Neural control of human airways in health and disease

Several aspects of airway function are under autonomic control: airway smooth muscle tone, submucosal gland secretion, epithelial cell function, bronchial vascular tone and permeability, and probably secretion from mast cells and other inflammatory cells. Neural control of human airways is more complex than previously recognized. In addition to afferent nerves and cholinergic adrenergic mechanisms (including circulating catecholamines), there are nonadrenergic, noncholinergic nerves that may be both excitatory and inhibitory. The neurotransmitters of this third nervous system are uncertain, but there is some evidence that neuropeptides may be involved. Several neuropeptides have recently been identified in human airways and, although they have potent effects, their pathophysiologic role is uncertain. There is much evidence that autonomic control of the airways may be abnormal in airway disease, particularly in asthma, but the precise role of neural mechanisms in the pathogenesis of air-flow obstruction and bronchial hyperresponsiveness remains to be defined.

VIP and PHM and their role in nonadrenergic inhibitory responses in isolated human airways

There is increasing evidence in many species that vasoactive intestinal peptide (VIP) may be a neurotransmitter in nonadrenergic inhibitory nerves. We have studied the effect of electrical field stimulation (EFS), exogenous VIP, and isoproterenol (Iso) on human airways in vitro. We have also studied a related peptide, peptide histidine methionine (PHM), which coexists with VIP in human airway nerves, and in separate experiments studied fragments of the VIP amino acid sequence (VIP1-10 and VIP16-28) for agonist and antagonist activity. Human airways were obtained at thoracotomy and studied in an organ bath. In bronchi EFS gave an inhibitory response that was unaltered by 10(-6) M propranolol but was blocked by tetrodotoxin, whereas in bronchioles there was little or no nonadrenergic inhibitory response. VIP, PHM, and Iso all caused dose-dependent relaxation of bronchi, VIP and PHM being approximately 50-fold more potent than Iso. VIP, but not Iso, mimicked the time course of nonadrenergic inhibitory nerve stimulation. In contrast bronchioles relaxed to Iso but not to VIP or PHM. Neither propranolol nor indomethacin altered the relaxant effects of VIP or PHM, suggesting a direct effect of these peptides on airway smooth muscle. Neither of the VIP fragments showed either agonist or antagonist activity. We conclude that VIP and PHM are more potent bronchodilators of human bronchi than Iso and that the association between the relaxant effects of these peptides and nonadrenergic inhibitory responses suggests that they may be possible neurotransmitters of nonadrenergic inhibitory nerves in human airways.

Reduction of nocturnal asthma by an inhaled anticholinergic drug

Although the mechanisms of nocturnal asthma are still uncertain, increased vagal cholinergic tone may be contributory factor. To examine this hypothesis, we have studied the effect of an anticholinergic drug, oxitropium bromide, on the early morning fall in peak expiratory flow (PEF) in patients with nocturnal asthma. Eighteen patients (aged 18 to 76 years; seven men) with documented nocturnal asthma were studied in a double-blind randomized cross-over study in which they received either oxitropium bromide (200 micrograms or 400 micrograms) or placebo in a single dose at night for two-week periods. With placebo the mean (+/- SE) fall in PEF (expressed as percentage of evening PEF) was 17.3 +/- 2.0 percent, which was significantly reduced to 10.3 +/- 3.3 percent after oxitropium (400 micrograms) (p less than 0.05; ANOVA). Closer analysis revealed that nine of the 18 patients had responded in a dose-dependent manner, with the mean percentage decreases with placebo, 200 micrograms, and 400 micrograms of oxitropium being 19.1 +/- 3.2, 11.5 +/- 4.4, and 5.0 +/- 4.5 percent, respectively (p less than 0.01 between each treatment). The remaining patients were unaffected by therapy. There were no differences between "responders" and "non-responders" in terms of age, atopic status, duration of asthma, severity of asthma, or bronchodilator response to albuterol (salbutamol). There were no differences in nocturnal symptoms between periods of treatment, and no side effects were recorded. We conclude that anticholinergic drugs may protect against nocturnal asthma in some patients, indicating the involvement of vagal cholinergic mechanisms.

Visualization of vasoactive intestinal peptide receptors in human and guinea pig lung

Binding of [125I]vasoactive intestinal peptide (VIP) to human and guinea pig lung sections was characterized and VIP receptors localized by light-microscopic autoradiography. Inhibition of [125I] VIP binding by unlabeled VIP, peptide histidine isoleucine, peptide histidine methionine, secretin and VIP fragment VIP10-28 indicated that [125I]VIP bound to specific receptors and binding was shown to be reversible. Scatchard analysis showed two binding sites; human lung and a dissociation constant (Kd) of 0.27 +/- 0.04 and 23.3 +/- 3.5 nM, with maximum binding capacities (Bmax) of 11.2 +/- 3.1 and 589 +/- 98 fmol mg-1 of protein for the high and low affinity sites, respectively. Guinea pig lung similarly had a high affinity Kd of 0.3 +/- 0.05 nM and Bmax of 226 +/- 61.1 fmol mg-1 of protein and a low affinity Kd value of 18.8 +/- 2.4 nM and Bmax of 1730 +/- 260 fmol mg-1 of protein. In both human and guinea pig lung there was labeling of cellular structures and no specific labeling pattern was found in sections incubated in the presence of an excess of unlabeled VIP. A high density of labeling was found over airway epithelium of all airways, submucosal glands and vascular smooth muscle. There was also labeling of airway smooth muscle of large, but not of small, airways. This localization corresponds to the pattern of VIPergic innervation. In addition, there was labeling of alveolar walls. This indicates that VIP has a physiological role in the lung.

Activated oxygen molecules generated by electrical stimulation affect vascular smooth muscle

Although electrical field stimulation has been used to evoke neural responses in blood vessels this technique can also generate activated oxygen molecules which may affect the experimental preparation. We evaluated this by stimulating 10 cc of aerated Krebs-Henseleit solution for a period of 20 s and then transferring the fluid to a pre-contracted segment of bovine pulmonary artery. The parameters required to achieve 50% maximal relaxation were a voltage of 4.0 +/- 0.3 volts (mean +/- S.E.), a frequency of 1.6 +/- 0.1 Hz and a pulse duration of 0.37 +/- 0.02 ms. Exposure to stimulated fluid caused greater amounts of relaxation in vessels pre-contracted with serotonin than in those pre-contracted with either histamine or potassium chloride. Relaxation was not endothelial dependent and it was not inhibited by adrenergic or cholinergic blockade. Since both ascorbate and catalase but not superoxide dismutase inhibited relaxation, it appears that hydrogen peroxide is involved. We conclude that even low intensity electrical field stimulation generates activated oxygen molecules in oxygenated physiologic buffer solution and that these molecules can relax blood vessels both directly and by altering the response to drugs in the muscle bath. These effects must be considered when using electrical field stimulation to study blood vessels.

Effect of infused vasoactive intestinal peptide on airway function in normal subjects

Vasoactive intestinal peptide, one of the putative neurotransmitters of non-adrenergic inhibitory nerves in human airways, is a potent relaxant of human airways in vitro. Previous in vivo studies of infused vasoactive intestinal peptide in asthmatic subjects have shown only a small bronchodilator effect, which may have been secondary to the cardiovascular effects of the peptide. The effect on airway function of infused vasoactive intestinal peptide was studied in normal subjects, who readily develop bronchodilation in response to a beta agonist. Separate experiments were designed to assess whether there is any synergy between this peptide and the beta agonist isoprenaline. Incremental doses of 1, 3, and 6 pmol/kg/min of vasoactive intestinal peptide were infused for 15 minutes. At 6 pmol/kg/min it caused a mean fall in systolic blood pressure from 108 to 88 mm Hg and a rise in heart rate from 71 to 95 beats/min. There was no significant change in specific airways conductance (sGaw) at any dose of vasoactive intestinal peptide. No significant changes were found with placebo. Isoprenaline (400 microgram) given by inhalation at the end of the infusion produced a mean increase in sGaw of 50%. Infused peptide caused no significant change in the cumulative dose-response curve for inhaled isoprenaline. The lack of effect of vasoactive intestinal peptide on airway responses in vivo may be due to rapid enzymatic breakdown of the peptide or to the fact that dosage has to be limited by the cardiovascular effects.

Vasoactive intestinal peptide in bovine pulmonary artery: localisation, function and receptor autoradiography

The role of vasoactive intestinal peptide (VIP) in the control of pulmonary vascular tone was investigated by functional response, immunocytochemical localisation and receptor autoradiography in bovine pulmonary arteries. VIP-immunoreactive nerve fibres were present at the adventitial-medial junction and in the media of the vessels. Exposure of precontracted bovine pulmonary artery segments to VIP in vitro resulted in almost complete (86 +/- 3%; mean +/- s.e.mean) relaxation, the concentration needed for 50% relaxation being 4.47 +/- 0.37 X 10(-9)M. VIP effects did not depend on the presence of intact endothelial cells. The distribution of VIP receptors was studied by autoradiography using [125I]-VIP. A high density of VIP receptors was found in arterial vascular smooth muscle, with a gradient of density from adventitia to luminal surface. There were no receptors on endothelial cells. These data show that VIP is a potent vasodilator of bovine pulmonary arteries, via direct activation of VIP receptors in vascular smooth muscle. VIP-immunoreactive nerves may influence pulmonary vascular tone directly and could, therefore, be important in regulating pulmonary blood flow.

Histamine is released from skin by substance P but does not act as the final vasodilator in the axon reflex

We have explored in man the hypothesis that histamine released from dermal mast cells by neurotransmitters from afferent nerves contributes to vasodilatation of the axon reflex. The ability of substance P to release histamine from human skin in vivo, and the effects of a histamine H1-receptor antagonist on capsaicin-induced axon reflex flares were studied. Intradermal injections of substance P (50 pmol) produced a weal and flare response which was associated with increased histamine concentration in blood draining the site (mean plasma histamine concentration before injection 0.17 +/- 0.02 ng ml-1 (+/- s.e.mean), concentration one minute after injection 1.26 +/- 0.28 ng ml-1, n = 6). Terfenadine, an H1-receptor antagonist, had no effect on the flare response to intradermal injection of capsaicin at a dose which inhibited by more than 60% the flare response to exogenous histamine and to histamine released from dermal mast cells by substance P. Substance P releases histamine from human skin in vivo. However, whatever the nature of the neurotransmitter released from afferent nerves during the axon reflex, it does not produce vasodilatation through release of histamine from dermal mast cells. Histamine may still contribute to the flare by initiation of the reflex.

Effects of inhaled platelet activating factor on pulmonary function and bronchial responsiveness in man

Platelet activating factor (PAF), a phospholipid inflammatory mediator, was given as an aerosol to eight normal subjects. PAF caused a dose-dependent bronchoconstriction in all subjects. This did not correlate well with responsiveness to methacholine. Some subjects showed tachyphylaxis to PAF-induced bronchoconstriction. No subject had a late bronchoconstriction response. Transient facial flushing and an increase in heart rate (mean 7 beats/min) occurred but there was no consistent change in blood pressure. Lyso-PAF, the inactive precursor and major metabolite of PAF, had no effect on pulmonary or cardiovascular responses. Six of the subjects took part in a double-blind, randomised, placebo-controlled, crossover study in which bronchial responsiveness to methacholine was measured over the 3 days after administration of PAF or lyso-PAF. PAF had a greater effect in raising responsiveness (p less than 0.01). Its maximum effect occurred at 3 days and returned to baseline in 1 to 4 weeks. PAF may contribute to the pathogenesis of bronchial hyperresponsiveness, which is the most characteristic abnormality in asthma.

Phosphatidylinositol response to cholinergic agonists in airway smooth muscle: relationship to contraction and muscarinic receptor occupancy

Hydrolysis of membrane phosphatidylinositol (PI) and polyphosphonoinositides (PPI) may be the coupling mechanism between receptor stimulation and the rise in intracellular calcium concentration that leads to smooth muscle contraction. In bovine tracheal smooth muscle, we correlated PI/PPI turnover, contraction and muscarinic receptor occupancy by carbamoylcholine (10(-9) to 10(-2) M). Inositol monophosphate formation after agonist stimulation, in the presence of lithium, provided a direct measurement of PI/PPI breakdown, and receptor occupancy was determined by [3H]quinuclidinyl benzylate binding. Carbamoylcholine caused a concentration-dependent contraction (EC50 = 7.4 X 10(-8) M), PI/PPI response (EC50 = 3.8 X 10(-5) M) and [3H]quinuclidinyl benzylate displacement (with high and low affinity binding sites have dissociation constants (Kd) of 3 X 10(-7) and 6 X 10(-4) M, respectively). This indicates the presence of spare receptors as maximal contraction is obtained when less than 20% of receptors are occupied. The concentration of carbamoylcholine inhibiting 50% of the PI/PPI response and 50% of maximal receptor occupancy (IC50) were similar for atropine (IC50 = 1 X 10(-9) and 5.3 X 10(-9) M, respectively), and for pirenzepine (IC50 = 3 X 10(-6) and 2.3 X 10(-6) M, respectively); the pA2 of the contraction was 8.3 +/- 0.12 for atropine and 7.2 +/- 0.08 for pirenzepine, indicating that M2 receptors may be largely predominant among bovine tracheal smooth muscle muscarinic receptors. Bovine tracheal smooth muscle may be a useful model to study the effects of other spasmogens, as it allows comparison of functional effects, PI breakdown and receptor occupancy in the same preparation.

cAMP immunocytochemistry provides evidence for functional VIP receptors in trachea

Vasoactive intestinal peptide (VIP), first isolated from porcine intestine (S. I. Said and V. Mutt, Science Wash. DC 169: 1217-1218, 1970), has been identified in postganglionic autonomic axons of many tissues. VIP has potent regulatory effects on the function of various cell types within these tissues, ranging from relaxation of smooth muscle to ion transport. Recently, VIP has been implicated in the regulation of mucus secretion in the respiratory tract, a process involving release of macromolecules from exocrine cells and transport of ions and water across the airway mucosa. However, because airway glands and mucosa both consist of mixed cell populations, it was unclear which specific cells contained VIP receptors and contributed to VIP-evoked responses. We identified these specific cells by using immunocytochemical techniques to monitor concentration changes in adenosine 3',5'-cyclic monophosphate (cAMP), the intracellular compound known to mediate VIP responses. Serous and mucous cells of ferret tracheal submucosal glands and ciliated and basal cells of dog tracheal mucosa all increased cAMP in response to VIP stimulation. We conclude that these cell types possess VIP receptors and thus participate in VIP-stimulated responses. In contrast, ferret tracheal epithelium and dog epithelial goblet cells showed little or no reactivity after VIP, and thus we believe that these cells lack VIP receptors.

Comparison of lung vascular and epithelial permeability indices in the adult respiratory distress syndrome

Measurements of pulmonary clearance of inhaled 99mTc-DTPA and transvascular 113mIntransferrin flux were made in 12 patients with established ARDS and 14 volunteer control subjects (7 smokers and 7 nonsmokers). Smokers had significantly increased 99mTc-DTPA clearance (clearance rate constant, 3.6 +/- 0.8; mean +/- SEM) compared with nonsmokers (1.2 +/- 0.1). All patients with ARDS had increased clearance of 99mTc-DTPA (5.2 +/- 0.9), but the finding was nonspecific in that increased clearance overlapped with the findings in normal smokers. Protein flux in smokers (protein flux units, 0.0 +/- 0.2) was similar to that in nonsmokers (0.3 +/- 0.2). In 9 of the 12 patients with ARDS, protein flux was increased, and as a group (3.2 +/- 1.0) they differed significantly (p less than 0.01) from the combined smoking and nonsmoking control subjects (0.2 +/- 0.1, n = 14). The parameters of DTPA clearance and transvascular protein flux correlated well in the patients with ARDS (Spearman's rank correlation = 0.71, p less than 0.01). Although 99mTc-DTPA clearance is a sensitive technique in ARDS, a single study in this context does not allow a diagnostic conclusion because of its non-specificity. Abnormal protein flux appears to be more specific for ARDS but was not a universal finding in the patients studied.

Airway receptors and asthma

Airway caliber is determined by a balance between many constrictor and dilator agents, which bring about their effects on the various target cells of the airway by activating specific cell surface receptors. Asthma and bronchial hyperresponsiveness may be viewed as an imbalance between excitatory and inhibitory receptor-mediated effects on the various target cells in the airway. Airway receptors can be grouped into those mediating the effects of the autonomic nervous system on the airways or those mediating the effects of the various mediators which may be generated in asthma. There is no convincing evidence that a fundamental defect in receptor function is involved in the pathogenesis of asthma, although minor abnormalities have been described. Recently there has been a considerable increase in our understanding of receptor function and control, which should throw light on the pathogenesis of airway obstruction and may lead to advances in asthma therapy. This may also lead to the development of novel drugs for treating asthma in the future.

Non-adrenergic non-cholinergic neural control of human airways

In addition to classical cholinergic and adrenergic neural mechanisms, a third division of autonomic control has been recognised in human airways. Non-adrenergic inhibitory nerves are the dominant inhibitory neural pathway in human airway smooth muscle and there is increasing evidence that VIP and a related peptide, PHM, may be the neurotransmitters. These peptides are probably cotransmitters of acetylcholine in the airways and may modulate cholinergic effects. A defect in this system could occur in asthma because inflammation may more rapidly inactivate these neurotransmitter peptides. Non-cholinergic excitatory nerves have also been described in animal airways, although their existence in human airways is less certain. The neurotransmitter may be substance P or a related peptide neurokinin A, which could be released by axon reflex. Another peptide, calcitonin gene-related peptide, is colocalized with substance P and appears to be much more potent in human airways. Non-adrenergic non-cholinergic mechanisms may also regulate mucus secretion and the bronchial microvasculature. The role of this nervous system in health and disease is still uncertain as there are no specific blockers available.

Asthma as an axon reflex

In asthma, damage to airway epithelium, possibly caused by eosinophil products, exposes C-fibre afferent nerve endings. Stimulation of these endings by inflammatory mediators such as bradykinin may result in an axon (local) reflex with antidromic conduction down afferent nerve collaterals and release of sensory neuropeptides such as substance P, neurokinin A, and calcitonin gene-related peptide. These peptides are potent inducers of airway smooth muscle contraction, bronchial oedema, extravasation of plasma, mucus hypersecretion, and possibly inflammatory cell infiltration and secretion. Thus, axon reflexes could account for at least some of the pathophysiology of asthma and this concept might lead to new strategies for treatment.

Prostaglandin D2 potentiates airway responsiveness to histamine and methacholine

In the bronchi of asthmatic subjects many bronchoconstrictor mediators and neurotransmitters might be released together, and therefore, potential interactions might occur that could be important in airway hyperreactivity. We have studied the effect of inhaled methacholine, bradykinin, and prostaglandin D2 (PGD2) on bronchial reactivity to inhaled histamine in 6 mild asthmatic subjects, 22 to 36 yr of age. All of the test spasmogens were given at equivalent bronchoconstricting concentrations. Simultaneous dosing with PGD2 caused a significant increase in reactivity to histamine, mean dose of histamine causing a 35% fall in specific airway conductance being 0.72 mumol before, and 0.32 mumol with, PGD2; (p less than 0.01). This was not seen with histamine itself, methacholine, or bradykinin. Prostaglandin D2 caused a similar increase in bronchial reactivity to inhaled methacholine, suggesting a postreceptor potentiation of airway smooth muscle contractility. This positive interaction between inflammatory mediators known to be released in asthma has important implications for understanding bronchial hyperreactivity.

Adrenergic and non-adrenergic, non-cholinergic control of airways

In addition to cholinergic neural mechanisms, airway tone is influenced by adrenergic mechanisms and by more recently described neural mechanisms which are non-adrenergic and non-cholinergic (NANC). Sympathetic innervation to human airways is very sparse and there is no functional adrenergic innervation of smooth muscle, although sympathetic fibres may supply ganglia, submucosal glands and bronchial vessels. Airway tone may be influenced by circulating adrenaline and there is some evidence that adrenaline secretion may be impaired in asthma. beta-Adrenoceptors (which are almost entirely of the beta 2-subtype) are localized to many cell types in airways and beta-agonist may be beneficial in airway obstruction, not only by directly relaxing airway smooth muscle (from trachea to terminal bronchioles), but by inhibiting mast cell mediator release, by modulating cholinergic nerves, by reducing bronchial oedema and by reversing the defect in mucociliary clearance. There is little evidence that beta-receptor function is impaired in asthma. Alpha-adrenoceptors, which are bronchoconstrictor, may be activated by inflammatory mediators and disease, and alpha-agonists cause bronchoconstriction in asthmatic patients. However, alpha-antagonists have little effect, which questions the role of alpha-receptors in asthma. NANC nerves which relax human airways have been demonstrated in vitro. Although the neurotransmitter is not certain, there is now convincing evidence that it may be vasoactive intestinal peptide (VIP) and a related peptide histidine methionine (PHM). VIP and PHM immuno-active nerves are found in human airways, and both peptides potently relay human airways in vitro (but not in vivo because of diffusion and metabolism problems).(ABSTRACT TRUNCATED AT 250 WORDS)

Chronobiology and asthma. I. Day-night differences in bronchial patency and dyspnea and circadian rhythm dependencies

The symptoms of allergic asthmatic patients typically worsen during the night, especially during the early morning hours. Although 24-hour variations in the environment contribute to the intensification of the asthmatic condition nocturnally, environmental changes themselves do not fully explain the temporal aspects of this disease. Circadian (about 24-hour) rhythms in critical bioprocesses constitute significant contributory factors. The exacerbation of asthma during the night represents the changing status of biological functioning due to circadian rhythms in bronchial patency; airways hyperreactivity to acetylcholine, histamine, and house dust; and plasma cortisol, epinephrine, histamine, and cyclic AMP, among others.

Circadian variation in airway function

Nocturnal asthma is a common and troublesome problem. Many possible mechanisms have been proposed, including exposure to allergens, sleep itself, the supine posture, withdrawal of bronchodilator drugs, gastric reflux, mucus plugging, and airway cooling. Although these may be contributory factors in individual patients, they cannot provide a universal explanation for the phenomenon of nocturnal and early morning wheezing. It now seems that nocturnal asthma may best be understood in terms of circadian rhythms. A circadian variation in airway caliber has been demonstrated in normal subjects; in asthmatic subjects, the same rhythm is present but with greater amplitude. The amplitude is magnified by bronchial hyper-responsiveness, a cardinal feature of asthma. Evidence now suggests that the fall in circulating epinephrine level at night removes an important defense against bronchoconstriction in asthmatic subjects, and this itself may be magnified by removal of the braking effect of epinephrine on mast cell mediator release. In addition, increased vagal reflex bronchoconstriction and the delayed effects of the fall in plasma cortisol level may also contribute to nocturnal wheezing. Thus, nocturnal asthma may be explained by a complex interaction of several coincident circadian rhythms, which produce only small changes in airway caliber in normal subjects; however, in asthmatic patients, these constrictor effects are magnified to produce bronchospasm severe enough to wake the patient.

Plasma histamine concentration during propranolol induced bronchoconstriction

The mechanism of propranolol induced bronchoconstriction in asthma is uncertain, as airway beta adrenoceptors are not innervated by sympathetic nerves and circulating adrenaline concentrations are not raised. Propranolol 10 mg was infused over 27 minutes in 14 subjects with mild asthma. Peak expiratory flow (PEF) decreased by 80-235 l/min (17-51% of baseline) in nine subjects, who were called "responders," and by less than 50 l/min (12% of baseline) in five "non-responders". These two groups did not differ in baseline ventilatory function or in any clinical characteristic. In "responders" mean PEF had decreased significantly from 440 to 390 l/min after infusion of propranolol 2.1 mg, though the maximum fall in PEF occurred during or within five minutes of the end of the infusion. In nine of the subjects (six "responders" and three "non-responders") the possibility that propranolol induced bronchoconstriction is due to blockade of mast cell beta receptors leading to increased mediator release was examined by measurement of plasma histamine concentration as an index of mast cell degranulation. There was no consistent change in plasma histamine concentration in either group. No evidence of increased mast cell mediator release has been found in association with propranolol induced bronchoconstriction.

Mediators and asthma

Several mediators produced by various inflammatory cells have now been implicated in asthma. They may contribute to the pathology of asthma by contracting airway smooth muscle, stimulating mucus secretion, and promoting bronchial oedema and chemotaxis of inflammatory cells. Because many mediators interact, antagonism of any single mediator is unlikely to have a major clinical effect.

The effect of inhaled nifedipine on bronchial reactivity to histamine in man

We have administered nifedipine by aerosol to six patients with mild asthma to determine whether local administration of a potent calcium channel blocker has any effect on resting airway tone or histamine reactivity. Subjects had their responsiveness to histamine measured and then received either nifedipine, 10 mg in 40% ethanol, or diluent alone in a randomized, double-blind fashion. Specific airway conductance, blood pressure, and heart rate did not change after either inhalation. Histamine reactivity was significantly reduced after the nifedipine aerosol, the geometric mean provocative concentration causing a 35% fall in specific airway conductance, rising from 5.0 to 10.9 mg/ml of histamine (p less than 0.05). In individuals this protective effect was variable but overall was no greater than that observed after sublingual nifedipine. Plasma nifedipine concentrations were measured in two subjects after administration of the aerosol and confirmed that inhaled nifedipine is absorbed across the bronchial mucosa.