INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

APPROXIMATELY 80% of subjects with chronic cervical spinal cord injury (tetraplegia) having no history of asthma or respiratory diseases before injury demonstrate airway hyperresponsiveness after inhalation of methacholine or histamine (10, 15, 31). Because methacholine induces bronchoconstriction by direct interaction with muscarinic receptors located on bronchial smooth muscle (19), hyperresponsiveness in these subjects suggests exaggerated airway irritability and/or increased baseline airway tone. With airway narrowing, bronchoconstrictor stimuli affect airflow resistance more significantly because changes in resistance to airflow are inversely related to the radius of the airway. Although histamine induces bronchoconstriction both directly by interaction with specific receptors on smooth muscle and indirectly by stimulating central and local neural reflexes (19), findings that bronchoconstriction caused by this agent in subjects with tetraplegia was not affected by pretreatment with ipratropium bromide also suggest direct action on airways (15).

A number of physical and physicochemical agents that act indirectly have been used to evaluate bronchial hyperresponsiveness (23). These include hypotonic or hypertonic aerosols, which appear to induce bronchospasm in susceptible individuals by altering osmolarity of the fluid lining the airways, thereby causing permeability changes in epithelial lining cells with associated release of mediators from inflammatory cells in or on the bronchial mucosa and/or amplified vagal afferent nerve activity (32). To further explore the pathogenesis of airway hyperreactivity, subjects with tetraplegia who previously demonstrated hyperreactivity to histamine were challenged with ultrasonically nebulized distilled water (UNDW). To define the role of the parasympathetic nervous system in the response, a second challenge was performed in which subjects were pretreated with an anticholinergic agent, ipratropium bromide, before exposure to UNDW.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The study group consisted of 15 male subjects with spinal cord injury, nine with tetraplegia (C_4-7) and six with paraplegia (T₉-L₁). All subjects were closely screened for history of pulmonary disease, allergies, and respiratory symptoms (34), and all denied having had recent or active respiratory infections. In addition, none of the subjects were receiving any medication known to alter airway tone or responsiveness. Participation was also contingent on a positive histamine bronchochallenge test for subjects with tetraplegia and a negative response for subjects with paraplegia. The Institutional Review Board for human studies of the Bronx Veterans Affairs Medical Center granted approval for the study. Informed consent of each subject was obtained before the investigation.

Spirometric measurements were obtained by using an automated pulmonary function laboratory (SensorMedics System 2200, Yorba Linda, CA) while the subjects were seated in their wheelchairs. Baseline values of forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV₁) were collected from each subject according to the current recommendations of the American Thoracic Society (2). Spirometry results are expressed as absolute values and percent predicted on the basis of the standards of Morris et al. (21).

The distilled water challenge was performed by using an ultrasonic nebulizer (Ultra-neb 99, DeVilbiss, Somerset, PA) generating particles with a diameter of 0.5-5 µm at an output (calibrated before each test) of 3.0 ml/min (SD = 0.1 ml/min). The aerosol from the ultrasonic nebulizer passed through tubing 67 cm in length with an internal bore size of 2.5 cm. Subjects inhaled UNDW during normal tidal volume breathing by using a two-way nonrebreathing valve (model 2700, Hans Rudolph, Kansas City, MO). Before each challenge, 50 ml of distilled water were placed into the reservoir of the nebulizer; after challenge, by measuring the distilled water remaining in the reservoir and tubing, but not in the valve, the output delivered to the valve was determined.

To begin each challenge subjects inhaled ultrasonically nebulized normal saline for 2 min. If the change in FEV₁ from baseline did not exceed 10%, UNDW was administered. Subjects were initially exposed to UNDW for 30 s; spirometry was performed immediately and again 2 min after exposure. The challenge continued by progressively increasing exposure time (1, 2, 4, and 8 min) until the test was terminated when either the FEV₁ decreased 20% or more from baseline (PD₂₀) or the maximal exposure time was reached (35). A PD₂₀ <24 ml was considered indicative of airway hyperresponsiveness and was calculated by using a computer program that generated its value by interpolation from a logarithmic dose-response curve. For those subjects not responding to the maximal dose of distilled water (24 ml), a PD₂₀ value of 24 ml was used in the calculation of mean PD₂₀. Ventilation and breathing frequency were obtained breath by breath during each exposure period by using a metabolic cart (SensorMedics System 2900, Yorba Linda, CA).

Five of the subjects with tetraplegia responding to the UNDW challenge returned within 20 days of the initial study to undergo a second challenge. The repeat challenge, in which identical methods as described above were used, was performed 30 min after the inhalation of aerosolized ipratropium bromide (2.5 ml of a 0.6% solution; Atrovent, Boehringer Ingelheim, Ridgefield, CT) administered via a DeVilbiss 646 nebulizer.

All data are expressed as means ± SD, and a geometric mean was calculated for PD₂₀ values. An unpaired Student's t-test was applied to determine whether PD₂₀ values differed between the groups with tetraplegia and paraplegia. A paired t-test was used to assess differences in spirometry and PD₂₀ values between the initial distilled water challenge and the ipratropium bromide rechallenge. Simple regression analysis was used to determine whether a significant relationship exists between the decline in FEV₁ after UNDW and baseline pulmonary function. The criterion for acceptance of statistical significance for all analyses was established at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Mean duration of injury and age did not differ significantly between subjects with tetraplegia and those with paraplegia [15 ± 9.93 vs. 14 ± 8.16 (SD) and 39 ± 11.02 vs. 40 ± 7.48 yr, respectively]. Histamine PD₂₀ (mg/ml) results used as a criterion for study participation are presented (Table 1). Of the total group, five subjects were never smokers, six were former smokers, and four were current smokers; medications are also reported (Table 1).

Mean baseline and percent predicted values of FVC and FEV₁ were significantly lower in the group with tetraplegia than in the group with paraplegia (P < 0.05), whereas mean FEV₁/FVC ratios did not differ between the two groups (Table 2). Eight of the nine subjects with tetraplegia had a significant bronchoconstrictor response to UNDW (geometric mean PD₂₀ = 7.76 ± 7.67 ml). In contrast, none of the subjects with paraplegia demonstrated a response (geometric mean PD₂₀ = 24 ml) (Table 2). The mean percent changes in FEV₁ from baseline for the two groups were 29 ± 8.20 and 2 ± 3.60%, respectively, which were also statistically different (P < 0.05). Regression analysis revealed no significant correlation between UNDW PD₂₀ and baseline FEV₁ percent predicted (data not shown).

Baseline spirometry values for the five subjects who returned for challenge with ipratropium bromide-UNDW did not differ significantly from those obtained before the original challenge (Table 3). These five subjects who initially responded to UNDW (geometric mean PD₂₀ = 5.99 ± 4.47 ml) were no longer responsive after pretreatment with ipratropium bromide (geometric mean PD₂₀ = 24 ml); the mean percent changes in FEV₁ from baseline values were 34 ± 8.32 and 7 ± 3.74%, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We found that eight of the nine subjects with chronic tetraplegia and known hyperresponsiveness to a pharmachological agent were also hyperreactive to UNDW, whereas none of the six subjects with paraplegia were responsive. Our findings cannot be attributed to preinjury respiratory disease or smoking status because the study subjects were critically screened and none presented with a history of asthma or atopic disorders, and current, former, and never smokers were represented in the entire group. Our findings are comparable to a study of subjects with asthma, in which 48 of 55 subjects demonstrated a PD₂₀ for UNDW of 9 ml, whereas all healthy controls had a PD₂₀ >24 ml (4). Other investigators (6, 22) have also reported that healthy subjects do not experience bronchoconstriction after inhalation of UNDW, whereas ~7% of the healthy population exhibit a positive response to methacholine (36). Current findings that subjects with tetraplegia are hyperreactive to a physicochemical agent that does not appear to act directly on airway smooth muscle (4, 16), combined with previous observations that these subjects are also hyperreactive to methacholine and histamine (10, 15, 31), agents with direct action on airway smooth muscle, indicate that overall airway responsiveness in these subjects is a nonspecific phenomenon similar to that observed in individuals with asthma. Support for two different stimuli-induced response pathways comes from additional findings that UNDW PD₂₀ obtained during the present study did not correlate with histamine-PD₂₀ from the same patients. Similarly, in subjects with asthma, responsiveness to UNDW does not correlate with that induced with methacholine or histamine (17, 33).

The specific mechanism(s) by which UNDW induces bronchoconstriction in susceptible individuals is not entirely evident. Findings, however, that alterations of osmolarity in bronchial fluids away from isosmolarity (13, 25) are associated with increased epithelial permeability in controls and subjects with asthma (18) suggest that the trigger for bronchoconstriction is not at the epithelial surface but is either in the paracellular spaces or submucosa (18). In the subepithelial space, activation of mast cells and basophils, which are present in higher numbers in airway tissue of subjects with asthma, could, in turn, result in the release of bioactive mediators (12, 20). In support of this hypothesis, an increase in neutrophil chemotactic activity is found in serum after exposure of asthmatic subjects to UNDW (26), and inhaled indomethacin prevents UNDW-induced hyperresponsiveness, presumably through inhibition of local prostaglandin synthesis (29). Moreover, in subjects with asthma, responsiveness to UNDW is greater in those having higher numbers of eosinophils and mast cells in epithelium, although the significance of the findings is made unclear by additional observations that these subjects also had more severe spontaneous airflow obstruction and symptoms and greater responsiveness to methacholine (9). Data failing to support a role of mast cells in UNDW-induced bronchoconstriction come from observations that inhaled furosemide prevents UNDW-induced bronchoconstriction in both atopic and nonatopic children with asthma (30). Also, although sodium cromoglycate, an agent that inhibits mast cell degranulation, attenuates UNDW-induced bronchoconstriction in subjects with asthma (1, 3, 6, 28), findings that the agent also inhibits bronchoconstriction caused by SO₂ (27), exercise (11), and cold air (7) suggest that cromolyn may work by other mechanisms (7). Finally, there is no evidence to suggest that subjects with chronic tetraplegia have an ongoing airway inflammatory process featuring an increase in the number of mast cells, eosinophils, and/or basophils, although this has not been investigated by invasive studies.

The present study also demonstrated that pretreatment of subjects with ipratropium bromide completely blocked UNDW-induced bronchoconstriction, indicating that hyperresponsiveness to UNDW is mediated through cholinergic pathways. Similar studies in subjects with asthma have been inconclusive: two studies using aerosolized atropine showed protection (14, 28), which was not merely a function of atropine's effect on baseline airway tone (28), whereas two additional studies using atropine given by aerosol (3) or intravenously (8) demonstrated no protection. An additional study evaluating ipratropium bromide reported no protection (1). Present findings support the hypothesis that bronchial hyperresponsiveness in subjects with chronic tetraplegia represents unopposed parasympathetic activity, potentially due to loss of sympathetic innervation of the lungs. Efferent stimulation of parasympathetic nerves results in bronchoconstriction, and baseline activation contributes to resting airway tone. Although human airway smooth muscle contains only rudimentary sympathetic innervation, sympathetic fibers may indirectly modulate bronchomotor tone by interacting with parasympathetic ganglia cells in the peribronchial plexa (5, 24). Despite the fact that ~45% of subjects with tetraplegia experience significant bronchodilation (>12% improvement in FEV₁) after inhalation of ipratropium bromide (80 µg), it seems unlikely that UNDW-induced bronchoconstriction and its prevention by ipratropium bromide can be attributed exclusively to increased baseline airway tone due to unopposed cholinergic activity. This is primarily because of contrasting findings that pretreatment with ipratropium bromide does not attenuate histamine-induced bronchoconstriction (15). In addition, the absence of correlation between baseline FEV₁ (percent predicted) and UNDW PD₂₀ suggests that responsiveness is not related to resting airway caliber. Definitive answers, however, are not available because resting airway tone has not been quantified in the spinal cord injury population.