aaDry powder inhalers (DPI) are increasingly used for aerosol delivery to the airways for the reasons that they are easy-to-use with no requirements of co-ordination, they are convenient and are without propellants or irritants. In addition, an ideal device should provide a high and predictable airway dose and a low pharyngeal dose. In all DPIs the inspiratory airflow of a patient provides the energy source to disperse the agglomerates of micronized powder and to move the respirable particles from the body of the inhaler to the lung. The drug is drawn from a holding bin of pure drug, or is available as drug particles mixed in a bulking agent such as lactose, which helps dispersion and improves the dose-to-dose reproducibility. It is a fundamental prerequisite for use that the patient can generate enough airflow energy through the device to achieve sufficient dispersion of the drug agglomerates. This interaction between the inhalation effort provided and the aerosolizing property is unique to the different designs of DPIs. The present study was undertaken to examine the likely lung dose, its reproducibility, and the effort dependency in children using two common DPIs: 1) the Turbuhaler is a widely used DPI which is easy-to-use and provides a high lung dose [1,2], yet apparently with flow dependent drug delivery [3][4][5]; and 2) the Diskus inhaler is the most recent addition to the range of multidose powder inhalers available for the delivery of β 2 -agonists or corticosteroids. It contains 60 factory-dispensed drug doses in a lactose carrier. The Diskus was designed as a low resistance device similar in aerosolizing property to the Diskhaler [6].We have studied the amount of aerosol obtained from the Diskus and Turbuhaler in asthmatic children of 4 and 8 yrs. The Turbuhaler is in most countries approved for use in children of >5 yrs, while the Diskus is approved for use down to 3 yrs of age. We therefore chose to consider children of 4 yrs as the lower limit for comparison, as this 1 yr difference to the recommended lower age is unlikely to affect the expected flow performance of the children. Our approach has been to record inhalation profiles from asthmatic children. We have used these profiles in an inhalation simulator [7] ( fig. 1) to draw doses from both the Diskus and Turbuhaler inhalers and subsequently to assess the particle size distribution of the drug using a cascade impactor. This method has enabled us to measure the emitted dose and the fractions of coarse and fine particles of drug that each child would be capable of drawing from these devices.
Methods
PatientsTwenty children aged 4 yrs and 18 children aged 8 yrs with asymptomatic asthma were recruited to the study. Informed consent was obtained from parents or guardians Full profiles of inhalation pressure versus time were recorded in 18 4 yr old and 18 8 yr old asthmatic children through Diskus and Turbuhaler inhalers. These data were used in an inhalation profile simulator to assess drug delivery from both a Diskus inhaler and a Turbuhaler inha...