Voluntary and reflex cough and the expiration reflex; implications for aspiration after stroke.

Aspiration is a common result of stroke, and may lead to lung infections and pneumonia. Cough may prevent this aspiration and thus prevent the pneumonia. We review the four types of cough usually used to assess aspiration risk: voluntary cough (VC), reflex cough (RC), the laryngeal expiration reflex (LER), and cough on swallow (CoS). VC is easy to test but starts with an inspiration that may cause aspiration, and is controlled by cortico-brainstem pathways that may not be involved in influencing aspiration. RC also starts with an inspiration, and requires instrumental intervention, but is more relevant to protecting the lungs. The LER starts with an expiration, so is 'anti-aspiration', and is easy to test, but its neural mechanisms have not been fully analysed. CoS can be tested at the same time as direct observations of aspiration, but little is known about its neural mechanisms. Each method has its advocates, and the purpose of the review is to discuss how each may be applied and how the information from each may be assessed and valued.

Clinical cough V: complementary and alternative medicine: therapy of cough.

We review the actions of complementary and alternative medicines (CAMs) in the treatment of cough and of the conditions associated with it; in particular asthma and upper respiratory tract infections. These therapies may work (1) peripherally, at the sites in the airways and lungs at which cough is being activated, (2) in the brainstem, where the neural "cough center" is situated, or (3) at the cerebral cortex, where cough can be initiated, suppressed or modified by conscious or unconscious controls. Of the large number of trials of CAMs against cough, most are inadequate in design. It may be difficult to randomize selection. Blinding is often impossible both for the patient and the therapist, and adequate placebo controls may be difficult to devise. The patient can usually identify the "active" treatment by the taste or smell of a medicine, or from the approach and apparatus being used. Pure chemicals can be extracted from many of the herbs used as antitussives, and can be shown to be effective in randomized, blind, and controlled trials, but it does not follow that the herb itself, used in the recommended formula and shown to be antitussive, acts by this agency unless a placebo effect is ruled out. A few herbs are identified where the evidence points to a true antitussive action. Of nonherbal treatments, the few positive results are usually outweighed by the larger number of negative ones. Thus, in general, CAMs for cough are welcomed enthusiastically by the patient but lack sound evidence for their efficacy. Antitussive chemicals can be extracted from many herbs, but it is no more than a reasonable hypothesis that the herb itself acts through this pathway.

Cough as a symptom.

The Third International Symposium on Cough focused on acute and chronic cough, both clinical and basic science. Cough is a defensive and clearing mechanism, and can cause discomfort and nuisance; it is also an important symptom of many chronic airway diseases. In this Symposium, several issues concerning the relationship of cough to disease processes, different types of cough, and the concept of idiopathic cough will be discussed. Characterization of cough receptor(s) and identification of peripheral and central mechanisms for cough sensitization are current areas of investigation for delineating the cause of chronic cough. Peripheral mechanisms may be most important for acute cough such as after viral infections. The role of pathological changes at the level of the airway mucosa and of cortical pathways will be reviewed. Finally, therapeutic inhibition of the cough reflex remains an area of active research.

Overview of neural pathways in allergy and asthma.

Three groups of airway sensory nervous receptor may be involved in the pathophysiological changes in asthma and allergy. Those most active will be the C-fibre receptors, the rapidly adapting receptors, and A delta-nociceptive receptors. All are stimulated or sensitised by the inflammatory and immunological changes. The C-fibre receptors may mediate the axon reflex neurogenic inflammation-bronchoconstriction, mucus secretion and mucosal hyperaemia due to neuropeptide release-but the evidence for this process in humans, unlike rodents, is scanty. Activation of the receptors will also cause central nervous reflexes. The pathways for these reflexes in the brainstem, where their interactions, and the chemical neurotransmitters involved, are beginning to be delineated. The resulting reflexes include bronchoconstriction, mucus secretion and mucosal vasodilatation, responses that will amplify any similar changes due to neurogenic inflammation. The cough reflex depends on the interaction of the three basic reflex pathways. The reflexes show plasticity at peripheral, ganglionic and central nervous levels, and it is unlikely that results in acute experiments on healthy animals will apply quantitatively to humans with asthma.

The pyrophysiology and sexuality of dragons.

To examine the means whereby dragons produce fire and steam, we have studied a related species, the desert-lizard Lacerta pyrophorus. Morphological studies showed that there were in the snout three distinctive features: (1) a dorsal swelling in the pharynx, the Organ of Feuerwerk, consisting of brown adipose tissue with an extensive sympathetic innervation; (2) greatly enlarged lachrymonasal ducts, the Ducts of Kwentsch; and (3) asbestos deposits in the nasal skin, the Bestos Bodies. Physiological studies show that the Organ of Feuerwerk can, when the animal is excited, produce extremely high temperatures. We discuss how these mechanisms can produce steam and fire, and how the snout is protected. We also discuss and offer a solution to the problem of how, since dragons are invariably male, the species can be propagated.

Modulation of airway sensitivity to inhaled irritants: role of inflammatory mediators.

Bronchopulmonary C-fiber endings and rapidly adapting pulmonary receptors (RARs) are primarily responsible for eliciting the defense reflexes in protecting the lungs against inhaled irritants. In anesthetized animals, inhalation of cigarette smoke, one of the common inhaled irritants, into the lungs elicits pulmonary chemoreflexes that are mediated through the stimulation of pulmonary C fibers. When the C-fiber conduction is selectively blocked in the vagus nerves, the same smoke inhalation triggered only augmented breaths, a reflex effect of activating RARs, in the same animals. Indeed, electrophysiologic study shows that inhaled smoke exerts a direct stimulatory effect on both types of afferents. Increasing evidence indicates that the excitability of these afferents and therefore their reflex actions are enhanced by airway mucosal inflammation; one such example is the airway hyperresponsiveness induced by acute exposure to ozone. Although the mechanism underlying the inflammation-induced hypersensitivity of C-fiber endings is not fully understood, the possible involvement of local release of certain inflammatory mediators, such as histamine and prostaglandin E(2) (PGE(2), should be considered. It is believed that changes in the membrane properties mediated by the activation of certain specific receptor proteins located on the membrane of these nerve terminals are involved, as the sensitizing effects of PGE(2) can be also demonstrated in cultured pulmonary C neurons.

Advances in understanding and treatment of cough.

Many different conditions and diseases cause cough. The commonest acute causes are pollution, including cigarette smoke, and upper respiratory tract infection. The commonest chronic causes are postnasal drip, asthma, chronic bronchitis and gastro-oesophageal reflux. Epidemiological studies give widely different patterns of incidence. The different conditions that cause cough have in common the fact that the cough is mediated via the vagus nerves, with sensory receptors in and under the epithelium from the larynx down to the smaller bronchi. These receptors are polymodal, responding to a large variety of stimuli, including mechanical and chemical irritants, inflammatory mediators, intraluminal material and large volume changes of the lungs. With irritation and inflammation, C fibre receptors release neurokinins such as substance P, which in turn stimulate cough receptors. The central nervous pathways for the cough reflex are poorly understood. They can be activated or inhibited voluntarily. Studies on the pharmacology of the central nervous pathways of coughing are opening up new therapeutic possibilities. Other new therapies include drugs acting on the sensory receptors for cough, thereby avoiding adverse central nervous effects.

Afferent receptors in the airways and cough.

The roles of airway rapidly adapting receptors (RARs) and of C-fibre receptors in the induction of cough are reviewed. It is concluded that, while there is substantial evidence that irritant receptors in the laryngeal wall and RARs in the tracheobronchial mucosa can cause cough, the evidence for such a similar direct role for C-fibre receptors is tenuous. Indeed there is accumulating evidence that the C-fibre receptors may cause apnoea and rapid shallow breathing, and also reflexly inhibit cough. However the C-fibre receptors may release tachykinins when stimulated, and these in turn may cause plasma extravasation from mucosal postcapillary venules. RARs are excited by increases in interstitial liquid volume, so C-fibre receptors may indirectly enhance cough via the RARs.

Autonomic regulation. i-NANC/e-NANC.

The excitatory and inhibitory nonadrenergic/noncholinergic (e-NANC, i-NANC) systems have been extensively studied. The terms excitatory and inhibitory apply to airway smooth muscle, but the neurotransmitters also act on other targets-blood vessels, glands, the epithelium-where individual actions may be the opposite. Thus, the nomenclature is unsatisfactory. Of the dozen or more putative NANC transmitters, criteria to establish their roles have been met only for vasoactive intestinal polypeptide (VIP), nitric oxide (NO), and substance P/neurokinin A (SP/NKA). VIP and NO co-localize in vagal motor nerves, but they are also found in sympathetic and sensory nerves. In general they have similar actions on target tissues, and their relative importance may vary with species. SP/NKA, released from sensory nerves, is thought to mediate neurogenic inflammation, a process that may include airway smooth muscle contraction, at least in rodents. The evidence for neurogenic inflammation in humans is weak. On the motor side, and also possibly on the sensory, different nerves seem to contain different selections of neurotransmitters, but it is not known if there are different motor controls for these nerves. Cotransmission presents a major conceptual and experimental problem, since the two or more transmitters may give opposite instructions to the target tissue. Inevitably most of the studies on the NANC systems are on isolated rodent tissues, and although quantitative, they indicate little of what happens in vivo, and certainly not in humans. The cocktail of mediators that must be released from nerves and associated cells in airway tissues during pathophysiologic processes may refresh physiologists, but little is known about the concentrations of the ingredients or about the strength of their actions and their interactions on different target tissues in the mucosa.

PAF-induced secretory hyperresponsiveness in the ferret trachea to bradykinin and its pharmacological inhibition.

Both PAF (10 microM) and bradykinin (0.1-10 microM) increased lysozyme (from submucosal gland serous cells (+138 and +45% for PAF, 10 microM, and bradykinin, 1 microM, respectively) and albumin (mainly active epithelial transport; +387 and +108%) outputs into the ferret tracheal lumen in vitro and reduced the negativity of the potential difference (PD: -33 and -17%) across the trachea. Since PAF can cause bronchial smooth muscle hyperresponsiveness, we tested whether these effects were interactive, and if PAF would increase the actions of bradykinin. The bradykinin-induced lysozyme and albumin outputs were more than trebled and the PD change was enhanced by PAF, after the immediate secretory effects of the latter had returned to baseline. The secretory and PD responses to PAF were all prevented by the PAF-antagonist WEB 2086 and by a combination of the free-radical scavengers catalase and SOD, indicating that PAF may act on specific receptors to release free-radicals. Nedocromil sodium inhibited the increase in lysozyme and albumin outputs produced by PAF, but had no effect on the PD response. None of the tracheal responses to bradykinin was modified by WEB 2086, catalase and SOD, or nedocromil sodium. The secretory and PD hyperresponsiveness to bradykinin caused by PAF was prevented by WEB 2086 and by catalase and SOD. Nedocromil sodium greatly inhibited the lysozyme and albumin hyperresponsiveness but had no effect on the PD response. Thus PAF may release more than one type of radical which have differential effects on serous cells and albumin transport compared with PD; nedocromil sodium may act only against the radical causing the secretory effects.

H2O2 increases sheep tracheal blood flow, permeability, and vascular response to luminal capsaicin.

Exogenous hydrogen peroxide (H2O2) causes airway epithelial damage in vitro. We have studied the effects of luminal H2O2 in the sheep trachea in vivo on tracheal permeability to low-molecular-weight hydrophilic (technetium-99m-labeled diethylenetriamine pentaacetic acid; 99mTc-DTPA) and lipophilic ([14C]antipyrine; [14C]AP) tracers and on the tracheal vascular response to luminal capsaicin, which stimulates afferent nerve endings. A tracheal artery was perfused, and tracheal venous blood was collected. H2O2 exposure (10 mM) reduced tracheal potential difference (-42.0 +/- 6.4 mV) to zero. It increased arterial and venous flows (56.7 +/- 6.1 and 57.3 +/- 10.0%, respectively; n = 5, P < 0.01, paired t-test) but not tracheal lymph flow (unstimulated flow 5.0 +/- 1.2 microliters.min-1.cm-1, n = 4). During H2O2 exposure, permeability to 99mTc-DTPA increased from -2.6 to -89.7 x 10(-7) cm/s (n = 5, P < 0.05), whereas permeability to [14C]AP (-3,312.6 x 10(-7) cm/s, n = 4) was not altered significantly (-2,565 x 10(-7) cm/s). Luminal capsaicin (10 microM) increased tracheal blood flow (10.1 +/- 4.1%, n = 5) and decreased venous 99mTc-DTPA concentration (-19.7 +/- 4.0, P < 0.01), and these effects were significantly greater after epithelial damage (28.1 +/- 6.0 and -45.7 +/- 4.3%, respectively, P < 0.05, unpaired t-test). Thus H2O2 increases the penetration of a hydrophilic tracer into tracheal blood and lymph but has less effect on a lipophilic tracer. It also enhances the effects of luminal capsaicin on blood flow and tracer uptake.

Sensory neurophysiology of the cough reflex.

The epithelium of the larynx, trachea, and larger bronchi contains sensory nerves that are responsible for cough. Their two main categories are rapidly adapting receptors (RARs) and C fiber receptors. Both types respond to a wide variety of mechanical and chemical irritants. The RARs are the main sensory complex responsible for cough. C fiber receptors cause neurogenic inflammation by the release of tachykinins such as substance P. The reflex action of C fiber receptors seems to be inhibition of cough. However, the released tachykinins can stimulate RARs and promote or enhance the cough response. The strength and pattern of cough depends on the sites of the airway that are stimulated and the local and central reflex interactions of the RARs and C fiber receptors. Tachykinins seem to be involved in cough, but their role needs further study.

Sodium metabisulphite causes epithelial damage and increases sheep tracheal blood flow and permeability.

Inhaled sodium metabisulphite (MBS) causes bronchoconstriction, cough and microvascular leakage. We have studied its effects on tracheal blood flow, potential difference (PD) and the permeability from tracheal lumen to venous blood of a low molecular weight hydrophilic tracer, 99mtechnetium-labelled diethylenetriamine penta-acetic acid (99mTc-DTPA) in anaesthetized sheep. Flow was measured in tracheal artery and blood from a cannulated tracheal vein collected for 5 min periods. The tracheal lumen was filled with Krebs-Henseleit solution (KH) containing 99mTc-DTPA for six to eight 15 min periods. During the third or fourth period, MBS (1, 20 or 100 mM) was washed into the tracheal lumen for 15 min. MBS increased tracheal blood flow (venous flow (Q'v), 5-10 min MBS exposure period: 1 mM -9 +/- 18% (n = 3); 20 mM +16 +/- 5% (n = 5; p < 0.05); 100 mM +43 +/- 13% (n = 5; p < 0.05). It decreased PD in a concentration-dependent way. Venous 99mTc-DTPA concentration increased progressively to +266 +/- 176 and +958 +/- 321% 25-30 min after exposure to 20 and 100 mM MBS, respectively (p < 0.05 for both). These effects were not blocked by luminal frusemide (3-7 mM) or flurbiprofen (100-500 microM). Histological sections showed changes to the epithelial cells and large intercellular spaces. Thus, luminal sodium metabisulphite increases tracheal blood flow, reduces transmural potential difference and causes tracheal epithelial damage, leading to an increase in 99mTc-labelled diethylenetriamine penta-acetic acid permeability.