BackgroundA positive bronchodilator response has been defined as a 12% increase in the forced expiratory volume in one second (FEV1) or forced vital capacity (FVC) from their respective pre-bronchodilator values, combined with at least a 0.2 L absolute change. Recent recommendations suggested the use of the percent change in FEV1 and FVC relative to their predicted normal values without having applied them in patients with airflow obstruction. The aim of the current study was to compare the two approaches over a wide range of pre-bronchodilator FEV1 and FVC values.MethodsA retrospective review of consecutive patients undergoing spirometry and bronchodilator testing was completed. The change in FEV1 and FVC with a bronchodilator was expressed relative to the pre-bronchodilator and predicted normal FEV1 and FVC.ResultsIn 1,040 patients with a non-paradoxical change in FEV1, 19.0% had a ≥ 12% change in FEV1 using their pre-bronchodilator value compared to 5.7% using their predicted normal value. For FVC, the respective values were 12.7% vs. 5.8%. The difference was retained in patients with a ≥ 0.2 L change in FEV1 or FVC. In unobstructed patients, the upper threshold (two standard deviations above the mean) of the bronchodilator response was 14% for FEV1 and 10% for FVC using predicted normal values.ConclusionsExpressing the percent change in FEV1 and FVC relative to predicted normal values reduces the over-estimation of the bronchodilator response, especially in patients with a very low pre-bronchodilator FEV1, including in those with a ≥ 0.2 L change in FEV1. Irrespective of pre-bronchodilator values, a ≥ 14% change in FEV1 and ≥ 10% change in FVC relative to the predicted normal values could be considered a positive bronchodilator response.