INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A gas mixture of helium and oxygen has a lower density than air. By converting regions of density-dependent turbulent airflow within the large airways to laminar flow, helium-oxygen (He-O₂) mixtures can improve expiratory flow and decrease the resistive work of breathing (1). In ambulatory patients with asthma, some investigators reported that inhalation of He-O₂ mixtures decreased dyspnea and pulsus paradoxus, and improved pulmonary gas exchange (5), whereas others reported no benefits (9). In spontaneously breathing patients with severe chronic obstructive pulmonary disease (COPD), inhalation of He-O₂ reduced arterial carbon dioxide tension (12). In mechanically ventilated patients with severe asthma, inhalation of He-O₂ decreased airway resistance and the alveolar-arterial oxygen gradient, and improved CO₂ elimination (13, 14). In patients receiving noninvasive ventilation for acute exacerbations of COPD, He-O₂ decreased dyspnea and work of breathing, and improved gas exchange (15, 16).

The pulmonary deposition of particles within an aerosol is influenced by the density of the gas being administered to the patient (17). Accordingly, He-O₂ is likely to influence pulmonary deposition of aerosolized bronchodilators, yet this interaction has been studied by surprisingly few investigators. Compared with air, Hess and colleagues (18) found that operating a nebulizer with He-O₂ decreased both the fraction of the nominal dose collected on a filter placed at the end of the T connector, and the aerosol's respirable mass in an in vitro model. Conversely, patients with asthma who breathed He-O₂ 80/20 displayed greater pulmonary deposition of monodisperse Teflon particles (~ 3.7 µm aerodynamic diameter) than when breathing air (19, 20). The influence, however, of He-O₂ on aerosol delivery during mechanical ventilation is not known.

In-line nebulizers and metered-dose inhalers (MDIs) are used for bronchodilator therapy in mechanically ventilated patients (21). The efficiency of these devices in delivering aerosols to the lower respiratory tract is less in mechanically ventilated patients than in ambulatory patients (21). Methods for enhancing aerosol delivery in such patients could improve clinical benefit and reduce costs. In a pediatric model of mechanical ventilation, albuterol delivery from a MDI was increased when the device was operated with a He-O₂ 70/30 mixture versus the same balance of nitrogen and oxygen (22). It is not known, however, how differing concentrations of helium might influence the generation of aerosols from MDIs and nebulizers nor their delivery to the lower respiratory tract of mechanically ventilated patients. To avoid unexpected hazard to critically ill patients, we used a mechanically ventilated tracheobronchial model to study the influence of various concentrations of He-O₂ on aerosol delivery from MDIs and nebulizers. In previous studies, we have shown that this in vitro model accurately reflects changes in in vivo delivery of aerosol in mechanically ventilated patients (23, 24).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Description of the Lung Model

Studies were conducted employing a tracheobronchial model that allows investigation of aerosol delivery during mechanical ventilation (Figure 1); this model is similar to that described in our previous publications (23, 24). A Siemens 900C ventilator (Elema AB, Solna, Sweden) was employed because it maintains accurate volumes and flows with He-O₂ mixtures (25). The ventilator provided controlled mechanical ventilation with a tidal volume of 800 ml, a frequency of 12 breaths/min, and a peak inspiratory flow of 40 L/min delivered with a square-wave configuration. A H-cylinder containing He-O₂ 80/20 and a 50 pound per square inch regulator (Puritan Bennett Corp., Lenexa, KS) was attached to the air inlet of a standard oxygen-air blender (Bird Corporation, Palm Springs, CA). The O₂ concentrations were continuously monitored with a fuel cell oxygen analyzer (Hudson, Temecula, CA), placed in the inspiratory limb between the ventilator and the MDI or between the nebulizer and endotracheal tube. To avoid erroneous readings of tidal volume and inspiratory air flow secondary to gas mixtures of varying density, a pneumotachometer (Ventrak, Wallingford, CT) was placed between the test lung and the filters (Figure 1). The pneumotachometer uses an algorithm that allows accurate readings of air flow and volume for specific concentrations of He-O₂. The accuracy of the measured volumes and the linearity of the response of each pneumotachometer was verified with a volumetric syringe containing various concentrations of He-O₂ and air-O₂. Tidal volume, measured by the pneumotachometer, was also compared with the volume displacement of the test lung (Michigan Instruments, Grand Rapids, MI), with measurements varying between the two devices by less than 3%. The quantity of albuterol was collected on a filter with a pore size less than 0.3 µm (Respirgard II bacterial/viral Filter, No. 303; Marquest Medical Products, Inc., Englewood, CO); the albuterol deposited on the filter was measured by spectrophotometry. Experiments with the ventilator were performed with inspired gas at ambient temperature (25 to 27° C) and a relative humidity (RH) of < 30%, or when heated to 35 ± 1° C and RH of 98 ± 2% (Fast-Response Digital Hygrometer/Thermometer; Curtin Matheson, Houston, TX).

View larger version (22K): [in this window] [in a new window]   Figure 1.   In vitro lung model of mechanical ventilation. A tank containing 80% helium and 20% oxygen (He-O2 80/20) was connected to the air inlet of a gas blender and O2 was connected to the O2 inlet. The concentration of He-O2 was adjusted by changing the flow rates of He-O2 80/20 or O2 through the blender. A jet nebulizer connected in the inspiratory limb of the ventilator circuit 45 cm from the endotracheal tube was operated with He-O2 mixtures or O2. A MDI and chamber spacer was placed in the ventilator circuit 15 cm from the endotracheal tube. The tracheobronchial model with filters placed at the distal ends of the main bronchi was used to collect albuterol. A gas analyzer in the inspiratory limb of the ventilator circuit was used to regulate He-O2 concentrations; inspiratory flow was continuously measured with a pneumotachometer. The ventilator circuit was connected to a test lung, and its volume displacement was used to confirm tidal volume.

Experiment 1: Influence of Gas Density on Aerosol Delivery with a MDI

MDI canisters (albuterol; Schering, Kenilworth, NJ; manufacturer estimated dose of 90 µg/puff) were warmed to hand temperature, well shaken, and primed, that is, each MDI was actuated 5 times before testing to ensure that the subsequent actuations provided a homogenous mixture of canister ingredients. Each subsequent actuation of the MDI was discharged into a MDI spacer (Aerovent; Monaghan Medical Inc., Plattsburgh, NY) in the inspiratory limb of the ventilator circuit at the onset of inspiration; successive doses from the MDI were actuated at 15-s intervals. Albuterol was administered from a set of four MDIs actuated in rotation.

To determine the influence of gas density on the delivery of albuterol, 8 puffs (720 µg) of albuterol were administered from a MDI into a chamber spacer during mechanical ventilation with the circuit containing air, O₂, or He-O₂ mixtures of 80/20, 70/30, 60/40, and 50/50. The tests were performed either at ambient conditions of temperature and humidity or with heated and humidified gas mixtures, and were repeated six times with each gas mixture. Albuterol deposition was measured on filters placed at the ends of the bronchi in the lung model (Figure 1).

To determine the effect of gas density on the location of aerosol deposition within the ventilator circuit, 8 puffs (720 µg) of albuterol were administered by MDI and chamber spacer into a dry, unheated ventilator circuit containing He-O₂ 80/20 or O₂. Albuterol deposition was measured on filters placed immediately distal to the chamber spacer, between the elbow connector and the endotracheal tube, and at the ends of the bronchi in the lung model (Figure 1) (n = 3 for each site).

Experiment 2: Influence of He-O₂ on Nebulizer Efficiency

A small-volume nebulizer (AeroTech II; CIS-US Inc., Bedford, MA) was filled with 5 ml of albuterol sulfate solution (Proventil; Schering, Kenilworth, NJ) at a concentration of 0.5 mg/ml (nominal dose 2.5 mg) and placed in the inspiratory limb of the ventilator circuit 45 cm from the circuit wye. Oxygen or He-O₂ 80/20, each at a flow rate of 6 L/ min, was used to generate aerosol continuously throughout the breathing cycle. A pressure-compensated oxygen flow meter (Timeter, St. Louis, MO) was used to adjust gas flow through the nebulizer, and appropriate calibration factors were applied to correct for the lower density of the He-O₂ mixtures (26). After 5 min of nebulization, albuterol deposition was measured on a filter placed immediately distal to the T-piece of the nebulizer.

To determine the effect of gas density on nebulizer efficiency, the nebulizer was operated at a flow rate of 6 L/min using air, O₂, and He- O₂ mixtures of 80/20, 60/40, 40/60, and 20/80. For each gas mixture, albuterol deposition was measured on a filter placed immediately distal to the T-piece of the nebulizer.

To determine the influence of gas flow rates on nebulizer efficiency, the nebulizer was operated separately with He-O₂ 70/30 and O₂ at flow rates of 5, 10, and 15 L/min. At each flow rate, albuterol deposition was measured on a filter placed immediately distal to the T-piece of the nebulizer.

Experiment 3: Optimizing Aerosol Delivery with a Nebulizer during Mechanical Ventilation

To determine which combination of gas mixtures delivers the most aerosol, the nebulizer was operated at three sets of gas composition and flow rates: O₂ at 6 L/min as recommended by the nebulizer manufacturer, He-O₂ 70/30 at 6 L/min, and He-O₂ 70/30 at 15 L/min. Each run was performed in a humidified ventilator circuit containing either O₂ or He-O₂ 70/30; of gases commercially available, He-O₂ 70/30 has the highest level of helium while also ensuring an O₂ concentration of more than 20%. With the nebulizer operated with O₂ at 6 L/min, the final concentration of He-O₂ in the ventilator circuit was reduced to 61/39 ± 2%. Albuterol deposition was measured on filters placed at the ends of the main bronchi in the lung model.

Assay Technique

On completion of each experiment, the filters were labeled, capped and filled with 5 ml of 0.1 M sodium hydroxide, and the albuterol eluted by gently shaking for 24 h. The volume recovered from each filter was recorded and the albuterol concentration determined at a wavelength of 246 nm (DU 64 spectrophotometer; Beckman Instruments, Fullerton, CA) using 0.1 M sodium hydroxide as the reference. Individual experiments were repeated three times on different days with single-blind analysis.

Data Analysis

All results were expressed as the absolute amount of drug in micrograms or as the fraction of nominal dose delivered (mean ± SE). Results were analyzed with repeated measure analysis of variance (ANOVA) using multivariate analysis, with Scheffe's F-test; p < 0.05 was considered significant. The regression coefficient (r) was calculated with statistical software (SuperANOVA; Abacus Concept, Berkeley, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiment 1: Influence of Gas Density on Aerosol Delivery with a MDI

Delivery of albuterol from a MDI and chamber spacer was greater when the ventilator circuit contained higher concentrations of He-O₂: 46.7 ± 3.3% and 43.9 ± 1.0% of the nominal dose for He-O₂ 80/20 and 70/30, respectively. The latter deliveries were higher than the drug deliveries observed with He-O₂ mixtures of 60/40 and 50/50: 39.0 ± 0.9% and 39.9 ± 1.3%, respectively (p < 0.04). Albuterol delivery was significantly greater with each of the He-O₂ mixtures than with air (30.2 ± 1.3%, p < 0.0001) or O₂ (29.1 ± 1.3%, p < 0.0001). Albuterol delivery was inversely related to gas density in the ventilator circuit (r = 0.98, p < 0.005) (Figure 2). For both He-O₂ or air, albuterol delivery was reduced significantly when the ventilator circuit was heated and humidified versus a dry circuit (Table 1). In a humidified ventilator circuit, delivery of albuterol remained higher with He-O₂ 80/20 than with air (p = 0.02).

View larger version (19K): [in this window] [in a new window]   Figure 2.   Albuterol delivery from a MDI to filters, placed at the ends of bronchi, as a function of gas density. Eight puffs (720 µg) of albuterol were administered into a chamber spacer during controlled ventilation in an unheated, dry ventilator circuit containing air, O2, or several mixtures of He-O2 (80/20, 70/30, 60/40, 50/50). Values of albuterol delivered are expressed as a percent of the nominal dose. Albuterol delivery was inversely related to gas density in the ventilator circuit (r = 0.98, p < 0.005)the highest aerosol delivery occurring with the lowest density gas (He-O2 80/20).

The density of gas within the ventilator circuit significantly altered the pattern of aerosol deposition (Figure 3). The enhancement in albuterol delivery to the lower airways with He- O₂ 80/20 over O₂ was mainly due to a decrease in aerosol deposition in the chamber spacer: mean 39.2% versus 55.2% (p < 0.001). Deposition of aerosol in the endotracheal tube was also decreased by He-O₂ 80/20mean 1.4 versus 4.2% for O₂ (p < 0.01) (Figure 3). Albuterol deposition in the ventilator circuit between the chamber spacer and endotracheal tube was comparable for He-O₂ 80/20 and O₂13.3% and 10.2%, respectively (p = 0.86) (Figure 3).

View larger version (23K): [in this window] [in a new window]   Figure 3.   Drug deposition, expressed as a percent of nominal dose of albuterol from a MDI, in the spacer chamber, the ventilator circuit, the endotracheal tube, and on filters at the bronchi under dry conditions (RH < 30%, 27° C) during controlled mechanical ventilation. When the ventilator circuit contained O2 or He-O2 80/20, aerosol deposition differed at the following sites: spacer, ventilator tubing, endotracheal tube, and bronchi. *p < 0.01; **p < 0.001.

Experiment 2: Influence of He-O₂ on Nebulizer Efficiency

When the nebulizer was operated at a flow rate of 6 L/min, the use of He-O₂ 80/20 reduced drug delivery to a filter placed immediately distal to the T-piece of the nebulizer by fivefold compared with O₂: 38.7 ± 0.7 µg versus 191.8 ± 36.4 µg (p = 0.0005). Albuterol delivery from the nebulizer was related to gas density (r = 0.944, p < 0.001), with the most dense gas, O₂, yielding the highest drug delivery (Figure 4). Albuterol delivery from the nebulizer was greater for O₂ and air than for He- O₂ mixtures of 80/20 and 60/40 (p < 0.001) (Figure 4).

View larger version (16K): [in this window] [in a new window]   Figure 4.   Albuterol delivery as a function of the density of the gas used for operating a small-volume nebulizer. Air, O2, and various mixtures of He-O2 (80/20, 60/40, 40/60,20/80) were used to operate a nebulizer at a constant flow rate of 6 L/min. Albuterol delivery (percent of the nominal dose of 2.5 mg) was measured on filters placed immediately distal to the nebulizer. Albuterol delivery from the nebulizer showed a positive correlation with gas density (r = 0.94, p < 0.0001), with the most dense gas (O2) producing the greatest delivery.

When the nebulizer was operated with He-O₂ 70/30, more albuterol was delivered to a filter placed immediately distal to the T-piece of the nebulizer when the flow rate was 10 L/min versus 5 L/min (p < 0.001). Likewise, when the nebulizer was operated with O₂, albuterol delivery was greater with a flow of 10 L/min versus 5 L/min (p < 0.01) (Table 2). At a gas flow of 5 L/min, nebulizer efficiency was 17-fold higher with O₂ than with He-O₂ 70/30 (p < 0.001); at a gas flow of 10 L/min, nebulizer efficiency was twofold higher with O₂ (p < 0.01) (Table 2). For He-O₂ 70/30, an increase in nebulizer flow from 10 to 15 L/min resulted in a 3.5-fold increase in nebulizer efficiency (p < 0.01). It was not possible to operate the nebulizer using O₂ at a flow rate of 15 L/min because a build-up of back-pressure disconnected the tubing from the nebulizer inlet. Albuterol delivery to a filter placed immediately distal to the T-piece of the nebulizer was almost doubled when the nebulizer was operated with He-O₂ 70/30 at 15 L/min than with O₂ at 10/L min (p < 0.0001) (Table 2).

Experiment 3: Optimizing Aerosol Delivery with a Nebulizer by Using Various Gas Mixtures during Mechanical Ventilation

When the nebulizer was operated with a flow of 6 L/min, albuterol delivery to the lower respiratory tract was approximately 50% higher with the use of O₂ versus He-O₂ 70/30 (Figure 5). To obtain aerosol delivery equivalent to that obtained with O₂ at a flow rate of 6 L/min, it was necessary to operate the nebulizer with He-O₂ 70/30 at flow rates of 15 L/min (Figure 5). The highest amount of albuterol was delivered to filters placed at the ends of the tracheobronchial model when the nebulizer was operated with O₂ and the ventilator circuit contained He-O₂ 70/30 (Figure 5).

View larger version (53K): [in this window] [in a new window]   Figure 5.   Determination of drug delivery to filters placed at the ends of the bronchi of the lung model when the nebulizer was operated with O2 at a flow of 6 L/min, with He-O2 70/30 at a flow of 6 L/min and He-O2 70/30 at a flow of 15 L/min, and while the ventilator circuit contained He-O2 70/30 (open bars) or O2 (hatched bars). Albuterol delivery was greatest when the nebulizer was operated with O2 and the ventilator circuit contained He-O2 70/30. Bars represent SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

He-O₂ mixtures can substantially enhance or reduce the efficiency of bronchodilator delivery during mechanical ventilation, depending on specific circumstances. Delivery of albuterol from a MDI to the lower respiratory tract was enhanced when the ventilator circuit contained He-O₂ mixtures as opposed to air or O₂ (Table 1). Conversely, when a nebulizer was operated with He-O₂, instead of O₂, albuterol delivery was reduced (Table 2). The greatest delivery of aerosol to the lower respiratory tract was achieved by operating the nebulizer with O₂ and entraining the aerosol into a ventilator circuit containing He-O₂. Further studies are needed to determine if the increased aerosol delivery with He-O₂ mixtures enhances bronchodilation in mechanically ventilated patients.

Effect of He-O₂ on MDI Delivery

Aerosol therapy is frequently administered to mechanically ventilated patients using a MDI and chamber spacer. We and other (23, 24, 27) have shown that several factors, including tidal volume, inspiratory flow rate, and circuit humidity influence the deposition of aerosol generated by a MDI during mechanical ventilation, but the influence of gas density on aerosol delivery was not known. In this study, we show that aerosol delivery was inversely related to the density of gas in the ventilator circuitdelivery being greatest with the least dense gas, i.e., He-O₂ 80/20 (Figure 2). With helium concentration of 70% or higher, albuterol delivery was increased by half or more over that achieved with air or O₂. When the ventilator circuit was humidified, delivery of albuterol to the main bronchi was decreased; the reduction was virtually identical for He-O₂ 80/20 and airslightly in excess of 60% (Table 1). In the presence of He-O₂ 80/20, aerosol delivery in a humidified circuit was comparable to aerosol delivery in a dry ventilator circuit being ventilated with air.

Mechanisms for Improved Aerosol Delivery with a MDI in Circuits Containing He-O₂

The deposition of aerosol particles in a ventilator circuit and the respiratory tract is mainly achieved through impaction and sedimentation. We previously showed that most of the aerosol emitted from a MDI impacts in the ventilator circuit before reaching the lungs (24). We now show that the amount of albuterol depositing in a chamber spacer and endotracheal tube is decreased when the ventilator circuit contains He-O₂ rather than O₂ (Figure 3), leaving a greater fraction to reach the tracheobronchial region. Given the nature of our model, the influence of He-O₂ on particle sedimentation cannot be determined.

Turbulence in the airways is associated with greater impaction of aerosol particles, reducing aerosol delivery to the lungs (28, 29). With high airflow, Reynolds number, which governs the type of flow within a tube of fixed diameter, can exceed the threshold for nonlaminar flow, i.e., 2,000 units, in the human trachea (30). For a similar flow and airway dimensions, Reynolds number for He-O₂ 80/20 is decreased to approximately one-third that with O₂ (20). This factor reasonably explains the reduction in aerosol deposition in the spacer device, endotracheal tube, and major airways when the ventilator circuit contained He-O₂ rather than O₂ (Figure 3).

The inspiratory flow rates during mechanical ventilation exceed those during spontaneous breathing, and are likely to contribute to turbulent flows with increased aerosol deposition in the central airways. When a radiolabeled aerosol was administered to mechanically ventilated patients by a MDI and chamber spacer, most of it deposited in the central zone of the lung (31). We previously reported that aerosol delivery to filters placed at the ends of the bronchi in our tracheobronchial model was twofold higher at an inspiratory flow rate of 40 L/min than at 80 L/min (24). That is, turbulence in the air stream produced by high ventilator flows appears to increase aerosol deposition within the ventilator circuit and major airways. The improvement in aerosol delivery with He-O₂ in the ventilator circuit may be a consequence of its ability to promote a more laminar pattern at any given flow rate.

Effect of He-O₂ on Nebulizer Efficiency

Nebulizers were originally employed in preference to MDIs for aerosol therapy in mechanically ventilated patients (32). We confirm the findings of Hess and coworkers who reported that both the nebulizer efficiency and the proportion of drug contained within respirable particles are decreased when nebulizers are operated with He-O₂ rather than O₂ at similar flow rates (18). We further show that the amount of drug leaving a nebulizer decreases linearly as a function of gas density (Figure 4). Entrainment of a solution by a gas and the creation of an aerosol during jet nebulization is governed by the Bernoulli principle, which describes the relationship between gas density and velocity with pressures generated at the jet orifice. Reducing gas density or velocity diminishes the pressure drop across a jet orifice, reducing the amount of aerosol generated. Accordingly, we suspected that operating a nebulizer with a higher flow of He-O₂ might increase the amount of aerosol being generated, compensating for the effects of reduced gas density. When the nebulizer was operated with He-O₂ flow of 15 L/min, nebulizer efficiency matched that achieved by operating the nebulizer with O₂ at a flow rate of 6 L/min (Table 2).

Nebulizer Operation with He-O₂ in Mechanically Ventilated Patients

During mechanical ventilation, the gas used in operating a nebulizer may be supplied intermittently, from the ventilator, or continuously from a secondary gas source, such as an air compressor or cylinder. Investigators have previously shown that the intermittent method of nebulizer operation is more efficient for aerosol delivery than continuous aerosol generation (21). Based on our findings, intermittent operation of a nebulizer with He-O₂ mixtures from the ventilator would result in markedly reduced efficiency; accordingly, it is preferable to operate the nebulizer continuously using He-O₂ mixtures. During continuous operation of the nebulizer, a He-O₂ of 15 L/ min was required to match the efficiency achieved with O₂ at 6 L/min. Use of such continuous high flows of He-O₂ through the circuit would waste gas, and also require readjustment of minute ventilation settings during nebulizer operation. Aerosol delivery to the lower respiratory tract of the tracheobronchial model could be optimized by operating the nebulizer continuously with O₂ at 6 L/min and entraining the aerosol in a ventilator circuit containing He-O₂. Although this method of nebulizer operation is tedious to set up, it increases aerosol delivery by approximately 60% over that achieved in a ventilator circuit containing O₂ (Figure 5). While nebulizer performance can be improved by increasing He-O₂ flow rates, nebulizer operation during mechanical ventilation with He-O₂ is less straightforward than use of a MDI. Careful attention to the details of nebulizer operation is required to achieve maximal nebulizer efficiency when using He-O₂ mixtures in the ventilator circuit.

In summary, He-O₂ in the ventilator circuit improved delivery of albuterol from both MDIs and nebulizers. The increase in aerosol delivery from a MDI was inversely correlated with the density of gas, that is, higher concentrations of helium in the ventilator circuit produced higher aerosol delivery. Conversely, efficiency of a nebulizer was markedly reduced when it was operated with helium. When using a He-O₂ 80/20 mixture to operate the nebulizer, 2.5 times higher flow of gas was needed to achieve nebulizer efficiency comparable to that achieved with O₂. Maximal efficiency was achieved when the nebulizer was operated with O₂ and the aerosol emitted was entrained into a ventilator circuit containing He-O₂. In conclusion, with a carefully executed technique of administration, use of He-O₂ mixtures in the ventilator circuit can increase delivery of aerosolized bronchodilators from both MDIs and nebulizers by as much as 50%; this degree of improvement could lead to significantly better outcomes in mechanically ventilated patients with severe airway obstruction.