Do inhaled corticosteroids affect perception of dyspnea during bronchoconstriction in asthma?

BACKGROUND Some of the disagreements on the perception of dyspnea (PD) during bronchoconstriction in asthma patients could depend on the interrelationships among the following: (1) the influence of baseline airflow obstruction on the patient's ability to detect any further increase in airway resistance; (2) the effect of eosinophilic inflammation on the airway; (3) bronchial hyperresponsiveness (BHR); and (4) the effect of inhaled corticosteroids (ICSs). OBJECTIVE We hypothesized that if the inflammation of the airway wall influences to some extent and in some way the PD in asthma patients, ICSs reverse the effect of airway inflammation on the PD. METHODS We studied 100 asthma patients who were divided into the following four groups: patients with obstruction who were either ICS-naive (group I) or were treated with ICSs (group II); and nonobstructed patients who were either ICS-naive (group III) or were treated with ICSs (group IV). PD on the visual analog scale (VAS) was assessed during a methacholine-induced FEV(1) decrease and specifically was quantified as the VAS slope and score at an FEV(1) decrease of 5 to 20%. BHR was assessed in terms of the provocative concentration of methacholine causing a 20% fall in FEV(1) (PC(20)). Eosinophil counts in induced sputum samples also were performed. Regression analysis, univariate analysis of variance, and factor analysis were applied for statistical evaluation. RESULTS For a 5 to 20% fall in FEV(1) from the lowest point after saline solution induction, VAS score was lowest in group II, slightly higher in group I, slightly higher still in group IV, and the highest in group III. In the patients as a whole, BHR related to PD, but age, clinical score, duration of the disease, and presence of baseline airway obstruction did not. In patients with obstruction who were treated with ICSs, eosinophil counts related to PD negatively. Factor analysis yielded the following four factors that accounted for 70% of the variance in the data: ICS; eosinophil counts; FEV(1); and PC(20) loaded on separated factors with PD loading on the same factors as PC(20). The post hoc analysis carried out dividing the patients into ICS-treated and ICS-naive, showed that in the former group eosinophil counts and BHR proved to be factors negatively associated with PD, while in the latter group eosinophil counts were positively associated with PD. CONCLUSIONS We have shown that eosinophilic inflammation of the airway wall may increase PD and that the association of eosinophil counts with ICSs may result in lessening the PD.