Technology - results

Turbohaler® is a reservoir dry powder inhaler with dose metering. The Pulmicort® Turbohaler® device contains 100 actuations of pure budesonide, housed in a small reservoir at the base of the inhaler. Each actuation is nominated to contain 200µg of budesonide.

In this particular experiment, a total of 77 doses were analysed by VariDose. The frequency distribution of the dose value for a single dose emitted from Turbohaler is in good agreement with the data (Meakin et. al., 1995), obtained by means of a twin-stage impinger.

A simultaneous analysis of the delivered dose using light obscuration (VariDose) and a twin-stage impinger (TSI) was performed. The VariDose device was placed between the mouthpiece of the inhaler and the entrance of the TSI.

The Bricanyl® Turbohaler®, containing 100 actuations of pure terbutaline sulphate, with each actuation nominated to contain 500µg of terbutaline sulphate, was used as the test inhaler. 20 individual actuations were simultaneously measured by TSI and VariDose.

The correlation coefficient between TSI data and VariDose data for the fine particle dose (FPD) was 0.94 (p<0.001). The coefficient variation for both methods is similar: 45% for the TSI measurements and 48% for VariDose. The observed large variation is consistent with typical dose variations delivered by the Turbohaler device.

A comparative evaluation between VariDose and the next generation pharmaceutical impactor (NGI) was performed. The experimental set-up includes an NGI impactor, VariDose and a Vectura prototype active inhaler device.

The dose actuation is facilitated by a vacuum pump at the flow rate of 60L/min. For this study, a drug formulation, which includes the micronised salbutamol sulphate (40%) and Respitose (60%), was used. Overall, there were 30 measurements by VariDose and 6 data sets of NGI measurements.

The correlation between the sets of data for FPD is statistically significant, with the correlation coefficient of 0.87 (p<0.023). The figure shows a possible calibration of VariDose measurements for the given experimental set-up.

Note that mean values of VariDose data are also characterised with variance (4% to 10%), which reflects the variation of the doses for a single actuation in each run of 5 shots. Unfortunately, NGI provides only average values for each run, thus no information for variation of single shots is available.

The best fit (solid) line, calculated using the least squares method is not crossing the origin. According to the calibration, the dose variation is within the dose range measured by NGI (dotted lines). This fact confirms that VariDose measurements may reliably characterise the FPD for a single actuation, although more experimental data would benefit the calibration precision.