Figures
- FIGURE 1
Flow diagram of the subjects screened and included in the full cohort at the start of year 1 and the number of exacerbations in year 1. RE: rarely eosinophilic; IE: intermittently eosinophilic; PE: predominantly eosinophilic.
- FIGURE 2
Receiver operating characteristic curve with area under the curve (AUC) for blood eosinophils (count and percentage) at enrolment predicting sputum eosinophilia >3% at enrolment (n=68). Blood eosinophil count: AUC 0.768, 95% CI 0.651–0.884; blood eosinophil percentage: AUC 0.851, 95% CI 0.750–0.951.
- FIGURE 3
Receiver operating characteristic curve with area under the curve (AUC) for blood eosinophils (count and percentage) at enrolment predicting the predominantly eosinophilic group over 12 months following enrolment (n=78). Blood eosinophil count: AUC 0.806, 95% CI 0.710–0.901; blood eosinophil percentage: AUC 0.841, 95% CI 0.755–0.928.
- FIGURE 4
Seasonal distribution of total and eosinophil-associated exacerbations. a) Number of total and exacerbations with blood eosinophils ≥2%. b) Proportion of exacerbations with blood eosinophils ≥2% defined as exacerbations with blood eosinophils ≥2% to total exacerbation rates in the predominantly eosinophilic, intermittent eosinophilic and rarely eosinophilic groups.
Tables
- TABLE 1
Baseline characteristics of the total chronic obstructive pulmonary disease (COPD) cohort
Total patients 127 Continuous variables Age years 66.8±8.61 Smoking history pack-years 50.3±28.2 Body mass index kg·m–2 27.0 (6.69) Fat-free body mass % 48.9 (21.5) (n=125) White blood cells count ×109 L–1 7.60 (2.20) (n=126) Eosinophils count ×109 L–1 0.20 (0.20) (n=126) Eosinophils % 2.94 (3.08) (n=126) Neutrophils count ×109 L–1 4.80 (1.70) (n=126) Fibrinogen g·L–1 4.80 (1.02) (n=114) C-reactive protein mg·L–1 5.00 (8.00) Eosinophils %# 1.93 (5.05) (n=83) Neutrophils %# 47.0 (70.4) (n=83) FEV1 % pred 46.7 (25.3) (n=126) ΔFEV1 % of baseline¶ 4.93 (21.1) (n=85) FEV1 reversibility % of pre-bronchodilator FEV1+ 11.2 (17.0) (n=105) KCO % 69.2 (30.7) (n=122) TLCO % 57.9 (29.5) (n=122) CAT score 16.0 (10.0) (n=126) 6MWT distance m 300 (170) (n=125) EXACT-PRO score 37.0 (12.0) (n=101) Exacerbation rate in year before study 2.00 (2.00) Exacerbation rate in first year of study 2.94 (3.91) Eosinophilic exacerbation rate in first year of study 0.99 (2.02) Follow-up in first year of study years 1.00 (0.02) Categorical variables Sex Male 68 (54) Female 59 (46) Current smoker§ Yes 54 (43) No 73 (57) Frequent exacerbators in year 0 Yes 99 (78) No 28 (22) Use of ICS at enrolmentƒ Yes 113 (89) No 14 (11) Sputum eosinophils >3%# Yes 27 (33) No 56 (67) Blood eosinophils ≥2% Yes 86 (68) No 40 (32) Bacteria present## Yes 57 (52) No 53 (48) Virus present¶¶ Yes 18 (17) No 85 (83) Blood eosinophils ≥200 cells·μL–1 Yes 90 (71.4) No 36 (28.6) Data are presented as n for patient numbers, mean±sd or median (interquartile range) for continuous variables, or n (%) for categorical variables. FEV1: forced expiratory volume in 1 s; KCO: transfer coefficient of the lung for carbon monoxide; TLCO: transfer factor of the lung for carbon monoxide; CAT: COPD Assessment Test; 6MWT: 6-min walk test; ICS: inhaled corticosteroid. #: sputum eosinophil and neutrophil percentage at baseline is reported (“baseline” is equal to enrolment if good quality data (squamous cell contamination <30%) is present at enrolment or the next (pre-exacerbation) stable visit with quality data within 4 months of enrolment); ¶: calculated as FEV1 at month 12×100/FEV1 at enrolment; +: calculated as (post-bronchodilator FEV1–pre-bronchodilator FEV1)/pre-bronchodilator FEV1×100; §: smoking status report derived from American Thoracic Society Division of Lung Disease questionnaire ATS-DLD-78A; ƒ: ICS use was coded as “Yes” if one of the following medications/inhalers was on the list: Symbicort, Seretide, QVAR, Fostair, beclomethasone, beclomethasone/formoterol, beclomethasone dipropionate, Clenil, fluticasone/salmeterol or budesonide/formoterol; ##: sputum sampling, measured by culture (includes Haemophilus influenzae, Moraxella catarrhalis, Streptococcus pneumoniae, Staphylococcus aureus and Pseudomonas aeruginosa); ¶¶: sputum sampling, measured by PCR (includes adenovirus, enterovirus, influenza, coronavirus, metapneumovirus, bocavirus, parainfluenza, respiratory syncytial virus and rhinovirus).
- TABLE 2
Baseline characteristics by longitudinal blood eosinophil group over 12 months for the 99 patients with sufficient data
Rarely eosinophilic Intermittently eosinophilic Predominantly eosinophilic p-value++ Patients 26 19 59 Continuous variables Age 62.7±8.63 67.9±6.87 (n=16) 68.0±9.05 (n=57) 0.034 Smoking history pack-years 46.3±22.8 64.9±43.2 (n=16) 51.0±28.6 (n=57) 0.098 Body mass index kg·m–2 28.2 (8.15) 27.9 (5.69) (n=16) 25.9 (4.73) (n=57) 0.280 Fat-free body mass % 53.5 (21.5) (n=25) 49.3 (18.3) (n=16) 48.8 (22.4) (n=56) 0.886 White blood cells count ×109 L–1 8.30 (3.30) (n=25) 7.65 (1.65) (n=16) 7.30 (2.00) (n=57) 0.060 Blood eosinophils count ×109 L–1 0.10 (0.10) (n=25) 0.20 (0.10) (n=16) 0.30 (0.20) (n=57) NA Blood eosinophils % 1.35 (1.45) (n=25) 2.12 (1.09) (n=16) 4.08 (3.05) (n=57) NA Blood neutrophils count ×109 L–1 5.20 (2.55) (n=25) 4.80 (0.50) (n=16) 4.30 (1.75) (n=57) 0.030 Fibrinogen g·L–1 4.70 (1.40) (n=23) 4.75 (0.97) (n=14) 4.80 (0.90) (n=51) 0.905 C-reactive protein mg·L–1 4.00 (7.50) 5.00 (5.75) (n=16) 5.00 (7.00) (n=57) 0.966 Sputum eosinophils %# 0.43 (2.89) (n=18) 0.82 (1.78) (n=7) 2.81 (7.40) (n=40) 0.070 Sputum neutrophils %# 9.77 (64.0) (n=18) 62.3 (63.4) (n=7) 38.0 (71.6) (n=40) 0.223 FEV1 % 47.7 (25.6) 43.9 (22.4) (n=16) 47.5 (26.0) (n=57) 0.661 ΔFEV1 % of baseline¶ 3.79 (27.3) (n=19) 10.8 (23.2) (n=15) 5.03 (21.4) (n=45) 0.375 FEV1 reversibility % of pre-bronchodilator FEV1+ 7.69 (15.0) (n=24) 17.9 (16.6) (n=13) 12.6 (12.5) (n=46) 0.355 KCO % 74.2 (31.0) (n=25) 68.3 (26.7) (n=16) 69.3 (30.3) (n=53) 0.526 TLCO % 62.0 (41.1) (n=25) 57.8 (28.8) (n=16) 56.3 (27.7) (n=53) 0.407 CAT score 15.5 (10.0) 20.5 (13.0) (n=16) 16.0 (9.00) (n=57) 0.101 6MWT distance m 300 (199) (n=25) 323 (172) (n=16) 327 (164) (n=56) 0.914 EXACT-PRO score 36.0 (14.0) (n=23) 36.5 (7.00) (n=14) 36.0 (15.0) (n=46) 0.607 Exacerbation rate in year before study 2.00 (3.00) 2.00 (3.00) (n=16) 2.00 (2.00) (n=57) 0.941 Exacerbation rate in first year of study 2.47 (3.22) 1.02 (2.00) (n=16) 2.04 (3.95) (n=57) 0.197 Eosinophilic exacerbation rate in first year of study 0.00 (0.99) 0.00 (0.99) (n=16) 1.38 (2.99) (n=57) <0.001 Follow-up in first year of study years 1.01 (0.02) 1.01 (0.01) (n=16) 1.01 (0.02) (n=57) 0.448 Categorical variables Sex Male 12 (46) 9 (56) 34 (60) 0.547 Female 14 (54) 7 (44) 23 (40) Current smoker§ Yes 15 (58) 4 (25) 24 (42) 0.125 No 11 (42) 12 (75) 33 (58) Use of ICS at enrolmentƒ Yes 21 (81) 15 (94) 51 (89) 0.468 No 5 (19) 1 (6) 6 (11) Sputum eosinophils >3%# Yes 2 (11) 0 (0) 19 (48) 0.003 No 16 (89) 7 (100) 21 (53) Blood eosinophils ≥2% Yes 7 (28) 9 (56) 55 (96) NA No 18 (72) 7 (44) 2 (4) Bacteria present## Yes 12 (55) 6 (50) 26 (50) 0.952 No 10 (45) 6 (50) 26 (50) Virus present¶¶ Yes 4 (19) 0 (0) 10 (20) 0.288 No 17 (81) 12 (100) 40 (80) Data are presented as n for patient numbers, mean±sd or median (interquartile range) for continuous variables, or n (%) for categorical variables, unless otherwise stated. FEV1: forced expiratory volume in 1 s; KCO: transfer coefficient of the lung for carbon monoxide; TLCO: transfer factor of the lung for carbon monoxide; CAT: COPD Assessment Test; 6MWT: 6-min walk test; ICS: inhaled corticosteroid; NA: not appropriate. #: sputum eosinophil and neutrophil percentage at baseline is reported (“baseline” is equal to enrolment if good quality data (squamous cell contamination <30%) is present at enrolment or the next (pre-exacerbation) stable visit with quality data within 4 months of enrolment); ¶: calculated as FEV1 at month 12×100/FEV1 at enrolment; +: calculated as (post-bronchodilator FEV1–pre-bronchodilator FEV1)/pre-bronchodilator FEV1×100; §: smoking status report derived from American Thoracic Society Division of Lung Disease questionnaire ATS-DLD-78A; ƒ: ICS use was coded as “Yes” if one of the following medications/inhalers was on the list: Symbicort, Seretide, QVAR, Fostair, beclomethasone, beclomethasone/formoterol, beclomethasone dipropionate, Clenil, fluticasone/salmeterol or budesonide/formoterol; ##: sputum sampling, measured by culture (includes Haemophilus influenzae, Moraxella catarrhalis, Streptococcus pneumoniae, Staphylococcus aureus and Pseudomonas aeruginosa); ¶¶: sputum sampling, measured by PCR (includes adenovirus, enterovirus, influenza, coronavirus, metapneumovirus, bocavirus, parainfluenza, respiratory syncytial virus and rhinovirus); ++: test for differences between the three longitudinal eosinophil groups (Kruskal–Wallis test for continuous variables; Chi-squared test for categorical variables).
- TABLE 3
Characteristics at first exacerbation overall# and by longitudinal blood eosinophil groups over 12 months¶
Overall cohort Rarely eosinophilic Intermittently eosinophilic Predominantly eosinophilic p-valueƒ Patients 108 26 13 48 Continuous variables White blood cells count ×109 L–1 8.00 (3.53) (n=102) 8.70 (4.25) (n=25) 8.95 (2.55) (n=12) 7.30 (3.23) (n=48) 0.349 Blood eosinophils count ×109 L–1 0.20 (0.20) (n=102) 0.10 (0.15) (n=25) 0.15 (0.10) (n=12) 0.30 (0.27) (n=48) <0.001 Blood eosinophils % 2.07 (2.78) (n=102) 1.19 (1.74) (n=25) 1.92 (0.88) (n=12) 3.39 (4.40) (n=48) <0.001 Blood neutrophils count ×109 L–1 5.30 (3.20) (n=102) 5.60 (4.05) (n=25) 5.55 (2.57) (n=12) 4.45 (2.98) (n=48) 0.045 Fibrinogen g·L–1 5.10 (1.40) (n=103) 4.80 (1.80) (n=25) 5.50 (2.25) (n=12) 5.10 (1.60) (n=46) 0.135 C-reactive protein mg·L–1 8.00 (14.0) (n=103) 7.00 (12.5) (n=25) 26.5 (49.8) (n=12) 7.00 (13.0) (n=47) 0.087 Sputum eosinophils % 1.20 (2.38) (n=61) 0.62 (2.29) (n=15) 1.40 (3.23) (n=9) 1.48 (3.56) (n=28) 0.273 Sputum neutrophils % 74.0 (47.1) (n=61) 76.8 (51.6) (n=15) 86.6 (40.3) (n=9) 74.1 (52.3) (n=28) 0.616 FEV1 % pred 42.7 (21.2) (n=97) 47.3 (30.0) (n=22) 41.5 (17.5) (n=12) 41.7 (20.0) (n=44) 0.395 CAT score 21.0 (11.0) (n=106) 22.0 (12.0) (n=26) 22.0 (17.0) (n=13) 19.0 (8.00) (n=47) 0.915 EXACT-PRO score 41.0 (11.0) (n=97) 43.0 (13.0) (n=23) 40.0 (10.0) (n=12) 39.5 (7.00) (n=44) 0.142 Categorical variables Bacteria present+ Yes 60 (61) 19 (86) 6 (55) 27 (57) 0.044 No 38 (39) 3 (14) 5 (45) 20 (43) Virus present§ Yes 40 (43) 12 (55) 4 (40) 14 (33) 0.235 No 53 (57) 10 (45) 6 (60) 28 (67) Sputum eosinophils >3% Yes 23 (30) 2 (11) 0 (0) 17 (47) 0.004 No 54 (70) 16 (89) 7 (100) 19 (53) Blood eosinophils ≥2% Yes 73 (68) 7 (28) 7 (54) 47 (98) <0.001 No 34 (32) 18 (72) 6 (46) 1 (2) Data are presented as n for patient numbers, median (interquartile range) for continuous variables or n (%) for categorical variables, unless otherwise stated. FEV1: forced expiratory volume in 1 s; CAT: COPD Assessment Test; #: n=108; ¶: n=87; +: sputum sampling, measured by culture sputum sampling, measured by culture (includes Haemophilus influenzae, Moraxella catarrhalis, Streptococcus pneumoniae, Staphylococcus aureus and Pseudomonas aeruginosa); §: sputum sampling, measured by PCR (includes adenovirus, enterovirus, influenza, coronavirus, metapneumovirus, bocavirus, parainfluenza, respiratory syncytial virus and rhinovirus); ƒ: test for differences between the three longitudinal eosinophil groups (Kruskal–Wallis test for continuous variables; Chi-squared test for categorical variables).
Supplementary Material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material ERJ-00853-2017_Supplement
Supplement to figure 4 ERJ-00853-2017_Figure_4
Supplementary Material
Author disclosures ERJ-00853-2017_Disclosures
