1. Top of page
  2. Abstract
  7. Acknowledgments


It has been suggested that immunologic events in the lungs may be involved in triggering immunity, in particular production of anti–citrullinated protein antibodies (ACPAs) during early phases of rheumatoid arthritis (RA). The aim of this study was to investigate the structural and immunologic features of the lungs in incident cases of early RA in relation to ACPA presence and smoking status.


High-resolution computed tomography (HRCT) was used to examine the lungs of 105 patients with early, untreated RA (70 with ACPA-positive RA and 35 with ACPA-negative RA) and 43 healthy individuals. Bronchoscopy with collection of bronchoalveolar lavage (BAL) fluid and mucosal bronchial biopsy specimens was performed in 23 RA patients. The presence of citrullinated proteins in the bronchial tissue was detected by immunohistochemical staining. ACPAs (detected with an anti–cyclic citrullinated peptide 2 test) and total Ig levels were determined in the sera and BAL fluid of RA patients.


HRCT imaging revealed that 63% of ACPA-positive RA patients had parenchymal lung abnormalities, compared with only 37% of ACPA-negative RA patients and 30% of healthy controls (each P < 0.05). These significant differences remained after adjustment for smoking status. Airway changes detected by HRCT were more frequent in RA patients than in healthy controls (66% versus 42%; P < 0.05), but there was no difference between ACPA-positive and ACPA-negative RA patients. Immunohistochemical studies of the bronchial tissue showed increased staining for citrullinated proteins in ACPA-positive RA patients compared with ACPA-negative RA patients (P < 0.05). ACPA levels were relatively higher in the BAL fluid as compared with the sera of ACPA-positive RA patients, suggesting that there is local production of ACPAs in the lungs of these patients.


The presence of ACPAs is associated with parenchymal lung abnormalities, site-specific citrullination, and antibody enrichment in the lungs early in the development of ACPA-positive RA.

There is increasing evidence to indicate that autoimmunity as well as inflammatory reactions occur systemically before development of any clinical sign of joint disease in patients with anti–citrullinated protein antibody (ACPA)–positive rheumatoid arthritis (RA) ([1, 2]). The fact that environmental agents such as smoking ([3]) contribute to the risk of developing ACPA-positive RA suggests that events in the lungs may occur early in the course of ACPA-positive RA and could contribute to the systemic autoimmunity that precedes the onset of RA.

There is, however, very little information available on whether changes in the lungs are present in the early stages of RA, and whether such changes might differ between ACPA-positive and ACPA-negative RA. Recently published data from a study of ACPA-positive healthy individuals at risk of RA suggested that structural pulmonary changes may indeed precede the onset of RA ([4]).

In the present study, we investigated the structural and functional pulmonary manifestations in a group of patients with incident RA. We also studied ACPAs in the bronchoalveolar lavage (BAL) fluid of patients with early RA to investigate whether lung abnormalities and local anticitrulline immunity are already present in the early stages after the onset of the first symptoms of RA.


  1. Top of page
  2. Abstract
  7. Acknowledgments


Patients attending the early arthritis clinic at Karolinska University Hospital in Stockholm, Sweden were invited to participate in a study investigating lung involvement in patients with newly diagnosed RA (the Lung Investigation in Newly Diagnosed RA [LURA] study). In total, 125 patients with early RA were invited to participate. Patients were considered eligible for the study if they had newly diagnosed RA according to the American College of Rheumatology 1987 classification criteria ([5]) and had not been previously treated with oral glucocorticoids and disease-modifying antirheumatic drugs (DMARDs). Additional criteria used for exclusion were pregnancy and alcohol and/or drug abuse. Of the 125 eligible patients, 18 declined to participate and 2 failed to complete the study protocol. Thus, 105 patients (84%) were available for the current study.

All patients were referred from primary care centers because of recent patient-reported symptoms of joint inflammation (median symptom duration 6 months, range 2–16 months). Treatment with nonsteroidal antiinflammatory drugs and joint injection with glucocorticoids, administered according to clinical indication, were allowed. The duration of the joint symptoms, as well as the extent of disease activity according to the Disease Activity Score in 28 joints using the erythrocyte sedimentation rate ([6]), presence of IgM and IgA rheumatoid factor (RF), presence of IgG ACPAs (determined with an Immunoscan RA Mark 2 anti–cyclic citrullinated peptide 2 [anti–CCP-2] kit; Euro-Diagnostica), smoking history, presence of respiratory symptoms (dyspnea and cough) during the last 12 months before inclusion, and self-reported history of pulmonary disease were assessed at inclusion. Self-reported history of pulmonary disease was defined as a history of asthma, chronic inflammation of the airways (bronchitis), emphysema, chronic obstructive pulmonary disease (COPD), or any other lung or airway disease reported by the patient.


For comparison, a control group of healthy individuals from the same geographic area, with a median age and sex distribution similar to that of the LURA cohort, was investigated using the same protocol. Controls were thus selected from the cohort of the Karolinska University Hospital Chronic Obstructive Pulmonary Disease and Smoking Proteomic (COSMIC) study. These subjects are recruited mainly by advertisement at the Lung Allergy Clinic at Karolinska University Hospital Solna ([7]). The COSMIC study investigates several aspects of COPD, using the same clinical protocol as that used in the LURA study. Thus, the control group consisted of 43 individuals without any signs of RA at the time of recruitment and from the same geographic area as the RA patients.

The patient group (LURA cohort) and control group (COSMIC cohort) were matched for numbers of never smokers (27% in the LURA cohort and 33% in the COSMIC cohort). Since former smokers were not included in the COSMIC study but were allowed in the LURA study, the cohorts were unbalanced in terms of former smokers, with 67% being current smokers in the COSMIC cohort, compared with 29% being current smokers and 44% former smokers in the LURA cohort. The analysis plan was adjusted to handle these differences as optimally as possible. Further exclusion criteria applied to the subjects from both the LURA and COSMIC study cohorts were pregnancy or planned pregnancy, drug and alcohol abuse, and severe debilitating diseases. Additional exclusion criteria, which were applied only to subjects in the COSMIC cohort, were presence of asthma, use of inhaled glucocorticoids, and use of theophylline. In the LURA cohort, 11 patients with self-reported asthma, 8 patients who had been treated with inhaled glucocorticoids, and 2 patients who had been treated with theophylline were included. The Regional Ethical Review Board in Stockholm approved both studies. Table 1 summarizes the characteristics of each cohort.

Table 1. Characteristics of the healthy controls and patients with rheumatoid arthritis (RA)*
 Healthy controlsPatients with RA
  1. IQR = interquartile range; RF = rheumatoid factor; ND = not done; ACPA = anti–citrullinated protein antibody; DAS28 = Disease Activity Score in 28 joints using the erythrocyte sedimentation rate; NA = not applicable.

No. of subjects43105
Age, median (range) years55 (44–65)56 (22–84)
Female sex, %7267
Smoking status, %  
Never smoker3326
Ever smoker6774
Current smoker6729
Median pack-years of smoking (IQR)27 (0–38)11 (0–23)
RF status, no. (%)  
IgM-RF positiveND74 (71)
IgA-RF positiveND59 (56)
Any RF positiveND76 (72)
ACPA positive, no. (%)1 (2)70 (67)
DAS28-ESR, mean ± SEMNA5.5 ± 0.1

High-resolution computed tomography (HRCT).

HRCT was performed in RA patients within 1 week after the diagnosis was first made at the early arthritis clinic. All individuals (cases and controls) were examined by HRCT using a Siemens Sensation CT instrument at full inspiration. Parameters were as follows: 0.625-mm collimator, rotation time 0.5 seconds, pitch 1,120 kV, and dose modulation (in all patients). Contiguous 2-mm axial images were reconstructed with a high-frequency filter of either B60f or B70f.

All images were reviewed for abnormalities, in accordance with the criteria included in the International Classification of HRCT for Occupational and Environmental Respiratory Diseases ([8]). The images were reviewed independently, in random order, by an experienced thoracic radiologist (SN) and a pulmonologist (RK), with patient identity masked. There was good consensus between the readers; the kappa statistic ranged from 0.5 to 0.8 for the individual radiologic patterns, with the exception of bronchiectasis. After the first run of independent evaluations, consensus on discrepant findings was reached, with evaluations still performed in a blinded manner. The HRCT findings were categorized as parenchymal abnormalities (nodules larger than 3 mm, ground-glass opacities, opacities, fibrosis, and emphysema) and airway abnormalities (bronchiectasis, air trapping, and bronchial wall thickening).

Evaluations of lung function

The forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) were measured using a spirometer (Vmax 229-6200; Legacy). The diffusing capacity for carbon monoxide (DLco) was measured using the single-breath method and corrected for hemoglobin. Lung function values are expressed as the percentage of predicted, in accordance with the European Community for Steel and Coal standards ([9]).

Bronchoscopy and retrieval of BAL fluid and biopsy specimens

All 105 RA patients included were asked whether they were willing to participate in bronchoscopy. Twenty-three (21%) of the recruited patients (11 women and 12 men, median age 58 years [interquartile range 23–69 years], 77% with ACPA-positive RA, 39% current smokers) agreed to undergo bronchoscopy. Bronchial mucosal biopsy specimens were obtained from the segmental and subsegmental septa in the left lung. In order to obtain BAL fluid, the bronchoscope (Olympus) was wedged in a middle lobe bronchus, and 5 aliquots (50 ml each) of phosphate buffered saline (PBS) were instilled. The fluid was gently suctioned back to a bottle kept on ice. Dwell-time was kept to a minimum, in accordance with the European Society of Pneumology Task Group guidelines ([10]). No aspirate was discarded. The bronchoscopy procedure and handling of the BAL fluid have been previously described in detail ([11, 12]). Bronchial biopsy specimens were snap-frozen and stored at −80°C until assessed.

Anti–CCP-2 purification and immunohistochemical analyses

Synovial fluid samples from patients with ACPA-positive RA (n = 26) were centrifuged, treated with hyaluronidase (Sigma-Aldrich), and precipitated using saturated ammonium sulfate. IgG were purified from the synovial fluid on HiTrap protein G HP columns (GE Healthcare), and eluted fractions were dialyzed and filtered before being applied to the CCP-2 affinity column (kindly provided by Euro-Diagnostica). Eluates (containing anti–IgG CCP-2 antibodies) and flow-through fractions (containing IgG-specific antibodies other than anti–IgG CCP-2) were collected. Protein concentrations were determined using a Bradford DC assay (Bio-Rad), and both fractions were biotinylated in vitro and used for immunohistochemical detection of citrullinated proteins.

For immunohistochemical analyses, formaldehyde–fixed, 7-μm–thick cryostat sections of mucosal bronchial tissue were blocked with hydrogen peroxide and washed, followed by incubation with biotinylated primary antibodies for 30 minutes at room temperature and then 60 minutes at 37°C. Slides were developed using Vectastain Elite ABC and DAB Substrate kits (Vector), followed by incubation and counterstaining with hematoxylin. The capacity of the eluted anti–CCP-2 antibodies to recognize the citrullinated form, but not the native form, of the proteins was first tested by Western blotting. We further titrated and validated these antibodies for immunohistochemistry using synovial biopsy tissue and synovial fluid–derived cells from RA patients (results available from the corresponding author upon request). Results of immunohistochemistry were evaluated (by GR and ME) using a double-blind, semiquantitative histologic evaluation system on a 3-point scale, with 0 representing absence of specific staining, 1 representing low amount of specific staining, and 2 representing high amount of specific staining for ACPAs. Representative staining results and histology scores are shown in Figure 1.


Figure 1. Patients with early, untreated rheumatoid arthritis (RA) with positivity for anti–citrullinated protein antibodies (ACPAs) exhibit higher levels of expression of the citrullinated protein in large bronchial biopsy tissue. A–F, Immunohistochemical staining of bronchial mucosal biopsy tissue from RA patients was performed using biotinylated anti–cyclic citrullinated peptide 2 (anti–CCP-2) eluates (A, C, and E) with flowthrough fractions as controls (B, D, and F), showing diaminobenzidine staining (brown) for the citrullinated protein in biopsy samples from ACPA-negative nonsmokers, ACPA-positive nonsmokers, and ACPA-positive smokers. G, The immunohistochemical staining results (using biotinylated anti–CCP-2 eluates) were assessed semiquantitatively to assign scores for the intensity of staining for the citrullinated proteins in bronchial mucosal biopsy tissue from ACPA-positive and ACPA-negative RA patients. Symbols represent individual patients; horizontal line shows the median (score of 1). ∗ = P < 0.05.

Download figure to PowerPoint

Antibody assays

Levels of IgG ACPAs in the serum and BAL fluid of RA patients were measured using a commercial anti–CCP-2 kit (Immunoscan RA Mark 2; Euro-Diagnostica) according to the manufacturer's protocol. Patients having serum anti–IgG CCP-2 antibody titers higher than 25 units/ml were defined as ACPA positive. All samples having anti–IgG CCP-2 serum titers equal to or lower than 25 units/ml were considered ACPA negative. Of the 21 serum samples tested, 3 had at least 1 duplicate in which the serum titer was above the detection limit of the kit (3,200 units/ml). All of these samples were diluted successively until the duplicate samples displayed titers below the upper detection limit, and only the corrected values were used.

Levels of IgA ACPAs in the serum and BAL fluid of RA patients were measured using the same commercial anti–CCP-2 kit (Immunoscan RA Mark 2; Euro-Diagnostica), with some modifications of the protocol provided by the manufacturer, as previously described ([13]). The presence of IgA in the standard provided with the kit was demonstrated using an in-house IgA enzyme-linked immunosorbent assay (ELISA). Briefly, microtiter plates (Nunc) were coated and incubated with AffiniPure F(ab′)2 fragment goat anti-human serum IgA, α-chain specific (Jackson ImmunoResearch) at a concentration of 2 μg/ml in PBS at 4°C overnight, and then washed and blocked for 1 hour with 1% bovine serum albumin. Coated plates were incubated with samples for 2 hours, washed, and then incubated with peroxidase-conjugated AffiniPure F(ab′)2 fragment goat anti-human serum IgA (Jackson ImmunoResearch), followed by detection using the chromogenic substrate 3,3′,5,5′-tetramethylbenzidine (Sigma-Aldrich). Plates were read at 450 nm, with 650 nm as reference. Sample concentrations were measured against the standard curve plotted using known concentrations of ChromoPure human serum IgA (Jackson ImmunoResearch). To measure anti–IgA CCP-2 antibodies, the secondary antibody provided in the anti–CCP-2 kit was replaced with horseradish peroxidase–conjugated goat anti-human IgA (Jackson ImmunoResearch), and the optimal concentration of secondary antibody required to detect the presence of anti–IgA CCP-2 antibodies was selected. The results were quantified using the standards provided by the kit. The anti–IgA CCP-2 ELISA was further validated using, as negative controls, the protein G HP column eluates and anti–CCP-2 column eluates containing only IgG.

The total concentrations (in gm/liter) of IgG and IgA in both the serum and BAL fluid were determined at the Clinical Immunology Laboratory of the Karolinska Hospital at Solna, using commercial radial immunodiffusion assays (Dade-Behring) and rate nephelometry (IMMAGE Immunochemistry System; Beckman Coulter). The assays were calibrated against the international standard CRM470. In all BAL fluid samples, the values were below the detection limit, and therefore the lower limit of detection of total IgG (0.07 gm/liter) and lower limit of detection of total IgA (0.06 gm/liter) were used.

Statistical analysis

Univariate analysis of categorical variables was performed using the chi-square test and Fisher's exact test, and for continuous variables, the Mann-Whitney U test and Wilcoxon's matched pairs signed rank test were used, where appropriate. Binomial logistic regression analysis was further performed to assess the relationship between the presence of HRCT abnormalities (either parenchymal or airway abnormalities) as the outcome and a set of predictor variables, including ACPA positivity, RF positivity, current smoking status, age ≥65 years, sex, self-reported history of lung disease, and symptom duration. The strength of associations was estimated using odds ratios (ORs) and 95% confidence intervals (95% CIs). In the subsequent analysis, all predictor variables were entered in a single model to examine which variables remained significant in a multivariate analysis. We also estimated the receiver operating characteristic curve and the corresponding area under the curve, as an index of the validity of our binomial regression model. P values less than 0.05 were considered significant. Statistical analyses were performed using Prism version 5 and Statistica version 10 software.


  1. Top of page
  2. Abstract
  7. Acknowledgments

Frequency of HRCT abnormalities in the lungs of RA patients, and presence at time of diagnosis

Both parenchymal and airway HRCT abnormalities were more frequent among RA patients than among healthy controls (54% versus 30% with parenchymal changes and 66% versus 42% with airway changes; each OR 2.7, P < 0.05). Fibrosis (12 [11%] of 105) and ground-glass infiltrates (7 [7%] of 105) were detected solely in RA patients. Parenchymal abnormalities were mainly identified in ACPA-positive RA patients, and to a lesser extent in ACPA-negative RA patients (63% versus 37%; OR 2.9, P = 0.02). In contrast, no differences in the distribution of airway abnormalities were observed between ACPA-positive and ACPA-negative RA patients (46 [66%] of 70 versus 23 [66%] of 35). A summary of the HRCT findings in healthy controls and RA patients is presented in Table 2.

Table 2. High-resolution computed tomography findings in the lungs of healthy controls and patients with rheumatoid arthritis (RA) stratified by anti–citrullinated protein antibody (ACPA) status*
 Healthy controlsACPA- negative RAACPA- positive RA
  1. Values are the number (%) of subjects. Ten of the 35 ACPA-negative RA patients were rheumatoid factor positive. Four of the 70 ACPA-positive RA patients were rheumatoid factor negative.

  2. a

    P < 0.05 versus healthy controls.

  3. b

    P < 0.05 versus healthy controls and ACPA-negative RA.

All subjects   
No. of subjects433570
Parenchymal findings13 (30)13 (37)44 (63)b
Emphysema7 (16)5 (14)13 (19)
Fibrosis04 (11)a8 (11)a
Ground-glass opacities02 (6)5 (7)
Opacities6 (14)5 (14)14 (20)
Nodules >3 mm3 (7)9 (25)24 (34)a
Airway findings18 (42)23 (66)a46 (66)a
Bronchiectasis2 (4.7)4 (11)14 (20)a
Wall thickening11 (26)11 (31)22 (31)
Air trapping14 (33)15 (43)30 (43)
Never smokers   
No. of subjects14820
Parenchymal findings1 (7)09 (45)b
Airway findings2 (14)6 (75)a14 (70)a

We then analyzed the contribution of smoking status to the observed changes by performing separate analyses for never smokers (Table 2). Both parenchymal and airway HRCT abnormalities were more frequent among RA patients who had never smoked as compared with healthy controls who had never smoked (32% versus 7% with parenchymal changes and 71% versus 14% with airway changes; P < 0.05 for both). In addition, among the never smokers, parenchymal abnormalities were more frequently observed in ACPA-positive RA patients compared with ACPA-negative RA patients (9 [45%] of 20 versus 0 of 8; P < 0.05), whereas no differences in the distribution of airway abnormalities were observed between ACPA-positive and ACPA-negative RA patients who had never smoked (14 [70%] of 20 versus 6 [75%] of 8).

No differences in lung function between RA patients and healthy controls

A slightly higher proportion of RA patients as compared with healthy controls had evidence of air flow limitation, defined as an FEV1/FVC <70% of predicted (35 [36%] of 98 [7 with missing data] versus 8 [21%] of 39 [4 with missing data]; P not significant). Similar proportions of RA patients and healthy controls had a reduced DLco, defined as a DLco <80% of predicted (51 [52%] of 98 [4 with missing data] versus 18 [45%] of 40 [7 with missing data]). No significant differences in the mean DLco values were observed between healthy controls (mean ± SD 82 ± 14% of predicted) and RA patients (mean ± SD 78 ± 16% of predicted). No correlation between the findings on HRCT and the findings on lung function tests was observed (results not shown). A slightly higher number of healthy controls as compared with RA patients reported experiencing respiratory symptoms of dyspnea (12 [28%] of 43 versus 13 [12%] of 105) and cough (5 [3%] of 43 versus 10 [1%] of 105) during the last 12 months before inclusion. No correlation between pulmonary symptoms and HRCT changes or lung function test results was observed.

Role of ACPA positivity as the major determinant of parenchymal HRCT abnormalities at the time of RA diagnosis

In the univariate analysis, parenchymal HRCT abnormalities in the lungs of patients with early, untreated RA were associated with the presence of ACPAs (OR 3.9, 95% CI 3.2–4.5, P < 0.05) and with age ≥65 years (OR 2.1, 95% CI 1.6–2.6) (Tables 3 and 4). In the multivariate binomial logistic regression analysis, ACPA positivity remained the only independent predictor of parenchymal lung abnormalities detected by HRCT (Table 4).

Table 3. Clinical characteristics of patients with early, untreated rheumatoid arthritis with or without high-resolution computed tomography (HRCT) findings of parenchymal abnormalities in the lung*
 HRCT findingsP
Parenchymal abnormalities (n = 57)No parenchymal abnormalities (n = 48)
  1. Parenchymal abnormalities were defined as the presence of at least one of the following: nodules larger than 3 mm, ground-glass opacities, opacities, fibrosis, or emphysema. ACPA = anti–citrullinated protein antibody; IQR = interquartile range; NS = not significant; RF = rheumatoid factor; DAS28-ESR = Disease Activity Score in 28 joints using the erythrocyte sedimentation rate.

ACPA positive, %74540.02
Age, median (IQR) years56 (42–63)61 (53–68)0.03
Ever smoker, %3421NS
Self-reported history of lung disease, %2519NS
No. female/no. male36/1135/23NS
Symptom duration, median (IQR) months6.5 (4.8–8.8)6 (4–8.3)NS
RF (IgG or IgA) positive, %81630.05
DAS28-ESR, mean ± SEM5.4 ± 0.15.6 ± 0.2NS
Table 4. Predictive factors for parenchymal lung abnormalities on high-resolution computed tomography in patients with early, untreated rheumatoid arthritis in the binomial logistic regression model*
 Odds ratio (95% CI)
  1. 95% CI = 95% confidence interval; ACPA = anti–citrullinated protein antibody; RF = rheumatoid factor; DAS28-ESR = Disease Activity Score in 28 joints using the erythrocyte sedimentation rate.

  2. a

    Significant in the multivariate analysis.

ACPA positive2.7 (2.1–3.4)a
Age ≥65 years2.1 (1.6–2.6)
History of smoking1.5 (1–2)
Self-reported history of lung disease1.4 (0.8–2)
RF (IgG or IgA) positive1.1 (0.4–1.8)
High DAS28-ESR (≥5.1)1 (0.6–1.5)
Female sex0.7 (0.2–1.2)
Symptom duration ≥6 months0.6 (0.1–1)

Local increase in citrullinated protein expression in the lungs of ACPA-positive RA patients

Since the presence of ACPAs was associated with the presence of parenchymal lung abnormalities, we further investigated the presence of citrullinated proteins in the lungs of RA patients according to ACPA status. Staining of the bronchial biopsy tissue using biotinylated anti–CCP-2 eluates, as well as biotinylated flowthrough fractions as controls (Figures 1A–F), revealed a significantly higher intensity of staining for citrullinated proteins in ACPA-positive RA patients (median histology score 1, range 0–2) as compared with ACPA-negative RA patients (median histology score 0, range 0–1) (P < 0.05) (Figure 1G). No correlation between these histologic findings and HRCT changes was observed.

Local enrichment of ACPAs in BAL fluid at the time of RA diagnosis

To further explore whether anticitrulline immunity may be generated locally in the lungs of patients with early RA, we investigated the presence of ACPAs in BAL fluid samples obtained from the patients. Overall, the relative titers of both IgG ACPAs and IgA ACPAs (corrected for total IgG and total IgA levels, respectively) were significantly higher in the BAL fluid as compared with paired serum samples from patients with early RA (Figure 2).


Figure 2. Enrichment of anti–citrullinated protein antibodies (ACPAs) in the lungs of ACPA-positive patients with early, untreated rheumatoid arthritis (RA). Normalized values of the anti–cyclic citrullinated peptide 2 tests for levels of IgG ACPAs (A) and IgA ACPAs (B) are shown in paired samples of serum and broncoalveolar lavage (BAL) fluid from RA patients with detectable ACPAs in both the serum and BAL fluid. Values were normalized to the levels of total IgG and total IgA. ∗ = P < 0.05.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  7. Acknowledgments

ACPA-positive RA is a complex disease. Signs of systemic immunity against citrullinated proteins are present before the onset of disease, in the absence of any clinical sign of joint inflammation. Our results demonstrated that lung abnormalities, increased protein citrullination, and ACPA enrichment in the lungs are present early after disease onset. Our observations strengthen the concept that the initiating event might be located outside the joint in patients with ACPA-positive RA.

The presence of lung abnormalities has been demonstrated in many previous studies of RA ([14-18]), but there are still many unresolved issues concerning the temporal relationships between various types of lung manifestations and various types of RA. A strength of our study is that it was performed in a population-based setting, where patients with suspected RA were referred directly from primary care to our rheumatology clinic early after the first symptoms of arthritis, and before receiving any disease-modifying drugs.

A further strength is that we were able to compare HRCT lung changes between patients with early, untreated RA and healthy controls and to account for smoking effects. One limitation is that our control population included only never and current smokers, but no former smokers, whereas the RA cohort included former smokers, which resulted in a much higher number of current smokers in the healthy population (67% of healthy individuals being current smokers, compared with 29% of RA patients). However, in the worst scenario, the bias introduced by this would result in detection of more HRCT findings in healthy individuals and a consequent underestimation of the differences between RA patients and controls.

The HRCT findings described herein do not necessarily represent clinically manifest lung disease, but could be an indirect sign of a subclinical inflammatory process. Preliminary results from our histologic analyses supported this hypothesis, despite the fact that the sample size of subjects with available peribronchial tissue was small, and despite the possibility of selection bias in the recruitment phase of volunteers for bronchoscopy. We did not observe any differences in lung function test results between the RA patients and healthy controls. This observation is consistent with previous studies in which pulmonary dysfunction was demonstrated to have no value in the prediction of RA development ([19]), and in which only a modest correlation was demonstrated between pulmonary function test results and HRCT findings in DMARD-treated patients with early RA ([20]).

A finding of particular interest in the present study was the presence of parenchymal changes at an early disease stage in patients with ACPA-positive RA, but not in those with ACPA-negative RA, compatible with a process in the lungs that precedes the symptoms in the joints. In contrast, the similar frequency of occurrence of airway disease in both ACPA-positive and ACPA-negative RA patients might point to a more general inflammation in the lungs during RA, rather than a specific feature associated with initiation/propagation of the ACPA-positive variant of the disease. In line with this hypothesis, a previous study demonstrated HRCT patterns indicative of a predominance of small airway disease in patients with longstanding RA. In contrast, interstitial abnormalities were frequently observed in early RA, even in the absence of respiratory symptoms ([21]).

Taken together, the findings suggest that early lung injury is the initiating event leading to local molecular changes resulting in neoepitope formation (such as citrullinated proteins) and generation of an immune response in genetically susceptible individuals, which subsequently spreads into the general lymphoid system to ultimately target the joints. Our demonstration of parenchymal HRCT changes occurring mainly in ACPA-positive RA patients and the increased expression of citrullinated proteins in the bronchial biopsy tissue of ACPA-positive RA patients, together with local enrichment of IgG ACPAs and IgA ACPAs in the lungs, are all findings that support this hypothesis. Earlier studies similarly found that ACPAs were present in the blood of RA patients before disease onset ([1, 22]) and in the absence of obvious synovial changes ([23]), while another study demonstrated evidence of antibody production in the lungs of RA patients with longstanding disease ([24]). It still remains to be determined whether the origin of the detected citrullinated proteins is the lung, and whether the joints and lungs have the same citrullinated proteins in common. However, our identification of citrullinated proteins in the bronchial tissue of patients with early RA, using anti–CCP-2 antibodies derived from the synovial fluid of RA patients, gives indirect proof that common citrullinated targets exist in the lungs and joints of RA patients. It is also possible that early lung changes in RA may be due to the secondary immune injury of the lungs that is associated with the presence of ACPAs, in which an initial triggering event occurs at sites different from both the lungs and joints. These 2 different scenarios explaining the HRCT changes are complementary, and not mutually exclusive.

Previous studies, performed with methods and clinical materials different from those used in the present study, yielded results that were partly, but not entirely, in line with this suggested scenario. Pulmonary involvement was present in RA patients recruited within 3 years after diagnosis ([21, 25]). A recent report described an increased frequency of both interstitial and airway disease in a cohort consisting of patients with early RA and patients with longstanding RA ([26]), with airway disease being more prominent in longstanding RA. Another study, in 60 patients with early RA, demonstrated that the presence of ACPAs was associated with bronchial wall thickening, and the presence of RF was associated with interstitial lung disease ([20]). The major limitation of these studies is that a large majority of the included patients were already treated with DMARDs, including methotrexate. An elegant, but small, study of ACPA-positive and/or RF-positive healthy individuals at risk of developing RA demonstrated that there was a higher incidence of HRCT airway abnormalities, but not parenchymal abnormalities, in these patients as compared with ACPA-negative healthy individuals ([4]).

In conclusion, the results from the present study in patients with newly diagnosed RA support the notion that early inflammatory events in the lungs may represent a critical initiating factor in the development of ACPA-positive RA. Thus, further studies on the relationships between inflammation in the lungs and inflammation in the joints are warranted, with the goals being to identify factors that may cause immune responses involved in joint-specific inflammation and to identify early events that can be targeted with preventive or therapeutic measures. Moreover, long-term followup studies to assess the potential clinical consequences of our findings are needed. Our current existing knowledge should nevertheless have direct consequences on the clinical management of the disease, such as underscoring the need for more active screening to identify lung disease in high-risk patients with RA (especially smokers with ACPA-positive RA), prompting implementation of antismoking strategies in the RA treatment program, and encouraging the development of interventional studies to suppress the local immune process in the lungs.


  1. Top of page
  2. Abstract
  7. Acknowledgments

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Catrina had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Reynisdottir, Karimi, Olsen, Grunewald, Nyren, Eklund, Klareskog, Sköld, Catrina.

Acquisition of data. Reynisdottir, Karimi, Joshua, Olsen, Hensvold, Harju, Engström, Grunewald, Nyren, Eklund, Klareskog, Catrina.

Analysis and interpretation of data. Reynisdottir, Karimi, Joshua, Olsen, Engström, Grunewald, Eklund, Klareskog, Sköld, Catrina.


  1. Top of page
  2. Abstract
  7. Acknowledgments

The technical assistance of Helene Blomqvist, Gunnel de Forest, Margitha Dahl, and Benita Dahlberg is greatly acknowledged. We also thank Michaela Larkin for editing help and advice.


  1. Top of page
  2. Abstract
  7. Acknowledgments
  • 1
    Rantapaa-Dahlqvist S, de Jong BA, Berglin E, Hallmans G, Wadell G, Stenlund H, et al.Antibodies against cyclic citrullinated peptide and IgA rheumatoid factor predict the development of rheumatoid arthritis.Arthritis Rheum2003;48:27419.
  • 2
    Nielen MM, van Schaardenburg D, Reesink HW, van de Stadt RJ, van der Horst-Bruinsma IE, de Koning MH, et al.Specific autoantibodies precede the symptoms of rheumatoid arthritis: a study of serial measurements in blood donors.Arthritis Rheum2004;50:3806.
  • 3
    Klareskog L, Padyukov L, Lorentzen J, Alfredsson L.Mechanisms of disease: genetic susceptibility and environmental triggers in the development of rheumatoid arthritis.Nat Clin Pract Rheumatol2006;2:42533.
  • 4
    Demoruelle MK, Weisman MH, Simonian PL, Lynch DA, Sachs PB, Pedraza IF, et al.Airways abnormalities and rheumatoid arthritis–related autoantibodies in subjects without arthritis: early injury or initiating site of autoimmunity?Arthritis Rheum2012:64:175661.
  • 5
    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 Rheum1988;31:31524.
  • 6
    Prevoo ML, van 't Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, van Riel PL.Modified disease activity scores that include twenty-eight–joint counts: development and validation in a prospective longitudinal study of patients with rheumatoid arthritis.Arthritis Rheum1995;38:448.
  • 7
    Kohler M, Sandberg A, Kjellqvist S, Thomas A, Karimi R, Nyren S, et al.Gender differences in the bronchoalveolar lavage cell proteome of patients with chronic obstructive pulmonary disease.J Allergy Clin Immunol2013;131:74351.
  • 8
    Kusaka Y, Hering KG, Parker JE, editors.International classification of HRCT for occupational and environmental respiratory diseases.Tokyo:Springer-Verlag;2005.
  • 9
    Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC.Lung volumes and forced ventilatory flows: report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society.Eur Respir J Suppl1993;16:540.
  • 10
    Clinical guidelines and indications for bronchoalveolar lavage (BAL): report of the European Society of Pneumology Task Group on BAL.Eur Respir J1990;3:93776.
  • 11
    Olsen HH, Grunewald J, Tornling G, Skold CM, Eklund A.Bronchoalveolar lavage results are independent of season, age, gender and collection site.PLoS One2012;7:e43644.
  • 12
    Karimi R, Tornling G, Grunewald J, Eklund A, Skold CM.Cell recovery in bronchoalveolar lavage fluid in smokers is dependent on cumulative smoking history.PLoS One2012;7:e34232.
  • 13
    Svard A, Kastbom A, Reckner-Olsson A, Skogh T.Presence and utility of IgA-class antibodies to cyclic citrullinated peptides in early rheumatoid arthritis: the Swedish TIRA project.Arthritis Res Ther2008;10:R75.
  • 14
    Turesson C, Jacobsson LT.Epidemiology of extra-articular manifestations in rheumatoid arthritis.Scand J Rheumatol2004;33:6572.
  • 15
    Metafratzi ZM, Georgiadis AN, Ioannidou CV, Alamanos Y, Vassiliou MP, Zikou AK, et al.Pulmonary involvement in patients with early rheumatoid arthritis.Scand J Rheumatol2007;36:33844.
  • 16
    Ayhan-Ardic FF, Oken O, Yorgancioglu ZR, Ustun N, Gokharman FD.Pulmonary involvement in lifelong non-smoking patients with rheumatoid arthritis and ankylosing spondylitis without respiratory symptoms.Clin Rheumatol2006;25:2138.
  • 17
    Bongartz T, Nannini C, Medina-Velasquez YF, Achenbach SJ, Crowson CS, Ryu JH, et al.Incidence and mortality of interstitial lung disease in rheumatoid arthritis: a population-based study.Arthritis Rheum2010;62:158391.
  • 18
    Nannini C, Medina-Velasquez YF, Achenbach SJ, Crowson CS, Ryu JH, Vassallo R, et al.Incidence and mortality of obstructive lung disease in rheumatoid arthritis: a population-based study.Arthritis Care Res (Hoboken)2013;65:124350.
  • 19
    Bergstrom U, Jacobsson LT, Nilsson JA, Berglund G, Turesson C.Pulmonary dysfunction, smoking, socioeconomic status and the risk of developing rheumatoid arthritis.Rheumatology (Oxford)2011;50:200513.
  • 20
    Wilsher M, Voight L, Milne D, Teh M, Good N, Kolbe J, et al.Prevalence of airway and parenchymal abnormalities in newly diagnosed rheumatoid arthritis.Respir Med2012;106:14416.
  • 21
    Mori S, Cho I, Koga Y, Sugimoto M.Comparison of pulmonary abnormalities on high-resolution computed tomography in patients with early versus longstanding rheumatoid arthritis.J Rheumatol2008;35:151321.
  • 22
    Brink M, Hansson M, Mathsson L, Jakobsson PJ, Holmdahl R, Hallmans G, et al.Multiplex analyses of antibodies against citrullinated peptides in individuals prior to development of rheumatoid arthritis.Arthritis Rheum2013;65:899910.
  • 23
    Van de Sande MG, de Hair MJ, van der Leij C, Klarenbeek PL, Bos WH, Smith MD, et al.Different stages of rheumatoid arthritis: features of the synovium in the preclinical phase.Ann Rheum Dis2011;70:7727.
  • 24
    Rangel-Moreno J, Hartson L, Navarro C, Gaxiola M, Selman M, Randall TD.Inducible bronchus-associated lymphoid tissue (iBALT) in patients with pulmonary complications of rheumatoid arthritis.J Clin Invest2006;116:318394.
  • 25
    Habib HM, Eisa AA, Arafat WR, Marie MA.Pulmonary involvement in early rheumatoid arthritis patients.Clin Rheumatol2011;30:21721.
  • 26
    Mori S, Koga Y, Sugimoto M.Different risk factors between interstitial lung disease and airway disease in rheumatoid arthritis.Respir Med2012;106:15919.