Pressurized Aerosol versus Jet Aerosol Delivery to Mechanically Ventilated Patients Comparison of Dose to the Lungs 1 - 3

H. D. FULLER, M. B. DOLOVICH, G. POSMITUCK, W. WONG PACK, and M. T. NEWHOUSE

Introduction

In spontaneously breathing normal subjects, the dose of aerosol deposited in the lungs is about 10070 of that released from a metered-dose inhaler (MDI) (1) and less if good inhalation technique is not practiced. A number of add-on devices (such as the Aerochamber") (2) have been developed for use with the MDI and these routinely enable lung deposition rates comparable to the best of those obtained using an MDI alone (1), but with simpler technique. Less work has been done assessing aerosol delivery to patients receiving intermittent positive-pressure ventilation (IPPV) from a mechanical ventilator. With different airway pressures and flow rate patterns, aerosol deposition can be expected to be different. Aerosol generation in this setting is usually achieved by means of jet nebulization of bronchodilator solution on the inspiratory side of the ventilator circuit. Using such a system, aerosol deposition to the lungs has been found to be no more than 1 to 3 0J0 of the original volume of solution (3-5). We have designed an aerosol holding chamber that is inserted in-line in the inspiratory arm of a ventilator circuit and allows the use of a pressurized aerosol MD I for bronchodilator treatment. By means of a randomized controlled trial, we have compared aerosol deposition to the lung from MDI and chamber with the conventional jet nebulization delivery. Methods Subjects wererecruited for the study from the patient population of our intensive care unit if they were being maintained on IPPV and were receivinginhaled bronchodilator as part of their treatment. Informed consent (approved by the hospital's Ethics Committee) was signed by the patient (or, if incompetent, by the next of kin). After recruitment, the subjects wererandomized into one of two groups by a person not otherwise involvedin the study

440

SUMMARY The purpose of this study was to compare deposition of aerosol to the lung from a metered-dose inhaler (MOl) and aerosol holding chamber and from a jet nebulizer in ventilatordependent patients. Twenty-one patients were entered into the study, all receiving assisted ventilation and inhaled bronchodilators because of airflow limitation. The average age was 68 yr; there were 10 men and 11 women. The patients were randomized to receive either 4 puffs (800 J.1g) of radiolabeled fenoterol by MOl of 1.75 ml (1,750 J.1g) of radiolabeled fenoterol solution by nebulizer. Imaging of lung fields was made by a portable scintillation camera at 5-min intervals during the study. Results showed that 20 patients completed the study, 9 receiving fenoterol by MOl, and 11 by jet nebulizer. Four we~ excluded from analysis because of previous pneumonectomy, two from each group. Lung depositl'on measured as a percent of given dose from either system was 5.65 ± 1.09 (mean ± SEM) for MOl plus extension chamber and 1.22 ± 0.35 for jet nebulizer (p < 0.001).Therefore, this trial shows significantly greater efficiency of aerosol deposition to the lung In ventilatordependent patients when using an MOl plus aerosol holding chamber than when using a jet nebulizer. AM REV RESPIR DIS 1990; 141:440-444

but with access to a randomization table. All subjects had their inhaled bronchodilators withheld for a minimum of 4 h before the study, and none were receiving parenteral bronchodilators. Subjects randomized to Group A received one puff (200 ug) of fenoterol labeled with technetium pertechnetate (99mTc04 -) (Berotee; Boehringer Ingleheim, Burlington, Ont.) at 5-min intervals for a total of 4 puffs (800 ug). The drug was administered through a lead-shielded in-line cylindrical chamber measuring 10 em long and 5 em in diameter (figure IA), towards the end of expiration, without end-expiratory breathhold. The chamber was positioned on the inspiratory limb of the ventilator circuit, 15 ern from the end of the endotracheal tube (figure IB). Subjects randomized to Group B received 1.75 ml (1,750 Jlg) of fenoterol solution to which 5 mCi of 99mTc sulfur colloid had been added. The volume of solution was then made up to 3 ml using 0.90/0 sodium chloride solution. Nebulization by means of a Bennett Twin-jet nebulizer (figure lC) (Bennett Respiration Products, Los Angeles, CA) was performed during the inspiratory phase for a total of 15 min and then removed from the ventilator circuit. Except for the D.S-ml "dead" volume remaining in the nebulizer, aerosolization of the solution was usually complete after this time. This time frame was therefore used for standardization of method.

Labeling of Fenoterol Radiolabeling of the MDI was performed using the technique of Kohler and coworkers (6). Briefly, the MDI was cooled to -70 0 C, and approximately 250 mCi of 99mTc-per_ technetate together with surfactant and freon 11 were added through a hole punctured in the canister bottom. The canister was then sealed and allowed to return passivelyto room temperature before use. The total MDI activity at the time of use was 53.3 ± 9.5 mCi, yielding 170 ± 32 u.Ciper puff. The aerosol particle size of several labeled canisters (actuated via MDI without in-line chamber) was determined using a cascade impactor (7) with sampling under isokinetic conditions (2). Impactor slides were assayed for both radioac-

(Received in originalform December 27, 1988 and in revised form June 26, 1989) I From the Departments of Medicine and Nuclear Medicine, S1. Joseph's Hospital, McMaster University, Hamilton, Ontario, Canada. 2 Supported by Trudell Medical, London, Ontario, and by Boehringer lngleheim (Canada) Ltd., Burlington, Ontario, Canada. 3 Correspondence and requests for reprints should be addressed to Dr. H. D. Fuller, Department of Medicine, S1. Joseph's Hospital, 50 Charlton Avenue East, Hamilton, Ontario L8N 4A6, Canada.

PRESSURIZED AEROSOL VERSUS JET AEROSOL DELIVERY TO VENTILATED PATIENTS

A

5

t

CM

--, +

. . - - 10

CM

---+

B EXPIRATORY LINE

ET TUBE

~ ~ C:~:S~ER ~

.. 20 eM ..

INSPIRATORY LINE

c EXPIRATORY LINE

ET TUBE

+---

70 eM ---+ INSPIRATORY LINE

Fig. 1. Details of the ventilator circuit. A. Dimensions of the aerosol holding chamber. B. Details of circuit for experiment in Group A. C. Details of ventilator circuit for experiments in Group B.

tivity using gamma camera and drug by UV spectrophotometry. There was no significant difference observed in mass median aerodynamic diameter (MMAD) between radioactivity (3.34 J.1m; geometric SD, 1.83) and drug (2.78 J.1m; GSD, 2.28)(MMAD p = 0.06, GSD p = 0.11), and the proportion of respirable particles within the aerosol, that is, with diameter < 5 jlm, microns was the same (p = 0.27). Therefore, it was assumed that the radiolabel and drug distributed in a similar way in vivo. Aerosols from the jet nebulizers were not sized, as three types of ventilators were used interchangeably, with settings varying between patients. Before each actuation, the canister was vigorously shaken to dispersethe drug throughout the labeled freon. Measurement of the radioactive delivered dose was made as follows: Group A: The labeled canister was weighed before use. Single puffs werecollected onto cotton balls stuffed into the actuator mouthpiece, which had been inserted into a surgical glove for ease of handling. Each cotton ball was removed from the actuator into a glove, and the radioactivity was measured in a dose calibrator. The first puff of each MDI used was considered a valve-priming puff, and its value was not included in the data. This was repeated approximately 8 to 10 times before each experiment and again 8 to 10 times after the completion of the experiment. Postexperimental puffs reflected a

441

lower number of microcuries per puff as a result of decay of the radiolabel 99mTc. When corrected for this, the values were not significantly different from the pre-experimental calibration values. The radioactive dose per puff given to each subject was calculated as the mean of all the pre-experiment calibration puffs. Weight loss per puff and total weight and radioactivity loss of the canister werealso determined. This confirmed the uniformity of each activation and lack of leakage of the canister, which, if present, would havereduced canister pressure and amount delivered to the lung. Group B: The radiolabeled fenoterol solution was counted in the dose calibrator before administration. All counts were timecorrected for decay of the 99mTc. Measurement of radioactivity deposited in the lungs was made by means of a portable gamma camera with diverging collimator (GE 300AT mobile Starcam; General Electric, Markham, Ont.) at the patient's bedside. One-minute real-time anterior images wereacquired as follows: For subjects in Group A, images were acquired immediately after each puff. For subjects in Group B, images were acquired at 5, 10, and 15 min after the start of the experiment. The lungs were outlined by manual positioning of the cursor, and the amount of radioactivity in cpm was determined for each image. In order to determine whether or not the dosage of fenoterol was adequate, peak inspiratory pressure was measured before each experiment at 5, 10, and 15 min into the experiment and at 30 min after the start of each experiment. Analysis Data for percent deposition of aerosol in the

lung was analyzed using the t test for independent groups. Inspiratory pressure data wereanalyzed using two-wayanalysis of variance (the method of unweighted means) with time and device as the two variables. Results

Twenty-one patients were recruited, of whom 20 had imaging completed as outlined in the protocol. The average age of the subjects was 68 yr; 15were being ventilated via an endotracheal tube and 5 via tracheostomy. Four were excluded (two from each group) because of pneumonectomy or lobectomy so as not to introduce the variable of altered area for gas exchange. The type of ventilator used and the ventilator settings were not standardized but rather left as they had been set in the patient's management in order to generalize the results to "normal usage." All patients were receiving assistcontrol ventilation at the time of study. The ventilators used in the experiment wereBennett MA-I and MA-2+ 2 and the Bayer2 volume-cycled ventilator. The subjects were randomized 9 to Group A and 11 to Group B. Demographic and diagnostic details are shown in table 1. Deposition of aerosol to the lung was measured as the cumulative count from four puffs of the 99mTc0 4 - fenoterol (Group A) or after a I5-min nebulization (Group B). It was expressed as a percentage of the total dose from four puffs (Group A) or of the total initial nebuliz-

TABLE 1 DEMOGRAPHIC, DIAGNOSTIC, AND DEPOSITION DATA OF THE 16 SUBJECTS ANALYZED Subject No.

Sex

VT

Deposition Route

Diagnosis

Deposition to Lungs as a % of Given Dose

M M F M F F M

650 450 500 700 400 600 800

Trach ETT ETT ETT ETT Trach ETT

Chronic airflow limitation Cardiogenic pulmonary edema Cardiogenic pulmonary edema Bronchogenic carcinoma Chronic airflow limitation Chronic airflow limitation Sepsis + ARDS

2.17 8.80 2.97 3.65 9.60 6.59 5.76

Age (yr)

Group A, MDI 1 71 2 92 3 72 4 68 5 70 6 65 7 67 Mean SEM Group B, nebulizer 1 70 2 69 3 80 4 66 5 70 6 52 7 55 8 67 9 65

5.76 1.09 F F M F M M M F F

700 500 700 600 700 800 800 700 600

ETT Trach Trach ETT ETT ETT ETT ETT ETT

Sepsis + ARDS Chronic airflow limitation Chronic airflow limitation Chronic airflow limitation Sepsis + ARDS Sepsis + ARDS Chronic airflow limitation Cardiogenic pulmonary edema Chronic airflow limitation

1.22 0.35

Mean SEM". Definition of abbreviations: Trach

1.15 0.80 3.68 0.59 0.59 0.50 1.70 0.37 1.59

=

tracheostomy; ETT

=

endotracheal tube; MOl

=

metered-dose inhaler.

442

FULLER, DOLOVICH, POSMITUCK, WONG PACK, AND NEWHOUSE

W

z

~

w

III

-e

.

5 0

III

:I

-5

a:: w

-10

< ::t

Z

-15

at

-20

0

IL I

...............

............. .

[] MOl (n.7) • NEBULIZER (n.9)

~

II

0

IL

ii: 1 puff

3 puffs

-25

5

t 10

I} 15

I ;--;0

TIME FROM START (mina)

..

Fig. 4. Peak inspiratory pressure (mean ± 1 SEM), represented as percentage change from baseline, shown over time with bronchodilation. There was no significant difference between the two groups .

..

... .............. 4 puffs

2 puffs

Fig. 2. Scintigraphs of aerosol deposition to the lungs of a representative patient using MOl and aerosol hold ing chamber. Cumulative depos ition from one to four puffs is shown.

er dose (Group B) (table I). As nebulizer "dead" volume varies between nebulizers and is not usually measured in the clinical setting, no correction was made for the amount of radioactivity remaining in the nebulizer "dead" volume at the end of nebulization. Representative scintiphotographs are shown in figure 2. There was a greater percent deposition from the MOl plus chamber (5.65 ± 1.090/0 [SEM)) than from the nebulizer (1.22 ± 0.35%[SEM)), and this difference was statistically significant (p < 0.001) (figure 3). Peak inspiratory pressure was mea-

o MDlln;7)

10

• NEBULIZER (n • 9)

Z

0

;::

8

iii 0

"w

6

ez

4

....I

2

C

::l

~

0

1 f

~

t·

2

3

4 PUFF (MOl)

5

10

15 min (NEBULIZER)

•

~.

Fig. 3. Deposition of aerosol to the lungs (mean ± 1 SEM) expressed as a percentage of delivered dose for patients in both groups. Statistical sign ificance between groups is seen after four puffs or at 15 min (p < 0.001).

sured on all patients at appropriate times, with the exception of two patients (one in each group) in whom the 30-min value was not taken. Analysis therefore excluded these missing values. Peak inspiratory pressure did not change significantly from baseline values in either group, and there was no difference between groups (figure 4). Discussion

This study has shown that 5.65 ± 1.09% of the dose of bronchodilator released from a MOl plus aerosol holding chamber is deposited in the lungs of patients receiving mechanical ventilation. This is considerably less than the amount deposited from MOl plus chamber in asthmatics with spontaneous ventilation, and presumably is due to the altered mechanics of volume-cycle mechanical ventilation (1). By contrast, however,the percent deposition from a jet nebulizer is considerably lower at 1.22 ± 0.35%. This result is similar to that found by MacIntyre and coworkers (4) who performed deposition experiments using jet nebulization in a variety of patients requiring bronchodilators and receiving mechanical ventilation. Flavin and coworkers (5) found between 0.2 and 1.5% deposition to rabbit lungs from various jet nebulizers. Fraser

and colleagues (3) reported a 5% deposition to a laboratory lung model using a jet nebulizer, but this is not directly comparable, and we have found in pilot studies using a similar laboratory model considerably greater deposition than in vivo. Therefore, it would seem that our results using the jet nebulizer are comparable to those found by others. To our knowledge, there has been no work previously done on aerosol deposition from a MDI and aerosol holding chamber in patients receiving mechanical ventilation. The difference between deposition from the two devices may be due to differences in aerosol droplet size, drug solubility, absorption of drug to ventilator tubing, etc., or to position of the device in the circuit. As can be seen in figure 1, the chamber is situated closer to the patient than the nebulizer. This was done deliberately because it is standard practice for the nebulizer to be positioned on the ventilator manifold. However, consideration could be given to moving the nebulizer to a position closer to the endotracheal tube. This has not been reported in vivo, but in a recent publication (8), a comparison of percent deposition in a lung model on a ventilator circuit found that when the nebulizer was positioned closer to the model, the .percent of given dose that reached the endotracheal tube was reduced 20 to 50%. Therefore, it would appear that repositioning the nebulizer closer to the patient would not improve percent deposition to the lungs, and it may adversely affect it. Support for there being a real increase in efficiency in the MDI plus chamber when compared with the nebulizer is given in separate experiments by Madsen and coworkers (9) and Pedersen and Bundgaard (10).

PRESSURIZED AEROSOL VERSUS JET AEROSOL DELIVERY TO VENTILATED PATIENTS

Madsen and coworkers found that during spontaneous ventilation an equivalent degree of bronchodilatation was achieved with approximately one-quarter the dose of terbutaline when administered by MOl plus chamber than when given by jet nebulizer. Pedersen and Bundgaard found that the maximal increase of FEV 1 after inhalation of 1 mg of terbutaline by MOl plus pear-shaped spacer was approximately three times that seen after inhalation of 1 mg of nebulized terbutaline. The difference in deposition between the two devices is probably due to more than one of the discussed mechanisms, but there is no evidence that different placement in the circuit of the devices that were tested was responsible for this difference. In addition to being a more efficient delivery device, the MOl plus chamber offers the following advantages over the jet nebulizer in ventilated patients. The drug is much more easily delivered as a MOl aerosol, requiring less technologist time. Dosing is more accurate and eliminates the necessity for individual preparation of solutions with associated potential errors. Another source of variability with the jet nebulizer is the determination of completion of aerosolization. When the nebulizer is visually empty, there is still a significant amount of drug left unaerosolized on the inside surface of the nebulizer reservoir and on the tubing. Thus, the total dose to the patient can only be estimated. Much of this variability is avoided by the use of MOL Because of the aqueous nature of the nebulized solution, there is potential for airway infection using a jet nebulizer, and this potential will be reduced with the use of a MOL Finally, there will be considerable cost saving because of the lower cost per delivered drug dose from MOl (the smaller amount of drug used) and the reduced time required of the respiratory technologist or nurse for administration of the drug (tables 2 and 3). The true cost saving resulting from use of MOl is probably greater than the conservative estimate in table 2, largely because many patients receive, in addition, ipratropium and/or higher doses of bronchodilators. The lack of change in peak inspiratory pressure after bronchodilation with either device may reflect one or more of the following points. First, the patient population receiving bronchodilators in the ICU may have only minimally reversible airflow obstruction during the acute exacerbation of their disease that require

443 TABLE 2

COST BENEFIT OF DELIVERING BRONCHODILATOR BY MOl VERSUS NEBULIZER (EXPRESSED IN CANADIAN DOLLARS) Cost of 1 x 200 dose fenoterol MOl Cost of four puffs fenoterol Hourly cost of one respiratory technologist Cost of RT time to administer four puffs (2 min)

11.39 0.23 15.00 0.50

Total cost of delivering four puffs Cost Cost Cost Cost

of of of of

0.73

20 ml x 0.1% fenoterol solution 1.75 ml fenoterol solution 3 ml saline diluent RT time to administer solution (4 min)

$0.73

12.23

Total cost of delivering 8 ml solution

1.07 0.31 1.00 $2.38

2.38

Savings per patient dose of MOl over nebulizer

$1.65

TABLE 3 STATISTICS FROM ST. JOSEPH'S HOSPITAL, HAMILTON FOR 1988 Ventilated patients, n Average ventilated days per patient Estimated proportion of ventilated patients receiving bronchodilators, % Usual number of treatments per day Estimated total number of treatments Estimated annual cost saving in our ICU using MOl rather than nebulizer

assisted ventilation. The fact that many of them are known to have been reversible in the past when relatively well and in stable condition does not exclude this possibility. Second, the measurement of peak inspiratory pressure may be an insensitive or inappropriate measure of airflow resistance. However, the peak inspiratory pressure reflects both elastic and resistiveproperties of the lung. Therefore, a reduction in resistance as might be expected after bronchodilator should lead to reduction in the peak inspiratory pressure. This is supported by a recent study (11) in which there was a 20070 reduction in peak inspiratory pressure after inhalation of 0.5 mg of nebulized ipratropium. This decrease is both clinically and statistically significant and suggests that we wereusing inadequate doses of bronchodilator in our study. Ruffin and coworkers (12)found that in patients with stable asthma, a dose of 25 ug of fenoterol to the lungs produced maximal bronchodilatation. The dose to the lungs by MOl plus chamber in our study was approximately 40 ug. It therefore seems likely that in acutely ill patients with severe airflow obstruction, a much larger dose to the lungs would be required in order to reverse the bronchoconstriction (13) as has been shown in severe acute asthma (14). Tfiis study has demonstrated that a

551 5 30 6 5,400 $8,910.00

MDI plus aerosol holding chamber delivers a nearly fivefold greater dose of aerosolized drug to the lungs in comparison with a jet nebulizer in patients receiving mechanical ventilation. A dose of 20 ug of fenoterol to the lungs requires nebulization of 1.7 mg (1.75 ml) of fenoterol solution, but it can be achieved by only 400 ug (two puffs) from a MDI plus chamber. In addition, it would appear likely that considerably larger doses of bronchodilator are necessary to produce dilatation in this population. Acknowledgment The writers wish to thank Lisa Gerrard for labeling of the Beroteccanisters;Carole Chambers for technical assistance; members of the Respiratory Therapy Department, St. Joseph's Hospital, Hamilton, for measurement of airway pressure; Colleen Saunders and Debra Anne Keay for manuscript preparation. References 1. Dolovich M, Ruffin R, Con D, Newhouse M. Clinical evaluation of simple demand inhalation MOl aerosol delivery device. Chest 1983;84:36-41. 2. Con D, Dolovich M, McCormack D, Ruffin R, Obminski G, Newhouse M. Design characteristics of portable breath actuated particle size selective medical aerosol inhaler. J Aerosol Sci 1982; 13:1-7. 3. Fraser I, Duvall A, Dolovich M, Newhouse M. Therapeutic aerosol delivery in ventilatory systems. Am Rev Respir Dis 1981; 123(Part 2:107). 4. Macintyre N, Silver R, Miller C, Schuler F,

444 Coleman E. Aerosol delivery in intubated, mechanically ventilated paients. Crit Care Med 1985; 13: 81-4. 5. Flavin M, MacDonald M, Dolovich M, Coates G, O'Brodovich H. Aerosol deliveryto the lung with an infant ventilator. Pediatr Pulmonol1986; 2:36-9. 6. Kohler D, Fleischer W, Matthys H. A new method for easy labelling of B2 agonists in the metered dose inhaler with 99 mTc. Respiration 1984; 46(Suppl 1:99). 7. Mercer T, Tilley M, Newton G. A multi-stage low flow-rate cascade impactor. J Aerosol Sci 1970; 1:9-157.

FULLER, DOLOVICH, POSMITUCK,

8. Hughes J, Saez J. Effects of nebulizer mode and position in a mechanical venilator circuit on dose efficiency. Respir Care 1987; 32:1131-5. 9. Madsen E, Bundgaard A, Hidinger K. Cumulative dose response study comparing terbutaline pressurizedaerosol administration via a pear-shaped spacer and terbutaline in a nebulized solution. Eur J Clin Pharmacol 1982; 23:27-30. 10. Pedersen J, Bundgaard A. Comparative efficacy of different methods of nebulizing terbutaline. Eur J Clin Pharmacol 1983; 25:739-42. 11. Legare M, Petrof B, Simkovitz P, Goldberg P, Gottfried S. Aerosolized ipratropium bromide

worm

PACK, AND NEWHOUSE

in mechanically ventilated COPD patients (abstract). Am RevRespir Dis 1988; 137(Part2 of 2:60). 12. Ruffin R, Kenworth M, Newhouse M. Response of ashmatic patients to fenoterol inhalation: a method of quantifying the airway bronchodilator dose. Clin Pharmacol Ther 1978; 23:338-45. 13. Jenkins S, Heaton R, Fulton T, et 01. Comparison of domiciliary nebulized salbutamol to a metered dose inhaler in stable chronic airflow limitation. Chest 1987; 91:804-7. 14. Morgan M, Singh B, Frame M, Williams S. Terbutaline aerosol given through pear spacer in acute severe asthma. Br Med J 1982; 285:849-50.