ABSTRACT
BACKGROUND: The existing research data are still not able to provide an answer to the issue of the correlation between dyspnea and inflammation in stable chronic obstructive pulmonary disease (COPD). The purpose of this study was to assess the possible relationship between a noninvasive medium called the exhaled breath condensate (EBC) and dyspnea in stable COPD patients. METHODS: A group of ten patients (five with and five without COPD) participated. The exhaled breath condensate was analyzed for the first time with a high-resolution device, the Accusizer 780SIS. The particle concentration of the EBC was measured and correlated with tools used for dyspnea assessment and clinical picture (modified Research Council dyspnea scale, mMRC scale; modified Borg dyspnea scale; and COPD Assessment Test, CAT scale). Because of the very small sample size (Ν = 10), bootstrapping method (applying 5000 bootstrap resamples with 95% confidence intervals) was used to derive robust estimates of standard errors and confidence intervals for estimates of means and correlation coefficients. Bootstrap works well in small sample sizes by ensuring the correctness of tests. RESULTS: The bootstrap means of EBC, mMRC, Borg, and CAT scales were 223863.43 (95% CI, 151308.58-297603.04), 1.30 (95% CI, 0.70-1.90), 1.55 (95% CI, 0.55-2.80), and 6.70 (95% CI, 4.80-8.60), respectively. The bootstrap Pearson's correlation coefficient (r) of EBC, mMRC, Borg, and CAT scales were 0.889 (95% CI, 0.716-0.979), 0.641 (95% CI, -0.542-0.887), and 0.569 (95% CI, -0.184-0.912), respectively. CONCLUSIONS: The effect size of the correlations is significantly high at the 0.01 level (two-tailed) between the EBC and the mMRC scale, less high at the 0.05 level (two-tailed) between the EBC and the Borg dyspnea scale and marginally with the CAT scale, respectively. Studies with larger samples will be needed to obtain more reliable results.
