Abstract
Background
The risk factors for interstitial lung disease (ILD) in rheumatoid arthritis (RA) are inconsistent among previous studies. Furthermore, the factors associated with the emergence of the recently defined progressive fibrosing (PF) phenotype are unknown. Herein, we analyze the risk factors for ILD in RA. We also analyze the factors associated with a PF phenotype.
Methods
We collected the clinical and laboratory details of subjects with RA with (cases) or without (controls) ILD. Scoring of high-resolution computed tomography (HRCT) features of ILD was performed. We identified the subgroup that developed the PF phenotype during follow-up. We analyzed the factors associated with ILD using logistic regression (primary objective). We also compared the characteristics of ILD subjects with or without the PF phenotype (secondary objective).
Results
We included 60 subjects (30 cases, 30 controls). Subjects with ILD had higher age, lower body mass index, longer duration of RA, and poorer lung function than the controls. Age (p = 0.007) and the duration of RA (p = 0.049) were the only significant predictors of ILD on univariate and multivariate analysis, respectively. Six (20%) subjects with RA-ILD developed a PF phenotype. These subjects were older, had greater frequency of honeycombing, and higher HRCT scores for honeycombing and aggregate fibrosis than those without the PF phenotype. Among subjects with honeycombing, 41.7% developed the PF phenotype.
Conclusions
RA-ILD was associated with the duration of RA and age. Subjects with the PF phenotype were older and had higher honeycombing and fibrosis scores on HRCT chest.
Keywords: Diffuse lung disease, Diffuse parenchymal lung disease, Interstitial pneumonia, Pulmonary fibrosis, Lung fibrosis, Connective tissue disease
Introduction
Rheumatoid arthritis (RA) is among the most common causes of connective tissue disease (CTD)-related interstitial lung disease (ILD).1 Clinically significant ILD occurs in about 5–10% of patients with RA, while another 20–30% may be identified only on a high-resolution computed tomography (HRCT) of the chest and lung function testing.2 The risk factors for ILD in patients with RA include older age, male gender, tobacco smoking, higher disease activity, the presence of other extra-articular manifestations, and the presence of anti-citrullinated protein antibodies (ACPAs) or high-titer rheumatoid factor (RF), although there are inconsistencies across studies.2, 3, 4, 5, 6, 7
A subgroup of patients with RA with a fibrotic ILD might develop a progressive course with declining lung function, despite immunosuppression.8 An ILD with such a disease course, when satisfying certain well-defined criteria for progression is known as progressive fibrosing (PF)-ILD.9 Recently, antifibrotic agents, namely nintedanib and pirfenidone, previously used for idiopathic pulmonary fibrosis (IPF) have been found useful in non-IPF PF-ILDs, including CTD-related ILDs.10, 11, 12, 13, 14, 15 However, the predictors of the development of the PF phenotype among patients with RA-ILD are unknown. Herein, we explore the risk factors for ILD in our patients with RA, and the characteristics associated with the PF phenotype.
Materials and methods
This was a case–control study of subjects with RA with or without ILD. The Ethics Review Committee of the Institute approved the study protocol. All subjects provided a written informed consent for participation. All guidelines as per the Declaration of Helsinki and good clinical practice guidelines were followed.
Study subjects
Consecutive subjects who visited our Chest Clinic between August 2017 and December 2018 were included as cases if they had RA with clinically significant ILD. Clinically significant ILD was defined by the presence of respiratory symptoms or an abnormal chest examination along with abnormalities suggestive of ILD on the HRCT. The controls with RA, without a clinically significant ILD were enrolled consecutively from the Rheumatology Clinic. Subjects were excluded from either study group if they had any of the following: (i) an overlap of RA with another CTD; (ii) the ILD could be attributed to causes other than RA such as drug toxicity or smoking related ILD; or (iii) they refused consent (Fig. 1). We suspected drug induced ILD if respiratory symptoms appeared while on methotrexate treatment (especially, when they appeared with a subacute onset and within the first couple of years of starting methotrexate), and the chest HRCT showed a pattern resembling hypersensitivity pneumonitis, nonspecific interstitial pneumonia (NSIP), or organizing pneumonia. Subjects presenting with an insidious onset of symptoms and having a usual interstitial pneumonia (UIP) pattern were in general considered to have RA-ILD.16 We considered smoking-related ILD if the chest HRCT showed centrilobular nodules (suggesting respiratory bronchiolitis ILD) or basal, peripheral ground-glass opacities with cystic changes (suggesting desquamative interstitial pneumonia) in the presence of history of tobacco smoking.
Fig. 1.
Flowchart depicting the methodology of the study. CTD, connective tissue disease; DAS28, 28-joint disease activity score; HRCT, high-resolution computed tomography; ILD, interstitial lung disease; PF, progressive fibrosing; RA, rheumatoid arthritis.
We diagnosed RA according to the 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) criteria.17 The presence of ILD and its pattern was identified on HRCT of the chest in symptomatic subjects. A multidisciplinary meeting between the pulmonologist, rheumatologist, and radiologist was held for classifying subjects, whenever there were competing diagnostic differentials.18
Objectives
The primary objective of the study was to evaluate the demographic, clinical, immunological and physiological factors associated with the presence of ILD. The secondary objective was to analyze the factors associated with the PF phenotype. We used the standard definition to define the PF phenotype; that is, the presence of HRCT signs of fibrosis (reticular abnormality with traction bronchiectasis, with or without honeycombing, with a disease extent of >10%) with at least one of the criteria suggestive of progression that developed over 24 months or less (a relative decline of ≥10% in the FVC; a relative decline in the FVC of 5%–10% along with a worsening of respiratory symptoms or an increased extent of fibrosis on high-resolution CT; or, a worsening of respiratory symptoms along with an increased extent of fibrosis).13,19
Study procedures
We collected the details of symptoms and their duration, comorbidities, history of smoke exposure and performed a detailed physical examination. All participants underwent laboratory investigations including RF, ACPA, antinuclear antibodies, chest radiograph, spirometry, and a 6-min walk test. We also calculated the 28-joint disease activity score (DAS28) for all study participants.20 In addition, we performed HRCT chest and assessed the diffusion capacity for carbon monoxide (DLCO) for the case subjects. We recorded the findings on echocardiography, if available. Bronchoscopy and lung biopsy were performed only if there was a clinical possibility of a cause of ILD other than RA itself. We excluded the presence of clinically significant ILD among the control subjects by the absence of respiratory symptoms and the presence of normal chest examination, chest radiograph, and spirometry. We followed up the control subjects every 6 months for up to 2 years with a screening for new-onset respiratory symptoms.
Computed tomography scans were obtained at end-inspiration with thin-collimation sections (2 mm or less) without the administration of contrast. We calculated the scores of various HRCT features of ILD using a system (Table 1) adapted from a validated system described by Salisbury et al.21,22 We made only minor changes in the system in that we did not score CT scans for centrilobular nodules, and we did not score for air trapping separately from mosaic attenuation. Two radiologists blinded to the clinical data interpreted each HRCT independently using the same images and recorded the findings on standardized scoring sheets. Using standard definitions, the radiologic features of traction bronchiectasis or traction bronchiolectasis, honeycombing, reticular pattern, ground glass, mosaic attenuation and/or air trapping were scored semi-quantitatively in each of the five lobes.23 For each feature, the lobe scores were summed and divided by 5 to obtain an average (scale of 0–5), indicative of the proportion of the total lung parenchyma showing the feature. We created a consensus semi-quantitative score as a two-reader mean of the average score for each feature. Furthermore, we created an aggregate fibrosis score by adding the scores for honeycombing and traction bronchiectasis and dividing the sum by 2. We also generated an individual reader and a consensus dichotomous variable for each feature. Thus, the feature was considered “present” if the respective individual or consensus semi-quantitative two-reader mean score was >0.5. Finally, the radiologists separately recorded the pattern of ILD in each case. In case, the patterns described by the two radiologists differed, a consensus was reached by discussion.
Table 1.
Details of the system for recording and scoring of high-resolution computed tomography characteristics of case subjects.
| Characteristic | Categories |
|---|---|
| Quality of HRCT scan | Excellent, diagnostic, or non-diagnostic |
| Craniocaudal distribution | Upper lung, lower lung, or diffuse |
| Axial distribution | Central, peripheral, subpleural, peribronchovascular, or diffuse |
| Abnormalities scored | Traction bronchiectasis or traction bronchiolectasis Honeycombing Reticular pattern Ground glass Mosaic attenuation and/or air trapping |
| Regions scored | Right upper lobe Right middle lobe Right lower lobe Left upper lobe/lingula Left lower lobe |
| Semi-quantitative scores | 0 indicates no involvement 1 represents <5% of the lobe involved (present but minimal) 2 reflects 5–25% 3 indicates 25–49% 4 signifies 50–75% 5 denotes >75% involvement |
HRCT, high-resolution computed tomography.
We followed up the subjects with RA-ILD in our Chest Clinic with clinical assessment, lung functions and other laboratory testing at visits planned at 6-month intervals. We identified ILD subjects, who developed the PF phenotype. We then compared the demographic, clinical, laboratory, and radiologic characteristics between subjects with or without PF-ILD.
Statistical analysis
Data were analyzed using the statistical package SPSS (version 23.0, IBM Inc., Armonk, NY, United States). This was an observational, exploratory study, so a formal sample size calculation was not performed. For the cases, we chose a convenience sample of 30 consecutive subjects with RA-ILD based on the number of subjects expected during the pre-specified study period. We enrolled control subjects in a 1:1 ratio to cases. All data are expressed as a number with percentage, mean with standard deviation (SD), or median with interquartile range, as appropriate. The differences between categorical and continuous variables were analyzed using the chi-square test (or the Fisher's exact test), and the Student's t-test (or Mann–Whitney U-test), respectively. We calculated the weighted Cohen's kappa (ĸ) to assess the interobserver agreement between the radiologists for the scoring of various CT abnormalities. Univariate and multivariate Cox regression analyses were performed to study the predictors of ILD and the odds ratio and adjusted odds ratio were calculated. Factors with p-values < 0.25 in the univariate analysis were entered into the multivariate analysis. Statistical significance was assumed at a p-value <0.05 for all results.
Results
Of the 67 subjects screened, we enrolled 30 cases and 30 control subjects. Four subjects with RA-ILD were excluded (two refused consent, one each had an overlap with systemic sclerosis (SSc) and inflammatory myositis, respectively). Three subjects were excluded from the control group (two refused consent and one had an overlap with Sjogren's syndrome). One of the two subjects who refused consent had breathlessness but a normal physical examination and chest radiograph. The subject was advised to undergo a chest CT and spirometry but did not follow up. One of the case subjects was diagnosed with ILD during evaluation for an initially expected inclusion into the control group. The cases were significantly older (mean age, 57.8 vs. 50.2 years; p = 0.004) than the controls (Table 2). Subjects with RA-ILD had significantly lower BMI (p = 0.002) and a longer duration of RA disease (mean duration, 10.1 years vs. 5.5 years; p = 0.04) than the controls. Numerically, more cases (89.3%) had a DAS28 score >3.2 suggesting active inflammatory disease than the controls (67.9%; p = 0.10). Age (p = 0.007) was the only factor associated with ILD on univariate regression (Table 3). On multivariate analysis, only the duration of RA was independently associated with ILD (p = 0.049) after adjusting for other factors including age, gender, and DAS28.
Table 2.
Comparison of the baseline characteristics and treatment between subjects with rheumatoid arthritis with or without interstitial lung disease (n = 60).
| Parameter | RA-ILD (n = 30) | RA without ILD (n = 30) | All subjects (n = 60) | p-value |
|---|---|---|---|---|
| Age, years | 57.8 ± 10.2 | 50.2 ± 9.2 | 54.0 ± 10.4 | 0.004 |
| Men, number (%) | 8 (26.7) | 4 (13.3) | 12 (20.0) | 0.33 |
| Body mass index, kg/m2 | 22.9 ± 4.1 | 26.3 ± 3.6 | 24.6 ± 4.2 | 0.002 |
| Any smoke exposure | 22 (73.3) | 24 (80.0) | 46 (76.6) | 0.54 |
| Tobacco smoking | 4 (13.3) | 4 (13.3) | 8 (13.3) | 1.00 |
| Biomass smoke | 19 (63.3) | 21 (70.0) | 40 (66.7) | 0.58 |
| Comorbid illnesses | ||||
| Any comorbidity | 10 (33.3) | 8 (26.7) | 18 (30.0) | 0.57 |
| Hypertension | 8 (26.7) | 5 (16.7) | 13 (21.7) | 0.35 |
| Diabetes mellitus | 2 (6.7) | 2 (6.7) | 4 (6.7) | 1.00 |
| Hypothyroidism | 0 | 1 (3.3) | 1 (1.7) | 1.00 |
| Coronary artery disease | 0 | 1 (3.3) | 1 (1.7) | 1.00 |
| Chronic kidney disease | 1 (3.3) | 0 | 1 (1.7) | 1.00 |
| Chronic liver disease | 0 | 0 | 0 | – |
| Cerebrovascular disease | 0 | 0 | 0 | – |
| Gastroesophageal reflux | 7 (23.3) | 10 (33.3) | 17 (28.3) | 0.39 |
| Duration of RA in years, mean (range) | 10.1 (0.2–40) | 5.5 (0.4–23) | 7.8 (0.2–40) | 0.04 |
| Autoantibodies | ||||
| Antinuclear antibody | 14 (53.8) | 11 (50.0) | 25 (52.1) | 0.79 |
| Rheumatoid factor | 0.99 | |||
| <ULN | 2 (7.4) | 2 (8.0) | 4 (7.7) | |
| 1–3 X ULN | 3 (11.1) | 3 (12.0) | 6 (11.5) | |
| >3 X ULN | 22 (81.5) | 20 (80.0) | 42 (80.8) | |
| Not available | 3 | 5 | 8 | |
| ACPA | 0.46 | |||
| <ULN | 3 (10.7) | 1 (4.0) | 4 (7.5) | |
| 1–3 X ULN | 3 (10.7) | 5 (20.0) | 8 (15.1) | |
| >3 X ULN | 22 (78.6) | 19 (76.0) | 41 (77.4) | |
| Not available | 2 | 5 | 7 | |
| Rheumatoid disease activity | (n = 28) | (n = 28) | (n = 56) | |
| DAS28 Score | 4.36 ± 0.79 | 3.99 ± 1.49 | 4.18 ± 1.19 | 0.25 |
| DAS28 ≥ 3.2 | 25 (89.3) | 19 (67.9) | 44 (78.6) | 0.10 |
| Spirometric parameters | ||||
| Type of abnormality | <0.001 | |||
| Normal | 11 (39.3) | 30 (100.0) | 41 (70.7) | |
| Restrictive defect | 17 (60.7) | 0 | 17 (29.3) | |
| Obstructive defect | 0 | 0 | 0 | |
| Could not perform | 2 | 0 | 2 | |
| FVC, litres | 1.90 ± 0.70 | 2.62 ± 0.65 | 2.27 ± 0.76 | <0.001 |
| FVC, % predicted | 71.3 ± 18.3 | 95.9 ± 11.9 | 83.9 ± 19.6 | <0.001 |
| FEV1, litres | 1.59 ± 0.58 | 2.08 ± 0.52 | 1.84 ± 0.59 | 0.001 |
| FEV1, % predicted | 78.7 ± 20.7 | 96.7 ± 13.8 | 88.0 ± 19.6 | <0.001 |
| Six-minute walk distance, meters | 352 ± 97 | 400 ± 68 | 377 ± 87 | 0.03 |
| Drugs received | ||||
| Methotrexate | 14 (46.7) | 28 (93.3) | 42 (0.7) | |
| Prednisolone | 28 (93.3) | 14 (46.7) | 42 (0.7) | |
| Hydroxychloroquine | 15 (50) | 19 (63.3) | 34 (56.7) | |
| Sulfasalazine | 7 (23.3) | 5 (16.7) | 12 (20.0) | |
| Leflunomide | 9 (30.0) | 3 (10.0) | 12 (20.0) | |
| Azathioprine | 6 (20.0) | 0 | 6 (10.0) | |
| Cyclophosphamide | 4 (13.3) | 0 | 4 (6.7) | |
| High-dose methylprednisolone | 1 (3.3) | 0 | 1 (1.7) | |
All values represent mean ± standard deviation or number (percentage), unless otherwise specified. ACPA, anti-citrullinated protein antibody; DAS28-28, joint disease activity score; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; ILD, interstitial lung disease; RA, rheumatoid arthritis; ULN, upper limit of normal.
Table 3.
Factors associated with interstitial lung disease in rheumatoid arthritis on univariate and multivariate logistic regression analysis.
| Characteristic | Odds ratio (95% CI) | P-value | Adjusted odds ratio (95% CI) | P-value |
|---|---|---|---|---|
| Age | 1.08 (1.02–1.15) | 0.007 | 1.06 (0.99–1.13) | 0.08 |
| Male gender | 2.36 (0.63–8.92) | 0.20 | 3.33 (0.72–15.43) | 0.12 |
| Smoking | 1.00 (0.23–4.43) | 1.00 | ||
| Duration of rheumatoid arthritis | 1.08 (0.99–1.16) | 0.052 | 1.09 (1.00–1.19) | 0.049 |
| DAS28 | 1.31 (0.83–2.07) | 0.24 | 1.18 (0.69–2.01) | 0.54 |
| High-titer RF | 1.39 (0.37–5.29) | 0.63 | ||
| High-titer ACPA | 1.16 (0.32–4.19) | 0.82 |
ACPA, anti-citrullinated protein antibody; DAS28, 28-joint disease activity score; RF, rheumatoid factor. Only the factors with a p-value < 0.25 on univariate analysis were entered into the multivariable regression model.
The interobserver agreement between the radiologists was moderate for honeycombing (ĸ = 0.66), traction bronchiectasis (ĸ = 0.63), ground-glass opacities (ĸ = 0.65), and mosaic attenuation (ĸ = 0.78), while it was weak for reticulation (ĸ = 0.37). Sixteen (53.3%), 11 (36.7%), and 1 (3.3%) subject with ILD had UIP, NSIP, and organizing pneumonia patterns, respectively, on the HRCT scans, while the pattern was indeterminate in two (6.6%). The subjects with a UIP pattern had a lower % predicted FVC compared with those with a non-UIP pattern (mean % predicted FVC, 74% vs. 58%; p = 0.04). Subjects with ILD were treated with prednisolone with or without other immunosuppressants (Table 2). Among the subjects with ILD, followed up for a median (IQR) duration of 23 (17–26) months, six (20%) developed the PF phenotype (Table 4). Those with this phenotype were older and had higher scores for honeycombing and fibrosis than those without (Table 5). Among 12 subjects with honeycombing, 41.7% developed this phenotype compared with only one (5.6%) of 18 with no honeycombing (p = 0.03). Five (83.3%) subjects with PF-ILD had a UIP pattern (4 definite, 1 probable). Two subjects with an NSIP pattern had worsening disease but did not fulfil the criteria for a fibrosing ILD. One of them died.
Table 4.
Details of subjects with rheumatoid arthritis-associated interstitial lung disease having the progressive fibrosing phenotype.
| Subject | Age, years | Treatment received | HRCT pattern of ILD | Parameters suggestive of physiological progression |
|---|---|---|---|---|
| 1 | 65 | Prednisolone, methotrexate | Definite UIP | FVC declined from 1.12 L to 1.01 L over 3 months |
| 2 | 79 | Prednisolone | Indeterminate | FVC declined from 2.41 L to 1.83 L over 11 months |
| 3 | 57 | Prednisolone, methotrexate, leflunomide, hydroxychloroquine, cyclophosphamide, sulfasalazine | Definite UIP | FVC declined from 1.44 L to 1.23 L and DLCO declined from 7.15 mL/min/mm Hg to 4.91 mL/min/mm Hg over 7 months |
| 4 | 68 | Prednisolone, leflunomide | Probable UIP | FVC declined from 2.46 L to 2.19 L and DLCO declined from 17.88 mL/min/mm Hg to 11.60 mL/min/mm Hg over 5 months |
| 5 | 60 | Prednisolone, azathioprine | Definite UIP | FVC declined from 1.74 L to 1.40 L over 15 months |
| 6 | 62 | Prednisolone, sulfasalazine | Definite UIP | FVC declined from 2.25 L to 1.84 L over 23 months |
DLCO, diffusion capacity for carbon monoxide; FVC, forced vital capacity; HRCT, high-resolution computed tomography; ILD, interstitial lung disease; L, liter; min, minute; mL, milliliter; RA, rheumatoid arthritis.
Table 5.
Comparison of characteristics between subjects with rheumatoid arthritis-related interstitial lung disease with or without a progressive fibrosing phenotype (n = 30).
| Parameter | PF phenotype (n = 6) | No PF phenotype (n = 24) | All RA-ILD subjects (n = 30) | p-value |
|---|---|---|---|---|
| Age, years | 65.2 ± 7.8 | 55.9 ± 10.1 | 57.8 ± 10.2 | 0.047 |
| Men, number (%) | 2 (33.3) | 6 (25.0) | 8 (26.7) | 0.65 |
| BMI, kg/m2 | 21.9 ± 4.8 | 23.3 ± 3.9 | 22.9 ± 4.1 | 0.48 |
| Any smoke exposure | 3 (50.0) | 19 (79.2) | 22 (73.3) | 0.30 |
| Tobacco smoking | 1 (16.7) | 3 (12.5) | 4 (13.3) | 1.00 |
| Biomass smoke | 3 (50.0) | 16 (66.7) | 19 (63.3) | 0.64 |
| Any comorbid illnesses | 2 (33.3) | 8 (33.3) | 10 (33.3) | 1.00 |
| Presence of GER | 2 (33.3) | 5 (20.8) | 7 (23.3) | 0.60 |
| Duration of RA, years | 1.1 (0.4–14.4) | 8.5 (0.9–17.1) | 6.8 (0.7–16.5) | 0.30 |
| Autoantibodies | ||||
| Antinuclear antibody | 3 (50.0) | 11 (55.0) | 14 (53.8) | 1.00 |
| High-titer RF | 4/5 (80.0) | 18/22 (81.8) | 22/27 (81.5) | 1.00 |
| High-titer ACPA | 3/5 (60.0) | 19/23 (82.6) | 22/28 (78.6) | 0.29 |
| RA disease activity | (n = 5) | (n = 23) | (n = 28) | |
| DAS28 Score | 4.25 ± 1.19 | 4.39 ± 0.71 | 4.36 ± 0.79 | 0.75 |
| DAS28 ≥ 3.2 | 4 (80.0) | 21 (91.3) | 25 (89.3) | 0.46 |
| Pulmonary function tests | ||||
| FVC, litres | 1.87 ± 0.60 | 1.91 ± 0.74 | 1.90 ± 0.70 | 0.90 |
| FVC, % predicted | 72.3 ± 17.3 | 71.0 ± 18.9 | 71.3 ± 18.3 | 0.88 |
| FEV1, litres | 1.58 ± 0.54 | 1.59 ± 0.59 | 1.59 ± 0.58 | 0.97 |
| FEV1, % predicted | 81.1 ± 20.2 | 78.0 ± 21.3 | 78.7 ± 20.7 | 0.75 |
| DLCO, % predicted | 59.5 ± 21.2 (n = 4) | 58.9 ± 27.4 (n = 20) | 59.0 ± 26.0 (n = 24) | 0.97 |
| HRCT features | ||||
| Pattern | 0.16 | |||
| Definite UIP | 4 (66.7) | 6 (25.0) | 10 (33.3) | |
| Probable UIP | 1 (16.7) | 5 (20.8) | 6 (20.0) | |
| NSIP | 0 | 11 (45.8) | 11 (36.7) | |
| OP | 0 | 1 (4.2) | 1 (3.3) | |
| Indeterminate | 1 (16.7) | 1 (4.2) | 2 (6.6) | |
| Semi-quantitative scores | ||||
| Honeycombing | 1.75 ± 1.19 | 0.43 ± 0.61 | 0.69 ± 0.30 | 0.04 |
| TBx | 1.13 ± 0.71 | 0.80 ± 0.80 | 0.87 ± 0.78 | 0.36 |
| Fibrosis score | 1.50 ± 0.85 | 0.62 ± 0.56 | 1.59 ± 1.42 | 0.004 |
| Reticulation | 1.05 ± 0.85 | 1.10 ± 0.72 | 1.09 ± 0.73 | 0.88 |
| Ground glass | 0.23 ± 0.35 | 0.97 ± 1.29 | 0.82 ± 1.19 | 0.19 |
| Mosaic attenuation | 0 | 0.12 ± 0.28 | 0.10 ± 0.26 | 0.05 |
| Dichotomous scores | ||||
| Honeycombing | 5 (83.3) | 7 (29.2) | 12 (40.0) | 0.03 |
| TBx | 6 (100.0) | 13 (54.2) | 19 (63.3) | 0.06 |
| Honeycombing or TBx | 6 (100.0) | 16 (66.7) | 22 (73.3) | 0.16 |
| Reticulation | 4 (66.7) | 20 (83.3) | 24 (80.0) | 0.57 |
| Ground glass | 2 (33.3) | 11 (45.8) | 13 (43.3) | 0.67 |
| Mosaic attenuation | 0 | 3 (12.5) | 3 (10.0) | 1.00 |
All values represent mean ± standard deviation, median (interquartile range), or number (percentage). BMI, body mass index; ACPA, anti-citrullinated protein antibody; DAS28, 28-joint disease activity score; DLCO, diffusion capacity for carbon monoxide; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; GER, gastroesophageal reflux; HRCT, high-resolution computed tomography; ILD, interstitial lung disease; NSIP, nonspecific interstitial pneumonia; OP, organizing pneumonia; PF, progressive fibrosing; RA, rheumatoid arthritis; TBx, traction bronchiectasis or traction bronchiolectasis, UIP, usual interstitial pneumonia.
There were four deaths (three cases, one control). All three case subjects (two had the PF-phenotype) developed an acute exacerbation of ILD. One control subject died of myocardial infarction. Two subjects with RA-ILD developed tuberculosis, while one each developed pneumonia and pulmonary aspergillosis, respectively. All these subjects recovered with appropriate therapy. All other RA-ILD subjects had stable symptoms at the time of the last follow-up. Of the 10 RA-ILD subjects, for whom echocardiography findings were available, five had pulmonary hypertension (three mild, one moderate, and one severe). No control subject developed any respiratory symptoms over a median follow-up of 17 months.
Discussion
We found the duration of RA to be the only significant factor associated with ILD in RA. A PF phenotype emerged in 20% patients and was associated with older age, the presence of honeycombing on HRCT chest, and higher scores for honeycombing and aggregate fibrosis. To our knowledge, this is the first study on the factors associated with the recently defined PF phenotype in RA-ILD. This is also the first study from India presenting a detailed profile and longitudinal follow-up of patients with RA-ILD.
The risk factors for ILD in patients with RA vary across studies (Supplementary Table 1).3, 4, 5, 6, 7,24, 25, 26, 27 We found the duration of RA to be the only significant factor associated with ILD in the multivariate analysis, although age was a significant factor in the univariate analysis. Our patients with RA-ILD also had a numerically higher frequency of active disease (DAS28 > 3.2) compared with those without ILD. Our results are like those of Restrepo et al, who also found the duration of RA and age to be significant factors independently associated with ILD.7 Their subjects with ILD also had higher DAS28 scores compared with controls. It has been suggested that ILD in RA occurs as a result of an immune response against post-translationally modified proteins occurring in the joints with subsequent cross-reactivity with similar antigen targets in the lungs28 A longer duration and higher activity of joint disease might make the triggering of such an immune response more likely. Our subjects with RA-ILD had a lower BMI which may be secondary to factors such as disease duration, disease activity, comorbidities, and drug adverse effects. We did not consider it as a covariate in the regression analysis because the differences in BMI may be secondary to the above factors. Moreover, too many covariates in a regression model with a small sample size may reduce the validity of the analysis.
We found that 6 (20%) of our subjects with RA-ILD developed the PF phenotype, consistent with previous data, which indicates that 20–30% of patients with ILD develop this phenotype.29,30 Importantly, among patients having honeycombing on the HRCT, about 40% developed the PF phenotype. In previous studies using various older definitions, 23–52% subjects were found to develop progressive fibrosis.8,31,32 Other studies have reported that older age, male gender, lower DLCO, RA disease activity score, high-titer RF, UIP pattern, presence of honeycombing and the extent of reticulation, traction bronchiectasis, honeycombing, and fibrosis are significant predictors of mortality in RA-ILD.8,31,33,34 Our study suggests that the same features of honeycombing and traction bronchiectasis might also be the risk factors for the PF phenotype. As this specific phenotype has been defined recently, no previous study has reported on it. Crucially, two of our subjects with an NSIP pattern also had severe lung disease; one of them died of an acute exacerbation. Both the subjects did not fit into the description of PF-ILD, as per the contemporary definition. Thus, some patients with a worsening ILD, who do not have clear signs of significant radiologic fibrosis such as traction bronchiectasis or honeycombing on the HRCT might not be identified as PF-ILD, yet may progress rapidly.
What are the clinical implications of our study? There are currently no large randomized clinical trials to guide the treatment of patients with RA-ILD. The prevailing clinical practice is to treat them with immunosuppressive drugs, as is done in SSc-ILD.2,35 In the recent SENSCIS trial, an antifibrotic agent, nintedanib was found to significantly reduce the rate of decline of lung function in patients with SSc-ILD.36 In the INBUILD trial, nintedanib reduced the decline in forced vital capacity in a diverse group of subjects with PF-ILDs, including those with various CTD-ILDs.13 The benefit was even more pronounced for subjects with a UIP pattern. Patients with RA-ILD are more likely to have a UIP pattern (about 43% patients) compared with those with SSc-ILD (about 10%).37,38 Therefore, we believe that RA-ILD, with its higher frequency of honeycombing and the UIP pattern may respond even more favorably to antifibrotic agents, especially, if a PF phenotype is present. The knowledge of the factors associated with the PF phenotype makes it easier to identify patients with an elevated risk of progression. Such patients may be monitored more closely and selected for an early institution of antifibrotic therapy, rather than waiting too long on immunosuppressive drugs alone. Notably, four of our subjects developed a lung infection (pneumonia, tuberculosis, or aspergillosis), which highlights the harms of the use of prolonged immunosuppressive therapy. It cannot be overemphasized that randomized controlled trials of antifibrotic agents for RA-ILD are the need of the hour.
This study has several limitations. It is a single-center study with a small sample size. We enrolled a small number of controls and did not match them to the cases for any baseline factors. Including only matched controls could have highlighted certain other differences between the study groups rather than age and disease duration. We did not perform HRCT chest in control subjects to confirm the absence of ILD. We also did not repeat detailed assessment of the control subjects during follow-up for the presence of new-onset ILD, except for symptom review. As the incidence of new-onset ILD among subjects with RA is low (1.3–5.0/1000 person-years), the probability of the control subjects developing new-onset ILD would be low.4,39 In addition, it is possible that some of our subjects with a stable course during the study period could develop a PF phenotype in future. We did not perform a multivariate analysis for the predictors of the PF phenotype in view of the small numbers. The current investigation is, thus, only a hypothesis-generating study. It is imperative that longer, larger, and multicentric studies are performed to identify the risk factors for the PF phenotype with greater certainty. Interestingly, a study is ongoing, with a plan to follow-up subjects with various fibrosing ILDs (including 50 patients with RA-ILD) longitudinally, to identify the biomarkers of progressive pulmonary fibrosis.40
In conclusion, the duration of RA was an important factor associated with ILD in our study population. Older age, the presence of honeycombing on HRCT chest, and higher scores for honeycombing and aggregate fibrosis were associated with the PF phenotype that emerged in 20% of the subjects with RA-ILD.
Patients/ Guardians/ Participants consent
Patients informed consent was obtained.
Ethical clearance
Institute/hospital ethical clearance certificate was obtained.
Source of support
Nil.
Disclosure of competing interest
The authors have none to declare.
Acknowledgment
None.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.mjafi.2022.08.004.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Dhooria S., Sehgal I.S., Agarwal R., et al. Incidence, prevalence, and national burden of interstitial lung diseases in India: estimates from two studies of 3089 subjects. PLoS One. 2022;17:e0271665. doi: 10.1371/journal.pone.0271665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dhooria S., Singh N. In: Rheumatology Clinics Rheumatoid Arthritis. Sharma Aman, Handa Rohini., editors. Evangel Publishing; 2022. Chapter 16: Rheumatoid arthritis and the lung. [Google Scholar]
- 3.Gabbay E., Tarala R., Will R., et al. Interstitial lung disease in recent onset rheumatoid arthritis. Am J Respir Crit Care Med. 1997;156:528–535. doi: 10.1164/ajrccm.156.2.9609016. [DOI] [PubMed] [Google Scholar]
- 4.Bongartz T., Nannini C., Medina-Velasquez Y.F., et al. Incidence and mortality of interstitial lung disease in rheumatoid arthritis: a population-based study. Arthritis Rheum. 2010;62:1583–1591. doi: 10.1002/art.27405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Koduri G., Norton S., Young A., et al. Interstitial lung disease has a poor prognosis in rheumatoid arthritis: results from an inception cohort. Rheumatology. 2010;49:1483–1489. doi: 10.1093/rheumatology/keq035. [DOI] [PubMed] [Google Scholar]
- 6.Kelly C.A., Saravanan V., Nisar M., et al. Rheumatoid arthritis-related interstitial lung disease: associations, prognostic factors and physiological and radiological characteristics–a large multicentre UK study. Rheumatology. 2014;53:1676–1682. doi: 10.1093/rheumatology/keu165. [DOI] [PubMed] [Google Scholar]
- 7.Restrepo J.F., del Rincon I., Battafarano D.F., Haas R.W., Doria M., Escalante A. Clinical and laboratory factors associated with interstitial lung disease in rheumatoid arthritis. Clin Rheumatol. 2015;34:1529–1536. doi: 10.1007/s10067-015-3025-8. [DOI] [PubMed] [Google Scholar]
- 8.Jacob J., Hirani N., van Moorsel C.H.M., et al. Predicting outcomes in rheumatoid arthritis related interstitial lung disease. Eur Respir J. 2019;53 doi: 10.1183/13993003.00869-2018. pii: 13993003.00869-2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Brown K.K., Martinez F.J., Walsh S.L.F., et al. The natural history of progressive fibrosing interstitial lung diseases. Eur Respir J. 2020;55 doi: 10.1183/13993003.00085-2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Richeldi L., du Bois R.M., Raghu G., et al. Efficacy and safety of nintedanib in idiopathic pulmonary fibrosis. N Engl J Med. 2014;370:2071–2082. doi: 10.1056/NEJMoa1402584. [DOI] [PubMed] [Google Scholar]
- 11.Dhooria S., Agarwal R., Sehgal I.S., et al. A real-world study of the dosing and tolerability of pirfenidone and its effect on survival in idiopathic pulmonary fibrosis. Sarcoidosis Vasc Diffuse Lung Dis. 2020;37:148–157. doi: 10.36141/svdld.v37i2.8718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Maher T.M., Corte T.J., Fischer A., et al. Pirfenidone in patients with unclassifiable progressive fibrosing interstitial lung disease: a double-blind, randomised, placebo-controlled, phase 2 trial. Lancet Respir Med. 2020;8:147–157. doi: 10.1016/S2213-2600(19)30341-8. [DOI] [PubMed] [Google Scholar]
- 13.Flaherty K.R., Wells A.U., Cottin V., et al. Nintedanib in progressive fibrosing interstitial lung diseases. N Engl J Med. 2019;381:1718–1727. doi: 10.1056/NEJMoa1908681. [DOI] [PubMed] [Google Scholar]
- 14.Acharya N., Sharma S.K., Mishra D., Dhooria S., Dhir V., Jain S. Efficacy and safety of pirfenidone in systemic sclerosis-related interstitial lung disease-a randomised controlled trial. Rheumatol Int. 2020;40:703–710. doi: 10.1007/s00296-020-04565-w. [DOI] [PubMed] [Google Scholar]
- 15.Dhooria S., Agarwal R., Gupta D. Is pirfenidone ready for use in non-idiopathic pulmonary fibrosis interstitial lung diseases? Lung India. 2015;32:4–5. doi: 10.4103/0970-2113.148396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Malaviya A.N. Does methotrexate cause interstitial lung disease in rheumatoid arthritis: what is the evidence? Int J Rheum Dis. 2020;23:713–716. doi: 10.1111/1756-185X.13828. [DOI] [PubMed] [Google Scholar]
- 17.Aletaha D., Neogi T., Silman A.J., et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League against Rheumatism collaborative initiative. Ann Rheum Dis. 2010;69:1580–1588. doi: 10.1136/ard.2010.138461. [DOI] [PubMed] [Google Scholar]
- 18.Dhooria S., Mehta R.M., Srinivasan A., et al. The safety and efficacy of different methods for obtaining transbronchial lung cryobiopsy in diffuse lung diseases. Clin Res J. 2018;12:1711–1720. doi: 10.1111/crj.12734. [DOI] [PubMed] [Google Scholar]
- 19.Cottin V., Hirani N.A., Hotchkin D.L., et al. Presentation, diagnosis and clinical course of the spectrum of progressive-fibrosing interstitial lung diseases. Eur Respir Rev. 2018;27 doi: 10.1183/16000617.0076-2018. pii: 27/150/180076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Disease Activity Score-28 for Rheumatoid Arthritis with ESR (DAS28-ESR) calculator available at: https://www.mdcalc.com/disease-activity-score-28-rheumatoid-arthritis-esr-das28-esr. Last accessed on 31st March 2020.
- 21.Salisbury M.L., Gu T., Murray S., et al. Hypersensitivity pneumonitis: radiologic phenotypes are associated with distinct survival time and pulmonary function trajectory. Chest. 2019;155:699–711. doi: 10.1016/j.chest.2018.08.1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Salisbury M.L., Gross B.H., Chughtai A., et al. Development and validation of a radiological diagnosis model for hypersensitivity pneumonitis. Eur Respir J. 2018;52:1800443. doi: 10.1183/13993003.00443-2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hansell D.M., Bankier A.A., MacMahon H., McLoud T.C., Muller N.L., Remy J. Fleischner Society: glossary of terms for thoracic imaging. Radiology. 2008;246:697–722. doi: 10.1148/radiol.2462070712. [DOI] [PubMed] [Google Scholar]
- 24.Zhang Y., Li H., Wu N., Dong X., Zheng Y. Retrospective study of the clinical characteristics and risk factors of rheumatoid arthritis-associated interstitial lung disease. Clin Rheumatol. 2017;36:817–823. doi: 10.1007/s10067-017-3561-5. [DOI] [PubMed] [Google Scholar]
- 25.Kiely P., Busby A.D., Nikiphorou E., et al. Is incident rheumatoid arthritis interstitial lung disease associated with methotrexate treatment? Results from a multivariate analysis in the ERAS and ERAN inception cohorts. BMJ Open. 2019;9:e028466. doi: 10.1136/bmjopen-2018-028466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Salaffi F., Carotti M., Di Carlo M., Tardella M., Giovagnoni A. High-resolution computed tomography of the lung in patients with rheumatoid arthritis: prevalence of interstitial lung disease involvement and determinants of abnormalities. Medicine (Baltim) 2019;98:e17088. doi: 10.1097/MD.0000000000017088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mena-Vazquez N., Perez Albaladejo L., Manrique-Arija S., et al. Analysis of clinical-analytical characteristics in patients with rheumatoid arthritis and interstitial lung disease: case-control study. Reumatol Clínica. 2021;17:197–202. doi: 10.1016/j.reuma.2019.06.001. [DOI] [PubMed] [Google Scholar]
- 28.Brito Y., Glassberg M.K., Ascherman D.P. Rheumatoid arthritis-associated interstitial lung disease: current concepts. Curr Rheumatol Rep. 2017;19:79. doi: 10.1007/s11926-017-0701-5. [DOI] [PubMed] [Google Scholar]
- 29.Wong A.W., Ryerson C.J., Guler S.A. Progression of fibrosing interstitial lung disease. Respir Res. 2020;21:32. doi: 10.1186/s12931-020-1296-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zamora-Legoff J.A., Krause M.L., Crowson C.S., Ryu J.H., Matteson E.L. Progressive decline of lung function in rheumatoid arthritis-associated interstitial lung disease. Arthritis Rheumatol. 2017;69:542–549. doi: 10.1002/art.39971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hyldgaard C., Ellingsen T., Hilberg O., Bendstrup E. Rheumatoid arthritis-associated interstitial lung disease: clinical characteristics and predictors of mortality. Respiration. 2019;98:455–460. doi: 10.1159/000502551. [DOI] [PubMed] [Google Scholar]
- 32.Li L., Gao S., Fu Q., et al. A preliminary study of lung abnormalities on HRCT in patients of rheumatoid arthritis-associated interstitial lung disease with progressive fibrosis. Clin Rheumatol. 2019;38:3169–3178. doi: 10.1007/s10067-019-04673-4. [DOI] [PubMed] [Google Scholar]
- 33.Assayag D., Lubin M., Lee J.S., King T.E., Collard H.R., Ryerson C.J. Predictors of mortality in rheumatoid arthritis-related interstitial lung disease. Respirology. 2014;19:493–500. doi: 10.1111/resp.12234. [DOI] [PubMed] [Google Scholar]
- 34.Singh N., Varghese J., England B.R., et al. Impact of the pattern of interstitial lung disease on mortality in rheumatoid arthritis: a systematic literature review and meta-analysis. Semin Arthritis Rheum. 2019;49:358–365. doi: 10.1016/j.semarthrit.2019.04.005. [DOI] [PubMed] [Google Scholar]
- 35.Naidu G., Sharma S.K., Adarsh M.B., et al. Effect of mycophenolate mofetil (MMF) on systemic sclerosis-related interstitial lung disease with mildly impaired lung function: a double-blind, placebo-controlled, randomized trial. Rheumatol Int. 2020;40:207–216. doi: 10.1007/s00296-019-04481-8. [DOI] [PubMed] [Google Scholar]
- 36.Distler O., Highland K.B., Gahlemann M., et al. Nintedanib for systemic sclerosis-associated interstitial lung disease. N Engl J Med. 2019;380:2518–2528. doi: 10.1056/NEJMoa1903076. [DOI] [PubMed] [Google Scholar]
- 37.Bendstrup E., Moller J., Kronborg-White S., Prior T.S., Hyldgaard C. Interstitial lung disease in rheumatoid arthritis remains a challenge for clinicians. J Clin Med. 2019;8 doi: 10.3390/jcm8122038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Perelas A., Silver R.M., Arrossi A.V., Highland K.B. Systemic sclerosis-associated interstitial lung disease. Lancet Respir Med. 2020;8:304–320. doi: 10.1016/S2213-2600(19)30480-1. [DOI] [PubMed] [Google Scholar]
- 39.Shidara K., Hoshi D., Inoue E., et al. Incidence of and risk factors for interstitial pneumonia in patients with rheumatoid arthritis in a large Japanese observational cohort, IORRA. Mod Rheumatol. 2010;20:280–286. doi: 10.1007/s10165-010-0280-z. [DOI] [PubMed] [Google Scholar]
- 40.Khan F., Stewart I., Howard L., et al. The its Not JUST Idiopathic pulmonary fibrosis Study (INJUSTIS): description of the protocol for a multicentre prospective observational cohort study identifying biomarkers of progressive fibrotic lung disease. BMJ Open Respir Res. 2019;6:e000439. doi: 10.1136/bmjresp-2019-000439. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.