The review by Brusasco et al 1 demonstrates that a simple causal hypothesis is insufficient to explain airway hyperresponsiveness in asthma. We have also refuted this hypothesis by identifying that airway eosinophilia can occur without airway hyperresponsiveness,2 and that in asthma airway hyperresponsiveness can persist despite suppresssion of airway inflammation and pancytopenia induced by chemotherapy.3 However, airway inflammation and airway hyperresponsiveness are so intrinsically linked in asthma that an alternative hypothesis must be provided.

My hypothesis is that airway inflammation can modulate the level of intrinsic airway responsiveness such that an increase in airway inflammation from, for example, allergen exposure will lead to an increase in airway responsiveness, and a decrease in airway inflammation from, for example, corticosteroid therapy will lead to a reduction in airway responsiveness. The absolute level of airway responsiveness, and hence whether it is categorised as normal or increased, will depend upon two factors—the starting level of airway responsiveness and the magnitude of change.

This hypothesis is consistent with the observed data. Exposure to respiratory sensitisers induces airway inflammation and increases the level of airway responsiveness. This occurs both in the “asthmatic range” and in the normal range of airway responsiveness (fig 1). Similarly, corticosteroid therapy reduces airway eosinophilia and this leads to an improvement in airway responsiveness, both in the normal range2 and in the asthmatic range (fig 1). It is interesting to note that the magnitude of change in airway responsiveness induced by changes in airway inflammation is similar whether it occurs within the asthmatic or the normal range.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Modulation of airway inflammation is associated with changes in airway responsiveness both within the asthmatic range (below the dotted line) and beyond this into the non-asthmatic range. Closed symbols are individual patient data, open symbols are group mean data. (A) Improvement in airway responsiveness following inhaled corticosteroid therapy in subjects with asthma and chronic cough; (B) deterioration in airway responsiveness following allergen challenge (closed triangles), seasonal allergen exposure (diamonds), or exposure to an occupational sensitiser (closed circles). Reproduced from Gibson4 with permission.

The mechanisms for the intrinsic airway hyperresponsiveness in asthma will differ from those resulting in a change in airway responsiveness induced by airway inflammation.

The consequences of airway remodelling such as airway wall thickening and alterations in smooth muscle mass and contractility may each be determinants of the underlying level of airway responsiveness, independent of any airway inflammation.

From a clinical perspective it is likely that airway inflammation will be suppressed by (low dose) corticosteroid therapy but airway hyperresponsiveness will persist. The optimal treatment in this circumstance remains to be defined, but the persistence of airway hyperresponsiveness in the absence of airway inflammation may explain the unanticipated benefit from the addition of long acting β₂ agonists to patients who remain symptomatic on inhaled steroids. It will be important to establish the relationship between changes in airway hyperresponsiveness, airway inflammation, and asthma symptoms and to use this information to assist in the rational choice of treatment in asthma.