Interstitial Lung Disease in Rheumatoid Arthritis Remains a Challenge for Clinicians

Interstitial lung disease (ILD) is a serious complication of rheumatoid arthritis (RA) contributing to significantly increased morbidity and mortality. Other respiratory complications, such as chronic obstructive pulmonary disease and bronchiectasis, are frequent in RA. Infections and drug toxicity are important differential diagnoses and should be considered in the diagnostic work-up of patients with RA presenting with respiratory symptoms. This review provides an overview of the epidemiology and pathogenesis of RA-ILD, the radiological and histopathological characteristics of the disease as well as the current and future treatment options. Currently, there is no available evidence-based therapy for RA-ILD, and immunosuppressants are the mainstay of therapy. Ongoing studies are exploring the role of antifibrotic therapy in patients with progressive fibrotic ILD, which may lead to a new treatment approach for subgroups of patients with RA-ILD.


Introduction
Rheumatoid arthritis (RA) is a systemic inflammatory disease primarily affecting synovial joints with possible involvement of other organs. It is most often diagnosed in the fourth and fifth decades of life, and women are affected three times as often as men. The lung is a common site of extra-articular disease. Rheumatoid arthritis-associated interstitial lung disease (RA-ILD) is one of the most feared manifestations and causes serious morbidity and increased mortality [1][2][3]. Rheumatoid arthritis is the most common of the connective tissue diseases, and a substantial number of patients with RA will consult general practitioners, rheumatologists, and pulmonologists with respiratory symptoms which, in some cases, may be caused by ILD. Evaluation of respiratory symptoms in patients with RA is challenging because of the many potential causes to be considered including ILD, chronic obstructive pulmonary disease (COPD), bronchiectasis, respiratory infections following immunosuppressive therapies, drug-induced pulmonary toxicity, and ischemic heart disease [4].
The aim of the present review was to provide an overview of epidemiology, diagnostics, treatment, and prognosis in RA-ILD.

Symptoms
Symptoms of ILD are indistinguishable from a number of more common lung diseases and include exertional dyspnea, cough, chest pain, and fatigue [28][29][30][31].
A clinical examination may show digital clubbing and/or Velcro-crackles on lung auscultation in patients with fibrotic ILD. Clubbing has been reported in up to 15% of patients with RA-ILD [32]. Bilateral basal crackles have been reported in 72%-100% of patients with RA-ILD. Crackles were also present in patients with RA without ILD but to a smaller extent [23,32]. The variability of the clinical appearance is most likely due to the heterogeneity of the disease and variability in high-resolution computed tomography (HRCT) patterns.

Imaging
Chest X-ray is an insensitive method to detect ILD in patients with RA. Up to 64% of patients with ILD on HRCT will not have evident interstitial changes on a chest X-ray [29,33]. Therefore, HRCT is a mandatory part of the diagnostic work up if ILD is suspected.

Imaging
Chest X-ray is an insensitive method to detect ILD in patients with RA. Up to 64% of patients with ILD on HRCT will not have evident interstitial changes on a chest X-ray [29,33]. Therefore, HRCT is a mandatory part of the diagnostic work up if ILD is suspected.
Lung ultrasonography (LUS) has been a suggested modality to identify ILD by detection of sonographic B-lines. Lung ultrasonography findings have a high sensitivity (89%-97%) but varying specificity (50%-97%) compared to HRCT in patients with RA-ILD [52,53]. However, the etiology of the B lines may be difficult to establish in clinical practice, and differentiation between inflammatory and fibrotic changes is not possible using LUS [54].

Pulmonary Function
Pulmonary function tests (PFTs), including diffusing capacity of the lung for carbon monoxide (DLCO), may detect pulmonary disease. However, the usefulness of PFTs as a screening tool for ILD in RA is limited by the variability within the normal range and by the presence of concomitant emphysema. The results of PFTs among patients with RA-ILD depend on the study cohorts and disease severity.
Abnormal PFT are seen in 45%-65% of patients with RA with or without respiratory symptoms [33,37,55]. The patterns include airway obstruction, restrictive patterns, and impaired DLCO. The prevalence of a restrictive pattern is 5%-25%; impaired DLCO is seen in approximately 20%-45% of patients with RA [55][56][57]. Although a large number of patients have abnormal PFTs, most abnormalities remain clinically insignificant and asymptomatic. There are no screening recommendations for lung disease among patients with RA, and it remains a challenge to decide how to manage minor pulmonary function impairment.

Bronchoscopy with Bronchoalveolar Lavage
Bronchoscopy with bronchoalveolar lavage (BAL) can be used to rule out infection as a cause of pulmonary infiltrates and to assess the profile of the inflammatory cells in the lungs. However, reports of cell counts have been ambiguous. Although increased BAL lymphocyte counts have been found, findings are not consistent [30,58,59]. However, these studies generally included few patients with RA-ILD, and larger studies are needed to determine the prognostic and therapeutic significance of BAL cytological differential counts.

Histopathology
The histopathological patterns in RA-ILD are diverse and all patterns of interstitial pneumonia may occur and even overlap. The classification of the idiopathic interstitial pneumonias [60] distinguishes among a number of specific histopathological patterns which are also seen in RA-ILD. A UIP pattern is characterized by heterogeneity with patchy fibrosis with honeycombing and fibroblast foci alternating with areas of normal lung tissue. Non-specific interstitial pneumonia has a uniform appearance with a thickening of the alveolar septa with varying degrees of inflammation and fibrosis. In respiratory bronchiolitis, desquamative interstitial pneumonia and follicular bronchiolitis, peribronchiolar inflammation, and fibrosis are present, whereas organizing pneumonia is dominated by intra-alveolar plugs of connective tissue. Diffuse alveolar damage is an acute form of lung damage similar to adult respiratory distress syndrome.
Rheumatoid arthritis-associated interstitial lung disease may be classified into two different radiological and histopathological phenotypes: (1) fibrotic disease with a UIP pattern and (2) predominantly inflammatory disease with a non-UIP pattern (e.g., NSIP and OP).
Traditionally, histopathology has been regarded as the gold standard in the diagnosis of ILD, but lung biopsies are associated with a significant risk related to the surgical procedure. However, the correlation between radiological and histopathological patterns is high in RA-ILD although studies are small [30,39,65]. Thus, biopsies in the diagnostic work up of RA-ILD should mainly be used in difficult cases with atypical clinical and radiological features. Transbronchial cryobiopsies have emerged as a new diagnostic tool and may be used more often in the future due to the lower associated mortality and high diagnostic yield [66].

Screening for RA-ILD
Screening for respiratory symptoms should be part of the follow up for RA, and patients should be referred for pulmonary evaluation if symptoms occur. Not only ILD, but also COPD, bronchiectasis, and recurrent infections must be identified and treated. There are no studies validating a specific screening program for RA-ILD.

Differential Diagnosis
In some cases, lung manifestations may precede the diagnosis of RA. Patients with idiopathic interstitial lung disease should be observed prospectively for signs of systemic disease [3,67,68]. Differential diagnostic considerations should also be part of the follow up and not only of the initial diagnostic work up. The evaluation of clinical or subclinical deterioration in patients with RA-ILD is often challenging, and infection and drug-induced pneumonitis are some of the differential diagnosis that should be considered apart from disease progression (Table 3). However, drug-induced ILD is relatively rare. A large meta-analysis estimated a prevalence of approximately 1% of patients treated with methotrexate, leflunomide, sulfasalazine or TNF-alpha inhibitors [69]. Disease-modifying anti-rheumatic drugs (DMARDs) are widely used in RA with methotrexate (MTX) as the first-line treatment. Methotrexate is also the drug that is most often suspected to cause pulmonary toxicity, mainly within 6-12 months from treatment initiation [70]. The radiologic pattern in MTX-induced pneumonitis typically resembles the pattern of hypersensitivity pneumonitis with diffuse centrilobular nodules and ground glass opacities. Other radiological ILD patterns are uncommon and more likely related to RA itself [68]. A meta-analysis demonstrated increased risk of lung disease, including infections, with MTX compared to other DMARDS and biological agents [70]. All cases of pneumonitis were reported in studies published prior to 2002. In a small study of 64 patients, Gochuico et al. [23] reported an increased risk of progression of preclinical interstitial changes in 21 patients with RA who received MTX compared to other DMARDS [23]. A systematic literature search found 21 prospective studies of MTX monotherapy but identified only 15 cases of pneumonitis among 3463 patients receiving low-dose MTX (0.43%) for up to 36.5 months [71]. Recently, a Mexican study showed improved survival among patients with RA-ILD treated with MTX [72]. Thus, the use of MTX in RA-ILD seems to have a fair long-term safety and may even improve survival; this is contrary to the results of older studies.
Biological agents such as TNF-alpha inhibitors are also associated with pulmonary toxicity [69]. Risk factors for lung injury are reported to be older age, concomitant immunosuppressive therapy, and pre-existing ILD [73].
Infections may complicate the use of all immunosuppressants therapies, and the risk increases with the burden of immunosuppression [74]. The dysfunctional immune system in RA and pre-existing lung disease further increase the risk of infections. In a retrospective study of 181 patients with RA-ILD, pneumonia was the most frequently reported condition (3.9 per 100 person-years) followed by opportunistic infections (1.5 per 100 person-years). The risk increased with prednisolone doses above 10 mg daily with or without DMARDs. Overall, infection rates were 7.4 per 100 person-years [75].
The risks of fungal and mycobacterial infections and reactivation of latent infections are increased by TNF-alpha inhibitors. Latent tuberculosis should be ruled out by interferon-gamma assays before initiation of therapy [76].
Other differential diagnoses should also be considered. Acute exacerbations of RA-ILD occur with a one-year incidence of 2.8% and may mimic severe respiratory infections [77]. Occasionally, environmental allergens may complicate the condition by causing hypersensitivity pneumonitis [78]. Patients with RA have an increased risk of heart disease including congestive heart failure with pulmonary edema [79]. Heart failure may present as ground glass opacities and smooth thickening of the intralobular septa on HRCT, complicating the assessment of lung involvement.
Patients with RA are at increased risk of lymphoma and lung cancer compared to the general population, and the risk is related to immunosuppression and autoimmunity [80]. Malignancy may mimic RA-ILD changes in the lung with OP-like consolidations (lung cancer, lymphoma) or nodular interlobular septal thickening (lymphangitis carcinomatosis). Chronic obstructive pulmonary disease is common among patients with RA [4], and other smoking-related lung diseases such as respiratory bronchiolitis-ILD, desquamative interstitial lung disease, and Langerhans cell histiocytosis may be seen in RA. Pleural effusions, bronchiectasis, and rheumatoid nodules are other pulmonary manifestations of RA that are seldom mistaken for ILD but may occur simultaneously.

Treatment
New treatment strategies for RA, including early and effective DMARDs, have resulted in reduced disease activity, but the optimal treatment of lung involvement is still a dilemma [81]. So far, no randomized controlled trials in RA-ILD have been conducted.
Corticosteroids, azathioprine, and mycophenolate are the most widely used therapeutic options, with rituximab or TNF-alpha inhibitors as alternatives in patients with refractory disease. Treatment response is often better in RA-ILD with an inflammatory pattern. Fibrotic disease, especially RA-UIP, tends to be less responsive and the course of disease is similar to that of idiopathic pulmonary fibrosis (IPF) [82].
It is a challenge to determine if watchful waiting or treatment initiation is the appropriate strategy for RA-ILD. Treatment is mainly initiated in clinically significant disease when there is evidence of disease progression. Symptom assessment and pulmonary function tests, including DLCO, should be part of the follow-up. Follow-up HRCT should be guided by symptoms and pulmonary function deterioration. The optimal follow-up interval is not known and will depend on disease course and treatment. Follow-up every 3-6 months in stable disease is often used in clinical practice [83].
It is unknown whether treatment can prevent disease progression in patients with subclinical disease.
Current treatment is empiric and essentially based on immunosuppression. Traditionally, corticosteroids are administrated as a daily oral dose or as pulse courses, and it is tapered over months depending on the tolerability and clinical response [84]. The effect of corticosteroids is believed to be limited to the inflammatory subtypes of RA-ILD such as NSIP and OP [65,85]. Other immunosuppressive drugs such as TNF-alfa-inhibitors, methotrexate, azathioprine, mycophenolate mofetil (MMF), and cyclophosphamide are used as maintenance therapy or in corticosteroid resistant cases [84].
In a study from 2013, MMF was shown to stabilize lung function in connective tissue disease-associated ILD (CTD-ILD) including a group of patients with RA (n = 18) with progressive decline in forced vital capacity (FVC). The MMF treatment resulted in a trend for FVC increase and was well tolerated [86] which was also reported in previous studies [87,88]. Saketkoo et al. [89] found that MMF improved symptoms and stabilized or improved PFTs in a small mixed CTD cohort that included RA patients. Furthermore, steroids could be tapered.
Treatment of fibrotic CTD-ILD with AZA and MMF resulted in stabilization of lung function in a study from 2016. Mycophenolate mofetil (MMF) seemed to be better tolerated as discontinuation of medication for non-pulmonary side effects was more common in patients on azathioprine (27%) than on mycophenolate (5%). The adverse incident rates were similar in the two groups and did not differ on the basis of HRCT patterns [90]. A retrospective study of 40 RA-ILD patients treated with high-dose glucocorticoids, leflunomide or methotrexate showed improvement in FVC after 4 months [91].
In summary, primarily MMF demonstrates some efficacy with stabilization or minor improvement of lung function combined with relatively low toxicity. Both AZA and MTX seem to have less efficacy. Larger clinical trials are needed to clarify their use in RA-ILD.
The use of cyclophosphamide in combination with methylprednisolone has shown some benefit in rapidly progressive, severe ILD and in RA-ILD with extensive UIP, but evidence is based on small retrospective case series [48,92].
Rituximab, a B-cell-depleting anti-CD20 antibody, was used in 700 patients with RA; 56 had RA-ILD prior to rituximab treatment. A total of 68% of these patients improved or had stable pulmonary function after rituximab [93]. Rituximab was safe and only three patients (0.4%) developed ILD after RTX treatment, which may point towards a protective effect of rituximab [93]. A similar response was seen in smaller case series [94]. Thus, rituximab may be useful in patients with CTD-ILD with severe ILD refractory to other forms of immunosuppression [95].
The use of TNF-alpha inhibitor therapy in RA-ILD is limited to small case series with reports of ILD stabilization [96,97].
Non-pharmacological initiatives should always be considered. As in any chronic lung disease, pulmonary rehabilitation is important, and it has been shown to improve walking distance, quality of life, and dyspnea [98]. Pulmonary physiotherapy, including breathing exercises, may be helpful in managing dyspnea and productive cough.
Influenza and pneumococcal vaccination and optimized hand hygiene should be emphasized as tools to prevent infections (345).
Smoking cessation is essential, and advice should be offered to all smokers with RA-ILD as well as those with only RA. Supplementary oxygen can be beneficial as part of palliative care for patients with hypoxemia.
Lung transplantation should be considered in the most severe and progressive cases. Selection of transplant candidates follows the general recommendations [99]. Extra-pulmonary RA manifestations may compromise the patient's functional ability, and side effects of immunosuppression may complicate or contraindicate transplantation. Studies have demonstrated similar one-year survival in patients with RA-ILD as in IPF and scleroderma-associated ILD [100,101].
Rheumatoid arthritis-associated interstitial lung disease with a UIP pattern shares radiological and histopathological similarities with IPF, and this raises the question of the role of antifibrotic therapies in the treatment for RA-ILD. A recently published study showed a positive effect of nintedanib on slowing the decline of lung function in a broad range of fibrosing interstitial lung diseases with a progressive phenotype including RA-ILD [102]. Ongoing studies are evaluating another antifibrotic drug, pirfenidone, in RA-ILD and progressive fibrotic ILD [103,104].

Prognosis
Survival in patients with RA has improved in recent years, but patients with RA-associated ILD still have significantly reduced survival compared to patients with RA alone [3,49,105]. Previous RA-ILD studies report a median survival of three to ten years [11,[106][107][108]. More recent studies tend to show longer survival which may be attributed to improved radiological techniques and increased awareness of RA-ILD leading to earlier diagnosis (Table 4).  Previous studies have reported UIP as the predominant radiological pattern and predictor of mortality [14,42]; however, recent studies report a UIP proportion of 40%-54% [40,47]. A recent study showed that radiology-based prediction models in RA-ILD can identify patients with a progressive fibrosis phenotype [112], and that the visual extent of ILD in combination with FVC values, as previously seen in scleroderma [113], predicts outcome in RA-ILD. Morisset et al. [47] showed that the addition of HRCT patterns to the ILD-GAP prediction model did not improve its performance.
Solomon et al. [107] showed that pulmonary physiology independently predicted mortality, but baseline HRCT patterns did not. These findings reflect the heterogeneous and often unpredictable disease behavior in RA-ILD.

Future Directions for RA-ILD
Rheumatoid arthritis-associated interstitial lung disease remains a challenge for clinicians due to the fact of its heterogeneity and variable course of disease. In recent years, our knowledge of risk factors, pathogenesis, and diagnostic approaches in RA-ILD has increased, but major gaps are still present (Table 5). Expert opinion, case series, and retrospective studies constitute the foundation for treatment recommendations. The first randomized controlled trials are ongoing with results expected in the end of 2019 and 2020, and this may result in the emergence of the first evidence-based treatment options in RA-ILD.

Conclusions
Rheumatoid arthritis-associated interstitial lung disease is a serious complication of RA contributing to significantly increased morbidity and mortality. Infections, drug toxicity, and other respiratory and cardiac diseases are important differential diagnoses and should be considered in the diagnostic work up of patients with RA. The predominant HRCT and histopathology patterns are UIP or NSIP. There is no evidence-based therapy in RA-ILD. Immunosuppressants remain the mainstay of therapy, but ongoing studies are exploring the role of antifibrotic therapy in patients with progressive fibrotic ILD.