Prevalence and risk factors of asymptomatic bronchiectasis in patients with rheumatoid arthritis at a tertiary care center in Saudi Arabia.

INTRODUCTION AND
OBJECTIVES: Bronchiectasis is a pulmonary manifestation that often occurs in individuals with rheumatoid arthritis (RA). Nevertheless, the prevalence of bronchiectasis in RA patients and predictors of its development/progression remain ill-defined. Our objective was to investigate the prevalence of bronchiectasis in a group of RA patients and examine possible clinical or biochemical risk factors that might contribute to its development.
METHODS: This was an observational study analyzing 100 RA patients with no pulmonary symptoms selected from King Abdulaziz University Hospital in the Western region of Saudi Arabia from October 2013 to 2014. Demographic, clinical and laboratory information were collected for all patients. Diagnosis was based on the 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) classification system, and disease activity was assessed using the 28-Joint Disease Activity Score Index with C-reactive protein; high-resolution computed tomography chest scans were performed. The prevalence of bronchiectasis was recorded and its association with different risk factors was examined using standard statistical methods.
RESULTS: All 100 patients fulfilled the ACR and EULAR classification criteria for RA diagnosis. Their mean age was 51.05 ± 13.5 years, disease duration was 6.19 ± 6.4 years and disease activity index was 4 ± 1.3 (moderate activity). A total of 35 (35%) patients developed bronchiectasis. Notably, we observed significant positive associations of bronchiectasis with age, disease duration and male gender (P < 0.001, P = 0.006, P = 0.028, respectively).
CONCLUSIONS: Asymptomatic bronchiectasis represents a common complication in moderately active RA patients within the Western Region of Saudi Arabia. Furthermore, several predictors of bronchiectasis development were identified, which can contribute to effective risk stratification in RA patients. Further prospective studies are needed to detect the prognosis of asymptomatic bronchiectasis in RA patients.

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by inflammatory polyarthritis and systemic features. RA diagnoses were previously based on clinical manifestations as well as the presence of IgM rheumatoid factor (RF), which is neither sensitive (60-80%) nor specific (80-90%).[1] However, over the past years, a number of novel autoantibodies have been described for use in RA detection. In this regard, assays detecting anti-cyclic citrullinated peptide (anti-CCP) antibodies have proven to be of great value, providing both diagnostic and prognostic information related to the development of extra-articular manifestations.[23] For this reason, new criteria have recently been developed, which incorporate anti-CCP testing.[4]

Bronchiectasis (BR) is a pulmonary manifestation that has been observed in RA patients.[5] It involves abnormal widening of airways of the lungs, leading to mucus build-up and vulnerability to infection. Although early studies suggested a BR prevalence of 1-10% in RA patients,[6] later reports indicated a prevalence of 20-58% using advanced detection methods (i. e., high resolution computed tomography [HRCT] of the lung).[7] The cause for this possible increased incidence of BR in RA patients remains unknown, but it has been suggested that its development could be predicted based on the presentation of certain risk factors (e. g., anti-CCP testing).

Currently, there are no reliable predictors for the development of BR in RA, although there have been some suggested predictors for the development of BR, which includes constitutional factors as male sex, prolonged disease duration, disease association, HLA-related shared epitope genes, nodular disease, presence of RF, positive anti-CCP as well as environmental factors such as smoking.

In the present study, we have investigated the prevalence of BR in RA patients at a tertiary center in the KSA. In addition, we worked to identify possible risk factors for BR development, placing special emphasis on assessing the value of anti-CCP testing.

Methods

Study design, population and setting

An observational study was conducted at a tertiary care center, King Abdulaziz University Hospital (KAUH). Between October 2013 and April 2014, a total of 293 RA patients were consecutively screened. The inclusion criteria were RA patients and age more than 17 years, and they were excluded if they had any respiratory symptoms (such as chronic cough, sputum production, or wheeze) or if they were known to have chronic lung diseases such as asthma, BR or COPD, or pulmonary fibrosis or tuberculosis. All patients were clinically and radiologically evaluated, and spirometry was performed. RA was diagnosed according to the 2010 ACR/EULAR classification system.[34] Out of the 293 patients, 100 patients fulfilled the criteria of selection. Moreover, this study was conducted in accordance with the Declaration of Helsinki and was approved by the Biomedical Ethics Research Committee of the Faculty of Medicine at KAUH. All study participants provided informed consent.

Data collection

In order to assess risk factors related to the development of BR, data were collected from the patients' clinical records, including demographic features (i. e., age, gender, nationality), clinical findings (e. g., disease duration, smoking history), and treatment history (i. e., disease modifying anti-rheumatic agents [DMARDs]: Methotrexate [MTX] and leflunamide; biological therapy). Disease activity was assessed using the 28-Joint Disease Activity Score Index with C-reactive protein (DAS28-CRP).[8] The following laboratory parameters were also considered: Erythrocyte sedimentation rate (ESR: Normal 0-20 mm/h), CRP levels (reference range: 0-3 mg/L as measured by immunonephelometry), RF (normal: 0-20 IU/L as measured by nephelometry), and anti-CCP (normal: 0-20 EU as measured via enzyme-linked immunosorbent assay [ELISA]).

HRCT of the chest without contrast medium was performed during end inspiration with the patient in a supine position (1-2 mm collimation cut at 1-2 mm intervals; SOMATOMA Definition, Siemens AG, Forchheim, Germany). A senior radiologist read images, and specific abnormalities were traced. A diagnosis of BR was given when at least two different airways in areas of non-consolidated lung met one or more of the following criteria: (a) inner diameter of the airway lumen was larger than the diameter of the accompanying pulmonary artery, (b) the airway was visible within 1 cm of the pleural edge/chest wall, (c) non-tapering of the airway for at least 2 cm beyond the last branch point.[9]

Statistical analysis

This study used descriptive statistics for ranges and means (± standard deviations [SD]) and frequency distributions for counts and percentages. Independent t-tests and chi-squared test were used to compare between patients with and without BR. When appropriate, the odd ratios were estimated with 95% confidence intervals. When variance did not present a normal distribution, Welch's t-test was used (i. e., disease duration). Results were considered to be significant when P-values were less than 0.05 (P < 0.05). All data analysis was performed using statistical package for social sciences (SPSS) software.

Results

Demographic and clinical information were collected for all patients [Table 1]. Their mean age was 51.05 ± 13.5 years (range: 19-79) and mean RA duration was 6.19 ± 6.4 years (range: 1-30). Using the DAS28-CRP, mean disease activity was moderate (4 ± 1.3 (range: 1.4-6.03)). Saudi patients made up 54% of the population, which mainly comprised females (85%). Also, the smoking status of the study participants was recorded: Active smokers (12%) and non-smokers (78%). All patients were taking DMARDs, with 77% receiving MTX and 15% receiving leflunamide. Moreover, 34% of subjects were on biological agents. The following mean antibody levels were observed in our patients: RF was 294 ± 950 IU/L (range: 5-8420) and anti-CCP was 654 ± 148 EU (range: 5-654). Overall, 71% of the subjects were positive for RF, whereas 66% displayed anti-CCP antibodies. High CRP levels were also observed in a majority of the study participants (86%).

Table 1

Demographic, clinical, and biochemical characteristics of the studied population

Our analysis revealed that 35% of the RA patients presented with BR. In this regard, the following variables were studied in relation to the development of BR: Age, gender, disease duration, smoking history, MTX use, DAS28-CRP score, ESR, CRP, RF, and anti-CCP levels [Table 2]. Notably, we observed statistically significant relationships of age and disease duration with BR. Indeed, patients with BR tended to be older (58.6 vs. 47 years) and had experienced approximately double the duration of RA. Additionally, males had a significantly higher rate of BR compared to females, displaying 3.4 times more risk for developing BR (P = 0.028).

Table 2

Odd ratios of different risk factors of bronchiectasis in patients with rheumatoid arthritis

Anti-CCP antibodies were also associated with BR. Indeed, patients with a positive anti-CCP test result had increased risk for BR (OR=3.66; P = 0.052). Nevertheless, mean anti-CCP serum levels in RA patients with and without BR were not significantly different (215.15 ± 273 and 165 ± 267 EU, respectively).

Discussion

Here, we have examined the prevalence of BR and potential risk factors associated with its development in moderately active RA patients within the Western Region of the KSA. Risk factors included age, disease duration, male gender, and positive anti-CCP serology.

Prevalence of BR in RA patients

For decades, BR has been known as a pulmonary manifestation that can occur in RA patients. Indeed, in 1967, Walker observed that the prevalence of BR in RA and osteoarthritis patients was 3.1% and 0.3%, respectively.[6] However, since the development of advanced detection methods (i. e., HRCT scan of the chest), this rate has increased. Nevertheless, few studies have been conducted with regard to the rate of BR in RA. In this respect, our study has detected a prevalence of 30%, which is much higher than that observed by Tunis et al., (16.6%).[10] On the other hand, our observed rate of BR in RA patients in the KSA is similar to the prevalence reported in the UK and France (25% and 30-41%, respectively)[111213] [Table 3].

Table 3

Previous studies assessing prevalence of BR in RA patients with sample size, setting, and limitation

The importance of BR in RA

It is well known that patients with BR have increased susceptibility to infection due to persistent colonization of bacteria in the lung,[14] which is a critical point to be considered in RA patients. In fact, lower respiratory tract infection has been shown to be more common in RA patients receiving biological therapy compared to non-biologic therapies with DMARDs (OR = 8.7; 95% CI: 1.7-43.4).[15] This is especially relevant in our patient cohort, as 34/100 patients were receiving biologic therapy, in which 12/34 (35%) displayed underlying BR. Importantly, it has even been hypothesized that BR might be associated with chronic active inflammatory cells in the lung, which may circulate and produce proinflammatory cytokines in the blood stream, thereby predisposing individuals to RA.[16] In this way, both anti-CCP positive antibodies and BR could precede the development of RA by 1.3 to 5 years.[17]

Predictors for development of BR

Since BR appears to be a common lung complication associated with RA, the obvious question arises as to whether we can predict its development based on patient characteristics. Indeed, risk stratification may be especially important prior to considering biological therapy. In this regard, although chest X-ray is often used a screening tool, it is not sensitive enough for BR diagnosis. Thus, there are currently no reliable predictors for the development of BR in RA. Nevertheless, some have suggested potential risk factors that might be associated with BR, including male gender, prolonged disease duration, disease association, human leukocyte antigen (HLA)-related shared epitope genes, smoking, nodular disease, and/or antibody positivity (anti-CCP and RF).[161819]

In particular, assays measuring anti-CCP autoantibodies have been found to be of great value in assessing RA patients. Indeed, anti-CCP is more specific for RA diagnosis, especially in those who are RF negative,[20] and can predict aggressive disease and joint erosion at 5 years better than RF.[21] Moreover, anti-CCP may represent a better indicator of active disease[22] and treatment outcome following biological therapy.[23] Importantly, our present findings have added to the benefit of using anti-CCP to assess RA patients, demonstrating a significant relationship between positive anti-CCP test results and BR.

Although anti-CCP antibodies are thought to originate from the joints, some data have suggested that the lung may represent another antigenic source. Indeed, when Fischer et al., evaluated 74 RA-free, anti-CCP-positive patients with non-specific respiratory complaints, they found that half of the patients (54%) developed lung diseases.[24] In addition, Quirke et al., compared the anti-CCP levels in two groups of RA patients (with BR and without lung disease), observing that the “RA with BR” group was more often positive for anti-CCP (87% vs. 48%) with significantly greater titers (P < 0.0001), despite a lower frequency of smoking (42% vs. 58%).[25] Nevertheless, smoking was linked to the development of both RA and anti-CCP antibodies during analyses of bronchoalveolar lavage fluid in smokers with RA, in which 28.5% expressed citrulline (compared to no expression in non-smoking RA patients).[26] That being said, there is a growing evidence to support that more patients with RA and BR are actually non-smokers.[1617] In this regard, our study did not show any relationship between smoking status and BR, as the majority of subjects with RA and BR were non-smokers (88.6%) and there was little difference in smoking between the two groups.

Study limitations

This study presented some limitations. Indeed, it is possible that sample size was insufficient for drawing strong conclusions. Moreover, the fact that this investigation was conducted at a single center may have introduced bias. Nevertheless, although further multicenter studies involving more patients from other regions of the KSA will be required to verify our findings, this study has contributed valuable information on the overall prevalence of BR in RA patients within the Western Region of the KSA. Moreover, the present study represents an important step in understanding the important risk factors associated with the development and progression BR, which can promote improved clinical management of RA patients both nationally and internationally.

Conclusions

In conclusion, this study has revealed that asymptomatic BR represents a common complication in RA patients within the Western Region of Saudi Arabia. Furthermore, several predictors of BR development were identified, including age, disease duration, male gender, and positive anti-CCP test results. These BR-associated factors could contribute to effective risk stratification in RA patients. Further prospective studies are needed to detect the prognosis of asymptomatic BR in RA patients.