INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Heparin is a polysulfated, complex glycosaminoglycan, which has been used clinically as an anticoagulant for over 50 yr (1). The basic polymeric structure of heparin is an alternating repeat sequence of disaccharide units comprising 1 4 linked L-iduronic acid and glucosamine residues (2, 3). The biological actions of heparin result from its polydispersity and its heterogenous molecular organization (1). The structural variability is often the basis of a wide variety of domain structures with a number of biological activities ascribed to heparin. Thus, in addition to its most widely known anticoagulant activity, the heparin molecule has multiple "nonanticoagulant" properties that include interaction with various growth factors (4, 5), regulation of cellular proliferation and angiogenesis (6, 7), and modulation of various proteases and enzymes (8, 9). Inhaled heparin also possesses anti-inflammatory and immunoregulatory properties (10), and has been shown to attenuate antigen-induced bronchoconstriction in allergic sheep (13), as well as to prevent the bronchoconstrictor response to exercise (14, 15) and antigen in asthmatic subjects (16, 17).

Because the biological actions of heparin are molecular weight dependent (18, 19), we have hypothesized that low- molecular-weight heparins (LMWHs) may have greater potency in inhibiting allergic bronchoconstrictor responses. We have recently demonstrated that the antiallergic activity of fractionated heparins is molecular weight dependent, and an inverse relationship between molecular weight and the antiallergic activity was observed (20). In order to further investigate the molecular weight dependence of its antiasthmatic activity, in this investigation we studied the comparative effects of inhaled unfractionated heparin and a LMWH, enoxaparin, in asthmatic subjects with history of exercise-induced bronchoconstriction (EIB).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Thirteen asthmatic subjects (four males, nine females) with a history of EIB symptoms and a documented bronchoconstrictor response to exercise participated in the study. All subjects (age range 16 to 42 yr, mean 27) were nonsmokers, asymptomatic at the time of the study, and without a history of bleeding diathesis, heart disease, or a recent history of upper respiratory infection. Subjects did not take any oral medicines or inhaled glucocorticosteroids, and beta-agonist inhalers were withheld for at least 24 h prior to each study day. Written informed consent was obtained from each subject before the study, and the research protocol was approved by the institutional review board for human investigation.

Pulmonary Function

Pulmonary function tests included spirometry for measurements of forced expired volume in one second (FEV₁). At each data point at least three reproducible measurements were obtained and the best FEV₁ was recorded.

Measurement of Inspired Volume per Minute (I)

The respiratory inductive plethysmograph (RIP; Respigraph, Non- Invasive Monitoring Systems, Miami Beach, FL) was used to measure inspiratory I during a baseline resting period and the exercise period (14, 15). Respibands were applied on the thorax just above the nipple line and on the abdomen at the level of the umbilicus. The subject stood on the treadmill and calibration of RIP was obtained using the Qualitative Diagnostic Calibration procedure, as described previously (14, 15). Data were collected for 5 min to obtain baseline measurements of average respiratory frequency, tidal volume (VT), and I. The subject then ran on the treadmill and data were collected again during the last 5 min of exercise. Toward the end of the exercise period, the subject performed tidal breathing on the spirometer for 5 to 7 breaths while maintaining his or her exercise level and validation of the accuracy of VT (RIP) to VT (spirometry) was compared; a calibration error of < 10% at rest and < 20% during exercise was considered acceptable (Table 1).

Exercise Testing

Exercise testing was performed on a treadmill and heart rate was monitored with an electrocardiogram (ECG). Exercise challenge was performed by gradually increasing the workload. The workload was augmented by increasing the speed and/or degree of inclination until 85% of predicted maximal heart rate was achieved. The subjects were then asked to continue exercise at that workload for 10 min. I was obtained before and during the last 5 min of exercise. The bronchoconstrictor response to exercise, was assessed by measurements of FEV₁ before, immediately after exercise, and then every 5 min for 30 min or until FEV₁ had returned to pre-exercise values.

Aerosol Delivery System

Solutions (4 ml) of heparin, enoxaparin, and placebo were administered as a constant-flow aerosol during tidal breathing, generated with a disposable raindrop medication nebulizer (Puritan Bennett, Lenexa, KS) during a 15- to 20-min period. The mass median aerodynamic diameter of droplets discharged from the nebulizer was estimated at 4.5 µm (geometric SD 2.1), as measured by a seven-stage Anderson cascade impactor. During tidal breathing, the subjects inhaled the aerosol from the nebulizer during inspiration, via a short mouthpiece. Based on the size distribution of the droplets and the tidal breathing pattern used, as well as volume actually nebulized (postnebulization wet nebulizer weight minus dry nebulizer weight), the lung deposition dose is expected to be 10 to 11% of the initial dose placed in the nebulizer, as demonstrated by us previously (21).

Measurements of Antifactor Xa

For estimation of antifactor Xa, 5 ml of venous blood was withdrawn and analyzed by a chromogenic assay.

Methacholine Challenge

Dilutions of methacholine were prepared fresh daily, and diluted in phosphate-buffered isotonic saline solution. Methacholine was delivered through a DeVilbiss No. 644 nebulizer (DeVilbiss Co., Somerset, PA; mass median aerodynamic diameter of 3.9 µm, geometric SD ± 2.4). For bronchial provocation the nebulizer was attached to a dosimeter, which consisted of a breath-activated solenoid valve and a source of compressed air (20 psi). Triggered by the subject's inspiratory effort, the solenoid valve was set to remain open for 0.6 s during inhalation to allow the compressed air to flow through the nebulizer, dispersing an average of 0.023 ml of the solution with each breath. The aerosolized material was delivered from end-expiratory position through the course of a submaximal inspiratory effort. After obtaining the baseline measurements of FEV₁, the subjects inhaled five breaths of the saline diluent, and the measurements were repeated after a 2-min interval. Dose-response curve for methacholine was then established by having the subjects take five inhalations from each of the increasing concentrations of the methacholine at intervals of 5 min. The first concentration was 0.075 mg/ml; the subsequent concentrations increased in an alternating twofold manner. Bronchoprovocation was stopped when FEV₁ had fallen by 20% from the postdiluent value or the maximal concentration of 5 mg/ml had been reached. FEV₁ was then plotted against the cumulative methacholine dose, expressed in breath units. One breath unit was defined as one inhalation of a 1 mg/ml concentration of methacholine. The results were expressed as the cumulative provocative dose of methacholine causing a 20% decrease in FEV₁ (PD₂₀). At the end of each experiment, the subjects took two inhalations of albuterol to reverse any remaining bronchoconstriction.

Agents

Heparin sodium (injection United States Parmacopeia [USP]; in bacteriostatic injection water) derived from porcine intestinal mucosa (Fujisawa USA, Inc., Deerfield, IL) was used undiluted as an aerosol in a concentration of 20,000 USP U/ml. Enoxaparin (Rhône-Poulenc Rorer Pharmaceuticals, Inc., Collegeville, PA) solution (0.5 mg/kg, 1 mg/ kg, 2 mg/kg) was reconstituted in 4 ml of bacteriostatic injection water. The nebulized volume of all solutions was kept constant at 4 ml.

Experimental Protocol

Exercise study (n = 13). Subjects were studied on seven different experiment days, at least 3 d apart. Exercise challenge was performed in an air-conditioned room with constant temperature and humidity. On experiment Day 1, after obtaining baseline pulmonary function tests, a control exercise challenge was performed to document the magnitude of EIB. The workload was gradually increased until 85% of the predicted maximal heart rate was achieved. Subjects then continued to perform exercise on the treadmill at that workload for 10 min. Only subjects with a > 15% decrease in FEV₁ were included in the study. For each subject the workload determined on the initial screening day was kept constant on subsequent experiment days. On experiment Day 2, venous blood was obtained for analysis of plasma antifactor Xa activity. On experiment Days 3 to 7, each subject performed a standardized exercise challenge for 10 min. Exercise challenge was performed 45 min after pretreatment with either 4 ml of aerosolized heparin (80,000 units = 7.5 mg/kg), enoxaparin (0.5 mg/kg, 1 mg/kg, 2 mg/ kg), or a placebo solution (vehicle, bacteriostatic injection water) in a double-blind, randomized, crossover design. Measurements of FEV₁ were obtained, before, 45 min after nebulization of placebo, enoxaparin, or heparin and serially for 30 min after the exercise challenge. The dose of heparin used (7.5 mg/kg) has previously been found to be the most potent dose in animal and human studies (13).

Plasma antifactor Xa activity (n = 5). In five subjects 5 ml of venous blood was obtained on experiment Day 2, before and 45 min after nebulization of 2 mg/kg dose of enoxaparin (total dose = 120 to 150 mg) for analysis of plasma antifactor Xa activity.

Methacholine challenge (n = 5). On two additional days, methacholine challenge was performed in five subjects 45 min after inhaled enoxaparin (2 mg/kg) or placebo in a double-blind, randomized, crossover design. PD₂₀ values of methacholine were determined on enoxaparin and placebo days for comparison.

Statistics. The data were expressed as mean ± SE. The area under the curve (AUC 0-30 min FEV₁ percent decrease) was calculated. The FEV₁ and AUC data were analyzed by a two-way analysis of variance with repeated measures, followed by the Newman-Keuls pairwise comparison for identifying significant differences between pairs. To further analyze the dose-response effect with enoxaparin, the slopes of the response curve were evaluated by linear regression analysis. A paired t test was used for comparing the effect of enoxaparin on methacholine PD₂₀. Significance was accepted when p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Baseline Pulmonary Functions (n = 13)

All subjects had normal pulmonary function parameters (Table 2). The FEV₁ values were comparable on different experiment days, and neither heparin nor different doses of enoxaparin had a significant effect on FEV₁ (p = not significant [NS]) (Table 3).

Antifactor Xa (n = 5)

Aerosolized enoxaparin did not change the plasma anti-Xa activity. Antifactor Xa was < 0.05 IU/ml before and 1 h after nebulization of enoxaparin.

Heart Rate and Ventilatory Parameters (n = 13)

The values of heart rate, respiratory frequency, VT, and I during rest and exercise were comparable on placebo, heparin, and enoxaparin days (Table 1). The workload on different experiment days was kept constant for each subject, with a mean ± SE value of 10 ± 1 metabolic equivalents of oxygen consumption (METS) (range 5 to 15). The mean ± SE values of heart rate during exercise were 162 ± 5 and 165 ± 5 beats/min on placebo and heparin days, and 164 ± 4, 166 ± 4, and 165 ± 5 beats/min on enoxaparin days, respectively. The mean ± SE values of I during exercise were 56 ± 5 and 57 ± 4 L/min on placebo and heparin days, and 58 ± 4, 64 ± 6, and 58 ± 6 L/ min on enoxaparin days, respectively.

Exercise-induced Changes in FEV₁ (n = 13)

Postexercise decreases in FEV₁ were reproducible on control and placebo days, with maximal decrease occurring within 5 min postexercise (Figure 1). The maximal decreases in FEV₁ (mean ± SE) were 30 ± 4% and 29 ± 5% on control and placebo days, respectively (p < 0.05). Pretreatment with inhaled heparin (80,000 units = 7.5 mg/kg) prevented the exercise- induced decreases in FEV₁ by 31% (Figure 1); the maximal decrease in FEV₁ was 20 ± 4%, which was significantly smaller than placebo (p < 0.05).

View larger version (20K): [in this window] [in a new window]   Figure 1.   Effect of pretreatment with aerosolized heparin (80,000 units = 7.5 mg/kg), LMWH (enoxaparin 2 mg/kg), or placebo on exercise-induced decreases in FEV1. Date are shown as mean ± SE percent decrease in FEV1 (n = 13). FEV1 was not significantly different from baseline at 30 min postexercise on heparin day, and at 20 min postexercise on enoxaparin day. BSL = Baseline; PD = Postdrug. *Significantly different from baseline (p < 0.05). +Significantly different from placebo (p < 0.05). ¶Enoxaparin significantly different from heparin (p > 0.05)

Enoxaparin caused a dose-dependent inhibition of postexercise decreases in FEV₁ (Figure 2). The maximal decreases in FEV₁ were inhibited by 28%, 38%, and 48% after pretreatment with 0.5 mg/kg (FEV₁ = 21 ± 5%), 1 mg/kg (FEV₁ = 18 ± 5%), and 2 mg/kg (FEV₁ = 15 ± 3%) doses of enoxaparin, respectively; these were significantly different from placebo (p < 0.05) (Figures 2 and 3). The inhibitory effect of 0.5 mg/kg dose of enoxaparin was comparable to 7.5 mg/kg heparin (p = NS), whereas 2 mg/kg dose of enoxaparin was significantly more potent than heparin, 0.5 mg/kg and 1 mg/kg doses of enoxaparin (p < 0.05) (Figure 3). The linear regression analysis also demonstrated a dose-response effect of enoxaparin, with slope of the response curve significantly different from zero (slope = 59.0 ± 27.5; p < 0.05).

View larger version (20K): [in this window] [in a new window]   Figure 2.   Effect of pretreatment with aerosolized placebo or enoxaparin (0.5 mg/kg, 1 mg/kg, and 2 mg/kg) on exercise-induced decreases in FEV1. Data are shown as mean ± SE % decrease in FEV1 (n = 13). FEV1 was not different from baseline at 20 min (enoxaparin 1 mg/kg and 2 mg/kg) and 25 min (enoxaparin 0.5 mg/kg) postexercise. BSL = Baseline; PD = Post drug. *Significantly different from baseline (p < 0.05). +Significantly different from enoxaparin (p < 0.05). ¶Enoxaparin 2 mg/kg dose significantly different from 0.5 mg/kg dose.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Maximal decreases in FEV1 after exercise following pretreatment with placebo, heparin (80,000 units = 7.5 mg/kg), or enoxaparin (0.5 mg/kg, 1 mg/kg, 2 mg/kg). Data are shown as mean ± SE maximal decreases in FEV1 (n = 13). *Significantly different from placebo (p < 0.05). +2 mg/kg dose is significantly different from heparin, 0.5 mg/kg, and 1 mg/kg enoxaparin (p < 0.05).

AUC was reduced by 34% by heparin, and by 32%, 46%, and 55% by 0.5 mg/kg, 1 mg/kg, and 2 mg/kg doses of enoxaparin, respectively (p < 0.05). The 2 mg/kg dose of enoxaparin caused a significantly greater reduction of AUC than heparin and 0.5 mg/kg dose of enoxaparin (p < 0.05).

The time to recovery was also shorter with 1 mg/kg and 2 mg/kg doses of enoxaparin (20 min), versus 0.5 mg/kg enoxaparin (25 min), heparin (30 min), or placebo (> 30 min) (Figures 1 and 2). Analysis of the individual data revealed that in three subjects enoxaparin (1 mg/kg and 2 mg/kg) was ineffective, while heparin and 0.5 mg/kg enoxaparin were ineffective in two additional subjects (n = 5 subjects).

Methacholine-induced Bronchoconstriction (n = 5)

Enoxaparin (2 mg/kg) had no effect on methacholine-induced bronchoconstriction. Mean ± SE PD₂₀ values of methacholine were 4 ± 3 breath units and 3 ± 2 breath units after placebo and enoxaparin treatments, respectively (p = NS).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have previously reported that inhaled heparin attenuates the acute bronchoconstrictor response to exercise (14, 15). The results of the present study extend our previous observations and demonstrate that inhaled LMWH, enoxaparin, attenuates EIB in a dose-dependent manner, and further, that the protective effect of inhaled enoxaparin on EIB is independent of its effect on plasma antifactor Xa activity.

Enoxaparin caused a dose-dependent attenuation of EIB. Inhaled doses of 0.5 mg/kg, 1 mg/kg, and 2 mg/kg caused 28%, 38%, and 48% inhibition of EIB as estimated by peak decreases in FEV₁. All doses of enoxaparin shortened the time to recovery of normal lung function and markedly reduced the time-response curve. Thus, AUC was reduced by 32%, 46%, and 55% by 0.5 mg/kg, 1 mg/kg, and 2 mg/kg doses of enoxaparin; in enoxaparin-responsive subjects even a greater inhibition was observed.

Many biological actions of heparin are molecular weight dependent (6, 18, 22). It is well established that both the degree of sulfation and molecular chain length influence the anticoagulant, antiproliferative, and elastase inhibitory activity of heparin (6, 18, 22). It has also been previously shown that the antiallergic activity of fractionated heparins in allergic sheep is inversely related to molecular weight (20). The results of the present study are consistent with this concept and demonstrate greater potency of LMWH, enoxaparin, in inhibiting EIB. The inhibitory effect of 0.5 mg/kg dose of enoxaparin was comparable to heparin (80,000 units = 7.5 mg/kg), whereas 2 mg/kg dose was more potent. The postexercise recovery time was also shorter with enoxaparin. The lower potency of inhaled heparin was not dose-related; and in animal and human studies the dose of heparin used in the present study was found to be the most potent dose (13). Perhaps the greater potency of enoxaparin is related to the presence of a higher percentage of oligosaccharide chains possessing antiallergic activity (20).

The magnitude of protection offered by inhaled enoxaparin against exercise-induced bronchoconstriction in this study is comparable with other agents. Various pharmacologic agents including beta-agonists, cromolyn sodium, atropine, calcium blockers, furosemide, and leukotriene inhibitors have been shown to inhibit EIB (23). Similarly, varying degrees of protection against bronchoconstrictor response induced by hyperventilation with cold dry air have also been observed (29). In most of the studies the protective effect of leukotriene antagonists and other agents against EIB has been partial. Leff and coworkers (30) showed that the leukotriene-receptor antagonist, montelukast, inhibited the exercise-induced peak decreases in FEV₁ by 31%, which is considerably lower than the 48% inhibition observed with 2 mg/kg enoxaparin observed in the present study. The protective effects of leukotriene antagonists, cromolyn, and other agents against EIB have shown considerable heterogeneity (23, 30). It has been observed that interruption of the leukotriene cascade results in inhibition of EIB in some patients, but has no effect in others. This indicates that the pathways leading to bronchoconstriction after exercise vary in different patients with asthma (33), and that in some subjects mediators other than the leukotrienes may be more important bronchoconstrictor agonists. We have previously observed varying degree of protection against EIB by inhaled heparin and cromolyn (14). Heparin completely or partly inhibited the EIB in 75% of subjects, whereas cromolyn was effective in 60% of subjects (14). Our present observations with LMWH are consistent with the previous data and demonstrate that enoxaparin offered some degree of protection in 77% of subjects. Although well-controlled comparative studies have not been conducted, in a direct comparison, the leukotriene antagonist SK&F 104353 was as effective as cromolyn sodium in preventing EIB (32), whereas inhaled heparin was found superior to cromolyn sodium (14, 15).

LMWHs are a diverse group of drugs derived from unfractionated heparin. LMWHs have a lower binding affinity to endothelial cells and reduced nonrenal cellular mechanism of clearance, resulting in increased bioavailability and prolonged in vivo biological activity (34). Although high doses of inhaled LMWH (> 270 mg) show prolonged anticoagulant activity (29), the bioavailability and elimination half-life for antiasthmatic activity of aerosolized LMWH is not known. Pharmacokinetic studies with aerosolized unfractionated heparin have demonstrated significant antiasthmatic activity lasting up to 6 to 12 h in allergic sheep (35), and 3 to 6 h in human subjects with EIB (15). It is possible that these molecular weight dependent properties may partly account for increased antiasthmatic activity observed with LMWH.

Previous studies in human subjects and sheep have suggested that the antiallergic activity of inhaled heparin is independent of its anticoagulant properties, as activated partial thromboplastin time (APTT) activity was not prolonged (14, 15, 17). Recent studies have confirmed this hypothesis and showed inhibition of allergic airway responses in the acute responder sheep by nonanticoagulant fractions of heparin (36, 37). Although high doses of inhaled LMWH (> 270 mg) have been shown to increase the plasma antifactor Xa activity (29), the highest dose of inhaled enoxaparin (150 mg) used in the present study had no effect on antifactor Xa activity.

The mechanism of inhibition of EIB by LMWH is not known. Enoxaparin failed to modify the bronchoconstrictor response to methacholine, thus excluding a direct effect on airway smooth muscle. Previous studies have suggested that inhaled unfractionated heparin prevents antigen-induced bronchoconstriction and airway hyperresponsiveness by modulation of mast cell mediator release, rather than by a direct effect on airway smooth muscle (13, 38, 39). This hypothesis was supported by in vivo and in vitro studies demonstrating inhibition of antigen-induced airway smooth muscle contraction and its failure to attenuate agonist-induced contractile responses (13, 40). Heparin also prevented the bronchoconstrictor responses induced by compound 48/80, a nonimmunologic mast cell degranulating agent (13), and inhibited the antigen-induced histamine release in bronchoalveolar lavage (BAL) (33, 41) and from isolated mast cells (38). Although the exact cellular mechanism of the antiasthmatic activity of unfractionated heparin is not clearly known, it has been proposed that the antiallergic activity of unfractionated heparin may be mediated by inhibition of inositol 1,4,5-triphosphate (IP₃)-mediated mast cell mediator release (39). LMWHs have also been shown to inhibit IP₃-induced Ca⁺⁺ release (42), attenuate anti-IgE induced mast cell degranulation in vitro (36), and to inhibit antigen-induced histamine release in BAL (41). Whether the antiasthmatic activity of enoxaparin in the present study is mediated by inhibition of mast cell mediator release is not known at present.