Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Nov 1.
Published in final edited form as: Arthritis Rheumatol. 2021 Sep 26;73(11):1994–2002. doi: 10.1002/art.41791

Relationship between rheumatoid arthritis and pulmonary function measures on spirometry in the UK Biobank

Lauren Prisco 1,*, Matthew Moll 2,3,4,*, Jiaqi Wang 1, Brian D Hobbs 2,3,4, Weixing Huang 1, Lily W Martin 1, Vanessa L Kronzer 5, Sicong Huang 1,2, Edwin K Silverman 2,3,4, Tracy J Doyle 2,3, Michael H Cho 2,3,4, Jeffrey A Sparks 1,2
PMCID: PMC8568623  NIHMSID: NIHMS1701389  PMID: 33982900

Abstract

Objective

We investigated the independent relationship of rheumatoid arthritis (RA) with type and severity of pulmonary patterns on spirometry compared to general population controls.

Methods

This cross-sectional study investigated the association of RA and pulmonary function measures on spirometry among subjects who had spirometry performed for research purposes in the UK Biobank. RA cases were identified by self-report and current DMARD/glucocorticoid use. General population controls denied RA. Outcomes included continuous % predicted forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC), type of spirometric pattern (restrictive or obstructive), and severity. We used multivariable regression to estimate the effects of RA cases versus controls, adjusting for age, sex, body mass index (BMI), and smoking status/pack-years.

Results

Among 350,776 analyzed subjects with spirometry performed (mean age 56.3 years, 55.8% female, and 45.5% ever smokers), we identified 2,008 cases of treated RA. In multivariable analyses, RA was associated with lower % predicted FEV1 (β −2.93, 95%CI −3.63,−2.24), % predicted FVC (β −2.08, 95%CI −2.72,−1.45), and FEV1/FVC (β −0.008, 95%CI −0.010,−0.005) than controls. RA was associated with restrictive (OR 1.36, 95%CI 1.21,1.52) and obstructive (OR 1.21, 95%CI 1.07,1.37) patterns independent of confounders. RA had the strongest associations for severe restrictive and obstructive patterns.

Conclusion

RA was associated with increased odds of restrictive and obstructive patterns, and this relationship was not explained by confounders including smoking. In addition to restrictive lung disease, clinicians should also be aware that airflow obstruction may be a pulmonary manifestation of RA.

Keywords: rheumatoid arthritis, pulmonary

INTRODUCTION

Pulmonary manifestations of rheumatoid arthritis (RA) are associated with high morbidity and mortality(1–7). Established pulmonary manifestations of RA include restrictive processes(2, 4, 8, 9) (such as interstitial lung disease [ILD]) and obstructive processes(3, 10–13) (such as bronchiectasis). Emerging research suggests that airways diseases may be common in RA and not explained by smoking(12, 14).

Pulmonary function testing (PFT), often obtained by spirometry, is an important tool for classifying lung diseases broadly into restrictive or obstructive lung processes as well as screening for and monitoring pulmonary diseases(15). Previous studies have investigated the relationship of RA with restrictive or obstructive pulmonary abnormalities compared to general population controls(9, 16–18). Only one study reported an association between RA and restrictive pattern on PFTs(17), and none found a significant association with obstructive pattern(9, 16, 17). However, as these were limited by small sample size and lack of detailed data on smoking(9, 16, 17), the relationship of RA with restrictive and obstructive ventilatory deficits independent of confounders, such as smoking, is unclear.

We aimed to investigate the type and severity of pulmonary patterns on spirometry in a large sample of the UK Biobank comparing patients with RA to general population controls. In the UK Biobank, spirometry was performed for research purposes (not only for those with suspected or known lung disease) with detailed data available on potential confounders, including smoking. We hypothesized that RA would be associated with increased risk for restrictive and obstructive patterns on spirometry.

METHODS

Study Population and Design

The UK Biobank is a prospective study of over 500,000 participants, aged 40–69 years. Details regarding the design of the UK Biobank study were reported previously in detail(19). Briefly, a random sample of adults was recruited between 2006 and 2010 from the National Health Service registry(20). Baseline visits were conducted at 22 sites across the United Kingdom. Questionnaires assessed smoking status/pack-years as well as medical history. Study staff measured height, weight, and spirometry among other measures(19). Data were also linked with the electronic health record to obtain administrative information such as read codes and medications. The overall UK Biobank received ethics approval from the North West Multicentre Ethics Research Committee. All subjects provided informed consent before participation. This secondary data analysis study was approved by the Mass General Brigham institutional review board.

We performed a cross-sectional analysis of the UK Biobank investigating the association of RA and pulmonary function measures on spirometry compared to general population controls. For this analysis, we only included subjects who had spirometry passing quality control for both FEV1 and FVC (see below). We also required smoking data since this was a key covariate. A flow diagram illustrating the analyzed study sample is included in Supplemental Figure 1.

Exposure: RA vs. Control

The primary exposure variable was RA compared with general population controls not reporting RA. As in previous UK Biobank studies, RA cases were identified by self-report and current use of disease-modifying antirheumatic drugs (DMARD) or systemic glucocorticoids(21, 22). A previous study using a similar RA case definition reported a positive predictive value of 88%(22). Current DMARD and/or glucocorticoid use was identified through the electronic health record. A subset of participants with RA had rheumatoid factor (RF) tested for research purposes from blood donated in the UK Biobank. We excluded participants from the analysis that self-reported RA but did not meet the RA case definition. General population controls had no self-reported history of RA.

Pulmonary Function Measures

Spirometry was performed for research purposes by trained respiratory therapists using a research protocol, as previously described(19). Spirometric measurements were made on a Pneumotrac 6800 spirometer (Vitalograph, Buckingham, UK). Spirometry was not performed on a small subset of participants with self-reported contraindications to spirometry (e.g., recent chest infection or myocardial infarction; recent chest, abdominal, or eye surgery; history of detached retina or pneumothorax). Bronchodilator medication was not administered. FEV1 and FVC were derived from spirometry volume-time series data, and identified as acceptable after reviewing automated spirometry acceptability designation, back extrapolated volume, and allowing 250 mL between the best measures for both FEV1 and FVC, as previously described(23). The best FEV1 and FVC values were taken from up to three consecutive blows, and the FEV1/FVC ratio was calculated from these measures. The percentage of predicted normal values was calculated adjusting for the age, sex, race, and height of the individual, as previously described(24). Among a subset with acceptable forced expiratory flow at 25–75% (FEF25–75%) measures, we calculated the percentage of predicted normal value as an additional indicator of expiratory airflow limitation and early airflow obstruction(24).

Primary Outcome: Type of Spirometric Abnormality

The co-primary outcomes were spirometric abnormalities: restrictive or obstructive pattern. We defined the outcomes as mutually exclusive. Restrictive pattern was defined as no obstructive pattern (FEV1/FVC ≥0.7) and FVC % predicted less than the calculated lower limit of normal (LLN) of FVC, consistent with prior clinical and research definitions for identifying restrictive patterns using only spirometry(25). Obstructive pattern was defined as FEV1/FVC <0.7, the standard cutpoint for both clinical practice and research studies(14, 25, 26).

Secondary Outcomes: Continuous Spirometric Results and Level of Severity

The secondary outcomes were continuous % predicted FEV1, FVC, FEV1/FVC, and FEF25–75% (among the subset with this measured), as well as the level of severity (mild, moderate, severe) for restrictive or obstructive pattern. Restrictive pattern (FVC % predicted < LLN and FEV1/FVC ≥0.7) and obstructive pattern (FEV1/FVC <0.7) severity were both defined based on degree of FEV1 impairment as: mild (FEV1 % predicted ≥70), moderate (FEV1 % predicted 50 to <70), and severe (FEV1 % predicted <50) according to standard clinical and research cutpoints (25, 27, 28).

Covariates

We considered covariates that have been associated with RA and/or pulmonary function abnormalities(29–34). Covariates were obtained at the same time of spirometry measurement in the UK Biobank. Sociodemographic variables were age (continuous in years) and sex. Lifestyle factors included measured height and weight to calculate BMI (continuous in kg/m2) as well as smoking status (never/past/current) and pack-years (continuous) by self-report. Presence of chronic respiratory illnesses (asthma, bronchiectasis, chronic obstructive pulmonary disease [COPD], ILD, and idiopathic pleural fibrosis [IPF]) was identified by self-report.

Statistical Analysis

Descriptive statistics for baseline covariates and spirometry outcomes were reported overall and according to RA status. We compared spirometric results of RA cases to controls using t-tests for continuous variables and chi-square tests for categorical variables.

We performed linear regression to estimate β coefficients and 95% confidence intervals (CIs) for the continuous spirometric values by RA status. The reference group in the models was general population controls. We adjusted the main multivariable models for potential confounders consisting of age, sex, smoking status and pack-years, and BMI. We additionally adjusted the multivariable model for history of chronic respiratory disease since RA is known to have pulmonary manifestations(1, 5). This model was considered exploratory since development of respiratory diseases may be on the causal pathway between RA and spirometric abnormalities. We also performed separate stratified analyses by sex, history of chronic respiratory disease, and smoking.

We performed logistic regression to estimate odds ratios (ORs) and 95%CIs for obstructive and restrictive spirometry patterns by RA status. We further adjusted the multivariable model for history of chronic respiratory disease and performed separate additionally stratified analyses by sex and history of chronic respiratory disease.

To further investigate the role of smoking and known chronic respiratory disease on RA status and risk of spirometric abnormalities, we constructed logistic regression models stratified by smoking status (never vs. ever smokers) and known chronic respiratory disease (absence vs. presence) and tested for interactions. Among the smoking subset, we performed an additional model adjusting for smoking status (current/past) and continuous pack-years. We also stratified by sex (male vs. female) and tested for interactions. For the ordinal severity analyses, we performed multinomial ordinal regression to obtain ORs and 95%CIs for severity of restrictive and obstructive pattern spirometric abnormalities by RA status adjusting for the same covariates as in the model. In a case-only analysis, we compared RF-positive RA to RF-negative for the co-primary outcomes of restrictive and obstructive patterns.

Statistical significance was defined as a two-sided p value less than 0.05. Analyses were performed using SAS 9.4 (SAS Institute, Cary, NC).

RESULTS

Study Sample Characteristics

The study sample was composed of 350,776 subjects who had spirometry performed for research purposes. Characteristics at the time of spirometry measurement for all subjects are shown in Table 1. The mean age was 56.3 (SD 8.1) years and 55.8% were female. We identified a total of 2,008 RA cases. Compared to general population controls, RA cases were more likely to be older, female, ever smokers, and to have higher smoking pack-years and a history of chronic respiratory disease (asthma, COPD, bronchiectasis, ILD, and IPF). Among n=900 RA cases that had RF tested for research purposes, 86.1% were seropositive.

Table 1.

Characteristics of rheumatoid arthritis subjects at time of spirometry, compared to general population controls in the UK Biobank (n=350,776).

RA cases (n=2,008) Controls (n=348,768)
Mean age, years (SD) 59.1 (7.0) 56.2 (8.1)
Female, n (%) 1,453 (72.4) 194,292 (55.7)
European ancestry, n (%) 1,865 (92.9) 321,285 (92.1)
Mean smoking pack-years (SD) 13.3 (17.0) 9.7 (14.6)
Smoking status
 Never, n (%) 911 (45.4) 190,369 (54.6)
 Past, n (%) 877 (43.7) 123,225 (35.3)
 Current, n (%) 220 (11.0) 35,174 (10.1)
Mean BMI, kg/m2 (SD) 27.7 (5.2) 27.3 (4.7)
Any self-reported chronic respiratory disease, n (%) 362 (18.0) 46,038 (13.2)
 Asthma, n (%) 295 (14.7) 42,398 (12.2)
 COPD, n (%) 81 (4.0) 4,844 (1.4)
 Bronchiectasis, n (%) 18 (0.9) 650 (0.2)
 ILD, n (%) 6 (0.3) 151 (0.04)
 IPF, n (%) 4 (0.2) 58 (0.02)
Open in a new tab

BMI, body mass index; COPD, chronic obstructive pulmonary disease; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis.

Spirometry Measures

Continuous spirometry results for the entire sample and stratified by RA status are shown in Table 2. The proportions of restrictive patterns in RA cases and general population controls were 18.1% and 14.1%, respectively. Obstructive pattern was found in 19.1% of RA cases and 13.9% of general population controls. Additional subgroup analyses of the continuous spirometry results are in Supplemental Tables 1–4.

Table 2.

Spirometry results of rheumatoid arthritis subjects, compared to general population controls in the UK Biobank (n=350,776).

RA cases (n=2,008) Controls (n=348,768) p-value
Mean FEV1 % predicted (SD) 87.5 (17.5) 91.5 (16.3) <0.0001
Mean FVC % predicted (SD) 90.0 (15.9) 93.0 (15.0) <0.0001
Mean FEV1/FVC (SD) 0.7 (0.1) 0.8 (0.1) <0.0001
Mean FEF25–75% % predicted (SD)* 82.3 (29.4) 89.3 (29.2) <0.0001
Restrictive pattern (FEV1/FVC ≥0.7 and (FVC <FVCLLN), n (%) 363 (18.1) 49,072 (14.1) <0.0001
 Mild (FEV1/FVC ≥0.7 and FEV1 % predicted ≥70) 228 (11.4) 33,657 (9.7) <0.0001
 Moderate (FEV1/FVC ≥0.7 and FEV1 % predicted 50 to <70) 122 (6.1) 14,206 (4.1)
 Severe (FEV1/FVC ≥0.7 and FEV1 % predicted <50) 13 (0.7) 1,209 (0.4)
Obstructive pattern (FEV1/FVC <0.7), n (%) 383 (19.1) 48,258 (13.8) <0.0001
 Mild (FEV1/FVC <0.7 and FEV1 % predicted ≥80) 127 (6.3) 21,323 (6.1) <0.0001
 Moderate (FEV1/FVC <0.7 and FEV1 % predicted 50 to <80) 220 (11.0) 23,558 (6.8)
 Severe (FEV1/FVC <0.7 and FEV1 % predicted <50) 36 (1.8) 3,377 (1.0)
Open in a new tab
*

There were n=162 RA cases and n=24,611 controls missing data on FEF25–75%.

FEF25–75%, forced expiratory flow at 25–75%; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; FVCLLN, predicted lower limit of normal of forced vital capacity; RA, rheumatoid arthritis.

RA vs. Controls: Continuous Spirometry Measures

Table 3 shows the associations of RA status with continuous spirometry results on spirometry performed for research purposes. RA was associated with a statistically significant lower % predicted FEV1 (β −2.93, 95%CI −3.63,−2.24), % predicted FVC (β −2.08, 95%CI −2.72,−1.45), FEV1/FVC (β −0.008, 95%CI −0.010,−0.005), and FEF25–75 (β −4.79, 95%CI −6.08,−3.49) compared to general population controls adjusted for age, sex, smoking status, smoking pack-years, and BMI.

Table 3.

Linear regression analyses of continuous spirometry results of rheumatoid arthritis subjects compared to general population controls in the UK Biobank (n=350,776).

FEV1 % predicted
β (95%CI)		 FVC % predicted
β (95%CI)		 FEV1/FVC
β (95%CI)		 FEF25–75% % predicted
β (95%CI)*
Unadjusted
Controls Ref Ref Ref Ref
RA cases −4.00 (−4.72, −3.29) −2.91 (−3.57, −2.26) −0.013 (−0.016, −0.010) −7.02 (−8.35, −5.68)
Multivariable **
Controls Ref Ref Ref Ref
RA cases −2.93 (−3.63, −2.24) −2.08 (−2.72, −1.45) −0.008 (−0.010, −0.005) −4.79 (−6.08, −3.49)
Open in a new tab
*

The study sample size for the FEF25–75% % predicted was n=326,603.

**

Adjusted for age, sex, smoking status (never/past/current), smoking pack-years (continuous), and BMI (continuous).

CI, confidence interval; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; RA, rheumatoid arthritis.

RA vs. Controls: Type of Spirometric Abnormality

The associations of RA with restrictive and obstructive patterns are shown in Table 4. RA had a multivariable OR of 1.36 (95%CI 1.21,1.53) for restrictive pattern and 1.31 (95%CI 1.16,1.47) for obstructive pattern compared to general population controls. These associations remained after further adjustment for known chronic respiratory illness.

Table 4.

Odds ratios for type of pulmonary patterns on spirometry, comparing rheumatoid arthritis subjects to general population controls in the UK Biobank (n=350,776).

Restrictive pattern
(FEV1/FVC ≥0.7 and FVC <FVCLLN)
OR (95%CI)		 Obstructive pattern
(FEV1/FVC <0.7)
OR (95%CI)
Unadjusted
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.35 (1.20, 1.51) 1.47 (1.31, 1.64)
Multivariable model 1 *
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.36 (1.21, 1.53) 1.31 (1.16, 1.47)
Multivariable model 2 **
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.36 (1.21, 1.52) 1.21 (1.07, 1.37)
Open in a new tab
*

Adjusted for age, sex, smoking status (never/past/current), smoking pack-years (continuous), and BMI (continuous).

**

Additionally adjusted for known respiratory illness.

CI, confidence interval; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; FVCLLN, predicted lower limit of normal of forced vital capacity; OR, odds ratio; RA, rheumatoid arthritis.

We assessed the impact of smoking status on the association of RA and type of pulmonary pattern using separate stratified models by smoking status (never/ever smokers) and known chronic respiratory disease (absence/presence, Table 5). In each smoking status stratum, significant associations remained between RA and restrictive and obstructive patterns. Among individuals without known chronic respiratory disease, significant associations remained between RA and restrictive and obstructive patterns. There were no significant interactions between RA status and smoking or known chronic respiratory disease for restrictive or obstructive patterns (p for interaction of case status and smoking=0.88 and 0.76, respectively; p for interaction, of case status and known chronic respiratory disease=0.40 and 0.84, respectively). Among individuals with no known respiratory disease, RA cases has significantly higher odds for both restrictive (OR 1.43, 95%CI 1.26,1.62) and obstructive patterns (OR 1.24, 95%CI 1.07,1.43). We also assessed the impact of sex (males/females) on the association of RA and type of spirometric abnormality using a stratified model. There were no significant interactions between RA status and sex for restrictive or obstructive patterns (Supplemental Table 5). Additional subgroup analyses of the type of patterns are in Supplemental Tables 6–8.

Table 5.

Odds ratios for type of pulmonary patterns on spirometry, comparing rheumatoid arthritis subjects to general population controls and stratified by smoking status in the UK Biobank (n=350,776).

Restrictive pattern
(FEV1/FVC ≥0.7 and FVC <FVCLLN)
OR (95%CI)		 Obstructive pattern
(FEV1/FVC <0.7)
OR (95%CI)
Among only never smokers (n=191,280)
Unadjusted
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.44 (1.22, 1.70) 1.26 (1.03, 1.53)
Multivariable *
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.43 (1.21, 1.70) 1.28 (1.05, 1.56)
Among only ever smokers (n=159,496)
Unadjusted
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.29 (1.10, 1.51) 1.46 (1.27, 1.68)
p for interaction 0.82 0.67
Multivariable *
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.32 (1.12, 1.55) 1.40 (1.21, 1.61)
p for interaction 0.88 0.76
Multivariable **
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.29 (1.10, 1.52) 1.31 (1.14, 1.52)
Among individuals with no known chronic respiratory disease (n=304,376)
Unadjusted
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.40 (1.23, 1.58) 1.38 (1.21, 1.59)
Multivariable **
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.43 (1.26, 1.62) 1.24 (1.07, 1.43)
Among individuals with known chronic respiratory disease (n=46,400)
Unadjusted
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.11 (0.84, 1.47) 1.36 (1.10, 1.69)
p for interaction 0.14 0.91
Multivariable **
Controls 1.00 (Ref) 1.00 (Ref)
RA cases 1.08 (0.81, 1.43) 1.16 (0.93, 1.46)
p for interaction 0.09 0.36
Open in a new tab
*

Adjusted for age, sex, and BMI (continuous).

**

Adjusted for age, sex, smoking status (current or past), smoking pack-years (continuous), and BMI (continuous).

CI, confidence interval; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; FVCLLN, predicted lower limit of normal of forced vital capacity; OR, odds ratio; RA, rheumatoid arthritis.

In the case-only analysis, RF-positive RA was associated with a HR of 2.30 (95%CI 1.20–4.42) for restrictive pattern compared to RF-negative RA, after multivariable adjustment. There was no association of RF-positive RA with obstructive pattern (OR 1.00, 95%CI 0.60–1.66) compared to RF-negative RA (Supplemental Table 9).

RA vs. Controls: Severity of Pulmonary Patterns

RA was significantly associated with increased odds of mild, moderate, and severe restrictive patterns compared to general population controls, with ORs 1.29 (95%CI 1.12,1.49), 1.45 (95%CI 1.20,1.75), and 1.86 (95%CI 1.08,3.23), respectively (Figure 1). RA was also significantly associated with increased odds of moderate and severe obstructive patterns compared to general population controls, with ORs of 1.49 (95%CI 1.29,1.72) and 1.54 (95% CI 1.10,2.16), respectively.

Figure 1.

Figure 1.

Open in a new tab

Multivariable odds ratios for type and severity of pulmonary patterns on spirometry, comparing rheumatoid arthritis subjects to general population controls in the UK Biobank (n=350,776). Adjusted for age, sex, smoking status (never/past/current), smoking pack-years (continuous), and BMI (continuous). CI, confidence interval; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; OR, odds ratio; RA, rheumatoid arthritis.

DISCUSSION

In this large study of over 350,000 subjects with spirometry performed for research purposes, we found that RA was strongly associated with increased odds of both restrictive and obstructive patterns. We showed that RA was associated with 36% increased odds of restrictive pattern and 21% increased odds of obstructive pattern, with the most marked risk for severe patterns. These results were not explained by possible confounding due to smoking, age, sex, or BMI. Our findings suggest that, in addition to restrictive lung diseases such as RA-ILD, obstructive lung diseases may be an important pulmonary manifestation of RA, even in never-smokers. Moreover, we found that RA patients without a history of known chronic respiratory disease had a higher risk of restrictive and obstructive patterns, suggesting that our findings are not explained by higher rates of known underlying lung disease among RA patients. These findings may be explained by RA disease pathogenesis, as obstructive lung diseases including COPD and asthma are highly associated with increased risk for developing RA(35). Moreover, either the RA disease process or its treatment may affect the thorax, lung parenchyma, and airways. However, it is important to note that smoking still contributes to pulmonary diseases and dysfunction in RA, since mucosal inflammation may produce RA-related autoantibodies prior to clinical RA onset(36, 37).

Prior studies investigated the relationship between prevalent RA and pulmonary function and physiology(16–18). One prospective study of 25 RA patients and 21 healthy controls found that RA patients had relatively normal spirometry results but had reduced aerobic capacity and respiratory muscle strength and endurance compared to controls(16). However, these results were unable to be adjusted for important confounders such as age, sex, and smoking and were limited by sample size(16). In contrast, our study was able to show a statistically significant association between RA and reduced lung function compared to general population controls even after adjusting for confounders. While our study evaluated the relationship between prevalent RA and pulmonary function, a large Swedish nested case-control study found no association between restrictive or obstructive pattern and subsequent incident RA(9).

RA is known to be associated with restrictive lung diseases such as RA-ILD(2, 4, 8, 9). A prospective cross-sectional study of 155 RA patients and 95 controls found that RA patients were more likely to have PFT abnormalities suggestive of restriction (e.g., reduced FVC and total lung capacity [TLC])(17). However, this association remained only among current smokers after stratification by smoking status. Unlike previous studies, we had detailed covariate data, including smoking status and pack-years, and were able to perform multivariable adjusted and stratified analyses to investigate the independent relationship between prevalent RA and pulmonary function abnormalities and were able to estimate the strength of association between RA and restrictive pattern on spirometry.

Emerging research suggests an association between RA and obstructive lung disease not explained by smoking(36, 38–40). However, previous studies reported inconsistent results on presence and severity of obstruction in RA(3, 6, 10–12, 16). A longitudinal study of 594 RA cases and 596 controls found that RA was associated with an increased risk of developing incident clinical obstructive lung disease (HR 1.54, 95%CI 1.01, 2.34)(3) during lengthy clinical follow-up, but not all patients had PFTs measured since clinical data were used. Another prospective study of 100 RA patients and 88 patients with other systemic rheumatic diseases found that obstructive lung disease was more frequent among patients with RA compared to rheumatologic disease controls(12). These studies provide support for our finding that prevalent RA is associated with increased odds of obstructive pattern, with the most marked risk for severe obstructive pattern.

Clinically detected forms of RA-ILD and other RA-related lung diseases are generally thought to be relatively uncommon, though emerging research shows that subclinical forms may be common(7, 18, 41). Inflammation at mucosal surfaces including the lungs may produce anti-citrullinated protein antibodies (ACPA) prior to clinical RA onset(36, 37). A prospective study of 50 ACPA-positive subjects (21 without arthritis, 10 early RA, and 17 long-standing RA) who underwent PFTs, cardiopulmonary exercise testing, and high-resolution computed tomography imaging showed that subclinical lung abnormalities occur early in RA pathogenesis(18). FEF25–75% is an early marker of obstruction(42–45). Previous studies have shown a statistically significant association between RA and reduced FEF25–75%, suggesting that obstruction may be common in RA(10, 46). Similarly, we found that RA was associated with a 4.79% lower FEF25–75% compared to controls after multivariable adjustment suggesting that small airway narrowing and early airflow obstruction may be clinically detectable in RA patients. Prior studies also show an association between seropositivity for both RF and ACPA with RA-related pulmonary diseases independent of smoking(14, 47). A cross-sectional study of 1,272 RA patients with PFTs performed for clinical purposes found that seropositive RA patients had a two-fold increased risk for PFT abnormalities compared to those with seronegative RA(14). Further research is needed to elucidate how the RA disease process and/or treatments may affect lung parenchyma and airways.

Strengths of our study include the large sample size and generalizability to the UK population along with access to both research data and electronic health record data. While our study sample was large, many from the entire UK Biobank were excluded mostly due to not having spirometry performed. Among the entire UK Biobank dataset, 0.57% met the RA case definition compared to 0.61% in our analyzed sample. In the entire UK Biobank, 54.2% are female compared to 55.7% in our study(26). Mean age of the entire UK Biobank is 56.5 years compared to 56.2 years for controls in our study(26). Therefore, we find it unlikely that a selection bias related to spirometry and RA case status explained our results. All analyzed subjects had spirometry performed for research purposes by trained respiratory therapists using a standard research protocol. Using spirometry performed for research purposes enabled us to assess for pulmonary abnormalities, and the association of RA with spirometric abnormalities remained among the subset without known chronic respiratory diseases. We also had access to detailed covariate data, including smoking status and pack-years as well as BMI and self-reported history of chronic respiratory disease.

Our study also has some limitations. Our definition of RA used self-report and current treatment with RA-specific medications. Self-report may be prone to misclassification so we additionally required current treatment, which has acceptable validity (positive predictive value 88%)(22). To further limit this possibility, we required controls to have no report of RA to lower the possibility that ambiguous or untreated cases may have been part of that group, but it is possible some had unrecognized RA. However, misclassification of the exposure status would be expected to bias results toward the null, so would not explain the results we report. The prevalence of RA in our study was 0.61%, consistent with two recent meta-analyses(48, 49). Moreover, 84% of the RA cases in the UK Biobank that we analyzed were RF+ based on banked blood and had demographic characteristics consistent with an RA population. Among RA cases, 86% were RF-positive, which is higher than typically observed in previous RA studies(14). While this further demonstrates the validity of our case definition of RA, there were relatively few cases with RF-negative RA. We were limited in our ability to compare RF-positive RA cases to RF-negative RA cases and our results may be most applicable to patients with seropositive RA. While we had spirometric measures on a large sample size performed for a research protocol with detailed quality control, we did not have complete PFT measures (such as lung volumes or diffusing capacity for carbon monoxide) that are necessary to definitively diagnose the presence of a restrictive lung disease or identify those with mixed pattern. Therefore, it is possible that some patients classified as obstructive pattern may also have restrictive pattern. Moreover, restrictive lung diseases, such as ILD, are a well-described RA complication and spirometry is the clinical standard to identify patients with obstructive lung disease. We were not able to analyze post-bronchodilator spirometry to assess reversible airflow obstruction in individuals with obstructive ventilatory defects. We did not have data concerning pulmonary symptoms or previous infections or RA characteristics such as disease activity, serostatus, medication history, severity, or duration. A subset of patients had RF tested for research purposes, but ACPA was not measured. Additionally, other inhalants such as passive smoking, air pollution, or e-cigarette use were not measured and may have impacted results. After excluding spirometry measures that did not pass quality control, our sample consisted mostly of individuals with European ancestry so may not be generalizable to more diverse populations. Our study was cross-sectional and data were collected over the years 2006–2010. We did not have access to the entire history of RA treatments that patients may have received and could affect lung health. Therefore, we were unable to investigate the impact of specific RA medications on spirometric abnormalities. Since the treatment landscape for RA has changed over time, there may also be secular changes over calendar time that we could not address. Longitudinal studies investigating the impact of specific RA medications on lung health are needed to address these issues.

In conclusion, we found that RA was associated with increased odds for restrictive and obstructive pattern abnormalities on spirometry. These associations were most marked for severe patterns and not explained by smoking. These results suggest that RA may be associated with obstructive lung disease in addition to the previously known effects on restrictive lung disease. Clinicians should be aware that both restrictive and obstructive abnormalities are more common in patients with RA and not due to smoking.

Supplementary Material

si1

ACKNOWLEDGEMENTS

We are grateful to the participants and staff of the UK Biobank. We thank Nick Shrine and colleagues for assistance in spirometric data.

Funding:

Dr. Sparks is supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (grant numbers K23 AR069688, R03 AR075886, L30 AR066953, P30 AR070253, and P30 AR072577), the Rheumatology Research Foundation (R Bridge Award), and the R. Bruce and Joan M. Mickey Research Scholar Fund. Dr. Doyle is supported by the National Institutes of Health (NIH)/National Heart, Lung, and Blood Institute (grant numbers K23 HL119558, R03 HL148484). Dr. Hobbs is supported by the NIH (grant numbers K08 HL136928, U01 HL089856, R01 HL147148, and R01 HL135142). The funders had no role in the decision to publish or preparation of this manuscript. The content is solely the responsibility of the authors and does not necessarily represent the official views of Harvard University, its affiliated academic health care centers, or the NIH.

REFERENCES

  • 1.England BR, Sayles H, Michaud K, Caplan L, Davis LA, Cannon GW, et al. Cause-Specific Mortality in Male US Veterans With Rheumatoid Arthritis. Arthritis Care Res (Hoboken). 2016;68(1):36–45. [DOI] [PubMed] [Google Scholar]
  • 2.Hyldgaard C, Hilberg O, Pedersen AB, Ulrichsen SP, Lokke A, Bendstrup E, et al. A population-based cohort study of rheumatoid arthritis-associated interstitial lung disease: comorbidity and mortality. Ann Rheum Dis. 2017;76(10):1700–6. [DOI] [PubMed] [Google Scholar]
  • 3.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(8):1243–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Doyle TJ, Dellaripa PF, Batra K, Frits ML, Iannaccone CK, Hatabu H, et al. Functional impact of a spectrum of interstitial lung abnormalities in rheumatoid arthritis. Chest. 2014;146(1):41–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Sparks JA, Chang SC, Liao KP, Lu B, Fine AR, Solomon DH, et al. Rheumatoid Arthritis and Mortality Among Women During 36 Years of Prospective Follow-Up: Results From the Nurses’ Health Study. Arthritis Care Res (Hoboken). 2016;68(6):753–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Pappas DA, Giles JT, Connors G, Lechtzin N, Bathon JM, Danoff SK. Respiratory symptoms and disease characteristics as predictors of pulmonary function abnormalities in patients with rheumatoid arthritis: an observational cohort study. Arthritis Res Ther. 2010;12(3):R104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Sparks JA JY, Cho SK, Vine S, Desai R, Doyle TJ, Kim SC. Prevalence, incidence, and cause-specific mortality of rheumatoid arthritis-associated interstitial lung disease among older RA patients. Rheumatology. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Raimundo K, Solomon JJ, Olson AL, Kong AM, Cole AL, Fischer A, et al. Rheumatoid Arthritis-Interstitial Lung Disease in the United States: Prevalence, Incidence, and Healthcare Costs and Mortality. J Rheumatol. 2019;46(4):360–9. [DOI] [PubMed] [Google Scholar]
  • 9.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(11):2005–13. [DOI] [PubMed] [Google Scholar]
  • 10.Geddes DM, Webley M, Emerson PA. Airways obstruction in rheumatoid arthritis. Ann Rheum Dis. 1979;38(3):222–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Collins RL, Turner RA, Johnson AM, Whitley NO, McLean RL. Obstructive pulmonary disease in rheumatoid arthritis. Arthritis Rheum. 1976;19(3):623–8. [DOI] [PubMed] [Google Scholar]
  • 12.Vergnenegre A, Pugnere N, Antonini MT, Arnaud M, Melloni B, Treves R, et al. Airway obstruction and rheumatoid arthritis. Eur Respir J. 1997;10(5):1072–8. [DOI] [PubMed] [Google Scholar]
  • 13.Halbert RJ, Isonaka S, George D, Iqbal A. Interpreting COPD prevalence estimates: what is the true burden of disease? Chest. 2003;123(5):1684–92. [DOI] [PubMed] [Google Scholar]
  • 14.Huang S, He X, Doyle TJ, Zaccardelli A, Marshall AA, Friedlander HM, et al. Association of rheumatoid arthritis-related autoantibodies with pulmonary function test abnormalities in a rheumatoid arthritis registry. Clin Rheumatol. 2019;38(12):3401–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ciancio N, Pavone M, Torrisi SE, Vancheri A, Sambataro D, Palmucci S, et al. Contribution of pulmonary function tests (PFTs) to the diagnosis and follow up of connective tissue diseases. Multidiscip Respir Med. 2019;14:17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Cimen B, Deviren SD, Yorgancloglu ZR. Pulmonary function tests, aerobic capacity, respiratory muscle strength and endurance of patients with rheumatoid arthritis. Clin Rheumatol. 2001;20(3):168–73. [DOI] [PubMed] [Google Scholar]
  • 17.Hyland RH, Gordon DA, Broder I, Davies GM, Russell ML, Hutcheon MA, et al. A systematic controlled study of pulmonary abnormalities in rheumatoid arthritis. J Rheumatol. 1983;10(3):395–405. [PubMed] [Google Scholar]
  • 18.Lucchino B, Di Paolo M, Gioia C, Vomero M, Diacinti D, Mollica C, et al. Identification of Subclinical Lung Involvement in ACPA-Positive Subjects through Functional Assessment and Serum Biomarkers. Int J Mol Sci. 2020;21(14). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. [3/30/2021];UK Biobank: Protocol for a large-scale prospective epidemiological resource. 2007 [Accessed. ]; Available from: https://www.ukbiobank.ac.uk/media/gnkeyh2q/study-rationale.pdf.
  • 20.Allen NSC, Downey P, Peakman T, Danesh J, Elliott P, Gallacher J, Green J, Matthews P, Pell J, Sprosen T, Collins R. UK Biobank: Current status and what it means for epidemiology. Health Policy and Technology. 2012;1(3):123–6. [Google Scholar]
  • 21.Siebert S, Lyall DM, Mackay DF, Porter D, McInnes IB, Sattar N, et al. Characteristics of rheumatoid arthritis and its association with major comorbid conditions: cross-sectional study of 502 649 UK Biobank participants. RMD Open. 2016;2(1):e000267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.McQueenie R, Nicholl BI, Jani BD, Canning J, Macdonald S, McCowan C, et al. Patterns of multimorbidity and their effects on adverse outcomes in rheumatoid arthritis: a study of 5658 UK Biobank participants. BMJ Open. 2020;10(11):e038829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Shrine N, Guyatt AL, Erzurumluoglu AM, Jackson VE, Hobbs BD, Melbourne CA, et al. New genetic signals for lung function highlight pathways and chronic obstructive pulmonary disease associations across multiple ancestries. Nat Genet. 2019;51(3):481–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med. 1999;159(1):179–87. [DOI] [PubMed] [Google Scholar]
  • 25.Pellegrino R, Viegi G, Brusasco V, Crapo RO, Burgos F, Casaburi R, et al. Interpretative strategies for lung function tests. Eur Respir J. 2005;26(5):948–68. [DOI] [PubMed] [Google Scholar]
  • 26.Johnson JD, Theurer WM. A stepwise approach to the interpretation of pulmonary function tests. Am Fam Physician. 2014;89(5):359–66. [PubMed] [Google Scholar]
  • 27.Dempsey TM, Scanlon PD. Pulmonary Function Tests for the Generalist: A Brief Review. Mayo Clin Proc. 2018;93(6):763–71. [DOI] [PubMed] [Google Scholar]
  • 28.Vogelmeier CF, Criner GJ, Martinez FJ, Anzueto A, Barnes PJ, Bourbeau J, et al. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Lung Disease 2017 Report: GOLD Executive Summary. Arch Bronconeumol. 2017;53(3):128–49. [DOI] [PubMed] [Google Scholar]
  • 29.Sparks JA, Karlson EW. The Roles of Cigarette Smoking and the Lung in the Transitions Between Phases of Preclinical Rheumatoid Arthritis. Curr Rheumatol Rep. 2016;18(3):15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Marchand NE, Sparks JA, Tedeschi SK, Malspeis S, Costenbader KH, Karlson EW, et al. Abdominal Obesity in Comparison with General Obesity and Risk of Developing Rheumatoid Arthritis in Women. J Rheumatol. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Xiao D, Chen Z, Wu S, Huang K, Xu J, Yang L, et al. Prevalence and risk factors of small airway dysfunction, and association with smoking, in China: findings from a national cross-sectional study. Lancet Respir Med. 2020. [DOI] [PubMed] [Google Scholar]
  • 32.Fletcher C, Peto R. The natural history of chronic airflow obstruction. Br Med J. 1977;1(6077):1645–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Costa D, Barbalho MC, Miguel GP, Forti EM, Azevedo JL. The impact of obesity on pulmonary function in adult women. Clinics (Sao Paulo). 2008;63(6):719–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Melo LC, Silva MA, Calles AC. Obesity and lung function: a systematic review. Einstein (Sao Paulo). 2014;12(1):120–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Kronzer VL, Westerlind H, Alfredsson L, Crowson CS, Nyberg F, Tornling G, et al. Respiratory diseases as risk factors for seropositive and seronegative rheumatoid arthritis and in relation to smoking. Arthritis Rheumatol. 2020. [DOI] [PubMed] [Google Scholar]
  • 36.Zaccardelli A, Liu X, Ford JA, Cui J, Lu B, Chu SH, et al. Elevated anti-citrullinated protein antibodies prior to rheumatoid arthritis diagnosis and risks for chronic obstructive pulmonary disease or asthma. Arthritis Care Res (Hoboken). 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Perry E, Kelly C, Eggleton P, De Soyza A, Hutchinson D. The lung in ACPA-positive rheumatoid arthritis: an initiating site of injury? Rheumatology (Oxford). 2014;53(11):1940–50. [DOI] [PubMed] [Google Scholar]
  • 38.Ford JA, Liu X, Chu SH, Lu B, Cho MH, Silverman EK, et al. Asthma, Chronic Obstructive Pulmonary Disease, and Subsequent Risk for Incident Rheumatoid Arthritis Among Women: A Prospective Cohort Study. Arthritis Rheumatol. 2020;72(5):704–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Friedlander HM, Ford JA, Zaccardelli A, Terrio AV, Cho MH, Sparks JA. Obstructive lung diseases and risk of rheumatoid arthritis. Expert Rev Clin Immunol. 2020;16(1):37–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Sparks JA, Lin TC, Camargo CA Jr., Barbhaiya M, Tedeschi SK, Costenbader KH, et al. Rheumatoid arthritis and risk of chronic obstructive pulmonary disease or asthma among women: A marginal structural model analysis in the Nurses’ Health Study. Semin Arthritis Rheum. 2018;47(5):639–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Robles-Perez A, Luburich P, Bolivar S, Dorca J, Nolla JM, Molina-Molina M, et al. A prospective study of lung disease in a cohort of early rheumatoid arthritis patients. Sci Rep. 2020;10(1):15640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Ciprandi G, Cirillo I, Klersy C, Marseglia GL, Vizzaccaro A, Pallestrini E, et al. Role of FEF25–75 as an early marker of bronchial impairment in patients with seasonal allergic rhinitis. Am J Rhinol. 2006;20(6):641–7. [DOI] [PubMed] [Google Scholar]
  • 43.Kwon DS, Choi YJ, Kim TH, Byun MK, Cho JH, Kim HJ, et al. FEF25–75% Values in Patients with Normal Lung Function Can Predict the Development of Chronic Obstructive Pulmonary Disease. Int J Chron Obstruct Pulmon Dis. 2020;15:2913–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Patterson GM, Wilson S, Whang JL, Harvey J, Agacki K, Patel H, et al. Physiologic definitions of obliterative bronchiolitis in heart-lung and double lung transplantation: a comparison of the forced expiratory flow between 25% and 75% of the forced vital capacity and forced expiratory volume in one second. J Heart Lung Transplant. 1996;15(2):175–81. [PubMed] [Google Scholar]
  • 45.Malerba M, Radaeli A, Olivini A, Damiani G, Ragnoli B, Sorbello V, et al. Association of FEF25–75% Impairment with Bronchial Hyperresponsiveness and Airway Inflammation in Subjects with Asthma-Like Symptoms. Respiration. 2016;91(3):206–14. [DOI] [PubMed] [Google Scholar]
  • 46.Chung MH, Lee HG, Kwon SS, Park SH, Kim KJ, Jung JI, et al. Airway obstruction in rheumatoid arthritis: CT manifestations, correlated with pulmonary function testing. Yonsei Med J. 2004;45(3):443–52. [DOI] [PubMed] [Google Scholar]
  • 47.Zhu J, Zhou Y, Chen X, Li J. A metaanalysis of the increased risk of rheumatoid arthritis-related pulmonary disease as a result of serum anticitrullinated protein antibody positivity. J Rheumatol. 2014;41(7):1282–9. [DOI] [PubMed] [Google Scholar]
  • 48.Almutairi K, Nossent J, Preen D, Keen H, Inderjeeth C. The global prevalence of rheumatoid arthritis: a meta-analysis based on a systematic review. Rheumatol Int. 2020. [DOI] [PubMed] [Google Scholar]
  • 49.Almutairi KB, Nossent JC, Preen DB, Keen HI, Inderjeeth CA. The Prevalence of Rheumatoid Arthritis: A Systematic Review of Population-based Studies. J Rheumatol. 2020. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

si1
NIHMS1701389-supplement-si1.pdf (257.1KB, pdf)

RESOURCES