Cough is a frequent symptom of many respiratory diseases and is often associated with underlying lung pathology and an abnormal CXR. Cough can occur in otherwise healthy people and is often a self-limiting symptom. Persistent coughing can be a socially disabling and distressing symptom, for which help is often sought. Cough syncope is loss of consciousness following violent coughing, a Valsalva-type manoeuvre, which impairs venous return to the heart and provokes bradycardia and vasodilatation (similar to an ordinary faint). Important as car drivers must cease driving until liability to cough syncope has ceased, confirmed by medical opinion; commercial drivers must cease driving and have no cough syncope or pre-syncope for 5y if they have a chronic respiratory condition, including smoking. If they have asystole due to cough, driving can be considered after pacemaker insertion.
• Acute cough = cough lasting <3 weeks, usually due to viral upper respiratory tract infection (URTI). May linger for 3–8 weeks as a ‘post-viral cough’ or subacute cough
• Chronic cough = cough lasting >8 weeks
• Patients with a normal CXR and persistent cough are often grouped under the heading ‘chronic cough’
• It can sometimes be difficult to determine the underlying cause
• Susceptible individuals have a heightened cough reflex (therefore coining the term ‘cough hypersensitivity syndrome’)
• Investigation is warranted, but successful response to therapeutic trials may aid determination of the underlying cause. Centres vary in their approach to this
• Specialist cough clinics suggest they achieve diagnosis and effective treatment in over 80% of patients referred with chronic cough.
In practice, over 90% of cases of chronic cough with a normal CXR are caused by one or more of:
• Cough variant asthma or eosinophilic bronchitis
• Gastro-oesophageal reflux disease (GORD)
• Post-nasal drip (or upper airway cough syndrome), due to perennial or allergic rhinitis, vasomotor rhinitis, or chronic sinusitis.
Full history can be unhelpful. Although cough is most commonly due to asthma, reflux, or post-nasal drip, there may be no specific symptoms to suggest these diagnoses.
• Duration of cough
• When it tends to occur—night or early morning, after exertion, on exposure to dust, pollen, aerosols, cold air (asthma), after meals or on sitting or bending over (GORD), nocturnal (post-nasal drip and asthma)
• Non-productive or productive and, if so, how much sputum and colour. Significant amounts of sputum usually indicate a 1° lung pathology
• Associated symptoms:
• Previous respiratory disease such as childhood asthma, eczema, or hay fever
• History of sinus disease or perennial rhinitis
• History of previous severe respiratory infections, such as whooping cough, that may have caused bronchiectasis
• Known cardiac disease or valvular heart disease
• Drug history ?ACE inhibitor
• Occupation ?Workplace irritants
• Smoker (common cause of persistent cough, dose-related, improves on stopping)
• Use of recreational drugs.
can also be unhelpful, as it is usually normal. Look for signs of underlying lung disease or other medical conditions such as heart failure, neurological disease (particularly bulbar involvement). Significant tonsillar enlargement should be excluded, as this is a recognized cause of cough which can respond to tonsillectomy.
• Ensure CXR is normal
• Spirometry may indicate restrictive or obstructive defect. Performance of spirometry may provoke cough and bronchospasm
• Methacholine challenge test (see p. [link]) provides the best positive predictive value for cough due to asthma. Lack of response means cough variant asthma is extremely unlikely. PC20 is normal in eosinophilic bronchitis
• Serial peak flow recordings twice daily for 2 weeks. >20% diurnal variation suggests asthma. Can be normal in cough variant asthma
• Induced sputum examination, if available, for eosinophil count, to suggest either asthma or eosinophilic bronchitis.
• Consider chest HRCT if any features suggestive of lung cancer or interstitial lung disease (ILD), as a small proportion may present with a normal CXR (central tumour)
• Consider ENT examination if predominantly upper respiratory tract disease, resistant to treatment. Consider sinus CT
• Consider bronchoscopy if foreign body possible, or history suggestive of malignancy, small carcinoid, endobronchial disease. Perform after CT to help guide bronchoscopist
• Consider 24h ambulatory oesophageal pH monitoring
• Consider oesophageal manometry for oesophageal dysmotility.
The initial treatment of patients with a chronic cough is determined by what the most likely underlying cause is, based on the history and investigations. The key is to give any drug treatment at a high enough dose, and for a long enough time (such as 2–3 months), to be effective.
Symptomatic treatment for cough
Over-the-counter medicines may provide relief, although there is little evidence of a specific pharmacological effect. Below is a list of possible treatments:
• Honey and lemon—home remedies
• Dextromethorphan—a non-sedating non-opiate. Component of many over-the-counter cough remedies. Dose response, with maximum cough reflex suppression at 60 mg (Benylin® preparations, Actifed® preparations, Vicks Vaposyrup® preparations, Sudafed Linctus®, Night Nurse®)
• Menthol—short-lived cough suppressant (Benylin® preparations, Vicks Vaposyrup® preparations)
• Sedative antihistamines—suppress cough but cause drowsiness. Good for nocturnal cough
• Codeine or pholcodine—opiate antitussives—codeine requires prescription. No greater efficacy than dextromethorphan and greater side effect profile
• Opiates—prescription. Low-dose morphine sulfate 5–10mg showed significant improvement in patients with intractable cough in randomized controlled trial (RCT) (Morice AH et al. Am J Crit Care Med 2007). Side effect profile of opiates, so should be used with caution
• Gabapentin—prescription. Neuromodulator used for chronic pain. Not yet licensed for cough, but RCT treatment success in Leicester Cough Questionnaire, with side effects mostly of nausea and fatigue in 31% on gabapentin, managed with dose reduction (Ryan NM et al. Lancet 2012). Starting dose 300mg/day, with gradual increases until cough suppressed, side effects, or maximum 600mg tds. Other side effects include diarrhoea, emotional lability, sleepiness, nystagmus, tremor, weakness, peripheral oedema
• Thalidomide—used in cough due to IPF, with RCT showing significant improvements in cough quality of life (QoL) questionnaires (Horton MR et al. Ann Intern Med 2012). Side effects in 74% on thalidomide vs 22% placebo, with constipation, dizziness, and fatigue (for IPF, see pp. [link]–[link]).
Assessing treatment response
Several measures have been developed and validated such as:
• Cough visual analogue scale
• Leicester Cough Questionnaire—cough-specific QoL
• Cough reflex sensitivity measurements—primarily a research tool; subjects inhale increasing doses of either capsaicin or citric acid, with the sensitivity recorded as the dose to cause two or five coughs.
• Infection: viral upper and lower respiratory tract infection, bacterial pneumonia, tuberculosis (TB), pertussis
• Chronic bronchitis
• Obstructive airways disease: COPD, asthma
• Cough variant asthma
• Eosinophilic bronchitis
• Obstructive sleep apnoea (OSA) (nocturnal only)
• Lung cancer
• Bronchiectasis, cystic fibrosis (CF)
• Airway irritants: smoking, dusts and fumes, acute smoke inhalation
• Airway foreign body.
• External compression of trachea by enlarged lymph nodes (e.g. lymphoma, TB)
• Mediastinal tumours/cysts/masses.
• Upper airway cough syndrome, including:
• Acute or chronic sinusitis
• Post-nasal drip due to perennial, allergic, or vasomotor rhinitis.
• Oesophageal dysmotility, stricture, or pharyngeal pouch causing repeated aspiration
• Oesophago-bronchial fistula.
Neurological disease affecting swallowing, causing repeated aspiration, such as stroke, multiple sclerosis, motor neurone disease (MND), or Parkinson’s disease.
or ‘cough variant asthma’, ‘cough-predominant asthma’. This represents one end of the asthma spectrum, with airway inflammation, but may have minimal bronchoconstriction. There is not always a typical asthma history, but ask about wheeze, atopy, hay fever, or childhood asthma or eczema. Cough may be the only symptom. Cough is typically worse after exercise, in cold air, after exposure to fumes or fragrances, or in the mornings.
• Spirometry may be normal, without evidence of airflow obstruction. There may be typical asthmatic diurnal peak flow variability of >20%, or peak flows may be stable
• Methacholine challenge should be positive for asthma but does not rule out a steroid-responsive cough. If negative, other causes of cough should be sought
• Treatment should be for at least 2 months, with high-dose inhaled steroids. Response may take days or weeks. Bronchodilators may make little difference. If inhaled steroid therapy has been tried unsuccessfully, ensure inhaler technique is optimal and a high dose has been used. Alternatively, prescribe a 2-week course of oral prednisolone 30mg/day, and assess response. If the cough improves, high-dose inhaled steroids should be continued and slowly reduced after about 2 months. There is a small trial showing leukotriene receptor antagonists decreased cough in people with cough variant asthma; consider using in patients who want to avoid inhaled steroids or in whom they are ineffective
• Eosinophilic bronchitis Airway eosinophilia, rarely with peripheral blood eosinophilia, causing heightened cough reflex, but no bronchial hyperresponsiveness/wheeze or peak flow variation. Diagnosis based on negative asthma investigations and induced sputum eosinophilia. Improves with inhaled corticosteroids, usually after 2–3 weeks, or trial of oral prednisolone. Sputum eosinophil count also reduces with treatment. If there is no response, the cough is unlikely to be due to eosinophilic airway inflammation.
Cough may be related to distal reflux at the lower oesophageal sphincter (LOS) or due to micro-aspiration of acid into the trachea. There may be associated oesophageal dysmotility. LOS reflux is often long-standing and is associated with a productive or non-productive daytime cough, and minimal nocturnal symptoms. It is worse after meals and when sitting down, due to increased intra-abdominal pressure being transmitted to the LOS. Micro-aspiration is associated with more prominent symptoms of reflux or dyspepsia, although these are not always present. Patients may have an intermittent hoarse voice, dysphonia, and sore throat. Cough may be the only symptom of reflux.
A trial of treatment
for both is recommended. This is with a high-dose proton pump inhibitor (PPI) for at least 2, usually 3, months, although longer treatment may be required to control cough. H2 receptor blockers are also effective, and prokinetics like metoclopramide may help as an addition if cough improves but has not gone completely. Other reflux avoidance measures should be carried out: avoiding caffeine, fatty foods, chocolate, excess alcohol, acidic drinks like orange juice, red wine, stop smoking, loose-fitting clothes, sleeping with an empty stomach (avoid eating <4h before bed), sleeping propped up, weight loss if overweight. Surgical fundoplication for reflux-associated cough resistant to drug therapy is not widely used but may be effective in carefully selected cases.
if required, due to either treatment failure or because of diagnostic uncertainty, is with 24h ambulatory pH monitoring, which determines the presence of reflux events, and event markers allow correlation with cough. These may not necessarily be responsible for the cough, so it is not a very specific or sensitive test. Oesophageal manometry can be used to measure the LOS pressure and oesophageal contractions after swallowing to determine the presence of oesophageal dysmotility.
Rhinitis and post-nasal drip
The term upper airway cough syndrome (UACS) is now being used to include all upper airway abnormalities causing cough and is replacing post-nasal drip. Rhinitis is defined as sneezing, nasal discharge, or blockage for >1h on most days for either a limited part of the year (seasonal) or all year (perennial). Rhinitis may be allergic (e.g. hay fever), non-allergic, or infective. The associated nasal inflammation may irritate cough receptors directly or produce a post-nasal drip. These secretions may pool at the back of the throat, giving a sensation of liquid dripping into the back of the throat, which requires frequent throat clearing, or drip directly into the trachea, initiating cough. There may be frequent nasal discharge. A history of facial pain and purulent nasal discharge suggests sinusitis, which can also predispose to post-nasal drip. Symptoms of cough can occur on lying but can be constant, regardless of position. Rhinosinusitis describes inflammation and infection within the nasal passages and paranasal sinuses, with chronic rhinosinusitis defined as symptoms persisting for more than 12 weeks.
may reveal swollen turbinates, ‘cobblestone’ nasopharyngeal mucosa, nasal discharge, or nasal polyps.
Nasal preparations should be taken by kneeling with the top of the head on the floor (‘Mecca’ position) or lying supine with the head tipped over the end of the bed. Improvements in cough should be found within 2 weeks. Duration of treatment is unclear.
• Non-allergic rhinitis Trials suggest the best results are with an initial 3 weeks of nasal decongestants with first-generation antihistamines (which have helpful anticholinergic properties) and pseudoephedrine. Alternatives are nasal ipratropium bromide or xylometazoline. This is then followed by 3 months of high-dose nasal steroids, which are ineffective when used as first-line treatment. Second-generation antihistamines (i.e. non-sedating) are of no use in non-allergic rhinitis
• Allergic rhinitis Second-generation oral antihistamine (e.g. cetirizine, loratadine, fexofenadine) and high-dose nasal steroids for 3 months at least
• Vasomotor rhinitis Nasal ipratropium bromide for 3 months; nasal steroids may also have a role
• Chronic rhinosinusitis Nasal steroids and saline lavage, which should have an effect by 4 weeks, and, if so, treatment should continue, although optimal duration unclear.
is an infection of the paranasal sinuses, which may complicate an URTI and is frequently caused by Haemophilus (H.) influenzae or Streptococcus (S.) pneumoniae. It causes frontal headache and facial pain. Chronic sinusitis may require further investigation with CXR or CT, which shows mucosal thickening and air-fluid levels. Surgery may be indicated.
• Chronic sinusitis Treat as for non-allergic rhinitis, but include 2 weeks of antibiotics active against H. influenzae such as doxycycline or co-amoxiclav.
Respiratory tract infections, especially if viral in nature, can cause cough. This may take weeks or months to resolve spontaneously, although most settle within 8 weeks. There may be a post-nasal drip contribution. The cough is related to a heightened cough reflex. Associated laryngospasm can occur, which is a sudden hoarseness, with associated stridulous inspiratory efforts and a sensation of being unable to breathe.
ACE inhibitor cough
occurs with any ACE inhibitor and is related to bradykinin not being broken down by angiotensin-converting enzyme and accumulating in the lung. Occurs in 10–15% of people on ACE inhibitors; more frequent in women. Can occur within weeks of starting the drug, but up to 6 months; the cough may be initiated by a respiratory tract infection but persists thereafter. Cough usually settles within a week of stopping the drug but may take months. Avoid all ACE inhibitors thereafter and may need to change to an angiotensin receptor antagonist. Stop ACE inhibitor in any patient with a troublesome cough.
Idiopathic chronic cough
accounts for 20% of referrals to a specialist cough clinic. It is diagnosed after a thorough assessment. Typically, there is lymphocytic airway inflammation, but there may also be a history of reflux cough. Typically, the patients are middle-aged women with a long-standing dry cough, often starting around the time of the menopause and triggered by an URTI. Organ-specific autoimmune disease is present in up to 30%, particularly hypothyroidism. Treatment is often ineffective.
BTS guidelines. Recommendations for the management of cough in adults. Thorax 2006;61(suppl 1). http://www.thoraxjnl.com.Find this resource:
Review series on cough. Morice A, Kastelik J. Thorax 2003;58:901–5;Find this resource:
Fontana GA, Pistolesi M. Thorax 2003;58:1092–5;Find this resource:
Dicpinigaitis PV. Thorax 2004;59:71–2;Find this resource:
McGarvey LP. Thorax 2004;59:342–6.Find this resource:
Birring SS et al. Development of a symptom specific health status measure for patients with chronic cough: Leicester Cough Questionnaire. Thorax 2003;58:339–43.Find this resource:
Birring SS et al. Eosinophilic bronchitis: clinical features, management and pathogenesis. Am J Resp Med 2003;2:169–73.Find this resource: