JOURNAL OF AEROSOL MEDICINE AND PULMONARY DRUG DELIVERY Volume 26, Number 0, 2013 Mary Ann Liebert, Inc.

. Pp. 110 DOI: 10.1089/jamp.2012.0996

A Novel Continuous Powder Aerosolizer (CPA) for Inhalative Administration of Highly Concentrated Recombinant Surfactant Protein-C (rSP-C) Surfactant to Preterm Neonates
1 1 G. Pohlmann, Dr rer nat, P. Iwatschenko, Dipl Ing,2 W. Koch, Prof, Dr rer nat, Dipl Ing, 1 3 4 H. Windt, Dipl Ing, M. Rast, Dipl Ing, M. Gama de Abreu, Prof, Dr med, F.J.H. Taut, Dr med,5 and C. De Muynck, Dr pharm3

Abstract

Background: In pulmonary medicine, aerosolization of substances for continuous inhalation is conned to different classes of nebulizers with their inherent limitations. Among the unmet medical needs is the lack of an aerosolized surfactant preparation for inhalation by preterm neonates, to avoid the risks associated with endotracheal intubation and surfactant bolus instillation. In the present report, we describe a high-concentration continuous powder aerosolization system developed for delivery of inhalable surfactant to preterm neonates. Methods: The developed device uses a technique that allows efcient aerosolization of dry surfactant powder, generating a surfactant aerosol of high concentration. In a subsequent humidication step, the heated aerosol particles are covered with a surface layer of water. The wet surfactant aerosol is then delivered to the patient interface (e.g., nasal prongs) through a tube. Results: The performance characteristics of the system are given as mass concentration, dose rate, and size distribution of the generated aerosol. Continuous aerosol ows of about 0.84 L/min can be generated from dry recombinant surfactant protein-C surfactant, with concentrations of up to 12 g/m3 and median particle sizes of the humidied particles in the range of 3 to 3.5 lm at the patient interface. The system has been successfully used in preclinical studies. Conclusion: The device with its continuous high-concentration delivery is promising for noninvasive delivery of surfactant aerosol to neonates and has the potential for becoming a versatile disperser platform closing the gap between continuously operating nebulizers and discontinuously operating dry powder inhaler devices. Key words: powder, dispersion, drug delivery, aerosolization, surfactant, preterm neonate, infant respiratory distress syndrome, inhalation

Introduction

T
1 2

o prevent or treat infant respiratory distress syndrome (IRDS) and further related complications, neonates can be successfully treated by airway instillation of surfactant suspension. However, instillation is invasive and requires procedures such as laryngoscopy and endotracheal intubation,

carrying a considerable risk of causing injury to babies,(13) and furthermore gives rise to the assumption that instillation of surfactant through an intratracheal tube is traumatic.(4) Neonatologists nowadays try to avoid intubation and positivepressure ventilation in preterm neonates(5) and favor use of noninvasive ventilator support. Furthermore, animal studies have shown nonuniform distribution of instilled surfactant

Fraunhofer Institute for Toxicology and Experimental Medicine ITEM, D-30625 Hannover, Germany. MTF MediTech Franken GmbH, 90542 Eckental, Germany. 3 Nycomed GmbH; Nycomed: a Takeda Company, D-78467 Konstanz, Germany. 4 Department of Anesthesiology and Intensive Care Therapy, University Hospital Dresden, Dresden, Germany. 5 rich, Switzerland. Takeda Pharmaceuticals International GmbH, Zu

2 material, resulting in poorer clinical response.(6) The idea to circumvent instillation is the aerosolization of surfactant. Previous attempts to generate surfactant aerosols were conned to nebulization of aqueous surfactant suspensions. Drawbacks of this route of administration are low lung deposition, long administration times, and insufcient evidence of efcacy.(1,7) The main reasons for failure are the low achievable concentration of the surfactant aerosol and the need for secondary aerosol dilution when the nebulizer is integrated into ventilator circuits. The maximum concentration achievable by nebulization of surfactant suspensions is principally limited by two factors: the low concentration of surfactant in the suspension, and the concentration of aerosolized particles that may lead to agglomeration when too high, and result in a shift in the size distribution, eventually causing reduced lung deposition. This study aimed at avoiding the drawbacks of nebulization by generating a highly concentrated surfactant aerosol directly from dry surfactant powder and feeding it into the ventilation system as proximal to the patient as possible, so as to avoid dilution by the gas ow in the ventilator circuit commonly used with noninvasive modes of ventilator support. With this new technology, we expected to achieve sufcient lung deposition of surfactant within reasonable administration time. Materials and Methods Continuous powder aerosolizer (CPA): general approach Recombinant surfactant protein-C (rSP-C) surfactant is a spray-dried ne powder with a mass median aerodynamic diameter (MMAD) of 1.7 lm and a geometric standard deviation (GSD) of 2.3. These distribution parameters make the

POHLMANN ET AL. powder ideally suited for direct inhalation administration after dry dispersion, because the MMAD is in the range where high peripheral lung deposition is expected. Dry aerosolization of the surfactant powder allows much higher aerosol concentrations compared with nebulization, because there is no dilution effect by suspending the surfactant in water. Hence, if the same initial aerosol concentration is assumed, nebulization of a suspension of 20% surfactant in liquid will automatically result in a nal concentration of airborne surfactant that is lower by a factor of ve compared with surfactant concentrations achievable by direct aerosolization of dry surfactant powder. Common practice in dispersion technology of dry powders is the induction of high shear stresses (see, e.g., Hinds(8)) into the powder. The degree of powder deagglomeration depends on its rheological properties and the efciency of energy input into the powder portions to be dispersed. Dry dispersion is normally facilitated in the high-speed section of nozzles or swirl chambers.(9) These methods usually require ow rates that are signicantly higher than the breathing minute volumes of preterm neonates, especially of neonates with lung diseases (Schmalisch et al.(10)), which are usually below 1 L/min. To avoid unnecessarily high ows and therefore spillage of surfactant, the dry dispersion device developed and used in this study is based on transient pulses of very small volumes of pressurized air (1030 mL) interacting with dened quantities (13 mg) of powder that are automatically fed from a reservoir into the dispersion zone. The actuation frequency determines the total mass ux and the volumetric ow rate related to aerosol generation. A small auxiliary ow helps to conduct the aerosol to the patient interface. This allows, in principle, the administration of high doses of surfactant material within a short time period

FIG. 1. Soft aggregate formed in the lower trachea after administration of approximately 1 g of dry surfactant aerosol, as has been seen at autopsy.

CONTINUOUS POWDER AEROSOLIZER (CPA) FOR SURFACTANT

FIG. 2. After administration of 1 g of wet aerosol, no aggregate formation in the lower trachea was found.

using only a small volume of air. This is particularly important in the treatment of children, infants, and newborns, where only a small fraction of the inhaled medication is deposited in the lungs.(11) Currently used surfactant suspensions for intratracheal instillation are dosed in the range of 100200 mg/kg.(12) It has been known for over two decades, however, that treatment with aerosolized surfactant improves lung function already at much lower doses of approximately 6 mg/kg.(13) In a rst preliminary study with preterm lambs,(14) we saw signs of efcacy at 7 mg/kg. Assuming a tidal volume of 5 mL for a preterm infant of 1 kg body weight(10) and a breathing frequency of 60/min, a preterm neonate may inhale 300 mL/ min. Assuming further a surfactant aerosol concentration of 8 mg/L, it would be possible to deliver approximately 2.4 mg/min surfactant to the lung. During 1 hr, such a setting could result in delivery of 144 mg of surfactant: Depending on both the particle size and the breathing pattern, a fraction of the delivered surfactant would deposit and remain in the lung. For obvious reasons, only little is known about deposition probabilities of aerosols in the airways of preterm neonates, but from the measurements de hler et al.,(15), deposition probabilities of aerosols scribed by Ko in preterm neonates from about 25% at a particle diameter of 3 lm to 10% at a diameter of 5 lm can be deduced. Therefore, a deposited surfactant dose of 14 to 36 mg/kg could be expected with such high surfactant concentrations. Foreseeable risks associated with the delivery of high quantities of dry material are the formation of occlusions in the delivery device and in the airways of the patient, as well as the dehydration of tissue due to the hygroscopicity of the surfactant powder (which has a residual moisture content of maximal 3% and is composed of rSP-C protein, phospholipids, a fatty acid, and salts). In a pilot study in anesthetized, mechanically ventilated pigs, conducted in the Clinic of Anesthesiology and Intensive Care Medicine, University Clinic Carl Gustav Carus, Dresden, we investigated tolerability of dry powder versus humidied powder surfactant aerosol administered via a thin catheter inserted through an endotracheal tube. The aerosol delivery was synchronized to

the breathing pattern and timed to occur during the early inspiration phase. After administration of approximately 1 g of surfactant aerosol, a soft aggregate had built up in the lower trachea, as was seen at autopsy (Fig. 1). With the same amount of wet aerosol, such an aggregate was not found (Fig. 2). This nding underlines that humidication is a safety prerequisite when applying the surfactant CPA technology. Accordingly, to overcome these adverse effects without compromising the high concentrations, a water layer is condensed in situ onto the surfactant particles immediately after the aerosolization from the powder. These water-coated particles behave like a uid when depositing on surfaces, and the surface layer can drain. Obviously, by adding sufcient water, adverse effects from the hygroscopicity of the surfactant can also be avoided. It is well known that bringing hygroscopic particles into humid air, for example by inhalation, will result in water uptake and growth of the particles.(16) Likewise, the extended retention time of the surfactant aerosol in the CPA system required by the humidication step will result in an additional growth of the particles by coagulation of the highly concentrated aerosol. Overall, a shift in the size distribution toward bigger particles will occur. On the other hand, growth by humidication would occur in the airways anyway. Data about deposition probability in preterm neonates are very hler et al.,(15) it can be deduced, that scarce. From the data of Ko the overall lung deposition probability in infants falls from 25% at a median diameter of 3 lm to 10% at 5 lm. Therefore, residence time of the aerosol in the humidier section should be as short as possible, in order to keep particle size to a minimum. In the following, we will describe the basic layout of the new two-step high-concentration CPA and present its operating characteristics. Technical description of the CPA system General. The CPA system (Fig. 3) consists of three main units: (1) the containerdisperserspacer (CDS) integrating a powder reservoir, a disperser/aerosolizer, and a spacer; (2)

POHLMANN ET AL. chamber and consists of a exible membrane that supports powder transport, so that even powders with bad ow characteristics can be dispersed.(17) In the dispersion chamber, short pulses of pressurized gas aerosolize a dened amount of surfactant into the spacer.(18) During each pulse, the dispersion chamber is reloaded with a dened amount of surfactant powder. This is accomplished by making use of the Venturi effect. A transient high-pressure pulse is injected through a small capillary into the dispersion chamber. This pulse generates a negative pressure in the chamber, drawing a certain amount of surfactant from the cone through a circular orice directly into the main gas stream. When the powder crosses the stream lines, it is exposed to very high shear forces. A small amount of gas on the order of 14 mL is sufcient to generate the dispersion pulse. Principally, different pulse frequencies may be used. The current application uses 10 pulses/min. The dispersion is very efcient, allowing deagglomeration of the powder directly from bulk to the primary particles. Initial aerosol concentrations are as high as 100 g/m3. To quench coagulation and therefore unwanted growth of the aerosol particles as far as possible, but still keep high concentrations, a continuous ow of auxiliary air (0.7 L/min) is added to dilute the aerosol and carry it forward (Fig. 5). Dilution and coagulation are both processes not desired. Without dilution, particles will grow too much while being conducted through the system. Diluting too much will have a negative impact on delivered dose. As both requirements cannot be fullled in parallel, a compromise has to be found. Therefore, a certain particle growth from the initial 1.7 lm has to be accepted. The auxiliary air enters the spacer through an annular array of openings,(17) acting also as a sort of sheath airstream that minimizes deposition of surfactant in the spacer. The particle cloud turns into a progressively continuous stream of surfactant aerosol as it is carried downstream through the spacer, the connection tubes, and the humidier. The spacer, the connection tube between dry powder

FIG. 3. Components of the CPA. The unit called CDS (container disperserspacer) is highlighted by the gray box. the aerosol conditioning/humidication device; and (3) the supply unit. The supply unit provides the necessary airows and energy. It also controls the temperatures of the humidier and the aerosol generated. The CDS is designed as a single-use device whose container part holds up to 3 g of spray-dried surfactant powder. The container is located above the disperser, which redisperses the powder into the miniature spacer located in front of it, generating an aerosol of high concentration. The system is designed such that the concentration of the aerosol leaving the spacer can be varied over a wide range up to about 24 g/m3. To cover the surfactant aerosol with water, the aerosol is conducted through a humidier chamber, where it is heated up to an adjustable temperature in an atmosphere of saturated water vapor. During transport from the heater to the patient interface, the surfactant is allowed to cool down to a physiological temperature, while the water condenses onto the particles. Dry-powder disperser. The supply unit provides pressurized gas with a dened oxygen fraction to the CDS unit. The container (see Fig. 4) sitting on a cone can hold up to 3 g of surfactant powder. The cone is located above a dispersion

FIG. 4. Principle of the dispersion of surfactant powder by pressure pulses. (Inset) A sketch of the real system.

FIG. 5.

Mixing of aerosol in the spacer.

CONTINUOUS POWDER AEROSOLIZER (CPA) FOR SURFACTANT

FIG. 6. Setup used to measure the aerodynamic diameter of the wet aerosol at the exit of the second connection tube.

disperser and humidier, as well as the humidier itself, serve as a mufer for the pressure pulse, so that at the exit of the second connection tubing pressure peaks are restricted to 1 mbar maximum when connected to a ventilatory circuit. Aerosol conditioning. In the humidier, the dry surfactant particles undergo a combined heating and humidication process.(19) Water is being heated in a chamber and transported through a water vaporpermeable membrane into a compartment through which the aerosol particle stream is conducted. Heat ux and ux of water vapor into the air space are sufcient to heat up the aerosol almost to the temperature of the water and to achieve saturation with water vapor. Further downstream, in the second connection tube, the aerosol cools down under controlled conditions and the water vapor condenses onto the particles. This condensation leads to particles with a water fraction sufcient to ensure draining away from surfaces as opposed to buildup of deposits and risk of occlusions. When suspensions are produced from powder consisting of small, agglomerated particles in the presence of gas, inclusion of gas is often not avoidable, thus forming bigger aggregates in the suspension. In our case, the water vapor condenses directly onto the

airborne particles, thereby preventing this inclusion and thus enhancing the aerosol behavior. The second connection tube is optimized in length and inner diameter to ensure user convenience while minimizing loss of material. The aerosol is delivered to the neonates nose via a set of modied prongs(20) or a mask at a physiological temperature of approximately 37C. Methods used for performance characterization Mass size distributions. The mass size distribution as a function of the aerodynamic diameter is determined using a New Generation Impactor (NGI, Copley Scientic AG, Therwil, Switzerland) operated at 15 L/min. The measurement setup is shown in Figure 6. The aerosol generated at a ow rate of 0.84 L/min (transient dispersion pulse and auxiliary air ow) is conducted directly into the induction port of the NGI. When measuring the dry aerosol emitted from the CDS, the impactor cups are pre- and postweighed without further conditioning. The wet aerosol is also conducted directly into the induction port of the NGI and collected into the impactor cups inside the NGI. These cups are preweighed prior to use, loaded with the wet aerosol, placed

FIG. 7. Setup used for diffraction spectrometer measurements. The device exit can be either the exit of the connection tube or the exit of the second connection tube.

POHLMANN ET AL.

FIG. 8. Measurement setup used for dose rate and efciency determination.

into a heated cabinet at 50C for 2 hr, and allowed to acclimatize for 2 hr at laboratory conditions before postweighing (balance used: Sartorius RC 210S-0D1). Laser diffraction measurements. Time-resolved size distributions are measured using a HELOS diffraction spectrometer (Sympatec GmbH, Clausthal-Zellerfeld, Germany; Fig. 7). This well known diffraction method allows in situ time-resolved measurements of particle size distributions. The surfactant aerosol to be measured is blown from the exit of the device directly into the measuring volume of the spectrometer. For this purpose, the HELOS system is used without further accessories. Dose rates and delivery efciency. Dose rate and efciency of the system are measured with the setup shown in

Figure 8. To determine the surfactant mass aerosolized, the CDS is weighed before and after aerosolization (Sartorius LP 3200D). The aerosol is sampled at the outlet of the second connection tubing. The ow rate drawn through the lter holder is higher than the aerosol ow rate delivered from the CPA. This ensures sampling of the entire aerosol. The sampling lters are preconditioned by placing them in the laboratory for more than 24 hr prior to use. After sampling of the wet aerosol, the lters are postconditioned as described above and then weighed. Results Mass size distributions Figure 9 shows an aerodynamic particle size distribution of the dry surfactant aerosol measured with the NGI at the

FIG. 9. Aerodynamic mass size distribution of the dry aerosol measured at the exit of the connection tube.

FIG. 10. Aerodynamic mass size distribution of the wet aerosol measured at the exit of the second connection tube.

CONTINUOUS POWDER AEROSOLIZER (CPA) FOR SURFACTANT

FIG. 11. Courses of the median diameter (X50 value, solid line, left axis) and the optical density (Copt, dashed line, right axis) of the dry aerosol at the exit of the connection tube. The optical density is an indicator of the concentration at the exit of the CDS. exit of the connection tube. The upper part of the gure gives the distribution frequency in discrete size classes. In the lower part, the corresponding cumulative distribution is shown on a log-probability scale. The solid line is a spline t to the data. From the spline t, the characteristic distribution parameters MMAD and GSD as a measure of the width of the distribution are determined. Average values of size distribution parameters measured at the exit of the CDS (dry aerosol) are MMAD = 3.2 lm and GSD = 2.1 (n = 2). The shapes of the size distributions measured at the exit of the second tube (wet aerosol; Fig. 10) are similar to the ones at the connection tube (dry aerosol; Fig. 9). The values of the size distribution parameters are MMAD = 3.2 lm and GSD = 2 (n = 3). There are particle losses of about 50% in the humidier and the transport tube to the patient interface (see also Dose rates and delivery efciency). It is important to note that humidication does not signicantly change the aerodynamic mass distribution of the aerosol delivered to the patient compared with what is generated from the dry powder (Fig. 9).

FIG. 13. Average size distribution of the dry aerosol at the exit of the connection tube. Laser diffraction measurements Figures 11 and 12 show the time-resolved parameter sequence for one of the multiple pressure pulses generating the surfactant aerosol. The solid (upper) curves show the median diameter (X50 value) of the size distributions, whereas the dashed (lower) curves show the course of the optical density of the aerosol as a qualitative marker for the aerosol concentration. At the exit of the rst connection tube (Fig. 11), the concentration rst rises to a maximum and, after 6 sec, nearly reaches zero. After the aerosol has passed the humidier and the second tube (Fig. 12), the form of the pulses has been altered by the residence time distribution of the system, resulting in a attened prole where the concentration does not reach zero during the pulse sequences. The mean diameters of the distributions (Figs. 13 and 14, Table 1) differ only slightly between dry and wet aerosol, reecting two counteracting physical phenomena that shape the distributions: while traveling through the system particles coagulate, shifting the size distribution toward bigger sizes, but

FIG. 12. Courses of the median diameter (X50 value, solid line, left axis) and the optical density (Copt, dashed, lower line, right axis) of the wet aerosol at the exit of the second connection tube. The optical density (dashed curve) is an indicator of the concentration at the exit of the second connection tube.

FIG. 14. Average size distribution of the wet aerosol at the exit of the second connection tube.

8 Table 1. Parameters of Measured Size Distribution of The Aerosol Diameter (lm) Percentage of cumulative curve 10 16 50 84 90 Dry 0.75 0.90 2.24 5.09 6.17 Wet 0.80 0.99 2.46 4.88 5.68

POHLMANN ET AL. 12 2 mg/min with regard to dry surfactant at the exit of the CDS and of 6.6 mg/min at the exit of the second connection tube. Discussion Current surfactant therapy routinely involves instillation of surfactant in liquid form directly into the trachea. This procedure is associated with many risks.(3) Past approaches to surfactant nebulization had little success due to inefcient aerosol devices, resulting in low intrapulmonary delivery of surfactant.(3) Due to agglomeration, the upper particle concentration stable in an aerosol is limited. Approaches with vibrating mesh nebulizers depend on the use of liquid surfactant formulations. To be used in this type of nebulizer, the active ingredients of the surfactant, namely phospholipids, have to be diluted notably before being nebulized. When aerosolizing dry surfactant powder, no such dilution is necessary. Consequently, dry aerosolization of the surfactant powder allows much higher aerosol concentrations compared with nebulization, because the additional dilution effect is avoided. For example, assuming the same initial aerosol concentration, nebulization of a suspension of 20% surfactant in liquid will automatically result in a nal concentration of airborne surfactant that is lower by a factor of ve compared with surfactant concentrations achievable by direct aerosolization of dry surfactant powder. Therefore, with the CPA system, we are able to generate surfactant aerosol of very high concentration. A factor directly inuencing the deposition of surfactant in the lung is the breathing pattern of the neonate. Schmalisch et al.(10) reported tidal breathing parameters for infants with chronic lung diseases and body weights between 1.95 and 3.80 kg at the time of measurement. They reported tidal volumes of 5.15 1.35 mL/kg and breathing rates of 55.4 14.2 min1, resulting in a minute volume of 285 148 mL/(min kg). Comparable data have been reported, for example, by Bhutani and Benitz,(21) for babies with weights as low as approximately 500 g. Taking these gures and calculating the mass of surfactant delivered to the lung during 1 hr using an aerosol concentration of 8 mg/L

at the same time particles are lost in the system by sedimentation and impaction processes, shifting the size distribution to smaller particles. Dose rates and delivery efciency To transfer sufcient heat and water vapor into the aerosol, an adequate residence time of the aerosol in the humidier is required. This leads to considerable losses of particles, especially when considering the high aerosol concentration. Therefore, most losses within the total system between the exit of the CDS and the outlet of the second tubeoccur in the humidier. By dening delivery efciency as the ratio of the mass leaving the second connection tube to the mass leaving the CDS, losses in the system downstream of the CDS can be characterized. Figure 15 shows the efciency as a function of the aerosol concentration at the CDS outlet. As can be expected, delivery decreases with increasing aerosol concentration. The three vertical lines in Figure 15 mark the intended working range of the device: lower limit (9.5 g/m3), operating point (16.2 g/m3), and upper limit (18.8 g/m3). Within this range, the mean efciency is 55 4%. The measured average concentration at the outlet of the CDS in the working range is 14.5 2.4 mg/L, with a mean efciency of 55%. This results in an average concentration at the outlet of the second connection tube of 8 mg/L. As during application the ow through the CDS is 0.84 L/min, this efciency results in a mean continuous dose rate of

FIG. 15. Efciency of aerosol delivery at the outlet of the second connection tube. The vertical lines indicate the operating point (dashed line) and the maximal permissible deviations from the operating point (solid lines).

CONTINUOUS POWDER AEROSOLIZER (CPA) FOR SURFACTANT Table 2. Comparison of CPA Features With Features of Existing Aerosolizer Technologies Features Continuous ow Humidied application possible Dry application possible Works with water-soluble substances Works with water-insoluble substances Large quantities applicable High concentration possible Usable in ventilators Nebulizer CPA DPI (liquid) (dry/humidied) (dry) x x x x x x x x x x x x x x x x

results in 137 71 mL/(kg hr). Assuming deposition probabilities in the lung of 10 to 25%(15) gives total deposited doses of 14 7 mg/kg to 34 18 mg/kg. Considering the mass median diameter measured, the value of 34 18 mg/kg is more likely. Comparing these estimated concentrations with the 7 mg/kg of our rst preliminary study with preterm lambs,(14) as well as with data reported by Lewis et al.,(13) it seems possible to successfully administer a sufcient amount of surfactant to preterm neonates by inhalation. The aerosol concentration at the outlet of the rst connection tube of the CPA can easily be adjusted by changing the frequency of pressure pulses per unit of time. The second parameter that is also easily adjustable is the ow of auxiliary air. With the described device, the concentration, dose rate, total ow, and water content of the aerosol delivered can thus be adjusted to different needs. Therefore, the CPA system is not restricted to be used for preterm neonates only. Many active pharmaceutical ingredients (APIs) are not or only poorly water-soluble. Some of these are attractive candidates for inhalation administration. At present, these APIs are made available for inhalation using formulations such as liposomes. These formulations are nebulized using continuously operating nebulizers.(22) It turns out that development of the liposomal suspensions for nebulization might be a very time-consuming task compared, for example, with the time spent to bring the liposomal formulation of amikacin into clinical trial.(23,24) The CPA system is capable of aerosolizing nonwatersoluble APIs(25) without the additional burden(22) often associated with formulation issues related to the nebulization of liposomes. Table 2 gives an overview of the typical features the new device facilitates and compares these with existing aerosolizer technologies for solutions (nebulizer) and dry powders [dry powder inhalers (DPI)]. The table shows that the new device combines the features of the different aerosolizer technologies, making available a new class of APIs for inhalation treatment. Conclusions The results presented here show that the novel CPA system is able to produce a continuous aerosol stream of highly concentrated rSP-C surfactant with median particle diameters in the size range of 3 lm. In preclinical trials, it was shown(14) that these high concentrations are well tolerated

when applied as wet aerosol. The new device is promising for administration of inhalable surfactant to preterm neonates on noninvasive ventilator support. From the measure hler et al.,(15) deposition probabilities ments described by Ko for aerosols in preterm neonates from about 25% at a particle diameter of 3 lm to 10% at a diameter of 5 lm can be deduced. As during 1 hr about 144 mg of surfactant (corresponding to a concentration of 14.5 mg/L at the CDS outlet, which results in 8 mg/L at the exit of the second connection tubing) can be safely delivered using our device, a deposited surfactant dose of 14 to 36 mg can be expected. Further investigations regarding the deposition probability of aerosols in preterm neonates are necessary. As no in vivo measurements are possible, experiments will be conducted with models of the airways. In pulmonary medicine, aerosolization of substances for continuous inhalation is conned to different classes of nebulizers(26) with their inherent limitations. The aim of developing the device presented here was to enable the generation of a highly concentrated humidied surfactant aerosol directly from dry surfactant powder. These efforts resulted in the CPA technology to be used as an aerosolizer not only for surfactant powder, but probably also suitable for almost all dry powders with particle diameters small enough to be inhaled. Comparison of the CPA system with existing technologies (Table 2) shows that it represents a new, versatile disperser platform that closes the gap between continuously operating nebulizers and the discontinuously operating DPI technology. A vast amount of substances that currently are not available for continuous inhalation treatment may in the future become available as inhalational therapy using the novel CPA technology platform. Acknowledgments This work was funded by Nycomed GmbH; Nycomed: a Takeda Company, Konstanz, Germany. Author Disclosure Statement The authors declare that they have jointly led patents on the device described in this report. The Fraunhofer Institute for Toxicology and Experimental Medicine ITEM and MTF MediTech Franken GmbH are contract development partners of Nycomed GmbH, Konstanz, Germany. Markus Rast and Christian De Muynck are full-time employees of Nyocmed GmbH; Friedemann Taut is a full-time rich, employee of Takeda Pharmaceuticals International, Zu Switzerland. References
1. Shah S: Exogenous surfactant: intubated present, nebulized future? World J Pediatr. 2011;7:1115. 2. Jorch G, Hartl H, Roth B, Kribs A, Gortner L, Schaible T, Hennecke H, and Poets C: Surfactant aerosol treatment of respiratory distress syndrome in spontaneously breathing premature infants [Letter]. Pediatr Pulmonol. 1997;24:222 224. 3. Pillow JJ, and Minocchieri S: Innovation in surfactant therapy II: Surfactant administration by aerosolization. Neonatology. 2012;101:337344. 4. Shangle CE, Haas RH, Vaida F, Rich WD, and Finer NN: Effects of endotracheal intubation and surfactant on a

10
3-channel neonatal electroencephalogram. J Pediatr. 2012;161:252257. Morley CJ, Davis PG, Doyle LW, Brion LP, Hascoet JM, and Carlin JB; COIN Trial Investigators: Nasal CPAP or intubation at birth for very preterm infants. N Engl J Med. 2008;358:700708. Mazela J, Merritt TA, and Finer NN: Aerosolized surfactants. Curr Opin Pediatr. 2007;19:155162. Finer NN, Merritt TA, Bernstein G, Job L, Mazela J, and Segal R: An open label, pilot study of Aerosurf combined with nCPAP to prevent RDS in preterm neonates. J Aerosol Med Pulm Drug Deliv. 2010;23:303309. Hinds WC: Aerosol Technology, 2nd edition. John Wiley & Sons, New York; 1999. Crowder M, and Donavan MJ: Science and technology of dry powder inhalers. In: HDC Smyth, and JA Hickey, (eds). Advances in Delivery Science and Technology: Controlled Pulmonary Drug Delivery. Springer, New York; pp. 203222, 2011. Schmalisch G, Wilitzki S, and Wauer GG: Differences in tidal breathing between infants with chronic lung diseases and healthy controls. BMC Pediatrics. 2005;5:36. Fink JB: Aerosol delivery to ventilated infant and pediatric patients. Respir Care. 2004;49:653665. Sweet HG, Carnielli V, Greisen G, Hallman M, Ozek E, Plavka R, Saugstad OD, Simeoni U, Speer CP, and Halliday HL; European Association of Perinatal Medicine: European consensus guidelines on the management of neonatal respiratory distress syndrome in preterm infants2010 update. Neonatology. 2010;97:402417. Lewis JF, Ikegami M, Jobe AH, and Tabor B: Aerosolized surfactant treatment of preterm lambs. J Appl Physiol. 1991;70:869876. Rahmel DK, Pohlmann G, Iwatschenko P, Volland J, Liebisch S, Kock H, Mecklenburg L, Maurer C, Kemkowski J, and Taut FJH: The non-intubated, spontaneously breathing, continuous positive airway pressure (CPAP) ventilated preterm lamb: a unique animal model. Reprod Toxicol. 2012;34;204215. hler E, Jilg G, Avenarius S, and Jorch G: Lung deposition Ko after inhalation with various nebulisers in preterm infants. Arch Dis Child Fetal Neonatal Ed. 2008;93:F275F279. Broday DM, and Georgopoulos PG: Growth and deposition of hygroscopic particulate matter in the human lungs. Aerosol Sci Technol. 2001;34:144159. Iwatschenko P, Pohlmann G, Windt H, Koch W, and Kist M: Improved apparatus for the aerosolization of large volumes of dry powder. 2010; WO 2010/122103 A1. Pohlmann G, Windt H, Nolte O, and Koch W: Device for dosing and dry nebulization. 2006; WO 2006/108558 A1.

POHLMANN ET AL.
19. Rahmel D, Taut F, De-Muynck C, Rast M, Iwatschenko P, Pohlmann G, Koch W, and Windt H: Humidied particles comprising a therapeutically active substance. 2012; WO 2012/025496 A1. 20. Iwatschenko P, Pohlmann G, Rahmel D, Taut F, De Muynck C, Koch W, and Windt H: Device and system for delivery of an aerosol to a patient on ventilatory support. 2012; WO 2012/020004 A1. 21. Butani VK, and Benitz WE: Pulmonary function and graphics. In: JP Goldsmith, and EH Karotkin, (eds). Assisted Ventilation of the Neonate, 5th ed. Elsevier, Philadelphia, PA; pp. 306320, 2011. 22. Zaru M, Mourtas S, Klepetsanis P, Fadda AM, and Antimisiaris SG: Liposomes for drug delivery to the lungs by nebulization. Eur J Pharm Biopharm. 2007;67:655666. 23. Schreier H, McNicol KJ, Ausborn M, Soucy DM, Derendorf H, Stecenko AA, Gonzalez-Rothi RJ: Pulmonary delivery of amikacin liposomes and acute liposome toxicity in the sheep. Int J Pharm. 1992;87:183193. 24. Li Z, Zhang Y, Wurtz W, Lee JK, Malinin VS, DurwasKrishnan S, Meers P, and Perkins WR: Characterization of nebulized liposomal amikacin (Arikace) as a function of droplet size. J Aerosol Med Pulm Drug Deliv. 2008;21:245253. 25. Pohlmann G, Iwatschenko P, Koch W, Windt H, Rast M, Taut F, and De-Muynck C: Continuous powder aerosolization (CPA): a new device to administer inhalable particle formulations in high concentrations. AAPS Annual Meeting and Exposition, Chicago, IL, October 1418, 2012; M1103. 26. Dolovich MB, and Dhand R: Aerosol drug delivery: developments in device design and clinical use. Lancet. 2011;377: 10321045.

5.

6. 7.

8. 9.

10.

11. 12.

13.

14.

Received on July 18, 2012 in nal form, December 10, 2012 Reviewed by: Gerhard Scheuch Ralph Niven

15.

16.

17.

Address correspondence to: Dr. Gerhard Pohlmann Fraunhofer Institute for Toxicology and Experimental Medicine (ITEM) Nikolai-Fuchs-Str. 1 D-30625 Hannover Germany E-mail: Gerhard.pohlmann@item.fraunhofer.de

18.