Study design After a 4-week washout period, during which patients were allowed to take only salbutamol (as needed), they were assigned according to a double-blind, double-dummy randomized crossover study to one of the two following 24-week treatments: 1) formoterol 12 µg plus beclomethasone HFA 200 µg (by MDI) b.i.d., or 2) salmeterol 50 µg plus fluticasone propionate 250 µg (diskus) b.i.d.. Daily doses of 400 µg beclomethasone HFA and 500 µg of fluticasone propionate are equivalent, according to the Global Initiative for Asthma (GINA) International guidelines6. At the end of first period and after another 4-week washout period, they received the second randomized 24-week treatment.
Adherence to the treatment was based on self-reports by each patient, on a declaration of willingness to continue to participate to the study and on the counter of drug doses disposed by inhalers. All patients have been trained to correctly inhale the drugs by different devices.
At the end of the washout periods and two treatments the subjects underwent Asthma Control Test (ACT), oxygen saturation (SaO2) measurement and Pulmonary Function Testing (PFT).
The subjects were enrolled after obtainment of written consent. The protocol has been approved by our local Ethic Committee.
Asthma Control Test Subjects completed an ACT questionnaire at the end of each washout and treatment period. Asthma control was assessed by the validated Italian version of the ACT. They subjectively evaluated the degree of impairment caused by their asthma during the preceding 4 weeks answering to five questions using a five-point-scale. The ACT is a reliable and validated tool, responsive to changes in asthma control over time. A cut-off score of 19 or less identifies subjects with poorly controlled asthma.
Pulmonary Function Testing A computerized water-sealed light-bell Stead-Wells spirometer (Biomedin, Padua, Italy) was used for measuring slow vital capacity (SVC), inspiratory capacity (IC), and maximal flow-volume curves. The operator was assisted during the test by software able to verify electronically both the acceptability and reproducibility of spirometric maneuvers. All the spirometric tests employed in the study fulfilled the recommendations of American Thoracic Society. Total lung capacity (TLC), functional residual capacity (FRC) and residual volume (RV) were obtained by using the plethysmographic method. The body plethysmograph was a variable flow, constant pressure-type with a frequency response flat until 30 Hz (Biomedin, Padua, Italy).
Determination of the slope of N2 phase III and closing volume (CV) was obtained during a single-breath nitrogen washout test (SBWN2)(Medical Graphics, St. Paul, MN, US). Subjects in sitting position after exhaling to residual volume (RV) inhaled slowly pure oxygen with a single breath until TLC and then expired slowly from TLC to RV. The expired nitrogen concentration was measured and plotted as function of expired lung volume. This relationship was fitted linearly in the phase III and the slope of N2 phase III was computed automatically by the software. The onset of phase IV was identified on the trace by software or by visual examination, if required. The average values of the N2 phase III slope and CV obtained from two acceptable and reproducible measurements (with expiratory slow vital capacity within 10%) were selected for analysis.
Expiratory maximal flow-volume curves breathing either air or heliox (oxygen 21% and helium 79%) were also performed. Breathing mixture was introduced in the bell of spirometer by a 3-way stopcock connected to the ambient air or to a cylinder containing humidified heliox at ambient temperature and pressure. In each occasion, after 3 slow deep breaths to total lung capacity, a forced expiratory maneuver was performed starting from TLC without end-inspiratory pause. Measurements of MEF50% with air and heliox and determination of their differences \[Delta(Heliox-air)MEF50%\] were obtained at baseline and subsequently at corresponding isovolume.
Statistical analysis All values are expressed as mean ± SD. We used the mean of the functional values of the two baselines for comparisons. Analysis of variance of the variables of interest among baseline and two therapeutic regimens (end of treatment periods) were performed by Friedman's test. Comparisons between groups were made according to Wilcoxon test, when allowed. A p value \<0.05 was considered significant. Statistical analysis was performed with GraphPad Prism Version 4 (GraphPad Software; San Diego, CA, USA).
