Association of Circulating Biomarkers with Disease Severity in Rheumatoid Arthritis: A Cross-Sectional Study

1. INTRODUCTION

Arthritis encompasses various musculoskeletal disorders [1]. Lyme arthritis (LA), triggered by Borrelia burgdorferi transmitted via infected tick bites (not mosquitoes) [2, 3], shares symptoms such as joint pain and swelling with rheumatoid arthritis (RA), a chronic systemic autoimmune disease affecting approximately 1% of the global population [4-6]. RA manifestations include joint pain, fatigue, swelling, morning stiffness, and rheumatoid nodules [7], with an associated reduction in life expectancy of 3–7 years [8]. Risk factors include female sex (nearly twice the incidence in women compared to men), smoking, family history, and advancing age; incidence in women aged 60–64 years exceeds that in those aged 18–29 years by more than sixfold [4, 5]. Rheumatoid factor (RF), detectable in ~80% of RA patients at higher titers than in other inflammatory or infectious conditions, correlates with severe complications [9, 10]. Although the precise etiology of RA remains unclear, dysregulated cellular and humoral immunity, alongside cytokine-mediated inflammation, plays a pivotal role in pathogenesis [11, 12]. Anemia, the most frequent extra-articular manifestation, arises from chronic inflammation and contributes to functional disability [13, 14]. These biomarkers were selected due to their established associations with RA inflammation and prognosis, as supported by ACR/EULAR criteria, where RF and anti-CCP aid in diagnosis, while CRP and ESR reflect acute-phase responses [15].

Anemia results from reduced red blood cell count or hemoglobin (Hb) concentration. Hb, an iron-containing protein in erythrocytes, primarily transports oxygen to tissues; normal levels vary by gender, altitude-related oxygen pressure, and respiratory conditions [16, 17]. In RA, inflammation diverts iron from Hb synthesis, suppresses bone marrow erythropoiesis, and blunts erythropoietin response, yielding anemia despite adequate reticuloendothelial iron stores [16]. Intracellular iron accumulation elevates ferritin in RA patients [18, 19]. Timely diagnosis critically influences prognosis, as early intervention halts destructive joint damage [20, 21]. Identifying at-risk individuals shortly after symptom onset enables initiation of disease-modifying therapy before irreversible lesions occur [22, 23]. Given RA’s unclear etiology, assessing blood biomarker fluctuations aids clinical management. This study, therefore, examined associations between circulating markers and disease severity in RA patients.

2. METHODS

This cross-sectional (descriptive-analytical) study was conducted using the census method on 148 patients referred to a specialized rheumatology clinic in Jiroft, Iran, in 2022. The census method involved enrolling all consecutive patients who met the inclusion criteria during the study period, without random sampling, to capture the entire eligible population. The study population consisted of all individuals with rheumatoid arthritis referred to the specialized rheumatology clinic in Jiroft, Iran, who were selected and evaluated based on predefined inclusion and exclusion criteria. Inclusion criteria were: informed consent to participate in the study, a definitive diagnosis of rheumatoid arthritis confirmed by a rheumatologist, and possession of a medical record at the specialized rheumatology clinic in Jiroft. Exclusion criteria included pregnancy, underlying conditions such as diabetes, chronic kidney disease, chronic liver disease, anemia, cardiovascular disorders, and use of iron supplements.

The sample size was determined a priori using G*Power software, assuming a medium effect size (Cohen's w=0.3) for chi-square associations between biomarkers and disease severity categories, with alpha=0.05 and power=0.80, yielding a minimum of 120 patients. We enrolled 148 to account for potential exclusions [24]. Post-hoc analysis confirmed >85% power for the observed associations. After obtaining ethical approval from the Research Vice-Chancellor of Jiroft University, the researchers entered the study setting to collect data. Data were collected through interviews, physical examinations, and reviews of medical records from patients referred to the rheumatology clinic. Patients were first assured of data confidentiality, after which a detailed physical examination was performed, and the number of painful and swollen joints was recorded on the researcher’s checklist. The collected information was then entered into the DAS-28 (Disease Activity Score-28) software [25, 26], and disease activity was analyzed. The DAS-28 score was calculated using the formula: DAS28 = 0.56 × √(T28) + 0.28 × √(SW28) + 0.7 × ln(ESR) + 0.014 × VAS, where T28 represents the number of tender joints, SW28 the number of swollen joints (both out of 28 assessed), ESR the erythrocyte sedimentation rate (mm/hour), and VAS the patient’s visual analogue scale for global health (0–100 mm). According to EULAR criteria, scores were categorized as follows: remission (<2.6, labeled 'recovery' in this study), low disease activity (2.6–3.2, labeled 'mild'), moderate disease activity (3.2–5.1), and high disease activity (>5.1). Moderate and high categories were combined as 'moderate or severe' due to low high-activity cases (n=5) for statistical power. Higher scores indicate greater severity [25, 27]. Data were analyzed using SPSS version 26. Descriptive statistics were reported as mean ± standard deviation (SD) for continuous variables and frequencies (percentages) for categorical variables. One-way analysis of variance (ANOVA) was used to compare means of quantitative variables (hemoglobin, ESR, AST, ALT, age) across the three disease severity groups (recovery, mild, moderate, or severe). When ANOVA was significant (p < 0.05), post-hoc comparisons were performed using Tukey's Honestly Significant Difference (HSD) test to identify pairwise differences while controlling for multiple comparisons. Pearson correlation coefficients were calculated to assess linear relationships between DAS-28 scores and continuous biomarkers. For categorical variables (RF, CRP, anti-CCP, sex), chi-square tests of independence were applied, with Fisher's exact test used when expected cell counts were <5. Multivariable logistic regression was used to estimate adjusted odds ratios (OR) for moderate/severe disease, controlling for age and gender. A two-sided p-value < 0.05 was considered statistically significant.

3. RESULTS

A total of 148 patients with rheumatoid arthritis were evaluated. The mean age of the participants was 51.84 ± 12.16 years, ranging from 19 to 81 years. Of these patients, 108 (73%) were female. Females had lower odds of having moderate-to-severe disease compared to males (OR = 0.7, 95% CI 0.3–1.5, p = 0.04). Among the 148 patients, 91 (61.5%) tested positive for rheumatoid factor (RF), while 57 (38.5%) had a negative RF test. Regarding C-reactive protein (CRP), 67 patients (45.3%) were positive, and 81 (54.7%) were negative. For anti-cyclic citrullinated peptide (anti-CCP) antibodies, 81 patients (54.7%) were positive, and 67 (45.3%) were negative. Patients with positive CRP results had significantly higher odds of having moderate-to-severe disease (OR = 3.3, 95% CI 1.7–6.4, p < 0.001). Similarly, positive anti-CCP results were strongly associated with greater disease severity (OR = 4.7, 95% CI 2.3–9.6, p < 0.001), as determined by logistic regression analysis (Table 1).

The mean Hb level in patients was 11.028 ± 1.47 g/dL, which was below the normal range; the highest Hb was 18.4 g/dL, and the lowest was 10.1 g/dL. Table 2 presents liver enzyme levels in the study patients. The mean aspartate aminotransferase (AST) was 25.05 ± 10.25 IU/L, within the normal range, with the highest value of 70 IU/L and the lowest of 11 IU/L. The mean alanine aminotransferase (ALT) was 23.86 ± 12.38 IU/L, also within the normal range, with the highest value of 85 IU/L and the lowest of 8 IU/L.

Table 2 also shows erythrocyte sedimentation rate (ESR) data; the mean ESR was 20.61 ± 16.26 mm/hour, within the normal range, with the highest value of 86 mm/hour and the lowest of 1 mm/hour. For ESR, ANOVA revealed a significant difference (F = 7.1, p < 0.001); post-hoc tests showed moderate/severe disease activity vs. remission (p < 0.001) and vs. mild activity (p = 0.04). Table 2 presents the mean DAS-28 score, which was 2.62 ± 2.43, indicating mild disease activity overall. The highest DAS-28 value was 5.37, and the lowest was 0.014 (Table 2).

Table 3 presents data on disease severity based on the DAS-28 score. Of the 148 patients examined, 79 (53.5%) were in remission, 15 (10.1%) had mild disease activity, and 54 (36.5%) had moderate or severe disease activity.

Of the 91 patients with a positive RF test, 45 (50%) were in remission, 6 (6.7%) had mild disease activity, and 40 (43.3%) had moderate-to-severe disease activity. Among the 57 patients with a negative RF test, 33 (57.9%) were in remission, 9 (15.8%) had mild disease activity, and 15 (26.3%) had moderate-to-severe disease activity. Chi-square analysis showed no significant association between RF positivity and disease severity (p = 0.07). Although patients with positive RF had higher odds of more severe disease, the difference was not statistically significant (OR = 1.9, 95% CI: 0.9–3.9, p = 0.07). Of the 67 patients with a positive CRP test, 27 (40.3%) were in remission, 6 (9%) had mild disease activity, and 34 (50.7%) had moderate or severe disease activity. Among the 80 patients with a negative CRP test, 52 (65%) were in remission, 9 (11.3%) had mild disease activity, and 19 (23.8%) had moderate or severe disease activity. There was a statistically significant association between disease severity and CRP positivity (p < 0.05). Of the 80 patients with a positive anti-CCP test, 33 (41.3%) were in remission, 6 (7.5%) had mild disease activity, and 41 (51.3%) had moderate or severe disease activity. Among the 67 patients with a negative anti-CCP test, 46 (68.7%) were in remission, 9 (13.4%) had mild disease activity, and 12 (17.9%) had moderate or severe disease activity. There was a statistically significant association between disease severity and anti-CCP positivity (p < 0.05).

One-way ANOVA revealed significant differences in mean hemoglobin levels across the three disease severity groups (F(2,145) = 4.21, p = 0.017). Post-hoc Tukey HSD tests indicated that the moderate or severe group had significantly lower mean hemoglobin (11.02 ± 1.33 g/dL) compared to the recovery group (12.19 ± 1.51 g/dL; p = 0.003), with no significant difference between mild and recovery groups (p = 0.98). The mean AST levels in the recovery, mild, and moderate-to-severe groups were 23.06, 23.93, and 28.28 U/L, respectively, with the highest mean observed in the moderate-to-severe group. However, the differences in AST levels among the three disease severity groups were not statistically significant (p > 0.05). The mean ALT levels in the remission, low-activity, and moderate-to-severe groups were 24.39, 23.26, and 24.23 IU/L, respectively, with the highest mean in the remission group; however, no statistically significant difference was observed across groups (p-value > 0.05). In contrast, mean ESR levels were 16.53, 20.33, and 26.67 mm/hour in the remission, low-activity, and moderate-to-severe groups, respectively, with the highest value in the moderate-to-severe group and a statistically significant difference across groups (F = 7.1, p-value < 0.001). The mean ages in the remission, low-activity, and moderate-to-severe groups were 52.48, 46.73, and 52.31 years, respectively, with the highest mean in the remission group; however, no statistically significant difference was observed across groups (p-value > 0.05). Among the forty male patients, twenty-three (57.5%) were in remission, and seventeen (42.5%) were in moderate-to-severe disease, whereas among the one hundred eight female patients, fifty-six (51.9%) were in remission, fifteen (13.9%) were in low activity, and thirty-seven (34.3%) were in moderate-to-severe disease. Multivariable logistic regression revealed a significant association between female gender and lower odds of moderate-to-severe disease (OR = 0.7, 95% CI 0.3–1.5, p-value = 0.04). Pearson correlations showed a negative association for Hb (r=-0.28, p=0.002) and a positive association for ESR (r=0.31, p<0.001) with DAS-28 scores. (Table 4).

4. DISCUSSION

RA, one of the most prevalent autoimmune diseases with a global prevalence of approximately 1% [4], affects women two to three times more frequently than men. Numerous studies have established ESR, CRP, and anti-CCP as key biomarkers for diagnosis and prognostic assessment in RA [15, 28, 29]. The findings revealed a significant association between RA disease severity and gender, with the majority of patients being women, consistent with the higher prevalence and often greater severity in females. West et al. reported, in a longitudinal study of Swedish patients, a significant difference in RA severity between women and men, with greater severity observed in women [30]. Hallert et al. similarly showed that disease severity was comparable between genders in the first year of follow-up but diverged significantly from the second year onward, with women exhibiting higher severity [31]. In a comprehensive review, Chen et al. identified gender as the strongest individual and environmental factor influencing RA [32]. Approximately 1.3 million adults in the United States are affected by RA, the majority of whom are women [33]. Additionally, Uhlig et al. found that the prevalence of RA in women was nearly twice that in men [34].

The findings demonstrated a statistically significant inverse relationship between RA disease severity and Hb levels: mean Hb was lower in the moderate-to-severe group compared to remission, consistent with anemia of chronic inflammation. Moghimi et al. reported positive correlations between RA severity, platelet count, Hb, and hematocrit [35]. In contrast, Milovanovic et al. observed a significant negative correlation between Hb and disease severity [36], while Yildirim et al. found a positive correlation between DAS-28 and Hb [37]. Multiple studies confirm lower Hb in RA patients [38], with anemia prevalence estimated at 30–70% [39]; a 2010 study indicated up to 50% of patients may be affected [40]. Mechanisms include macrophage iron sequestration, reduced transferrin availability, ferritin storage, and inadequate erythropoietin response relative to anemia degree-all orchestrated by inflammatory cytokines impairing hematopoiesis [41]. Ariaeian et al. similarly reported significantly reduced Hb in RA patients versus controls during antioxidant enzyme assessment [42].

The findings revealed that 54.7% of patients were anti-CCP positive, with a significant association between anti-CCP positivity and higher RA severity. Papadopoulos et al. reported that anti-CCP-positive patients exhibited more tender and swollen joints alongside elevated disease activity indices [43]. Similarly, Esalatmanesh et al. observed significantly higher mean disease activity in anti-CCP-positive versus anti-CCP-negative patients, confirming a strong correlation between DAS-28 and anti-CCP status, with greater severity in positive cases [44]. In contrast, Glasnović et al. found no significant association between anti-CCP titers and DAS-28-based disease activity, though titers predicted erosive joint changes [45]. Quinn et al. demonstrated that anti-CCP-positive patients experienced more radiological progression and joint destruction than anti-CCP-negative individuals [46]. Anti-CCP plays a critical role in RA diagnosis and prognosis, with positivity rates reaching 84.4% in some cohorts [47]; its predictive value for disease onset, severity, and joint damage [48, 49] supports its integration into treatment stratification per ACR/EULAR guidelines.

Additional key findings included significant associations between RA severity and both CRP positivity and elevated ESR. CRP, a hallmark acute-phase reactant, is consistently elevated in RA patients [50]; high-sensitivity CRP assays now aid in detecting low-grade inflammation and predicting future cardiovascular events, including myocardial infarction [51, 52]. Heydari et al. reported significant differences in CRP concentrations between severe active disease and both inactive and moderately active states, with a linear correlation to DAS-28; multivariable regression confirmed an independent effect, whereby each 1-unit increase in CRP added 0.01 to DAS-28 after adjusting for age, gender, disease duration, and RF [53]. Similarly, Kolarz et al. attributed ESR elevations in RA to local inflammatory cytokines [54], while Skoumal et al. demonstrated concurrent increases in osteoprotegerin (OPG) and ESR driven by IL-1 and macrophage colony-stimulating factor, with a significant correlation between ESR and DAS-28 [55].

5. LIMITATIONS

This study had several limitations. The lack of accurate diagnosis in the early stages of the disease, along with the challenges of conducting examinations during this period, led to the exclusion of a large number of mild cases. Additionally, the study was limited to patients referred to a single specialized rheumatology clinic in Jiroft, Iran. As a cross-sectional study, it is unable to establish causality. Given these limitations, future studies should include a larger sample size and be conducted across multiple hospitals to enhance the reliability of the findings.

CONCLUSION

This study revealed significant associations between rheumatoid arthritis severity and female gender, positive CRP and anti-CCP, lower hemoglobin, and elevated ESR. At the same time, no links were found with age, RF, or liver enzymes. A substantial proportion of patients exhibited moderate to severe disease activity, highlighting ongoing inflammatory challenges. These findings underscore the value of integrating clinical assessments with key laboratory markers for accurate risk stratification and monitoring, in line with established guidelines. Effective RA management demands a multidisciplinary approach-encompassing rheumatologists, therapists, nurses, and pharmacists-to optimize disease control and patient outcomes. Promoting structured multidisciplinary teamwork is essential for translating these insights into improved clinical practice and health policy.

RECOMMENDATIONS FOR HEALTH POLICY MAKERS

Health policymakers should encourage multidisciplinary care models that involve rheumatologists, primary care physicians, therapists, nurses, pharmacists, and allied professionals, while implementing policies that incentivize integration to improve outcomes. Access to self-management programs should be expanded, with emphasis on enhancing self-efficacy, education, and psychological support. Evidence-based, patient-centered guidelines for difficult-to-treat RA should be developed, prioritizing individualized goals based on disease history, comorbidities, and patient preferences. Healthcare organizations should ensure infrastructure for collaboration, such as care coordinators, shared electronic records, and regular team meetings. Investment in in-person and digital patient education programs is recommended to empower active participation in goal-setting and disease management. Patient-reported outcomes, disease activity, and quality measures should be systematically monitored to drive continuous improvement. A patient-centered approach with regular goal adjustments is advised, alongside leveraging multidisciplinary expertise for coordinated care and providing comprehensive education on treatments and self-management to build patient confidence. These strategies, when implemented, can significantly enhance RA management and quality of life.

REFERENCES