CA2531251A1 - Nebulizer for aerosol drug delivery - Google Patents

Abstract

A method for the capture, return, and reaerosolization of undelivered aerosol with a neonatal jet nebulizer is described. The method involves the placement of an entraining jet and an impaction surface into the expiratory line of a neonatal ventilator-nebulizer setup, where the jet entrains unused medical aerosol and deposits it onto an impaction surface.
The deposited solution is then returned to the nebulizer reservoir in a feedback tube. As a result, more of the initial dosage is delivered to the patient. The fraction of the dosage delivered to a filter connected to a passive neonatal test lung was compared between the situations with and without feedback. A statistically significant improvement was observed for the situation with feedback, with up to 60% more aerosol delivered for a ventilator setting of 7.5 cm H 2 0 [0.74 kPa] (.apprxeq. 6 mL tidal volume) and 40 breaths per minute (inspiratory-to-expiratory ratio of 3:7). Further device improvements may enhance the clinical efficacy of jet nebulizers in ventilated newborns.

Key Words: neonatal, respiratory drug delivery, inhaled pharmaceutical aerosol, intubated, pediatric intensive care

Description

NEBULIZER FOR AEROSOL DRUG DELIVERY

A major difficulty with the use of jet nebulizers in neonatal settings is caused by neonates having extremely small tidal volumes when compared with required nebulizing air flow rates, e.g. tidal volumes of approximately 0.5 L/min while nebulizing air flow rates usually range from 4 to 6 L/min (the flow rates typically needed to achieve optimal aerosol particle sizes). As a result, during inspiration, only a small fraction of the nebulized volume of medicine is delivered to the neonate (Fink 2004). Clearly the nebulizer efficiency is extremely low in these circumstances, with nebulizer efficiency defined as the fraction of the initial dosage of medicine output by the nebulizer to the patient. In contrast, an adult would generally have a tidal volume that is large enough to include all of the nebulized flow rate, yielding much higher efficiency than in neonates.

One possible way to improve jet nebulizer efficiency for neonates is to reduce the amount of aerosol lost in the outflow. Since the nebulizing air rate cannot be reduced without negatively altering the aerosol particle sizes, another method for lowering the amount of aerosol lost to the outflow is needed. As a result, a method for recycling as much of the aerosol outflow as possible was desired.

Numerous authors have examined nebulizer efficiency with existing commercial devices in ventilated pediatric settings (see Fink 2004 for a review). There have been only a few efforts to overcome the low efficiency of jet nebulizers when applied to neonates. One method is to cycle the nebulizing air flow on and off in synchronization with a neonate's breathing pattern, with the nebulizer active only when the neonate is inhaling. An improvement of a factor of 1.6 was observed with this approach (Pelkonen et al. 1997).
Another method is use an alternate method to generate the nebulized aerosol, e.g. Aerogen, Inc. has developed micropump nebulizers that employ a vibrating mesh to nebulize solution, which leads to high efficiencies for neonates, typically > 10% (Bartram 2005, Dubus et al.
2005). Ultrasonic nebulizers, which also employ vibrating meshes for nebulization, have also been shown to have greater efficiencies than typical jet nebulizers for some solutions (Fok et al. 1998, Wagner et al. 2000). Finally, there are numerous studies comparing the efficiencies of metered-dose inhalers (MDI) and spacers with jet nebulizers in neonates (Gappa et al.
1997, Fok et al. 1998, Avent et al. 1999, Khalaf et al. 2001, Lugo et al.
2001). MDI and spacer combinations were shown to have the same effect on neonatal lung compliance and resistance as jet nebulizers while requiring smaller dosages than the jet nebulizers in clinical trials. However, jet nebulizers are still used clinically in neonatal ventilation, and it may be useful to consider ways to improve their efficiency in neonatal drug delivery.

2 Methods 2.1 Entrainment, Impaction, and Feedback of Outflow Aerosol To improve jet nebulizer delivery in neonatal ventilation, here we consider a method to recycle the aerosol. For this purpose, aerosol particles in the outflow were collected and returned to the nebulizer reservoir. Once the aerosol mass collected from the outflow is returned to the nebulizer reservoir, the medication can then be reaerosolized.
With this approach, nebulizer efficiency is improved while the total nebulization time given a certain dosage would be extended when compared with a normal nebulizer.

The collection of aerosol particles in the outflow involved two steps: 1) entrainment, and 2) impaction (Figure 1). Inclusion of a narrow coaxial "entraining jet" in a rigid outflow tube in the expiratory line allowed for entrainment, while the inclusion of a solid surface perpendicular to the entraining jet's axis allowed for impaction.

Entrainment occurs when a high-speed air flow exits the narrow jet tube and entrains air in front of the jet exit. An aerosol moving in the outflow is drawn, or "entrained", into the conically-expanding jet. As the aerosol particles are being entrained, they are accelerated to the same high velocities as the air from the jet. Note that since the jet expands, the velocity of the jet air and the entrained aerosol decreases with distance.

Impaction is simply the deposition of particles on a surface when particles strike it. In order for impaction to occur, particles of a given mass require a minimum impaction velocity; hence, the requirement of an entraining jet to accelerate the aerosol particles.
Particles approaching the surface with velocities below that of the impaction velocity simply avoid depositing on the surface.

Once the solution is collected on the impaction surface, it can be returned to the nebulizer reservoir through a gravity feed (in this case, a narrow tube).

Figure 1 shows a schematic of the collection and return mechanism.
2.2 Modification of the Nebulizer To add the entraining jet, impaction surface, and feedback to the nebulizer, a modification to the nebulizer top was necessary. Since the feedback tube needed to be connected to the nebulizer, it was desirable for the entraining jet and impaction surface to be connected to the nebulizer as well. With this in mind, the nebulizer top was extended to include a section of the expiratory branch of the ventilator, which ultimately also required the inclusion of connections to the inspiratory line and output to the patient (Figure 2).

3 Hudson Neb-U-Mist II nebulizers (Temecula, California) were used in these experiments, and each nebulizer reservoir was used for less than 10 runs.

A hole was cut into the nebulizer top to allow for the insertion of the feedback tube.
(This was the only irreversible modification to the nebulizer.) The feedback tube was connected to the nebulizer baffle and a "feedback connector", which had a conically- sloped interior to allow for the return of collected droplets of medical solution without pooling effects. The feedback tube was made of Tygon R-3603 tubing (Saint-Gobain, Paris, France), and the feedback connector was custom-made from Teflon tubing in the machine shop of the Department of Mechanical Engineering at the University of Alberta.

The feedback connector was connected to a 61404 tee piece (B&F Medical, Toledo, Ohio) which acted as the impaction surface. It also allowed for connections to the expiratory line and to the tee piece that held the entraining jet (with a rubber 0-ring). The remaining connections (as well as the aforementioned modifications) are shown in Figure 2.

It should be noted that all of the tee pieces had approximately 0.75-inch [1.9-cm] inner diameters and 0.88-inch [2.2-cm] outer diameters, and they were each approximately 2.0 inches [5.1 cm] long and 1.2 inches [3.0 cm] tall. The extension tube had a 0.88-inch [2.2-cm] inner diameter and a 1.0-inch [2.5-cm] outer diameter. The feedback tube had an inner diameter of about 0.24 inches [6.0 mm] and an outer diameter of approximately 0.35 inches [9.0 mm]. The entraining jet tube had an inner diameter of roughly 0.04 inches [1 mm], while it had an outer diameter of roughly 0.12 inches [3.0 mm].

Additionally, it should be noted that the nebulizer end of the feedback tube extended to a length such that it would be submerged in the nebulizer solution in the nebulizer reservoir. This was done to prevent nebulized droplets from exiting directly to the

4 expiratory line by traveling up the feedback tube.

The jet exit was approximately 1.2 inches [3.0 cm] from the furthest point on the wall of the tee piece acting as the impaction surface. This distance was large enough to meet the requirement for the entrainment of a large fraction of the aerosol in the outflow but was small enough to prevent the velocities of most of the particles from dropping below impaction velocity.

2.3 Additional Equipment and Materials Used A Sechrist Infant Ventilator Model IV-100B (Anaheim, California) was used to supply the ventilation. A Pari Proneb Ultra air compressor (Pari, Germany) was used to provide the nebulizing air flow for the Hudson nebulizer with a measured flow rate of

5.5 L/min (Singer American Meter Company DTM-115 dry test gas flowmeter, Wellesley, Massachusetts). The air flow controller and rotameter, for the entraining jet, was Omega Engineering's FL-3440C-HRV rotameter with calibration 044-40C (Manchester, United Kingdom). The air flow for the entraining jet was provided by a wall feed.
Vital Signs' Respirgard II filters (Totowa, New Jersey) were used to collect the aerosol.

A passive neonatal test lung with a compliance of 8 f 2 mL/kPa was used.

The solute was ciprofloxacin hydrochloride, whose absorbance in solution was directly proportional to concentration for concentrations ranging from 200 ng/mL to 16 g/mL.
The ciprofloxacin was dissolved in 0.9% saline solutions. Absorbance of standards and dilutions of the ciprofloxacin was measured using UV spectroscopy (Hewlett Packard 8452A
Diode Array Spectrophotometer, Palo Alto, California).

2.4 Experimental Procedure The modified nebulizer was connected to the ventilator as indicated in Figure 2. A filter was placed between the outflow and the ventilator's expiratory line in order to prevent ciprofloxacin from being released into open air. The output to the test lung was connected to a filter, which in turn was connected to the passive neonatal test lung.

For each run, 7 mL of ciprofloxacin solution of known concentration was placed into the nebulizer reservoir. Each run was either a run where the entraining jet was off with no feedback; a run where the entraining jet was on with no feedback; or a run where the entraining jet was on with feedback. Each time the entraining jet was run, the rotameter was used to set a measured dry air flow rate of 8 L/min (Singer American Meter Company DTM-115 dry test gas flowmeter, Wellesley, Massachusetts). It should be noted that the flow rate could not be set arbitrarily high in order to boost the jet exit velocity due to the increased significance of evaporation effects.

The difference in the ventilator's inspiratory pressure and the expiratory pressure ranged from OP az~ 6 cm H20 to OP -- 7.5 cm H20 between runs, which resulted in tidal volumes ranging from approximately 5 mL to 6 mL. The data was adjusted to a uniform OP of 7.5 cm H20 assuming linear variation of efficiency versus OP. A bias flow of 2 1 L/min was used, and 40 breaths per minute with a 3:7 inspiration-to-expiration ratio were simulated. The nebulizer was allowed to run for as long as it could continue to aerosolize solution. At the end of each run, the filter (and its casing) connected to the test lung was assayed by distilled water to extract the captured ciprofloxacin. The extract was diluted to mL. To ensure the thorough extraction of the ciprofloxacin from the filter, a second extract was performed. (Each second extract was also diluted to 10 mL.)

6 For each dilution, the absorbance at A=272 nm was recorded as the peak, while the absorbance at A=298 nm was recorded as the baseline.

The nebulizer efficiency was calculated as:

77neb = m,f2lter x 100% (1) mdose where:
Tnfilte,. = mass of ciprofloxacin deposited on the filter and filter casing mdose = dosage of ciprofloxacin placed in the nebulizer T7neb = nebulizer efficiency In order to determine whether differences between sets of data were statistically significant, the t-test for independent samples was used to obtain a p-value for each comparison (Microsoft Excel, Redmond, Washington). The null hypothesis was accepted to be true for p> 0.05, while the alternative hypothesis was accepted for p <
0.05.

3 Results and Discussion Table 1 shows the results of our testing. Three cases are presented: case 1) the apparatus was run with no entraining jet and no feedback; case 2) the apparatus was run with the entraining jet active but with no feedback; and case 3) the apparatus was run with both the entraining jet and feedback. The unmodified nebulizer (case 1) shows a delivery efficiency of 1%, similar to that seen by previous authors (Pelkonen 1997, Fink 2004).

Case 1 and case 2 do not differ significantly (p = 0.12 > 0.05), indicating no significant effect of the entraining jet when no feedback was present, as expected.

Initially, it was feared that the entraining jet might entrain aerosol that would have normally been delivered to the patient as well as the outflow aerosol, which would have

7 decreased the efficiency since less of the aerosol would have been output to the test lung. (If this were the case, the feedback may or may not have made up the decrease in efficiency.) Apparently, the confinement of the jet in a fairly narrow tube limited the range of entrainment of aerosol so that the entraining jet alone had the desired lack of effect on efficiency.

Comparing the entraining jet with and without feedback (case 2 versus 3), a statistically significant increase in efficiency due to feedback was found (Table 1).

Although the improvement seen from the inclusion of the feedback was expected by us, large variations in the measured efficiencies for the runs with both the entraining jet and feedback active were unforeseen. A likely explanation is that the entraining jet was somewhat loose, and orientation of the entraining jet probably had a relatively large effect on the efficiency. If the jet was at an angle to the expected axis, more deposition would have occurred on the tube walls around the jet. The deposition on these walls would not have been returned to the nebulizer store. Tighter control of the entraining jet would be expected to reduce the variability but future work is needed to confirm this hypothesis.

Given the above results, it is unsurprising that a statistically significant improvement in efficiency is seen in Table 1 going from a situation with no entraining jet and no feedback to a situation with both an entraining jet and feedback. It should be noted that the run times increased by approximately 40% with feedback, with a concomitant increase in drug concentration near the end of nebulization. However, the improvement in delivery by a factor of 1.6 when an entraining jet with feedback is used is encouraging.
Combining this approach with other improvements to jet nebulizer delivery may make jet nebulizers a more attractive delivery method in neonates.

8 Although a reasonable effort was spent in an attempt to determine a geometry that would yield ideal results, more optimized geometries should still be possible, since the test apparatus was limited by the fact that the modified nebulizer was comprised primarily of commercially-available tee pieces rather than custom designed parts.

9 Table 1: Nebulizer efficiency measured in vitro with a neonatal ventilator (mean ~
standard deviation) Case # of runs Nebulizer efficiency 1. No entraining jet and no feedback 4 0.96% 0.09%

2. Entraining jet with no feedback 6 1.08% f 0.18%
3. Enraining jet with feedback 8 1.58% f 0.42%

Product Sources B&F Medical 61404 tee pieces (B&F Medical, Toledo, Ohio) Hewlett Packard 8452A Diode Array Spectrophotometer (Hewlett Packard, Palo Alto, California) Hudson Neb-U-Mist II nebulizers (Temecula, California) Microsoft Excel (Microsoft, Redmond, Washington) Omega Engineering FL-3440C-HRV rotameter (Omega Engineering, Manchester, United Kingdom) Pari Proneb Ultra air compressor (Pari, Germany) Sechrist Infant Ventilator Model IV-100B (Sechrist, Anaheim, California) Singer DTM-115 dry gas flowmeter (Singer American Meter Company, Wellesley, Massachusetts) Tygon R-3603 tubes (Saint-Gobain, Paris, France) Vital Signs Respirgard II filters (Vital Signs, Totowa, New Jersey) References [1] Fink JB. Aerosol Delivery to Ventilated Infants and Pediatric Patients.
Respir Care 2004;49:653-665.

[2] Pelkonen AS, Nikander K, Turpeinen M. Jet nebulization of budesonide suspension into a neonatal ventilator circuit: Synchronized versus continuous nebulizer flow.
Pediatr P ulm onol 1997; 24: 282-286.

[3] Bartram P. Use of a micropump nebulizer for aerosolized medication delivery in a ventilated infant. Neonatal Intensive Care 2005;18(1):14.

[4] Dubus JC, Vecellio L, DeMonte M, Fink JB, Grimbert D, Montharu J, et al.
Aerosol deposition in neonatal ventilation. Pediatr Res 2005;58(1):10-14.

[5] Fok TF, Lam K, Ng PC, Leung TF, So HK, Cheung KL, et al. Delivery of salbutamol to nonventilated preterm infants by metered-dose inhaler, jet nebulizer, and ultrasonic nebulizer. Eur Respir J 1998;12(1):159-164.

[6] Wagner MH, Wiethoff S, Friedrich W, Mollenhauer I, Obladen M, Boenick U.
Ultrasonic surfactant nebulization with different exciting frequencies. Biophys Chem 2000;84(1):35-43.

[7] Gappa M, Gartner M, Poets F, vonderHardt H. Effects of salbutamol delivery from a metered dose inhaler versus jet nebulizer on dynamic lung mechanics in very preterm infants with chronic lung disease. Pediatr Pulmonol 1997;23(6):442-448.

[8] Avent ML, Gal P, Ransom JL, Brown YL, Hansen CJ, Ricketts WA, et al.
Evaluating the delivery of nebulized and metered-dose inhalers in an in vitro infant ventilator lung model. Ann Pharmacother 1999;33(2):144-148.

[9] Khalaf MN, Hurley JF, Bhandari V. A prospective controlled trial of albuterol aerosol delivered via metered dose inhaler-spacer device (MDI) versus jet nebulizer in ventilated preterm neonates. Am J Perinatol 2001;18(3):169-174.

[10] Lugo RA, Kenney JK, Keenan J, Ballard J, Ward RM. Albuterol delivery in a neonatal ventilated lung model: Nebulization versus chlorofluorocarbon- and hydrofluoroalkane-pressurized metered dose inhalers. Pediatr Pulmonol 2001;31(3):247-254.

Claims (5)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY OR
PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. Apparatus, comprising:
a nebulizer connected to an inspiratory line and having a corresponding expiratory line;
and an aerosol collector in the expiratory line.

2. The apparatus of claim 1 further comprising a feedback tube for returning aerosol collected in the expiratory line to the nebulizer.

3. The apparatus of claim 1 or 2 in which the aerosol collector comprises a jet tube directed at an impact surface.

4. The apparatus of claim 3 in which the jet tube is arranged to entrain aerosol in the expiratory line.

5. The apparatus of claim 3 or 4 in which the jet tube is oriented to deliver a jet directed at a right angle to the impact surface.