Implementation of an Interstitial Lung Abnormality Clinic and Multidisciplinary Discussion

Introduction

Interstitial lung abnormalities (ILAs) are abnormalities identified on computed tomography (CT) that are potentially compatible with interstitial lung disease (ILD) in patients without clinical suspicion for ILD. Per the 2020 Fleischner Society Position Paper, ILAs are purely an imaging finding and diagnosis does not necessitate the absence of respiratory symptoms or pulmonary function impairment; however, when these features are present, a diagnosis of ILD should be considered.¹ Adherence to this recommendation is critical for improving early identification of ILD but is challenged by the lack of expert-based consensus criteria differentiating ILAs from ILD. Further complicating this issue, ILAs are often unrecognized and underreported by radiologists and infrequently prompt pulmonary referral for evaluation of possible ILD.² Additionally, existing ILD clinics with associated multidisciplinary discussions may not have the time or expertise to systemically evaluate and longitudinally follow each patient with ILAs and discuss their imaging findings in detail. Creating the infrastructure to support increased recognition and timely referral of ILAs could be key to improving outcomes within ILD.

Our objective was to establish an ILA clinic and multidisciplinary discussion to (1) identify and characterize ILAs, (2) identify modifiable factors associated with development of ILD, and (3) tailor follow-up testing based on perceived risk of ILA progression.

Methods

We established a biweekly ILA clinic and monthly multidisciplinary discussion in July 2022 at the University of Chicago Medicine and report the first year of findings. Our study was formally reviewed and determined to be quality improvement; therefore, institutional review board review was not required and informed consent was waived, per institutional policy.

Clinic

A total of 20 clinic sessions were held over the course of the year. Referrals were solicited through correspondence and educational presentations to departments and divisions where providers were likely to order CT scans in which ILAs could be identified (internal medicine, hospital medicine, cardiology, oncology). To complement this approach, beginning in February 2023, referrals were also solicited by directly contacting providers of patients identified by a thoracic radiologist (LC) as having potential ILAs on studies obtained for indications other than suspected ILD (n = 19, between February 2023–June 2023), including lung cancer screening, pulmonary nodule follow-up, surveillance of known malignancy, concern for pulmonary embolism, and others. As part of her clinical work flow, LC is responsible for interpreting a broad array of chest CTs, including high-resolution non-contrasted studies, high-resolution contrasted studies with various protocols, and low-dose studies protocolized for lung cancer screening; potential ILA cases were not restricted to a particular chest CT specification. Additionally, providers with ILD experience (pulmonary, rheumatology) often directly referred patients they identified as having possible ILAs.

The clinic was staffed by a senior ILD fellow (KS) and either of 2 pulmonary attendings (RJ, MS) with a combined 30 years of experience managing ILD. At the initial visit, a detailed history was taken, including comorbidities, symptoms, inhalational exposures (including tobacco use and high-risk occupational and environmental exposures such as asbestos, silica, lead, birds, and others), and family history. Subjects were examined for auscultatory crackles, digital clubbing, or stigmata of connective tissue disease (CTD). Pulmonary function tests (PFTs) were obtained if not available within 3 months of initial visit, and serologies were ordered if symptoms or exam findings suggested possible CTD.

Multidisciplinary Discussion

The multidisciplinary discussion was attended by a committee of ILD-expert pulmonary fellows and attendings, a thoracic radiologist, and a coordinator. All subjects seen in clinic were reviewed by the committee, in addition to subjects with ILAs followed by alternative providers. Cases were reviewed systematically as follows: (1) history, (2) symptoms, (3) exposures, (4) physical exam findings, (5) laboratory evaluation, (6) PFTs, and (7) most recent chest CT and prior CTs with complete or partial lung views. After review, a multidisciplinary diagnosis of ILA, ILD, or other was established. ILAs were characterized by feature, distribution, presence of fibrosis, and extent.¹ Individualized surveillance intervals for PFTs (every 3, 6, or 12 months) and imaging (every 6, 12, or 24–36 months) were established for subjects with ILAs based on risk of progression. High risk for progression was defined as recent/ongoing inhalational exposures, family history of ILD, subpleural fibrotic ILAs, reticular opacities, or a usual interstitial pneumonia pattern.^1,3-6 ILD was defined as ILAs plus respiratory symptoms that could not be attributed to comorbidities and one of the following: (1) total lung capacity < 80% of predicted or (2) diffusion capacity of carbon monoxide < 80% of predicted that was not explained by comorbidities (eg, obesity, neuromuscular disease, emphysema). Additionally, subjects with ILAs and a history of rheumatoid arthritis, systemic sclerosis, familial ILD, and/or pleuroparenchymal fibroelastosis were classified as having ILD regardless of symptoms and lung function.¹

Results

Of 39 subjects evaluated, the average age was 70.3 y; the majority were female (n = 25, 64.1%), and over one third were Black (n = 14, 37.8%). The majority of subjects were referred by pulmonary and primary care providers (33.3% and 30.8%, respectively), followed by oncology (15.4%). The most common comorbidities were gastroesophageal reflux disease (38.5%), coronary artery disease (30.8%), and obstructive sleep apnea (23.1%).

At the multidisciplinary discussion, ILD-focused history, physical exam, and imaging for each subject were reviewed (Table 1). Ten subjects (25.6%) received a multidisciplinary diagnosis of ILD. Of those, 2 had a family history consistent with familial ILD; one had morphea on exam and was subsequently diagnosed with CTD, and one reported a history of liver cirrhosis that prompted genetic testing revealing short-telomere syndrome. The remaining 6 subjects had respiratory symptoms, impaired PFTs, and either an inhalational exposure, drug exposure, or diagnostic imaging finding that led to a diagnosis of ILD.

Table 1.

Review of Interstitial Lung Disease–Focused History and Exam by Multidisciplinary Committee

Twenty-four subjects (61.5%) received a multidisciplinary diagnosis of ILA. Of those, 2 had a history of CTD (small-vessel vasculitis [n = 1], psoriatic arthritis [n = 1]), and 3 had a family history of ILD that did not meet criteria for familial ILD. The majority (91.7%) reported respiratory symptoms. Eighteen (75%) reported a recent/ongoing inhalational exposure, and 10 (41.7%) had a clinically important serology titer, as defined by the criteria for interstitial pneumonia with autoimmune features.⁷ PFTs were normal in subjects with ILAs. The most common ILA feature was reticulation (100%), and the majority of subjects had limited (ie, occupying < 3 lung zones¹) subpleural nonfibrotic (ie, ground glass opacities, reticulation, and/or non–emphysematous cysts that do not abut the pleura¹) ILAs. ILAs were documented in the radiology report final impression in only 9 (39.1%) subjects. Of subjects with prior CTs available (n = 16), all but one had ILAs on the prior CT; and of those, 8/15 (53.3%) had progression. The majority of subjects with ILAs were recommended to have surveillance PFTs performed every 6 months (n = 14, 58.3%) and ILD-protocolized chest CT yearly (n = 18, 75.0%). See Figure 1.

Fig. 1.

Multidisciplinary committee diagnostic recommendations for patients with interstitial lung abnormalities. PFT = pulmonary function test; ILD = interstitial lung disease; CT = computed tomography.

Five subjects received other diagnoses, including atelectasis (n = 2), bronchial wall thickening (n = 1), focal ground glass opacities (n = 1), and smoking-related lung disease (n = 1).

Discussion

We established a dedicated ILA clinic and multidisciplinary discussion and discovered that one quarter of subjects with suspected ILA actually had early ILD. This work sheds light on an urgent need to systematically evaluate patients with ILAs earlier and identify factors that, if remediated, may impact the natural history of disease. Our findings are of particular importance within the context of our unique community on the South Side of Chicago, which houses a large percentage of Black patients in whom ILD outcomes are significantly worse.⁸

Over the past decade, there has been a rapid expansion in the scientific investigation of ILAs that has led to a growing understanding that ILAs and ILD exist on a spectrum. There have been efforts to discriminate between ILAs and early ILD by accounting for PFTs, histopathologic findings, and/or degree of fibrosis;^9-11 but at present, no clear consensus criterion exists. Using a similar approach to our colleagues in the field, we chose to define ILD by the presence of respiratory symptoms and lung function impairment, while also accounting for important historical factors and imaging findings that define particular subtypes of ILD. By systemically evaluating all subjects with suspected ILAs in our clinic and multidisciplinary discussion, we identified a handful of asymptomatic or minimally symptomatic subjects who met our proposed criteria for early ILD based on factors that were either previously unreported (ie, family history of ILD), missed (ie, subtle physical exam and imaging findings), and critically, amenable to treatment. Historically, a minority of patients with ILAs were evaluated in ILD clinic at our institution prior to the establishment of our ILA clinic; however these patients often did not undergo regimented follow-up and were frequently excluded from the ILD multidisciplinary discussion in favor of discussing more complex cases. By identifying these patients earlier in the course of disease and following them longitudinally, we have a unique and exciting opportunity to elucidate how risk factor modification and early treatment affects disease course.

In addition to identifying subjects with early ILD, we also discovered that all subjects diagnosed with ILAs had at least one imaging feature associated with a high risk of radiographic progression (40% with subpleural fibrotic ILAs, 100% with reticulation).^4-5 Our high rate of reticulation is in direct contrast to other studies where ground-glass opacities, a nonfibrotic feature without a known association with progression, are most common.⁵ This may reflect a multitude of socioeconomic factors unique to our diverse patient population but alternatively could be explained by our limited sample size. Many subjects also had recent or ongoing inhalational exposures that potentially heightened their risk for progression, prompting a recommendation for PFTs every 6 months and yearly imaging in most.

Finally, we found that fewer than half of subjects with ILAs had this finding documented in the final impression of the radiology report. In 2018, Oldham and colleagues² demonstrated similarly low rates of ILA reporting in a study leveraging lung cancer screening CTs, suggesting a lack of improvement over the past 5 years.

A limitation in our clinic’s impact has been the relatively low referral rates, with an average of only 2 new patients per clinic session; this likely underestimates the true incidence of ILAs at our large, academic medical center. Moving forward, we hope to see improvement in this metric through a sustained partnership with radiology and utilization of tools such as natural language processing and quantitative imaging analysis to augment radiology review and reporting. As our referral rate expands with utilization of the aforementioned tools, we anticipate a need for additional support staff to accommodate scheduling demands, clinicians and clinic space to evaluate patients, and ILD-trained thoracic radiologists to participate in the ILA multidisciplinary discussion, approximating 20 person-hours/month. An additional limitation includes our short follow-up time period of one year. We hope to gain additional insights into the natural history of ILAs as our existing clinic infrastructure and follow-up duration continue to grow.

By establishing a dedicated ILA clinic and multidisciplinary discussion, our institution has taken an important step in increasing the recognition and evaluation of ILAs, which often reveals a diagnosis of ILD. This undertaking calls attention to an exciting opportunity to move the needle of ILD management away from treatment of advanced fibrosis and toward prevention of fibrogenesis.