`
`OPEN ACCESS
`
`
`
`pharmaceutics
`
`ISSN 1999-4923
`www.mdpi.com/journal/pharmaceutics
`
`Review
`Analytical Challenges and Regulatory Requirements for
`Nasal Drug Products in Europe and the U.S.
`
`Sabrina Trows 1, Klaus Wuchner 2, Rene Spycher 2 and Hartwig Steckel 1,*
`
`1 Department of Pharmaceutics and Biopharmaceutics, Kiel University, Grasweg 9a, 24118 Kiel,
`Germany; E-Mail: strows@pharmazie.uni-kiel.de
`2 Janssen Pharmaceuticals, Pharmaceutical Development & Manufacturing Sciences,
`Johnson & Johnson, Hochstrasse 201, 8205 Schaffhausen, Switzerland;
`E-Mails: kwuchne1@its.jnj.com (K.W.); rspycher@its.jnj.com (R.S.)
`
`* Author to whom correspondence should be addressed; E-Mail: hsteckel@pharmazie.uni-kiel.de;
`Tel.: +49-431-880-1330; Fax: +49-431-880-1352.
`
`Received: 20 December 2013; in revised form: 24 March 2014 / Accepted: 31 March 2014 /
`Published: 11 April 2014
`
`
`Abstract: Nasal drug delivery can be assessed by a variety of means and regulatory
`agencies, e.g., the Food and Drug Administration (FDA) and the European Medicines
`Agency (EMA) have published a set of guidelines and regulations proposing in vitro test
`methods for the characterization of nasal drug products. This article gives a summary of
`the FDA and EMA requirements regarding the determination of droplet size distribution
`(DSD), plume geometry, spray pattern and shot weights of solution nasal sprays and
`discusses the analytical challenges that can occur when performing these measurements.
`In order to support findings from the literature, studies were performed using a standard
`nasal spray pump and aqueous model formulations. The aim was to identify possible
`method-, device- and formulation-dependent influencing factors. The literature review, as
`well as the results from the studies show that DSD, plume geometry and spray pattern are
`influenced by, e.g., the viscosity of the solution, the design of the device and the actuation
`parameters, particularly the stroke length, actuation velocity and actuation force.
`The dominant factor influencing shot weights, however, is the adjustment of the actuation
`parameters, especially stroke length and actuation velocity. Consequently, for routine
`measurements assuring, e.g., the quality of a solution nasal spray or, for in vitro
`bioequivalence studies, the critical parameters, have to be identified and considered in
`method development in order to obtain reproducible and reliable results.
`
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 1 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`
`Keywords: nasal drug delivery; regulatory aspects; test methods; nasal sprays
`
`196
`
`
`
`1. Introduction
`
`The most prominent way of intranasal drug delivery is the administration of locally acting drugs in
`order to treat nasal congestion, infections and allergic rhinitis [1]. However, the nasal route can also be
`used for the systemic delivery of drugs for the therapy of various diseases, like osteoporosis and
`migraine, as well as for pain management and also for the administration of vaccines [2]. It is a
`painless, non-invasive delivery route, resulting in a rapid drug onset of action, due to the high
`vascularization of the nose and high permeability of the nasal mucosa under avoidance of first pass
`metabolism [3]. These advantages lead to high patient convenience and compliance.
`For nasal drug delivery, there are several dosage forms available. The most popular examples are
`nasal sprays and nasal drops for which the drug can be formulated as a solution or suspension.
`Alternative dosage forms are the pressurized nasal aerosols and nasal powders. Typically, aqueous
`nasal spray formulations contain the drug, as well as bioadhesive polymers, surfactants, tonicity agents
`and, in some cases, penetration enhancers [4]. Bioadhesive polymers, like sodium carboxymethyl
`cellulose, are often used to increase the viscosity of the formulation in order to stabilize the suspension
`or to increase the residence time in the nasal cavity to modify drug absorption [4,5]. Surfactants
`can be included in the formulation to solubilize the drug in case of poor solubility or to increase the
`wettability [6].
`Besides the formulation, also the delivery device plays an important role in nasal drug delivery, and
`only the combination of both, device and formulation, determines the properties of the final nasal drug
`product. This makes the development of nasal drug products more complex, since the variability of the
`formulation and the device have to be taken into account [5]. Therefore, the analytical requirements
`for the approval of nasal drug products exceed those for solid dosage forms [7]. For the in vitro
`characterization of nasal drug products in the development phase, as well as for quality control and
`bioavailability/bioequivalence studies, regulatory agencies, like the Food and Drug Administration
`(FDA) and the European Medicines Agency (EMA), have published guidelines and regulations
`proposing various test methods [8–11]. Tables 1 and 2 give a summary of the recommended tests for
`the different nasal drug products. However, in order to obtain reliable results, the test methods need to
`be validated, and in this context, it is essential to know the factors that can influence the
`measurements. In some studies, it could be shown that the spray characteristics can be influenced by
`the design of the device, by the formulation properties, like viscosity and surface tension, and by the
`handling of the device, i.e., the actuation parameters [5,12–18]. Additionally, the selected technique
`and the set-up of the measurements can also have an effect on the results and have to be considered
`during method development.
`
`
`
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 2 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`
`197
`
`Table 1. Tests recommended for the finished drug product specification by the Food and
`Drug Administration (FDA) and the European Medicines Agency (EMA) (standard quality
`tests are not listed).
`
`Test
`
`Pressurized
`metered dose
`nasal sprays
`Specifications for the drug product
`yes #
`Pump/valve delivery
`Delivered dose/content
`uniformity
`Dose content uniformity
`through container life
`Content
`uniformity/uniformity of
`dosage units
`
`
`
`yes #
`
`no *
`
`Mean delivered dose
`
`Spray pattern
`Particle/droplet size
`distribution
`Particle size distribution
`of API
`Microscopic evaluation
`Particulate matter
`Microbial limits
`Preservative content
`Preservatives and stabilizing
`excipients assay
`
`Sterility
`
`Net content/minimum fill
`Number of actuations
`per container
`Weight loss (stability)
`Leachables (stability)
`Osmolality
`Viscosity
`Appearance and color of
`content and container
`closure system
`Water or moisture content
`
`Dehydrated alcohol content
`
`Leak rate
`
`
`
`yes *
`
`yes #
`
`yes
`
`yes, for suspensions #
`
`yes, for suspensions #
`yes #
`yes
`no *
`
`
`
`no *
`
`yes #
`
`yes *
`
`
`yes #
`
`
`
`yes #
`
`yes
`yes, if used as a
`cosolvent #
`yes
`
`Nasal
`powders
`
`Single and
`multiple use
`nasal drops
`
`yes, for multiple
`use drops
`
`
`
`Single and
`multiple use
`nasal sprays
`
`yes #
`yes, for multiple
`use sprays
`
`yes #
`
`yes, for single use
`drops *
`
`yes, for single use
`sprays *
`
`yes, for multiple use
`drops *
`
`yes, for multiple
`use sprays *
`yes #
`
`yes
`
`
`
`no *
`
`yes *
`
`yes
`
`no
`
`yes
`
`
`
`
`
`yes
`no *
`
`
`
`no *
`
`yes *
`
`
`
`
`
`
`
`
`yes
`
`
`
`no
`
`
`
`
`
`yes
`yes, if present *
`
`
`
`yes, if product is
`sterile *
`
`
`
`
`
`
`
`
`
`
`no
`
`
`
`no
`
`yes, for
`suspensions #
`
`yes #
`yes
`yes, if present *
`
`yes #
`
`yes, if product is
`sterile *
`yes #
`yes, for multiple
`use sprays *
`yes #
`yes #
`yes #
`yes #
`
`
`
`no
`
`
`
`no
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 3 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`
`Table 1. Cont.
`
`198
`
`Test
`
`Nasal
`powders
`
`Single and
`multiple use
`nasal drops
`
`Single and
`multiple use
`nasal sprays
`
`Pressurized
`metered dose
`nasal sprays
`Specifications for the drug product
`yes, if cosolvent or more
`than one propellant is used #
`Explanatory note: “yes”, the test is recommended for the particular drug product; “no”, the particular drug
`product is excluded from the test; blank, no specific details in the guidelines are available; # FDA only
`requirement; * EMA only requirement; API, active pharmaceutical ingredient.
`
`Pressure testing
`
`
`
`
`
`
`
`This article gives an overview of the regulatory requirements regarding the determination of droplet
`size distribution (DSD), plume geometry, spray pattern and shot weights. These tests, among others,
`are required in development and in vitro bioequivalence studies, as well as in quality control matters.
`In addition to the regulatory requirements, analytical challenges and possible influencing factors
`related to the device, formulation composition and selected method/technique that affect nasal spray
`characteristics are reviewed. In order to support findings from the literature, studies comprising the
`determination of DSD, plume geometry, spray pattern and shot weights were performed using model
`formulations and a standard nasal spray pump.
`
`Table 2. Tests recommended for nasal drug product characterization/development studies
`by the FDA and the EMA.
`Pressurized
`metered dose
`nasal sprays
`Drug product characterization/development studies
`
`Nasal
`powders
`
`Single and
`multiple use
`nasal drops
`
`Single and
`multiple use
`nasal sprays
`
`Test
`
`Physical characterization
`
`yes, for suspensions * yes *
`
`yes, for
`suspensions *
`
`yes, for suspensions *
`
`Priming and repriming
`(in various orientations)
`Plume geometry
`Microscopic evaluation
`Effect of resting time
`
`yes
`
`no
`
`no
`
`yes #
`yes, for suspensions #
`yes #
`
`
`
`
`
`Shaking requirements
`
`yes, for suspensions
`
`no
`
`Minimum fill justification
`Extractables/leachables
`Performance after
`temperature cycling
`Effect of environmental
`moisture
`
`Cleaning instructions
`
`
`
`
`yes *
`yes *
`
`yes
`
`yes *
`
`yes
`
`
`yes *
`no *
`
`no
`
`yes *
`
`yes
`
`
`
`
`yes, for
`suspensions
`yes *
`yes *
`
`no
`
`no *
`
`yes
`
`yes #
`
`
`
`yes, for suspensions
`
`yes *
`yes *
`
`yes
`
`no *
`
`yes, for multiple
`use drops
`
`yes, for multiple
`use sprays
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 4 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`
`Table 2. Cont.
`
`199
`
`Test
`
`Pressurized
`metered dose
`nasal sprays
`Drug product characterization/development studies
`Device robustness
`yes
`Profiling of sprays near
`container exhaustion
`(tail off characteristics)
`Delivered dose uniformity
`through container life
`
`yes #
`
`yes *
`
`Effect of storage on PSD
`
`yes, for
`suspensions #
`
`Particle/droplet size distribution yes
`
`Preservative effectiveness
`(and sterility maintenance)
`
`no
`
`Photostability
`
`yes #
`
`yes #
`
`yes
`yes #
`
`yes, if drug
`is exposed
`to light #
`Actuator/mouthpiece deposition yes
`Determination of appropriate
`storage conditions
`Stability of primary
`(unprotected) package
`Delivery device development
`Microbial challenge
`Effect of dosing orientation
`
`Nasal
`powders
`
`Single and
`multiple use
`nasal drops
`
`Single and
`multiple use
`nasal sprays
`
`yes
`
`
`
`yes *
`
`
`
`yes
`
`no
`
`yes
`
`
`
`yes
`
`yes #
`
`yes, for multiple
`use drops *
`
`
`
`no
`
`yes, for multiple
`use sprays *
`yes, for
`suspensions #
`yes, for multiple
`use sprays
`
`yes, if present
`
`yes, if present
`
`yes, if drug
`is exposed
`to light #
`yes
`
`yes, if drug
`is exposed
`to light #
`no
`
`yes, if drug
`is exposed
`to light #
`yes *
`
`
`
`
`yes
`
`
`
`
`yes
`
`
`
`yes #
`
`yes
`
`In vitro dose proportionality
`
`yes, for suspensions
`in multiple
`strengths #
`
`
`
`
`
`yes #
`yes, for
`suspensions in
`multiple
`strengths #
`no *
`no *
`no *
`yes *
`Low temperature performance
`Explanatory note: “yes”, the test is recommended for the particular drug product; “no”, the particular drug
`product is excluded from the test; blank, no specific details in the guidelines are available; # FDA only
`requirement; * EMA only requirement; and PSD, particle size distribution.
`
`2. Experimental Section
`
`2.1. Materials
`
`Mechanical nasal spray pumps delivering 100 (cid:541)L of formulation per actuation were provided by
`Aptar (Radolfzell, Germany). Water was used
`in double-distilled quality (FinnAqua 75,
`San Asalo-Sohlberg Corp., Helsinki, Finland). Sodium carboxymethyl cellulose (Tylopur C 30 G) was
`obtained from Clariant (Muttenz, Switzerland) and polysorbate 80 from Uniqema (Snaith, UK).
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 5 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`2.2. Model Formulations
`
`200
`
`The basic formulation was water. For investigating formulation-dependent variables, the viscosity
`was varied by adding 1%–5% sodium carboxymethyl cellulose (NaCMC) and the surface tension,
`respectively, by the addition of 0.0001%–0.1% polysorbate 80.
`
`2.3. Determination of Viscosity and Surface Tension
`
`Viscosity was measured using a Vibro viscosimeter (A&D Company Ltd., Tokyo, Japan) at room
`temperature, and the surface tension of the water and model formulations was determined using a plate
`tensiometer (Processor Tensiometer K 12, Krüss GmbH, Hamburg, Germany).
`
`2.4. Determination of Droplet Size Distribution
`
`The droplet size distribution (DSD) was determined by laser diffraction using HELOS with
`SPRAYER-module and ROTOR, as well as the force and traject actuator, respectively (Sympatec
`GmbH, Clausthal-Zellerfeld, Germany). The spraying angle was varied between 0° and 90°; the
`actuation force between 20 and 100 N. The distance to the measuring zone ranged from 3 to 7 cm, and
`the stroke length was set between 1 and 7 mm. Time-resolved measurements were performed, and data
`were analyzed according to the Fraunhofer theory. All determinations were performed in triplicate.
`
`2.5. Determination of Plume Geometry
`
`For the determination of plume geometry, an Imager E-lite CCD-camera (charge-coupled device
`camera) and sheet light (LaVision, Göttingen, Germany) were used. The images were corrected for
`distortion, due to the skewed camera perspective, and plume angle was determined manually using
`CorelDraw X6 software (Corel, Ottawa, ON, Canada).
`
`2.6. Determination of Shot Weights/Validation of Pump Delivery
`
`In order to determine the shot weights, the nasal sprays were filled with 10.0 mL of the respective
`formulation. The device was actuated with an automated actuator (SPRAYER-module, Sympatec), and
`after each actuation, the device was weighed on an analytical balance (A 200 S, Sartorius, Göttingen,
`Germany) to determine the delivered mass. The actuation parameters were set as follows: the actuation
`force ranged from 40 to 100 N, and the stroke length was varied between 1 and 7 mm.
`
`3. Results and Discussion
`
`3.1. Droplet Size Distribution
`
`The DSD of a nasal spray is a critical parameter, since it significantly influences the in vivo
`deposition of the drug in the nasal cavity [19]. The droplet size is hereby mainly influenced by the
`design and handling, e.g., the actuation parameters, of the device, as well as by the formulation, and
`the prevalent median droplet size is between 30 and 120 (cid:541)m [20]. If the droplets are too large
`(>120 (cid:541)m), deposition takes place mainly in the anterior parts of the nose, and if the droplets are too
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 6 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`small (<10 (cid:541)m), they can possibly be inhaled and reach the lungs [4,20], which should be avoided
`because of safety reasons.
`
`201
`
`3.1.1. Regulatory Aspects
`
`In order to determine the DSD of a nasal spray, the FDA and the EMA [8,9,11] recommend making
`use of laser diffraction, which has already become the standard technique in the industry for droplet
`and particle size analysis [4]. Laser diffraction is a fast and efficient method that measures the
`geometric size of droplets and particles in real-time based on two common light scattering principles,
`which are Mie- or Fraunhofer-theory [19,21]. When determining
`the DSD,
`time-resolved
`measurements should be performed, i.e., the droplet size and obscuration or transmission are recorded
`at defined time intervals, e.g., every 1 ms, over the entire spray event. On the basis of time history
`profiles (obscuration/DSD versus time), the spray event can then be characterized by three distinct
`phases: the formation phase, which is indicated by a rapid increase in obscuration and a decrease in
`droplet size, followed by the fully developed phase, where obscuration and droplet size attain a
`plateau, and, finally, the dissipation phase, designated by a rapid decrease in obscuration and an
`increase in droplet size [9,20]. For in vitro equivalence purposes, the FDA recommends determining
`the time history profiles of droplet sizes and obscuration over the complete life of the single spray at
`two distances ranging from 2 to 7 cm from the nozzle tip, with the two distances separated by 3 cm or
`more [9]. For new drug applications (NDAs), only one distance within this range is requested [8,21].
`In both cases, the data to report should be collected only during the fully developed phase and should
`comprise the droplet size expressed as D10, D50 and D90, as well as the span defined as (D90 (cid:237) D10)/D50
`as an indicator for the width of the distribution and, for NDAs, additionally, the fraction of droplets
`smaller than 10 μm [8,9]. It has to be defined and stated by the applicant at which region of the plateau
`phase the droplet size data were determined. For this, the FDA suggests three different variants of data
`analysis: “the average of all scans over the entire plateau, the data of a single scan only at the
`maximum obscuration or the average of a specified range of scans around this obscuration” [9].
`Additionally, the FDA recommends using an automated actuator to minimize variability.
`In contrast to the FDA, the EMA gives less information about the measurement of the droplet size
`distribution of nasal sprays in their “guideline on the pharmaceutical quality of inhalation and nasal
`products”, e.g., how to analyze the data. However, the agency also requests that limits for the median
`droplet size and the fraction <10 μm should be given [11].
`
`3.1.2. Analytical Aspects
`
`The droplet size distribution can be influenced by various factors, which have to be considered
`when performing droplet size measurements. Those influencing factors comprise formulation-related
`properties, like viscosity and surface tension, the design of the device, the actuation parameters, e.g.,
`actuation force, stroke length, actuation velocity, but also method-dependent variables, like the
`spraying angle or the distance between the nozzle and the laser beam.
`
`
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 7 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`3.1.2.1. Formulation Dependent Variables
`
`202
`
`The formulation of a nasal spray, among other factors, plays an important role in nasal drug
`delivery. In order to improve nasal drug delivery, some solution-based formulations contain
`bioadhesive polymers to increase the residence time of the formulation on the mucosa and to control
`drug absorption. The addition of such polymers also leads to an increase in viscosity, which
`consequently has an impact on droplet size distribution and drug deposition in the nasal cavity [18].
`Harris et al. [18] have studied the influence of the addition of methylcellulose to a solution containing
`desmopressin and noticed an increase in droplet size with an increasing amount of polymer from
`53 (water) to 200 μm (0.5% methylcellulose). Dayal et al. [14] have investigated the influence of the
`addition of NaCMC and Carbopol, respectively, on the droplet size distribution of an aqueous nasal
`spray in comparison to water. For solutions with NaCMC, they observed an increase in droplet size
`with an increasing amount of polymer for different spray pumps, whereas the D90 underwent a more
`significant change than the D10. Additionally, they could show a linear relationship between the
`viscosity and the corresponding D50. For Carbopol solutions, Dayal et al. also reported an overall
`increase in droplet size with an increasing amount of polymer compared to water, but the effect was
`less pronounced than for the solutions with NaCMC. Moreover, the difference in droplet sizes between
`the two tested Carbopol solutions was very small. The difference between the two polymers can be
`explained by their differing rheological behavior: while the carboxymethyl cellulose (CMC) solution
`behaved like a Newtonian fluid at the low concentration used, the Carbopol solutions exhibited a
`shear-thinning behavior, and hence, the viscosity is reduced when the liquid is atomized by the spray
`pump, leading to smaller droplet sizes than initially expected. Pennington et al. [22] have also
`measured the DSD of solutions containing Newtonian viscosity modifiers, namely polyethylene glycol
`(PEG), propylene glycol (PG) and glycerin in different concentrations. They also discovered that with
`increasing amount of modifier, the viscosity and the droplet size increases, but they did not detect a
`linear, but a logarithmic relationship between those parameters. The fact that an increase in viscosity
`leads to an increase in droplet size was also evaluated statistically using a Box–Behnken experimental
`design by Guo et al. [5]. They also investigated aqueous solutions of NaCMC and showed that
`viscosity is a statistically significant factor influencing DSD. A Box–Behnken experimental design
`was also used by Dayal et al. [13] in order to evaluate the effect of the surface-active polymers,
`hydroxyethylcellulose (HEC) and polyethylene-oxide (PEO), in combination with ionic excipients
`(sodium chloride and calcium chloride) on the viscosity and DSD of nasal sprays. They found that the
`viscosity of the solutions was increased with an increasing amount of polymer, showing a slight shear
`thinning behavior, and that the addition of ionic excipients significantly affected the viscosity of
`the polymer solutions. For HEC and PEO solutions, Dayal et al. reported an increase in D50 in a
`concentration-dependent manner, with PEO showing larger droplets than HEC, as well as an
`interaction of polymer and electrolytes, resulting in different DSD, due to changes in viscosity.
`Alongside viscosity modifiers and other excipients nasal spray formulations can also contain
`surfactants, e.g., to solubilize the active pharmaceutical ingredient (API) or to enhance drug
`absorption. However, surfactants also have an impact on the surface tension, which can possibly
`influence DSD. Two groups [5,14] have studied the influence of polysorbate 80 on the droplet size of
`nasal sprays. Dayal et al. showed that an increase in surfactant resulted in a decrease in droplet size,
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 8 of 25
`
`
`
`203
`
`Pharmaceutics 2014, 6
`
`but this effect was very slight. Guo et al., however, could not detect a significant influence of the
`surfactant concentration on the DSD. In our own studies, we also investigated the influence of
`polysorbate 80 on DSD. The surfactant was varied between 0.0001% and 0.1%, resulting in surface
`tensions between 66.5 and 38.4 mN/m. Water showing a surface tension of 72.6 mN/m was also
`measured as a comparison. The results are shown in Figure 1. There is no significant influence on the
`D10, D50 and D90, which supports the findings by Guo et al., and only a slight effect on the span was
`observed, which shows some fluctuations.
`Generally, it can be said that the viscosity has a major influence on the DSD, leading to an increase
`of the droplet size of the nasal sprays, whereas the surface tension only has a minor to no effect
`on DSD.
`
`Figure 1. The effect of surface tension on the droplet size distribution (DSD) and span
`measured at 7 cm from the nozzle and an actuation force of 60 N. Results are presented as
`the average ± SD of three actuations.
`
`2.00
`1.90
`1.80
`1.70
`1.60
`1.50
`1.40
`1.30
`1.20
`1.10
`1.00
`
`span
`
`x10
`
`x50
`
`x90
`
`span
`
`60.0
`
`50.0
`
`40.0
`
`30.0
`
`20.0
`
`10.0
`
`0.0
`
`droplet size, μm
`
`80.0
`
`70.0
`
`60.0
`50.0
`40.0
`surface tension, mN/m
`
`30.0
`
`
`
`3.1.2.2. Device-Dependent Variables
`
`Multi-dose nasal spray pumps, meanwhile, are available from a variety of different manufacturers.
`Essentially, they are all based on the same principle, i.e., the dose is divided in a metering chamber and
`forced through a nozzle to disperse the liquid into fine droplets. However, mechanical spray pumps can
`differ in spray performance, due to modifications of the swirl chamber and inlet channels, altering the
`dimensions and geometry of the orifice diameter or differences in the pressure that build up in the
`volume chamber prior to dispensing [23].
`Suman et al. have compared two different nasal spray pumps regarding droplet size. In one study,
`they found statistical differences in D50 values at varied distances between the nozzle and laser
`beam [24]. In a second study, they compared nasal spray pumps differing in their mechanical operation
`and could not detect differences in D10 and D50 values, but significant differences in D90 [15]. In this
`case, the droplet sizes differed by more than 13%. Dayal et al. [14] have also evaluated the influence
`of pump design on the DSD and concluded that the nozzle orifice has an impact on the droplet size that
`is emitted, since alterations in the diameter, shape and length will affect the compression forces of the
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 9 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`liquid, friction and spray velocity. However, in a statistical evaluation, they stated that changes in the
`formulation have a greater impact on the D50 than the design of the device.
`
`204
`
`3.1.2.3. Method- and Actuation-Dependent Variables
`
`The handling of a nasal spray device, i.e., the actuation parameters, like stroke length, actuation
`force, actuation velocity or hold time, as well as the set-up of the measurements (time point, distance
`to measuring zone, spraying angle) can significantly influence the DSD and other in vitro tests.
`Generally, correct actuation parameters have to be determined for a selected device [4] and should
`mimic hand actuation [25].
`As described in Section 3.1.1, the FDA suggests determining the DSD during the fully developed
`phase of the spray event, which leads to the highest degree of reproducibility [26] and most stable
`droplet size values [27]. Choosing an inaccurate time point can result in an under- or over-estimation
`of the DSD, and presently, the FDA does not give recommendations on how to determine the stable
`phase. Guo and Doub [17] conclude that the obscuration thresholds should be defined as close to the
`maximum obscuration as possible in order to avoid extreme droplets, which are present at the
`beginning and end of the spray event. They set their lower threshold to 90% of the maximum
`obscuration in order to define the fully developed phase. Eck et al. [26], however, defined the stable
`phase at 25%–30% absolute obscuration and determined the droplet size at the maximum obscuration.
`This is only possible when there are no fluctuations present in the time-history plot; otherwise, the data
`have to be averaged over the entire plateau of the spray [14]. It has also to be taken into account that
`obscuration values vary with changes in the actuation parameters, the device and the distance to the
`measuring zone [14,17], and hence, obscuration thresholds have to be defined for every selected device
`and measurement set-up.
`For the measurement set-up, the FDA requests to select one or two distances for in vitro
`equivalence purposes in the range of 2 to 7 cm (refer to Section 3.1.1). It is known that the distance
`between the nozzle and the laser beam affects the DSD measurement, due to different settling
`velocities of the droplets, the plume dynamics and the varied representation of the true DSD in the
`measurement zone [14,17,26]. In the literature, there are controversial findings regarding the influence
`of the distance reported. Eck et al. [26] determined the DSD of a commercial nasal spray solution at 1,
`2.5 and 5 cm and found that an increase in distance led to a decrease in D10, D50 and D90 values for the
`stable phase. In a study by Guo and Doub using water as a model formulation [17], the actuation
`distance was varied between 1 and 9 cm, and they also observed significant differences in D50. In the
`range of 1 to 3 cm, the D50 decreased, showing the smallest value at 3 cm, and then increased with
`increasing distance, showing the largest overall value at 9 cm. In our own studies, we have varied the
`distance to the measuring zone between 3 and 7 cm and observed an increase in D10 and D50, as well as
`a decrease in span with increasing actuation distance (Figure 2a) using water as a model formulation.
`The reason might be that the determination at higher distances provides more time for the plume to
`develop and, hence, smaller droplets are able to coalesce and form larger droplets, which leads to an
`overall increase in the droplet size. Dayal et al. [14] have performed measurements at a distance of 1.5,
`3 and 6 cm using commercial nasal sprays and reported a decrease in DSD with increasing distance to
`the measuring zone. This was explained by the assumption that with greater distances, part of the
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 10 of 25
`
`
`
`Pharmaceutics 2014, 6
`
`droplets escape from the measuring zone, resulting in a larger percentage of droplets, and,
`consequently, the measured data is not representative of the entire DSD of the spray. Therefore, Dayal
`et al. suggest that actuation at a short distance may provide a better representation of DSD. However,
`measurements at a short distance may lead to multiple scattering, due to the high density of droplets in
`the measuring zone, which can result in an underestimation of droplet size and, consequently, a
`distance has to be chosen based on the obscuration levels that reduce multiple scattering events [14].
`
`205
`
`Figure 2. The effect of distance to the measuring zone (a) and spraying angle (b) on the
`DSD and span of the model formulation (water) measured at an actuation force of 60 N.
`The results represent the average ± SD of three actuations.
`
`1.6
`
`1.4
`
`.2
`
`11
`
`x10
`
`x50
`
`x90
`
`span
`
`span
`
`0.8
`
`0.6
`
`0.4
`
`.2
`
`00
`
`10
`
`30
`
`70
`
`90
`
`50
`angle, °
`(b)
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`
`droplet size, μm
`
`2.0
`1.8
`1.6
`1.4
`1.2
`1.0
`0.8
`0.6
`0.4
`0.2
`0.0
`
`span
`
`8
`
`x10
`
`x50
`
`x90
`
`span
`
`4
`
`6
`distance, cm
`(a)
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`
`droplet size, μm
`
`2
`
`A second set-up parameter for DSD measurements besides the actuation distance is the spraying
`angle. Most routine measurements are performed at a fixed angle, which has to be defined, but in order
`to determine, e.g., a position (in)dependency of DSD, determinations should be performed at a variety
`of angles. In this study, the position angle of the nasal spray was varied between 0° (horizontal
`set-up) and 90° (upright position) to determine an adequate angle for routine measurements and to
`evaluate the position dependency of the DSD. For the 0° and 10° angles, no data could be obtained,
`since no dose was released from the device after actuation. However, the results for measurements at
`angles between 20° and 90° show that there is no significant change in DSD and span with varying
`spraying angles (Figure 2b). Consequently, the DSD generated by this particular device is not
`dependent on the position of the nasal spray once a dose is metered. For routine measurements, an
`angle of 70°–80° is suggested, since obscuration values are more stable in this region than for smaller
`angles (data not shown), and this range comes close to in-use conditions when patients administer a
`nasal spray.
`As mentioned before, actuation parameters can also influence DSD and other spray characteristics.
`Consequently, it is crucial to know their influence and to select them carefully for every device
`considering the age (pediatric, adult, geriatric settings) and gender of the target group [4,25].
`For in vitro tests, automated actuation systems should be used in order to control actuation, to increase
`the reproducibility of measurements and to minimize operator bias. More information about automated
`actuation systems can be found by Guo and Doub [17] and Dayal et al. [14]. In a study, Guo and Doub
`have evaluated the influence of several actuation parameters on the DSD and found out that the hold
`
`
`
`Nalox1053
`Nalox-1 Pharmaceuticals, LLC
`Page 11 of 25
`
`
`