Comorbidities, Cardiovascular Therapies, and COVID-19 Mortality: A Nationwide, Italian Observational Study (ItaliCO)

Introduction

Italy, after Japan, tops the list of the world's oldest countries, with over 22% of its population aged 65 or older (1). Italy has been one of the hardest hit countries during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic. As of September 1 2020 over 35,500 persons had died due to Coronavirus disease 2019 (COVID-19), especially in the northern regions of the country, with 84.5% of deaths occurring in patients age 70 or older (Istituto Superiore di Sanitá, ISS, https://www.epicentro.iss.it/coronavirus/), and a crude case fatality rate in the region Lombardy of 18.3% (2). Previous study showed that comorbid conditions play a relevant role in increasing the risk of death in patients with COVID-19 (3–9). In particular, hypertension and underlying cardiovascular diseases (CVDs) have been strongly associated with death in COVID-19 inpatients (7, 10, 11), and case fatality rates tend to be high in older people and hypertensive individuals (12). Indeed, the prevalence of CVDs in COVID-19 patients across studies ranges from 8 to 42% (13). Hypertension, heart failure (HF) and ischemic heart disease are often treated with renin-angiotensin-aldosterone system (RAAS) blockers such as angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs). The use of ACE inhibitors/ARBs in patients with COVID-19 or at risk of infection with the virus is currently a subject of intense debate (14, 15), due to the evidence that SARS-CoV-2 uses the ACE2 receptor for entry into target cells (16). ACE2 and its related axis are an endogenous counter-regulatory system, with effects opposite to those of the ACE axis (17, 18). Nonetheless, ACE inhibitor/ARBs have also been associated with a reduction of mortality and re-hospitalization in patients with cardiovascular diseases due to their anti-thrombotic and anti-inflammatory effects, and protective effects against endothelial dysfunction (13). Whether the use of ACE inhibitors and ARB affects the mortality of COVID-19 patients has been debated. The only and largest survey published so far assessing the association between comorbidities, use of ACE inhibitors/ARBs, and COVID-19 death included 4,480 patients from Denmark (12). The authors found no evidence that either ACE inhibitors or ARB increased the risk for death among persons hospitalized for COVID-19. However, the COVID-19-related death burden in Denmark has been tremendously lower than Italy (628 vs. 35,595, respectively, as of September 1 2020) which poses questions on the heterogeneity of the Italian and Danish populations and the ways COVID-19 hit and was handled by the two countries. A second study from China also excluded subjects aged ≥75 years (19). To date, there is no study available of the relation between chronic use of ACE inhibitors and ARB, considered separately, and mortality in hospitalized COVID-19 patients that includes sufficient patients in the older age group, and that accounts for concomitant cardiovascular therapies or comorbid conditions. Moreover, due to higher fatality of COVID-19 infection in patients affected by CVDs, there is an unmet need to understand the link between cardiovascular therapies, CVD and COVID-19 severity and mortality.

In this study, we describe baseline characteristics and factors associated with death among 3,179 patients hospitalized for COVID-19, who were either discharged or died, and who were residents of 19 out of Italy's 21 regions, including the main islands. In assessing potential risks for death, we specifically assessed the comorbidities and the role of pre-hospitalization ACE inhibitors and ARBs and other commonly used CVD medications (such as beta-blockers, calcium channel blockers and diuretics). By age distribution, our sample is representative of older people, that are most severely affected by COVID-19 since the beginning of the pandemic (5, 6).

Methods

Patient Inclusion

Data for 5,625 patients hospitalized for COVID-19 and with a positive nasopharyngeal swab for SARS-CoV-2 virus were manually extracted from medical charts from 61 hospitals across Italy. Patients were included in this analysis if they had been either discharged or had died at the time of ascertainment (n = 3,179, Figure 1, 56 sites). The status for each patient was reported at the time of data collection by the local investigators and represents an assessment of the patient's condition between March 25 and April 22, 2020. All the patients' information was obtained by manual review of the medical charts by the attending physician or nurse during their shifts. Each participating center was provided, upon enrollment, with a database to fill with patients' demographic, social, and clinical information and detailed instructions about the data collection. Smoking history was manually extracted from the chart for each patient. Information about smoking was not available for 316 patients. The collection and analysis of data in the registry have been deemed exempt from ethics review.

Figure 1. Italian Cartographic representation of the study subjects: Cartographic representation of the patients in this study cohort, with the area of each red circle proportional to the combined number of patients from each compact metropolitan area.

Comorbidities

Investigators manually extracted information about preexisting comorbidities known or suspected to be associated with COVID-19 mortality from the chart of each patient that was still hospitalized in their hospital or discharged within 30 days from the collection of the data. Information was available for atrial fibrillation, blood cancer, organ cancer, coronary artery disease, cardiomyopathy, chronic heart failure, chronic obstructive pulmonary disease (COPD), chronic renal failure, diabetes, hypertension, obesity, and stroke. We used a count of the reported number of comorbidities for each patient to assess their combined effect on mortality. Patients missing comorbidity information were excluded from these analyses (n = 17, Figure 2).

Figure 2. Flow chart of patient sample sizes.

Cardiovascular Medications

For this study, we specifically targeted extraction of detailed information from the patient's chart regarding use of ACE inhibitors and ARB at the time of admission. We also extracted information about other medications usually prescribed for hypertension (beta-blockers, diuretics, and Ca-antagonists).

Statistics

A generalized linear mixed model, mixed-effects logistic regression, was used to assess the relations of sex, age, comorbidity count and hypertension medication use to death relative to recovery (STATA 16, StataCorp, College Station, TX, USA). The primary outcome was inpatient mortality. Since data were clustered by hospital site, site was included in the models as a random effect to account for potential within site correlation of patient characteristics. The number of patients contributed by each hospital site varied, ranging from 2 to 242 patients (Supplementary Figure 1). A dummy category for those patients missing smoking information was included in the model for Figure 3.

Figure 3. Risk factors for mortality—all risk factors were included in the model, clustered by site (n = 2,868). ARB, Angiotensin receptor blocker; ACEi, Angiotensin converting enzyme inhibitor; BB, Beta-blocker; Di, Diuretic; CA, Ca-antagonist.

Results

There were 3,179 patients with complete data for sex, age, status, and comorbidities (Table 1); 2,282 (71.8%) had been discharged from the hospital and 897 (28.2%) had died. The median age was 69.0 years, with an interquartile range of 57 to 78 years (Supplementary Figure 2).

Table 1. Characteristics of all patients, recovered patients and deceased patients.

Risk Factors for Death

The relation of age and death was non-linear; very few patients under the age of 50 died (Table 1). Males were more likely to die than females and patients aged ≥65 years were over six times more likely to die compared to younger patients (Table 2). Current smoking was unrelated to death in either univariate or multivariable analyses (Figure 3). Hypertension (47.2%) was the most frequent comorbidity, followed by diabetes (16.3%), coronary artery disease (11.3%), atrial fibrillation (8.1%), and obesity (6.9%) (Table 1). In a model including all comorbidities and adjusted for sex, age and site, the comorbidities cardiomyopathy, COPD, chronic renal failure, hypertension, obesity, organ cancer, and stroke were all independent risk factors for death (Table 3). Interestingly, while only 5.9% of COVID-19 inpatients had COPD, the latter was highly and significantly associated with increased risk of death (Table 3). In addition to evaluating the relation of the individual comorbidities to death, the count of comorbidities reported for each patient was strongly associated with risk of death (Table 2), with the odds for death increasing by 35% for each additional comorbidity (evaluated as an ordinal count; adjOR =1.35 [1.2, 1.5] p < 0.001), after adjusting for sex, age, and site.

Table 2. Proportion of patient groups who died vs. recovered at the time of data collection.

Table 3. Multivariable model for the risk of death associated with each comorbidity after adjustment for sex and age, clustering for site as a random effect.

Analysis of Variability Across Geographic Regions and Hospitals

There were differences in the number of comorbidities across the hospitals and regional areas that provided data (Supplementary Table 1). Patients from the Central Italy were reported to have the fewest number of comorbidities and those from the Northeastern region to have the most. Therefore, in order to assess whether the geographic/hospital variation in diagnostic labeling of comorbidities could have influenced the meaning of the associations made between comorbidities and mortality, in addition to the random effect adjustment for hospital site, we added regional area to the multivariable model shown in Table 2. There was no appreciable change in the relation of any risk factor to death, including comorbidity count, after this additional adjustment (data not shown).

Risk of Death by Hypertension Medication Use

Of the 3,179 patients, 2,868 had complete information for hypertension medication use. Patients with no comorbidities were less likely to use ARBs and ACE inhibitors but there was no trend for increased use of ACE inhibitors or ARBs among patients with one or more comorbidities (Supplementary Table 2). The use of diuretics, beta-blockers and Ca-antagonists increased significantly with the number of comorbidities reported for each patient. Comorbidities were strongly associated with age but not with sex (Supplementary Table 2).

Most of the 951 patients taking either ACE inhibitors or ARB at admission had hypertension, 87.9 and 90.3%, respectively, and beta-blocker use was reported for 29.9%, diuretic use for 24.3% and Ca-antagonist use for 22.8%. After adjustment for age, sex, number of comorbidities, smoking and site, we found no increased risk of death associated with the use of ACE inhibitors, ARBs, beta-blockers or Ca-antagonists (Figure 3 and Table 4). There was a marginal negative association between ARB use and a marginal positive association between diuretic use and death (Figure 3, p = 0.025 and p = 0.020, respectively).

Table 4. List of risk factors included in the model, clustered for site as a random effect.

Discussion

This is the first and largest Italian countrywide study to date of COVID-19 inpatients, the majority of which was aged over 65, who either died or were discharged from hospital in 19 out of the 21 Italian regions, including the major islands (Figure 1). We found, in line with previous publications (3–9), that pre-existing comorbidities are major risk factors for death in COVID-19 patients. We report that the number of comorbidities is linearly and strongly associated with the risk of COVID 19-related death. However, after adjusting for comorbidities, age, and sex, we report that the lack of an association between risk of inpatient death due to COVID-19 and use of CVD medications reported in younger patients (3–9), extends to the older population at highest risk for COVID-19. Diuretics were associated with a marginal increased risk of death, and ARBs with a marginal decreased risk of death in this sample. Each comorbidity increased mortality risk independently from age and gender. Among CVD, cardiomyopathies and hypertension were related to a poor outcome. These findings are in line with Inciardi et al. (20) who showed that, in the Northern Italian population, COVID-19 patients with pre-existing CVD had an increased rate of death compared to patients without CVD. The underlying systemic inflammation in patients with CVD (21) might contribute to the increase immune responses and inflammatory cascade known to lead to a worse prognosis in COVID-19 patients (22).

Prior therapy with ACE inhibitors/ARBs was not related to worse prognosis in this cohort. The use of ACE inhibitors and ARB in patients with COVID-19 has been called into question by some (14), due to the evidence that SARS-CoV-2 uses the ACE2 receptor for entry into target cells (16). On the other hand, it has been recently shown that treatment with ACE inhibitors and ARBs does not increase ACE2 plasma levels in patients with heart failure (23). A recent study by Fosbol et al. (12) showed no association between risk of mortality in the Danish population and ACE inhibitor/ARB use. However, the COVID-19-related death burden in Denmark has been ~57-fold lower than Italy (https://coronavirus.jhu.edu/data/mortality), which calls for further studies investigating the characteristics of the Italian COVID-19 population. Additionally, Fosbol et al. did not compare the use of ACE inhibitors to the use of ARB, and the data were computed from national electronic health record review. Here we report that a similar conclusion is applicable to the Italian population. As the north of Italy, and especially the region Lombardia, suffered from one of the highest COVID-19 mortality rates worldwide (2), our data (carefully collected by manual review of medical charts from Italian 56 hospital distributed throughout the peninsula) are particularly important in order to understand the characteristics of the Italian COVID-19 patients with regional specificity. Also, in our study, a higher number of inpatients taking either ACEi or ARB than the Danish cohort was enrolled, thus allowing a stronger statistical power to rule out the individual effects of ACE inhibitors and ARB treatments. In another study, Reynolds et al. reported that previous treatment with ACE inhibitors or ARBs was not associated with a higher risk of testing positive for COVID-19 (10). Similarly, Mancia et al. recently showed that the use of ARBs and ACE inhibitors was more frequent among patients from one Italian region (Lombardy) who were infected with SARS-CoV-2 than among a large population of controls who were matched for age, sex, and place of residence (9). However, in the same study, neither ACE inhibitors nor ARBs showed an independent association with COVID-19 in patients with mild-to-moderate disease or in those with severe disease. Neither of these two latter studies had mortality as primary outcome, but rather the likelihood of the subjects of testing positive for COVID-19, or experiencing severe manifestations of COVID-19.

In a study of 5,700 patients hospitalized with COVID-19 in the New York City area, the mortality rates for patients with hypertension not taking an ACE inhibitors or ARBs at admission, taking an ACE inhibitor, or taking an ARB were comparable (8). Moreover, among 1,128 hospitalized COVID-19 patients with hypertension from Hubei, China, the inpatient use of ACE inhibitors/ARB was reported to be associated with lower risk of all-cause mortality compared with ACE inhibitors/ARB non-users (19). However, in the first study (8) the results were unadjusted for known confounders, including age, sex, race, ethnicity, and comorbidities. In the second study (19), the sample-size included only 188 patients who received ACE inhibitors/ARB, and thus it did not have the power to test the effects of ACE inhibitors and ARBs separately.

This is the first study assessing the safety of CVD medications, and in particular ACE inhibitors and ARB studied separately, in a nation-wide representative sample of COVID-19 inpatients, mostly aged >65. We provide strong evidence suggesting that, regardless of a person's risks for COVID19, the five drugs most frequently used for the treatment of CVDs in outpatient settings are not associated with increased inpatient mortality due to COVID-19.

Among CVD therapies, diuretics were associated with a marginally significant increased risk of mortality in our final models. However, the proportion of patients who were using diuretics at the time of admission increased markedly with the number of comorbidities in each patient. Although we adjusted for number of comorbidities, it is likely that residual confounding may be present. Similarly, use of ARBs was associated with a small decrease in death rates after adjustment for confounders, but the effect was marginally significant and may have been influenced by unaccounted confounding.

Our study has several strengths: (1) it is the first Italian country-wide study describing the characteristics of a highly heterogeneous cohort of COVID-19 inpatients both in terms of severity of the clinical manifestations, and also in terms of geographical distribution. We believe that the analyses performed in this study are clinically informative given that the north of Italy suffered from one of the highest mortalities for COVID-19 worldwide; and (2) the availability of a sample representative of the older people (>65 years of age) at highest risk for morbidity and mortality due to COVID-19. Limitations include: (1) the observational nature of the study, and the fact that data on comorbidities were collected by manual review of the medical charts with no objective assessment; this methodological limitation did not allow the disentanglement of the independent associations of antihypertensive medications with mortality (i.e., adequate control for disease related-, hypertension-, and other medication use- related variables); (2) lack of other outcome measures apart from death or hospital discharge, and of information about the cause of death of the patients (such as severe respiratory distress, acute kidney injury, myocardial infarction, pulmonary, and systemic thromboemobilsm); (3) the absence of information about specific treatments received during hospitalization and in-hospital ACE inhibitors/ARBs continuation or discontinuation; (4) the lack of a severity score for COVID-19 patients during hospitalization; and (5) the potential bias introduced by excluding patients that were still hospitalized at the time of data collection. However, we felt that including the latter could have introduced an even bigger bias, because these individuals could have had associations between the variables analyzed and the outcome that could be different from patients that were included because they had an outcome.

In summary, the results of our nationwide Italian study of a population of COVID19 inpatients, that suffered from one of the highest mortality rates worldwide, confirms that the number of comorbidities appears to be independently associated with increased COVID-19-related death. Among cardiovascular comorbidities both hypertension and pre-existing cardiomyopathy were associated with COVID-19 risk of death. Nonetheless, we provide reassuring evidence that use of commonly-used CVD medications is not associated with increased risk of death due to COVID-19, regardless of the person's risk for the disease due to age, sex, or comorbid conditions. Our findings show that there is no need to interrupt treatments for CVD in COVID-19 patients; in particular, the treatment with ACE inhibitors/ARBs should be continued in order to reduce potential cardiovascular derangement in COVID-19 patients. Further studies are needed in order to shed light onto the relationship between CVD therapies, underlying CVD, and prognosis in COVID-19 patients.

Data Availability Statement

All datasets generated for this study are included in the article/Supplementary Material.

Ethics Statement

The studies involving human participants were reviewed and approved by University of Arizona IRB waiver #2003521629. The ethics committee approvals/waivers were also obtained from each of the participating hospitals. The ethics committee waived the requirement of written informed consent for participation.

Author Contributions

FP, GR, EB, MB, MCa, BC, MCo, AC, FD'A, ED'E, GF, SGa, SGu, SH, MK, LM, AP, RP, PP, VP, MP, CT, and RT collected the data. FP, DAS, SGu, JCW, and FDiM analyzed the data and wrote the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This work was funded by the FAMRI and the Asthma and Airway Disease Research Center, University of Arizona research funds.

Conflict of Interest

MB has served on advisory boards and has received travel fundings, honoraria for speaking from Angelini, Astra Zeneca, bayer, Cubist, Pfizer, Menarini. MCo has received personal fees from Chiesi, AstraZeneca, Boehringer-Ingelheim, Alk-Abello, GSK, Novartis, Zambon, and scientific grants from Chiesi and University of Ferrara, Italy. FDiM has received personal fees from Chiesi, AstraZeneca, Boehringer-Ingelheim, GSK, Novartis, Zambon, Guidotti/Malesci, Menarini, Mundipharma, TEVA, Almiral, Levante Pharma, Sanophi, and scientific grants from AstraZeneca, Boehringer-Ingelheim, GSK, Novartis. AP has received board membership and consultancy fees, payment for lectures, grants for research, travel expenses reimbursements from GSK, AZ, Boehringer Ingelheim, Chiesi Farmaceutici, TEVA < Mundipharma, Zambon, Novartis, Menarini, Sanofi,Roche, Edmondpharma, Fondazione Maugeri, Fondazione Chiesi. JW has received investigator initiated research funding from Vertex pharmaceuticals.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

We thank all the investigators of the ItaliCO study group for collecting the patients' data included in this manuscript.

ITALICO: Italian National Study on Risk Factors Associated With COVID-19

List of Contributors

F. Polverino MD PhD¹, D. Stern MS¹, M. Polverino MD², F. D'Amico MD³, E. D'Elia, MD PhD⁴, A. Agarossi MD⁵, S. Agati MD⁶, E. Agosteo MD⁷, F. Ando' MD⁸, M. Andreoni MD⁹, IF. Angelillo DDS MPH¹⁰, G. Arcoleo MD¹¹, C. Arena MD¹², P. Baiamonte MD¹¹, E. Balestro MD¹³, L. Ball MD, PhD¹⁴, P. Banfi MD¹⁵, G. Bartoletti MD¹⁶, R. Bartolotta RN¹⁷, M. Bassetti MD PhD¹⁴, D. Battaglini MD¹⁴, M. Bellan MD PhD¹⁸, I. Benzoni MD PhD¹⁹, R. Bertolini MD²⁰, M. Bevilacqua MD²¹, M. Bezzi²² MD, A. Bianco MD²³, A. Bisbano MD²⁴,F. Bobbio MD¹⁸, G. Bocchialini MD²², F. Bonetti MD²⁰, F. Boni MD²⁵, M. Bonifazi MD²⁶, G. Borgonovo MD²⁵, S. Borre' MD²⁷, M. Bosio MD²⁸, G. Brachini MD²⁹, I. Brunetti MD¹⁴, L. Calagna³, F. Calò¹⁰, M. Candelli MD PhD³⁰, A. Capuozzo MD², T. Carr MD¹, A. Castellani MD²², F. Catalano MD PhD¹¹, G. Catania MD³¹, E. Catena MD⁵, M. Cattaneo^{32, 33}, A. Cattelan MD¹³, V. Ceruti MD²¹, F. Chiumiento MD³⁴, G. Cicchitto MD², B. Cirillo MD²⁹, M. Confalonieri MD³⁵, P. Confalonieri MD³⁵, M. Contoli MD PhD³⁶, N. Coppola MD PhD¹⁰, A. Corsico MD²⁸, R. Cosentina MD³, R. Costantino MD³⁷, C. Crimi MD PhD³⁸, A. Currà MD³⁹, M. D'Abbraccio MD⁴⁰, A. Dalbeni MD²¹, F. Daleffe MD²², R. Davide MD⁴¹, M. Del Donno MD⁴², F. Di Marco, MD PhD⁴³, F. Di Pastena MD⁴⁴, F. Di Perna MD⁴⁵, Z. Di Rosa MD⁴⁶, A. Di Sabatino MD²⁸, O. Elesbani MD⁴⁰, D. Elia MD³², V. Esposito MD PhD⁴⁷, L. Fabiani MD⁴⁸, G. Falco MD²⁵, G. Falo MD³⁵, C. Fanelli MD⁴⁹, A. Fantin MD⁵⁰, F. Ferrigno MD⁵¹, G. Fiorentino MD⁴⁶, F. Franceschi MD PhD³⁰, M. Fronza MD⁵², G. Gardini Gardenghi MD⁴⁰, S. Gasparini MD²⁶, D.R. Giacobbe MD¹⁴, C. Giannotti MD⁵, G. Giannotti MD¹⁹, A. Gidari MD⁵³, F. Giovanardi MD PhD²⁵, P. Gnerre MD³¹, F. Gonnelli MD²⁶, M. Graziano MD⁵⁴, S. Greco MD⁴⁷, A. Grosso PhD²⁸, S. Guarino MD³², S. Guerra MD PhD¹, S. Harari MD³², A. Iannarelli MD⁵⁵, P. Imitazione MD³⁹, F. Inglese MD⁵⁶, V. Iodice MD⁴⁶, A. Izzo MD⁴³, C. La Greca MD¹⁶, M. Kraft MD¹, A. Lax MD¹⁵, F. Legittimo MD²⁰, A. Leo MD⁵⁷, S. Leone MD⁵³, V. Lepidini MD⁹, M. Leto RN³⁶, F. Licata MD²³, F. Locati MD³, L. Lorini MD⁴, B. Lucchetti MD²⁰, I. Maida MD⁴⁸, M. Macera MD¹⁰, E. Manzillo MD⁴⁶, A. March MD⁴⁹, D. Mascheroni MD⁵⁸, A. Mastroianni MD¹⁷, I. Mauro MD², M. Mazzitelli MD²³, E. Mazzuca MD¹¹, L. Mennella MD¹⁶, C. Micheletto MD¹², A. Mingoli MD²⁹, P. Minuz MD²¹, M. Moioli MD¹⁴, L. Monti MD⁵⁸, R. Morgagni, MD PhD⁹, L. Mucci MD⁵⁹, M. Muselli MD⁴⁷, S. Negri MD⁶, C.G.A. Nobile MD⁶⁰, S. Oldani MD⁶¹, C. Olivieri MD²⁷, A. Papi MD³⁵, G. Parati MD⁶², L. Parodi MD³¹, R. Parrella MD PhD⁴⁶, E. Pastorelli MD²⁷, V. Patruno MD⁴⁹, F. Pellegrino MD⁷, P. Pelosi MD FERS¹⁴, M.F. Pengo, MD PhD⁶², D. Pepe MD²³, A. Perotti MD⁵, R. Petrino MD²⁷, M. Petrucci MD³⁰, R.M. Piane MD¹⁶, G. Pignataro MD PhD³⁰, M. Pino MD²³, M. Pirisi MD¹⁸, V. Poletti MD⁶¹, F. Porru MD⁶³, F. Pugliese MD²⁹, R. Punzi MD⁴⁶, D.A. Ramaroli MD¹², C. Robba MD PhD¹⁴, R. Rostagno MD²⁷, G. Ruocco MD⁶⁴, U. Sabatini MD²⁸, P.P. Sainaghi MD PhD¹⁸, F. Salton MD³⁴, C. Salzano MD⁶⁵, A. Sanduzzi MD³⁹, S. Sanduzzi Zamparelli MD¹⁰, V. Sangiovanni MD⁴⁶, D. Santopuoli MD⁴⁵, P. Sapienza MD²⁹, L. Sarmati MD⁹, E. Schiaroli MD⁶⁶, F. Scienza MD²⁷, M. Senni MD⁴, L. Serchisu PhD⁵¹, S. Sgherzi MD⁵⁷, D. Soddu MD¹⁸, D. Soranna MD⁶¹, C. Sorino MD PhD⁶, S. Spadaro MD³⁵, E. Stirpe MD⁴⁹, C. Tana MD⁵⁸, S. Tardivo MD¹², S. Tartaglia MD²⁷, E. Teopompi MD²⁰, R. Terribile MD²⁷, M. Tomchaney¹, E. Torelli MD³⁰, C. Torlasco MD⁶¹, C. Torti MD²³, E. Tupputi MD⁵⁶, C. Ugolinelli MD⁵, A. Vatrella MD⁶⁸, A.G. Versace MD⁸, M. Villani MD⁶⁸, L. Vincenzo MD⁴⁰, C.A. Volta MD³⁵, N. Voraphani MD¹, J.C. Woods PhD⁶⁹, E. Zekaj MD⁷⁰, R. Zoppellari MD³⁵, F.D. Martinez MD¹

¹Asthma and Airway Disease Research Center, University of Arizona

²Ospedale Scarlato, Scafati

³ASST Bergamo Est, Seriate

⁴ASST Papa Giovanni XXIII, Bergamo

⁵Ospedale Sacco, Milano

⁶ASST Lariana, Ospedale Sant'Anna di Como

⁷Clinica San Carlo- Paderno Dugnano, Milano

⁸Policlinico di Messina, Messina

⁹Policlinico Universitario Tor Vergata, Roma

¹⁰Università Vanvitelli, Napoli

¹¹A.O.O.R. Villa Sofia Cervello, Palermo

¹²Università di Verona, Verona

¹³Azienda Ospedaliera Universitaria di Padova, Padova

¹⁴Policlinico San Martino and Universita' di Genova, Genova

¹⁵IRCCS Fondazione Don Carlo Gnocchi, Milano

¹⁶Azienda USL Toscana Nord Ovest, Ospedale di Lucca

¹⁷zienda Ospedaliera di Cosenza, Cosenza

¹⁸AOU Maggiore della Carità and Università del Piemonte Orientale UPO, Novara, Italy

¹⁹Ospedale di Cremona, Cremona

²⁰AUSL-IRCCS RE Ospedale di Guastalla

²¹Policlinico GB Rossi, Verona

²²Spedali Civili di Brescia, Brescia

²³Università degli studi Magna Graecia, Catanzaro

²⁴Ospedale San Giovanni di Dio, Crotone

²⁵AUSL-IRCCS, Reggio Emilia

²⁶Polytechnic University of Marche Region, Azienda Ospedali Riuniti, Ancona

²⁷Ospedale S Andrea, Vercelli

²⁸IRCCS Policlinico San Matteo Foundation, Pavia

²⁹Universita La Sapienza, Roma

³⁰Fondazione Universitaria Policlinico Gemelli - IRCCS. Università Cattolica del Sacro Cuore di Roma, Roma

³¹Ospedale San Paolo, Savona

³²Department of Medical Sciences, San Giuseppe Hospital MultiMedica IRCCS, Milan, Italy

³³Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy

³⁴Ospedale di Eboli, Eboli

³⁵Ospedale di Trieste, Trieste

³⁶Universita' di Ferrara, Ferrara

³⁷Azienda Ospedaliera Pugliese Ciaccio, Catanzaro

³⁸Azienda Ospedaliera Universitaria Policlinico-Vittorio Emanuele, Catania

³⁹Ospedale Giulio Jazzolino, Vibo Valentia

⁴⁰Ospedale Monaldi and Federico II University, Napoli

⁴¹ASST del Garda, Ospedale di Desenzano del Garda

⁴²Ospedale Rummo, Benevento

⁴³ASST Papa Giovanni XXIII, Bergamo, and University of Milan

⁴⁴Ospedale Dono Svizzero, Formia

⁴⁵Ospedale di Caserta, Caserta

⁴⁶Ospedale Cardarelli, Campobasso

⁴⁷Ospedale Cotugno, Napoli

⁴⁸Universita' de L'Aquila

⁴⁹Ospedale di Sassari, Sassari

⁵⁰Ospedale Universitario di Bolzano

⁵¹ASL Salerno, Scafati

⁵²Ospedale di Bolzano, Bolzano

⁵³Ospedale di Perugia, Perugia

⁵⁴Azienda Sanitaria Locale and San Giuseppe Moscati Hospital, Avellino

⁵⁵Ospedale Santa Maria Goretti, Latina

⁵⁶ASST Mantova, Mantova

⁵⁷ASL BT. Pneumologia Andria

⁵⁸IC Villa Aprica, Gruppo San Donato, Como

⁵⁹ASL Lanciano Vasto Chieti, Chieti, Italy

⁶⁰Università della Calabria, Cosenza

⁶¹Ospedale Morgagni, Forli'

⁶²IRCCS Istituto Auxologico Italiano, Ospedale San Luca, Milano

⁶³Erasmus MC, Rotterdam

⁶⁴Ospedale Regina Montis Regalis, Mondovì, Cuneo

⁶⁵Hospital “Buon Consiglio-Fatebenefratelli,” Napoli

⁶⁶Ospedale di Perugia, Perugia

⁶⁷Universita' di Salerno

⁶⁸Ospedale di Crema, Crema

⁶⁹Cincinnati Children's Hospital and University of Cincinnati

⁷⁰IRCCS Istituto Ortopedico Galeazzi Milano

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcvm.2020.585866/full#supplementary-material

References