Children using a spacer device rather than another device for delivering inhaled corticosteroids (ICS) has been identified as a risk factor for cough immediately after inhalation. The aim of this study was to point out the different factors influencing the occurrence of such lateral side-effects. We studied this local side-effect in 402 asthmatic children (55.6 +/- 34.9 months; 65.6% boys) treated for at least 1 month with beclomethasone dipropionate (n = 331), budesonide (n = 47) or fluticasone propionate (n = 24) delivered from pressurized metered-dose inhalers and small (75.1%) or large volume (24.8%) spacer devices mainly used with face mask (90.7%). A total of 219 patients (54.5%), treated with either high doses of ICS or ICS and long-acting beta2-agonist, were considered as having severe asthma. Cough was reported after each inhalation of corticosteroids in 216 patients (53.7%). Among them, about 30% also complained of cough with beta2-agonists. Despite different propellants and dispersants, all corticosteroids induced cough similarly. Cough was not linked with asthma severity, but was significantly related to therapy duration and use of long-acting beta2-agonist. Type and volume of the spacer device, use of a face mask or mouthpiece were not influencing factors. Cough after inhalation of corticosteroids delivered from spacer devices is a frequent local side-effect in children with asthma. This side effect can greatly alter compliance. A practitioner must be sought at each visit.