared to Caucasian patients. Some affect socio-demographics, such as younger age at presentation (almost ten years earlier) and an extreme female preponderance [17]. Others extend to patients’ perceptions and views of particular outcomes primarily influenced by nationality, ethnicity, and cultural background [18]. In addition, patients from non-developed countries experienced substantial delays in diagnosis, which is usually observed as several months but might extend to years in Latin American countries, and results in worse outcomes and patterns of radiographic progression, which begin early and persist after a prolonged follow-up [19, 20]. The differential disease expressions observed among (primarily) seropositive RA patients from LATAM might extend to patients with negative autoantibodies and justify the current study.

In 2004, we began assembling a recent-onset dynamic cohort of RA patients at a national referral center for rheumatic diseases in Mexico City, the Instituto Nacional de Ciencias Médicas y Nutrición Salvador-Zubirán (INMCyN-SZ). Patients included in the cohort had complete periodic assessments and were considered appropriate candidates for achieving the following study objectives:

  1. - To compare the referral path, baseline characteristics, and the first two years of follow-up clinical outcomes between SNPs from a recent-onset RA dynamic cohort and SPPs.
  2. - To compare the two groups’ baseline and annual radiographic outcomes during the first five years of follow-up.
  3. - To identify predictors of incidental erosive disease, with a particular interest in the baseline presence of RF and ACPA.

Materials and methods

Setting and study population: The recent-onset dynamic RA cohort

Patients whose data were analyzed were identified from the recent-onset dynamic RA cohort initiated in 2004 at the INCMyN-SZ.

Patients who had ≤ 12 months of symptom duration, at least one swollen joint, and no other rheumatic diagnosis but RA entered the recent-onset cohort [21]. At the baseline evaluation, the primary rheumatologist recorded the patient’s real-world referral path/journey to the cohort and a complete medical history, including demographic data, RF, and ACPA. At baseline and follow-ups, all the patients had standardized rheumatic assessments (66/68 swollen/tender joint counts, DAS28-ESR [22], and physician-overall disease visual analog scale [VAS]), patient-reported outcomes measures (PROMs) recorded (disability as per the Health Assessment Questionnaire disability index [HAQ-DI], quality of life (QoL) as per the Short-Form 36 [SF-36], fatigue [as per SF-36], pain [as per a pain-VAS], and patient overall disease status [overall-VAS]) [23], assessment of comorbidities [24], medication behavior [21] and treatment; in addition, laboratory parameters were obtained.

All the patients in the cohort were evaluated at baseline and every two months during the first two years of follow-up. After that, visits were scheduled every two, four, or six months (mandatory for all the patients), depending on the patients and disease characteristics. The rheumatologist in charge of patients’ follow-ups prescribed treatment. It was Treat-to-Target (T2T) oriented and primarily with conventional disease-modifying anti-rheumatic drugs (DMARDs) with/without glucocorticoids.

Finally, radiographs of hands and feet were scheduled at baseline and annually after that.

Study design and data collection

The current study has a cohort design. Through March 2023, the cohort comprised 237 RA patients. One hundred eighty-eight patients (79.3%) had at least two years of follow-up, the minimum time required for a significant clinical patient´ follow-up. Also, 188 (79.3%), 187 (78.9%), 157 (66.2%), 161 (67.9%), and 144 (60.8%) patients had consecutive annual radiographs and their data will be used to compare erosive disease prevalence and annual incidences during the first five years of follow-up between RA subsets (objective two) and to examine predictors of incidental erosive disease (objective three).

Up to the cut-off, 10 (5.3%) patients were deceased, 43 (22.9%) were lost to follow-up, and 135 (71.8%) were currently active in the cohort (Fig 1). Interestingly, there were differences between seronegative and seropositive groups in the number (%) of patients currently active (11 [45.8%] vs. 124 [75.6%]), lost to follow-up (13 [54.2%] vs. 30 [18.3%]) and dead (0 vs. 10 [6.1%]) (p≤0.0001).

thumbnail
Download:
Fig 1. Study flow-chart.

https://doi.org/10.1371/journal.pone.0304953.g001

Ending in May 2023 and up to the last follow-up relevant to the study objectives or death, all the charts were retrospectively reviewed by a single trained data abstractor who corroborated the integrity of the data collected.

Definitions

A senior X-ray technician was in charge of performing digitized images of radiographs of the hands (posterior-anterior and oblique views) and feet (posterior-anterior view) (Digitized, CR900 Direct view, Serie number:3172897, 120V, 10A, Kodak). Hand and foot radiographs were read chronologically by a radiologist and a rheumatologist, who were neither blinded to the patient’s clinical context nor to each other evaluation. RA was classified as erosive if both physicians identified at least one unequivocal cortical bone defect or break after carefully reviewing radiographs, and no formal validated scoring system was used. Disagreement was resolved by consensus. For the current report, the erosive disease was evaluated at baseline and then annually for up to five years of follow-up. Incidental erosive disease was defined in baseline erosive disease-free patients who developed erosions at follow-ups.

RF levels were determined by nephelometry and ACPA by second-generation (from 2004 to 2012) and third-generation (from 2013 onwards) ELISA. At the baseline evaluation, SNPs were defined when serum RF and ACPA titers were below local reference values. At the baseline evaluation, SPPs were defined when serum RF and/or ACPA titers were above local reference values.

Sample size estimation

Data from all the patients from the cohort up to March 2023, with recent-onset RA and at least two-year follow-up information, were analyzed.

We performed a post hoc power analysis to achieve the study objectives. We retrospectively estimated the power achieved considering the probability of error α equal to 0.05 and the results obtained: a difference in cumulative swollen joints and radiographic outcomes between SNPs and SPPs and the hazard ratio (HR) for ACPA, substantial morning stiffness and cumulative CRP to predict incidental erosive disease. In all cases, the power achieved with the 188 patients whose data were analyzed was ≥ 0.80. However, the power achieved for baseline ESR HR to predict incidental erosive disease was 0.57.

We used G power and Epidat software for post-hoc power analysis.

Statistical analysis

Descriptive statistics were used with frequencies and percentages for dichotomous variables and median (percentile [p] 25–75) for continuous variables (non-normal distribution).

The patients’ characteristics at referral to the cohort, at the baseline evaluation, and during follow-ups were compared between SNPs and SPPs and between patients with incidental erosive disease and patients erosive disease-free at the last follow-up, using appropriate tests: the X2 test for the categorical variables and the Mann-Whitney U test for continuous variables (non-normal distribution).

Cumulative clinical outcomes were restricted for up to the first five years of follow-ups. The following outcomes were considered: disease activity (DAS28-ESR), pain (pain-VAS), patient-overall disease-VAS, fatigue (SF-36 corresponding component), disability (HAQ-DI), QoL (SF-36), treatment (glucocorticoids use and the number of DMARDs/patient), persistence on therapy and Charlson score [24]. Cumulative clinical outcomes behavior (for continuous variables) was summarized as the Area Under the Curve (AUC), calculated by the trapezoid method and presented standardized by the length of the study period evaluated [25]. Dichotomous variables (prednisone use [or equivalent], persistence on therapy, and morning stiffness duration greater than 15 min) were summarized if present during cumulative follow-ups.

Multivariate Cox regression analysis estimated HRs (95% confidence interval [CI]) to define variables predictive of incidental erosive disease. We conceived models that considered variables at cohort entry (Model 1), cumulative variables (Model 2), and a combination of baseline and cumulative variables (Model 3). Variables’ inclusion was based on their statistical significance in the univariate analysis and clinical relevance. Considering that more than fifty variables were tested, the probability of obtaining spurious associations was higher than the nominal alpha value of 0.05 for each test. Bonferroni adjustment was considered to protect against this risk (p≤0.0009). The variables included in the model were considered simultaneous independent variables after revising collinearity (one variable was excluded if rho≥0.70). Models were repeated after RF and ACPA were forced. A test-based backward selection procedure defined the variables significantly associated with incidental erosive disease (dependent variable). The Log Likelihood function is reported as a measure of the model goodness of fit.

Missing data varied from 0% (socio-demographics, baseline characteristics, and cumulative follow-ups) to 13% (real-life referral journey), and no imputation was performed.

All statistical tests were two-sided and evaluated at the 0.05 significance level. All analyses were performed using SPSS (version 21.0, IBM Corp., Armonk, NY, USA).

Ethics

The Institutional Review Board approved the study (Comités de Ética e Investigación del INCMyN-SZ: IRE-274-10/11-1). When entering the clinic, all the patients provided written informed consent for clinical follow-ups. They also offered additional written permission to review each patient´s chart and present their data in scientific publications and forums.

Results

Population characteristics (Tables 1 and 2)

thumbnail
Download:
Table 1. Sociodemographic characteristics, patient´s referral path, baseline RA-related outcomes in the overall population, and their comparison between SNPs and SPPs.

https://doi.org/10.1371/journal.pone.0304953.t001

thumbnail
Download:
Table 2. First two-year cumulative outcomes and their comparison between SNPs and SPPs.

https://doi.org/10.1371/journal.pone.0304953.t002

Overall, patients were primarily middle-aged females (167 [88.8%]), with (median, p25-75) 12 (9–15) years of formal education and a medium-low socioeconomic level (168 [89.4%]). Patients mentioned a short time from symptoms onset to first physician evaluation (18.5 [2–59.8] days), performed in the minority by a specialist (36 [22.2%]). Half the patients were indicated DMARDs by the first physician who evaluated them, and 58 (30.9%) glucocorticoids. Among the 94 patients on DMARDs, 64 (68%) were receiving methotrexate, 14 (15%) chloroquine, and two each (2.1%) leflunomide and sulfasalazine. In contrast, only 12 (12.8%) patients were on combined DMARDs.

At the baseline evaluation in the recent-onset cohort, patients had a short disease duration (5 months [3–6.8]) and substantial clinical and serological disease activity, translating into a significant impact on the PROMs. In addition, 187 patients (99.5%) received combined DMARDs and 98 (52.1%) glucocorticoids (Table 1). The most frequent DMARD combinations were methotrexate and chloroquine in 148 (78.8%) patients, methotrexate, chloroquine and sulfasalazine in 20 (10.6%), methotrexate and leflunomide in 4 (2.1%) and methotrexate and sulfasalazine in 2 (1.1%) patients (data available for 174 patients).

Finally, the number of American College of Rheumatology (ACR) classification criteria for RA was 5 (5–6) [26].

Table 2 summarizes cumulative outcomes during the first two years of follow-up and highlights a significant improvement in joint counts, acute reactant phase determinations, and PROMs, translating into mild disease activity, the absence of disability, pain under control, and good QoL. Meanwhile, all the patients received increased combined DMARDs (2.6 [2–3]) and 114 (60.6%) glucocorticoids. The most frequent DMARD combinations included two DMARDs prescribed in 110 (63.2%) patients, among whom the most frequent combination was methotrexate and chloroquine in 60 patients. Also, 60 (35.4%) additional patients received three DMARDs combined (35 received methotrexate, chloroquine, and sulfasalazine). Finally, four (2.2%) patients received four DMARDs combined (methotrexate, chloroquine, sulfasalazine, and leflunomide) (data available for 174 patients).

Finally, 114 (60.6%) patients were persistent in therapy during the period evaluated.

Comparison of the sociodemographic characteristics, referral path, baseline disease-related characteristics, and first two years of follow-up clinical outcomes between SNPs and SPPs (Tables 1 and 2)

There were 17 patients (9%) classified as seronegative and 171 (91%) as seropositive. Baseline socio-demographic characteristics were similar between groups. Meanwhile, the time from symptoms onset to the first physician evaluation (days) was longer in SPPs (20 [5–60] vs. 11.5 [1.8–33.5], p = 0.015), who also visited a higher number of physicians before the first evaluation at the cohort (1 [0–2] vs. 0 [0–1], p = 0.001), had more frequently glucocorticoids at cohort referral (95 [55.6%] vs. 3 [17.6%], p = 0.004) and a higher number of DMARDs/patient (2 [2–2] vs. 1 [1–2], p≤0.0001), compared to their counterparts. Baseline clinical outcomes, acute reactant-phase determinations, and PROMs were similar between both groups, although some tendencies were observed toward more deteriorated outcomes among SPPs. Also, as expected, SNPs differed from SPPs in the number of ACR 1987 classification criteria for RA: 4 (4–5) vs. 5 (5–6), p≤0.001 (Table 1).

During the first two years of follow-up, SPPs showed a higher number of (0–66) swollen joints (4.6 [3–6.8] vs. 3 [1.5–4.8], p = 0.012) and received a higher number of DMARDs/patient (2.6 [2–3] vs. 2.2 [1.4–2.5], p≤0.001) compared to SNPs. Also, more SPPs received glucocorticoids (109 [63.7%] vs. 5 (29.6%), p = 0.008) and were persistent on therapy (105 [63.2%] vs. 6 [35.3%], p = 0.036), as summarized in Table 2.

Radiographic outcomes (Table 3)

thumbnail
Download:
Table 3. Radiographic outcomes in the overall population and comparison between SNPs and SPPs.

https://doi.org/10.1371/journal.pone.0304953.t003

At baseline and follow-ups, erosive disease was detected only in SPPs. Table 3 summarizes the number (%) of patients with erosions in the overall population and compares baseline and annual incident radiographic outcomes between SNPs and SPPs. Overall, the number (%) of patients with erosive disease increased over follow-ups, from 10.1% at baseline to 36.1% at five-year radiographic assessment. At any time, erosions were detected more frequently in the feet. Differences in incidental erosive disease between SNPs and SPPs were statistically significant at the feet location.

Predictors of incidental erosive disease: RF and ACPA attribution

Fifty-three (31.4%) patients developed incidental erosive disease; their data were compared with those of 116 patients who were erosive disease-free and are summarized in Table 4. Briefly, socio-demographics and real-life referral journey characteristics were similar. However, patients with incidental erosive disease had more tender and swollen joint counts, higher acute reactant phase determination and DAS28, scored worse on the physician-VAS and PROMs (but pain and fatigue), had more frequently substantial morning stiffness and met a higher number of ACR 1987 classification criteria, all at the baseline evaluation. Cumulative outcomes confirmed greater disease activity (tender and joint counts, acute reactant phase determinations levels, DAS28, and physician-VAS), more deteriorated PROMs (but SF36, fatigue, and substantial morning stiffness), and more intensive treatment among the patients with incidental erosive disease.

thumbnail
Download:
Table 4. Comparison of baseline characteristics, referral path, baseline clinical outcomes, and cumulative outcomes between patients with incidental erosive disease and patients erosive disease-free.

https://doi.org/10.1371/journal.pone.0304953.t004

Table 5 summarizes the multivariate Cox regression analysis results to define incidental erosive disease predictors. The table presents the variables included in each model, significant predictors of incidental erosive disease and significant predictors when RF and ACPA were forced into the models. In model 1, baseline ESR, substantial morning stiffness, and ACPA predicted incidental erosive disease. In model 2, cumulative CRP and physician-VAS predicted incidental erosive disease. Finally, in model 3, ESR, ACPA, substantial morning stiffness (all at baseline), and cumulative CRP predicted incidental erosive disease.

thumbnail
Download:
Table 5. Multivariate Cox regression analysis to predict incidental erosive disease.

https://doi.org/10.1371/journal.pone.0304953.t005

Discussion

The current study was performed in a well-characterized dynamic cohort of recent-onset RA patients, representative of real-life outpatients. All of them had prospective complete rheumatologic assessments that included physician evaluation, PROMs, biologic variables, and radiographic assessments on an annual basis. In addition, a directed evaluation at cohort entry of the real-life referral path to the cohort was also obtained, including time from symptoms onset, previous physicians’ evaluation, and treatment. Data related to real-life referral paths are exceptionally retrieved but impact the clinical status of the patients at cohort enrollment. At the same time, it has been described that the level of disease activity at the baseline evaluation impacts future outcomes [27].

We first observed that 9% of the patients from the cohort were SNPs, and these patients showed a shorter time from symptoms onset to the first physician evaluation, visited fewer physicians, and received less intensive treatment in terms of glucocorticoids and DMARDs than SPPs, at the baseline evaluation and during the first two years of follow-up. Clinical outcomes and PROMs were similar at the baseline and during follow-up but for the cumulative swollen joints (0–66), which were higher among SPPs who also referred to being more frequently persistent on therapy.

Our prevalence figure of seronegative RA is in the low range of that described in historical cohorts, 10–40% [28]. More recent studies, which include a collaboration of 16 registries, suggest a higher prevalence, up to 20–30% [4, 29], attributable to an aging population and its impact on RA epidemiology [4–6, 30]. Interestingly, a lower prevalence of seronegative RA has been observed in countries with low age at disease diagnosis [29], recognized as a differential characteristic of RA in LATAM populations [17]. However, data from early-onset cohorts highlight that up to 50–60% of the patients who fulfill RA classification criteria lack disease-specific autoantibodies [31, 32]. Differences observed in the estimated prevalence can be attributable to criteria selection cohorts and milder phenotypes described in some forms of seronegative RA [1]. We observed a lower percentage of SNPs currently active in the cohort. In comparison, a higher rate of patients was lost to follow-up, compared to SPPs, which might reflect self-remitting statuses in the former patients who might drop out of the cohort already described.

The natural history of seronegative RA is generally characterized by an abrupt onset, with a pattern of symptoms development that increases rapidly and is accompanied by muscle weakness [11, 12]. This might translate to the shorter time observed in our SNPs, from the symptom onset to the first physician evaluation, and the fewer physicians visited before cohort referral. Previous studies have confirmed longer, patient- and physician-related delays in SPPs, attributed to a more gradual symptom-onset and often “come-and-go” initial symptoms [11, 33–35]. Also, in agreement with our result of higher cumulative swollen joints (0–66) in SPPs, there is published evidence that in ACPA-negative patients, symptoms start in the upper extremities [11]. In a large, prospective early arthritis cohort, no differences were observed in the first symptoms nor the signs found in the physical examination at initial presentation between ACPA-positive patients and negative patients; however, during follow-up, ACPA-positive patients had more swollen joints [36]. Finally, the differences observed between SPPs and SNPs in the treatment received and the treatment-related behavior can be explained by physicians’ perception that seronegative RA is frequently milder or self-remitting and has lower DMARD requirements [1, 37, 38], more convincing benefits of some DMARDs in SPPs [1, 39], and lower retention rates of some DMARDs observed in SNPs [1, 9, 10].

Second, erosive disease was detected only in SPPs, and annual incidences increased during follow-up. At any time, erosions were detected more frequently in the feet, followed by the hands.

It has been well documented that SNPs achieve more frequent suppression of joint and systemic inflammation and remain erosive-free compared to SPPs [1, 38], confirmed in patients with recent-onset disease [15] and when bone erosions are detected with ultrasound instead of using conventional radiographs of hands and feet [40]. This association has been consistent in patients with ACPA seronegative status [36, 41–43]. At the same time, it has been observed that ACPA-positive patients develop erosions earlier and more abundantly than their counterparts [44]. Finally, similar to our results, previous studies had reported more erosions in the feet than in the hands among patients with recent-onset disease. Also, erosions developed earlier in the feet [45–47].

Baseline ACPA, ESR levels, substantial morning stiffness, and cumulative CRP were consistent predictors of incidental erosive disease in the different models tested.

Several studies have identified disease-specific autoantibodies as significant predictors of erosive disease [36, 43, 48, 49], which has also been confirmed among patients with recent-onset disease [40, 50, 51]. In their most recent report [51], Hetland et al. confirmed that positive ACPA (in addition to bone marrow edema) were independent predictors of radiographic progression after 11 years in 120 RA patients with early disease (< 6 months of symptoms duration). We did not identify RF as a significant predictor, similar to what has been published [52], and the differences observed might be due to the RA classification criteria used, time for erosive status assessment, how erosive status is assessed, and its characteristics (for instance, stable erosive status vs. progressive erosive status).

Baseline ESR has been identified as an independent predictor of radiographic progression. Syversen et al. [43] followed a cohort of 238 patients with RA longitudinally for ten years, among whom 125 patients had radiographic assessments of hands and feet at baseline and 10-year follow-up. ACPA, IgM RF, ESR, and female gender were independent predictors of radiographic progression and could be combined into an algorithm for better prediction.

(Significant) Morning joint stiffness has been proposed to be a marker of active disease that reflects the circadian pattern in the RA disease process, characterized by elevated nocturnal levels of pro-inflammatory cytokines that are insufficiently suppressed by endogenous nocturnal cortisol [53]. Interestingly, IL 6 seems instrumental in developing morning stiffness [53–55] and radiographic damage [56], which might explain our result. In addition, morning symptoms, including morning stiffness duration and severity, have been correlated with measures of disease activity such as DAS28, ACR20, and HAQ-DI, with the severity of morning stiffness showing greater effect size and less variability [57]. Ultimately, the level of disease activity at the onset is a significant negative prognostic factor [58, 59]. Our result has practical implications as morning stiffness-related information is a component of the medical history, easily obtained from the patient.

Finally, cumulative CRP predicted incidental erosive. Our findings support the established link between the prolonged effects of disease activity, as judged by CRP levels, and the progression of RA, as evidenced by radiologic assessments. This association has been observed in established [60–62] and early [63, 64] RA patients. A study by Van Leeuwen et al. [63] followed a cohort of 149 patients with symptoms duration of ≤1 year and found a correlation coefficient of 0.66 between radiologic progression and time-integrated CRP. Notably, CRP levels were more closely associated with radiologic progression than joint counts. The researchers also tried to account for individual variations in CRP levels by determining a mathematical constant that linked CRP levels to radiologic progression in each patient [64]. This allowed them to create a model that predicted joint damage at six years based on CRP and radiologic variables measured during the first six months of disease. It is important to note that CRP levels reflect an indirect measure of events in the synovium and that the total volume of synovial inflammation is what triggers the production of CRP.

Limitations of the study need to be addressed. First, this is a single-center study developed in a population of patients with particular characteristics, which limits results generalization. However, we consider it to contribute to current knowledge of the topic, which has been conceived based on studies performed mainly on Caucasians. Second, erosive disease status was established neither according to published definitions [52] nor using validated scoring methods but reflecting real-world physician evaluations; however, a radiologist and a rheumatologist read radiographs, and the increased prevalence of erosive disease observed during follow-ups, particularly in the feet location, is described when validated scoring systems are used. Third, non-erosiveness was established based on the absence of any erosion on conventional X-rays, and no additional techniques were used to confirm non-erosiveness status. Fourth, seronegative RA was defined based on the absence of serum RF and ACPA values above the local normal range; however, additional autoantibodies had been implicated in radiographic damage [3]. Fifth, the recent-onset cohort was initiated in 2004 when 2010 ACR/EULAR classification criteria were not available, and these perform better for recent-onset RA but still have insufficient sensitivity and specificity for seronegative forms [1]. Sixth, our definition of substantial morning stiffness was based on its duration over 15 minutes, and there is evidence that it might not be a marker of disease activity in early RA patients [65]. Seventh, the study was underpowered for baseline ESR to predict incidental erosive disease. Finally, a substantial number of patients were lost during follow-up, which was more evident in the seronegative group, which might have biased the results.

In conclusion, 9% of Mexican Mestizo patients from a recent onset dynamic RA cohort had concomitant negative RF and ACPA serum titers at baseline. These patients differed mildly in the real-life referral path, treatment received, and clinical outcomes from SPPs. However, only patients with disease-specific autoantibodies developed erosions over follow-up, which was more evident in the feet. In the cohort, the incidental erosive disease was predicted by baseline clinical (substantial morning stiffness), serologic (ESR and ACPA), and cumulative (CRP) variables.

Supporting information

S1 Checklist. PLOS ONE clinical studies checklist.

https://doi.org/10.1371/journal.pone.0304953.s001

(DOCX)

S2 Checklist. STROBE checklist for cross sectional studies for paper “Differences in referral path, clinical and radiographic outcomes between seronegative and seropositive patients: A case control-study design within a dynamic recent-onset rheumatoid arthritis cohort”.

https://doi.org/10.1371/journal.pone.0304953.s002

(DOCX)

References

  1. 1. De Stefano L, D’Onofrio B, Gandolfo S, Bozzalla Cassione E, Mauro D, Manzo A, et al. Seronegative rheumatoid arthritis: one year in review 2023. Clin Exp Rheumatol 2023; 41(3):554–564. pmid:36971084
  2. 2. Bugatti S, Manzo A, Montecucco C, Caporali R. The clinical value of autoantibodies in rheumatoid arthritis. Front Med (Lausanne). 2018; 3(5):339. pmid:30560132
  3. 3. Sokolova MV, Schett G, Steffen U. Autoantibodies in rheumatoid arthritis: Historical background and novel findings. Clin Rev Allergy Immunol. 2022; 63:138–151. pmid:34495490
  4. 4. Courvoisier DS, Chatzidionysiou K, Mongin D, Lauper K, Mariette X, Morel J, et al. The impact of seropositivity on the effectiveness of biologic anti-rheumatic agents: results from a collaboration of 16 registries. Rheumatology (Oxford). 2021; 60:820–8. pmid:32810263
  5. 5. Myasoedova E, Davis J, Matteson EL, Crowson CS. Is the epidemiology of rheumatoid arthritis changing? Results from a population-based incident study, 1985–2014. Ann Rheum Dis. 2020; 79:440–444.
  6. 6. Matthijssen XME, Huizinga TWJ, van der Helm-van Mil AHM. Increasing incidence of autoantibody-negative RA is replicated and is partly explained by an aging population. Ann Rheum Dis. 2022 (81): e69.
  7. 7. Hahn J, Malspeis S, Choi MY, Stevens E, Karlson EW, Lu B, et al. Association of healthy life-style behaviors and the risk of developing rheumatoid arthritis among women. Arthritis Car Res (Hoboken). 2023; 75:272–276.
  8. 8. Marchand NE, Sparks JA, Malspeis S, Yoshida K, Prisco L, Zhang X, et al. Long-Term weight changes and risk of rheumatoid arthritis among women in a prospective cohort: a marginal structural model approach. Rheumatology (Oxford). 2022; 61:1430–1439. pmid:34247242
  9. 9. Greenwood M, Shipa M, Yeoh SA, Roussou E, Mukerjee D, Ehrenstein MR. Methotrexate reduces withdrawal rates of TNK inhibitors due to ineffectiveness in rheumatoid arthritis but only in patients who are seropositive. Ann Rheum Dis. 2020; 79:1516–1517.
  10. 10. Hernández-Breijo B, Brenis CM, Plasencia-Rodríguez C, Martínez-Feito A, Novella-Navarro M, Pascual-Salcedo D, et al. Methotrexate reduces the probability of discontinuation TNF inhibitors in seropositive patients with rheumatoid arthritis: a real-world data analysis. Front Med (Lausanne). 2021 Jun 29; 8:692557.
  11. 11. Burgers LE, van Steenbergen HW, Ten Brinck RM, Huizinga TW, van der Helm-van Mil AH. Differences in the symptomatic phase preceding ACPA-positive and ACPA-negative RA. A longitudinal study in arthralgia during progression to clinical arthritis. Ann Rhem Dis. 2017; 76:1751–174.
  12. 12. van Boheemen L, Ter Wee MM, Falahee M, Filer A, van Beers-Tas M, Finckh A, et al. The symptoms in persons at risk of rheumatoid arthritis (SPARRA) questionnaire: preceding clinical arthritis development. Scan J Rheumatol. 2022;29: 1–8.
  13. 13. den Hollander NK, Verstappen M, Sidhu N, van Mulligen E, Reijnierse M, van der Helm-van Mil AHM. Hand and foot MRI in contemporary and undifferentiated arthritis: in which patients is MRI valuable to detect RA early? A large prospective study. Rheumatology (Oxford). 2022; 61:3963–3973.
  14. 14. Verstappen M, van Steenbergen HW, de Jong PHP, van der Helm-van Mil AHM. Unraveling heterogeneity within ACPA-negative rheumatoid arthritis: the subgroup of patients with a strong clinical and serological response to initiation of DMARD treatment favor disease resolution. Arthritis Res Ther. 2022; 24:4. pmid:34980246
  15. 15. Svensson B, Andersson MLE, Gjertsson I, Hafström I, Ajeganova S, Forslind K. Erosion-free rheumatoid arthritis: clinical and conceptional implications-a BARFOT study. BMC Rheumatol. 2022 Dec 30; 6:88. pmid:36581910
  16. 16. Bugatti S, De Stefano L, D’Onofrio B, Nicrosini A, Mauric E, di Lernia M, et al. Inflammatory correlates of the patient global assessment of disease activity vary in relation to disease duration and autoantibody status in patients with rheumatoid arthritis. Ann Rheum Dis. 2022; 81:1206–1213. pmid:35623639
  17. 17. Cardiel MH, Pons-Estel BA, Sacnun MP, Wojdyla D, Saurit V, Marcos JC, et al. Treatment of early rheumatoid arthritis in a multinational inception cohort of Latin-American patients: The GLADAR experience. J Clin Rheumatol. 2012;18:327–335. pmid:23047532
  18. 18. Ledón-LLanes LA, Contreras-Yáñez I, Guaracha-Basáñez GA, Valverde-Hernández SS, González-Marín A, Ballinas-Sánchez ÁJ, et al. How do Mexican patients with rheumatoid arthritis define sexual and reproductive health? a content analysis study. J Clin Rheumatol. 2022 Sep 1; 28:305–313. pmid:35612587
  19. 19. Govoni M, Sciré CA, Manara M, Malavolta N, Mancarella L, Sandri G et al. Does early arthritis clinic organization improve outcomes? What evidence is there? A systematic review. Clin Exp Rheumatol. 2013;31:443–451.
  20. 20. Rodríguez-Polanco E, Al Snih S, Kuo YF, Millán A, Rodríguez MA. Lag time between onset of symptoms and diagnosis in Venezuelan patients with rheumatoid arthritis. Rheumatol Int. 2011 May;31(5):657–65. pmid:20066425
  21. 21. Contreras-Yanez I, Ponce-De Leon S, Cabiedes J, Rull-Gabayet M, Pascual-Ramos V. Inadequate therapy behavior is associated to disease flares in patients with rheumatoid arthritis who have achieved remission with disease-modifying anti-rheumatic drugs. Am Med Sci. 2010; 340:282–290.
  22. 22. van Gestel AM, Prevoo ML, van ’t Hof MA, van Rijswijk MH, van de Putte LB, van Riel PL. Development and validation of the European League Against Rheumatism response criteria for rheumatoid arthritis. Comparison with the preliminary American College of Rheumatology and the World Health Organization/International League Against Rheumatism Criteria. Arthritis Rheum. 1996 Jan; 39:34–40. pmid:8546736
  23. 23. Barnabe C, Wattiaux A, Petkovic J, Beaton D, Shea B, Greer-Smith R, et al. Validation studies of rheumatoid arthritis patient-reported outcome measures in populations at risk for inequity: A systematic review and analysis using the OMERACT summary of measurement properties equity table. Semin Arthritis Rheum. 2022 Aug;55:152029. pmid:35640489
  24. 24. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987; 40:373–383. pmid:3558716
  25. 25. Matthews JN, Altman DG, Campbell MJ, Royston P. Analysis of serial measurement in medical research. BMJ.1990; 300:230–5.
  26. 26. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988 Mar;31:315–324. pmid:3358796
  27. 27. Agelii M Leu, Andersson MLE, Jones BL, Sjöwall C, Kastbom A, Hafström I, et al. Disease activity trajectories in rheumatoid arthritis: a tool for prediction of outcome, Scand J Rheumatol. 2021;50:1, 1–10. pmid:32856510
  28. 28. Curtis JR, Jain A, Askling J, Bridges SL Jr, Carmona L, Dixon W, et al. A comparison of patients characteristics and outcomes in selected European and US rheumatoid arthritis registries. Semin Arthritis Rheum. 2012, Aug;40:2–14.
  29. 29. Maassen JM, Bergstra SA, Chopra A, Govind N, Murphy EA, Vega-Morales D, et al. Phenotype and treatment of elderly onset compared with younger onset rheumatoid arthritis patients in international daily practice. Rheumatology (Oxford). 2021; 60:4801–4810.
  30. 30. Kaipiainen-Seppanen O, Kautiainen H. Declining trend in the incidence of rheumatoid factor-positive rheumatoid arthritis in Finland 1980–2000. J Rheumatol. 2006; 33:2132–8. pmid:17014003
  31. 31. Brinkmann GH, Norvang V, Norli ES, Grøvle L, Haugen AJ, Lexberg ÅS, et al. Treat to target strategy in early rheumatoid arthritis versus routine care- A comparative clinical practice study. Semin Arthritis Rheum. 2019 Apr; 48:808–814. pmid:30205983
  32. 32. Balduzzi S, Scirè CA, Sakellariou G, Benaglio F, Bugatti S, Montecucco C, et al. In early inflammatory polyarthritis, more intensive management according to the 2010 ACR/EULAR criteria leads to higher rates of clinical remission: comparison of two cohorts treated according to different treat-to-target protocols. Clin Exp Rheumatol. 2017; 35:401–5. pmid:27974097
  33. 33. Derksen VF, Ajeganova S, Trouw LA, van der Helm-van Mil AH, Hafström I, Huizinga TW, et al. Rheumatoid arthritis phenotype at presentation differs depending on the number of auto-antibodies present. Ann Rheum Dis. 2017, Apr;76(4):716–720. pmid:28283528
  34. 34. van der Linden MP, le Cessie S, Raza K, van der Woude D, Knevel R, Huizinga TW, et al. Long-term impact of delay in assessments of patients with early arthritis. Arthritis Rheum. 2010; 62:3537–3546.
  35. 35. Stack RJ, van Tuyl LH, Sloots M, van de Stadt LA, Hoogland W, Maat B, et al. Symptoms complexes in patients with seropositive arthralgia and in patients newly diagnosed with rheumatoid arthritis: a qualitative exploration of symptoms development. Rheumatology (Oxford). 2014; 53:1646–53.
  36. 36. van der Helm-van Mil AHM, Verpoort KN, Breedveld FC, Toes REM, Huizinga TWJ. Antibodies to citrullinated proteins and differences in clinical progression of rheumatoid arthritis. Arthrit Res Ther. 2005; 7:R949–R958 pmid:16207336
  37. 37. Carbonell-Bobadilla N, Soto-Fajardo C, Amezcua-Guerra LM, Batres-Marroquín AB, Vargas T, Hernández-Diazcouder A et al. Patients with seronegative rheumatoid arthritis have different phenotype than seropositive patients: A clinical and ultrasound study. Front Med (Lausanne). 2022 Aug 16;9:978351.
  38. 38. Willemze A, Trouw LA, Toes RE, Huizinga TW. The influence of ACPA status and characteristics on the course of RA. Nat Rev Rheumatol 2012; 8:144–52. pmid:22293763
  39. 39. Duong SQ, Crowson CS, Athreya A, Atkinson EJ, Davis JM 3rd, Warrington KJ, et al. Clinical predictors of response to methotrexate in patients with rheumatoid arthritis: a machine learning approach using clinical trial data. Arthritis Res Ther. 2022; 24:162. pmid:35778714
  40. 40. Lin D, Zhao M, Zhang Y, Xie Y, Cao J, Pan Y. Seronegative rheumatoid arthritis has milder inflammation and bone erosion in an ultrasound study of disease-modifying anti-rheumatic drugs (DMARDs)-naïve Chinese cohort. Ann Transl Med. 2022; 10:661.
  41. 41. Jansen LM, van Schaardenburg D, van der Horst-Bruinsma I, van der Stadt RJ, de Koning MH, Dijkmans BA. The predictive value of anti-cyclic citrullinated peptide antibodies in early arthritis. J Rheumatol. 2003; 30:1691–5. pmid:12913923
  42. 42. Kroot EJ, de Jong BA, van Leeuwen MA, Swinkels H, van den Hoogen FH, van’t Hof M, et al. The prognostic value of anti-cyclic citrullinated peptide antibody in patients with recent-onset rheumatoid arthritis. Arthritis Rheum. 2000; 43:1831–5. pmid:10943873
  43. 43. Syversen SW, Gaarder PI, Goll GL, Ødegård S, Haavardsholm EA, Mowinckel P, et al. High anti-cyclic citrullinated peptide levels and an algorithm of four variables predict radiographic progression in patients with rheumatoid arthritis: results from a 10-year longitudinal study. Ann Rheum Dis. 2008; 67:212–7. pmid:17526555
  44. 44. Machold KP, Stamm TA, Nell VP, Pflugbeil S, Aletaha D, Steiner G, et al. Very recent onset rheumatoid arthritis: clinical and serological patient characteristics associated with radiographic progression over the first years of disease. Rheumatology (Oxford) 2007; 46:342–9. pmid:16899498
  45. 45. van der Heijde DM, van Leeuwen MA, van Riel PL, van de Putte LB. Radiographic progression on radiographs of hands and feet during the first 3 years of rheumatoid arthritis measured according to Sharp’s method (van der Heijde modification). J Rheumatol 1995; 22(9):1792–6. pmid:8523365
  46. 46. Hulsmans HM, Jacobs JW, van der Heijde DM, van AlbadaKuipers GA, Schenk Y, Bijlsma JW. The course of radiologic damage during the first six years of rheumatoid arthritis. Arthritis Rheum. 2000; 43:1927–40. pmid:11014342
  47. 47. Plant MJ, Jones PW, Saklatvala J, Ollier WE, Dawes PT. Patterns of radiological progression in early rheumatoid arthritis: results of an 8-year prospective study. J Rheumatol. 1998; 25:417–26. pmid:9517757
  48. 48. Forslind K, Ahlmén M, Eberhardt K, Hafström I, Svensson B; BARFOT Study Group. Prediction of radiological outcome in early rheumatoid arthritis in clinical practice: role of antibodies to citrullinated peptides (anti-CCP). Ann Rheum Dis. 2004; 63:1090–5. pmid:15308518
  49. 49. Amaya-Amaya J, Calixto OJ, Saade-Lemus S, Calvo-Paramo E, Mantilla RD, Rojas-Villarraga A, et al. Does non-erosive rheumatoid arthritis exist? A cross-sectional analysis and a systematic literature review. Semin Arthritis Rheum. 2015 Apr;44(5):489–498. pmid:25440525
  50. 50. Hetland ML, Stengaard-Pedersen K, Junker P, Østergaard M, Ejbjerg BJ, Jacobsen S et al; CIMESTRA study group. Radiographic progression and remission rates in early rheumatoid arthritis—MRI bone oedema and anti-CCP predicted radiographic progression in the 5-year extension of the double-blind randomised CIMESTRA trial. Ann Rheum Dis. 2010 Oct;69(10):1789–95.
  51. 51. Hetland ML, Østergaard M, Stengaard-Pedersen K, Junker P, Ejbjerg B, Jacobsen S et al, and the CIMESTRA Study Group*. Anti-cyclic citrullinated peptide antibodies, 28-joint Disease Activity Score, and magnetic resonance imaging bone oedema at baseline predict 11 years’ functional and radiographic outcome in early rheumatoid arthritis. Scand J Rheumatol. 2019 Jan;48(1):1–8.
  52. 52. Nordberg LB, Lillegraven S, Aga AB, Sexton J, Olsen IC, Lie E, et al. Comparing the disease course of patients with seronegative and seropositive rheumatoid arthritis fulfilling the 2010 ACR/EULAR classification criteria in the treat-to-target setting: 2-year data from the ARCTIC trial. RMD Open. 2018 Nov 16;4(2): e000752.
  53. 53. Sierakowski S, Cutolo M. Morning symptoms in rheumatoid arthritis: a defining characteristic and marker of active disease. Scand J Rheumatol Suppl. 2011;125:1–5. pmid:21529303.
  54. 54. Cutolo M, Seriolo B, Craviotto C, Pizzorni C, Sulli A. Circadian rhythms in RA. Ann Rheum Dis 2003;62:593–6. pmid:12810417
  55. 55. Kirwan JR, Clarke L, Hunt LP, Perry MG, Straub RH, Jessop DS. Effect of novel therapeutic glucocorticoids on circadian rhythms of hormones and cytokines in rheumatoid arthritis. Ann N Y Acad Sci 2010;1193:127–33. pmid:20398018
  56. 56. Kondo Y, Kaneko Y, Sugiura H, Matsumoto S, Nishina N, Kuwana M et al. Pre-treatment interleukin-6 levels strongly affect bone erosion progression and repair detected by magnetic resonance imaging in rheumatoid arthritis patients. Rheumatology (Oxford). 2017 Jul 1;56(7):1089–1094. pmid:28340008.
  57. 57. Boers M, Buttgereit F, Saag K, Alten R, Grahn A, Storey D et al. What Is the Relationship Between Morning Symptoms and Measures of Disease Activity in Patients With Rheumatoid Arthritis? Arthritis Care Res (Hoboken). 2015 Sep;67(9):1202–1209. pmid:25807939.
  58. 58. Leu Agelii M, Andersson M, Jones BL, et al. Disease activity trajectories in rheumatoid arthritis: a tool for prediction of outcome. Scand J Rheumatol. 2021;50:1–10. pmid:32856510
  59. 59. Albrecht K, Zink A. Poor prognostic factors guiding treatment decisions in rheumatoid arthritis patients: a review of data from randomized clinical trials and cohort studies. Arthritis Res Ther. 2017;19:68. Published 2017 Mar 23. pmid:28335797
  60. 60. Carrier N, de Brum-Fernandes AJ, Liang P, Masetto A, Roux S, Biln NK et al. Impending radiographic erosive progression over the following year in a cohort of consecutive patients with inflammatory polyarthritis: prediction by serum biomarkers. RMD Open. 2020 May;6(1):e001191. pmid:32371434.
  61. 61. Plant MJ, Williams AL, O’Sullivan MM, Lewis PA, Coles EC, Jessop JD. Relationship between time-integrated C-reactive protein levels and radiologic progression in patients with rheumatoid arthritis. Arthritis Rheum. 2000 Jul;43(7):1473–7. pmid:10902748
  62. 62. Hassell AB, Davis MJ, Fowler PD, Clarke S, Fisher J, Shadforth ME, et al. The relationship between serial measures of disease activity and outcome in rheumatoid arthritis. QJM 1993;86:601–7. pmid:8255975
  63. 63. Van Leeuwen MA, van der Heijde DMFM, Rijswijk MH, Houtman PM, van Riel PLCM, van de Putte LBA, et al. Interrelationships of outcome measures and process variables in early rheumatoid arthritis—a comparison of radiological damage, physical disability, joint counts and acute phase reactants. J Rheumatol 1994;21:425–9.
  64. 64. Van Leeuwen MA, van Rijswijk MH, Sluiter WJ, van Riel PL, Kuper IH, van de Putte LB, et al. Individual relationship between progression of radiological damage and the acute phase response in early rheumatoid arthritis: towards development of a decision support system. J Rheumatol 1997;24:20–7.
  65. 65. Yazici Y, Pincus T, Kautiainen H, Sokka T. Morning stiffness in patients with early rheumatoid arthritis is associated more strongly with functional disability than with joint swelling and erythrocyte sedimentation rate. J Rheumatol. 2004 Sep;31(9):1723–6. pmid:15338490.