Chapter 36: Pulmonary Pharmacolog

Particle Size

The size of particles for inhalation is of critical importance in determining the site of deposition in the respiratory tract. The optimum size for particles to settle in the airways is 2-5 μm mass median aerodynamic diameter (MMAD). Larger particles settle out in the upper airways, whereas smaller particles remain suspended and are therefore exhaled. There is increasing interest in delivering drugs to small airways, particularly in COPD and severe asthma (Sturton et al., 2008). This involves delivering drug particles of ∼1 μm MMAD, which is now possible using drugs formulated in hydrofluoroalkane (HFA) propellant.

Pharmacokinetics

Of the total drug delivered, only 10-20% enters the lower airways with a conventional pressurized metered-dose inhaler (pMDI). The fate of the inhaled drug is poorly understood. Drugs are absorbed from the airway lumen and have direct effects on target cells of the airway. Drugs may also be absorbed into the bronchial circulation and then distributed to more peripheral airways. Whether drugs are metabolized in the airways is often uncertain, and there is little understanding of the factors that may influence local absorption and metabolism of inhaled drugs. Drugs with higher molecular weights tend to be retained to a greater extent in the airways. Nevertheless, several drugs have greater therapeutic efficacy when given by the inhaled route. The inhaled corticosteroid ciclesonide is a prodrug activated by esterases in the respiratory tract to the active principle des-ciclesonide. More extensive pulmonary distribution of a drug with a smaller MMAD increases alveolar deposition and thus is likely to increase absorption from the lungs into the general circulation resulting in more systemic side effects. Thus, although HFA pMDIs deliver more inhaled corticosteroid to smaller airways, there is also increased systemic absorption, so that the therapeutic ratio may not be changed.

Delivery Devices

Several ways of delivering inhaled drugs are possible (Virchow et al., 2008).

Pressurized Metered-Dose Inhalers. Drugs are propelled from a canister with the aid of a propellant, previously with a chlorofluorocarbon (Freon) but now replaced by a hydrofluoroalkane (HFA) that is "ozone friendly." These devices are convenient, portable, and typically deliver 100-400 doses of drug. It is necessary to coordinate inhalation with activation of the device, so it is important that patients are taught to use these devices correctly. Many patients find this difficult despite instruction.

Spacer Chambers. Large-volume spacer devices between the pMDI and the patient reduce the velocity of particles entering the upper airways and the size of the particles by allowing evaporation of liquid propellant. This reduces the amount of drug that impinges on the oropharynx and increases the proportion of drug inhaled into the lower airways. The need for careful coordination between activation and inhalation is also reduced because the pMDI can be activated into the chamber and the aerosol subsequently inhaled from the one-way valve. Perhaps the most useful application of spacer chambers is in the reduction of the oropharyngeal deposition of inhaled corticosteroids and the consequent reduction in the local side effects of these drugs. Large volume spacers also reduce the systemic side effects of drugs because less is deposited in the oropharynx, and therefore swallowed. It is the swallowed fraction of the drug absorbed from the GI tract that makes the greatest contribution to the systemic fraction. This is of particular importance in the use of certain inhaled steroids, such as beclomethasone dipropionate, which can be absorbed from the GI tract. Spacer devices are also useful in delivering inhaled drugs to small children who are not able to use a pMDI. Children as young as 3 years of age are able to use a spacer device fitted with a face mask.

Dry Powder Inhalers. Drugs may also be delivered as a dry powder using devices that scatter a fine powder dispersed by air turbulence on inhalation. These devices may be preferred by some patients (Chan, 2006) because careful coordination is not as necessary as with the pMDI, but some patients find that the dry powder is an irritant. Children <7 years of age find it difficult to use a dry powder inhaler (DPI) because they may not be able to generate sufficient inspiratory flow. DPIs have been developed to deliver peptides and proteins, such as insulin (e.g., exubera, afresa), systemically.

Nebulizers. Two types of nebulizer are available. Jet nebulizers are driven by a stream of gas (air or oxygen), whereas ultrasonic nebulizers use a rapidly vibrating piezo-electric crystal and thus do not require a source of compressed gas. The nebulized drug may be inspired during tidal breathing, and it is possible to deliver much higher doses of drug compared with pMDI. Nebulizers are therefore useful in treating acute exacerbations of asthma and COPD, for delivering drugs when airway obstruction is extreme (e.g., in severe COPD), for delivering inhaled drugs to infants and small children who cannot use the other inhalation devices, and for giving drugs such as antibiotics when relatively high doses must be delivered. Small handheld nebulizers (soft mist inhalers) are now also available.

Oral Route

Drugs for treatment of pulmonary diseases may also be given orally. The oral dose is much higher than the inhaled dose required to achieve the same effect (typically by a ratio of ∼20:1), so that systemic side effects are more common. When there is a choice of inhaled or oral route for a drug (e.g., β₂ agonist or corticosteroid), the inhaled route is always preferable, and the oral route should be reserved for the few patients unable to use inhalers (e.g., small children, patients with physical problems such as severe arthritis of the hands). Theophylline is ineffective by the inhaled route and therefore must be given systemically. Corticosteroids may have to be given orally for parenchymal lung diseases (e.g., in interstitial lung diseases), although it may be possible in the future to deliver such drugs into alveoli using specially designed inhalation devices with a small particle size.

Parenteral Route

The intravenous route should be reserved for delivery of drugs in the severely ill patient who is unable to absorb drugs from the GI tract. Side effects are generally frequent due to the high plasma concentrations.

Chemistry. The development of β₂ agonists is based on substitutions in the catecholamine structure of norepinephrine and epinephrine (Chapter 12). The catechol ring consists of hydroxyl groups in the 3 and 4 positions of the benzene ring (Figure 36–4). Norepinephrine differs from epinephrine only in the terminal amine group; in general, further modification at this site confers β receptor selectivity. Further substitution of the terminal amine resulted in β₂ receptor selectivity, as in albuterol (salbutamol) and terbutaline. Exogenous catecholamines are rapidly metabolized by catechol-O-methyl transferase (COMT), which methylates in the 3-hydroxyl position, and accounts for the short duration of action of catecholamines. Modification of the catechol ring, as in albuterol and terbutaline, prevents this degradation and therefore prolongs the effect. Catecholamines are also broken down in sympathetic nerve terminals and in the GI tract by monoamine oxidase (MAO), which cleaves the side chain. Isoproterenol, which is a substrate for MAO, is metabolized in the gut, making absorption variable. Substitution in the amine group confers resistance to MAO and ensures reliable absorption. Many other β₂-selective agonists have now been introduced and, although there may be differences in potency, there are no clinically significant differences in selectivity. Inhaled β₂-selective drugs in current clinical use (apart from rimiterol [not available in the U.S.] that is broken down by COMT) have a similar duration of action (3-6 hours). The inhaled long-acting inhaled β₂ agonists salmeterol and formoterol have a much longer duration of effect, providing bronchodilation and bronchoprotection for >12 hours (Kips and Pauwels, 2001). Formoterol has a bulky substitution in the aliphatic chain and has a moderate lipophilicity, which appears to keep the drug in the membrane close to the receptor, so it behaves as a slow-release drug. Salmeterol has a long aliphatic chain, and its long duration may be due to binding within the receptor binding cleft ("exosite") that anchors the drug in the binding cleft. Once-daily β₂ agonists, such as indacaterol, with a duration of action >24 hours, are now in development (Cazzola and Matera, 2008).

Mode of Action

Occupation of β₂ receptors by agonists results in the activation of the G_s-adenylyl cyclase-cAMP-PKA pathway, resulting in phosphorylative events leading to bronchial smooth muscle relaxation (Figure 36–5). β Agonists produce bronchodilation by directly stimulating β₂ receptors in airway smooth muscle, and in vitro relax human bronchi and lung strips (indicating an effect on peripheral airways). In vivo there is a rapid decrease in airway resistance. β₂ Receptors have been localized to airway smooth muscle of all airways by direct receptor binding techniques and autoradiographic mapping studies.

The molecular mechanisms by which β agonists induce relaxation of airway smooth muscle include:

• Lowering of [Ca²⁺]_i concentration by active removal of Ca²⁺ from the cytosol into intracellular stores and out of the cell

• Acute inhibition of the PLC-IP₃ pathway and its mobilization of cellular Ca²⁺ (prolonged use can upregulate expression of PLC_β1)

• Inhibition of myosin light chain kinase activation

• Activation of myosin light chain phosphatase

• Opening of a large conductance Ca²⁺-activated K⁺ channel (K_Ca), which repolarizes the smooth muscle cell and may stimulate the sequestration of Ca²⁺ into intracellular stores. β₂ Receptors may also couple to K_Ca via G_s so that relaxation of airway smooth muscle may occur independently of an increase in cAMP.

Several actions of β₂ agonists are mediated not by PKA but by other cAMP-regulated proteins, such as the exchange protein activated by cAMP (EPAC) (Holz et al., 2006). β₂ Agonists act as functional antagonists and reverse bronchoconstriction irrespective of the contractile agent. This is an important property for the treatment of asthma because many bronchoconstrictor mechanisms (neurotransmitters and mediators) are likely to be contributory. In COPD the major mechanism of action is likely to be reduction of cholinergic neural bronchoconstriction.

β₂ Receptors are localized to several different airway cells, where they may have additional effects. Thus, β₂ agonists may cause bronchodilation in vivo not only via a direct action on airways smooth muscle, but also indirectly by inhibiting the release of bronchoconstrictor mediators from inflammatory cells and of bronchoconstrictor neurotransmitters from airway nerves. These mechanisms include:

• Prevention of mediator release from isolated human lung mast cells (via β₂ receptors).

• Prevention of microvascular leakage and thus the development of bronchial mucosal edema after exposure to mediators, such as histamine and leukotriene D₄.

• Increase in mucus secretion from submucosal glands and ion transport across airway epithelium; these effects may enhance mucociliary clearance, and thereby reverse the defective clearance found in asthma.

• Reduction in neurotransmission in human airway cholinergic nerves by an action at presynaptic β₂ receptors to inhibit acetylcholine release. This may contribute to their bronchodilator effect by reducing reflex cholinergic bronchoconstriction.

Although these additional effects of β₂ agonists may be relevant to the prophylactic use of these drugs against various challenges, their rapid bronchodilator action is probably attributable to a direct effect on airway smooth muscle.

Anti-Inflammatory Effects

Whether β₂ agonists have anti-inflammatory effects in asthma is controversial. The inhibitory effects of β₂ agonists on mast cell mediator release and microvascular leakage are clearly anti-inflammatory, suggesting that β₂ agonists may modify acute inflammation. However, β₂ agonists do not appear to have a significant inhibitory effect on the chronic inflammation of asthmatic airways, which is suppressed by corticosteroids. This has now been confirmed by several biopsy and bronchoalveolar lavage studies in patients with asthma who are taking regular β₂ agonists (including long-acting inhaled β₂ agonists), that demonstrate no significant reduction in the number or activation in inflammatory cells in the airways, in contrast to resolution of the inflammation that occurs with ICS. This may be related to the fact that effects of β₂ agonists on macrophages, eosinophils, and lymphocytes are rapidly desensitized.

Oral β₂ agonists are occasionally indicated as an additional bronchodilator. Slow-release preparations (e.g., slow-release albuterol and bambuterol [a prodrug that slowly releases terbutaline but is not commercially available in the U.S.]) may be indicated in nocturnal asthma; however, these agents are less useful than inhaled β agonists because of an increased risk of side effects. Several short-acting β₂ agonists are available. With the exception of rimiterol (which retains the catechol ring structure and is therefore susceptible to rapid enzymatic degradation), they are resistant to uptake and enzymatic degradation by COMT and MAO. There is little to choose between the various short-acting β₂ agonists currently available; all are usable by inhalation and orally, have a similar duration of action (∼3-4 hours; less in severe asthma), and similar side effects. Differences in β₂ selectivity have been claimed but are not clinically important. Drugs in clinical use include albuterol (salbutamol), levalbuterol, metaproterenol, terbutaline, as well as several not available in the U.S., fenoterol, tulobuterol, rimiterol, and pirbuterol.

Tolerance to the bronchodilator effect of formoterol and the bronchoprotective effects of formoterol and salmeterol have been demonstrated, but this is a small loss of protection that does not appear to be progressive and is of doubtful clinical significance. Although both drugs have a similar duration of effect in clinical studies, there are differences. Formoterol has a more rapid onset of action and is an almost full agonist, whereas salmeterol is a partial agonist with a slower onset of action. These differences might confer a theoretical advantage for formoterol in more severe asthma, whereas it may also make it more likely to induce tolerance. However, no significant clinical differences between salmeterol and formoterol have been found in the treatment of patients with severe asthma (Nightingale et al., 2002).

In COPD, LABA are effective bronchodilators that may be used alone or in combination with anticholinergics or ICS. LABA improve symptoms and exercise tolerance by reducing both air trapping and exacerbations. In asthma patients, LABA should never be used alone because they do not treat the underlying chronic inflammation; rather, LABA should always be used in combination with ICS (preferably in a fixed-dose combination inhaler). LABA are an effective add-on therapy to ICS and are more effective than increasing the dose of ICS when asthma is not controlled at low doses.

Recently, a combination inhaler that contains formoterol and budesonide was shown to be more effective for relieving acute symptoms than either terbutaline or formoterol alone, suggesting that the inhaled corticosteroids may also be contributing to the benefit (Rabe et al., 2006). This may make it possible to control asthma with a single inhaler both for maintenance and relief of symptoms.

Stereoselective β₂ Agonists. Albuterol is a racemic mixture of active R- and inactive S-isomers. Animal studies suggest that the S-isomer may increase airway responsiveness, providing a rationale for the development of R-albuterol (levalbuterol). Although the R-isomer is more potent than racemic R/S-albuterol in some studies, careful dose responses show no advantage in terms of efficacy and no evidence that the S-albuterol is detrimental in asthmatic patients (Lotvall et al., 2001). Because levalbuterol is more expensive than normally used racemic albuterol, this therapy has no clear clinical advantage (Barnes, 2006b). Stereoselective formoterol (R,R-formoterol, arformoterol) has now been developed as a nebulized solution but also appears to offer no clinical advantage over racemic formoterol (Madaan, 2009).

β₂ Receptor Polymorphisms. Several single nucleotide polymorphisms and haplotypes of the human β₂ adrenergic receptor gene (ADRβ2), which affect the structure of β₂ receptors, have been described. The common variants are Gly¹⁶Arg and Gln²⁷Glu, which have in vitro effects on receptor desensitization, but clinical studies have shown inconsistent effects on the bronchodilator responses to short- and long-acting β₂ agonists (Hawkins et al., 2008). Some studies have shown that patients with the common homozygous Arg¹⁶Arg variant have more frequent adverse effects and a poorer response to short-acting β₂ agonists than heterozygotes or Gly¹⁶Gly homozygotes, but overall these differences are small, and there appears to be no clinical value in measuring ADRβ2 genotype. No differences have been found with responses to LABA between these genotypes (Bleecker et al., 2007).

Side Effects. Unwanted effects are dose related and due to stimulation of extrapulmonary β receptors (Table 36–1 and Chapter 12). Side effects are not common with inhaled therapy but quite common with oral or intravenous administration.

• Muscle tremor due to stimulation of β₂ receptors in skeletal muscle is the most common side effect. It may be more troublesome with elderly patients and so is a more frequent problem in COPD patients.

• Tachycardia and palpitations are due to reflex cardiac stimulation secondary to peripheral vasodilation, from direct stimulation of atrial β₂ receptors (human heart has a relatively high proportion of β₂ receptors; see Chapter 12), and possibly also from stimulation of myocardial β₁ receptors as the doses of β₂ agonist are increased. These side effects tend to disappear with continued use of the drug, reflecting the development of tolerance. There can be a dose-related prolongation of the corrected QT interval (QTc).

• Hypokalemia is a potentially serious side effect. This is due to β₂ receptor stimulation of potassium entry into skeletal muscle, which may be secondary to a rise in insulin secretion. Hypokalemia might be serious in the presence of hypoxia, as in acute asthma, when there may be a predisposition to cardiac arrhythmias (Chapter 29). In practice, however, significant arrhythmias after nebulized β₂ agonists are rarely observed in acute asthma or COPD patients.

• Ventilation-perfusion (V/Q) mismatch due to pulmonary vasodilation in blood vessels previously constricted by hypoxia, resulting in the shunting of blood to poorly ventilated areas and a fall in arterial oxygen tension. Although in practice the effect of β₂ agonists on PaO₂ is usually very small (<5 mm Hg fall), occasionally in severe COPD it can be large, although it may be prevented by giving additional inspired oxygen (O₂).

• Metabolic effects (increase in free fatty acid, insulin, glucose, pyruvate, and lactate) are usually seen only after large systemic doses.

Tolerance. Continuous treatment with an agonist often leads to tolerance (desensitization, subsensitivity), which may be due to down-regulation of the receptor (Chapter 12). For this reason there have been many studies of bronchial β receptor function after prolonged therapy with β₂ agonists. Tolerance of non-airway β₂ receptor–mediated responses, such as tremor and cardiovascular and metabolic responses, is readily induced in normal and asthmatic subjects. Tolerance of human airway smooth muscle to β₂ agonists in vitro has been demonstrated, although the concentration of agonist necessary is high and the degree of desensitization is variable. In normal subjects, bronchodilator tolerance is not a consistent finding after high-dose inhaled albuterol. In asthmatic patients tolerance to the bronchodilator effects of β₂ agonists has not usually been found. However, tolerance develops to the bronchoprotective effects of β₂ agonists, and this is more marked with indirect bronchoconstrictors that activate mast cells (e.g., adenosine, allergen, and exercise) than with direct bronchoconstrictors, such as histamine and methacholine. The reason for the relative resistance of airway smooth muscle β₂ responses to desensitization remains uncertain but may reflect the large receptor reserve: >90% of β₂ receptors may be lost without any reduction in the relaxation response. The high level of ADRβ2 expression in airway smooth muscle compared with peripheral lung may also contribute to the resistance to tolerance because a high rate of β receptor synthesis is likely. In addition, the expression of GRK2, which phosphorylates and inactivates occupied β₂ receptors, is very low in airway smooth muscle (McGraw and Liggett, 1997). By contrast there is no receptor reserve in inflammatory cells, GRK2 expression is high, and tolerance to β₂ agonists rapidly develops at these sites.

Experimental studies have shown that corticosteroids prevent the development of tolerance in airway smooth muscle, and prevent and reverse the fall in pulmonary β receptor density (Mak et al., 1995). However, ICS fail to prevent the tolerance to the bronchoprotective effect of inhaled β₂ agonists, possibly because they do not reach airway smooth muscle in a high enough concentration.

Long-Term Safety. Because of a possible relationship between adrenergic drug therapy and the rise in asthma deaths in several countries during the early 1960s, doubts were cast on the long-term safety of β agonists. A causal relationship between β agonist use and mortality has never been firmly established, although in retrospective studies this would not be possible. A particular β₂ agonist, fenoterol, was linked to the rise in asthma deaths in New Zealand in the early 1990s because significantly more of the fatal cases were prescribed fenoterol than the case-matched control patients. An epidemiological study examined the links between drugs prescribed for asthma and death or near death from asthma attacks, based on computerized records of prescriptions. There was a marked increase in the risk of death with high doses of all inhaled β₂ agonists (Spitzer et al., 1992). The risk was greater with fenoterol, but when the dose is adjusted to the equivalent dose for albuterol, there is no significant difference in the risk for these two drugs. The link between high β₂ agonist usage and increased asthma mortality does not prove a causal association because patients with more severe and poorly controlled asthma, who are more likely to have an increased risk of fatal attacks, are more likely to be using higher doses of β₂ agonist inhalers and less likely to be using effective anti-inflammatory treatment. Indeed, in the patients who used regular inhaled steroids there was a significant reduction in risk of death (Suissa et al., 2000).

Regular use of inhaled β₂ agonists has also been linked to increased asthma morbidity. Regular use of fenoterol was associated with worse asthma control and a small increase in airway hyperresponsiveness compared with patients using fenoterol "on demand" for symptom control over a 6-month period (Sears et al., 1990). However, this was not found in a study with regular albuterol (Dennis et al., 2000). There is some evidence that regular inhaled β₂ agonists may increase allergen-induced asthma and sputum eosinophilia (Mcivor et al., 1998). A possible mechanism is that β₂ agonists upregulate expression of PLC_β1, resulting in augmentation of the bronchoconstrictor responses to cholinergic agonists and mediators (McGraw et al., 2003). Short-acting inhaled β₂ agonists should only be used on demand for symptom control, and if they are required frequently (more than three times weekly), an ICS is needed.

The safety of LABA in asthma remains controversial. A large study of the safety of salmeterol showed an excess of respiratory deaths and near deaths in patients prescribed salmeterol, but these deaths occurred mainly in African Americans living in inner cities who were not taking ICS (Nelson et al., 2006). Similar data have also raised concerns about formoterol. This may be predictable because LABA do not treat the underlying chronic inflammation of asthma. However, concomitant treatment with ICS appears to obviate such risk, so it is recommended that LABA should only be used when ICS are also prescribed (preferably in the form of a combination inhaler so that the LABA can never be taken without the inhaled corticosteroids) (Jaeschke et al., 2008). All LABA approved in the U.S. carry a black box warning cautioning against overuse. Studies are underway to examine their long-term safety profile, especially in children with asthma. There are less safety concerns with LABA use in COPD. No major adverse effects were reported in a large study over 3 years in COPD patients and in several other studies (Calverley et al., 2007; Rodrigo et al., 2008).

Future Developments

β₂ Agonists will continue to be the bronchodilators of choice for asthma in the foreseeable future because they are effective in all patients and have few or no side effects when used in low doses. It would be difficult to find a bronchodilator that improves on the efficacy and safety of inhaled β₂ agonists. Although some concerns have been expressed about the long-term effects of short-acting inhaled β₂ agonists, when used as required for symptom control, inhaled β₂ agonists appear safe. Use of large doses of inhaled β₂ agonists indicates poor asthma control; such patients should be assessed and appropriate controller medication used. LABA are a very useful option for long-term control in asthma and COPD. In asthma patients LABA should probably only be used if the patient is receiving concomitant ICS. There is little advantage to be gained by improving β₂ receptor selectivity because most of the side effects of these agents are due to β₂ receptor stimulation (muscle tremor, tachycardia, hypokalemia). Several once daily inhaled β₂ agonists, such as indacaterol and carmoterol, are now in clinical development (Cazzola and Matera, 2008).

Chemistry. Theophylline is a methylxanthine similar in structure to the common dietary xanthines caffeine and theobromine. Several substituted derivatives have been synthesized, but only two appear to have any advantage over theophylline: the 3-propyl derivative, enprofylline, which is more potent as a bronchodilator and may have fewer toxic effects because it does not antagonize adenosine receptors; and doxofylline (7-[1,3-dioxalan-2-ylmethyl] theophylline), a novel methylxanthine available in some countries (Dini and Cogo, 2001). Doxofylline, which has a dioxolane group at position 7, has an inhibitory effect on phosphodiesterases (PDEs) similar to that of theophylline but is less active as an adenosine antagonist and may have a more favorable side-effect profile. Many salts of theophylline have also been marketed; the most common is aminophylline, which is the ethylenediamine salt used to increase its solubility at neutral pH. Other salts either do not have any advantage (e.g., oxtriphylline [choline theophyllinate]) or are virtually inactive (e.g., acepifylline). Thus, theophylline remains the major methylxanthine in clinical use.

Mechanism of Action. The mechanisms of action of theophylline are still uncertain. In addition to its bronchodilator action, theophylline has many nonbronchodilator effects that may be relevant to its effects in asthma and COPD (Figure 36–6). Many of these molecular effects are seen only at high concentrations that exceed the therapeutic range, however. Several molecular mechanisms of action have been proposed:

• Inhibition of phosphodiesterases. Theophylline is a nonselective PDE inhibitor, but the degree of inhibition is relatively minimal at concentrations of theophylline that are within the so-called therapeutic range. PDE inhibition and the concomitant elevation of cellular cAMP and cyclic guanosine monophosphate (cGMP) almost certainly account for the bronchodilator action of theophylline, but PDE inhibition is unlikely to account for the nonbronchodilator effects of theophylline that are seen at lower concentrations. Inhibition of PDE should lead to synergistic interaction with β agonists through an increase in cAMP (Figure 36–5), but this has not been convincingly demonstrated in clinical studies. Several isoenzyme families of PDE have now been recognized and those important in smooth muscle relaxation include PDE3, PDE4, and PDE5. Theophylline is a weak inhibitor of all PDE isoenzymes.

• Adenosine receptor antagonism. Theophylline antagonizes adenosine receptors at therapeutic concentrations. Adenosine causes bronchoconstriction in airways from asthmatic patients by releasing histamine and leukotrienes. Adenosine antagonism is unlikely to account for the anti-inflammatory effects of theophylline but may be responsible for serious side effects, including cardiac arrhythmias and seizures through the antagonism of A₁ receptors. Of interest may be the adenosine A_2B receptor on mast cells, which is activated by adenosine in asthmatic patients (Wilson, 2008).

• Interleukin-10 release. IL-10 has a broad spectrum of anti-inflammatory effects, and there is evidence that its secretion is reduced in asthma. IL-10 release is increased by theophylline, and this effect may be mediated via inhibition of PDE activities, although this has not been seen at the low doses that are effective in asthma (Oliver et al., 2001).

• Effects on gene transcription. Theophylline prevents the translocation of the pro-inflammatory transcription factor NF-κB into the nucleus, potentially reducing the expression of inflammatory genes in asthma and COPD. Inhibition of NF-κB appears to be due to a protective effect against the degradation of the inhibitory protein I-κBα (Ichiyama et al., 2001). However, these effects are seen at high concentrations and may be mediated by inhibition of PDE.

• Effects on apoptosis. Prolonged survival of granulocytes due to a reduction in apoptosis may be important in perpetuating chronic inflammation in asthma (eosinophils) and COPD (neutrophils). Theophylline promotes apoptosis in eosinophils and neutrophils in vitro. This is associated with a reduction in the anti-apoptotic protein Bcl-2 (Chung et al., 2000). This effect is not mediated via PDE inhibition, but in neutrophils may be mediated by antagonism of adenosine A_2A receptors (Yasui et al., 2000). Theophylline also induces apoptosis in T lymphocytes, reducing their survival; this effect appears to be mediated via PDE inhibition (Ohta and Yamashita, 1999).

• Histone deacetylase activation. Recruitment of histone deacetylase-2 (HDAC2) by glucocorticoid receptors switches off inflammatory genes. Therapeutic concentrations of theophylline activate HDAC, thereby enhancing the anti-inflammatory effects of corticosteroids (Cosio et al., 2004; Ito et al., 2002). This mechanism is independent of PDE inhibition or adenosine receptor antagonism and appears to be mediated by inhibition of PI3-kinase-δ, which is activated by oxidative stress (Marwick et al., 2009). The anti-inflammatory effects of theophylline are inhibited by a HDAC inhibitor, trichostatin A. Low doses of theophylline increase HDAC activity in bronchial biopsies of asthmatic patients and correlate with the reduction in eosinophil numbers in the biopsy.

• Other effects. Several other effects of theophylline have been described, including an increase in circulating catecholamines, inhibition of calcium influx into inflammatory cells, inhibition of prostaglandin effects, and antagonism of tumor necrosis factor (TNF)α. These effects are generally seen only at concentrations of theophylline that are above the therapeutic range in asthma and are therefore unlikely to contribute to the anti-inflammatory actions of theophylline observed in asthmatics.

Nonbronchodilator Effects. Theophylline has clinical benefit in asthma and COPD at plasma concentrations <10 mg/L, sufficiently low that such effects are unlikely to be explained by theophylline's bronchodilator action. There is increasing evidence that theophylline has anti-inflammatory effects in asthma (Barnes, 2003). For example, chronic oral treatment with theophylline inhibits the late response to inhaled allergen and reduces infiltration of eosinophils and CD4⁺ lymphocytes into the airways after allergen challenge. In patients with mild asthma, low doses of theophylline (C_p ∼5 mg/L) reduce the numbers of eosinophils seen in bronchial biopsies, bronchoalveolar lavage, and induced sputum (Lim et al., 2001); conversely, in severe asthma, withdrawal of theophylline results in increased numbers of activated CD4⁺ cells and eosinophils in bronchial biopsies (Kidney et al., 1995). In patients with COPD, theophylline reduces the total number and proportion of neutrophils in induced sputum, the concentration of IL-8, and neutrophil chemotactic responses (Culpitt et al., 2002). Theophylline withdrawal in COPD patients results in worsening of disease. In vitro theophylline is able to increase responsiveness to corticosteroids and to reverse corticosteroid resistance in cells from COPD subjects (Ito et al., 2002).

Pharmacokinetics and Metabolism. There is a close relationship between improvement in airway function and serum theophylline concentration. Below 10 mg/L, therapeutic effects in terms of bronchodilation are small; above 20 mg/L additional benefits are outweighed by side effects, so that the therapeutic range was generally taken to be 10-20 mg/L (55-110 μM). Even within that range, toxicity can be observed, making theophylline a difficult drug to use. However, recent studies suggest that theophylline has anti-asthma effects other than bronchodilation below 10mg/L, so the therapeutic range is now taken as 5-15 mg/L. The dose of theophylline required to give these therapeutic concentrations varies among subjects, largely because of differences in clearance of the drug. In addition, there may be differences in bronchodilator response to theophylline; furthermore, with acute bronchoconstriction, higher concentrations may be required to produce bronchodilation. Theophylline is rapidly and completely absorbed, but there are large interindividual variations in clearance, due to differences in hepatic metabolism (Table 36–3). Theophylline is metabolized in the liver, mainly by CYP1A2, and a large number of factors are known to influence hepatic metabolism of theophylline (Zhang and Kaminsky, 1995):

• Increased clearance is seen in children (1-16 years of age), and in cigarette and marijuana smokers, due to induction of CYP1A2. Concurrent administration of phenytoin, phenobarbital, and rifampin increases activity of CYPs, especially CYP1A2, resulting in increased metabolic breakdown, so that higher doses may be required.

• Reduced clearance occurs in liver disease, pneumonia, and heart failure; doses must be reduced to half and plasma levels monitored carefully. Decreased clearance is also seen with coadministration of certain drugs, including erythromycin, certain quinolone antibiotics (ciprofloxacin, but not ofloxacin), allopurinol, cimetidine (but not ranitidine), fluvoxamine, the 5′-lipoxygenase inhibitor zileuton, and the leukotriene receptor antagonist zafirlukast, all of which interfere with CYP function. Thus, if a patient on maintenance theophylline requires a course of erythromycin, the dose of theophylline should be halved. Viral infections and vaccination may also reduce clearance.

Because of these variations in clearance, individualization of theophylline dosage is required and plasma concentrations should be measured 4 hours after the last dose with slow-release preparations when steady state has been achieved. There is no significant circadian variation in theophylline metabolism, although there may be delayed absorption at night related to supine posture.

Preparations and Routes of Administration. Intravenous aminophylline has been used for many years in the treatment of acute severe asthma. The recommended dose is 6 mg/kg given intravenously over 20-30 minutes, followed by a maintenance dose of 0.5 mg/kg per hour. If the patient is already taking theophylline, or there are any factors that decrease clearance, these doses should be halved and the plasma level checked more frequently. Nebulized β₂ agonists are now preferred over intravenous aminophylline for acute exacerbations of asthma and COPD.

Oral immediate-release theophylline tablets or elixir, which are rapidly absorbed, give wide fluctuations in plasma levels, and are not recommended. Several sustained-release preparations are now available that are absorbed at a constant rate and provide steady plasma concentrations over a 12- to 24-hour period. Both slow-release aminophylline and theophylline are available and are equally effective (although the ethylene diamine component of aminophylline has been implicated in allergic reactions). For continuous treatment twice daily therapy (∼8 mg/kg twice daily) is needed. Recently, once-daily preparations have gained popularity. For nocturnal asthma a single dose of slow-release theophylline at night is often effective. Once optimal doses have been determined, routine monitoring of plasma concentrations is usually not necessary, unless a change in clearance is suspected or evidence of toxicity emerges.

Clinical Use. In patients with acute asthma, intravenous aminophylline is less effective than nebulized β₂ agonists and should therefore be reserved for those patients who fail to respond to, or are intolerant of, β agonists. Theophylline should not be added routinely to nebulized β₂ agonists because it does not increase the bronchodilator response and may increase their side effects (Parameswaran et al., 2000). Indeed, the role of theophylline in the contemporary management of asthma and COPD has been questioned. However, there is good evidence that theophylline may provide an additional bronchodilator effect even when maximally effective doses of β₂ agonist have been given (Rivington et al., 1995). Thus, if adequate bronchodilation is not achieved by a β agonist alone, theophylline may be added to the maintenance therapy with benefit. Addition of low-dose theophylline to either a high or low dose of ICS in patients who are not adequately controlled provides better symptom control and lung function than doubling the dose of inhaled steroid (Evans et al., 1997; Lim et al., 2000; Ukena et al., 1997), although it is less effective as an add-on therapy than a LABA (Wilson et al., 2000). Theophylline may be useful in patients with nocturnal asthma because slow-release preparations are able to provide therapeutic concentrations overnight, although it is less effective than a LABA (Shah et al., 2003). Studies have also documented the steroid-sparing effects of theophylline. Although theophylline is less effective than a β₂ agonist and corticosteroids, a minority of asthmatic patients appears to derive unexpected benefit; even patients on oral steroids may show a deterioration in lung function when theophylline is withdrawn (Brenner et al., 1988; Kidney et al., 1995). Theophylline has been used as a controller in the management of mild persistent asthma, although it is usually found to be less effective than low doses of inhaled corticosteroids (Seddon et al., 2006). Although LABA are more effective as an add-on therapy, theophylline is considerably less expensive and may be the only affordable add-on treatment when the costs of medication are limiting (Bateman et al., 2008).

Theophylline is still used as a bronchodilator in COPD, but inhaled anticholinergics and β₂ agonists are preferred (Rabe et al., 2007). Theophylline tends to be added to these inhaled bronchodilators in more severe patients and has been shown to give additional clinical improvement when added to a long-acting β₂ agonist (Ram et al., 2002). Intravenous aminophylline is less effective in treating acute exacerbations than nebulized β₂ agonists and has a much higher incidence of adverse effects (Barr et al., 2003; Duffy et al., 2005). As in asthmatics, patients with severe COPD deteriorate when theophylline is withdrawn from their treatment regimen. A theoretical advantage of theophylline is that it its systemic administration may have effects on small airways, resulting in reduction of hyperinflation and thus a reduction in dyspnea.

Side Effects. Unwanted effects of theophylline are usually related to plasma concentration and tend to occur at C_p > 15 mg/L. However, many patients develop side effects even at low plasma concentrations. To some extent, side effects may be reduced by gradually increasing the dose until therapeutic concentrations are achieved. The most common side effects are headache, nausea, and vomiting (due to inhibition of PDE4), abdominal discomfort, and restlessness (Table 36–3). There may also be increased acid secretion (due to PDE inhibition) and diuresis (due to inhibition of adenosine A₁ receptors). Theophylline may lead to behavioral disturbance and learning difficulties in schoolchildren, although it is difficult to design adequate controls for such studies. At high concentrations, cardiac arrhythmias may occur as a consequence of inhibition of cardiac PDE3 inhibition and antagonism of cardiac A₁ receptors. At very high concentrations, seizures may occur due to central A₁ receptor antagonism. Use of low doses of theophylline, targeting plasma concentrations of 5-10 mg/L, largely avoids side effects and drug interactions and makes it less necessary to monitor plasma concentrations (unless checking for compliance).

In animals and man there is a small degree of resting bronchomotor tone that is probably due to tonic vagal nerve impulses that release acetylcholine in the vicinity of airway smooth muscle because it can be blocked by anticholinergic drugs. Acetylcholine may also be released from other airway cells, including epithelial cells (Wessler and Kirkpatrick, 2008). The synthesis of acetylcholine in epithelial cells is increased by inflammatory stimuli (such as TNF-α), which increase the expression of choline acetyltransferase, which could contribute to cholinergic effects in airway diseases. Muscarinic receptors are expressed in airway smooth muscle of small airways that do not appear to be innervated by cholinergic nerves; these receptors may be a mechanism of cholinergic narrowing in peripheral airways that could be relevant in COPD, responding to locally synthesized, non-neuronal ACh.

Myriad mechanical, chemical, and immunological stimuli elicit reflex bronchoconstriction via vagal pathways, and cholinergic pathways may play an important role in regulating acute bronchomotor responses in animals. These observations suggested that cholinergic mechanisms might underlie airway hyperresponsiveness and acute bronchoconstrictor responses in asthma, with the implication that anticholinergic drugs would be effective bronchodilators. These drugs may afford protection against acute challenge by sulfur dioxide, inert dusts, cold air, and emotional factors, but they are less effective against antigen challenge, exercise, and fog. This is not surprising because anticholinergic drugs will only inhibit reflex ACh-mediated bronchoconstriction and have no blocking effect on the direct effects of inflammatory mediators, such as histamine and leukotrienes, on bronchial smooth muscle. Furthermore, cholinergic antagonists probably have little or no effect on mast cells, microvascular leak, or the chronic inflammatory response.

Theoretically, anticholinergics may reduce airway mucus secretion and reduce mucus clearance, but this is generally not observed in clinical studies. Oxitropium bromide (not available in the U.S.) in high doses reduces mucus hypersecretion in patients with COPD with chronic bronchitis.

Clinical Use. In asthmatic patients, anticholinergic drugs are less effective as bronchodilators than β₂ agonists and offer less efficient protection against bronchial challenges. These drugs may be more effective in older patients with asthma in whom there is an element of fixed airway obstruction. Anticholinergics are currently used as an additional bronchodilator in asthmatic patients not controlled on a LABA. Nebulized anticholinergic drugs are effective in acute severe asthma but less effective than β₂ agonists. Nevertheless, in the acute and chronic treatment of asthma, anticholinergic drugs may have an additive effect with β₂ agonists and should therefore be considered when control of asthma is not adequate with nebulized β₂ agonists. A muscarinic antagonist should be considered when there are problems with theophylline or when inhaled β₂ agonists cause a troublesome tremor in elderly patients.

In COPD, anticholinergic drugs may be as effective as or even superior to β₂ agonists. Their relatively greater effect in COPD than in asthma may be explained by an inhibitory effect on vagal tone, which, although not necessarily increased in COPD, may be the only reversible element of airway obstruction and one that is exaggerated by geometric factors in the narrowed airways of COPD patients (Figure 36–7). Anticholinergic drugs reduce air trapping and improve exercise tolerance in COPD patients.

Therapeutic Choices. Ipratropium bromide (atrovent, others) is available as a pMDI and nebulized preparation. The onset of bronchodilation is relatively slow and is usually maximal 30-60 minutes after inhalation, but may persist for 6-8 hours. It is usually given by MDI three to four times daily on a regular basis, rather than intermittently for symptom relief, in view of its slow onset of action.

Oxitropium bromide (not available in the U.S.) is a quaternary anticholinergic bronchodilator that is similar to ipratropium bromide in terms of receptor blockade. It is available in higher doses by inhalation and may therefore have a more prolonged effect. Thus, it may be useful in some patients with nocturnal asthma.

Combination inhalers of an anticholinergic and β₂ agonist, such as ipratropium/albuterol (combivent, duoneb, others), are popular, particularly among patients with COPD. Several studies have demonstrated additive effects of these two drugs, thus providing an advantage over increasing the dose of β₂ agonist in patients who have side effects.

Tiotropium bromide is a long-acting anticholinergic drug that is suitable for once-daily dosing as a DPI (spiriva) or via a soft mist mini-nebulizer device (not available in the U.S.) (Barr et al., 2006). Tiotropium binds to all muscarinic receptor subtypes but dissociates very slowly from M₃ and M₁ receptors, giving it a degree of kinetic receptor selectivity for these receptors compared with M₂ receptors, from which it dissociates more rapidly. Thus, compared with ipratropium, tiotropium is less likely to antagonize M₂-mediated inhibition of ACh release (the resulting increase in ACh could counteract the blockade of M₃ receptor–mediated bronchoconstriction) (Chapter 9). It is an effective bronchodilator in patients with COPD and is more effective than ipratropium four times daily without any loss of efficacy over a 1-year treatment period. Over a 4-year period, tiotropium improves lung function and health status and reduces exacerbations and all-cause mortality, although there is no effect on disease progression (Tashkin et al., 2008). As a result, tiotropium is becoming the bronchodilator of choice for COPD patients.

Adverse Effects. Inhaled anticholinergic drugs are generally well tolerated. On stopping inhaled anticholinergics, a small rebound increase in airway responsiveness has been described, but the clinical relevance of this is uncertain. Systemic side effects after ipratropium bromide and tiotropium bromide are uncommon during normal clinical use because there is little systemic absorption. Because cholinergic agonists can stimulate mucus secretion, there has been concern that anticholinergics may reduce secretion and lead to more viscous mucus. However, ipratropium bromide, even in high doses, has no detectable effect on mucociliary clearance in either normal subjects or in patients with airway disease. A significant unwanted effect is the unpleasant bitter taste of inhaled ipratropium, which may contribute to poor compliance. Nebulized ipratropium bromide may precipitate glaucoma in elderly patients due to a direct effect of the nebulized drug on the eye. This may be prevented by nebulization with a mouthpiece rather than a face mask. Reports of paradoxical bronchoconstriction with ipratropium bromide, particularly when given by nebulizer, were largely explained as effects of the hypotonic nebulizer solution and by antibacterial additives, such as benzalkonium chloride and ethylenediaminetetraacetic acid (EDTA). This problem has not been described with tiotropium. Occasionally, bronchoconstriction may occur with ipratropium bromide given by MDI. It is possible that this is due to blockade of prejunctional M₂ receptors on airway cholinergic nerves that normally inhibit acetylcholine release. Tiotropium causes dryness of the mouth in 10-15% of patients, but this usually disappears during continued therapy. Urinary retention is occasionally seen in elderly patients.

Future Developments. Anticholinergics are the bronchodilators of choice in COPD and therefore have a growing market. Several long-acting muscarinic antagonists are now in clinical development, including glycopyrrolate and aclidinium (Cazzola, 2009; Hansel et al., 2005). Dual-action drugs that are both muscarinic antagonists and β₂ agonist (MABA) are also in clinical development (Ray and Alcaraz, 2009).

Magnesium Sulfate. Increasing evidence indicates that magnesium sulfate (MgSO₄) is useful as an additional bronchodilator in patients with acute severe asthma. Intravenous or nebulized MgSO₄ benefits adults and children with severe exacerbations (forced expiratory volume in 1 second [FEV₁] <30% of predicted value), giving improvement in lung function when added to nebulized β₂ agonist and a reduction in hospital admissions (Mohammed and Goodacre, 2007). The treatment is cheap and well tolerated, although the clinical benefit appears small. Side effects include flushing and nausea but are usually minor. Magnesium sulfate appears to act as a bronchodilator and may reduce cytosolic Ca²⁺ concentrations in airway smooth muscle cells. The concentration of magnesium is lower in serum and erythrocytes of asthmatic patients than in normal controls and correlates with airway hyperresponsiveness (Emelyanov et al., 1999), although the improvement in acute severe asthma after magnesium does not correlate with plasma concentrations. More studies are needed before intravenous and inhaled MgSO₄ are routinely recommended for the management of acute severe asthma. There are too few studies in acute exacerbations of COPD to make any firm recommendation (Skorodin et al., 1995).

K⁺ Channel Openers. K⁺ channels are involved in recovery of excitable cells after depolarization and are important in stabilization of cells. K⁺ channel openers such as cromakalim or levcromakalim (the levo-isomer of cromakalim) open ATP-dependent K⁺ channels in smooth muscle, leading to membrane hyperpolarization and relaxation of airway smooth muscle (Black et al., 1990). This suggests that K⁺ channel activators may be useful as bronchodilators (Pelaia et al., 2002). Clinical studies in asthma, however, have been disappointing, with no bronchodilation or protection against bronchoconstrictor challenges. The cardiovascular side effects of these drugs (postural hypotension, flushing) limit the oral dose; furthermore, inhaled formulations are problematic. New developments include K⁺ channel openers that open Ca²⁺-activated large conductance K⁺ channels (maxi-K channels) that are also opened by β₂ agonists; these drugs may be better tolerated. Maxi-K channel openers also inhibit mucus secretion and cough, and they may be of particular value in the treatment of COPD.

Atrial Natriuretic Peptides. Atrial natriuretic peptide (ANP) activates membrane-bound guanylyl cyclase and increases cellular cyclic GMP, leading to bronchodilation by mechanisms similar to those of NO on smooth muscle (Chapter 3). ANP and the related peptide urodilatin are bronchodilators in asthma and give effects comparable to those of β₂ agonists (Angus et al., 1993; Fluge et al., 1999). Because ANP and congeners work via a mechanism distinct from that of β₂ agonists, they may give additional bronchodilation that may be useful in acute severe asthma when β₂ receptor function might be impaired.

Vasoactive Intestinal Polypeptide Analogs. Vasoactive intestinal polypeptide (VIP) is a 28 amino acid peptide that binds to two GPCRs, VPAC₁ and VPAC₂, both of which couple primarily to G_s to stimulate the adenylyl cyclase-cAMP-PKA pathway leading to relaxation of smooth muscle. VIP is a potent dilator of human airway smooth muscle in vitro but is not effective in patients because it is rapidly metabolized (plasma t_1/2 ∼2 minutes); in addition, VIP causes vasodilator side effects. More stable analogs of VIP, such as Ro 25-1533, which selectively stimulates VIP receptors in airway smooth muscle (via the VPAC₂ receptor), have been synthesized. Inhaled Ro 25-1533 has a rapid bronchodilator effect in asthmatic patients but is not as prolonged as formoterol (Linden et al., 2003).

Other Inhibitors of Smooth Muscle Contraction. Agents that inhibit the contractile machinery of airway smooth muscle, including rho kinase inhibitors, inhibitors of myosin light chain kinase, and myosin inhibitors, are also in development. Because these agents also cause vasodilation, it will be necessary to administer them by inhalation.

Chemistry. The adrenal cortex secretes cortisol (hydrocortisone) and, by modification of its structure, it was possible to develop derivatives, such as prednisone, prednisolone, and dexamethasone, with enhanced corticosteroid effects but with reduced mineralocorticoid activity (Chapter 42). These derivatives with potent glucocorticoid actions were effective in asthma when given systemically but had no anti-asthmatic activity when given by inhalation. Further substitution in the 17α ester position resulted in steroids with high topical activity, such as beclomethasone dipropionate, triamcinolone, flunisolide, budesonide, and fluticasone propionate, which are potent in the skin (dermal blanching test) and were later found to have significant anti-asthma effects when given by inhalation (Figure 36–8).

Mechanism of Action. Corticosteroids enter target cells and bind to glucocorticoid receptors (GR) in the cytoplasm (Chapter 42). There is only one type of GR that binds corticosteroids and no evidence for the existence of subtypes that might mediate different aspects of corticosteroid action (Barnes, 2006a). The steroid-GR complex moves into the nucleus, where it binds to specific sequences on the upstream regulatory elements of certain target genes, resulting in increased (or rarely, decreased) transcription of the gene, with subsequent increased (or decreased) synthesis of the gene products. GR may also interact with protein transcription factors and coactivator molecules in the nucleus and thereby influence the synthesis of certain proteins independently of any direct interaction with DNA. The repression of transcription factors, such as activator protein-1 (AP-1) and NF-κB, is likely to account for many of the anti-inflammatory effects of steroids in asthma. In particular, corticosteroids reverse the activating effect of these pro-inflammatory transcription factors on histone acetylation by recruiting HDAC2 to inflammatory genes that have been activated through acetylation of associated histones (Figure 36–9). GRs are acetylated when corticosteroids are bound and bind to DNA in this acetylated state as dimers, whereas the acetylated GR has to be deacetylated by HDAC2 in order to interact with inflammatory genes and NF-κB (Ito et al., 2006).

There may be additional mechanisms that are also important in the anti-inflammatory actions of corticosteroids. Corticosteroids have potent inhibitory effects on MAP kinase signaling pathways through the induction of MKP-1, which may inhibit the expression of multiple inflammatory genes (Clark, 2003).

Anti-Inflammatory Effects in Asthma. The mechanisms of action of corticosteroids in asthma are still poorly understood, but their efficacy is most likely related to their anti-inflammatory properties. Corticosteroids have widespread effects on gene transcription, increasing the transcription of several anti-inflammatory genes and suppressing transcription of many inflammatory genes. Steroids have inhibitory effects on many inflammatory and structural cells that are activated in asthma and prevent the recruitment of inflammatory cells into the airways (Figure 36–10). Studies of bronchial biopsies in asthma have demonstrated a reduction in the number and activation of inflammatory cells in the epithelium and submucosa after regular ICS, together with a healing of the damaged epithelium. Indeed, in patients with mild asthma the inflammation may be completely resolved after inhaled steroids.

Steroids potently inhibit the formation of cytokines (e.g., IL-1, IL-3, IL-4, IL-5, IL-9, IL-13, TNF-α, and granulocyte-macrophage colony-stimulating factor, GM-CSF) that are secreted in asthma by T-lymphocytes, macrophages, and mast cells. Corticosteroids also decrease eosinophil survival by inducing apoptosis. Corticosteroids inhibit the expression of multiple inflammatory genes in airway epithelial cells, probably the most important action of ICS in suppressing asthmatic inflammation. Corticosteroids also prevent and reverse the increase in vascular permeability due to inflammatory mediators in animal studies and may therefore lead to resolution of airway edema. Steroids have a direct inhibitory effect on mucus glycoprotein secretion from airway submucosal glands, as well as indirect inhibitory effects by down-regulation of inflammatory stimuli that stimulate mucus secretion.

Corticosteroids have no direct effect on contractile responses of airway smooth muscle; improvement in lung function after ICS is presumably due to an effect on the chronic airway inflammation and airway hyperresponsiveness. A single dose of ICS has no effect on the early response to allergen (reflecting their lack of effect on mast cell mediator release) but inhibits the late response (which may be due to an effect on macrophages, eosinophils, and airway wall edema) and also inhibits the increase in airway hyperresponsiveness.

ICS have rapid anti-inflammatory effects, reducing airway hyperresponsiveness and inflammatory mediator concentrations in sputum within a few hours (Erin et al., 2008). However, it may take several weeks or months to achieve maximal effects on airway hyperresponsiveness, presumably reflecting the slow healing of the damaged inflamed airway. It is important to recognize that corticosteroids suppress inflammation in the airways but do not cure the underlying disease. When steroids are withdrawn there is a recurrence of the same degree of airway hyperresponsiveness, although in patients with mild asthma it may take several months to return.

Effect on β₂ Adrenergic Responsiveness. Corticosteroids increase β adrenergic responsiveness, but whether this is relevant to their effect in asthma is uncertain. Steroids potentiate the effects of β agonists on bronchial smooth muscle and prevent and reverse β receptor desensitization in airways in vitro and in vivo (Barnes, 2002; Giembycz et al., 2008). At a molecular level, corticosteroids increase the transcription of the β₂ receptor gene in human lung in vitro and in the respiratory mucosa in vivo and also increase the stability of its messenger RNA. They also prevent or reverse uncoupling of β₂ receptors to G_s. In animal systems, corticosteroids prevent down-regulation of β₂ receptors.

β₂ Agonists also enhance the action of GR, resulting in increased nuclear translocation of liganded GR receptors and enhancing the binding of GR to DNA. This effect has been demonstrated in sputum macrophages of asthmatic patients after an ICS and inhaled LABA (Usmani et al., 2005). This suggests that β₂ agonists and corticosteroids enhance each other's beneficial effects in asthma therapy.

Pharmacokinetics. The pharmacokinetics of oral corticosteroids are described in Chapter 42. The pharmacokinetics of inhaled corticosteroids are important in relation to systemic effects (Barnes et al., 1998b). The fraction of steroid that is inhaled into the lungs acts locally on the airway mucosa but may be absorbed from the airway and alveolar surface. Thus, a portion of an inhaled dose reaches the systemic circulation. Furthermore, the fraction of inhaled steroid that is deposited in the oropharynx is swallowed and absorbed from the gut. The absorbed fraction may be metabolized in the liver (first-pass metabolism) before reaching the systemic circulation (Figure 36–3). The use of a spacer chamber reduces oropharyngeal deposition and therefore reduces systemic absorption of ICS, although this effect is minimal in corticosteroids with a high first-pass metabolism. Mouth rinsing and discarding the rinse have a similar effect, and this procedure should be used with high-dose dry powder steroid inhalers with which spacer chambers cannot be used.

Beclomethasone dipropionate and ciclesonide are prodrugs that release the active corticosteroid after the ester group is cleaved by esterases in the lung. Ciclesonide is available as a MDI (alvesco) for asthma and as a nasal spray for allergic rhinitis (omnaris). Budesonide and fluticasone propionate have a greater first-pass metabolism than beclomethasone dipropionate and are therefore less likely to produce systemic effects at high inhaled doses.

Routes of Administration and Dosing

Inhaled Corticosteroids in Asthma. Inhaled corticosteroids are recommended as first-line therapy for all patients with persistent asthma. They should be started in any patient who needs to use a β₂ agonist inhaler for symptom control more than twice weekly. They are effective in mild, moderate, and severe asthma and in children as well as adults (Barnes et al., 1998b). Although it was recommended that ICS be initiated at a relatively high dose and then the dose reduced once control was achieved, there is no evidence that this is more effective than starting with the maintenance dose. Dose-response studies for ICS are relatively flat, with most of the benefit derived from doses <400 μg beclomethasone dipropionate or equivalent (Adams et al., 2008). However, some patients (with relative corticosteroid resistance) may benefit from higher doses (up to 2000 μg/day).

For most patients, ICS should be used twice daily, a regimen that improves compliance once control of asthma has been achieved (which may require four-times daily dosing initially or a course of oral steroids if symptoms are severe). Administration once daily of some steroids (e.g., budesonide, mometasone, and ciclesonide) is effective when doses ≤400 μg are needed. If a dose 800 μg daily via pMDI is used, a spacer device should be employed to reduce the risk of oropharyngeal side effects. ICS may be used in children in the same way as in adults; at doses ≤400 μg/day there is no evidence of significant growth suppression (Pedersen, 2001). The dose of ICS should be the minimal dose that controls asthma; once control is achieved, the dose should be slowly reduced (Hawkins et al., 2003). Nebulized corticosteroids (e.g., budesonide) are useful in the treatment of small children who are not able to use other inhaler devices.

Inhaled Corticosteroids in Chronic Obstructive Pulmonary Disease. Patients with COPD occasionally respond to steroids, and these patients are likely to have concomitant asthma. Corticosteroids have no objective short-term benefit on airway function in patients with true COPD, although these agents often produce subjective benefit because of their euphoric effect. Corticosteroids do not appear to have any significant anti-inflammatory effect in COPD; there appears to be an active resistance mechanism, which may be explained by impaired activity of HDAC2 as a result of oxidative stress (Barnes, 2009). ICS have no effect on the progression of COPD, even when given to patients with presymptomatic disease; additionally, ICS have no effect on mortality (Calverley et al., 2007; Yang et al., 2007). ICS reduce the number of exacerbations in patients with severe COPD (FEV₁ <50% predicted) who have frequent exacerbations and are recommended in these patients, although there is debate about whether these effects are due to inappropriate analysis of the data (Suissa et al., 2008). Oral corticosteroids are used to treat acute exacerbations of COPD, but the effect is very small (Niewoehner et al., 1999).

Patients with cystic fibrosis, which involves inflammation of the airways, are also resistant to high doses of ICS.

Systemic Steroids. Intravenous steroids are indicated in acute asthma if lung function is <30% predicted and in patients who show no significant improvement with nebulized β₂ agonist. Hydrocortisone is the steroid of choice because it has the most rapid onset (5-6 hours after administration), compared with 8 hours with prednisolone. The required dose is uncertain; it is common to give hydrocortisone 4 mg/kg initially, followed by a maintenance dose of 3 mg/kg every 6 hours. Methylprednisolone is also available for intravenous use, but there is no evidence that the high doses previously used (1 g) are more effective. Intravenous therapy is usually given until a satisfactory response is obtained, and then oral prednisolone may be substituted. Oral prednisolone (40-60 mg) has a similar effect to intravenous hydrocortisone and is easier to administer. A high dose of inhaled fluticasone propionate (2000 μg daily) is as effective as a course of oral prednisolone in controlling acute exacerbations of asthma in a family practice setting and in children in an emergency department setting, although this route of delivery is more expensive (Levy et al., 1996; Manjra et al., 2000).

Prednisolone and prednisone are the most commonly used oral steroids. Clinical improvement with oral steroids may take several days; the maximal beneficial effect is usually achieved with 30-40 mg prednisone daily, although a few patients may need 60-80 mg daily to achieve control of symptoms. The usual maintenance dose is ∼10-15 mg/day. Short courses of oral steroids (30-40 mg prednisolone daily for 1-2 weeks) are indicated for exacerbations of asthma; the dose may be tapered over 1 week after the exacerbation is resolved (the taper is not strictly necessary after a short course of therapy, but patients find it reassuring). Oral steroids are usually given as a single dose in the morning because this coincides with the normal diurnal increase in plasma cortisol and produces less adrenal suppression than if given in divided doses or at night. Alternate-day treatment has the advantage of less adrenal suppression, although in many patients control of asthma is not optimal on this regimen.

Adverse Effects. Corticosteroids inhibit ACTH and cortisol secretion by a negative feedback effect on the pituitary gland (Chapter 42). Hypothalamic-pituitary-adrenal (HPA) axis suppression depends on dose and usually only occurs with doses of prednisone >7.5-10 mg/day. Significant suppression after short courses of corticosteroid therapy is not usually a problem, but prolonged suppression may occur after several months or years. Steroid doses after prolonged oral therapy must be reduced slowly. Symptoms of "steroid withdrawal syndrome" include lassitude, musculoskeletal pains, and, occasionally, fever. HPA suppression with inhaled steroids is usually seen only when the daily inhaled dose exceeds 2000 μg beclomethasone dipropionate or its equivalent daily.

Side effects of long-term oral corticosteroid therapy include fluid retention, increased appetite, weight gain, osteoporosis, capillary fragility, hypertension, peptic ulceration, diabetes, cataracts, and psychosis. Their frequency tends to increase with age. Very occasionally adverse reactions (such as anaphylaxis) to intravenous hydrocortisone have been described, particularly in aspirin-sensitive asthmatic patients.

The incidence of systemic side effects after ICS is an important consideration, particularly in children (Barnes et al., 1998b) (Table 36–4). Initial studies suggested that adrenal suppression occurred only with inhaled doses >1500-2000 μg/day. More sensitive measurements of systemic effects include indices of bone metabolism, such as serum osteocalcin and urinary pyridinium cross-links, and in children, knemometry, which may be increased with inhaled doses as low as 400 μg/day beclomethasone dipropionate in some patients. The clinical relevance of these measurements is not yet clear, however. Nevertheless, it is important to reduce the likelihood of systemic effects by using the lowest dose of inhaled steroid needed to control the asthma, and by use of a large-volume spacer to reduce oropharyngeal deposition.

Several systemic effects of inhaled steroids have been described and include dermal thinning and skin capillary fragility (relatively common in elderly patients after high-dose inhaled steroids). Other side effects, such as cataract formation and osteoporosis, are reported but often in patients who are also receiving courses of oral steroids. There has been particular concern about the use of inhaled steroids in children because of growth suppression (Pedersen, 2001). Most studies have been reassuring that doses ≤400 μg/day have not been associated with impaired growth; on the contrary, there may even be a growth spurt as asthma is better controlled. There is some evidence that use of high-dose ICS is associated with cataract and glaucoma, but it is difficult to dissociate the effects of ICS from the effects of courses of oral steroids that these patients usually require.

ICS may have local side effects due to the deposition of inhaled steroid in the oropharynx. The most common problem is hoarseness and weakness of the voice (dysphonia) due to atrophy of the vocal cords following laryngeal deposition of steroid; it may occur in up to 40% of patients and is noticed particularly by patients who need to use their voices during their work (lecturers, teachers, and singers). Throat irritation and coughing after inhalation are common with MDI and appear to be due to additives because these problems are not usually seen if the patient switches to a DPI. There is no evidence for atrophy of the lining of the airway. Oropharyngeal candidiasis occurs in ∼5% of patients. There is no evidence for increased lung infections, including tuberculosis, in patients with asthma. Growing evidence suggests that high doses of ICS increase the risk of pneumonia in patients with COPD (Singh et al., 2009); although this is reported with high doses of fluticasone propionate, a similar increase in pneumonia has not been found with budesonide, which may be explained by its lower systemic effects (Sin et al., 2009).

It may be difficult to extrapolate systemic side effects of corticosteroids using data from normal subjects. In asthmatic patients, systemic absorption form the lung is reduced, presumably because of reduced and more central deposition of the inhaled drug in more severe patients (Brutsche et al., 2000; Harrison et al., 2001); most of the inhaled drug deposits in larger airways, thereby limiting effects in the smaller airways where inflammation is also found, especially in patients with severe asthma. Corticosteroid MDIs with HFA propellants produce smaller aerosol particles and may have a more peripheral deposition, making them useful in treating patients with more severe asthma.

Cromones

Cromolyn sodium (sodium cromoglycate) is a derivative of khellin, an Egyptian herbal remedy and was found to protect against allergen challenge without any bronchodilator effect. A structurally related drug, nedocromil sodium, which has a similar pharmacological profile to cromolyn, was subsequently developed. Although cromolyn was popular in the past because of its good safety profile, its use has sharply declined with the more widespread use of the more effective ICS, particularly in children. Neither cromolyn nor nedocromil is available in the U.S. for respiratory indications. For further information, consult earlier editions of this text.

Antihistamines. Histamine mimics many of the features of asthma and is released from mast cells in acute asthmatic responses, suggesting that antihistamines may be useful in asthma therapy. There is little evidence that histamine H₁ receptor antagonists provide any useful clinical benefit, as demonstrated by a meta-analysis (van Ganse et al., 1997). Newer antihistamines, including cetirizine and azelastine, have some beneficial effects, but this may be unrelated to H₁ receptor antagonism. Antihistamines are not recommended in the routine management of asthma.

Anti-leukotrienes. There is considerable evidence that cysteinyl-leukotrienes (LTs) are produced in asthma and that they have potent effects on airway function, inducing bronchoconstriction, airway hyperresponsiveness, plasma exudation, mucus secretion, and eosinophilic inflammation (Figure 36–11; also see Chapter 33 and Peters-Golden and Henderson, 2007). These data suggested that blocking the leukotriene pathways with leukotriene modifiers could be useful in the treatment of asthma, leading to the development of 5′-lipoxygenase (5-LO) enzyme inhibitors (of which zileuton [zyflo] is the only drug marketed) and several antagonists of the cys-LT₁ receptor, including montelukast (singulair), zafirlukast (accolate), and pranlukast (not available in the U.S.).

Clinical Studies. Anti-leukotrienes have been intensively investigated in clinical studies (Calhoun, 2001). Leukotriene antagonists inhibit the bronchoconstrictor effect of inhaled LTD₄ in normal and asthmatic volunteers, and they also inhibit bronchoconstriction induced by a variety of challenges, including allergen, exercise, and cold air, by approximately 50%. With aspirin challenge, in aspirin-sensitive asthmatic patients, there is almost complete inhibition of the response (Szczeklik and Stevenson, 2003). Similar results have been obtained with the 5-LO inhibitor zileuton. This suggests there may be no advantage in blocking LTB₄ in addition to cysteinyl-LTs, in treating asthma. These drugs are active by oral administration, possibly an important advantage in chronic treatment.

In patients with mild to moderate asthma, anti-leukotrienes cause a significant improvement in lung function and asthma symptoms, with a reduction in the use of rescue inhaled β₂ agonists. Several studies show evidence for a bronchodilator effect, with an improvement in baseline lung function, suggesting that leukotrienes are contributing to the baseline bronchoconstriction in asthma, although this varies among patients. However, anti-leukotrienes are considerably less effective than inhaled corticosteroids in the treatment of mild asthma and cannot be considered the treatment of first choice (Ducharme, 2003). Anti-leukotrienes are indicated as an add-on therapy in patients who are not well controlled on ICS. The added benefit is small, equivalent to doubling the dose of ICS, and less effective than adding a LABA (Ducharme et al., 2006).

In patients with severe asthma who are not controlled on high doses of ICS and LABA, anti-leukotrienes do not appear to provide any additional benefit (Robinson et al., 2001). Theoretically, anti-leukotrienes should be of particular value in patients with aspirin-sensitive asthma because they block the airway response to aspirin challenge; however, their benefit is no greater here than in other types of asthma (Dahlen et al., 2002). Anti-leukotrienes are effective in preventing exercise-induced asthma, with efficacy similar to that of LABA (Coreno et al., 2000). Anti-leukotrienes also have a weak effect in rhinitis that may be additive with the effects of an antihistamine (Nathan, 2003).

Studies have demonstrated weak anti-inflammatory effects of anti-leukotrienes in reducing eosinophils in sputum and biopsies (Minoguchi et al., 2002), but this is much less marked than with ICS and there is no additional anti-inflammatory effect when added to an ICS (O'Sullivan et al., 2003). Anti-leukotrienes therefore appear to act mainly as anti-bronchoconstrictor drugs, and they are clearly less broadly effective than β₂ agonists because they antagonize only one of several bronchoconstrictor mediators.

In COPD, Cys-LTs are not elevated in exhaled breath condensate, and cys-LT1 receptor antagonists have no role in the therapy of COPD (Barnes, 2004). By contrast, LTB₄, a potent neutrophil chemoattractant, is elevated in COPD, indicating that 5-LO inhibitors that inhibit LTB₄ synthesis may have some potential benefit by reducing neutrophil inflammation. However, a pilot study failed to indicate any clear benefit of a 5-LO inhibitor in COPD patients (Gompertz and Stockley, 2002).

Adverse Effects. Zileuton, zafirlukast, and montelukast are all associated with rare cases of hepatic dysfunction; thus liver-associated enzymes should be monitored. Several cases of Churg-Strauss syndrome have been associated with the use of zafirlukast and montelukast. Churg-Strauss syndrome is a very rare vasculitis that may affect the heart, peripheral nerves, and kidney and is associated with increased circulating eosinophils and asthma. It is uncertain whether the reported cases are due to a reduction in oral or inhaled corticosteroid dose or are a direct effect of the drug. Cases of Churg-Strauss syndrome have been described in patients on anti-leukotrienes who were not on concomitant corticosteroid therapy, suggesting there is a causal link (Nathani et al., 2008).

Future Developments. One of the major advantages of anti-leukotrienes is their effectiveness in tablet form. This may increase compliance with chronic therapy and makes treatment of children easier. Montelukast is effective as a once-daily preparation (10 mg in adults, 5 mg in children). In addition, oral administration may treat concomitant allergic rhinitis. However, the currently available clinical studies indicate a relatively modest effect on lung function and symptom control, which is less for every clinical parameter measured than with ICS. This is not surprising: There are many mediators besides cys-LTs involved in the pathophysiology of asthma, and it is unlikely that antagonism of a single mediator would be as effective as steroids, which inhibit multiple aspects of the inflammatory process in asthma. Similarly, if anti-leukotrienes are functioning in asthma as bronchodilators and anti-constrictors, it is unlikely that they will be as effective as a β₂ agonist, which will counteract bronchoconstriction regardless of the spasmogen. It is likely that anti-leukotrienes will be used less in the future because combination inhalers are the mainstay of asthma therapy.

An interesting feature of the clinical studies is that some patients appear to show better responses than others, suggesting that leukotrienes may play a more important role in some patients. The variability in response to anti-leukotrienes may reflect differences in production of or responses to leukotrienes in different patients, and this in turn may be related to polymorphisms of 5-LO, LTC₄ synthase, or cys-LT₁ receptors that are involved in the synthesis of leukotrienes (Tantisira and Drazen, 2009).

It is unlikely that further advances can be made in cys-LT₁-receptor antagonists if montelukast, a once-daily drug, is giving maximal receptor blockade. Cys-LT₂ receptors may be important in vascular and airway smooth muscle proliferative effects of cys-LT and are not inhibited by current cys-LT₁-receptor antagonists (Back, 2002). The role of the cys-LT₂ receptor in asthma is unknown, as is the value of its blockade. 5-LO inhibitors have some potential in COPD, cystic fibrosis, and possible severe asthma as LTB₄ may contribute to neutrophil chemoattraction in the lungs of these conditions.

Clinical Use. Omalizumab is used for the treatment of patients with severe asthma. The antibody is administered by subcutaneous injection every 2-4 weeks, and the dose is determined by the titer of circulating IgE. Omalizumab reduces the requirement for oral and inhaled corticosteroids and markedly reduces asthma exacerbations. It is also beneficial in treating allergic rhinitis. Because of its very high cost, this treatment is generally used only in patients with very severe asthma who are poorly controlled even on oral corticosteroids and in patients with very severe concomitant allergic rhinitis (Walker et al., 2006). It may not be useful in concomitant atopic dermatitis due to the high levels of circulating IgE that cannot be neutralized. It may also be of value in protecting against anaphylaxis during specific immunotherapy. The major side effect of omalizumab is an anaphylactic response, which is uncommon (<0.1%) (Cox et al., 2007).

The cellular mechanisms of specific immunotherapy are of interest. Specific immunotherapy induces the secretion of the anti-inflammatory cytokine IL-10 from regulatory helper T-lymphocytes (T_reg) and this blocks co-stimulatory signal transduction in T cells (via CD28) so that they are unable to react to allergens presented by antigen-presenting cells (Larche et al., 2006). Applying an understanding of the cellular processes involved might lead to safer and more effective approaches in the future. More specific immunotherapies may be developed with cloned allergen epitopes, T-cell peptide fragments of allergens, CpG oligonucleotides, and vaccines of conjugates of allergen and toll-like receptor (TLR)-9 to stimulate T_H1 immunity and suppress T_H2 immunity (Broide, 2009).

Novel Mediator Antagonists. Blocking the receptors or synthesis of inflammatory mediators is a logical approach to the development of new treatments for asthma and COPD. However, in both diseases many different mediators are involved, and therefore blocking a single mediator is unlikely to be very effective, unless it plays a unique and key role in the disease process (Barnes, 2004; Barnes et al., 1998a). Several specific mediator antagonists have been found to be ineffective in asthma, including antagonists/inhibitors of thromboxane, platelet-activating factor, bradykinin, and tachykinins. However, these blockers have often not been tested in COPD, in which different mediators are involved. A number of other approaches are under study, as noted next.

CRTh2 Antagonists. Chemotactic factor for T_H2 cells has been identified as prostaglandin D₂, which acts on a DP₂ receptor (Chapter 33). Several CRTh2 antagonists are now in development for asthma with promising initial results (Pettipher et al., 2007).

Endothelin Antagonists. Endothelin has been implicated in some of the structural changes that occur in asthma and COPD. Endothelin antagonists are approved for the treatment of pulmonary hypertension and might be useful in treating the structural changes that occur in asthma and COPD, but so far they have not been tested.

Antioxidants. Oxidative stress is important in severe asthma and COPD and may contribute to corticosteroid resistance. Existing antioxidants include vitamins C and E and N-acetyl-cysteine. These drugs have weak effects, but more potent antioxidants are in development (Kirkham and Rahman, 2006).

Inducible Nitric Oxide Synthase Inhibitors. NO production is increased in asthma and COPD as a result of increased inducible NO synthase (iNOS) expression in the airways (Chapter 3). NO and oxidative stress generate peroxynitrite anion (O=N-O-O⁻), which may nitrate proteins, leading to altered cell function. Several selective iNOS inhibitors are now in development (Hobbs et al., 1999), and one of these, L-N⁶-(1-imminoethyl)lysine (L-NIL), gives a profound and long-lasting reduction in the concentrations of NO in exhaled breath (Hansel et al., 2003). An iNOS inhibitor was found to be ineffective in asthma, however (Singh et al., 2007).

Cytokine Modifiers. Cytokines play a critical role in perpetuating and amplifying the inflammation in asthma and COPD, suggesting that anti-cytokines may be beneficial as therapies (Barnes, 2008a). Although most attention has focused on inhibition of cytokines, some cytokines are anti-inflammatory and may have therapeutic potential.

IL-5 plays a pivotal role in eosinophilic inflammation and is also involved in eosinophil survival and priming. It is an attractive target in asthma because it is essential for eosinophilic inflammation, there do not appear to be any other cytokines with a similar role, and lack of IL-5 in gene knockout mice does not have any deleterious effect. Anti-IL-5 antibodies inhibit eosinophilic inflammation and airway hyperresponsiveness in animal models of asthma, and they markedly reduce circulating and airway eosinophils in asthma patients but have no effect on airway hyperresponsiveness or clinical features of asthma (Flood-Page et al., 2007; Leckie et al., 2000). In carefully selected patients with severe asthma and persistent eosinophilia despite high doses of corticosteroids, there is a significant reduction in exacerbations (Haldar et al., 2009). Blocking IL-4 and IL-13, which determine IgE synthesis (Figure 36–12), has so far proved to be ineffective in clinical studies.

TNF-α may play a key role in amplifying airway inflammation, through the activation of NF-κB, AP-1, and other transcription factors. TNF-α production is increased in asthma and COPD, and it may be associated with the cachexia and weight loss that occur in some patients with severe COPD. In COPD patients and in patients with severe asthma, anti-TNF-α blocking antibodies have been ineffective, at the expense of increasing infections and malignancies (Rennard et al., 2007; Wenzel et al., 2009).

Chemokine Receptor Antagonists. Many chemokines are involved in asthma and COPD and play a key role in recruitment of inflammatory cells, such as eosinophils, neutrophils, macrophages, and lymphocytes into the lungs. Chemokine receptors are attractive targets because they are hepta-spanning membrane proteins; small molecule inhibitors are now in development (Donnelly and Barnes, 2006; Viola and Luster, 2008). In asthma, CCR3 antagonists, which should block eosinophil recruitment into the airways, are the most favored target, but several small molecule CCR3 antagonists have failed in development because of toxicity. In COPD, CXCR2 antagonists, which prevent neutrophil and monocyte chemotaxis due to CXC chemokines such as CXCL1 and CXCL8, have been effective in animal models of COPD and in neutrophilic inflammation in normal subjects, and are now in clinical trials in COPD patients.

Protease Inhibitors

Several proteolytic enzymes are involved in the chronic inflammation of airway diseases. Mast cell tryptase has several effects on airways, including increasing responsiveness of airway smooth muscle to constrictors, increasing plasma exudation, potentiating eosinophil recruitment, and stimulating fibroblast proliferation. Some of these effects are mediated by activation of the proteinase-activated receptor PAR2. Tryptase inhibitors have so far proved to be disappointing in clinical studies.

Proteases are involved in the degradation of connective tissue in COPD, particularly enzymes that break down elastin fibers, such as neutrophil elastase and matrix metalloproteinases (MMP), which are involved in emphysema. Neutrophil elastase inhibitors have been difficult to develop, and there are no positive clinical studies in COPD patients. MMP9 appears to be the predominant elastolytic enzyme in emphysema, and several selective inhibitors are now in development.

New Anti-Inflammatory Drugs

Phosphodiesterase Inhibitors. The preservation and elevation of cellular cyclic AMP in inflammatory cells often reduces cell activation and release of inflammatory mediators. PDE4 is the predominant PDE isoform in inflammatory cells, including mast cells, eosinophils, neutrophils, T lymphocytes, macrophages, and structural cells such as sensory nerves and epithelial cells (Houslay et al., 2005), suggesting that PDE4 inhibitors could be useful as an anti-inflammatory treatment in both asthma and COPD. In animal models of asthma, PDE4 inhibitors reduce eosinophil infiltration and responses to allergen, whereas in COPD they are effective against smoke-induced inflammation and emphysema. Several PDE4 inhibitors have been tested clinically, but with disappointing results; their usefulness has been limited by side effects, particularly nausea, vomiting, headaches, and diarrhea. In COPD, an oral PDE4 inhibitor, roflumilast, has some effect in reducing exacerbations and improving lung function, but side effects also limit its efficacy (Fabbri et al., 2009). Of the four subfamilies of PDE4, PDE4D is the major form whose inhibition is associated vomiting; inhibition of PDE4B is important for anti-inflammatory effects. Thus, selective PDE4B inhibitors may have a greater therapeutic index. Inhaled PDE4 inhibitors to reduce systemic absorption and adverse responses have proved to be ineffective. PDE5 inhibitors are vasodilators that are used in the treatment of pulmonary hypertension.

NF-κB Inhibitors. NF-κB plays an important role in the orchestration of chronic inflammation (Figure 36–9); many of the inflammatory genes that are expressed in asthma and COPD are regulated by this transcription factor (Barnes and Karin, 1997). This has prompted a search for specific blockers of these transcription factors. NF-κB is naturally inhibited by the inhibitory protein IκB, which is degraded after activation by specific kinases. Small molecule inhibitors of the IB kinase IKK2 (or IKKβ) are in clinical development (Karin et al., 2004; Kishore et al., 2003). These drugs may be of particular value in COPD, where corticosteroids are largely ineffective. However, there are concerns that inhibition of NF-κB may cause side effects such as increased susceptibility to infections, which has been observed in gene disruption studies when components of NF-κB are inhibited.

Mitogen-Activated Protein Kinase Inhibitors. There are three major MAP kinase pathways, and there is increasing recognition that these pathways are involved in chronic inflammation (Delhase et al., 2000). There has been particular interest in the p38 MAP kinase pathway that is blocked by a novel class of drugs, such as SB203580 and RWJ67657. These drugs inhibit the synthesis of many inflammatory cytokines, chemokines, and inflammatory enzymes (Cuenda and Rousseau, 2007). The p38 MAPK inhibitors are in development for the treatment of asthma (they inhibit T_H2 cytokine synthesis) and for COPD (they inhibit neutrophilic inflammation and signaling of inflammatory cytokines and chemokines). However, clinical studies have revealed marked adverse effects and toxicities; inhaled delivery is being explored.

Mucoregulators

Many pharmacological agents may influence the secretion of mucus in the airways, but there are few drugs that have demonstrably useful clinical effects on mucus hypersecretion (Rogers and Barnes, 2006). Mucus hypersecretion occurs in chronic bronchitis, COPD, cystic fibrosis, and asthma. In chronic bronchitis, mucus hypersecretion is related to chronic irritation by cigarette smoke and may involve neural mechanisms and the activation of neutrophils to release enzymes such as neutrophil elastase and proteinase-3 that have powerful stimulatory effects on mucus secretion. Mast cell–derived chymase is also a potent mucus secretagogue. This suggests that several classes of drugs may be developed to control mucus hypersecretion.

A major problem in assessing the effects of drugs on mucus hypersecretion is the difficulty in accurately quantifying airway mucus production, quality, and clearance. Several current drugs for airway disease might affect mucus production. Systemic anticholinergic drugs appear to reduce mucociliary clearance, but this is not observed with either ipratropium bromide or tiotropium bromide, presumably reflecting their poor absorption from the respiratory tract. β₂ Agonists increase mucus production and mucociliary clearance and have been shown to increase ciliary beat frequency in vitro. Because inflammation leads to mucus hypersecretion, anti-inflammatory treatments should reduce mucus hypersecretion; ICS are very effective in reducing increased mucus production in asthma.

Sensory nerves and neuropeptides are important in the secretory activities of the submucosal gland (which predominates in proximal airways) and goblet cell (more notable in peripheral airways). Opioids and K⁺ channel openers inhibit mucus secretion mediated via sensory neuropeptide release; peripherally acting opioids may be developed to control mucus hypersecretion due to irritants in the future (Rogers, 2002).

Mucolytics

Several agents can reduce the viscosity of sputum in vitro. One group are derivatives of cysteine that reduce the disulfide bridges that bind glycoproteins to other proteins such as albumin and secretory IgA. These drugs also act as antioxidants and may therefore reduce airway inflammation. Only N-acetylcysteine (mucomyst, others) is available in the U.S.; carbocysteine, methylcysteine, erdosteine, and bromhexine are available elsewhere. Orally administered, these agents are relatively well tolerated, but clinical studies in chronic bronchitis, asthma, and bronchiectasis have been disappointing. A systematic review of several small studies showed a small benefit in terms of reducing exacerbations, with most of the benefit from N-acetylcysteine; whether this relates to the mucolytic activity of N-acetylcysteine or to its action as an antioxidant is unclear (Poole and Black, 2001). A large controlled study of oral N-acetylcysteine in COPD patients showed no effect in disease progress or in preventing exacerbations, although there was some benefit in the patients not treated with inhaled corticosteroids (Decramer et al., 2005), as confirmed in a subsequent study of carbocysteine in COPD patients not treated with other medications (Zheng et al., 2008). N-acetylcysteine is not currently recommended for COPD management.

DNAse (dornase alfa, pulmozyme) reduces mucus viscosity in sputum of patients with cystic fibrosis and is indicated if there is significant symptomatic and lung function improvement after a trial of therapy (Henke and Ratjen, 2007). There is no evidence that dornase alfa is effective in COPD or asthma, however.

The epidermal growth factor receptor (EGFR) plays a critical role in airway mucus secretion from goblet cells and submucosal glands and appears to mediate the mucus secretory response to several secretagogues, including oxidative stress, cigarette smoke, inflammatory cytokines, and activated TLRs (Burgel and Nadel, 2004). Small molecule inhibitors of EGFR kinase, such as gefitinib and erlotinib, have been developed for use as anticancer therapies and are currently being assessed as treatments for mucus hypersecretion in COPD patients.

Expectorants

Expectorants are oral drugs that are supposed to enhance the clearance of mucus. Although expectorants were once commonly prescribed, there is little or no objective evidence for their efficacy. Such drugs are often emetics that are given in sub-emetic doses on the basis that gastric irritation may stimulate an increase in mucus clearance via a reflex mechanism. However, there is no good evidence for this assumption. Lacking evidence for their efficacy, the FDA has removed most expectorants from the market in a review of over-the-counter drugs. With the exception of guaifenesin, no agents are approved as expectorants in the U.S. In patients who find it difficult to clear mucus, adequate hydration and inhalation of steam may be of some benefit.

Whenever possible, treat the underlying cause, not the cough. Asthma commonly presents as cough, and the cough will usually respond to ICS. A syndrome characterized by cough in association with sputum eosinophilia but no airway hyperresponsiveness and termed eosinophilic bronchitis also responds to ICS (Birring et al., 2003). Nonasthmatic cough does not respond to ICS but sometimes responds to anticholinergic therapy. The cough associated with postnasal drip of sinusitis responds to antibiotics (if warranted), nasal decongestants, and intranasal steroids. The cough associated with ACE inhibitors (in ∼15% of patients treated) responds to lowering the dose or withdrawal of the drug and substitution of an AT₁ receptor antagonist (Chapter 26). Gastroesophageal reflux is a common cause of cough through a reflex mechanism and occasionally as a result of acid aspiration into the lungs. This cough may respond to suppression of gastric acid with an H₂ receptor antagonist or a proton pump inhibitor (Chapter 45), although even large doses may not always be effective (Chang et al., 2006). Some patients have a chronic cough with no obvious cause, and this chronic idiopathic cough may be due to airway sensory neural hyperesthesia (Haque et al., 2005).

Opiates. Opiates have a central mechanism of action on μ opioid receptors in the medullary cough center, but there is some evidence that they may have additional peripheral action on cough receptors in the proximal airways. Codeine and pholcodine (not available in the U.S.) are commonly used, but there is little evidence that they are clinically effective, particularly on postviral cough; in addition, they are associated with sedation and constipation (Dicpinigaitis, 2009a). Morphine and methadone are effective but indicated only for intractable cough associated with bronchial carcinoma. A peripherally acting opioid agonist, 443C81, does not appear to be effective for cough.

Dextromethorphan. Dextromethorphan is a centrally active N-methyl-D-aspartate (NMDA) receptor antagonist. It may also antagonize opioid receptors. Despite the fact that it is in numerous over-the-counter cough suppressants and used commonly to treat cough, it is poorly effective. In children with acute nocturnal cough, it is not significantly different from placebo in reducing cough (Dicpinigaitis, 2009a). It can cause hallucinations at higher doses and has significant abuse potential.

Benzonatate. (tessalon, others), a local anesthetic, acts peripherally by anesthetizing the stretch receptors located in the respiratory passages, lungs, and pleura. By dampening the activity of these receptors, benzonatate may reduce the cough reflex at its source. The recommended dose is 100 mg, three times/day, and up to 600 mg/day, if needed. Although clinical studies shortly after its approval showed some efficacy, benzonatate, 200 mg, was not effective in suppressing experimentally-induced cough in a recent clinical trial (Dicpinigaitis et al., 2009b). Side effects include dizziness and dysphagia. Seizures and cardiac arrest have occurred following an acute ingestion. Severe allergic reactions have been reported in patients allergic to para-aminobenzoic acid, a metabolite of benzonatate.

Other Drugs. Several other drugs reportedly have small benefits in protecting against cough challenges or in reducing cough in pulmonary diseases (Dicpinigaitis, 2009a). These drugs include moguisteine (not available in the U.S.), which acts peripherally and appears to open ATP-sensitive K⁺ channels; baclofen, a GABA_B-selective agonist; and theobromine, a naturally occurring methylxanthine. Although the expectorant guaifenesin is not typically known as a cough suppressant, it is significantly better than placebo in reducing acute viral cough and inhibits cough-reflex sensitivity in patients with upper respiratory tract infections (Dicpinigaitis et al., 2009b).

Novel Antitussives. There is clearly a need to develop new more effective therapies for cough, particularly drugs that act peripherally in order to avoid sedation. There are close analogies between chronic cough and sensory hyperesthesia, so new therapies are likely to arise from pain research (Barnes, 2007).

Transient Receptor Potential V1 Antagonists. TRPV1 (previously called the vanilloid receptor) is a member of the transient receptor potential (TRP) family of ion channels activated by capsaicin, H⁺, and bradykinin, all of which are potent tussive agents. TRPV1 antagonists block cough induced by capsaicin and bradykinin and are effective in some models of cough (McLeod et al., 2008). A side effect of these drugs is loss of temperature regulation and hyperthermia, which has prevented clinical development.

Transient Receptor Potential A1 Antagonists. TRPA1 is emerging as a more promising novel target for antitussives. This channel is activated by oxidative stress and many irritants and may be sensitized by inflammatory cytokines (Taylor-Clark et al., 2009). Several selective TRPA1 antagonists are now in development.

Doxapram (dopram, others). At low doses (0.5 mg/kg IV), doxapram stimulates carotid chemoreceptors; at higher doses it stimulates medullary respiratory centers. Its effect is transient; thus, intravenous infusion (0.3-3 mg/kg per minute) is needed for sustained effect. Unwanted effects include nausea, sweating, anxiety, and hallucinations. At higher doses, increased pulmonary and systemic pressures may occur. Both the kidney and the liver participate in the clearance of doxapram, which should be used with caution if hepatic or renal function is impaired. In COPD, the infusion of doxapram is restricted to 2 hours. The use of doxapram to treat ventilatory failure in COPD has now largely been replaced by noninvasive ventilation (Greenstone and Lasserson, 2003).

Almitrine. Almitrine bismesylate is a piperazine derivative that appears to selectively stimulate peripheral chemoreceptors and is without central actions (Winkelmann et al., 1994). It is ineffective in patients with surgically removed carotid bodies. Almitrine stimulates ventilation only when there is hypoxia. Long-term use of almitrine is associated with peripheral neuropathy, limiting its availability in most countries.

Acetazolamide. The carbonic anhydrase inhibitor acetazolamide (Chapter 25) induces metabolic acidosis and thereby stimulates ventilation, but it is not widely used because the metabolic imbalance it produces may be detrimental in the face of respiratory acidosis. It has a very small beneficial effect in respiratory failure in COPD patients (Jones and Greenstone, 2001). The drug has proved useful in prevention of high altitude sickness (Basnyat and Murdoch, 2003).

Naloxone. Naloxone is a competitive opioid antagonist that is indicated only if ventilatory depression is due to overdose of opioids.

Flumazenil. Flumazenil is a benzodiazepine receptor antagonist that can reverse respiratory depression due to overdose of benzodiazepines (Gross et al., 1996).

Primary (idiopathic and familial) pulmonary hypertension (PPAH), where there is no identifiable cause, is uncommon. Most cases of pulmonary hypertension are associated with connective tissue disorders, such as systemic sclerosis, or they are secondary to hypoxic lung diseases, such as interstitial lung disease and COPD, where chronic hypoxia leads to hypoxic pulmonary vasoconstriction. Secondary pulmonary hypertension in COPD rarely requires specific therapy. In secondary pulmonary hypertension due to chronic hypoxia, the initial treatment is correction of hypoxia using supplementary O₂ therapy, including ambulatory oxygen, with the aim of increasing O₂ saturation to >90%. Right heart failure is treated initially with diuretics. Anticoagulants are indicated for the treatment of pulmonary hypertension secondary to chronic thromboembolic disease, but they may also be indicated for patients with severe pulmonary hypertension who have an increased risk of venous thrombosis.

Prostacyclin. Prostacyclin (PGI₂; Epoprostenol) is produced by endothelial cells in the pulmonary circulation and directly relaxes pulmonary vascular smooth muscle cells by increasing intracellular cyclic AMP concentrations (see Chapter 33). Reduced prostacyclin production in PAH has led to the therapeutic use of epoprostenol and other stable prostacyclin derivatives (Gomberg-Maitland and Olschewski, 2008). Functionally and physiologically, PGI₂ opposes the effects of TXA₂.

Intravenous epoprostenol (flolan, others) is effective in low-ering pulmonary arterial pressures, improving exercise performance, and prolonging survival in PPAH. Because of its short plasma t_1/2, prostacyclin must be administered by continuous intravenous infusion using an infusion pump. Common side effects are headache, flushing, diarrhea, nausea, and jaw pain. Continuous intravenous infusion is inconvenient and very expensive. This has led to the development of more stable prostacyclin analogs. Treprostinil (remodulin) is given by continuous subcutaneous infusion or as an inhalation (tyvaso), consisting of four daily treatment sessions with nine breaths per session. Oral beraprost sodium appears to be somewhat less effective and its effect seems to diminish after 3 months of therapy, so it has not been approved in most countries. Iloprost (ventavis) is a stable analog that is given by inhalation, but it needs to be given by nebulizer six to nine times daily. It is associated with the vasodilator side effects of prostacyclin, including syncope. It may also cause cough and bronchoconstriction because it sensitizes airway sensory nerves.

Endothelin Receptor Antagonists

Endothelin-1 (ET-1) is a potent pulmonary vasoconstrictor that is produced in increased amounts in PAH. ET-1 contracts vascular smooth muscle cells and causes proliferation mainly via ET_A receptors. ET_B receptors mediate the release of prostacyclin and NO from endothelial cells. Several endothelin antagonists are now on the market for the treatment of PPAH (see Chapter 26).

Bosentan (tracleer) was the first ET antagonist developed and is a antagonist of both ET_A and ET_B receptors. Several long-term clinical trials have now established the efficacy of oral bosentan in reducing symptoms and improving mortality in PPAH. Starting dose is 62.5 mg twice daily for 4 weeks, then increasing to the maintenance dose of 125 mg twice daily. The drug is generally well tolerated. Adverse effects include abnormal liver function tests, anemia, headaches, peripheral edema, and nasal congestion. Given the risk of serious liver injury, liver aminotransferases should be monitored monthly. A class effect is a risk of testicular atrophy and infertility; bosentan is potentially teratogenic.

Ambrisentan (letairis) is selective ET_A receptor antagonist. It is given orally once daily at a dose of 5-10 mg. The theoretical advantage of blocking only ET_A receptors is that ET_B receptors may continue to stimulate release of PGI₂ and NO, giving a greater therapeutic effect. However, its clinical efficacy is similar to that of bosentan, as are its adverse effects. Use of ambrisentan also requires monthly monitoring of liver aminotransferases. Sitaxsentan (not available in the U.S.) is similar to ambrisentan but may be less likely to cause liver dysfunction.

Phosphodiesterase 5 Inhibitors

Nitric oxide activates soluble guanylate cyclase to increase cyclic GMP, which is hydrolyzed to 5′GMP by PDE5 (Chapter 27). Elevation of cGMP in smooth muscle causes relaxation (Chapter 3), which the inhibition of PDE5 prolongs and accentuates. In the pulmonary bed, inhibition of PDE5 induces vasodilation.

Sildenafil. Sildenafil (revatio) is a selective PDE5 inhibitor that is given at a dose (20 mg three time daily orally) that is lower than used for erectile dysfunction (100 mg; Chapter 27). It is effective in lowering pulmonary resistance and improving exercise tolerance in patients with PAH. Side effects include headache, flushing, dyspepsia, and visual disturbances.

Tadalafil. Tadalafil (adcirca) has a longer duration of action than sildenafil so may be suitable for once-daily dosing. It is approved for World Health Organization group 1 PAH to improve exercise ability.