In older children and adults, bronchodilator (BD) dose-effect relationship is part of the characteristics of asthma disease. There are no data on BD dose-response relationship in wheezy preschool children whose disease pathophysiology is poorly understood, but may, in part, takes on the characteristics of asthma.

The investigators assume that 1) in young children interrupter resistance (Rint) could be used to measure a BD effect 2) the response to BD may vary depending on the dose used 3) the dose-response relationship could depend on the environment and gene polymorphism ADBR2.

This is a prospective phase II study on dose-response relationship and description of the dose-response curve design using a "sparse" and a modeling approach MCP-Mod.

The dose-response relationship will be modeled by sparse data. The investigators will test two doses per child in four designs that will be drawn. These doses will be assessed using Rint technique by a person blinded to the actual dose delivered to the child.

Measurements of 90 children will estimate E0, Imax and D50 (pharmacokinetic constants) with an accuracy of 3.5%, 8.9% and 25.7% respectively.

The bronchodilator used in the study is the Salbutamol as Ventolin ® (GSK) suspension for inhalation as an aerosol at a dose of 100μg per puff. Ventolin ® is used as part of the MA (No. 344 387-3)

In older children and adults, bronchodilator (BD) dose-effect relationship is part of the characteristics of asthma disease. There are no data on BD dose-response relationship in wheezy preschool children whose disease pathophysiology is poorly understood, but may, in part, takes on the characteristics of asthma.

We assume that 1) in young children interrupter resistance (Rint) could be used to measure a BD effect 2) the response to BD may vary depending on the dose used 3) the dose-response relationship could depend on the environment and gene polymorphism ADBR2.

This is a prospective phase II study on dose-response relationship and description of the dose-response curve design using a "sparse" and a modeling approach MCP-Mod.

The dose-response relationship will be modeled by sparse data. We will test two doses per child in four designs that will be drawn. These doses will be assessed using Rint technique by a person blinded to the actual dose delivered to the child.

Measurements of 90 children will estimate E0, Imax and D50 (pharmacokinetic constants) with an accuracy of 3.5%, 8.9% and 25.7% respectively.

The bronchodilator used in the study is the Salbutamol as Ventolin ® (GSK) suspension for inhalation as an aerosol at a dose of 100μg per puff. Ventolin ® is used as part of the MA (No. 344 387-3) clinical implications The demonstration, first, of the possibility for Rint to detect a dose-response to BD will lead, secondly, to the determination of the minimum dose required for the detection of a reversibility in young children using Rint. It will end a long-standing debate about whether, when no Rint change is observed after BD administration in a young child, the child has actually no reversibility or the BD dose used was not sufficient to demonstrate one.

Moreover, the demonstration of a BD dose-response relationship in young children will suggest similarity between wheezy young children and older children and adults with asthma bronchial behaviour that may have possible therapeutic implications.

Inclusion Criteria:

• Any patient between 2 years 6 months and 6 years 11 months of age referred for pulmonary function testing with bronchodilator test, due to recurrent wheezing ,at least three times in the past year, to LFT laboratories at Armand Trousseau, Robert Debré, both in Paris and Arnaud de VILLENEUVE, in Montpellier hospitals
• No use of bronchodilator in the previous 12 hours before the test
• Parents gave their signed consent for the study

Exclusion Criteria:

Any patient aged 2 years 6 months and 6 years 11 months :

• with another chronic lung disease (bronchopulmonary dysplasia, chronic bronchitis, viral sequel, pathology of inhalation, thoracic-pulmonary malformation, tracheomalacia), anatomical or functional abnormality of the pharyngolaryngeal tract (tonsils touching or in contact with the uvula, laryngomalacia, subglottic stenosis, vocal cord paralysis, laryngeal obstruction)
• taking regular treatment including leukotrienes receptor antagonist during the week before the test.
• treated with oral glucocorticosteroids within 15 days before the test.
• without social security insurance
• with opposition of the family
• known intolerant to Salbutamol