Statistical Analysis
In Study 1, assuming a two-sided α of 0.05 and β of 0.20, we estimated at least 21 cases per each group, based on the difference of 2.12 mm2 and a standard deviation of 2.39 mm2 of Ai between asthmatic patients and control subjects16. In Study 2, based on the difference in Ai before and after inhalation of bronchodilator17, assuming a difference of 54.9% and a standard deviation of 22.5% for airway volume, with a two-sided significance level of 5% and a power of 80%, a total sample size of 8 cases would be required.
Statistical analysis was performed using SPSS version 22.0 (SPSS Inc., Chicago, USA) and GraphPad Prism 5.0 (GraphPad Inc., USA). Data were expressed as mean ± standard deviation. In Study 1, the differences in Ai/BSA and Aw% between the RB8 and RB9 segment were determined by paired t-tests. Paired t-test was applied to compare the spirometric and EB-OCT data before and after salbutamol inhalation. Differences among groups were analyzed using one-way analysis-of-variance or Kruskal-Wallis test where appropriate. Bonferroni correction was applied to adjust for multiple comparisons. The association between EB-OCT and spirometric parameters, and sputum cell counts were determined with Pearson’s or Spearman’s correlation model when appropriate. P<0.05 was deemed statistically significant unless otherwise stated.