The physiologic function of cough is to remove foreign substances and mucus from the respiratory tract. It is a three-phase mechanical process that involves a deep inspiration, increasing lung volume, and muscular contraction against a closed glottis and sudden opening of the glottis. The maneuver produces and sustains a high linear air velocity to expel material from the respiratory tree.

Cough is a reflex response that is mediated by the medulla but is subject to voluntary control. The afferent limb may involve receptors in the larynx, respiratory tree, pleura, acoustic duct, nose, sinuses, pharynx, stomach, or diaphragm. The receptors respond to mechanical, inflammatory, or irritant stimuli. The trigeminal, glossopharyngeal, phrenic, and vagus nerves can carry the afferent signal. The efferent limb of the cough reflex involves the recurrent laryngeal, phrenic, and spinal motor nerves, which innervate the respiratory muscles.The most common cause of chronic cough is cigarette smoking, which may trigger the cough reflex by direct bronchial irritation; alternatively, smoking may induce inflammatory changes and the production of mucus, which stimulates a self-propagating productive cough. Chronic bronchitis may ensue. Chronic cough and decreased flow rates have been observed in adolescents after only 3 to 5 years of smoking. Pipe and cigar smoking cause lesser degrees of difficulty.

Environmental irritants play a major role in the production of cough in patients living in industrialized urban areas. Pollutants that are frequently involved are heavy smog, sulfur dioxide, nitrous oxide, and industrial gases such as ammonia. In Great Britain, the relationship between air quality and production of cough has been documented. The dusts and particulate matter that are capable of producing pneumoconioses contribute to the problem (see Chapter 39). The excessive drying of normal airway moisture that takes place in centrally heated homes (humidity may fall below 10% unless a humidifier is utilized) can result in a persistent dry cough during the winter months.

Inflammation anywhere along the upper or lower respiratory tract is capable of producing cough; receptors capable of transmitting impulses that stimulate cough are believed to be distributed throughout the respiratory system. The greater the inflammatory stimulus, the greater is the white cell response and the more purulent is the sputum. (The green coloration of very purulent sputum is caused by the degeneration of white cells.) A number of patients experience a dry, persistent cough after an upper respiratory infection; these postinfectious coughs commonly last more than 3 weeks and may last more than 8 weeks. The pathophysiology may be unrelated to postnasal drip or airway hyperactivity and is believed to be related to airway epithelial damage. In some populations, infection with Bordatella species has proven to be a relatively common explanation for highly prolonged cough associated with infection.

When airway hyperactivity is present, cough is a common symptom. Most patients with classical asthma complain of cough, and in some cases cough is the symptom that predominates in the clinical picture. Studies of asthmatic patients have emphasized that cough can occur in the absence of wheezing or abnormalities on routine pulmonary function testing. The cough is characteristically worse at night and can be triggered or exacerbated by exposure to environmental irritants, allergens, or cold. Exercise is a common stimulant. In such cases, the bronchorrheal component of asthma predominates, but methacholine or carbachol challenge will often unmask the obstructive manifestations (see Chapter 48). Chronic bronchitis due to smoking is among the most common causes of chronic cough and sputum production. The condition is defined clinically as the presence of a productive cough that persists for at least 3 months for 2 consecutive years. A morning cough is often prominent, and bronchospasm is a frequent accompaniment (see Chapter 47). Bronchiectasis is also characterized by cough and sputum production, but it differs clinically from bronchitis in that repeated bouts of hemoptysis and pneumonia are more likely to occur. Copious amounts of purulent sputum are often produced. Chronic cough and sputum production commonly persist between episodes of pneumonia. Focal destruction of supporting lung tissue leads to dilation of
bronchi and focal findings of rhonchi and wheezes on physical examination. A history of suppurative pneumonia in childhood is sometimes elicited. Eosinophilic bronchitis in the absence of asthma has also been associated with chronic cough in 10% to 15% of cases in both primary care and specialist clinics. Its pathophysiology is largely undefined, though the cough does respond to inhaled corticosteroids.

Carcinoma of the lung, more often than not in the smoker, may present with cough in its early stages, particularly when an endobronchial lesion is present. Often, the cigarette smoker notes a change in the pattern of a chronic “cigarette cough.” Hemoptysis is noted in about 5% to 10% of early cases. Other clues are localized wheezing and purulent sputum suggestive of obstruction. In later stages, cough is present in conjunction with weight loss, anorexia, and dyspnea. In some instances, a systemic syndrome (e.g., inappropriate secretion of antidiuretic hormone, hypertrophic pulmonary osteoarthropathy, dermatomyositis, peripheral neuropathy) may precede the appearance of tumor.

Interstitial pathology including fibrosis and pulmonary edema may stimulate mechanical receptors and result in a nonproductive cough. Congestive heart failure with chronic interstitial pulmonary edema is associated with nocturnal cough because venous return is increased at night, which worsens heart failure (see Chapter 32). When failure is severe, frothy pink or blood-tinged sputum may be noted. Extraluminal compression of bronchi also stimulates mechanical receptors; examples of compressing lesions include hilar adenopathy, aortic aneurysm, and neoplasm.

Because receptors of the afferent limb of the cough reflex are found in the nose, pharynx, sinuses, and acoustic ducts, common afflictions in these areas have been found to be common causes of cough. Chronic allergic rhinitis (see Chapter 222) with resultant postnasal drip ranks as one of the leading causes of chronic cough in specialty clinic populations. The nasal mucosa may be edematous and the pharyngeal mucosal “cobblestoned” in appearance. Similarly, sinusitis (see Chapter 219) may be associated with a persistent cough and sputum production secondary to excessive retropharnygeal drainage of mucus. It accounts for up to one third of patients with postnasal drip syndrome. Even impacted cerumen and external otitis have been implicated in stimulating the cough reflex (see Chapter 218).

Because there are afferent limb receptors for the cough reflex in the stomach and lower esophagus, it is not surprising that a condition as common as gastroesophageal reflux is associated with chronic cough. In fact, it is among the three most common causes identified in case series of patients with persistent chronic cough. Mechanisms include (a) esophageal irritation with stimulation of an esophageal–tracheobronchial reflex and (b) nocturnal aspiration of gastric juices. Cough may be the only presenting symptom.

The use of angiotensin-converting-enzyme (ACE) inhibitors has been associated with an unexpectedly high incidence of dry nocturnal cough, with reports of 10% to 15% of patients being affected. First reported with use of enalapril, the cough has been associated with most long-acting ACE inhibitor preparations. Patients complain of an irritated feeling. The cough usually does not respond to a switch to another ACE inhibitor, although reducing the dose may help. In about 50% of instances, the cough is so annoying that ACE inhibitor therapy must be terminated. The pathophysiology of ACE inhibitor-induced cough is not entirely understood, but it appears to be an increase in sensitivity to the cough reflex. Therefore, ACE inhibitors may be unmasking subclinical cough associated with one of the aforementioned mechanisms.

Psychogenic cough has been described as more prevalent in children, but it may occur in adults; characteristically, it is nonproductive, occurs at times of emotional stress, and ceases during the night. The prevalence of psychogenic cough in reported series varies inversely with the attention to systematic evaluation and search for the foregoing mechanisms.

The common causes of chronic cough are listed in Table 41.1. In an often-cited series of 139 consecutive cases of chronic cough encountered in the community setting, the cause was hyperactive airway disease in 21%, postnasal drip in 19%, postinfectious status in 9%, chronic bronchitis in 4%, gastroesophageal reflux in 4%, and, in a few cases, occupational lung disease and psychiatric illness. In one referral setting study, a postnasal drip syndrome accounted for 41% of cases, asthma for 24%, esophageal reflux for 21%, and chronic bronchitis for 5%. Cough was the sole presentation of asthma in 28% of asthmatic patients and of reflux in 43% of patients with reflux. In one fourth of cases, more than one cause was identified. Sinusitis accounted for 38% of cases of postnasal drip. Insight from a growing number of series of patients from specialty cough clinics using systematic approaches to diagnosis suggests that when a patient who is not a smoker and not taking ACE inhibitors presents with chronic cough and has a normal chest radiograph, it is highly likely that the etiology will be related to asthma, gastroesophageal reflux disease (GERD), a postnasal drip syndrome, or some
combination of these three entities. This has been termed the pathogenetic triad of chronic cough of relatively obscure origin. Rare but noteworthy causes of chronic cough include irritation of the pleura, diaphragm, or pericardium. Osteophytes of the cervical spine and pacemaker malfunction have been reported as truly rare causes of cough.

Although in some cases the cause of a chronic cough is readily apparent, presentations of even the common underlying conditions may be subtle, so that careful investigation is necessary. During the initial workup, the physician should consider serious causes (cancer, TB, heart failure) while checking for the much more common treatable causes (asthma, esophageal reflux, postnasal drip). In a detailed study of workup for chronic cough, history offered the highest yield, with 70% of patients having a true-positive finding; physical examination was second, with 49%, and laboratory studies were third, with an average of 22% of patients having true-positive findings.

The most effective means of stopping the cough is to identify and treat the underlying cause (see Chapters 47, 49, 61, and 222). An empiric trial of an etiologic therapy can be highly effective in providing a diagnosis as well as relief of symptoms (see prior discussion); however, certain etiologic therapies should not be used empirically, especially antibiotics in the absence of proven infection. Symptomatic management is distinguished from empiric etiologic therapy. It is directed at eliminating precipitants and suppressing the cough. The goal is to prevent the complications that may result from prolonged forceful coughing, such as sleeplessness, musculoskeletal pain, rib fractures, pneumothorax, exhaustion, pneumomediastinum, posttussive syncope (see Chapter 24), and rupture of subconjunctival or nasal veins. The occurrence of any of these complications may be a reason for occasionally suppressing a cough that has not been completely diagnosed.

The first priority and simplest manipulation is to remove or reduce irritants. Of paramount importance is cessation of smoking and passive exposure to cigarette smoke; this alone eliminated cough in 77% and reduced it in another 17% of patients within 1 month. Second, an appropriate humidity should be maintained. If a humidifier is used, it should be kept clean because it can become colonized with bacteria or fungi and cause infection or hypersensitivity pneumonitis. Third, adequate internal hydration should be encouraged, with at least 1,500 mL of fluid consumed daily. These simple measures alone may abolish cough in many patients.

The patient with a chronic cough secondary to established underlying lung disease requires careful education. The patient must be informed that sputum should be expectorated when possible. Patients with chronic bronchitis or bronchiectasis can be taught how to cough with quiet, forceful expirations and how to perform postural drainage to promote removal of mucus from the bronchioles. Postural drainage is best performed
before meals and at bedtime. Ipratropium is sometimes helpful in reducing nighttime cough in the patient with chronic obstructive pulmonary disease (see Chapter 47).

Patients with chronic cough often request temporary cough suppression to allow uninterrupted sleep; such suppression is also required when complications of cough arise. A wide variety of agents have been used to treat cough. The most effective are the narcotic antitussives, which act centrally to suppress the medullary cough center. Other preparations are expectorants or mucolytic agents, which merely help to mobilize sputum. They can also have a placebo effect. When cough significantly interferes with sleeping or eating, a narcotic cough suppressant should be used. Codeine is the drug of choice. It should be given in relatively small doses of 8 to 15 mg at intervals of 3 to 4 hours, according to the patient's needs. In many instances, a dose before bedtime will suffice. Liquid and tablet preparations are equally effective. If a small dose does not suppress the cough, doses of up to 60 mg every 3 to 4 hours may be tried. It is worth noting that many patients expect to use a syrup for cough suppression; prescribing the drug in syrup form may provide some psychological benefit. Patients for whom a narcotic antitussive is prescribed should be given small quantities and followed closely to ensure that the cough resolves and excessive use does not result. The obvious exception to this precaution is the patient with incurable lung cancer or other terminal illness, who should receive the doses necessary to provide relief from the discomfort of persistent cough.

Nonnarcotic antitussives lack addiction potential but are not as effective as codeine. The most popular and effective over-the-counter cough suppressant is dextromethorphan, which has a mild suppressant effect. Many over-the-counter preparations contain alcohol, sympathomimetics, and antihistamines. The mucolytic effects of alcohol are minimal. The sympathomimetics are of little use except in patients whose cough derives from chronic vasomotor rhinitis (see Chapter 222). The antihistamines are most useful for patients with allergic upper airway disease (see Chapter 222) and are a helpful adjunct for inducing sleep when taken at bedtime. Some over-the-counter agents dull the peripheral sensory receptors; this is the rationale for putting mild topical anesthetics in sprays, syrups, and cough lozenges. They are of questionable utility.

Expectorants are heavily consumed. More than 60 preparations containing guaifenesin are available; terpin hydrate is another popular expectorant. These agents are often combined with an effective cough suppressant and, as such, produce a beneficial effect, but by themselves they have no proven effect and represent an unnecessary expense. They are given when the patient insists on something for cough but clear indications for cough suppression are lacking, or when the patient believes expectorants will help. Systematic review of randomized trials of over-the-counter remedies in the setting of acute cough suggests either no effect or effect size small enough to be of doubtful clinical significance.

Patients with cough secondary to asthma respond to inhaled topically active corticosteroids and bronchodilators (see Chapter 48). Topical steroid therapy is effective for patients with eosinophilic bronchitis and may also help in allergic rhinitis (see Chapter 222). Patients with persistent cough after a recent respiratory tract infection and no signs of pneumonitis may benefit from a short course of inhaled steroid therapy, which presumably lessens residual inflammatory changes. Time is another effective therapy. If Bordatella pertussis infection is suspected, treatment with erythromycin or (if the patient is allergic) trimethoprim–sulfamethoxazole for 14 days should be considered.

Patients with suspected reflux should respond to a course of antireflux therapy with antacids, histamine₂ blockers, and proton-pump inhibitors (see Chapter 61), although, as noted, the benefits may not become apparent for several weeks.

Although endobronchial cancer is feared, it is not a common cause of chronic cough in the absence of other findings, especially in the patient with normal chest radiographic findings. However, for the patient without a diagnosis who has risk factors for cancer (smoking, occupational exposure), a chest CT may be indicated. A consultation for consideration of bronchoscopy may also be warranted, particularly if the CT findings are abnormal. The patient without risk factors and normal chest radiographic findings, or the patient with risk factors and a normal chest CT, can be followed expectantly without resort to bronchoscopy because the likelihood of a positive study result is very small.