This was an observational, cross-sectional study to evaluate the differences in the production of CANO and JawNO and the relationship of these variables with clinical variables in a consecutive series of COPD patients who performed complete lung function testing (measurement of lung volumes and diffusion) in our respiratory outpatient clinic. The study was performed between November 2014 and May 2015.

The study population comprised adult patients over 40 years of age, smokers, or non-smokers with an accumulated pack-year index of at least 10, and a diagnosis of COPD according to national and international guidelines and recommendations.1 Exclusion criteria were the presence of any respiratory disease other than COPD that might significantly affect the examination (including a history of bronchial asthma), a history of COPD exacerbation in the 4 weeks before the test, inability to perform the study procedures or complete the questionnaires, and participation in any other clinical trial or research study.

Clinical variables: for each patient, data were collected on their respiratory disease, including time since onset, toxic habits, comorbidities, baseline dyspnea measured according to the modified Medical Research Council (mMRC) scale, COPD Assessment Test (CAT®) questionnaire, and history of exacerbations in the previous year (classified as moderate if treated with systemic corticosteroids and/or antibiotics in an outpatient setting, and severe in the case of admission > 24h to a hospital or emergency department).

Blood tests: before lung function tests were performed, peripheral blood was obtained for the determination of absolute eosinophil counts and percentages, and for the measurement of C-reactive protein (CRP).

Measurement of the alveolar (CANO) and bronchial (JawNO) component of nitric oxide in exhaled air: before lung function tests were performed, patients performed 3 FENO maneuvers at 50mL/s (FENO50) followed by additional determinations at 100mL/s, 200mL/s, and 350mL/s (at least 2 each) in order to obtain CANO and JawNO levels, according to international guidelines.23,24 An NO chemiluminescence analyzer (HypAir FENO®, Medisoft, Belgium) was used.

Lung function variables: patients performed spirometry at baseline and after inhaling salbutamol 400μg in accordance with national and international guidelines.25,26 Lung volumes and diffusing capacity of the lung for carbon monoxide were also determined according to applicable recommendations.27,28 All measurements were performed on the same lung function testing equipment (MasterLab, Jaeger GmbH, Germany).

The study was conducted in compliance with the principles of the Declaration of Helsinki. Patients who were invited to participate signed informed consent forms. The study was approved by the reference clinical research ethics committee. All patient records generated in the database were confidential and handled in compliance with the Personal Data Protection Law 15/1999 of 13 December 1999.

A descriptive statistical analysis was made of the study variables, using absolute and relative frequencies in the case of qualitative variables. Depending on whether distribution was normal or non-normal (applying the appropriate normality test for the sample size), quantitative variables were summarized by mean, standard deviation (Md ± SD), and range (minimum and maximum) or P50 (P25 - P75] (median, interquartile range), respectively.

Quantitative variables were compared by study groups using the Student's t-test for independent samples or the Kruskal–Wallis H (the appropriate test was selected from previously obtained results). Post hoc tests were conducted in the case of ANOVA; otherwise, the Mann–Whitney U-test with the Bonferroni correction was used. To determine the relationship between CANO, JawNO, and the clinical variables, a multiple linear regression analysis was performed with CANO and JawNO as dependent variables, respectively, using the forward stepwise method. The level of statistical significance was set at a p-value of 0.05. The statistical analysis was performed using the Statistical Package for Social Sciences (SPSS Inc., Chicago, IL, USA), version 24.0.

Ninety-five patients were invited to participate, 73 of whom finally completed all procedures. Fig. 1 is a STROBE diagram of the study participants.

Figure 1.

STROBE diagram of study participants.

(0.11MB).

Baseline patient characteristics are shown in Table 1. In summary, this is a series of patients with predominantly severe COPD, mean FEV1 < 50%, and a high burden of symptoms and exacerbations in the previous year. About 25% of the study population were active smokers.

Table 2 shows patient characteristics according to the eosinophil count criteria for defining ACO (≥2% eosinophils, ≥3% eosinophils, or ≥300cells/μL). Clinical parameters did not differ among the groups according to the criterion for defining ACO, except for nitric oxide measurements (FENO50, CANO, and JawNO). Figs. 2 and 3 show CANO and JawNO levels, respectively, using different eosinophil cut-off points for the definition of ACO (2%, 3%, and > 300 eosinophils, respectively). In all cases, CANO levels were higher for patients defined as having ACO according to the eosinophil count, while differences were greater when the criteria of >3% and >300 eosinophils were used. JawNO levels were higher only in those patients with ACO defined by >3% or >300 eosinophils. The use of more restrictive criteria (>300 eosinophils and >3%) did not improve the detection of differences between patients classified as ACO and other patients.

Figure 2.

CANO values by ACO criterion. (a) ACO defined by ≥300 eosinophils/μL. (b) ACO defined by ≥2% eosinophils. (c) ACO defined by ≥3% eosinophils. Horizontal lines show the mean value, figures show the individual values, the box shows the IQ range, and the whiskers show the mean standard error.

(0.13MB).

Figure 3.

JawNO by ACO criterion. (a) ACO defined by ≥300 eosinophils/μL. (b) ACO defined by ≥2% eosinophils. (c) ACO defined by ≥3% eosinophils.

(0.13MB).

Multiple linear regression analyses aimed at evaluating the associations between CANO and JawNO and different clinical, biological, and lung function variables are shown in Tables 3 and 4, respectively. Variables associated with alveolar inflammation (CANO), and which predicted its finding, were CRP and total eosinophil count in peripheral blood, with an R2 of 0.412. Variables that showed the closest association with bronchial inflammation (JawNO), and which predicted its finding, were smoking, number of moderate exacerbations in the previous year, and eosinophil count in peripheral blood, with an R2 of 0.477. Appendix B (additional material) depicts CANO and JawNO values compared with the predicted values according to these models (expressed as regressions of the standardized predicted value).