Severe centrilobular emphysema in a patient without airflow obstruction.

We welcome the comments of Dr. Lipworth regarding the comparison between exhaled nitric oxide (NO) and methacholine reactivity in separating doses of inhaled steroid. Our study looked at the short-term effects of administering doses, increased weekly, of inhaled beclomethasone on exhaled NO and PC20 methacholine, in a small number of asthmatic subjects. First, there are important differences in the time courses of response of these two measures of asthma outcome. Exhaled NO is a rapidly responding marker, as seen in a previous study of ours where the maximal fall of NO after 1,000 g inhaled beclomethasone was complete after only one week.1 This was the basis for selecting periods of 1 week for each dose level in this study. Thus, it is difficult to compare exhaled NO with PC20 methacholine, where the maximal response takes several months. Second, we agree that, with dose intervals of 1 week, there may have been carry-over effects from the previous dose. We mentioned this in the “Discussion” section. Third, it is true that, in all subjects combined, the exhaled NO response plateaus at 400 g, while in the publications referred to by Dr. Lipworth, PC20 methacholine plateaus at 1,600 g. However, because the potential for a parameter to change depends on its absolute value, exhaled NO plateaus at 800 g/d in those subjects with very high baseline exhaled NO, as shown in Figure 1 of our manuscript. Our study is limited, however, by a small number of subjects. A dose-response relationship for an asthma outcome measure could lead to the ability to tailor the dose of inhaled steroids. Because exhaled NO reacts rapidly, this could be performed in a short period, whereas with PC20 methacholine, several months might be necessary. Before closing the chapter comparing dose responses between exhaled NO and PC20, there is a need to perform further studies of inhaled steroid dose-response in a larger number of subjects, looking at exhaled NO and PC20 methacholine over longer periods of time, while controlling for baseline values as covariates.