Background: There is currently disagreement on the way of expressing the reversibility of airflow obstruction, with some evaluations based on the initial FEV₁ while others use predicted FEV₁ (according to age, gender and height). Objectives: To test the relevance of expressing bronchodilator response as a percentage of predicted FEV₁, we evaluated the influence of morphological data on the response to bronchodilators in a population with a large range of predicted values. Methods: We measured the change in FEV₁ after inhalation of 200 µg of salbutamol in 30 asthmatic subjects (15 adults and 15 children) in whom predicted FEV₁ ranged between 1.13 and 4.10 liters and analyzed the respective influence of initial and predicted FEV₁ on bronchodilator response. Results: We have shown a significant relationship between the absolute variation in FEV₁, in liters, and predicted FEV₁ (p = 0.0019). There was also a significant relationship between the absolute variation in FEV₁ and initial FEV₁, in liters (p = 0.02). This relationship was no longer significant (p = 0.8) when the variation and initial FEV₁ were both expressed as percentages of predicted FEV₁. In addition, multiple regression analysis showed that predicted FEV₁ was the only independent variable correlating with the response to bronchodilators. Conclusion: This study showed the influence of predicted FEV₁ on bronchodilator response. This result provides an additional argument for expressing bronchodilator response as a percentage of predicted FEV₁.