key: cord-0961597-uf5t368k
authors: Lee, Hyun; Choi, Hayoung; Yang, Bumhee; Lee, Sun-Kyung; Park, Tai Sun; Park, Dong Won; Moon, Ji-Yong; Kim, Tae-Hyung; Sohn, Jang Won; Yoon, Ho Joo; Kim, Sang-Heon
title: Interstitial lung disease increases susceptibility to and severity of COVID-19
date: 2021-04-21
journal: Eur Respir J
DOI: 10.1183/13993003.04125-2020
sha: ab6006243fa1212edd2c428d678d7d1cd8a56b28
doc_id: 961597
cord_uid: uf5t368k

BACKGROUND: There are limited data regarding the relationship between interstitial lung disease (ILD) and the natural course of coronavirus disease 2019 (COVID-19). In this study, we investigate whether patients with ILD are more susceptible to COVID-19 than those without ILD and evaluate the impact of ILD on disease severity in patients with COVID-19. METHODS: A nationwide cohort of patients with COVID-19 (n=8070) and a 1:15 age-, sex-, and residence-matched cohort (n=121 050) were constructed between January 1, 2020 and May 30, 2020 in Korea. We performed a nested case-control study to compare the proportions of patients with ILD between the COVID-19 cohort and the matched cohort. Using the COVID-19 cohort, we also evaluated the risk of severe COVID-19 in patients with ILD versus those without ILD. RESULTS: The proportion of patients with ILD was significantly higher in the COVID-19 cohort than in the matched cohort (0.8% versus 0.4%, p<0.001). The odds ratio [OR] of having ILD was significantly higher in the COVID-19 cohort than in the matched cohort (adjusted OR=2.02, 95% confidence interval [CI]=1.54–2.61). Among patients in the COVID-19 cohort, patients with ILD were more likely to have severe COVID-19 than patients without ILD (49.3% versus 13.1%), including mortality (13.4% versus 2.8%) (all p<0.01). The risk of severe COVID-19 was significantly higher in patients with ILD than in those without ILD (adjusted OR=2.32, 95% CI=1.24–4.01). CONCLUSION: The risks of COVID-19 and severe presentation were significantly higher in patients with ILD than in those without ILD.

The coronavirus disease 2019 (COVID-19) pandemic currently imposes a serious healthrelated burden worldwide [1] . The disease severity of COVID-19 varies widely according to the underlying conditions of infected patients [2] . Of the underlying diseases, chronic underlying respiratory comorbidities have significant impacts on the course of COVID-19 [3] [4] [5] [6] . Chronic obstructive pulmonary disease (COPD) [7, 8] and thoracic malignancy [9 -11] have been shown to be risk factors for COVID-19, and these are associated with more severe presentations and poor prognoses of COVID-19. Asthma is suggested to impose no major additional risk of susceptibility and severity of COVID-19 [12, 13] . However, there are very limited data on the natural course of COVID-19 in patients with interstitial lung disease (ILD).

Although it seems logical to assume that patients with ILD would be more susceptible to COVID-19 than patients without ILD, no studies have clearly revealed such a relationship [14, 15] . Regarding the disease course [16] , patients with ILD seem more likely to experience respiratory failure and death compared to those without ILD [17, 18] . However, previous studies were performed in single centres. Accordingly, the question of whether patients with ILD are more susceptible to COVID-19 and experience more severe clinical courses than those without ILD remains unanswered. In addition, the susceptibility to and severity of COVID-19 in patients with ILD compared to other viral infections, such as seasonal influenza virus infection, have also not been well elucidated to date.

In this study, we investigate whether patients with ILD are more susceptible to COVID-19 than those without ILD and evaluate the impact of ILD on the severity of COVID-19. Furthermore, we investigated whether patients with ILD are more susceptible to influenza virus infection than those without ILD and evaluate the impact of ILD on the severity of influenza virus infection.

The Korean government provided the researchers with anonymised national patient data for the evaluation of COVID-19. This large cohort comprises Korea National Health Insurance Service (NHIS) claims made between January 1, 2015, and May 30, 2020, and consists of three sub-cohorts: (1) the COVID-19 cohort comprised of patients who had confirmed diagnoses of COVID-19 (positive real-time reverse-transcriptase polymerase chain reaction

[PCR] results for SARS-CoV-2) after January 2020 (n = 8,070); (2) an inspection control group comprised patients who were finally confirmed not to have COVID-19 (negative realtime reverse-transcriptase PCR results for SARS-CoV-2) (n = 222,257)-claims for COVID-19 tests are made using a special -public crisis‖ code (MT043) and can thus identify all individuals tested for COVID-19 in Korea; and (3) a matched cohort that included a 15-fold population matched with the COVID-19 cohort by age, sex, and residential area (excluding confirmed patients and test controls) (n = 121,050) [11, 19] . Of these data, we used data for the COVID-19 cohort and the matched cohort in the present study ( Figure 1 ).

The data in this cohort were combined with claims-based data from the NHIS for the period between January 1, 2015, and May 30, 2020. South Korea has a single-payer universal health system; the NHIS maintains claims data on all reimbursed inpatient and outpatient visits, procedures, and prescriptions. We extracted information on age, sex, and region of residence from insurance eligibility data. Therefore, the datasets analysed in this study include personal data, healthcare records of inpatients and outpatients from the past 5 years (including healthcare visits, prescriptions, diagnoses, and procedures), pharmaceutical visits, COVID-19-related outcomes, and death records.

The institutional review board of our institution approved this study (application no.

HYUH 2020-06-029). Since the NHIS database was constructed after anonymisation, the requirement for informed consent from participants was waived.

Laboratory confirmation of SARS-CoV-2 infection was defined as a positive result on a realtime reverse-transcriptase PCR assay of nasal or pharyngeal swabs, under the guidelines for laboratory diagnosis of COVID-19 in Korea [17] . ILD was defined as ≥ 1 claim under the Comorbidities were defined as ≥ 2 claims under ICD-10 diagnosis codes as major diagnoses during the study period (from January 1, 2015, to May 30, 2020), as follows:

angina pectoris (I20), myocardial infarction (I21, I22, or I25.2), cerebrovascular disease (G45-G46, I60-I69, or H34.0), diabetes mellitus (E10-E14), hypertension (I10-I15), heart failure (I43, I50, I09.9, I11.0, I25.5, I13.0, I13.2, I42.0, I42.5-I42.9, or P29.0), lung cancer (C34), and malignancy other than lung cancer (C00-C97, except for C34) [20] . Severe COVID-19 disease was defined as cases requiring supplemental oxygen therapy only, oxygen therapy, intensive care unit (ICU) admission, mechanical ventilation, or extracorporeal membrane oxygenation treatment in addition to individuals who died after a confirmed COVID-19 diagnosis [5, 10, 21] .

Patients with CTD who used biologic agents were defined as those who used infliximab, etanercept, adalimumab, abatacept, anakinra, tocilizumab, golimumab, certolizumab, rituximab, tofacitinib, baricitinib, upadacitinib, ustekinumab, secukinumab, ixekizumab, or guselkumab.

The main objectives of this study were to evaluate whether patients with ILD are more susceptible to COVID-19 than those without ILD and whether COVID-19 patients with ILD suffer from more severe COVID-19 disease course than those without ILD.

Categorical variables are presented as numbers (%) and were compared using the chi-squared test or Fisher's exact test, as appropriate. To evaluate the effect of ILD on susceptibility to COVID-19, the odds ratio (OR) for the presence of ILD in the COVID-19 cohort relative to that in the age-, sex-, and the residence-matched cohort was evaluated using univariable and multivariable logistic regression analyses. In multivariable analysis, covariates which not only may affect the relationship between ILD and the susceptibility of COVID-19, but also are significant in univariable analysis (p < 0.05) were adjusted: the type of insurance, reflecting socioeconomic status [22] and comorbidities (lung cancer, CTD, hypertension, diabetes mellitus, cardiovascular disease [angina pectoris, myocardial infarction, and heart failure], and cerebrovascular disease. Additionally, to evaluate the effect of ILD on the occurrence of severe COVID-19, we evaluated the OR for severe COVID-19 in patients with ILD relative to those without ILD using the COVID-19 cohort. In multivariable analysis, we adjusted for potential confounders which may affect the relationship between ILD and severe COVID-19 and are significant in univariable analysis (p < 0.05): age, sex, and comorbidities (lung cancer, CTD, hypertension, diabetes mellitus, cardiovascular disease, cerebrovascular disease, and malignancy other than lung cancer). All statistical analyses were performed using SAS ® 9.4 (SAS Institute, Cary, NC, USA). Graphs were complied with the use of GraphPad Prism version 9.0.2 (GraphPad Software, San Diego, CA, USA).

The data source for analysing the impact of ILD on influenza virus infection was the 2017 Health Insurance Review and Assessment Service, National Patient Sample (HIRA-NPS), which is nationally representative and open to the public for research purposes [23] . The database includes approximately 1,400,000 individuals each year drawn by 3% stratified random sampling by age and sex from the entire population who had claims records during the year [20, 24] . We used the HIRA-NPS, which was different from the dataset used for analysing the impact of COVID-19 on patients with ILD, because there was dramatically decreased seasonal influenza activity during the COVID-19 pandemic [25] . The reason why we used the HIRA-NPS, which was different from the dataset used for analysing the impact of ILD on COVID-19, was dramatically decreased seasonal influenza activity during the COVID-19 pandemic. In addition, as the current dataset was composed of the COVID-19 and matched cohorts, we could not establish the influenza cohort and matched cohorts using this database.

Using the 2017 HIRA-NPS database, we assessed patients with influenza virus infection (influenza cohort) (n = 27,745) and a matched cohort that included a 10-fold sample of age-, sex-, and residence-matched patients (n = 277,450). The above-mentioned definitions and statistical analysis were applied to the analysis of influenza virus infection like those of COVID-19. Table 1 depicts the characteristics of the study population. The COVID-19 and matched cohorts were well balanced in terms of age and sex. The proportion of patients who received medical aid was higher in the COVID-19 cohort than in the matched cohort (7.3% vs 3.7%, p < 0.001). Regarding pulmonary comorbidities, the rates of ILD (0.8% vs 0.4%, p < 0.001), and lung cancer (0.3% vs. 0.5%, p = 0.002) were higher in the COVID-19 cohort than in the matched cohort. The proportions of all extra-pulmonary comorbidities, except for malignancy other than lung cancer (p = 0.716), were also significantly higher in the COVID-19 cohort than in the matched cohort.

As shown in Table 2 , the adjusted OR for the presence of ILD in the COVID-19 cohort relative to the matched cohort was 2.02 (95% CI = 1.54-2.61), and the adjusted OR for the presence of ILD was 5.65 (95% CI = 4.24-7.45) in the influenza cohort to the matched cohort.

Subgroup analyses stratified by age, sex, and CTD are summarised in Table 2 . Table 3) .

Among the COVID-19 cohort, COVID-19 patients with ILD had more severe disease courses than those without ILD, requiring the following treatments: ICU admission (10.5% vs 2.9%, p < 0.001), oxygen therapy only (mortality cases were not included) (46.3% vs 12.3%, p < 0.001), mechanical ventilation treatment except for ECMO (mortality cases were not included) (11.9% vs 1.9%, p < 0.001). COVID-19 patients with ILD also demonstrated significantly higher mortality than those without ILD (49.3% vs 13.1%, p < 0.001) (Figure 3 ). 

Using a nationwide COVID-19 database and an age-, sex-, and region-matched cohort, in this study we showed that patients with ILD were 2.40-fold more likely to have COVID-19 compared with those without ILD, even after adjustments for potential confounding factors.

This study revealed that ILD is associated with higher susceptibility to COVID-19 disease, and the effect was especially significant in males and younger patients. Additionally, we found that COVID-19 patients with ILD had more severe clinical courses, including higher mortality than those without ILD. Whereas increased susceptibility to COVID-19 was observed regardless of the presence or absence of CTD, the presence of CTD was not associated with severe COVID-19 in patients with ILD. In contrast, IPF was not only associated with increased susceptibility of COVID-19 but also shows the highest risk of developing severe COVID-19.

It was previously demonstrated that pulmonary comorbidities including asthma, COPD, and lung cancer are associated with increased susceptibility to COVID-19 [7, 8, 10, [26] [27] [28] [29] . However, unfortunately, there was very limited information on whether patients with ILD have a higher risk of COVID-19 than those without ILD. Two previous studies evaluated this issue [14, 15] . However, since these studies were performed in single centres by indirectly comparing rates of COVID-19 with that of the general population, the results were inconclusive. Considering the limited generalisability of previous studies, our study has the advantage of using a large nationally representative dataset composed of a COVID-19 cohort and age-, sex-, and region-matched cohort. In our study, the increased risk of COVID-19 in patients with ILD persisted even after adjusting for potential confounding factors, including the type of insurance (representing household income) and other comorbidities (hypertension, diabetes mellitus, and lung cancer), a demonstration that supports the validity of our findings.

Furthermore, this study has another advantage because we discovered that the increasing trend of odds for COVID-19 was more pronounced in those of younger age (<60 years) and among males.

Previous studies exploring the disease severity of COVID-19 in patients with ILD have yielded conflicting results. Whereas one study suggested that patients with ILD did not show the increased occurrence of severe COVID-19 [14] , the other two studies showed that patients with ILD have more severe disease and poor prognosis than those without ILD [15, 18] . In the former study [14] , patients who were not hospitalised were not allowed to undergo specific PCR testing for SARS-CoV-2, which was solely reserved for the most severe cases during this pandemic. Thus, the number of overall ILD patients with COVID-19 could not be accurately evaluated, which limits the scope of this study. In contrast, the latter two studies compared the severity of COVID-19 according to the presence or absence of ILD and showed that COVID-19 patients with ILD had a more severe presentation of COVID-19 than those without ILD [15, 18] . Similarly, in this study, we observed that patients with ILD suffer from more severe COVID-19 than those without ILD. Furthermore, whereas previous studies simply described poor outcomes (i.e., mortality) in ILD patients with COVID-19 [15] , our study has advantage to provide more detailed information about the clinical course of severe COVID-19 that includes oxygen therapy, mechanical ventilation, ECMO, and ICU admission.

Another notable aspect of this study is our evaluation of the relationships between subgroups of ILD patients. In this study, the risk of severe COVID-19 was higher in males and younger patients compared to females and older patients, respectively. Previous studies revealed that males and young patients are more prone to acute exacerbations, of which the major aetiologies are viral infections than females [30] and these findings are in line with our study results. However, as older patients have a higher risk of severe presentation of COVID-19 than younger patients, our findings should be confirmed in future studies [31] . In this study, whereas an increased risk of severe COVID-19 was observed in ILD patients without coexisting CTD, it was not found in ILD patients with CTD. Several previous studies have revealed that the risk of disease severity of COVID-19 is increased in patients with CTD [17, 32, 33] and ILD [15, 34] . Accordingly, CTD patients with ILD may have a higher risk of severe COVID-19 compared to those without ILD, contrary to our findings. Because the number of ILD or CTD patients in our study was small, future studies are needed to further explore the issue.

Interestingly, the OR (7.50) for the presence of ILD in the influenza cohort relative to the matched cohort was higher than that (2.40) in the COVID-19 cohort than in the matched cohort. The results are very surprising, considering the higher contagious characteristics of SARS-CoV-2 compared to influenza. Although the reasons for this are not clear, there may be some explanations for this phenomenon. From the beginning of the COVID-19 pandemic, there have been warnings that patients with chronic lung disease may be more susceptible to COVID-19, and their prognosis would be worse than those without chronic lung disease [35] .

Accordingly, it is very likely that patients with ILD had exaggerated fear of highly contagious characteristics of COVID-19 and concerns about poor prognosis, which might lead to very high compliance with public health interventions, including social distancing, avoiding outdoor activities, and universal personal preventive measures (e.g., using face masks and handwashing) in patients with ILD [25, 36] . Such adherence to public health interventions might have substantially attenuated the impact of COVID-19 on patients with ILD, although those with ILD revealed higher susceptibility to and severity of COVID-19 than those without ILD in this study.

The major strength of this study is that we demonstrated that patients with ILD are more susceptible to COVID-19 and experience more severe COVID-19 compared with those without ILD using nationally representative data, which reflects real-world clinical outcomes.

However, this study also has limitations that should be acknowledged. First, since the Korean national insurance claims dataset was used, data on the severity of ILD, including pulmonary function test and treatment duration, were limited. Hence, this study could not elucidate whether the poor outcomes of ILD patients with COVID-19 were related to ILD alone.

Future research is warranted to address this issue. Second, the small number of ILD patients using biologics did not allow us to investigate whether biologics lead to adverse outcomes in ILD patients with COVID-19. Third, since we used ICD-10 codes for diagnosis, there is a possibility of misclassification. In addition, the coding in the early phase of COVID-19 (January to May 2020), which is the study period of the current study, might be different from that in the current phase (after May 2020). However, as all Korean hospitals were obligated to report all confirmed COVID-19 cases to the Korean Centres for Disease Control and Prevention from the first case in January 2020, we carefully suggest that the coding error might be insignificant. Fourth, although we adjusted for potential confounding factors which can influence the association between ILD and COVID-19 susceptibility, there were factors not adjusted for in our model. For example, we could not adjust for smoking history and body mass index as the NHIS database did not provide these variables. Fifth, there is a possibility that patients with ILD had more concern about COVID-19 and received tests more (or repeated tests), which was followed by the increased prevalence of COVID-19 in patients with ILD compared to their counterparts. The negative test rates of COVID-19 in ILD patients were higher than in those without ILD (98.4% vs 96.5%), which might have affected our study results. Sixth, the NHIS dataset lacked detailed information, including nonrespiratory conditions or specific causes of death. Thus, we could not specify non-pulmonary COVID-19 pathology (e.g., acute kidney injury) when assessing severe COVID-19.

In conclusion, patients with ILD are more susceptible to COVID-19 and experience more severe COVID-19 compared with those without ILD. Whereas the susceptibility of COVID-19 was not different by the type of ILD (except for hypersensitivity pneumonitis), ILD was associated with especially higher risk of severe COVID-19. Clinicians should be aware of the increased risk of COVID-19 in their ILD patients and manage or educate them appropriately amid this COVID-19 pandemic. The adjusted model was adjusted for age, sex, and comorbidities (lung cancer, CTD, hypertension, diabetes mellitus, cardiovascular disease [angina pectoris, myocardial infarction, and heart failure], cerebrovascular disease, and malignancies other than lung cancer). OR, odds ratio; CI, confidence interval; COVID-19, coronavirus disease 2019; ILD, interstitial lung disease; CTD, connective tissue disease; IPF, idiopathic pulmonary fibrosis; HP, hypersensitivity pneumonitis. 

Clinical Characteristics of Coronavirus Disease 2019 in China

Estimating excess 1-year mortality associated with the COVID-19 pandemic according to underlying conditions and age: a populationbased cohort study

Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study

COVID-19 in patients with thoracic malignancies (TERAVOLT): first results of an international, registry-based, cohort study

Baseline Characteristics and Outcomes of 1591 Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region

The Impact Of The COVID-19

Pandemic On Hospital Admissions In The United States

Severity and Mortality associated with COPD and Smoking in patients with COVID-19: A Rapid Systematic Review and Meta-Analysis

Expression of ACE2 in airways: Implication for COVID-19 risk and disease management in patients with chronic inflammatory respiratory diseases

Cancer patients in SARS-CoV-2 infection: a nationwide analysis in China

Risk of Coronavirus Disease Incidence, Severe Presentation, and Mortality in Patients with Lung Cancer

Asthma among hospitalized patients with COVID-19 and related outcomes

The impact of COVID-19 on patients with asthma

Single-center experience of patients with interstitial lung diseases during the early days of the COVID-19 pandemic

Clinical characteristics of COVID-19 in patients with preexisting ILD: A retrospective study in a single center in Wuhan

Presentation, diagnosis and clinical course of the spectrum of progressive-fibrosing interstitial lung diseases

Clinical features of rheumatic patients infected with COVID-19 in Wuhan, China

Increased Odds of Death for Patients with Interstitial Lung Disease and COVID-19: A Case-Control Study

Association of prescribed medications with the risk of COVID-19 infection and severity among adults in South Korea

Population-based prevalence of bronchiectasis and associated comorbidities in South Korea

Impact of bronchiectasis on susceptibility to and severity of COVID-19: a nationwide cohort study

The Impact of Socioeconomic Status on the Clinical Outcomes of COVID-19; a Retrospective Cohort Study

Demographics, treatment trends, and survival rate in incident pulmonary artery hypertension in Korea: A nationwide study based on the health insurance review and assessment service database

A guide for the utilization of Health Insurance Review and Assessment Service National Patient Samples

Impact of Public Health Interventions on Seasonal Influenza Activity During the SARS-CoV-2 Outbreak in Korea

Effect of Asthma and Asthma Medication on the Prognosis of Patients with COVID-19

Asthma control, self-management, and healthcare access during the COVID-19 epidemic in Beijing

Allergic disorders and susceptibility to and severity of COVID-19: A nationwide cohort study

inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

COVID-19 in rheumatic diseases in Italy: first results from the Italian registry of the Italian Society for Rheumatology

Attitudes, and Practice Regarding COVID-19 among Patients with Musculoskeletal and Rheumatic Diseases in Nepal: A Web-Based Cross-Sectional Study

Outcome of Hospitalization for COVID-19 in Patients with Interstitial Lung Disease: An International Multicenter Study

Prevalence of comorbidities and its effects in patients infected with SARS-CoV-2: a systematic review and meta-analysis

Face masks and containment of COVID-19: experience from South Korea

Data are presented as numbers and percentages

Data are presented as numbers and percentages

COVID-19: coronavirus disease 2019; ILD: interstitial lung disease