Spirometric criteria for asthma: adding further evidence to the debate.

BACKGROUND: Objective assessments of pulmonary function are considered essential for the diagnosis of asthma. The degree of reversibility of FEV(1) considered supportive of asthma varies between international asthma guidelines.

OBJECTIVE: We sought to compare the relative performance of international guideline reversibility criteria for identifying impairment in persons with a significant bronchodilator response (SBR) without an asthma diagnosis.

METHODS: The North West Adelaide Health (Cohort) Study, a population biomedical study of 4060 subjects, conducted spirometry according to American Thoracic Society criteria. SBR was defined as postbronchodilator FEV(1) responses of at least 12% or 15% of baseline values, 9% of predicted values, or 400 mL. A self-completed questionnaire assessed current asthma (CA), respiratory symptoms, and participant demographics.

RESULTS: The prevalence of CA was 9.4% (n = 380), whereas 1.3% (>/=400 mL) to 4.5% (>/=9% of predicted value) of participants demonstrated an SBR in the absence of CA. With the exception of the 9% predicted criterion, prebronchodilator mean FEV(1) (percent predicted) in those demonstrating an SBR but no CA was significantly worse than that in the CA group. Significantly more respiratory symptoms were experienced by the SBR groups than the group without asthma. Logistic regression analyses identified different characteristics of those classified by the following criteria: 12% and 15%, age of 40 years or greater and household income of less than $40,000; 9% predicted, household income of less than $40,000; 400 mL, male sex (odds ratio, 4.5; 95% CI, 2.1-9.3).

CONCLUSIONS: Different criteria identify different persons, but SBR by any criteria was associated with significant respiratory impairment, some of which might be attributable to asthma. Postbronchodilator change as a percentage of predicted value was the least biased of the criteria.