Bronchodilator Response: Another Piece in the Asthma Mosaic

Considering that it is such a common disease, the diagnosis and severity classification of asthma is extraordinarily difficult. Gross put it well: “It’s like love, we all know what it is, but who would trust anybody else’s definition?”1 The National Asthma Education and Prevention Program (NAEPP) guidelines suggest the diagnosis of asthma should be in large part based upon the medical history and physical examination.2 However, these guidelines rightly go on to point out that “patients with asthma are heterogeneous and present signs and symptoms that vary widely from patient to patient as well as within each patient over time.” Subtle variations in the interpretation of an individual patient’s signs and symptoms may greatly influence not only whether the diagnosis of asthma is made, but also affect the aggressiveness of the ensuing treatment. Thus, we can think of asthma as a complex mosaic, with the history and exam as only two tiles in a much larger diagnostic picture.

In order to provide more objective diagnostic criteria, the NAEPP guidelines advise routine use of spirometry to aid in the diagnosis of asthma. Indeed, the three facets of history, exam and low baseline forced expiratory volume in 1 second (FEV1) remain the standard for diagnosing asthma in adults. However, in children there are convincing data showing that baseline FEV1 is not a good measure of the presence of asthma or its severity.

In the Childhood Asthma Management Program study, which evaluated 1041 children with mild to moderate asthma, more than 50% of the patients had moderate persistent asthma as defined by frequency of symptoms.3 Asthma was well documented in these patients over the 5-year life of the study, yet the prebronchodilator FEV1 at the start of the study was clearly normal at 94% of predicted.

A much quoted paper by Fuhlbrigge et al4 found those children with FEV1 values of <60% predicted had a 70% likelihood of having an asthma exacerbation in the following year. In those with FEV1 values >80%, the likelihood of experiencing an exacerbation was reduced to 25% to 30%. These data have been interpreted as showing that attacks can be predicted based on percent predicted FEV1. Yet it is perhaps more important to note that 94% of the FEV1 values in this population of asthmatic children were normal, meaning that 80% of asthma attacks occurred in children with a normal baseline FEV1. The limits of using FEV1 alone in the assessment of childhood asthma were also demonstrated by Bacharier et al,5 who found a lack of association among asthma symptom severity, intensity of medication therapy, and percent predicted FEV1 in asthmatic children.

Bronchial lability may be a more useful measure in the diagnosis of childhood asthma. Proving the existence of airway hyperreactivity in the context of clinical symptoms begins to add more details to our asthma mosaic. This is not a new concept. Exaggerated bronchodilation followed by bronchoconstriction in response to exercise—the so-called “bronchial lability index”—was described in asthmatic children by Jones.6 This and other early studies observed that pediatric asthmatic airways are remarkably labile in both directions when appropriately stimulated.

The study by Gallant et al7 published in this issue of The Journal shows that detecting broncholability by measuring the response to an inhaled bronchodilator can aid in the diagnosis of asthma in children. The authors demonstrate that using 9% as a distinct cutoff value for improvement in FEV1 after inhaled albuterol (either 180 μg by metered dose inhaler or 2.5 mg by nebulizer) can distinguish a group of known asthmatic children from those who are normal by history.

The findings of Gallant et al support the earlier study by Dundas et al8 that determined that a 9% cutoff for the bronchodilator response (BDR) to 400 μg of salbutamol (albuterol) provided the greatest balance between sensitivity and specificity in separating wheezers from nonwheezers in a group of London schoolchildren (race not described). However as stated by Dundas et al, the diagnostic value of a 9% BDR cutoff will vary with the prevalence of wheezing in the study population. Gallant et al studied a group of clinically diagnosed asthmatic children with a presumed incidence of wheezing of 100%. This exaggerated the difference between this study group and the comparator group to some degree. As a diagnostic test, BDR will be used in populations in which the incidence of wheezing may be much lower and the distinction between asthmatics and nonasthmatics is less clear. Without a prospective assessment of the 9% BDR cutoff value in an unselected cohort of subjects, the findings of Gallant et al still leave us several steps away from implementing BDR as a diagnostic test for asthma.

Although the ethnic composition of Gallant et al’s population is described as primarily Hispanic, the racial composition is not fully described. As recognized by the authors, extrapolating the 9% BDR cutoff to similarly aged African-American, Caucasian, or mixed populations is difficult, especially in light of the fact that different genetic groups respond to bronchodilator medications differently.9 The diagnostic BDR cutoff point certainly may be lower in children with less sensitivity to beta-agonist medications than the general population.

The effect of baseline lung function on BDR measurement also must be considered, as Gallant et al acknowledge. This relationship was described by Sly,10 who noted that the greatest percent increase in peak expiratory flow rate (PEFR) with treadmill exercise was seen in those asthmatic children with the lowest baseline PEFR. Where a child stands in relation to his or her maximum lung function on the day of testing will contribute to his or her ability to respond. A child already at his or her personal maximum for FEV1 would not exhibit a response to a bronchodilator even if he or she were asthmatic. In a disease as variable as asthma, this may prove a difficult hurdle to cross to use BDR as a diagnostic criterion.

The present study used 2 distinct methods of delivering the albuterol medication—some subjects used a metered dose inhaler, whereas others received the medication via wet nebulization. Although the quantitative difference in medication delivery between these 2 methods of medication administration may be small, the effect on BDR is not known. Using a single delivery method may have resulted in different outcomes. The dose and mode of delivery of beta-agonists are likely to play some role in the degree of observed bronchodilation and will need to be standardized to make this a clinically helpful test.

Gallant et al have presented a very sound idea for helping pediatric clinicians diagnose asthma. They have shown that BDR distinguishes between asthmatics and nonasthmatics better than baseline FEV1 alone, and that a combination of a high BDR and a low FEV1 is best for discriminating asthmatics from nonasthmatics (although these characteristics may be linked). Finally, and perhaps most importantly, they have demonstrated another way in which lung function testing can be helpful in the difficult process of diagnosing and managing asthma in children. Spirometric evaluation is relatively simple to perform in many preschool-age and nearly all school-age children.11 We believe that with further study, BDR testing will prove to be an important tool in our efforts to complete the asthma mosaic.