Management of chronic cough

Summary

Cough that remains unexplained after basic clinical assessment is a common reason for referral to secondary care. Much of the evidence about management of isolated chronic cough is derived from case series; this evidence suggests that isolated chronic cough is usually due to asthma, gastro-oesophageal reflux disease, and upper airway conditions, and that it can be cured in most people by treatment of these conditions. However, there is increasing recognition that satisfactory control of chronic cough is not achieved in a substantial number of patients seen in secondary care. Moreover, there is a concern that perpetuation of the belief that chronic cough is solely due to the effects of comorbid conditions is inhibiting research into the pathophysiology of an abnormally heightened cough reflex, and jeopardising development of improved treatments. We advocate a change in emphasis, which makes a clear distinction between cough due to corticosteroid-responsive eosinophilic airway diseases and corticosteroid-resistant non-eosinophilic cough. We recommend that some factors with weak evidence of an association with cough are best viewed as potential aggravating factors of an intrinsic abnormality of the cough reflex, rather than the cause. We call for more research into the basic mechanisms and pharmacological control of an abnormally heightened cough reflex, and recommend ways to assess the effects of potentially antitussive treatments.

Introduction

Chronic cough is defined as cough lasting for more than 8 weeks.¹ This definition is based on evidence that a cough lasting longer than this duration is unlikely to be due to a respiratory tract infection.² Chronic cough is a common symptom of almost all chronic respiratory, and some non-respiratory conditions. Several recognisable causes of chronic cough, such as chronic obstructive pulmonary disease (COPD), chronic bronchitis, lung cancer, an inhaled foreign body, pulmonary tuberculosis, sarcoidosis, idiopathic pulmonary fibrosis, and heart failure will be obvious after clinical assessment, spirometry, and chest radiography. The diagnostic approach and management of these conditions is outside the scope of this Series. We will focus on the diagnosis and management of chronic cough in adults when no obvious cause can be identified by such an assessment. Cough is the primary focus of referral in 38% and the sole focus in 10% of patients seen in a typical adult respiratory clinic.³

Patients referred with isolated chronic cough1, 4, 5 are assessed on the basis of the anatomic, diagnostic protocol originally described by Irwin and colleagues⁶ more than 25 years ago. This protocol is so named because the emphasis is on the importance of conditions—especially, asthma, rhinosinusitus, and gastro-oesophageal reflux—that affect structures within the anatomical distribution of vagal afferent nerves.6, 7, 8 Important modifications to the protocol are the recognition that eosinophilic bronchitis without asthma is an important cause of chronic cough, and acceptance that non-invasive assessment of airway inflammation is a desirable step in the assessment of patients with chronic cough.9, 10 Although this protocol has undoubtedly been an important advance in the management of chronic cough, it remains largely based on expert opinion, with an absence of evidence from randomised, double-blind, placebo-controlled trials supporting central tenets of the protocol. Nevertheless, there are many reports of successful management of cohorts of patients with chronic cough, by use of variants of the anatomic, diagnostic approach to investigation.3, 6, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 The findings in these reports should be tempered by those from others, in which management has not been as successful.19, 20, 21 There is increasing recognition that up to 46% of patients seen in secondary care with chronic cough have an unexplained cough, despite extensive investigation and treatment trials (see the preceding part of this Series), suggesting that aspects of the anatomic, diagnostic investigation protocol need to be modified. The most important difficulties with present management protocols apply especially to patients whose cough is not due to an obvious major factor such as smoking or treatment with an angiotensin-converting enzyme (ACE) inhibitor, and who do not have evidence of an eosinophilic airway disease (panel).

We propose that when treating patients in whom cough is not due to an obvious pulmonary disease or a major factor such as smoking or treatment with ACE inhibitors, a distinction should be drawn between cough due to eosinophilic airway diseases and non-eosinophilic chronic cough. This distinction is supported by the substantial differences in epidemiology, pathology, and expectation of treatment responses between these groups (table 1). We believe that no preconceptions should exist about the underlying causes of non-eosinophilic cough. Moreover, extrapulmonary factors such as rhinitis and gastro-oesophageal reflux, in which the evidence of a causal association with chronic cough is weak, are best viewed as potential aggravating factors of any underlying abnormality of the cough reflex, especially prevalent in middle-aged women, rather than causes. The model in figure 1 has the advantage of providing a basis for an understanding of the incomplete response to treatment seen in many patients. This is important because it will stimulate rather than inhibit research into other causes of a heightened cough reflex, which in turn could result in the development of improved treatment.³⁰