INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Heparin may be conveniently administered by inhalation, and as it has anti-inflammatory properties besides its anticoagulant and antithrombotic properties, the effects of inhaled heparin on bronchial asthma have been the subject of several studies (1). However, the results of these studies have been inconsistent possibly because various combinations of heparin formulations and concentrations have been used in a variety of ultrasonic and jet nebulizers without quantifying the dose of heparin deposited in the lower respiratory tract (LRT). Thus, the doses administered to the LRT are likely to have varied considerably from study to study.

Furthermore, different populations were studied and the effects of heparin evaluated using bronchial provocation with exercise, allergens, histamine, methacholine, adenosine, and metabisulfite (1) making accurate comparisons difficult.

We previously characterized the particle size distribution and output of several formulations and concentrations of heparin nebulized by jet and ultrasonic nebulizers (8). We found that sodium heparin in the highest concentration, nebulized at a driving flow rate of 10 L/min with a Sidestream jet nebulizer gave the greatest estimated LRT dose. Of a loading dose of 4 ml of sodium heparin solution (80,000 international units), 15,000 IU was deposited on an inhalation filter. The mass median diameter (MMD) was 2.01 µm and the geometric standard deviation (GSD) 2.46. This corresponded to an estimated dose of 10,000 IU of heparin having a high probability of reaching the LRT in normal, healthy adults (8).

The aim of this study was to determine in normal healthy volunteers: (1) the percentage of radiolabeled heparin deposited in the LRT after a single nebulization; (2) the distribution of inhaled heparin in peripheral, intermediate, and central airways; and (3) the kinetics of heparin aerosol clearance.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nebulizers and Heparin

Sodium heparin in a concentration of 25,000 IU/ml (Leo Pharmaceutical, Copenhagen, Denmark) and five Sidestream jet nebulizers (Medic-Aid, Bognor Regis, UK) from the same lot were used.

Radiolabeling Methods

To label the sodium heparin, 2 ml of ^99mTc-labeled heparin was prepared by mixing 1.6 ml heparin (40,000 IU) with 0.2 ml of stannous chloride (0.4 mg) and 0.2 ml ^99mTc (10 to 18 mCi). The radiolabeling was carried out by the Radiopharmaceutical Department of McMaster University, Hamilton, Ontario, Canada. A volume of 2 ml of sodium heparin was added, resulting in 4.0 ml of ^99mTc-heparin in a concentration of 22,500 IU/ml. The total nebulizer charge was thus 90,000 IU heparin. The quality of labeling was previously evaluated by paper chromatography (Whatmann No. 1, ethylenediamine tetraacetic acid [EDTA] buffer and methyl ethyl ketone), and uniformity of labeling by gel column chromatography (Sephacryl S-200 HiPrep 16/60; Pharmacia Biotech, Freiburg, Germany), 0.5 M NaCl, linear flow rate 21 cm/h). Less than 5% of free pertechnetate (TcO₄) and less than 10% of Tc-colloids were found with paper chromatography, and sodium heparin was labeled uniformly with ^99mTc (9).

To examine the in vivo stability of the radiolabeled heparin, plasma from two subjects was drawn 1 h after termination of inhalation, and subjected to paper (Whatmann No. 1) chromatography with methyl ethyl ketone as solvent. This yields the fraction of free technetium (pertechnetate) in the blood.

Validation of Radiolabeling Methods

Experiments were performed to evaluate whether the radiolabeling process had any effect on the particle size distribution of heparin from the Sidestream jet nebulizer. Particle characterization was undertaken with a seven-stage nonviable Andersen Cascade Impactor (Graseby Andersen, Smyrna, GA) calibrated at its designated flow of 28.3 L/ min. A volume of 4 ml of ^99mTc-heparin was loaded in the Sidestream jet nebulizer, driven at a flow rate of 10 L/min, and nebulized for 1 min into the cascade impactor (n = 4). The aerosol was fractionated into the mass of heparin and amount of ^99mTc on the inlet throat, and on the seven impactor stages. The amount of ^99mTc on each impaction stage was determined with a gamma camera (MaxiCamera; General Electrics, Plainville, CT). The concentration of heparin was determined as antifactor IIa activity by Leo Pharmaceutical. Measurements of particle size distributions were compared with determinations made by a laser diffraction method (Malvern MasterSizer X; Malvern Instruments, Malvern, Worcestershire, UK) with both labeled and unlabeled heparin.

Study Population

Fifteen healthy volunteers (10 females, 5 male) were included in the study (mean age 36 yr), all of whom had normal lung function (FEV₁ 94.4 ± 3% predicted, FVC 92.9 ± 2.4% predicted, and FEV₁/FVC 103.1 ± 2.3% predicted). All volunteers had coagulation variables within the normal limits (activated partial thromboplastin time [APTT] 28.4 ± 3.3 [predicted 25 to 40] s; prothrombin time, international ratio [INR] 1.03 ± 0.05 [predicted 1.0 to 1.3]; platelets 228 ± 44 [predicted 150 to 300 · 10⁹/L]). Exclusion criteria were pulmonary diseases, pregnancy or breast feeding, smoking, heparin allergy, anticoagulant therapy, and hemorrhagic disorders. None of the volunteers had any concurrent diseases, except one female with Raynaud's syndrome. Each volunteer provided written informed consent. The study was approved by the local ethics committee.

Administration of Radiolabeled Aerosol

A volume of 4 ml of ^99mTc-labeled sodium heparin (concentration 22,500 IU/ml, total 90,000 IU) was loaded in the Sidestream jet nebulizer operated at a flow rate of 10 L/min. Subjects were fitted with a noseclip and then inhaled for 15 min through a mouthpiece until visible aerosol generation stopped.

All subjects were asked to breathe from the nebulizer with "normal" tidal respiration. The respiratory pattern was monitored via a breathing circuit set up as a one-way, partially closed system (10). Because of the driving flow rate of 10 L/min, no diluting air was needed. The aerosol was inhaled through a one-way valve, and exhaled through an absolute bacterial filter (Respigard II; Marquest Medical Products, Inc., Englewood, CO). The inspiratory and expiratory lines passed into a 5-L "bag-in-box" system. The inspired volume was thus displaced from the box by the balloon. A respiratory flow transducer (Model HP47304A; Hewlett-Packard, Palo Alto, CA) was connected to the box, and the inspiratory and expiratory flow rates and respiratory frequency were measured with a pneumotachygraph (A. Fleisch, Lausanne, Switzerland) on a chart recorder. The pneumotachygraph was calibrated before each study with a rotameter (Brooks Instrument Division, Emerson Electric Company, Hatfield, PA). The respiratory frequency and peak inspiratory flow were read from the tracing for each subject.

Scintigraphic Measurements

A transmission scan was obtained before the first study day to determine an attenuation correction factor, and to obtain outlines of the lungs for subsequent definition of central, intermediate, and peripheral regions of interest. The regions of interest were defined by area (Figure 1). Scintigraphic images of the chest (anterior and posterior), oropharynx (lateral), and the stomach (anterior) were recorded using a gamma camera, immediately after heparin inhalation. Scintigraphic images of the residual heparin in the nebulizer, the exhalation filter, and tubing were also obtained.

View larger version (22K): [in this window] [in a new window]   Figure 1.   The regions of interest in the lungs. The boundaries between the peripheral, intermediate, and central zones have been drawn so that the zones are of equal area. Similarly, the upper and lower zones also have the same area.

Regions of interest were drawn around the lungs, the oropharynx, and the stomach. The counts obtained within these regions were corrected for background radioactivity, radioactive decay, and for tissue attenuation; in regions where both anterior and posterior images were recorded, the geometric mean of counts in both images was calculated before correcting for tissue attenuation. The nebulizer charge was fractionated into the percentages of activity at the mouth, in the lungs, oropharynx, stomach, retained in the nebulizer residual, in the tubing, and on the exhalation filter. Subsequent lung scintigrams (anterior and posterior) were recorded 0.25, 0.5, 1.0, 1.5, 3, 4, 6, and 24 h after conclusion of inhalation. Five subjects had additional scintigrams taken 22 h after inhalation. The percentage radioactivity at each time point was calculated as a percentage of the maximal counts recorded immediately after heparin inhalation.

Dynamic posterior lung scintigrams (30-s frames) during inhalation of ^99mTc-heparin were obtained for five of the subjects.

Blood Collection, Processing, and Analysis

Blood samples were drawn from the first 10 subjects at baseline, after 5 and 10 min during nebulization, and 1, 3, 5, 10, 20, 30, 45, 60, 90 min, 2, 3, 4, 6, and 24 h after conclusion of heparin aerosol inhalation, through a catheter placed in an arm vein. Blood samples from the last five subjects were drawn at baseline, 5, 10, 15, 30, 60, 90 min, and 4, 6, 22, and 24 h after start of inhalation, by venipuncture in the antecubital vein. Blood was collected into citrated vacutainer tubes (Becton Dickinson, San Jose, CA). The first 3 ml of blood was discarded and 3 ml of blood was then drawn for measurement of radioactivity, followed by 5 ml of blood for coagulation assays. Within 1 h after sampling, the red blood cells were sedimented by centrifugation at 3,000 g for 15 min. The supernatant from the 5-ml tubes was spun for another 10 min at 3,000 g, and the resulting platelet-poor plasma was stored at 70° C for later analysis. Radioactivity in 1 ml of plasma was determined with a well counter (Model 300C Nuclear Spectrometer; Specialties Electronics Co. Inc., NJ).

The amount of heparin present in the blood over 24 h was calculated as the area under the curve of the profile of activity in the plasma (converted to units of heparin); the plasma volume was estimated from body weights using a value of 43 ml/kg for women, and 48.9 ml/kg for men (11).

The following coagulation assays were performed: P-thrombin fragment 1 + 2 (F₁₊₂) (Enzygnost F1 + 2 ELISA microkit; Behringwerke, Marburg, Germany), P-thrombin-antithrombin (TAT) (Enzygnost TAT ELISA microkit; Behringwerke), factor Xa-antithrombin III (FXa-ATIII) (12), and factor Xa-tissue factor pathway inhibitor (FXa-TFPI) (13).

Statistics

Statistical analyses were performed using SOLO for DOS (BMDP Statistical Software, Los Angeles, CA), and EXCEL for Windows (Microsoft, Redmond, WA). All values are expressed as mean ± SD.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Validation of Radiolabeling Method

The labeling of heparin aerosols with ^99mTc did not affect the aerosol characteristics. Deposition on the impaction stages is shown in Figure 2. Mass median aerodynamic diameter (MMAD) determined by the Andersen cascade impactor using the radiolabeled heparin was 1.74 ± 0.10 µm (GSD 1.94 ± 0.06), and with heparin antifactor IIa activity 1.76 ± 0.12 µm (GSD 1.98 ± 0.03). The percentage of particles below 3 µm was 82 ± 3% and 79 ± 7%, respectively. MMD (GSD) was 2.48 ± 0.04 µm (2.14 ± 0.10) determined by laser diffraction. 61 ± 1% of particles were below 3 µm. These values are comparable to previous studies with nonlabeled heparin (8). Plots of the cumulative percentages of the heparin concentration (antifactor IIa) and the radioactivity showed that MMAD as well as the particle size distribution were unchanged by radiolabeling. Thus, ^99mTc-labeled heparin aerosols and unlabeled heparin aerosols behave similarly and the former is thus suited for studies in humans.

View larger version (35K): [in this window] [in a new window]   Figure 2.   Summary of the radiolabeling validation data for the Sidestream jet nebulizer. Mean (SD) particle size distribution of labeled drug and radiolabel are shown. The figure demonstrates close agreement between the heparin assay and 99mTc-heparin. Stages marked on y-axis refer to the Anderson Cascade Impactor. The cutoff diameters are mentioned after the stage number.

Inhalation Details

All subjects were encouraged to breathe with "normal" tidal breathing. Their mean respiratory frequency was 13.33 ± 4.85 breaths/min. Three subjects had frequencies of 18, 21, and 25 breaths/min. The average peak inspiratory flow was 29.42 ± 3.07 L/min. There was no correlation between the respiratory frequency/peak inspiratory flow and the regional deposition in the lung, or between the percentage of ^99mTc-heparin presented at the mouth, deposited on the exhalation filter, or remaining in the nebulizer residual.

Deposition Data

The deposition data expressed as a percentage of nebulizer charge are summarized in Table 1. Mean lung deposition was 7.62 ± 1.76% of the nebulizer charge, and 14.84 ± 2.82% of the nebulizer charge reached the mouth. The residual in the nebulizer was 47.91 ± 5.59%, 32.36 ± 3.61% was found on the exhalation filter, and 4.91 ± 1.78% in the tubing. 6.37 ± 2.07% of the initial amount of ^99mTc-heparin could not be accounted for. This was not caused by leakage of the breathing circuit, because no radioactivity could be detected in the tubing beyond the exhalation filter. The subjects produced slight amounts of saliva, but the radioactivity in the saliva was minute.

Of the ^99mTc-heparin reaching the mouth, 52.29 ± 10.43% was found in the lung. The dynamic lung scintigrams showed that the deposition of ^99mTc-heparin in the lungs increased linearly in the first 10 min, then plateaued.

The clearance curve of ^99mTc-heparin from the lungs was biexponential with an initial, fast phase assumed to represent distribution in the body, and a later, slower elimination phase (Figure 3). Of the ^99mTc-heparin initially deposited, 39.27 ± 7.84% remained in the lungs 24 h after the beginning of inhalation.

View larger version (18K): [in this window] [in a new window]   Figure 3.   The time course of the elimination of inhaled 99mTc-heparin from the lungs. The graph shows the individual and mean (SD) values (% of maximal count per minute [CPM]).

The initial distribution of inhaled ^99mTc-heparin in the peripheral, intermediate, and central zones of the lungs was 31.33 ± 6.24%, 42.69 ± 4.78%, and 25.98 ± 8.35%, respectively (Figure 4). The figure also demonstrates that the clearance rates were similar in all three zones. Determination of the deposition of ^99mTc-heparin in upper and lower lung zones showed that 45.4 ± 3.44% was found in the upper zone, and 55.39 ± 3.95% in the lower zone. The ratio of the upper/lower deposition was 0.91. 

View larger version (24K): [in this window] [in a new window]   Figure 4.   The initial distribution of inhaled 99mTc-heparin in the peripheral, intermediate, and central lung zones. The distribution is seen to be uniform; clearance rates are similar in all three zones.

Blood Analyses

Radioactivity. Figure 5 shows the average number of heparin units as a function of the time of blood sampling. Absorption of ^99mTc-heparin started immediately after the beginning of inhalation. The time of the peak concentration, T_max, was 61.07 ± 24.59 min after the start of inhalation. The clearance curve of ^99mTc-heparin from the blood was similar to the clearance curve from the lungs showing an initial, fast phase, and a later, slower phase.

View larger version (19K): [in this window] [in a new window]   Figure 5.   The time course of 99mTc-heparin in the blood following inhalation of 99mTc-heparin. The graph shows individual and mean (SD) values (heparin units).

The area under the curve for heparin in the blood was 687 ± 310 IU. Expressed as percentage of nebulizer charge, and of LRT dose the values were 0.76 ± 0.35% and 10.00 ± 3.40%, respectively.

Coagulation. Analyses of F₁₊₂ and TAT from the first 10 subjects were high and varied widely, indicating that coagulation had been activated during blood collection, presumedly owing to the use of a plastic catheter (F₁₊₂: 0.4 to 4.3 nM [predicted 0.6 to 1.5 nM], TAT: 1 to 2,440 pM [predicted 15 to 45 pM]) (14). Thus, these analyses were considered erroneous and discarded.

The study was repeated in five additional subjects. F₁₊₂ was still elevated above 2.0 nM in some of the samples despite careful venipuncture. The FXa-ATIII levels in the postinhalation plasmas did not exceed the preinhalation plasmas by 25% or more. FXa-TFPI was not increased in any of the postinhalation samples, and thus there was no evidence of heparin- induced TFPI release. Thus, a single LRT dose of heparin did not cause significant anticoagulation.

In vivo quality control of labeling. The paper chromatography of the serum from two subjects showed 9.8% and 9.3% of free technetium (pertechnetate).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that the LRT dose after a single nebulization of unfractionated heparin is 8% of the nebulizer charge or approximately 7,000 IU. In a previous study, the dose of heparin reaching the LRT was estimated to approximately 10,000 IU (8). The 8% found in this study is similar to the results of studies with nebulization of ₂-agonists in jet nebulizers (15, 16). Similarly, in our previous study, we found that 18% of the nebulizer charge was deposited on inhalation filters at the mouth, 27% on exhalation filters, and 55% remained in the nebulizer residual as dead volume. In the present study, the corresponding percentages were 15%, 32%, and 53%, respectively.

Our results emphasize the importance of the choice of nebulizer used for clinical studies. Although the Sidestream jet nebulizer is considered one of the most efficient nebulizers with respect to LRT dose, we were able to deliver only 8% of the nebulizer charge to the LRT. The greater efficiency compared with ultrasonic nebulizers is mainly due to a better particle size distribution as demonstrated in our previous study and corroborated in the present study. Many of the previous studies of inhaled heparin employed ultrasonic nebulizers without knowledge of either particle size or LRT dose. Based on our present knowledge, estimates of LRT dose from these studies are 400 to 3,000 IU which are likely to be too small to be clinically relevant. Such small and variable doses are the most likely explanation for the contradictory results of previous clinical studies with inhaled heparin.

The present information on aerosol kinetics depends upon the in vivo stability of ^99mTc-heparin. The finding of less than 10% of free technetium in the plasma 1 h after inhalation indicates to us that this is satisfactory. Similar results were reported by Esquerré and coworkers, who found less than 5% free technetium (17). A preliminary report of poor in vivo stability of Tc-heparin resulting from protein binding of technetium has not been confirmed by other investigators (18). The kinetics of inhaled free technetium aerosol are much faster with a T_max of 10 min (19). Furthermore, heparin instilled into the trachea of experimental animals has shown similar clearance rates from the lungs (20). Thus, the time course we have described is clearly not that of inhaled free technetium, and we have confidence in our results on aerosol kinetics.

The present study does not distinguish between pulmonary tissue and blood. However, the amount of heparin found in the blood shows that radioactivity attributable to labeled heparin in pulmonary blood is negligible.

The above mentioned kinetics have given rise to the hypothesis that heparin is stored in the lungs, probably in or on the surface of the vascular endothelium and in mast cells (20- 22). The amount of heparin excreted or metabolized has not been determined, but some of the difference between the amount transferred to the blood and the amount present in the blood over 24 h could be explained by similar storage in systemic vascular endothelium, mast cells, and macrophages. Support for this hypothesis is found in animal experiments, which have shown a concentration ratio of 100:1 between heparin bound to the endothelium and heparin in the blood (23, 24).

Should inhaled heparin find therapeutic usage such kinetics are advantageous because once a day administration may provide anti-inflammatory activity for 24 h. Documentation of this will require determination of sensitive clinical and laboratory markers of anti-inflammation such as induced sputum. The fact that the preventive effect of heparin on exercise-induced asthma lasts less than 6 h (1) does not exclude a worthwhile anti-inflammatory effect of longer duration.

The fact that ^99mTc-heparin can be detected in the blood very early after the start of inhalation might indicate that some heparin may be absorbed from the oropharyngeal and gastric mucous membranes. Considering the minimal amount of heparin in the blood, the point of uptake becomes of no clinical relevance.

Despite the sensitive coagulation assays, we failed to demonstrate any anticoagulant effect of inhaled heparin in the blood. This is not surprising, because only 700 IU of heparin was found in the blood over 24 h. Thus, a single nebulization of 90,000 IU heparin in a Sidestream jet nebulizer delivering 7,000 IU to the LRT can safely be administered for anti-inflammatory purposes. The small amount of heparin in the blood not only makes inhalation of heparin safe, but also indicates that local mechanisms are responsible for any anti-inflammatory effect in the lungs. Fortunately, the anti-inflammatory properties of heparin are independent of its anticoagulant properties (25).

The present study was conducted in healthy volunteers. Generally, particle deposition is increased in patients with obstructive lung disease, and particles are deposited more centrally (26, 27). Therefore, the lung deposition and clearance of inhaled heparin in patients with asthma should be studied.

In conclusion, in normal adults 8% of the nebulizer charge of heparin delivered by a Sidestream jet nebulizer reached the lungs after a single 90,000 IU nebulization and was uniformly distributed. ^99mTc-heparin was cleared slowly from the lungs, 39% remaining after 24 h. 700 IU of ^99mTc-heparin was absorbed into the blood, with a T_max of 61 min after the start of inhalation. No anticoagulation could be detected with this single heparin inhalation.