- Email: [email protected]
Evaluation of Impaction Force of Nasal Sprays and Metered-Dose Inhalers Using the Texture Analyser☆
Evaluation of Impaction Force of Nasal Sprays and Metered-Dose Inhalers Using the Texture Analyser CHANGNING GUO, WEI YE, JOHN KAUFFMAN, WILLIAM H. DOUB U.S. Food & Drug Administration, Division of Pharmaceutical Analysis, 1114 Market Street, Room 1002, St. Louis, Missouri 63101
Received 27 March 2008; revised 7 October 2008; accepted 4 November 2008 Published online 18 December 2008 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.21648
ABSTRACT: The impaction force from an inhalation product is an important characteristics by which to characterize the spray plume. It is one of the plume characteristics that can be perceived by a patient, and is expected to be good measures of local delivery equivalence for inhalation drugs. A Stable Micro Systems TA-XT.plus Texture Analyser equipped with 750 g load cell was used to measure the impaction force of several nasal sprays and metered-dose inhalers (MDIs). A survey of several commercial nasal spray and MDI products shows that impaction forces of these products varies from 1.5 to 6.5 g force and are significantly different from each other. A 3-level, 4-factor Box–Behnken design was applied to the study of impaction force of nasal sprays using placebo solutions. The influences of four factors: actuation stroke length, actuation velocity, concentration of gelling agent, and concentration of surfactant, were investigated. Of those factors examined here, actuation velocity exerts the greatest effect on impaction force. Impaction force is a discriminative parameter for in vitro testing of nasal spray and MDI products. Since impaction force is more directly related to patient sensation and aerosol deposition in the nasal mucus than other, more traditional parameters, it may provide a better way to evaluate in vitro equivalence in support of abbreviated new drug applications (ANDAs) for orally inhaled and nasal drug products. ß 2008 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 98:2799–2806, 2009
Keywords: impaction force; nasal spray; MDI; Texture Analyser; design of experiment
INTRODUCTION Inhalation drug products are usually characterized via measurement of shot weight, spray pattern, plume geometry, and droplet size distribution (DSD). These parameters are recommended by the FDA1–3 and widely used by the pharmaceutical industry for assessment of equivalence between two nasal spray products. In addition to the abovementioned parameters, the impaction force of the aerosol plume is also an important characteristics by which the spray The views presented in this paper do not necessarily reflect those of the Food and Drug Administration. Correspondence to: Changning Guo (Telephone: 314-5393852; Fax: 315-539-2113; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 98, 2799–2806 (2009) ß 2008 Wiley-Liss, Inc. and the American Pharmacists Association
plume may be evaluated. Impaction force is also among the most noticeable characteristics perceived by a patient. For nasal spray products, a forceful plume can be especially uncomfortable to the patient while, for metered-dose inhaler (MDI) products, the situation is more critical. In the case of an MDI product, a forceful blast of cold, liquid propellant impacting on the back of the patient’s throat normally results in two adverse conditions, high throat deposition4 and the cold-Freon effect.5,6 In addition to being uncomfortable for the user, a high-velocity, forceful blast of formulation exiting the device can result in an inconsistent or nonexistent dose delivered to the lung. To our knowledge, there have been only two papers published where impaction force
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2799
2800
GUO ET AL.
measurements have been made for inhalation drug spray plumes. In an early study, Gabrio et al.7 constructed a device consisting of a precision miniature load cell and a quick-response thermocouple which they used to determine the impaction force and plume temperature of various MDIs. Impaction force measurements were made for 28 marketed products consisting of bronchodilators, steroids, press-and-breathe, breath- actuated and nasal inhalers, all using manual actuations. They concluded that there are significant differences in the spray force and plume temperatures of MDIs, and that impaction force is affected by both the type of propellant and the actuator orifice diameter. Spray force measurements were shown to be good indicators of the amount of drug deposition occurring in the throat. In a previous study by our group,8 a method to measure the impaction force for nasal sprays and MDIs was developed using a commercially available instrument. A 20-mL Pfeiffer nasal spray pump filled with water and a Flovent1 HFA 44 mg MDI (GSK) were used as test products. The results showed that the maximum impaction force varied significantly at different spray distances. The maximum impaction forces were observed at around 3 and 6 cm for the nasal spray and the MDI, respectively. In this study, impaction forces of several commercially available nasal spray and MDI products were surveyed. Automated actuators were used to obtain optimal and consistent actuation performance for each tested product. Design of experiments (DOE) methodology was used to gain understanding of the influence of actuation parameters and formulation properties on the impaction force of nasal sprays. A 3-level, 4-factor Box–Behnken design was applied to the study of the impaction force of nasal sprays using placebo solutions. The influences of actuation parameters and formulation properties on the impaction force were investigated.
was recorded as the measured impaction force of the actuation. SprayVIEW NSx and MDx automated actuators (Proveris Scientific Corporation, Sudbury, MA) were used to mechanically control the actuation of nasal spray and MDI products, respectively, in order to achieve consistent actuation performance. Viota software (Proveris Scientific Corporation) was used to characterize different products and establish appropriate actuation parameters. For all the surveyed nasal sprays and MDIs, the auto characterization method was used, with the force of contact set at 0.3 kg and force at end of stroke set at 6 kg. Actuation profiles for all the tested products were symmetric with actuation velocity of 50 mm/s, actuation acceleration of 4000 mm/s2, and hold time of 100 ms.
METHODS A Stable Micro Systems TA-XT.plus Texture Analyser (Texture Technologies Corp., Scarsdale, NY) equipped with a 750 g load cell and a TA-40 probe (10.0 cm diameter) was used to measure the impaction force. The load cell was sampled every 5 ms (200 data points per second) which allowed the generation of force versus time curves. The maximum value in the force versus time profile JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
Figure 1. Instrument setup for nasal spray (a) and MDI (b) impaction force measurements. DOI 10.1002/jps
DOI 10.1002/jps
API
Ipratropium bromide and albuterol sulfate Fluticasone propionate Fluticasone propionate Albuterol Albuterol Beclomethasone Dipropionate Albuterol
API
Beclomethasone dipropionate monohydrate Desmopressin acetate Fluticasone propionate Cromolyn sodium Flunisolide Mometasone furoate monohydrate None
Flovent HFA 44 Flovent HFA 220 Proventil Proventil_HFA Qvar (BDP HFA) Ventolin
Combivent
MDI
DDAVP Flonase NasalCrom Nasarel Nasonex Water
Beconase AQ
Nasal Spray
HFA HFA CFC HFA HFA CFC
CFC
Propellant
Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT GlaxoSmithKline, Research Triangle Park, NC GlaxoSmithKline, Research Triangle Park, NC Schering Corporation, Kenilworth, NJ Key Pharmaceuticals, Inc., Kenilworth, NJ IVAX Laboratories, Inc., Miami, FL Allen & Hanburys, Division of Glaxo, Inc., Research Triangle Park, NC
Manufacturer
4.6 3.5 5.3 3.0 3.8 4.4
4.9
Average Force (g)
6.5 6.1 4.4 1.5 1.8 5.2
3.7
Average Force (g)
Rhone-Poulenc Rorer Pharmaceuticals, Inc., Fort Washington, PA Allen & Hanburys, Division of Glaxo, Inc., Research Triangle Park, NC Pharmacia Consumer Healthcare, Peapack, NJ IVAX Laboratories, Inc., Miami, FL Schering Corporation, Kenilworth, NJ Pfeiffer pump filled with DI water
Glaxo Wellcome, Inc., Research Triangle Park, NC
Manufacturer
0.2 0.2 0.2 0.1 0.1 0.1
0.2
SD (g)
0.2 0.1 0.2 0.2 0.1 0.2
0.1
SD (g)
2.8 2.5 4.7 6.5 6.6 2.1
5.2
RSD (%)
4.1 2.9 3.2 5.6 2.6 4.4
2.9
RSD (%)
Table 1. Impaction Force of Several Commercially Available Nasal Spray and MDI Products at 3 and 6 cm Distance, Respectively, with Manual Activation (n ¼ 6)
IMPACTION FORCE OF NASAL SPRAY AND MDI
2801
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2802
GUO ET AL.
Figure 1 shows the instrument setup for impaction force measurement of nasal sprays and MDIs. For nasal spray impaction force measurements, the nasal spray pump was actuated upward using a SprayVIEW NSx automated actuator; For MDI impaction force measurements, the Texture Analyser was placed on its side with the MDI being discharged horizontally using a SprayVIEW MDx automated actuator. Six nasal sprays and seven MDI products were used as test products in this study. The product names and manufacturers are listed in Table 1. Pfeiffer (PFE) 0.10 mL nasal spray pumps (material number 62602, dip tube length 58 mm) and 20 mL bottles (material number 34473) filled with placebo solutions were used in both survey and DOE studies. The PFE nasal spray units were provided by Pfeiffer of America (Princeton, NJ) and are claimed to deliver 100 mL per actuation, equivalent to 100 mg water per actuation. DOE methodology was used to gain an understanding of the influence of actuation parameters and formulation properties on the impaction force of nasal sprays. Due to a limited ability to make different MDI formulations, we did not run a similar DOE study on MDIs. A 3-level, 4-factor Box–Behnken design was selected for the nasal spray study, because it can evaluate quadratic interactions between pairs of factors while minimizing the number of required experiments. JMP 5.1.1 software (SAS Institute, Cary, NC) was used to generate the DOE matrix and analyze the response surface models. The PFE nasal spray unit was filled with 18 mL placebo solutions prior to testing and the first six actuations were fired to waste as priming shots. The placebo formulations used in this study are aqueous solutions of extra-low-viscositygrade carboxymethylcellulose (CMC, Hercules, Inc., Wilmington, DE) and Tween-80 (Pharmacia, Piscataway, NJ) in different concentrations. CMC and Tween-80 are both common pharmaceutical excipients. CMC is normally used as an additive to increase solution viscosity, and Tween-80 is normally used as a surfactant to adjust the surface tension of solutions.
RESULTS Impaction Forces of Some Commercial Nasal Spray and MDI Products Figure 2 shows the impaction force versus time profiles from a PFE nasal spray pump (filled JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
Figure 2. Impaction force versus time profile from a nasal spray pump filled with water at 3 cm spray distance (a) and a Flovent1 HFA 44 mcg MDI at 6 cm spray distance (b).
with water) and a Flovent1 HFA 44 mcg MDI as examples. For the impaction force measurements performed in this study, signal-to-noise ratios were typically greater than 10. The baseline noise level could be due to building vibration. A higher noise level was observed for the nasal spray impaction force measurement than for the MDI, which can be explained by the different orientations of the instrument setup. For the nasal spray setup, the impaction plate is suspended such that it may be more sensitive to external vibration sources. For MDI’s, the apparatus lies on its side and external vibrations may be damped. In this study, impaction forces for nasal sprays and MDIs were measured from 3 and 6 cm from nozzle tip or mouth piece, respectively, where the maximum impaction forces were observed according to an earlier study.8 Table 1 shows the DOI 10.1002/jps
IMPACTION FORCE OF NASAL SPRAY AND MDI
measured impaction force for several nasal spray and MDI products. Optimization of actuation parameters for each nasal spray and MDI product was performed using the SprayVIEW NSx/MDx automated actuators. The relative standard deviation (RSD) for six replicates measurements of impaction force varied from 2.1% to 6.6% for all tested products, indicating good repeatability for the method. The impaction force of nasal spray products varied from 3.0 to 4.9 g, and those of MDI products ranged from 1.5 to 6.5 g. T-test was performed between each pair of means for nasal spray group and MDI group separately. Observed differences between the different products were statistically significant at the 95% confidence level, with the exception of DDAVP versus water in Pfeiffer pump, which suggests that impaction force is a discriminative parameter for in vitro testing of nasal spray and MDI products.
2803
Influence of Actuation Parameters and Formulation Properties on Nasal Spray Performance A 3-level, 4-factor Box–Behnken design was used for this study. Four factors, actuation stroke length (3.5, 4.4, 5.3 mm), actuation velocity (30, 50, 70 mm/s), concentration of CMC (0%, 1%, 2%), and concentration of Tween-80 (0%, 2.5%, 5%) were investigated for their influences on the impaction force of nasal spray at 3 cm distance. Selection of the range of these factors is based on previous studies.9,10 The concentration of CMC and Tween-80 are the dominant factors which influence the formulation viscosity and surface tension, respectively. Table 2 shows the resultant DOE input parameter and data table for the 3-level, 4-factor Box–Behnken design used here. Nine placebo formulations with different viscosities and surface tensions were prepared as aqueous solutions by
Table 2. DOE Input Parameter Table with Response for the 3-Level, 4-Factor Box–Behnken Design (Total of Nine Different Formulations) Experiment # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 DOI 10.1002/jps
Pattern
Stroke Length (mm)
Velocity (mm/s)
CMC (%)
Tween-80 (%)
Impaction Force (g)
00– 00 00 þ00 0þ0 00þ 00 00 þ0-0 0þ0 –00 þ00 0000 0000 0000 þ00 þþ00 0þ0 0þ0 þ0þ0 0þþ0 00þ 00þ 00þ þ00þ 0þ0þ 00þþ
4.4 4.4 3.5 5.3 4.4 4.4 4.4 3.5 5.3 4.4 3.5 5.3 4.4 4.4 4.4 3.5 5.3 4.4 3.5 5.3 4.4 4.4 4.4 3.5 5.3 4.4 4.4
50 30 50 50 70 50 30 50 50 70 30 30 50 50 50 70 70 30 50 50 70 50 30 50 50 70 50
0 1 1 1 1 2 0 0 0 0 1 1 1 1 1 1 1 2 2 2 2 0 1 1 1 1 2
0 0 0 0 0 0 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 5 5 5 5 5 5
3.6 2.1 4.3 5.3 4.4 4.5 2.5 4.7 5.1 5.2 2.3 2.4 4.6 4.6 4.7 5.3 6.7 2.3 4.4 5.0 5.2 4.7 2.4 4.6 5.2 5.3 4.7
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2804
GUO ET AL.
Table 3. The Coefficients of the Quadratic Model and Effect Tests for Impaction Force of Nasal Spray at 3 cm Distance Term
Coefficient Estimate STD Errors p-Value
Intercept S V C T SV SC VC ST VT CT SS VV CC TT
b0 b1 b2 b3 b4 b5 b6 b7 b8 b9 b10 b11 b12 b13 b14
4.63 0.34 1.51 0.03 0.23 0.33 0.05 0.05 0.10 0.15 0.23 0.34 0.81 0.09 0.19
0.18 0.09 0.09 0.09 0.09 0.16 0.16 0.16 0.16 0.16 0.16 0.14 0.14 0.14 0.14
<0.0001 0.0028 <0.0001 0.7891 0.0300 0.0626 0.7576 0.7576 0.5395 0.3621 0.1808 0.0300 <0.0001 0.5354 0.1966
employing different concentrations of CMC and Tween-80. A total of 27 experiments using the factor values described in Table 2 were performed, and responses were measured for each experiment. The coded design patterns shown in Table 3 symbolize the scaled factor values (high (þ), middle (0) and low ()) used in each run, in the order of stroke length, velocity, concentration of CMC and Tween-80, respectively. The measured impaction force (response) is also tabulated in Table 2. Each value is the average of four replicates. Scaled factors are traditionally used in a DOE prediction equation for a response from selected input parameters. A scaled factor (range 1 to þ1) is a parameter centered on the mean and scaled by parameter range/2. Fscaled ¼
2ðF Fcenter Þ ðFmax Fmin Þ
(1)
In Eq. (1), F is the factor value (experimental parameter), Fscaled is the scaled factor value, Fmax, Fmin, and Fcenter are the maximum, minimum, and center point parameter values used in the DOE. Factor scaling normalizes the influence of each factor on the prediction response. The scaled factors for the four factors used in this study are given by the following expressions. S¼
2 ðStroke Length 4:4Þ ð5:3 3:5Þ
(2)
2 ðVelocity 50Þ ð70 30Þ
(3)
V¼
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
C¼
T¼
2 ðCMC concentration 1%Þ ð2 0%Þ
2 ðTween80 concentration 2:5%Þ ð5 0%Þ
(4)
(5)
S, V, C, and T are the scaled factor values for stroke length, velocity, CMC concentration, and Tween-80 concentration, respectively. A nonlinear quadratic model of nasal spray impaction force generated by regression of the experimental factor values on the measured response is expressed as follows: R ¼ b0 þ b1 S þ b2 V þ b3 C þ b4 T þ b5 SV þ b6 SC þ b7 VC þ b8 ST þ b9 VT þ b10 CT þ b11 S2 þ b12 V 2 þ b13 C2 þ b14 T2
(6)
where R is the measured response, and the bis are the coefficients that estimate the contribution of each scaled parameter term to the response. Terms containing more than one factor represent interaction terms, and higher order terms indicate a quadratic (nonlinear) nature of relationship. A positive sign for R indicates a synergistic effect, whereas a negative sign represents an antagonistic effect. Because the parameter values are scaled, the magnitudes of the coefficients reflect the importance of the term to the measured response. The regression coefficients and effect test results of the quadratic DOE model (Eq. 6) are presented in Table 3. The effect tests show that, statistically, the measured impaction force of the nasal spray delivery is significantly affected by stroke length ( p ¼ 0.0028), actuation velocity ( p < 0.0001), and concentration of Tween-80 ( p ¼ 0.03) independently, with no significant interaction effects between pairs of the four factors ( p > 0.05 for all interaction terms). However, the scaled estimates indicate that actuation velocity is the dominant factor influencing impaction force, while the actuation stroke length and concentration of Tween-80 have relatively smaller influences on impaction force. Surprisingly, the concentration of gelling agent, which has significant effects on many other nasal spray characteristics, such as spray pattern, plume geometry and DSD,11 has little, if any, influence on impaction force. The positive signs of S terms indicate that increasing stroke length will lead to the production of a larger impaction force. For the V and DOI 10.1002/jps
IMPACTION FORCE OF NASAL SPRAY AND MDI
2805
T terms, the scaled estimates of the first-order terms have positive signs, while the ones of the second-order terms have negative signs. Since the scaled factor values are between 1 and 1 in most of situations, the first-order term has more influence than the second-order term, thus increasing actuation velocity and/or concentration of surfactant will also lead a larger impaction force. By eliminating all insignificant terms from Eq. (6), an optimized impaction force model (shown in Eq. 7 and Tab. 4) was recalculated using JMP. R ¼ 4:45 þ 0:34S þ 1:51V þ 0:23T þ 0:41S2 0:74V 2
(7)
Figure 3 shows a plot of actual versus predicted impaction force from the simplified quadratic model. The quadratic model shows good fit with correlation coefficient (R2) of 0.94. The DOE study helps to identify the source of variability in nasal spray product performance, thus give us a better understanding of how to control the variability. For example, stronger plume with higher impaction force may be obtained by increasing the formulation surface tension and/or via a selection of a pump with a longer stroke length. Moreover, the quadratic models developed from the DOE study quantitatively describe the inherent relationships between the factors and response, which will assist the product design to achieve desired impaction force.
CONCLUSIONS We have developed a method to measure impaction force for inhalation drug products (nasal sprays and MDIs). This new technique exhibits Table 4. The Coefficients of the Simplified Quadratic Model and Effect Tests for Impaction Force of Nasal Spray at 3 cm Distance Term
Coefficient Estimate STD Errors p-Value
Intercept S V T SS VV DOI 10.1002/jps
b0 b1 b2 b4 b11 b12
4.45 0.34 1.51 0.23 0.41 0.74
0.11 0.09 0.09 0.09 0.13 0.13
<0.0001 0.0014 <0.0001 0.0246 0.0044 <0.0001
Figure 3. Correlation plots for predicted and actual impaction force for nasal sprays, where the horizontal broken line is the mean of the Y variable, the solid line is the line of fit, and the two broken curves describe the confidence region where the test is significant at the 95% level relative to the line of fit. (RSq: R2, correlation coefficient; RMSE: root mean square error.)
good repeatability as evidenced by a low RSD for repeated measurements. A survey of several commercial nasal spray and MDI products shows that spray impaction forces for these products differ significantly from each other over a range of 1.5–6.5 g, which indicates that impaction force can be used as a discriminating parameter for in vitro testing for these product types. The DOE study demonstrates that the impaction force is influenced by both formulation properties and actuation parameters, especially the actuation velocity. An optimized quadratic model generated from the DOE results shows good predictive power for impaction force. The distance from a nasal spray nozzle tip or MDI mouth piece to the local delivery site may differ significantly from person to person, and is therefore unrelated to the distance at which the device develops highest maximum impaction force. However, the highest maximum impaction force and its corresponding distance, or maximum impaction forces at 2 or more distances, could be used for quality control or comparison purposes. Since impaction force is more directly related to patient sensation and aerosol deposition than other more traditional, parameters, it may provide a better way to evaluate in vitro equivaJOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2806
GUO ET AL.
lence in support of abbreviated new drug applications (ANDAs) for orally inhaled and nasal drug products. Due to its decent discriminating power, this technique may also provide a good evaluation of manufacturing changes.
REFERENCES 1. U.S. Food and Drug Administration. 1998. Draft Guidance for Industry: Metered Dose Inhaler (MDI) and Dry Powder Inhaler (DPI) Drug Products. 2. U.S. Food and Drug Administration. 2002. Guidance for Industry: Nasal Spray and Inhalation Solution, Suspension, and Spray Drug Products—Chemistry, Manufacturing, and Controls Documentation. 3. U.S. Food and Drug Administration. 2003. Draft Guidance for Industry: Bioavailability and Bioequivalence Studies for Nasal Aerosols and Nasal Sprays for Local Action. 4. Newman SP, Pavia D, Moren F, Sheahan NF, Clarke SW. 1981. Deposition of pressurised aerosols in the human respiratory tract. Thorax 36:52–55. 5. Crompton GK. 1982. Problems patients have using pressurized aerosol inhalers. Eur J Respir Dis Suppl 119:57–65.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
6. Pedersen S, Frost L, Arnfred T. 1986. Errors in inhalation technique and efficiency in inhaler use in asthmatic children. Allergy 41:118– 124. 7. Gabrio BJ, Stein SW, Velasquez DJ. 1999. A new method to evaluate plume characteristics of hydrofluoroalkane and chlorofluorocarbon metered dose inhalers. Int J Pharm 186:3–12. 8. Guo C, Doub WH. 2006. Development of a novel technology to measure impaction force of nasal sprays and metered dose inhalers using the Texture Analyser. In: Dalby RN, Byron PR, Peart J, Suman JD, Farr SJ, editors. Respiratory drug delivery, Vol. 2. 2006. River Grove, IL: Davis Healthcare International Publishing. pp 621– 624. 9. Guo C, Doub WH. 2006. The influence of actuation parameters on in vitro testing of nasal spray products. J Pharm Sci 95:2029–2040. 10. Dayal P, Shaik MS, Singh M. 2004. Evaluation of different parameters that affect droplet-size distribution from nasal sprays using the Malvern Spraytec. J Pharm Sci 93:1725–1742. 11. Guo C, Stein KJ, Kauffman JF, Doub WH. 2008. Assessment of the influence factors on in vitro testing of nasal sprays using Box–Behnken experimental design. Euro J Pharm Sci 35:417–426.
DOI 10.1002/jps
Received 27 March 2008; revised 7 October 2008; accepted 4 November 2008 Published online 18 December 2008 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.21648
ABSTRACT: The impaction force from an inhalation product is an important characteristics by which to characterize the spray plume. It is one of the plume characteristics that can be perceived by a patient, and is expected to be good measures of local delivery equivalence for inhalation drugs. A Stable Micro Systems TA-XT.plus Texture Analyser equipped with 750 g load cell was used to measure the impaction force of several nasal sprays and metered-dose inhalers (MDIs). A survey of several commercial nasal spray and MDI products shows that impaction forces of these products varies from 1.5 to 6.5 g force and are significantly different from each other. A 3-level, 4-factor Box–Behnken design was applied to the study of impaction force of nasal sprays using placebo solutions. The influences of four factors: actuation stroke length, actuation velocity, concentration of gelling agent, and concentration of surfactant, were investigated. Of those factors examined here, actuation velocity exerts the greatest effect on impaction force. Impaction force is a discriminative parameter for in vitro testing of nasal spray and MDI products. Since impaction force is more directly related to patient sensation and aerosol deposition in the nasal mucus than other, more traditional parameters, it may provide a better way to evaluate in vitro equivalence in support of abbreviated new drug applications (ANDAs) for orally inhaled and nasal drug products. ß 2008 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 98:2799–2806, 2009
Keywords: impaction force; nasal spray; MDI; Texture Analyser; design of experiment
INTRODUCTION Inhalation drug products are usually characterized via measurement of shot weight, spray pattern, plume geometry, and droplet size distribution (DSD). These parameters are recommended by the FDA1–3 and widely used by the pharmaceutical industry for assessment of equivalence between two nasal spray products. In addition to the abovementioned parameters, the impaction force of the aerosol plume is also an important characteristics by which the spray The views presented in this paper do not necessarily reflect those of the Food and Drug Administration. Correspondence to: Changning Guo (Telephone: 314-5393852; Fax: 315-539-2113; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 98, 2799–2806 (2009) ß 2008 Wiley-Liss, Inc. and the American Pharmacists Association
plume may be evaluated. Impaction force is also among the most noticeable characteristics perceived by a patient. For nasal spray products, a forceful plume can be especially uncomfortable to the patient while, for metered-dose inhaler (MDI) products, the situation is more critical. In the case of an MDI product, a forceful blast of cold, liquid propellant impacting on the back of the patient’s throat normally results in two adverse conditions, high throat deposition4 and the cold-Freon effect.5,6 In addition to being uncomfortable for the user, a high-velocity, forceful blast of formulation exiting the device can result in an inconsistent or nonexistent dose delivered to the lung. To our knowledge, there have been only two papers published where impaction force
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2799
2800
GUO ET AL.
measurements have been made for inhalation drug spray plumes. In an early study, Gabrio et al.7 constructed a device consisting of a precision miniature load cell and a quick-response thermocouple which they used to determine the impaction force and plume temperature of various MDIs. Impaction force measurements were made for 28 marketed products consisting of bronchodilators, steroids, press-and-breathe, breath- actuated and nasal inhalers, all using manual actuations. They concluded that there are significant differences in the spray force and plume temperatures of MDIs, and that impaction force is affected by both the type of propellant and the actuator orifice diameter. Spray force measurements were shown to be good indicators of the amount of drug deposition occurring in the throat. In a previous study by our group,8 a method to measure the impaction force for nasal sprays and MDIs was developed using a commercially available instrument. A 20-mL Pfeiffer nasal spray pump filled with water and a Flovent1 HFA 44 mg MDI (GSK) were used as test products. The results showed that the maximum impaction force varied significantly at different spray distances. The maximum impaction forces were observed at around 3 and 6 cm for the nasal spray and the MDI, respectively. In this study, impaction forces of several commercially available nasal spray and MDI products were surveyed. Automated actuators were used to obtain optimal and consistent actuation performance for each tested product. Design of experiments (DOE) methodology was used to gain understanding of the influence of actuation parameters and formulation properties on the impaction force of nasal sprays. A 3-level, 4-factor Box–Behnken design was applied to the study of the impaction force of nasal sprays using placebo solutions. The influences of actuation parameters and formulation properties on the impaction force were investigated.
was recorded as the measured impaction force of the actuation. SprayVIEW NSx and MDx automated actuators (Proveris Scientific Corporation, Sudbury, MA) were used to mechanically control the actuation of nasal spray and MDI products, respectively, in order to achieve consistent actuation performance. Viota software (Proveris Scientific Corporation) was used to characterize different products and establish appropriate actuation parameters. For all the surveyed nasal sprays and MDIs, the auto characterization method was used, with the force of contact set at 0.3 kg and force at end of stroke set at 6 kg. Actuation profiles for all the tested products were symmetric with actuation velocity of 50 mm/s, actuation acceleration of 4000 mm/s2, and hold time of 100 ms.
METHODS A Stable Micro Systems TA-XT.plus Texture Analyser (Texture Technologies Corp., Scarsdale, NY) equipped with a 750 g load cell and a TA-40 probe (10.0 cm diameter) was used to measure the impaction force. The load cell was sampled every 5 ms (200 data points per second) which allowed the generation of force versus time curves. The maximum value in the force versus time profile JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
Figure 1. Instrument setup for nasal spray (a) and MDI (b) impaction force measurements. DOI 10.1002/jps
DOI 10.1002/jps
API
Ipratropium bromide and albuterol sulfate Fluticasone propionate Fluticasone propionate Albuterol Albuterol Beclomethasone Dipropionate Albuterol
API
Beclomethasone dipropionate monohydrate Desmopressin acetate Fluticasone propionate Cromolyn sodium Flunisolide Mometasone furoate monohydrate None
Flovent HFA 44 Flovent HFA 220 Proventil Proventil_HFA Qvar (BDP HFA) Ventolin
Combivent
MDI
DDAVP Flonase NasalCrom Nasarel Nasonex Water
Beconase AQ
Nasal Spray
HFA HFA CFC HFA HFA CFC
CFC
Propellant
Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT GlaxoSmithKline, Research Triangle Park, NC GlaxoSmithKline, Research Triangle Park, NC Schering Corporation, Kenilworth, NJ Key Pharmaceuticals, Inc., Kenilworth, NJ IVAX Laboratories, Inc., Miami, FL Allen & Hanburys, Division of Glaxo, Inc., Research Triangle Park, NC
Manufacturer
4.6 3.5 5.3 3.0 3.8 4.4
4.9
Average Force (g)
6.5 6.1 4.4 1.5 1.8 5.2
3.7
Average Force (g)
Rhone-Poulenc Rorer Pharmaceuticals, Inc., Fort Washington, PA Allen & Hanburys, Division of Glaxo, Inc., Research Triangle Park, NC Pharmacia Consumer Healthcare, Peapack, NJ IVAX Laboratories, Inc., Miami, FL Schering Corporation, Kenilworth, NJ Pfeiffer pump filled with DI water
Glaxo Wellcome, Inc., Research Triangle Park, NC
Manufacturer
0.2 0.2 0.2 0.1 0.1 0.1
0.2
SD (g)
0.2 0.1 0.2 0.2 0.1 0.2
0.1
SD (g)
2.8 2.5 4.7 6.5 6.6 2.1
5.2
RSD (%)
4.1 2.9 3.2 5.6 2.6 4.4
2.9
RSD (%)
Table 1. Impaction Force of Several Commercially Available Nasal Spray and MDI Products at 3 and 6 cm Distance, Respectively, with Manual Activation (n ¼ 6)
IMPACTION FORCE OF NASAL SPRAY AND MDI
2801
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2802
GUO ET AL.
Figure 1 shows the instrument setup for impaction force measurement of nasal sprays and MDIs. For nasal spray impaction force measurements, the nasal spray pump was actuated upward using a SprayVIEW NSx automated actuator; For MDI impaction force measurements, the Texture Analyser was placed on its side with the MDI being discharged horizontally using a SprayVIEW MDx automated actuator. Six nasal sprays and seven MDI products were used as test products in this study. The product names and manufacturers are listed in Table 1. Pfeiffer (PFE) 0.10 mL nasal spray pumps (material number 62602, dip tube length 58 mm) and 20 mL bottles (material number 34473) filled with placebo solutions were used in both survey and DOE studies. The PFE nasal spray units were provided by Pfeiffer of America (Princeton, NJ) and are claimed to deliver 100 mL per actuation, equivalent to 100 mg water per actuation. DOE methodology was used to gain an understanding of the influence of actuation parameters and formulation properties on the impaction force of nasal sprays. Due to a limited ability to make different MDI formulations, we did not run a similar DOE study on MDIs. A 3-level, 4-factor Box–Behnken design was selected for the nasal spray study, because it can evaluate quadratic interactions between pairs of factors while minimizing the number of required experiments. JMP 5.1.1 software (SAS Institute, Cary, NC) was used to generate the DOE matrix and analyze the response surface models. The PFE nasal spray unit was filled with 18 mL placebo solutions prior to testing and the first six actuations were fired to waste as priming shots. The placebo formulations used in this study are aqueous solutions of extra-low-viscositygrade carboxymethylcellulose (CMC, Hercules, Inc., Wilmington, DE) and Tween-80 (Pharmacia, Piscataway, NJ) in different concentrations. CMC and Tween-80 are both common pharmaceutical excipients. CMC is normally used as an additive to increase solution viscosity, and Tween-80 is normally used as a surfactant to adjust the surface tension of solutions.
RESULTS Impaction Forces of Some Commercial Nasal Spray and MDI Products Figure 2 shows the impaction force versus time profiles from a PFE nasal spray pump (filled JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
Figure 2. Impaction force versus time profile from a nasal spray pump filled with water at 3 cm spray distance (a) and a Flovent1 HFA 44 mcg MDI at 6 cm spray distance (b).
with water) and a Flovent1 HFA 44 mcg MDI as examples. For the impaction force measurements performed in this study, signal-to-noise ratios were typically greater than 10. The baseline noise level could be due to building vibration. A higher noise level was observed for the nasal spray impaction force measurement than for the MDI, which can be explained by the different orientations of the instrument setup. For the nasal spray setup, the impaction plate is suspended such that it may be more sensitive to external vibration sources. For MDI’s, the apparatus lies on its side and external vibrations may be damped. In this study, impaction forces for nasal sprays and MDIs were measured from 3 and 6 cm from nozzle tip or mouth piece, respectively, where the maximum impaction forces were observed according to an earlier study.8 Table 1 shows the DOI 10.1002/jps
IMPACTION FORCE OF NASAL SPRAY AND MDI
measured impaction force for several nasal spray and MDI products. Optimization of actuation parameters for each nasal spray and MDI product was performed using the SprayVIEW NSx/MDx automated actuators. The relative standard deviation (RSD) for six replicates measurements of impaction force varied from 2.1% to 6.6% for all tested products, indicating good repeatability for the method. The impaction force of nasal spray products varied from 3.0 to 4.9 g, and those of MDI products ranged from 1.5 to 6.5 g. T-test was performed between each pair of means for nasal spray group and MDI group separately. Observed differences between the different products were statistically significant at the 95% confidence level, with the exception of DDAVP versus water in Pfeiffer pump, which suggests that impaction force is a discriminative parameter for in vitro testing of nasal spray and MDI products.
2803
Influence of Actuation Parameters and Formulation Properties on Nasal Spray Performance A 3-level, 4-factor Box–Behnken design was used for this study. Four factors, actuation stroke length (3.5, 4.4, 5.3 mm), actuation velocity (30, 50, 70 mm/s), concentration of CMC (0%, 1%, 2%), and concentration of Tween-80 (0%, 2.5%, 5%) were investigated for their influences on the impaction force of nasal spray at 3 cm distance. Selection of the range of these factors is based on previous studies.9,10 The concentration of CMC and Tween-80 are the dominant factors which influence the formulation viscosity and surface tension, respectively. Table 2 shows the resultant DOE input parameter and data table for the 3-level, 4-factor Box–Behnken design used here. Nine placebo formulations with different viscosities and surface tensions were prepared as aqueous solutions by
Table 2. DOE Input Parameter Table with Response for the 3-Level, 4-Factor Box–Behnken Design (Total of Nine Different Formulations) Experiment # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 DOI 10.1002/jps
Pattern
Stroke Length (mm)
Velocity (mm/s)
CMC (%)
Tween-80 (%)
Impaction Force (g)
00– 00 00 þ00 0þ0 00þ 00 00 þ0-0 0þ0 –00 þ00 0000 0000 0000 þ00 þþ00 0þ0 0þ0 þ0þ0 0þþ0 00þ 00þ 00þ þ00þ 0þ0þ 00þþ
4.4 4.4 3.5 5.3 4.4 4.4 4.4 3.5 5.3 4.4 3.5 5.3 4.4 4.4 4.4 3.5 5.3 4.4 3.5 5.3 4.4 4.4 4.4 3.5 5.3 4.4 4.4
50 30 50 50 70 50 30 50 50 70 30 30 50 50 50 70 70 30 50 50 70 50 30 50 50 70 50
0 1 1 1 1 2 0 0 0 0 1 1 1 1 1 1 1 2 2 2 2 0 1 1 1 1 2
0 0 0 0 0 0 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 5 5 5 5 5 5
3.6 2.1 4.3 5.3 4.4 4.5 2.5 4.7 5.1 5.2 2.3 2.4 4.6 4.6 4.7 5.3 6.7 2.3 4.4 5.0 5.2 4.7 2.4 4.6 5.2 5.3 4.7
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2804
GUO ET AL.
Table 3. The Coefficients of the Quadratic Model and Effect Tests for Impaction Force of Nasal Spray at 3 cm Distance Term
Coefficient Estimate STD Errors p-Value
Intercept S V C T SV SC VC ST VT CT SS VV CC TT
b0 b1 b2 b3 b4 b5 b6 b7 b8 b9 b10 b11 b12 b13 b14
4.63 0.34 1.51 0.03 0.23 0.33 0.05 0.05 0.10 0.15 0.23 0.34 0.81 0.09 0.19
0.18 0.09 0.09 0.09 0.09 0.16 0.16 0.16 0.16 0.16 0.16 0.14 0.14 0.14 0.14
<0.0001 0.0028 <0.0001 0.7891 0.0300 0.0626 0.7576 0.7576 0.5395 0.3621 0.1808 0.0300 <0.0001 0.5354 0.1966
employing different concentrations of CMC and Tween-80. A total of 27 experiments using the factor values described in Table 2 were performed, and responses were measured for each experiment. The coded design patterns shown in Table 3 symbolize the scaled factor values (high (þ), middle (0) and low ()) used in each run, in the order of stroke length, velocity, concentration of CMC and Tween-80, respectively. The measured impaction force (response) is also tabulated in Table 2. Each value is the average of four replicates. Scaled factors are traditionally used in a DOE prediction equation for a response from selected input parameters. A scaled factor (range 1 to þ1) is a parameter centered on the mean and scaled by parameter range/2. Fscaled ¼
2ðF Fcenter Þ ðFmax Fmin Þ
(1)
In Eq. (1), F is the factor value (experimental parameter), Fscaled is the scaled factor value, Fmax, Fmin, and Fcenter are the maximum, minimum, and center point parameter values used in the DOE. Factor scaling normalizes the influence of each factor on the prediction response. The scaled factors for the four factors used in this study are given by the following expressions. S¼
2 ðStroke Length 4:4Þ ð5:3 3:5Þ
(2)
2 ðVelocity 50Þ ð70 30Þ
(3)
V¼
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
C¼
T¼
2 ðCMC concentration 1%Þ ð2 0%Þ
2 ðTween80 concentration 2:5%Þ ð5 0%Þ
(4)
(5)
S, V, C, and T are the scaled factor values for stroke length, velocity, CMC concentration, and Tween-80 concentration, respectively. A nonlinear quadratic model of nasal spray impaction force generated by regression of the experimental factor values on the measured response is expressed as follows: R ¼ b0 þ b1 S þ b2 V þ b3 C þ b4 T þ b5 SV þ b6 SC þ b7 VC þ b8 ST þ b9 VT þ b10 CT þ b11 S2 þ b12 V 2 þ b13 C2 þ b14 T2
(6)
where R is the measured response, and the bis are the coefficients that estimate the contribution of each scaled parameter term to the response. Terms containing more than one factor represent interaction terms, and higher order terms indicate a quadratic (nonlinear) nature of relationship. A positive sign for R indicates a synergistic effect, whereas a negative sign represents an antagonistic effect. Because the parameter values are scaled, the magnitudes of the coefficients reflect the importance of the term to the measured response. The regression coefficients and effect test results of the quadratic DOE model (Eq. 6) are presented in Table 3. The effect tests show that, statistically, the measured impaction force of the nasal spray delivery is significantly affected by stroke length ( p ¼ 0.0028), actuation velocity ( p < 0.0001), and concentration of Tween-80 ( p ¼ 0.03) independently, with no significant interaction effects between pairs of the four factors ( p > 0.05 for all interaction terms). However, the scaled estimates indicate that actuation velocity is the dominant factor influencing impaction force, while the actuation stroke length and concentration of Tween-80 have relatively smaller influences on impaction force. Surprisingly, the concentration of gelling agent, which has significant effects on many other nasal spray characteristics, such as spray pattern, plume geometry and DSD,11 has little, if any, influence on impaction force. The positive signs of S terms indicate that increasing stroke length will lead to the production of a larger impaction force. For the V and DOI 10.1002/jps
IMPACTION FORCE OF NASAL SPRAY AND MDI
2805
T terms, the scaled estimates of the first-order terms have positive signs, while the ones of the second-order terms have negative signs. Since the scaled factor values are between 1 and 1 in most of situations, the first-order term has more influence than the second-order term, thus increasing actuation velocity and/or concentration of surfactant will also lead a larger impaction force. By eliminating all insignificant terms from Eq. (6), an optimized impaction force model (shown in Eq. 7 and Tab. 4) was recalculated using JMP. R ¼ 4:45 þ 0:34S þ 1:51V þ 0:23T þ 0:41S2 0:74V 2
(7)
Figure 3 shows a plot of actual versus predicted impaction force from the simplified quadratic model. The quadratic model shows good fit with correlation coefficient (R2) of 0.94. The DOE study helps to identify the source of variability in nasal spray product performance, thus give us a better understanding of how to control the variability. For example, stronger plume with higher impaction force may be obtained by increasing the formulation surface tension and/or via a selection of a pump with a longer stroke length. Moreover, the quadratic models developed from the DOE study quantitatively describe the inherent relationships between the factors and response, which will assist the product design to achieve desired impaction force.
CONCLUSIONS We have developed a method to measure impaction force for inhalation drug products (nasal sprays and MDIs). This new technique exhibits Table 4. The Coefficients of the Simplified Quadratic Model and Effect Tests for Impaction Force of Nasal Spray at 3 cm Distance Term
Coefficient Estimate STD Errors p-Value
Intercept S V T SS VV DOI 10.1002/jps
b0 b1 b2 b4 b11 b12
4.45 0.34 1.51 0.23 0.41 0.74
0.11 0.09 0.09 0.09 0.13 0.13
<0.0001 0.0014 <0.0001 0.0246 0.0044 <0.0001
Figure 3. Correlation plots for predicted and actual impaction force for nasal sprays, where the horizontal broken line is the mean of the Y variable, the solid line is the line of fit, and the two broken curves describe the confidence region where the test is significant at the 95% level relative to the line of fit. (RSq: R2, correlation coefficient; RMSE: root mean square error.)
good repeatability as evidenced by a low RSD for repeated measurements. A survey of several commercial nasal spray and MDI products shows that spray impaction forces for these products differ significantly from each other over a range of 1.5–6.5 g, which indicates that impaction force can be used as a discriminating parameter for in vitro testing for these product types. The DOE study demonstrates that the impaction force is influenced by both formulation properties and actuation parameters, especially the actuation velocity. An optimized quadratic model generated from the DOE results shows good predictive power for impaction force. The distance from a nasal spray nozzle tip or MDI mouth piece to the local delivery site may differ significantly from person to person, and is therefore unrelated to the distance at which the device develops highest maximum impaction force. However, the highest maximum impaction force and its corresponding distance, or maximum impaction forces at 2 or more distances, could be used for quality control or comparison purposes. Since impaction force is more directly related to patient sensation and aerosol deposition than other more traditional, parameters, it may provide a better way to evaluate in vitro equivaJOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
2806
GUO ET AL.
lence in support of abbreviated new drug applications (ANDAs) for orally inhaled and nasal drug products. Due to its decent discriminating power, this technique may also provide a good evaluation of manufacturing changes.
REFERENCES 1. U.S. Food and Drug Administration. 1998. Draft Guidance for Industry: Metered Dose Inhaler (MDI) and Dry Powder Inhaler (DPI) Drug Products. 2. U.S. Food and Drug Administration. 2002. Guidance for Industry: Nasal Spray and Inhalation Solution, Suspension, and Spray Drug Products—Chemistry, Manufacturing, and Controls Documentation. 3. U.S. Food and Drug Administration. 2003. Draft Guidance for Industry: Bioavailability and Bioequivalence Studies for Nasal Aerosols and Nasal Sprays for Local Action. 4. Newman SP, Pavia D, Moren F, Sheahan NF, Clarke SW. 1981. Deposition of pressurised aerosols in the human respiratory tract. Thorax 36:52–55. 5. Crompton GK. 1982. Problems patients have using pressurized aerosol inhalers. Eur J Respir Dis Suppl 119:57–65.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 8, AUGUST 2009
6. Pedersen S, Frost L, Arnfred T. 1986. Errors in inhalation technique and efficiency in inhaler use in asthmatic children. Allergy 41:118– 124. 7. Gabrio BJ, Stein SW, Velasquez DJ. 1999. A new method to evaluate plume characteristics of hydrofluoroalkane and chlorofluorocarbon metered dose inhalers. Int J Pharm 186:3–12. 8. Guo C, Doub WH. 2006. Development of a novel technology to measure impaction force of nasal sprays and metered dose inhalers using the Texture Analyser. In: Dalby RN, Byron PR, Peart J, Suman JD, Farr SJ, editors. Respiratory drug delivery, Vol. 2. 2006. River Grove, IL: Davis Healthcare International Publishing. pp 621– 624. 9. Guo C, Doub WH. 2006. The influence of actuation parameters on in vitro testing of nasal spray products. J Pharm Sci 95:2029–2040. 10. Dayal P, Shaik MS, Singh M. 2004. Evaluation of different parameters that affect droplet-size distribution from nasal sprays using the Malvern Spraytec. J Pharm Sci 93:1725–1742. 11. Guo C, Stein KJ, Kauffman JF, Doub WH. 2008. Assessment of the influence factors on in vitro testing of nasal sprays using Box–Behnken experimental design. Euro J Pharm Sci 35:417–426.
DOI 10.1002/jps