Aerodynamic characteristics of a dry powder inhaler at low inhalation flows using a mixing inlet with an Andersen Cascade Impactor

Aerodynamic characteristics of a dry powder inhaler at low inhalation flows using a mixing inlet with an Andersen Cascade Impactor

European Journal of Pharmaceutical Sciences 39 (2010) 348–354 Contents lists available at ScienceDirect European Journal of Pharmaceutical Sciences ...

619KB Sizes 0 Downloads 26 Views

European Journal of Pharmaceutical Sciences 39 (2010) 348–354

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps

Aerodynamic characteristics of a dry powder inhaler at low inhalation flows using a mixing inlet with an Andersen Cascade Impactor夽 Dinesh Kumar Nadarassan a , Khaled H. Assi a , Henry Chrystyn b,∗ a b

Institute of Pharmaceutical Innovation, University of Bradford, Bradford BD7 1DP, UK Division of Pharmacy and Pharmaceutical Sciences, School of Applied Sciences, University of Huddersfield, Huddersfield HD1 3DH, UK

a r t i c l e

i n f o

Article history: Received 23 July 2009 Received in revised form 5 January 2010 Accepted 9 January 2010 Available online 20 January 2010 Keywords: Low inhalation flow Mixing inlet Dose emission Dry powder inhaler

a b s t r a c t The aerodynamic characteristics of the dose emitted from a dry powder inhaler (DPI) are inhalation flow dependent but have not been determined for low flows. We have designed novel methodology to measure these at <28.3 l min−1 . The original Andersen Cascade Impactor (ACI) designed for use at 60 l min−1 was adapted to include a mixing inlet (MIXINLET) which allows inhalation flows through the DPI from 5 to 60 l min−1 . The mean fine particle dose (FPD) from a formoterol Turbuhaler using the MIXINLET method at 10, 20, 28.3, 40 and 60 l min−1 was 0.55, 1.39, 1.80, 2.88 and 5.86 ␮g and the mass median aerodynamic diameter (MMAD) was 6.6, 6.0, 5.4, 5.1 and 2.8 ␮m. Similarly, the FPD using the ACI method was 0.13, 0.69, 1.50, 2.48 and 5.42 ␮g and MMADs were 12.2, 7.4, 5.5, 4.8 and 2.7 ␮m. The accuracy of the original ACI <28.3 l min−1 is unknown. The ACI with the mixing inlet allows the determination of the in vitro dose emission properties of DPIs at flows <28.3 l min−1 whilst maintaining a constant flow through the ACI. This methodology, therefore, can help to focus attention to the lowest inhalation flow required for a DPI. © 2010 Elsevier B.V. All rights reserved.

1. Introduction The aerodynamic characteristics of the dose emitted from a dry powder inhaler (DPI) are determined by the degree of deaggregation of the metered dose inside the inhaler during the inhalation manoeuvre. Turbulent energy occurs inside a DPI by the interaction between the inhalation flow and the inhaler’s resistance (Clark and Hollingworth, 1993). This leads to flow dependent dose emission that has been reported to be a property of all passive DPIs with some more prone to this phenomenon than others (Hindle and Byron, 1995; Ross and Schultz, 1996; Hill and Slater, 1998; Palander et al., 2000). When using a DPI, Everard et al. (1997) have shown that a failure to inhale deeply and forcibly at the start of the inhalation means that the drug particles generated are not sufficiently de-aggregated to enter the lungs and are simply deposited in the mouth and oropharynx where they have no clinical efficacy. Standard pharmacopoeia methodologies for the in vitro dose emission properties from a DPI use an inhalation flow corresponding to a pressure differential of 4 kPa across the inhaler and an inhalation volume of 4 l (BP, 2005; EP, 2005; USP, 2005). Many patients

夽 Work carried out at the Institute of Pharmaceutical Innovation, University of Bradford, Bradford BD7 1DP, UK. ∗ Corresponding author. Tel.: +44 0 1484 472178; fax: +44 0 1484 472182. E-mail addresses: [email protected] (D.K. Nadarassan), [email protected] (K.H. Assi), [email protected] (H. Chrystyn). 0928-0987/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ejps.2010.01.002

are not able to achieve the high inhalation flows required to create a pressure drop of 4 kPa (Pedersen et al., 1990; Broeders et al., 2003; Al-Showair et al., 2007). During an acute exacerbation inhalation flows are reduced (Pedesen, 1986; Broeders et al., 2004) so the emitted dose will be reduced at a time when the patient requires extra relief from their medication. The Anderson Cascade Impactor (ACI) is usually the method of choice to characterise the quality of the dose emitted from an inhaler. The ACI has been designed to operate at an inhalation flow of 28.3 l min−1 with recent modifications to allow determinations at 60 and 90 l min−1 . However since DPIs have a different resistance depending on their design (Clark and Hollingworth, 1993) then inhalation flows required to achieve the compendial recommended pressure difference of 4 kPa will vary. When using non-standard flows with the ACI the cut-off diameters for the stages have to be recalculated (Oort, 1995; BP, 2005; EP, 2005; USP, 2005). This has not been applied to flows <28.3 l min−1 with the ACI due to the decrease in the efficiency of impaction. Hence, reports of the aerodynamic characteristics of DPIs below inhalation flows of 28.3 l min−1 are not available. However identification of the dose emission characteristics of the emitted dose at low flows will be more clinically relevant than identifying these properties using optimal conditions as recommended by the Pharmacopoeias. At low flows if the de-aggregation is not efficient and the particles that are emitted have no or little potential for lung deposition then it is most likely that the patient will receive no therapeutic benefit even though they have performed the best inhalation manoeuvre that

D.K. Nadarassan et al. / European Journal of Pharmaceutical Sciences 39 (2010) 348–354

349

Fig. 3. Andersen Cascade Impactor methodology incorporating the mixing inlet (MIXINLET method). Fig. 1. The mixing inlet (Copley Scientific, UK; reproduced with permission).

they are capable of doing. This information for each DPI needs to be determined. To be able to measure dose emission properties from DPIs at low inhalation flows we have adapted the ACI methodology by using a mixing inlet. The mixing inlet is shown in Fig. 1 with a cross section in Fig. 2. The central tube has the same internal dimensions as the induction port and is surrounded by a sheath into which the supplementary air is introduced. The thickness of the central tube tapers towards the bottom such that it ends as a knife sharp edge. This together with the internal shallow angles ensures minimal turbulence where the two flows meet.

We have used the ACI designed to be operated at 60 l min−1 with supplementary air introduced via the side arm of the mixing inlet (see Section 2 for full details). The difference between these is the inhalation flow drawn through the DPI. Using this method we have measured the total drug output and particle size distribution of formoterol from an Oxis Turbuhaler (AstraZeneca, UK) at varying flows (in particular below 30 l min−1 ). To highlight the validity of our method we have compared these results to those measured using the standard ACI methodology. We have also measured pressure changes at different flows using the standard ACI methodology and our adaptation with the mixing inlet to provide further verification. 2. Materials and method 2.1. Anderson Cascade Impactor with the mixing inlet valve (MIXINLET method)

Fig. 2. Cross section of the mixing inlet.

The ACI designed to be used at 60 l min−1 was used so stages 0 and 7 were replaced by −1 and −0 (Copley Scientific, UK). The collection plates were sprayed with silicone lubricant (Pro-Power Silicone Lubricant, Premier Farnell plc, UK) and dried for 30 min prior to analysis. The Anderson Cascade Impactor (ACI) stages were assembled and a GF/A filter (Whatman plc, UK) was placed in the final stage of the ACI impactor. The pre-separator was filled with 10 ml of 60% methanol (in water). Pharmacopoeias (BP, 2005; EP, 2005; USP, 2005) recommend the use of the pre-separator for dry powder inhalers (DPIs) to entrain large particles usually >10 ␮m (Mitchell and Nagel, 2003). The mixing inlet (Copley Scientific, UK) was placed between the induction port and the pre-separator as illustrated in Fig. 3. This figure shows how a flow of 20 l min−1 was drawn through the Turbuhaler with the ACI operated at 60 and 40 l min−1 of supplementary air provided into the mixing inlet through its side arm. The ACI was connected to a Critical Flow Controller Model TK2000 (Copley Scientific, UK) to ensure sonic flow and provide the required inhalation flow and volume. The vacuum flow was provided by a HCP5 High Capacity Pump (Copley Scientific, UK). This will be referred to as the MIXINLET method. We

350

D.K. Nadarassan et al. / European Journal of Pharmaceutical Sciences 39 (2010) 348–354

12 ␮g formoterol fumarate per dose and nominal emitted of 9 ␮g; AstraZeneca, UK) were determined at inhalation flows of 10, 20, 28.3, 40 and 60 l min−1 all with an inhalation volume of 4 l. The respective supplementary flow introduced through the side arm of the mixing inlet was, therefore, 50, 40, 31.7, 20 and 0 l min−1 . For each determination 10 consecutive doses were discharged into the ACI and each dose was prepared according to the manufacturer’s recommended patient instructions. Ten separate doses were used to help with assay sensitivity and overcome the variability of dose emission from the Turbuhaler (Tarsin et al., 2004). The mean of 10 doses, therefore, limits any influence from erratic dose emission. Three separate determinations were made for each flow. A total of 4 inhalers were used and the order of determinations was randomised. ACI determinations using the STANDARD method were made at inhalation flows of 10, 20, 28.3, 40 and 60 l min−1 and an inhalation volume of 4 l. The original version of the ACI was used for all the determinations except for those at 60 l min−1 were stages 0 and 7 were replaced by −1 and −0. For 10, 20 and 40 l min−1 the cut-off diameters were adjusted for inhalation flows of 10, 20 and 40 l min−1 (Oort, 1995). Again three determinations were made using 10 consecutive doses and the order was randomised. 2.4. Quantification of formoterol Fig. 4. Schematic design of the adapted MIXINLET method to measure the pressure changes in the Turbuhaler at different inhalation flows.

have also used the ACI operated alone, without the mixing inlet, as described in the compendial methodologies (BP, 2005; EP, 2005; USP, 2005) and so this will be referred to as the STANDARD method. 2.2. Measurement of pressure changes in the Turbuhaler at different flows The MIXINLET, described in Fig. 3, and the STANDARD methods were both modified, using an adaptation of methodology described by Clark and Hollingworth (1993) to measure the pressure drops across inhalers at different flows. The modification to the MIXINLET method is described in Fig. 4. The standard mouthpiece adapter for the ACI method was replaced by a specially made adapter to allow an outlet to be connected to the MKS Baraton Type 223B pressure transducer (MKS Instruments, GmbH). The Turbuhaler was encased in a specially made chamber which also had a connection to the pressure transducer. Hence the absolute pressure changes in the Turbuhaler could be measured at different inhalation flows. Leak testing to ensure airtight seals was carried out. The inhaler chamber was attached to a MKS Type 1500 Mass-Flow Meter (MKS Instruments, GmbH) to measure the mass flow applied through the inhaler. The ACI was operated at 60 l min−1 with supplementary air introduced, via the side arm of the mixing inlet, at flows of 55, 50, 45, 40, 30, 20, 10 and 0 l min−1 . Hence the respective airflow drawn through the Turbuhaler was 5, 10, 15, 20, 30, 40, 50 and 60 l min−1 . Three determinations of the pressure drop (P) were made for each setting. These measurements were also made without the mixing inlet – this was the STANDARD method with the adaptation to measure the pressure changes. Inhalation flow through the ACI and Turbuhaler was controlled by the Critical Flow Controller. Three measurements were made at the same flows as the MIXINLET method. 2.3. Determination of the aerodynamic characteristics of the emitted dose Using the MIXINLET method the aerodynamic characteristics of formoterol from an Oxis Turbuhaler (nominal labelled dose of

After each determination all the plates and stages were washed with 60% methanol (in water). Similarly the induction port, mixing inlet and pre-separator were also washed. The amount of formoterol fumarate from these washings was determined using high performance liquid chromatography (HPLC). The chromatographic separations were carried out at 25 ◦ C on a C18 Spherisorb® (250 mm × 4.6 mm × 5 ␮m) analytical column (Waters, USA). The analytical column was protected with a C18 (4 mm × 3 mm i.d.) security cartridge system (Phenomenex, USA). The mobile phase was acetonitrile:5 mM disodium hydrogen orthophosphate buffer (70:30 v/v) adjusted to pH 3 with orthophosphoric acid. Before use the mobile phase was filtered through a membrane filter (47 mm diameter, pore size 0.25 ␮m) and sonicated under vacuum for 10 min. The mobile phase was delivered at a flow of 1.0 ml min−1 and the injection volume was 100 ␮l. UV detection at 214 nm was used. The calibration curves were linear (r2 = 0.99) over formoterol concentrations ranging from 50 to 500 ng ml−1 (n = 6). The method had an accuracy of >99% and intra- and inter-day precision CV of <1.9% and <1.2%, respectively, at three different concentrations (50, 300, 500 ng ml−1 ). The limits of detection (LOD) and quantitation (LOQ) of formoterol were 15 and 50 ng ml−1 , respectively. 2.5. Data analysis The total emitted dose (TED) for the MIXINLET method was the amounts deposited in the induction port (USP Throat), the mixing inlet, the pre-separator and all the stages of the ACI. The TED from the STANDARD method was the same except that the mixing inlet was not used. The TED has also been expressed as a percentage of the nominal (labelled) dose. For the MIXINLET method the combined amounts deposited in the induction port, mixing inlet and pre-separator were calculated – emitted dose pre-impactor (EDPI). For the STANDARD method the EDPI amounts were those deposited in the induction port and the pre-separator. The Copley Inhaler Testing Data Analysis Software (CITDAS version 2.0) was used to calculate the aerodynamic dose emission parameters. For each determination the software confirmed that the spread of each aerodynamic particle size distribution was unimodal and log normal such that the parameters could be calculated.

D.K. Nadarassan et al. / European Journal of Pharmaceutical Sciences 39 (2010) 348–354

351

Table 1 The mean (S.D.) pressure change (P) in the Turbuhaler at different flows using the MIXINLET and STANDARD methods (n = 3). Flow through Turbuhaler (l min−1 )

MIXINLET

STANDARD −1

Flow through mixing valve (l min 5 10 15 20 30 40 50 60

)

55 50 45 40 30 20 10 0

For each determination a plot of the logarithm of the percentage less than a stated size on a probability scale against the logarithm of the effective cut-off diameter of the stage was made according to that recommended in the Pharmacopoeias. From this plot the fine particle dose (FPD) was the amount calculated from the cumulative percentage drug mass associated with particles <5 ␮m. The fine particle fraction (FPF) was the FPD divided by the nominal (labelled) dose and expressed in %. The mass median aerodynamic diameter (MMAD) was the size corresponding to the 50th percentile of the cumulative mass-weighted distribution. MMAD values were determined by interpolating the data points closest to the 50th mass percentile. Since the bulk of the aerosol was present at this region of the distribution, then the estimates of the FPD and MMAD are as precise as possible for this method. The geometric standard deviation (GSD) was determined as the square root of the ratio of the 84.1 to 15.9 mass percentiles of the aerodynamic particle size distribution (Hinds, 1999). Statistical analysis to compare the two methods was made using a two-way ANOVA through an adaptation of the GLM Multivariate Option in SPSS V16.

P Turbuhaler (kPa)

P Turbuhaler (kPa)

0.19 (0.04) 0.38 (0.12) 0.46 (0.09) 0.78 (0.09) 2.56 (0.08) 3.69 (0.11) 4.50 (0.06) 5.36 (0.09)

0.18 (0.09) 0.36 (0.06) 0.48 (0.08) 0.77 (0.06) 2.60 (0.12) 3.74 (0.10) 4.52 (0.14) 5.24 (0.09)

Fig. 5. The mean (S.D.) aerodynamic particle size distribution of the dose emitted from a formoterol Turbuhaler at different inhalation flows measured by the MIXINLET method (n = 3 at each flow rate).

3. Results 3.1. Validation of pressure changes at different flows Table 1 shows that the pressure changes (P) within the Turbuhaler for different flows using the MIXINLET and STANDARD methods were the same. This table confirms that for the MIXINLET method the inhalation flow drawn through the Turbuhaler was as expected. Statistical analysis revealed no difference in the pressure changes between the MIXINLET and STANDARD methods. 3.2. Aerodynamic characteristics of the emitted dose 3.2.1. ACI with mixing inlet (MIXINLET method) Fig. 5 describes the cumulative size distribution of the dose emitted at flows from 10 to 60 l min−1 when the mixing inlet was used with the ACI. From these distributions Table 2 summarises the effect of inhalation flow on the aerodynamic characteristics of

Fig. 6. The mean (S.D.) MMAD from a formoterol Turbuhaler measured by the MIXINLET and STANDARD methods at different inhalation flows (n = 3).

Table 2 The mean (S.D.) aerodynamic characteristics of the emitted dose from a formoterol Turbuhaler with respect to each 12 ␮g (nominal) dose using the MIXINLET method (n = 3). Aerodynamic parameter

Induction port (␮g) Mixing inlet (␮g) Pre-separator (␮g) FPD (␮g) TED (␮g) MMAD (␮m) GSD

Inhalation flow (l min−1 ) 10

20

28.3

40

60

1.54 (0.11) 0.97 (0.03) 1.33 (0.06) 0.55 (0.05) 5.90 (0.19) 6.60 (0.10) 2.13 (0.06)

1.34 (0.03) 0.87 (0.09) 1.21 (0.04) 1.39 (0.03) 6.91 (0.14) 5.97 (0.06) 2.13 (0.08)

1.28 (0.08) 0.82 (0.03) 1.16 (0.06) 1.80 (0.12) 7.43 (0.13) 5.37 (0.25) 1.73 (0.06)

1.23 (0.09) 0.74 (0.10) 1.12 (0.06) 2.88 (0.08) 7.93 (0.38) 5.13 (0.06) 1.63 (0.10)

0.98 (0.16) 0.52 (0.05) 0.86 (0.13) 5.86 (0.20) 8.98 (0.11) 2.77 (0.06) 1.50 (0.10)

352

D.K. Nadarassan et al. / European Journal of Pharmaceutical Sciences 39 (2010) 348–354

Table 3 The mean (S.D.) aerodynamic characteristics of the emitted dose from a formoterol Turbuhaler with respect to each 12 ␮g (nominal) dose using the STANDARD method (n = 3). Aerodynamic parameter

Induction port (␮g) Pre-separator (␮g) FPD (␮g) TED (␮g) MMAD (␮m) GSD

Inhalation flow (l min−1 ) 10

20

28.3

40

60

1.73 (0.10) 1.86 (0.05) 0.13 (0.03) 5.91 (0.23) 12.17 (0.3) 1.7 (0.1)

1.85 (0.07) 1.95 (0.05) 0.69 (0.05) 6.82 (0.38) 7.37 (0.02) 1.7 (0.01)

1.65 (0.07) 1.84 (0.05) 1.50 (0.19) 7.09 (0.21) 5.48 (0.06) 2.18 (0.24)

1.64 (0.11) 1.82 (0.12) 2.48 (0.32) 7.38 (0.12) 4.80 (0.10) 1.80 (0.10)

0.80 (0.10) 1.07 (0.22) 5.42 (0.20) 8.67 (0.14) 2.67 (0.21) 1.67 (0.08)

the emitted dose from a formoterol Turbuhaler. This table together with Figs. 6 and 7 highlight the flow dependent dose emission from this product between 30 and 60 l min−1 and more importantly the effect of flow below 30 l min−1 . The data reveals that at low inhalation flows the aerodynamic properties of the emitted dose suggest a decreased potential for lung deposition and that the MMAD drops below 5 ␮m only above 40 l min−1 . Fig. 7 shows that the mean (S.D.) FPF at 10, 20, 28.3, 40 and 60 l min−1 is 4.6% (0.5), 11.6% (0.3), 15.0% (1.2), 24.0% (0.8) and 48.8% (1.7). Fig. 7 also confirms that the increase in the TED is inhalation flow dependent with respective amounts of 49.2% (1.9), 57.6% (1.4), 61.9% (1.3), 66.1% (3.8) and 74.8% (1.1) of the nominal dose. The mean (S.D.) emitted dose pre-impactor (EDPI) was 3.83 ␮g per dose (0.08), 3.42 ␮g per dose (0.11), 3.26 ␮g per dose (0.11), 3.08 ␮g per dose (0.20) and 2.36 ␮g per dose (0.22). 3.2.2. ACI without the mixing inlet (STANDARD method) The data in Table 3 and Figs. 6 and 7 also confirms that when using the traditional ACI method the aerodynamic characteristics from the formoterol Turbuhaler are inhalation flow dependent. The mean (S.D.) FPF at 10, 20, 28.3, 40 and 60 l min−1 was 1.1% (0.2), 5.8% (0.4), 12.5% (1.9), 20.7% (3.2) and 45.2% (1.7). The TED was 49.2% (1.9), 56.8% (3.2), 59.0% (2.1), 61.5% (1.2) and 72.3% (2.2) of the nominal dose as shown in Fig. 7. The mean (S.D.) emitted dose pre-impactor (EDPI) was 3.59 ␮g per dose (0.11), 3.80 ␮g per dose (0.12), 3.49 ␮g per dose (0.11), 3.46 ␮g per dose (0.22) and 1.87 ␮g per dose (0.30). 3.2.3. Comparison of the ACI with and without the mixing valve Statistical analysis revealed no difference between the MIXINLET and STANDARD methods for the TED and the EDPI. As highlighted in Figs. 6 and 7 statistical analysis revealed a differ-

Fig. 8. A comparison of the aerodynamic particle size distributions of the dose emitted from a formoterol Turbuhaler obtained from the MIXINLET and STANDARD methods at inhalation flows of (a) 28.3 l min−1 and (b) 60 l min−1 (n = 3).

ence (p < 0.5) in the MMAD and the FPD between the two methods at flows below 28.3 l min−1 . Similar aerodynamic dose emission characteristics between the two methods at flows >28.3 l min−1 are demonstrated by the data in Fig. 8a and b. 4. Discussion Fig. 7. The mean (S.D.) FPF and the TED% from a formoterol Turbuhaler measured by the MIXINLET and STANDARD methods (n = 3).

Previous measurements of pressure and flow have revealed no fluctuations during operation when similar methodology was used

D.K. Nadarassan et al. / European Journal of Pharmaceutical Sciences 39 (2010) 348–354

(Miler et al., 2000). Similar pressure changes at different inhalation flows through the Turbuhaler with and without the introduction of supplementary air using the mixing inlet (presented in Table 1) demonstrate that the inhalation conditions inside the Turbuhaler are the same for the two methods. When using flows >28.3 l min−1 the cut-off diameters of the original ACI stages can be re-calculated (Oort, 1995). However, the accuracy of the aerodynamic data when using the original ACI alone at flows <28.3 l min−1 is unknown and hence the aerodynamic characteristics of DPIs below these flows have not been studied. This is the reason why the mixing inlet was introduced. At these low flow rates the Reynolds number becomes too low, and other factors come into play such as gravity and maintenance of turbulent airflow. All these combine to decrease the efficiency of impaction in the ACI at low inhalation flows. This is highlighted by the bigger discrepancy between the aerodynamic data of the MIXINLET and STANDARD methods below 28.3 l min−1 . However the TED and the EDPI (the amount emitted in the induction port and pre-separator plus the mixing inlet for the MIXINLET) measured by both methods were similar. This together with the same pressure drops inside the DPI confirms the similar inhalation conditions inside the Turbuhaler. Fig. 6, however, suggests that rather than the inefficiency of the ACI at low flows there may be de-aggregation of the dose inside the ACI when using low flows with the MIXINLET method. Using similar methodology it has been reported that the aerodynamic characteristics emitted from an active DPI where the same but those of a passive DPI were different at flows of 20 and 50 l min−1 (Miler et al., 2000). Hence de-aggregation is not occurring in the ACI. This is further supported by clinical data reported by Pedersen et al. (1990) describing the bronchodilator response of terbutaline from a Turbuhaler at different inhalation flows. There was a significant increase in the FEV1 of these 14 asthmatic children, compared to baseline, after the inhalation of 250 ␮g terbutaline using flows of 13, 22, 30 and 60 l min−1 . The degree of response obtained at 13 and 22 l min−1 would be more consistent with the FPD and MMAD values from the MIXINLET method at the low flows. It is very unlikely that the very low FPD and high MMAD for the inhalation flows <30 l min−1 , determined by the STANDARD method would produce the level of bronchodilator response in these children. The methodology we have used does not restrict measurements up to 60 l min−1 because this can be extended to 90 l min−1 (for DPIs with a low resistance) by using the ACI with the appropriate stages for this rate. Using the same inhalation flow with the ACI means that cut-off diameters are not changed and that the conditions inside the impactor are always the same irrespective of the flow through the inhaler. An alternative would be to use the Next Generation Impactor (NGI). The NGI has been calibrated for operation at flow rates of 30, 60 and 100 l min−1 (Marple et al., 2003) and also at 15 l min−1 for nebulisers (Marple et al., 2004). Like the ACI the cut-off diameters have to be re-calculated at different flows. Also at the lower flow rate it is recommended that the pre-separator is not used and that the MOC is ineffective due to the very low Reynolds number at the nozzles of the MOC (Marple et al., 2004). How these recommendations would effect DPI determinations has not been studied. The NGI could be operated with the mixing inlet in the same way that we have used it with the ACI. This would allow operation of the NGI at the same flow and thus the cut-off stages would not need to be adjusted. Previous in vitro methods to measure the aerodynamic dose emission characteristics from a DPI have used a holding chamber to draw a dose from the DPI and then transfer the emitted dose into an ACI operated at 28.3 l min−1 . The electronic lung has been used to determine the dose emission characteristics using simulations

353

of patient inhalation profiles of children with asthma (Bisgaard et al., 1998) and adults with severe asthma (Tarsin et al., 2006). A different approach, by Kamin et al. (2002), used a 1 l volume drawn at different flows and acceleration rates to emit a dose from a DPI into a 3 l vessel that was immediately sampled by the ACI. A third method has used methodology similar to ours in that supplementary air was used to determine tiotropium dose emission at low flows through the HandiHaler (Chodosh et al., 2001). All these like the method we report have their merits. Our method does not require the extra step of emitting the dose into a holding chamber and allows the ACI to be used at the same flow. At present we are extending our methodology to varying the inhalation volume because a 4 l inhalation is a large volume for a patient to achieve. Overall the results highlight that with respect to the fine particle dose and the MMAD the minimum rate for use with the Turbuhaler is approximately 30–40 l min−1 . In vivo determination of the lung dose following inhalation of terbutaline (Newman et al., 1991) and budesonide (Borgstrom et al., 1994) has been reported to double from 30 to 60 l min−1 . Our in vitro aerodynamic characteristics suggest that there would be a doubling of the lung dose from 40 to 60 l min−1 . However it has been shown that the Turbuhaler provides a similar effect at 30 and 60 l min−1 (Pedersen et al., 1990) but this could be due to making measurements at the plateau of the dose response relationship. Nevertheless 30 l min−1 is regarded as the minimum value for use with the Turbuhaler. Although the MMAD at this rate is still above 5 ␮m the fine particle dose is about 15% of the nominal dose. Overall, therefore, there is a distinct possibility of a link between the aerodynamic characteristics of the emitted dose, lung deposition and clinical response but it may not be a direct relationship. Our FPD data from the MIXINLET method is consistent with the bronchodilator response obtained when asthmatic children inhaled terbutaline at flows ≤30 l min−1 from a Turbuhaler (Pedersen et al., 1990). It is doubtful if a similar comparison can be made for inhaled steroids because a high MMAD may be useful for a bronchodilator whereas for a steroid a smaller MMAD may be better. Hence it is important to use both in vitro and in vivo response to determine the lowest flow required for an inhaler and that these may be different for bronchodilators and anti-inflammatory agents. This should extend to different strengths in the same device because of the de-aggregation requirements for different drug:excipient ratios. The results show that at 60 l min−1 the fine particle dose emitted from the Turbuhaler is about 50% of the nominal dose which may explain why attention in the past has focussed on optimal inhalation flows. At 60 l min−1 when using the ACI with and without the mixing inlet the total emitted dose was 9 and 8.7 ␮g, respectively. This compares to the labelled emitted dose of 9 ␮g per metered dose. Peak inhalation flow and the optimal flow for each inhaler has traditionally been the focus of attention. However, as long as a dose with the potential to deposit into the lungs is emitted then a clinical response will be obtained. Since control is titrated to dose then it is not clinically relevant that a particular patient can only achieve a low flow (and hence a small amount of turbulent energy inside the inhaler) because the dose (even though it is low) is adjusted accordingly. The clinically relevant issue is the minimum flow below which the emitted dose does not have the necessary characteristics to be deposited into the lungs. This would be more important during an acute exacerbation because inhalation flows through an inhaler are reduced and recover slowly (Broeders et al., 2004). It is important, therefore, that the dose emission properties of all inhalers are determined at low flows. The results show how our method can be used to identify these minimum flows for DPIs and thus focus attention to this clinically relevant issue rather than current emphasis that is directed towards optimal flows.

354

D.K. Nadarassan et al. / European Journal of Pharmaceutical Sciences 39 (2010) 348–354

5. Conclusion Our ACI methodology with the mixing inlet was able to measure the dose emission characteristics of the Turbuhaler at very low flows. This new method also allows measurement of the aerodynamic characteristics of the emitted dose without having to alter the cut-off diameters of the impactor when using different flows. Our methodology can now be used to identify the dose emission properties of all DPIs to indicate the lowest flow required to use the device. This is more clinically important than a debate about what is the optimal flow. Since the turbulent energy to de-aggregate the dose is a function of the formulation, the inhaler’s resistance and the patient’s flow then these minimum flows will be different for each type of DPI and therapeutic class. Disclosure statement Henry Chrystyn has no shares in any pharmaceutical companies. He has received sponsorship to carry out studies, together with some consultant agreements and honoraria for presentations, from several pharmaceutical companies that market inhaled products. These include AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, Innovata Biomed, Meda, Napp Pharmaceuticals, Omron, Orion Teva, Trinity Chiesi, Truddell and UCB. Research sponsorship has also been received from grant awarding bodies (EPSRC and MRC). The work in this paper was supported by the PhD tuition fees of the University of Bradford. Conflict of interest None of the authors have a conflict of interest. References Al-Showair, R.A., Tarsin, W.Y., Assi, K.H., Pearson, S.B., Chrystyn, H., 2007. Can all patients with COPD use the correct inhalation flow with all inhalers and does training help? Respir. Med. 101, 2395–2401. Bisgaard, H., Klug, B., Sumby, B.S., Burnell, P.K.P., 1998. Fine particle mass from the Diskus inhaler and Turbuhaler inhaler in children with asthma. Eur. Respir. J. 11, 1111–1115. Borgstrom, L., Bondesson, E., Moren, F., Trofast, E., Newman, S.P., 1994. Lung deposition of budesonide inhaled via Turbuhaler® : a comparison with terbutaline sulphate in normal subjects. Eur. Respir. J. 7, 69–73. British Pharmacopoeia (BP), 2005. Preparations for Inhalation. Aerodynamic Assessment of Fine Particles—Fine Particle Dose and Particle Size Distribution (Ph. Eur. Method 2.9.18). British Pharmacopoeia, pp. A277–A290. Broeders, M.E.A.C., Molema, J., Hop, W.C.J., Folgering, H.Th.M., 2003. Inhalation profiles in asthmatics and COPD patients: reproducibility and effects of instruction. J. Aerosol Med. 16, 131–141. Broeders, M.E., Molema, J., Hop, W.C., Vermue, N.A., Folgering, H.T., 2004. The course of inhalation profiles during an acute exacerbation of obstructive lung disease. Respir. Med. 98, 1173–1179. Chodosh, S., Flanders, J.S., Kesten, S., Serby, C.W., Hochrainer, D., Witek Jr., T.J., 2001. Effective delivery of particles with the HandiHaler dry powder inhalation system over a range of chronic obstructive pulmonary disease severity. J. Aerosol Med. 14, 309–315. Clark, A.R., Hollingworth, A.M., 1993. The relationship between powder inhaler resistance and peak inspiratory conditions in healthy volunteers—implications for in vitro testing. J. Aerosol Med. 6, 99–110.

European Pharmacopoeia (EP), 2005. 2.9.18 Preparation for Inhalation: Aerodynamic Assessment of Fine Particles, 5.0 edition. European Pharmacopoeia, pp. 245–253. Everard, M.L., Devadason, S.G., Souëf, P.N.L., 1997. Flow early in the inspiratory manoeuvre affects the aerosol particle size distribution from a Turbuhaler. Respir. Med. 91, 624–628. Hill, L.S., Slater, A.L., 1998. A comparison of the performance of two modern multidose dry powder asthma inhalers. Respir. Med. 92, 105–110. Hindle, M., Byron, P.R., 1995. Dose emission from marketed dry powder inhalers. Int. J. Pharm. 116, 169–177. Hinds, W.C.L., 1999. Aerosol Technology: Properties, Behaviour, and Measurement of Airborne Particles, 2nd edition. Wiley–Interscience, NY, USA. Kamin, W.E., Genz, T., Roeder, S., Scheuch, G., Trammer, T., Juenemann, R., Cloes, R.M., 2002. Mass output and particle size distribution of glucocorticosteroids emitted from different inhalation devices depending on various inspiratory parameters. J. Aerosol Med. 15, 65–73. Marple, V.A., Olson, B.A., Santhanakrishan, K., Mitchell, J.P., Murray, S.C., HudsonCurtis, B.L., 2003. Next Generation Pharmaceutical Impactor (a new impactor for pharmaceutical inhaler testing). Part II: Archival calibration. J. Aerosol. Med. 16, 301–324. Marple, V.A., Olson, B.A., Santhanakrishan, K., Roberts, D.L., Mitchell, J.P., HudsonCurtis, B.L., 2004. Next Generation Pharmaceutical Impactor: A new impactor for pharmaceutical inhaler testing. Part III. Extension of archival calibration to 15 L/min. J. Aerosol. Med. 17, 335–343. Miler, N.C., Maniaci, M.J., Dwivedi, S., Ward, G.W., 2000. Aerodynamic sizing with simulated inhalation profiles: total dose capture and measurement. In: Dalby, R.N., Byron, P.R., Farr, S.J., Peart, J. (Eds.), Respiratory Drug Delivery VII, vol. 1. Serentec Press, Raleigh, NC, pp. 191–196. Mitchell, J.P., Nagel, N.W., 2003. Cascade impactors for the size characterization of aerosols from medical inhalers: their uses and limitation. J. Aerosol Med. 16, 341–376. Newman, S.P., Morén, F., Trofast, E., Talaee, N., Clarke, S.W., 1991. Terbutaline sulphate Turbuhaler: effect of inhaled flow rate on drug deposition and efficacy. Int. J. Pharm. 74, 209–213. Oort, M.V., 1995. In vitro testing of dry powder inhalers. Aerosol Sci. Technol. 22, 364–373. Palander, A., Mattila, T., Karku, M., Muttonen, E., 2000. In vitro comparison of three salbutamol-containing multidose dry powder inhalers. Clin. Drug Invest. 20, 25–33. Pedesen, S., 1986. How to use a Rotahaler. Arch. Dis. Child. 61, 11–14. Pedersen, S., Hansen, O.R., Fuglsang, G., 1990. Influence of inspiratory flow rate upon the effect of a Turbuhaler. Arch. Dis. Child. 65, 308–319. Ross, D., Schultz, R., 1996. Effect of inhalation flow rate on the dosing characteristics of dry powder inhaler (DPI) and metered dose inhaler (MDI) products. J. Aerosol. Tarsin, W., Assi, K.H., Chrystyn, H., 2004. In-vitro inhalation flow rate dependent dosage emission from a combination of budesonide and eformoterol in a dry powder inhaler. J. Aerosol Med. 17, 25–32. Tarsin, W.Y., Pearson, S.B., Assi, K.H., Chrystyn, H., 2006. Emitted dose estimates from Seretide Diskus and Symbicort Turbuhaler following inhalation by severe asthmatics. Int. J. Pharm. 316, 131–137. United States Pharmacopeia (USP), 2005. Aerosols, Nasal Sprays, Metered Dose Inhalers and Dry Powder Inhalers, vol. 28. The United States Pharmacopeia, pp. 2359–2377.

Glossary FPD: Fine particle dose, amount containing particles <5 ␮m. FPF: Fine particle fraction; the FPD divided by the nominal (labelled) dose which is 12 ␮g. EDPI: Emitted dose pre-impactor. For the MIXINLET method this is the amount deposited in the induction port, the mixing inlet and the pre-separator, for the STANDARD method this is the amount deposited in the induction port and the pre-separator. Nominal dose: This is the dose on the label of the Oxis Turbuhaler (in the UK) which in this case is the amount that is metered (12 ␮g). TED: Total emitted dose; the amount deposited in all parts of the impactor equipment. TED%: The total emitted dose divided by the nominal dose and expressed in %.