Systemic sclerosis (SSc) is a heterogeneous autoimmune disease characterized by microvascular damage and generalized fibrosis that can have potential major complications depending on the internal organ involvement.1 Interstitial lung disease (ILD) is one of the most frequent complications, with a prevalence ranging from 44% to 50%.2 ILD invariably leads to decreased lung function, worsening symptoms, impaired quality of life, high healthcare costs, and early mortality.1,3 Individuals with advanced age, longer duration of SSc, diffuse SSc subtype, decreased baseline forced vital capacity (FVC), positive anti-topoisomerase I (anti-Scl70) antibodies, negative anti-centromere antibodies or raised inflammatory markers have an increased risk of SSc-ILD.2,4

Considering the high prevalence of ILD in SSc patients, systematic screening for SSc-ILD, especially in those at high risk, is supported by consensus-based recommendations.5–7 Because of its sensitivity, high-resolution computed tomography (HRCT) of the chest is recommended for screening.6,8 Pulmonary function tests (PFT) have lower sensitivity for diagnosing ILD, but they support diagnosis, assess severity, and diagnose progression at follow-up.8,9

Progression of SSc-ILD predicts mortality; some of its risk factors are male sex, higher modified Rodnan skin score, extent of ILD on HRCT and gastroesophageal reflux disease.4,10 In the EUSTAR database, 23–27% of patients with SSc-ILD showed progression in each 12-month follow-up period; however, the behavior of progression is highly heterogeneous and may vary in consecutive follow-up periods.11,12

Assessment of ILD needs a multidisciplinary approach of clinical, functional, imaging, and histopathological data to integrate a clinical probability for a diagnosis.13 However, due to limited knowledge of SSc, barriers to access to diagnostic tests, and late referral to specialists, many cases are misdiagnosed, increasing morbidity and mortality.14,15 This situation has been documented in Latin America, where availability of PFT, diagnostic imaging, and ILD referral centers is low.16

In Colombia, there is a lack of clinically relevant data and expert-based consensus to diagnose and manage ILDs, including SSc-ILD. It is likely that barriers to healthcare access promote heterogenous practices in the different regions of the country for the diagnosis, follow-up, and treatment of these diseases.17 Therefore, we conducted a national survey to assess the current practice for the diagnosis and follow-up of SSc-ILD patients by pulmonologists and rheumatologists in Colombia.

This survey is the first study that provides an overview of how specialists assess and monitor patients with SSc-ILD in the Colombian context, describing screening, rescreening, diagnosis, and follow-up practices. While the availability of diagnostic tests, including PFTs and imaging, in the working environment of the participants is high, access to specialized radiologists or pathologists and multidisciplinary team discussion is low, despite the fact that integrated approach in multidisciplinary teams is the gold standard for diagnosis of ILDs.18

Most patients with SSc-ILD are respiratory asymptomatic, especially when the disease is mild, so close follow-up is important as clinical behavior could be unpredictable.11,12 Interdisciplinary expert consensuses recommend baseline HRCT in all SSc patients for ILD screening, with greater agreement for patients with respiratory symptoms and those with high-risk factors, as PFTs have low sensitivity for early detection of ILD.5–7 In our study, screening with HRCT of the chest is performed by 78.9% of the participants, which is high compared to previous surveys. Nevertheless, there are disparities in diagnostic approaches between pulmonologists and rheumatologists due to uncertainty among physicians about how to perform an appropriate evaluation. Some participants rely more on clinical risk factors, symptoms, or functional impairment to order a HRCT.

In a global survey conducted in 2020 only 64.9% of respondents screened for ILD with HRCT in patients with newly diagnosed SSc, while 30.7% ordered HRCT only on clinical suspicion (fine crackles on auscultation, FVC<80% of the predicted value, FVC or DLCO decline, or dyspnea).19 It seems that the screening rate has increased due to consensuses recommendations, as a previous survey had described that only 51% of general rheumatologists and 66% of SSc experts routinely ordered HRCT in newly diagnosed SSc patients.20

We realized that one of the participants’ reasons for not performing screening HRCT included the lack of sufficient clinical evidence beyond expert-based recommendation. To favor clinical decision making on routine screening, further studies should evaluate its impact and cost-effectiveness in different settings. Another potential drawback of HRCT is the radiation risk; tools such as lung ultrasound, low-dose or reduced number of slices HRCT protocols may be helpful in a screening and rescreening algorithm, although they need to be validated and cannot replace HRCT as the imaging of choice for the diagnosis of ILD.21–23

Despite there are no recommendations on rescreening, 56.8% of our participants would repeat a HRCT annually, especially rheumatologists. Pulmonologists rely more on symptoms and functional worsening. In the last global survey, only 14.1% of the participants would perform an annual HRCT in the rescreening, since they do so more on clinical suspicion (64.9%).19 Decisions in this aside should probably be guided by risk factors for developing SSc-ILD and new onset of symptoms or functional impairment, balancing healthcare costs and impact on SSc prognosis.

In our study, participants scheduled PFTs (spirometry and DLCO) mainly every 6 months, but rheumatologists prefer a longer interval for follow-up. Currently, it is suggested to monitor the progression of fibrosing ILDs with PFTs at least every 3–4 months during the first year, whereas serial HRCT is often required on a case-by-case basis, less frequently than clinical and functional assessment.24 Repeat HRCT of the chest annually to detect progression was a less common decision for our participants (44.2%) compared with repeat PFTs. Although HRCT may be a more sensitive method than spirometry for monitoring disease progression in early SSc-ILD, there is no consensus regarding HRCT intervals of repetition.6,25 Repeat imaging can be performed according to the individual patient's risk factors for progression or if complementary information to PFTs is needed, with an annual follow-up interval or less frequently if the patient is clinically stable or improving.24

Participants of our survey estimated that 30% of their patients progress over a year, which is consistent with what has been reported.11 Progression of ILD can be identified by combinations of increasing respiratory symptoms, PFTs decline, and increasing fibrosis on HRCT, with no alternative explanation for worsening.24,26 For ILD secondary to autoimmunity, OMERACT (Outcome Measures in Rheumatology) has proposed to standardize the concept of clinically significant progression by a relative decline in FVC of ≥10% or by a relative decline in FVC of ≥5% to <10% and a relative decline in DLCO of ≥15%.27 In SSc-ILD, progression is proposed to be defined by involvement in more than one of the following domains: spirometry and gas exchange, respiratory symptoms, and quantitative HRCT.28 Some of the individual predictors of poor outcomes in SSc-ILD may help prioritize populations requiring closer monitoring; composite measures could be used in clinical practice to design follow-up algorithms.29,30 Our participants chose the extent of ILD on HRCT, baseline PFTs impairment, desaturation during 6MWT, baseline dyspnea, antibody profile, duration of SSc and SSc subtype as the main factors they considered in assessing progression.

There are embedded limitations in survey research. Participants were not selected by random sampling; this may lead to a volunteer effect among participants and a lack of representativeness that may not accurately reflect the opinions and characteristics of the broader population of physicians treating SSc-ILD. Since more than half of the participants treat less than five SSc-ILD patients per month, we may have excluded more experienced specialists due to convenience sampling. However, given that the participants were members of the only two associations of pulmonologists and rheumatologists in Colombia, they could be representative of the population subset where the findings are applicable. Although we used reminders to increase the participation rate there was a low response, so the conclusions drawn from this sample are limited. The design of the questionnaire could have caused bias, since the questions selected could have affected how participants evaluated their own clinical practice. To manage for response bias, the questionnaire consisted mostly of closed multiple-choice questions, so participants possibly did not have to speed through it to finish it quickly. We tested the survey to be completed in 10–15min.

The reasons why each pulmonologist or rheumatologist is more likely to make any specific clinical decision in the SSc-ILD setting can’t be explained by the design of this study and should be explored by qualitative research. The results of this survey may help to generate hypotheses for further studies and to adapt or develop consensus recommendations in the specific context of Colombia.7

In conclusion, the results of this survey provide insights into the real-life practice of pulmonologists and rheumatologists in the screening and monitoring of SSc-ILD in Colombia. Our findings help to identify areas with a lack of consensus by both specialists and provide directions for future research. Strategies should be tailored and designed based on local data to meet health needs and support clinical decision making to homogenize approaches of care.

Study conceptualization and design: all authors. Data collection: all authors. Statistical analysis: JLG, OMG. Interpretation of results: all authors. Manuscript preparation: all authors. All authors read and approved the final manuscript. The corresponding author attests that all listed authors meet authorship criteria.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no conflict of interests regarding the publication of this article.