Miravitlles et al. Respiratory Research (2022) 23:49 https://doi.org/10.1186/s12931-022-01965-3 RESEARCH Open Access Determinants of blood eosinophil levels in the general population and patients with COPD: a population‑based, epidemiological study Marc Miravitlles1,2* , Juan José Soler‑Cataluña2,3, Joan B. Soriano2,4, Francisco García‑Río2,5, Pilar de Lucas6, Inmaculada Alfageme7, Ciro Casanova8, José Miguel Rodríguez González‑Moro9, M. Guadalupe Sánchez‑Herrero10, Julio Ancochea2,4 and Borja G. Cosío2,11 Abstract Background: Blood eosinophils are considered a biomarker for the treatment of chronic obstructive pulmonary disease (COPD). Population-based studies are needed to better understand the determinants of the blood eosinophil count (BEC) in individuals with and without COPD. Methods: EPISCAN II is a multicentre, cross-sectional, population-based epidemiological study aimed at investigating the prevalence and determinants of COPD in Spain. Study subjects were randomly selected from the general popula‑ tion, and COPD was defined by a post-bronchodilator ­FEV1/FVC < 0.7. For the pre-specified outcomes related to BEC, the first 35 COPD and 35 non-COPD subjects were consecutively recruited in 12 of the participating centres with the objective of analysing 400 individuals in each group. Baseline BEC and its association with demographic, clinical and functional variables were analysed. Results: A total of 326 COPD and 399 non-COPD subjects were included in the analysis. The mean age (standard deviation [SD]) was 63.2 years (11.0), 46.3% were male, and 27.6% were active smokers. BEC was significantly higher in individuals with COPD [192 cells/μL (SD: 125) vs. 160 cells/μL (SD: 114); p = 0.0003]. In a stepwise multivariate model, being male, active smoker and having a previous diagnosis of asthma were independently associated with having a higher BEC. Conclusions: This population-based study estimated the distribution of eosinophils in the healthy adult population and concluded that COPD patients have a significantly higher BEC. Male sex, active smoking and concomitant asthma were significantly associated with a higher BEC. Keywords: Epidemiology, Computed tomography, Imaging population, Biomarkers Background Chronic obstructive pulmonary disease (COPD) is a het- erogeneous disease with multiple phenotypes and endo- *Correspondence: marcm@separ.es types, which may be associated with different prognoses 1 Pneumology Department, Hospital Universitari Vall d’Hebron/Vall and response to treatment [1, 2]. One of the most widely d’Hebron Institut de Recerca (VHIR), Vall d’Hebron Barcelona Hospital studied phenotypes of COPD is that associated with Campus, P. Vall d’Hebron 119‑129, 08035 Barcelona, Spain Full list of author information is available at the end of the article increased eosinophilic inflammation. Although there is © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Miravitlles et al. Respiratory Research (2022) 23:49 Page 2 of 10 still controversy about the role of the blood eosinophil The study was approved by the ethics committees of count (BEC) as a marker of risk of exacerbation [3–8], each of the participating centres, and all participants pro- there is wide consensus about the role of the BEC as a vided informed consent. The EPISCAN II protocol is reg- biomarker to identify patients who will present better istered at https://​clini​caltr​ials.​gov (NCT03028207) and at response to inhaled corticosteroids (ICS) [9–11]. In fact, www.​gsk-​clini​calst​udyre​gister.​com/​study/​205932. the most recent recommendations of pharmacologic treatment include the BEC as a guide for the indication Variables and procedures of treatment with ICS, particularly in patients at risk of Demographic information on age, sex, level of educa- exacerbations [12, 13]. tion, family conditions, comorbidities, weight, height In order to better understand the role of blood eosino- and smoking were collected. The number of respira- phils as a biomarker in COPD it is important to generate tory exacerbations in the last 12 months was also col- new data about the distribution of blood eosinophil val- lected. Forced spirometry was performed using a ues in healthy adult populations and different groups of pneumotachograph (Vyntus Spiro, Carefusion, Ger- patients with COPD according to different phenotypes, many), according to standardized procedures [21], and degree of severity and treatments. While the distribution Global Lung Function Initiative equations were used of blood eosinophils has been examined in several United as the reference value [22]. Spirometry was performed States [3, 14] or North and Central Europe-based COPD before and 15–30 min after inhalation of 400 μg of sal- populations [14–18], data from Southern European butamol. COPD was defined as a post-bronchodilator countries are limited. forced expiratory volume in one second/forced vital EPISCAN II is a population-based, epidemiological capacity ­(FEV1/FVC) ratio < 0.7. The study popula- study the main objective of which was to investigate the tion was divided into two cohorts, depending on the prevalence and determinants of COPD in all autonomous results of the post-bronchodilator spirometry: patients communities in Spain [19]. One of the pre-specified sec- with COPD ­(FEV1/FVC < 0.7) and non-COPD individu- ondary objectives of the EPISCAN II study was to analyse als ­(FEV1/FVC ≥ 0.7). Symptoms were assessed by the the distribution of BECs in a subsample of participants COPD Assessment Test (CAT) questionnaire. Physi- with and without COPD in order to estimate BECs in the cal activity was measured by the Yale Physical Activity general adult Spanish population and to investigate fac- Questionnaire (YPAS) validated for the Spanish popu- tors associated with an elevated BEC [19]. The current lation and the elderly population. This questionnaire article presents the results of the analysis of BECs and reflects the amount, frequency, and intensity of physi- their determinants in a subgroup of participants in the cal activity that can be used to estimate the effects of EPISCAN II study with and without COPD from centres physical activity as a continuous parameter, even at the distributed throughout Spain. low levels of activity that might be expected in COPD patients [23, 24]. The 6-min walking test (6MWT) was Methods performed according to the American Thoracic Society EPISCAN II is a national, multicentre, cross-sectional, (ATS) guidelines [25]. population-based epidemiological study the main objec- Absolute values and percentage of diffusing capac- tive of which was to investigate the prevalence of COPD ity of the lung for carbon monoxide (DLCO) and alveo- in Spain. The full protocol, the fieldwork and all the lar volume were collected by single-breath CO diffusing methods have been described previously [19, 20]. Briefly, capacity (MasterScreen diffusion, Carefusion, Germany), 20 hospitals from the 17 Spanish autonomous commu- according to the ATS/European Respiratory Society nities conducted the study from April 2017 to February (ERS) recommendations [26]. Twenty ml of venous blood 2019. Subjects from the general population who were were collected from each participant for routine blood resident in the postal code areas nearest the participating analysis, and determination of C-reactive protein (CRP) hospitals were selected. A list of random telephone num- and fibrinogen. The collection procedure was standard- bers was obtained, stratified according to these postal ised, and each centre stored the samples at − 80 °C until codes and quotas for sex and age groups. The inclusion shipping for centralised analysis. criteria were as follows: men or women aged 40 years or Computed tomography (CT) images were acquired more with no physical or cognitive difficulties that would during inhalation, without contrast and with low- prevent them from completing spirometry or any of the dose radiation. The images obtained underwent semi- study procedures. automatic post-processing for determination of the Miravitlles et al. Respiratory Research (2022) 23:49 Page 3 of 10 percentage of emphysema, areas of extension, airway Regarding blood analysis, COPD individuals had no thickness, and other lung parenchyma attenuation and significant differences in red blood cells, platelets, fibrin- airway wall thickness parameters [27–29]. ogen (g/L) or CRP (mg/dL) values but did have higher concentrations of total leukocytes, neutrophils and Statistical analysis eosinophils (p < 0.05) (Table 1). Considering the 10.2% prevalence of COPD found in On CT imaging, COPD participants showed a higher the previous EPISCAN study [28] a priori sample size percentage of areas of emphysema [10% (11.9) vs. 4.7% calculation estimated that with an accuracy of ± 0.6% (7.5); p < 0.0001] and higher percentages of airway wall and a 10% dropout rate, approximately 10,200 eli- area in both primary and secondary bronchi (both gible individuals should be included in the study p < 0.0001) (Table 1). for the primary outcome, that is, the prevalence of COPD in Spain. For secondary outcomes related to Blood eosinophil counts according to the different subject CT scan and blood analysis, the first 35 COPD and characteristics 35 non-COPD subjects were consecutively invited The mean BEC was 175 cells/μL (SD: 120; 95% confi- to participate in 12 of the participating centres with dence interval [CI]: 166;184) and median 150 cells/ the objective of recruiting a total of approximately μL. BECs were significantly higher in individu- 400 individuals in each group. Categorical variables als with COPD (192 cells/μL (SD: 125; median: 177 were presented as numbers with percentages, and cells/μL) vs. 160 cells/μL (SD: 114; median: 132 cells/ continuous variables as mean with standard devia- μL); p = 0.0003). The distribution of blood eosino- tion (SD). The characteristics of the population in the phils in both groups of subjects is depicted in Fig. 1. diagnosis of COPD groups were compared using the There were no differences in the proportion of sub- Student’s t-test, or χ2 test. The comparisons of blood jects with more than 300 eosinophils/μL (11.5% in eosinophil levels according to clinical, functional and non-COPD vs. 16% in COPD, p = 0.08), but there were CT characteristics were tested for significance using more non-COPD subjects with less than 100 eosino- the Student’s t-test and analysis of variance. Correla- phils/μL compared with COPD patients (32.8% vs. tions between continuous variables were evaluated 20.6%, p = 0.0002). In the 159 subjects with airflow using Pearson’s correlation coefficient. Univariate and obstruction according to the lower limit of normal of multivariate linear regression analyses were used to FEV1/FVC the mean BEC was 204 cells/μL (SD: 135; investigate factors associated with blood eosinophil 95% confidence interval [CI]: 182;225) and median levels. Only variables with a level of significance < 0.1 180 cells/μL. BECs were also significantly higher in in the univariate analysis were included in a stepwise men, smokers, individuals with more advanced stages multivariate linear regression model. The R2 coeffi- of COPD and in those with a previous diagnosis of cient of determination was calculated for the model. asthma. Data were analysed with the Statistical Analysis Sys- The differences in BEC according to CT scan tem (SAS) Enterprise Guide 7.15, considering a statis- parameters did not achieve statistical significance tical significance (p) of 0.05 for all the statistical tests (Table 2). This analysis was repeated separately for the performed. COPD patients and the non-COPD participants, and neither subgroup showed significant differences in BEC between individuals with or without significant Results emphysema or airway thickness (data not shown). Population Since the majority of the COPD patients identified The 12 participating centres recruited a total of 326 were mild, we changed the cut off of emphysema COPD and 399 non-COPD subjects, who constituted from 10 to 5%, but again no significant differences the population of our study. The mean age of the partici- in BEC were observed (189.5 cells/μL (SD: 134.2) in pants was 63.2 years (SD): 11.0, and 46.3% were male, and patients with 0–5% emphysema compared to 202.8 27.6% were active smokers. Most clinical characteristics cells/μL (SD: 129.1) in patients with > 5% emphysema; were significantly different between COPD and non- p = 0.18). COPD participants. COPD subjects were older, more fre- Figure 2 shows the mean and 95%CI of blood eosino- quently male, less frequently never smokers, with lower phil levels in the different subgroups of participants lung function parameters, less exercise tolerance and classified according to the variables showing significant higher CAT scores (Table 1). differences in univariate analysis. Miravitlles et al. Respiratory Research (2022) 23:49 Page 4 of 10 Table 1 Demographic, clinical, blood, and imaging characteristics of participants according to the presence of airflow limitation compatible with COPD COPD (n = 326) Non-COPD (n = 399) All (n = 725) P value Demographic Age, years 66.9 (10.5) 60.1 (10.5) 63.2 (11.0) < 0.0001 Sex, male, % 182 (55.8%) 154 (38.6%) 336 (46.3%) < 0.0001 BMI 27.1 (4.4) 27.1 (4.8) 27.1 (4.6) 0.986 Smoker 114 (35.0%) 86 (21.6%) 200 (27.6%) < 0.0001 Former smoker 151 (46.3%) 152 (38.1%) 303 (41.8%) Never smoker 61 (18.7%) 161 (40.4%) 222 (30.6%) Clinical Asthma diagnosis 42 (12.9%) 32 (8.0%) 74 (10.2%) 0.031 CAT​ 9.2 (6.9) 7.0 (5.9) 8.0 (6.5) < 0.0001  ­FEV1, L 2.2 (0.7) 2.9 (0.7) 2.6 (0.8) < 0.0001  ­FEV1, % 82.0 (18.7) 104.6 (14.6) 94.4 (20.0) < 0.0001 DLCO, % 89.9 (23.1) 98.8 (18.4) 94.8 (21.1) < 0.0001  ­FEV1/FVC 62.0 (8.4) 79.8 (5.1) 71.8 (11.1) < 0.0001 6 MWD, m 478 (108) 518 (98) 501 (105) < 0.0001 YPAS, score 45.7 (23.1) 48.4 (21.6) 47.2 (22.3) 0.111 Exacerbations last year 0.27 (0.82) 0.05 (0.27) 0.14 (0.60) < 0.0001 Treatment with ICS 61 (18.7%) 16 (4.0%) 77 (10.6%) < 0.0001 Blood Erythrocyte, cells/µl 4.95 (0.52) 4.92 (0.44) 4.93 (0.47) 0.796 Leukocyte, cells/µl 7273 (2194) 6746 (1838) 6983 (2022) 0.0005 Neutrophils, cells/µl 4284 (1603) 3925 (1426) 4086 (1518) 0.001 Neutrophils, % 58.5 (9.2) 57.3 (8.3) 57.9 (8.8) 0.068 Eosinophils, cells/µl 192 (125) 160 (114) 175 (120) 0.0003 Eosinophils < 100 cell/μL 67 (20.6%) 131 (32.8%) 198 (27.3%) 0.0002 Eosinophils > 300 cells/μL 52 (16.0%) 46 (11.5%) 98 (13.5%) 0.080 Eosinophils, % 2.8 (1.8) 2.4 (1.6) 2.6 (1.7) 0.008 Platelets, cells/µl 233 (58) 241 (62) 238 (61) 0.070 Fibrinogen (g/L) 3.9 (0.9) 3.8 (0.9) 3.8 (0.9) 0.208 CRP (mg/dL) 2.0 (4.1) 1.5 (2.8) 1.7 (3.4) 0.0514 Imaging Emphysema, % 10.0 (11.9) 4.7 (7.5) 7.0 (10.0) < 0.0001 Airway wall area, % (Primary bronchi) 18.4 (2.9) 16.9 (3.4) 17.6 (3.3) < 0.0001 Airway wall area, % (Secondary bronchi) 27.5 (5.7) 24.6 (5.6) 25.9 (5.8) < 0.0001 BMI body mass index, CAT​COPD assessment test, FEV1 forced expiratory volume in the first second, FVC forced vital capacity, DLCO diffusion capacity of the lung for carbon monoxide, 6MWD 6-min walk distance, YPAS Yale Physical Activity Questionnaire, CRP C-reactive protein Factors independently associated with BEC ICS treatment, platelets, fibrinogen, CRP, emphysema A stepwise multivariate model was developed with the (% fixed threshold), and airway wall area (for central following variables: age, sex, body mass index, smoking, and peripheral airways in %). Only five variables had a p previous diagnosis of asthma, COPD, CAT scores, FEV1 value < 0.10 in the final model, with three being signifi- (%), DLCO (%), YPAS score, 6MWD, exacerbations, cant at a level of < 0.05. Being male, active smoker, and having a previous diagnosis of asthma were significantly Miravitlles et al. Respiratory Research (2022) 23:49 Page 5 of 10 45% 40% 35% 30% percentage 25% 20% 15% 10% 5% 0% <100 101-200 201-300 >300 COPD Non-COPD Fig. 1 Distribution of BEC in individuals with and without COPD and independently associated with a higher BEC in COPD patients (192 cells/μL). The concentrations (Table 3). observed in our population of non-COPD individuals were intermediate compared to those observed in an Discussion Austrian population described by Hartl et al. [17] with a Given the potential for BEC as a biomarker in COPD geometric mean of 128 cells/μL, or those reported in a [9, 11, 30], it is important to understand the distribu- study in the UK (140 cells/μL in smokers and 120 cells/μL tion and determinants of BEC in the general adult pop- in non-smokers) [18], and the concentrations observed ulation and in COPD patients. Our population-based in the UK Clinical Practice Research Datalink (182 cells/ sample of non-COPD adults recruited in the EPISCAN μL) [15]. However, they were very similar to the values II study had a mean BEC of 160 cells/μL, being sig- reported in Copenhagen, with a median value of 150 nificantly higher for COPD patients. We also observed cells/μL in adult never smokers and 160 cells/μL in ever a significantly higher BEC in males, active smokers, smokers [16]. and individuals with previous history of asthma and Our data replicate previous results showing a sig- increased levels of eosinophils with increasing severity nificantly higher BEC in subjects with COPD com- of airflow obstruction. pared with non-COPD controls. In the previously Since the EPISCAN II was a population-based epi- mentioned study by Kolsum et al. [18], the mean BEC demiological study, it is important to note that the was 209 cells/μL in COPD subjects, and in the UK majority of our population of COPD patients had mild database study, blood eosinophils were a mean of 8% disease, with 56% of GOLD stage I, and 75% were undi- higher in COPD compared with non-COPD subjects agnosed and untreated [20]. [15]. A previous population-based database study The mean BEC in non-COPD subjects was 160 cells/ in Spain in + 57,000 patients with COPD reported a μL, which was significantly lower than the mean count mean eosinophil count of 253 cells/μL [5], with 28.4% Miravitlles et al. Respiratory Research (2022) 23:49 Page 6 of 10 Table 2 Blood eosinophil levels according to clinical, functional of individuals with counts > 300 eosinophils/μL, being and CT characteristics clearly higher than the 192 cells/μL observed in our Blood eosinophil counts P value study with only 16% with BEC > 300 cells/μL. Simi- larly, in other Spanish cohorts, Cosio et al. [31] found Mean (SD) (95%CI) n a mean BEC of 240 cells/μL in a sample of 706 COPD Sex patients without asthma-COPD overlap, and Soler- Male 194.8 (127.7) (181.1; 208.5) 336 < 0.0001 Cataluña et al. [8] reported a mean BEC of 256 cells/μL Female 158.1 (111.7) (147.0; 169.3) 388 in 233 COPD patients, 38.2% with ≥ 300 eosinophils/ Age μL, while the mean BEC in asthmatic patients was 402 40–50 166.2 (123.4) (139.9; 192.5) 87 0.848 cells/μL. These higher values may be explained by the 50–60 178.5 (126.8) (161.1; 196.1) 204 different characteristics of the participants, since the 60–70 173.0 (118.6) (157.3; 188.8) 219 subjects in these studies had diagnosed COPD with a > 70 177.7 (116.4) (162.0; 193.4) 214 mean FEV1(%) of between 57 and 64%, and 74% to 85% BMI were men [5, 8, 30] compared with a mean F ­ EV1(%) of Underweight 166.8 (132.6) (44.2; 289.6) 7 0.928 82% and 56% males in our population. Actually, some Normal 171.7 (136.8) (154.4; 189.0) 243 of these characteristics have been associated with Overweight 178.3 (112.7) (165.6; 191.0) 306 BEC; in our study, male sex, active smoking, previ- Obese 173.9 (108.5) (157.2; 190.6) 165 ous history of asthma and impaired F ­ EV1(%) were sig- Smoking nificantly and independently associated with a higher Smoker 189.4 (123.2) (172.2; 206.6) 200 0.025 BEC. Moreover, in previous studies, male sex [16, 17], Former smoker 178.1 (123.3) (164.2; 192.1) 303 active smoking [17, 18] and asthma diagnosis [15, Never smoker 158.2 (113.1) (143.2; 173.2) 221 17] have been described as being associated with a Asthma diagnosis higher BEC. Other factors associated with increased Yes 204.5 (142.0) (171.6; 237.4) 74 0.027 eosinophils described in large database study are: No 171.8 (117.7) (162.8; 180.9) 650 age < 18 years, positive skin prick test, atopy, a positive COPD bronchodilator test, metabolic syndrome and obesity No 160.7 (114.7) (149.4; 172.0) 399 0.001 [16, 17]. Our study found not only a significant differ- GOLD I 185.8 (115.9) (168.9; 202.8) 183 ence in BEC between COPD and non-COPD individu- GOLD II 197.4 (133.4) (173.3; 221.5) 120 als, but also an increase in BEC from non-COPD to GOLD III/IV 227.0 (158.0) (157.0; 297.1) 22 GOLD stage III/IV COPD, and similarly, a significant DLCO negative correlation between eosinophils and F ­ EV1(%) > 80% 170.8 (118.5) (160.8; 180.8) 545 0.202 indicating higher levels in more severe disease. < 80% 184.2 (116.9) (166.2; 202.3) 163 Interestingly, we did not observe any significant asso- Exacerbations ciation between blood eosinophils and exacerbations 0 173.5 (119.2) (164.4; 182.5) 667 0.222 or ICS treatment, probably due to the characteristics of 1 or more 195.0 (137.1) (158.6; 231.4) 57 our population which was mainly made up of mild and YPAS score untreated patients. Our results concur with those of a < 51 sedentary 180.3 (122.1) (168.5; 192.0) 416 0.207 previous large database study in patients at low risk of ≥ 51 non-sedentary 169.2 (118.5) (155.6; 182.7) 295 exacerbations in Primary Care that did not observe any Emphysema influence of treatment with ICS on BEC, and no asso- 0–10% 173.0 (123.7) (161.9; 184.1) 480 0.084 ciation was found between blood eosinophils and the > 10% 190.3 (130.5) (167.9; 212.8) 132 frequency of exacerbations [5]. Airway Wall thickness (Primary Bronchi) Blood eosinophils were not associated with percent airway wall area for either the central or peripheral air- 0–20% 179.9 (124.3) (168.9; 191.0) 485 0.071 ways. Apparently, there was a trend towards a higher > 20% 164.4 (128.4) (141.9; 187.0) 127 BEC in patients with COPD and more emphysema on Airway Wall thickness (Secondary Bronchi) CT, but the differences were not significant. On the 0–30% 173.9 (119.5) (163.3; 184.5) 489 0.192 contrary, Papaioannou et al. [32] found that emphy- > 30% 190.4 (146.0) (164.2; 216.7) 121 sema was associated with low blood eosinophil counts BMI body mass index, GOLD Global Initiative for Obstructive Lung Disease, COPD chronic obstructive pulmonary disease, YPAS Yale Physical Activity in a group of 98 patients with COPD. Again, these dif- Questionnaire, DLCO diffusion capacity of the lung for carbon monoxide, CI ferences may be due to the different characteristics of confidence interval, SD standard deviation Miravitlles et al. Respiratory Research (2022) 23:49 Page 7 of 10 GOLD III/IV GOLD II COPD GOLD I p<0.0001 No No Asthma p=0.0257 Yes Never smoker Smoking Former smoker p=0.0274 Smoker Female p=0.0015 Sex Male 100 150 200 250 300 350 Fig. 2 Mean and 95% confidence intervals of BEC of individuals included in the study according to different characteristics. Vertical dotted line denotes mean blood eosinophil value for the whole population the populations analysed; they defined emphysema as despite these characteristics, there were clear and signifi- the presence of emphysema lesions in > 15% of the pul- cant differences in their characteristics compared with monary parenchyma, and their emphysema patients non-COPD individuals, including a significantly higher had a mean F ­ EV1(%) of 43%. In contrast, there was concentration of blood eosinophils. The cross-sectional an underrepresentation of patients with moderate design did not allow evaluation of the possible prognostic and severe COPD in our sample, and the mean per- value of BEC for outcomes such as exacerbations. How- cent emphysema area was 10% in our COPD patients. ever, the design of the study had the strength to allow the A recent Canadian study described that patients with analysis of an unselected population of COPD and non- COPD and an elevated BEC had thickened central air- COPD individuals from all geographic areas of Spain. way walls and a reduction in the total number of vis- Finally, we did not systematically assess the presence of ible airways, indicating airway remodelling [33]. More possible causes of hypereosinophilia, although they have studies are needed to clarify the relationship between a low prevalence in the general population. BEC and the classic phenotypes of chronic bronchitis and emphysema in COPD. Our study has several limitations, since it was a pop- Conclusions ulation-based study of the general population, patients The results of our population-based study provide an identified with COPD were predominantly mild, undi- estimate of the distribution of eosinophils in the healthy agnosed and paucisymptomatic, with underrepresenta- adult population in Spain and has demonstrated that tion of patients with other degrees of severity. However, COPD patients, even with milder stages, have a signifi- cantly higher BEC. In addition to COPD, male sex, active Miravitlles et al. Respiratory Research (2022) 23:49 Page 8 of 10 Table 3 Multivariate analysis of the factors associated with the BEC Parameter Univariate Multivariate P value Estimate (standard error) P value Intercept 208.09 (31.01) < 0.0001 Age (years) 0.4555 Sex Female Reference Reference Male < 0.0001 30.98 (10.65) 0.0038 BMI (kg/m2) 0.8615 Smoking Never smoker Reference Reference Former smoker 0.0620 10.97 (12.73) 0.3892 Active smoker 0.0081 31.21 (13.68) 0.0229 Asthma No Reference Reference Yes 0.0274 36.91 (16.83) 0.0288 CAT score 0.1159 FEV1 (%) 0.0002 − 0.53 (0.28) 0.0543 DLCO (%) 0.0474 YPAS score 0.0493 6 MWD, m 0.3057 Exacerbations last year 0.0331 COPD No Reference Yes 0.0003 Treatment with ICS 0.0099 Platelets, cells/µl 0.1026 Fibrinogen (g/L) 0.9357 CRP (mg/dL) 0.0831 Total emphysema volume (%) 0.0082 0.94 (0.51) 0.0660 % Airway wall area (Primary Bronchi) 0.4094 % Airway wall area (Secondary bronchi) 0.1527 BMI body mass index, CAT​COPD assessment test, FEV1 forced expiratory volume in the first second, FVC forced vital capacity, DLCO diffusion capacity of the lung for carbon monoxide, 6MWD 6-min walk distance, YPAS Yale Physical Activity Questionnaire, CRP C-reactive protein. Values in bold are statistically significant. R-Square = 0.067 smoking, the presence of asthma and a worse ­FEV1(%) José Soler-Cataluña, Francisco García-Río, Ciro Casanova, José Miguel Rod‑ ríguez González-Moro, Borja G. Cosio, Guadalupe Sánchez, and Joan B Soriano. are significantly associated with a higher BEC. Principal Investigators, collaborators and participating centres: MADRID: H. La Princesa, Julio Ancochea Bermudez (IP) / Elena García Castillo / Claudia Valenzuela / Joan B Soriano. CASTILLA LEÓN: H. U. de Burgos, Ana Pueyo Abbreviations Bastida (IP) / Lourdes Lázaro Asegurado / Luis Rodríguez Pascual / Mª José ATS: American Thoracic Society; BEC: Blood eosinophil count; CAT​: COPD Mora. ARAGÓN: H. Gral. San Jorge, Luis Borderias Clau (IP) / Lourdes Arizón assessment test; COPD: Chronic obstructive pulmonary disease; CT: Computed Mendoza / Sandra García. EXTREMADURA: H. San Pedro de Alcántara, tomography; CRP: C-reactive protein; DLCO: Diffusing capacity of the lung Juan Antonio Riesco Miranda (IP) / Julián Grande Gutiérrez / Jesús Agustín for carbon monoxide; EPISCAN: Epidemiological study of COPD in Spain; ERS: Manzano / Manuel Agustín Sojo González. CASTILLA LEÓN: H. Clínico U. de European Respiratory Society; FEV1: Forced expiratory volume in 1 s; FVC: Salamanca, Miguel Barrueco Ferrero (IP) / Milagros Rosales. GALICIA: H. Álvaro Forced vital capacity; GOLD: Global initiative for obstructive lung disease; ICS: Cunqueiro, José Alberto Fernández Villar (IP) / Cristina Represas / Ana Priegue Inhaled corticosteroids; SAS: Statistical analysis system; SD: Standard deviation; / Isabel Portela Ferreño / Cecilia Mouronte Roibás / Sara Fernández García. 6MWT: Six-minute walking test; UK: United Kingdom; YPAD: Yale Physical Activ‑ I. BALEARES: H. Son Espases, Borja G Cosío (IP) / Rocío Cordova Díaz / Nuria ity Questionnaire. Toledo Pons / Margalida Llabrés. ARAGÓN: H. U. Miguel Servet, José María Marín Trigo (IP) / Marta Forner / Begoña Gallego / Pablo Cubero / Elisabet Vera. Acknowledgements C. VALENCIANA: H. Arnau de Vilanova (Valencia), Juan José Soler Cataluña (IP) The collaboration of Monica Sarmiento and Neus Canal of IQVIA, and Carolina / Mª Begoña Picurelli Albero / Noelia González García. ANDALUCÍA: H. Virgen Peña and José Julio Jiménez from GSK is explicitly acknowledged. The fol‑ de la Macarena, Agustín Valido Morales (IP) / Carolina Panadero / Cristina lowing investigators have participated in EPI SCAN II: Scientific Committee: Benito Bernáldez/ Laura Martín -Bejarano y Maria Velar. MURCIA: H. Gral. U. Inmaculada Alfageme, Pilar de Lucas, Julio Ancochea, Marc Miravitlles, Juan Santa Lucía (Cartagena), Antonio Santa Cruz Siminiani (IP) / Carlos Castillo Miravitlles et al. Respiratory Research (2022) 23:49 Page 9 of 10 Quintanilla / Rocío Ibáñez Meléndez / José Javier Martínez Garcerán / Desirée Author details 1 Lozano Vicente / Pedro García Torres / Maria del Mar Valdivia. NAVARRA: Pneumology Department, Hospital Universitari Vall d’Hebron/Vall d’Hebron Clínica Universidad de Navarra, Juan Pablo de Torres Tajes (IP) / Montserrat Institut de Recerca (VHIR), Vall d’Hebron Barcelona Hospital Campus, P. Vall Cizur Girones / Carmen Labiano Turrillas. LA RIOJA: H. de San Pedro (Logroño), d’Hebron 119‑129, 08035 Barcelona, Spain. 2 CIBER de Enfermedades Respira‑ Carlos Ruiz Martínez (IP)/ Elena Hernando / Elvira Alfaro / José Manuel García torias (CIBERES), Instituto de Salud Carlos III (ISCIII), Madrid, Spain. 3 Servicio / Jorge Lázaro. PAÍS VASCO: H. Santiago Apóstol (H. Txagorritxu), David Bravo de Neumología, Hospital Arnau de Vilanova‑Lliria, Universidad de la Laguna, (IP) /Laura Hidalgo / Silvia Francisco Terreros / Iñaki Zorrilla / Ainara Alonso Valencia, Spain. 4 Servicio de Neumología, Hospital Universitario la Princesa, Colmenero. ASTURIAS: H. Central de Asturias, Cristina Martínez González (IP) / Universidad Autónoma de Madrid, Madrid, Spain. 5 Servicio de Neumología, Susana Margon/ Rosirys Guzman Taveras/ Ramón Fernández / Alicia Álvarez. Hospital Universitario la Paz-IdiPAZ, Madrid, Spain. 6 Servicio de Neumología, CANTABRIA: H. de Valdecilla, José Ramón Agüero Balbín (IP) / Juan Agüero Hospital General Gregorio Marañon, Madrid, Spain. 7 Unidad de Gestión Clínica Calvo. CATALUÑA: H. U. Vall d’Hebron, Jaume Ferrer Sancho (IP) / Esther Rod‑ de Neumología, Hospital Universitario Virgen de Valme, Universidad de Sevilla, ríguez González/ Eduardo Loeb. CASTILLA LA MANCHA: H. U. de Guadalajara, Sevilla, Spain. 8 Servicio de Neumología, Hospital Universitario Nuestra Señora José Luis Izquierdo Alonso (IP) / Mª Antonia Rodríguez García. I. CANARIAS: de Candelaria, Tenerife, Spain. 9 Servicio de Neumología, Hospital Universitario H. U. de Tenerife, Juan Abreu González (IP)/ Candelaria Martín García/ Rebeca “Príncipe de Asturias”, Alcalá de Henares, Universidad de Alcalá, Madrid, Spain. 10 Muñoz/ Haydée Martín García. ASTURIAS: H. U. de San Agustín (Avilés), Miguel Departamento Médico, GSK, Tres Cantos, Madrid, Spain. 11 Servicio de Neu‑ Angel Martínez Muñiz (IP) / Andrés Avelino Sánchez Antuña / Jesús Allende mología, Hospital Universitario Son Espases-IdISBa, Palma de Mallorca, Spain. González / Jose Antonio Gullón Blanco / Fernando José Alvarez Navascues / Manuel Angel Villanueva Montes / María Rodríguez Pericacho / Concepción Received: 1 October 2021 Accepted: 20 February 2022 Rodríguez García / Juan Diego Alvarez Mavárez. Authors’ contributions The study concept and design: MM, JJS-C, JBS, FG-R, PL, IA, CC, JMG-M, GS, JA, BGC. The acquisition of the data: JJS-C, JA, BGC. Analysis or interpretation References of the data: MM, JJS-C, JBS, BGC. The drafting of the manuscript: MM. Critical 1. Cosio BG, Soriano JB, López-Campos JL, Calle M, Soler JJ, de Torres JP, revision and approval for submission: MM, JJS-C, JBS, FG-R, PL, IA, CC, JMG-M, et al. Distribution and outcomes of a phenotype-based Approach to GS, JA, BGC. All authors read and approved the final manuscript. guide COPD management: results from the CHAIN cohort. PLoS ONE. 2016;11(9):e0160770. Funding 2. Miravitlles M, Calle M, Soler-Cataluña JJ. Clinical phenotypes of COPD. The EPI SCAN II study was funded by an unrestricted grant from Identification, definition and implications for guidelines. Arch Bronconeu‑ GlaxoSmithKline. mol. 2012;48:86–98. 3. Yun JH, Lamb A, Chase R, Singh D, Parker MM, Saferali A, et al. Blood Availability of data and materials eosinophil count thresholds and exacerbations in patients with Information on the GSK data sharing commitments and requesting access to chronic obstructive pulmonary disease. J Allergy Clin Immunol. anonymized individual participant data and associated documents can be 2018;141(6):2037–47. found at www.​clini​calst​udyda​tareq​uest.​com. 4. Kerkhof M, Chaudhry I, Pavord ID, Miravitlles M, Kook Rhee C, Halpin DMG, et al. Blood eosinophil count predicts treatment failure and hospi‑ tal readmission for COPD. ERJ Open Res. 2020;6(4):00188–2020. Declarations 5. Miravitlles M, Monteagudo M, Solntseva I, Alcázar B. Blood eosino‑ phil counts and their variability and risk of exacerbations in COPD: a Ethics approval and consent to participate population-based study. Arch Bronconeumol. 2021;57:13–20. The study was approved by the ethics committees of each of the participating 6. Golpe R, Dacal D, Sanjuán-López P, Martín-Robles I, Pérez-de-Llano LA. centres, and all participants provided informed consent. Plasma eosinophil count and patient-centered events in chronic obstruc‑ tive pulmonary disease in real-life clinical practice. Arch Bronconeumol. Consent for publication 2020;56(2):129–30. Not applicable. 7. Singh D, Wedzicha JA, Siddiqui S, de la Hoz A, Xue W, Magnussen H, et al. Blood eosinophils as a biomarker of future COPD exacerbation risk: Competing interests pooled data from 11 clinical trials. Respir Res. 2020;21(1):240. Marc Miravitlles has received speaker or consulting fees from AstraZeneca, 8. Soler-Cataluña JJ, Novella L, Soler C, Nieto ML, Esteban V, Sánchez-Toril F, Bial, Boehringer Ingelheim, Chiesi, Cipla, CSL Behring, Laboratorios Esteve, et al. Clinical characteristics and risk of exacerbations associated with dif‑ Gebro Pharma, Kamada, GlaxoSmithKline, Grifols, Menarini, Mereo Biopharma, ferent diagnostic criteria of asthma-COPD overlap. Arch Bronconeumol. Novartis, pH Pharma, Palobiofarma SL, Rovi, TEVA, Spin Therapeutics, Verona 2020;56(5):282–90. Pharma and Zambon, and research grants from Grifols. Juan José Soler-Cata‑ 9. Pavord ID, Lettis S, Locantore N, Pascoe S, Jones PW, Wedzicha JA, et al. luña has received speaker fees from AstraZeneca, Bial, Boehringer Ingelheim, Blood eosinophils and inhaled corticosteroid/long-acting β-2 agonist Chiesi, Esteve, Ferrer, GlaxoSmithKline, Menarini, Novartis and Teva, and and efficacy in COPD. Thorax. 2016;71:118–25. consulting fees from AstraZeneca, Bial, Boehringer Ingelheim, GlaxoSmith‑ 10. Brusselle G, Pavord ID, Landis S, Pascoe S, Lettis S, Morjaria N, et al. Blood Kline, Ferrer and Novartis. Francisco García-Río has received speaker or con‑ eosinophil levels as a biomarker in COPD. Respir Med. 2018;138:21–31. sulting fees from AstraZeneca, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, 11. Singh D, Bafadhel M, Brightling CE, Sciurba FC, Curtis JL, Martinez FJ, et al. Menarini, Novartis, Pfizer and Rovi, and research grants from Chiesi, Esteve, Blood eosinophil counts in clinical trials for chronic obstructive pulmo‑ Gebro Pharma, GlaxoSmithKline, Menarini and TEVA. Julio Ancochea has nary disease. Am J Respir Crit Care Med. 2020;202(5):660–71. received speaker or consulting fees from Actelion, Air Liquide, Almirall, Astra‑ 12. Global Strategy For The Diagnosis, Management, And Prevention Of Zeneca, Boehringer Ingelheim, Carburos Médica, Chiesi, Faes Farma, Ferrer, Chronic Obstructive Pulmonary Disease. 2021 Report. https://​goldc​opd.​ GlaxoSmithKline, InterMune, Linde Healthcare, Menarini, MSD, Mundipharma, org/​wp-​conte​nt/​uploa​ds/​2020/​11/​GOLD-​REPORT-​2021-​v1.1-​25Nov​20_​ Novartis, Pfizer, Roche, Rovi, Sandoz, Takeda y Teva. Inmaculada Alfageme has WMV.​pdf [Accessed May 16th, 2021. no conflicto of interest. Ciro Casanova has received speaker or consulting 13. Miravitlles M, Calle M, Molina J, Almagro P, Gómez JT, Trigueros JA, et al. fees from AstraZeneca, Bial, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, Spanish COPD Guidelines (GesEPOC) 2021: updated pharmacological Menarini, Novartis, and research grants from GlaxoSmithKline, Menarini and treatment of stable COPD. Arch Bronconeumol 2022;58:69–81. https://​ AstraZeneca. M. Guadalupe Sánches-Herrero is a GSK employee within the doi.​org/​10.​1016/j.​arbres.​2021.​03.​005. Medical Department. Borja G Cosio has received speaker or consulting fees 14. Vogelmeier CF, Kostikas K, Fang J, Tian H, Jones B, Morgan CL, et al. from AstraZeneca, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, Menarini, Evaluation of exacerbations and blood eosinophils in UK and US COPD Novartis, Sanofi, TEVA, and research grants from Menarini, AstraZeneca and populations. Respir Res. 2019;20(1):178. Boehringer-Ingelheim. Miravitlles et al. Respiratory Research (2022) 23:49 Page 10 of 10 15. Landis S, Suruki R, Maskell J, Bornar K, Hilton E, Compton C. Demographic and clinical characteristics of COPD patients at different blood eosinophil levels in the UK clinical practice research datalink. COPD. 2018;15:177–84. 16. Çolak Y, Afzal S, Nordestgaard BG, Marott JL, Lange P. Combined value of exhaled nitric oxide and blood eosinophils in chronic airway disease: the Copenhagen General Population Study. Eur Respir J. 2018;52(2):1800616. 17. Hartl S, Breyer MK, Burghuber OC, Ofenheimer A, Schrott A, Urban MH, et al. Blood eosinophil count in the general population: typical values and potential confounders. Eur Respir J. 2020;55(5):1901874. 18. Kolsum U, Southworth T, Jackson N, Singh D. Blood eosinophil counts in COPD patients compared to controls. Eur Respir J. 2019;54(4):1900633. 19. Alfageme I, de Lucas P, Ancochea J, Miravitlles M, Soler-Cataluña JJ, García-Río F, et al. 10 Years After EPISCAN: a new study on the prevalence of COPD in Spain—a summary of the EPISCAN II protocol. Arch Bron‑ coneumol. 2019;55:38–47. 20. Soriano JB, Alfageme I, Miravitlles M, de Lucas P, Soler-Cataluña JJ, García- Río F, et al. Prevalence and Determinants of COPD in Spain: EPISCAN II. Arch Bronconeumol. 2021;57:61–9. 21. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al. ATS/ERS Task Force. Standardization of spirometry. Eur Respir J. 2005;26:319–38. 22. Quanjer PH, Stanojevic S, Cole TJ, Baur X, Hall GL, Culver BH, ERS Global Lung Function Initiative, et al. Multi-ethnic reference values for spirom‑ etry for the 3–95-yr age range: the global lung function 2012 equations. Eur Respir J. 2012;40:1324–43. 23. De Abajo S, Larriba R, Márquez S. Validity and reliability of the Yale Physical Activity Survey in Spanish elderly. J Sports Med Phys Fitness. 2001;41:479–85. 24. Donaire-González D, Gimeno-Santos E, Serra I, Roca J, Balcells E, Rodríguez E, et al. Validación del cuestionario de actividad física de Yale en pacientes con enfermedad pulmonar obstructiva crónica. Arch Bron‑ coneumol. 2011;47:552–60. 25. American Thoracic Society ATS Statement: Guidelines for the Six-Minute Walk Test. Am J Respir Crit Care Med 2002; 166: 111–117. 26. MacIntyre N, Crapo RO, Viegi G, Johnson DC, van der Grinten CPM, Brusasco V, et al. Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur Respir J. 2005;26:720–35. 27. Owrangi AM, Etemad-Rezai R, McCormack DG, Cunningham IA, Parraga G. Computed tomography density histogram analysis to evaluate pulmo‑ nary emphysema in ex-smokers. Acad Radiol. 2013;20:537–45. 28. Galban C, Han M, Boes J, Chughtai K, Meyer C, Johnson T, et al. Computed tomography-based biomarker provides unique signature for diagnosis of COPD phenotypes and disease progression. Nat Med. 2012;18:1711–5. 29. Nambu A, Zach J, Schroeder J, Jin G, Kim SS, Kim YI, et al. Quantitative computed tomography measurements to evaluate airway disease in chronic obstructive pulmonary disease: relationship to physiological measurements, clinical index and visual assessment of airway disease. Eur J Radiol. 2016;85:2144–51. 30. Bafadhel M, Pavord ID, Russell REK. Eosinophils in COPD: just another biomarker? Lancet Respir Med. 2017;5(9):747–59. 31. Cosio BG, Soriano JB, López-Campos JL, Calle-Rubio M, Soler-Cataluna JJ, de Torres JP, et al. Defining the asthma-COPD overlap syndrome in a COPD cohort. Chest. 2016;149(1):45–52. 32. Papaioannou AI, Kostikas K, Papaporfyriou A, Angelakis L, Papathanasiou E, Hillas G, et al. Emphysematous phenotype is characterized by low blood eosinophils: a cross-sectional study. COPD. 2017;14:635–40. 33. Tan WC, Bourbeau J, Nadeau G, Wang W, Barnes N, Landis SH, et al. High Ready to submit your research ? Choose BMC and benefit from: eosinophil counts predict decline in FEV1: results from the CanCOLD study. Eur Respir J. 2021;57(5):2000838. • fast, convenient online submission • thorough peer review by experienced researchers in your field Publisher’s Note • rapid publication on acceptance Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions