Approach to chronic cough in children

INTRODUCTION — Cough is the most common reason why patients acutely seek medical consult in countries where data are available such as in the United States [1] and Australia [2]. Cough as a symptom is sometimes trivialized by health professionals but is often distressing to a child's family or caregivers, irrespective of whether the cough is acute [3] or chronic [4].

Careful assessment of children with chronic cough is important to identify any underlying disease. The approach outlined in this topic review is designed for use in all patients up to 18 years of age. However, for adolescents 15 years and older, it is reasonable to use adult guidelines and algorithms since the etiologies of chronic cough in adolescents are fairly similar to those in adults. (See "Causes and epidemiology of subacute and chronic cough in adults" and "Evaluation and treatment of subacute and chronic cough in adults".)

An approach to the diagnosis and management of chronic cough in children is presented here, preceded by a brief overview of the epidemiology, physiology, and pathophysiology. Approaches to wheezing and stridor in children are presented separately, although these symptoms may present concurrently. (See "Evaluation of wheezing in infants and children" and "Assessment of stridor in children".)

DEFINITIONS

Chronic cough — Chronic cough in children 14 years and younger usually is defined as a daily cough lasting four or more weeks [5]. This definition is based upon expert consensus, as expressed in guidelines from the American College of Chest Physicians and Thoracic Society of Australia and New Zealand [5] and the European Respiratory Society [6]. The rationale is that most acute respiratory infections in children resolve within this interval and that the presence of chronic cough impairs quality of life [7] and may represent sinister underlying etiologies [8]. The four-week threshold was selected to permit relatively early diagnosis of serious underlying illness such as missed foreign body [9] and bronchiectasis [10] to prevent future respiratory morbidity. However, these guidelines also envision watchful waiting for certain patients with chronic cough when specific symptoms and/or signs (specific cough pointers) are absent [5].

In contrast, the British Thoracic Society utilizes a threshold of eight weeks duration [11], which is the threshold used to define chronic cough in adults [12]. However, the guideline also encourages individualized management by including the caveat that a "relentlessly progressive prolonged acute cough [>three weeks]…may warrant investigation before eight weeks" [11].

Specific cough — Specific cough refers to a chronic cough that is ultimately attributable to an underlying physiologic cause (which is usually but not always of pulmonary origin). The most important causes of specific cough in children are outlined in the table (table 1). In most cases, an initial work-up consisting of a focused history, physical examination, chest radiography, and spirometry will identify features that serve as clues to the underlying disease (sometimes termed "specific cough pointers") (table 2). These include classically recognized cough sounds and characteristics, such as whooping, honking, or absence of cough during sleep, which point to a specific cause (these are sometimes termed "classical" cough characteristics) (table 3).

Nonspecific cough — Conversely, a "nonspecific" cough is defined as a chronic cough that does not have an identifiable cause after a reasonable evaluation. A chronic cough is more likely to be nonspecific if it is dry and there are no abnormalities identified on initial evaluation (ie, no "specific cough pointers" and likely to resolve without medications (table 2)).

CAUSES — There are many causes of cough. The majority originate in the lungs, but there are also nonpulmonary etiologies, some proven and others controversial. The causes are presented in the table (table 1) and further discussed in another topic review. (See "Causes of chronic cough in children".)

Etiologies of chronic cough in children are different from those of adults and, thus, clinicians should use pediatric-specific cough guidelines [5,6]. As an example, gastroesophageal reflux and upper airway cough syndrome (formerly known as postnasal drip syndrome) are thought to be common causes of chronic cough in adults but are controversial as common causes of chronic cough in children [5,13,14].

EPIDEMIOLOGY

Prevalence — The reported prevalence of chronic cough varies depending on many factors, including the definition of chronicity used in the studies, age of children, and how cough is determined. Epidemiology questionnaires on isolated chronic cough are generally poorly reproducible; reproducibility is generally better for chronic cough associated with other symptoms [15]. The reporting of cough severity depends upon the population characteristics [16]. As an example, the accuracy of cough reporting by urban Australian parents is excellent compared with an objective measure [17]. By contrast, the accuracy of parent-reported cough is worse if the parent smokes [18], is a member of an indigenous Australian population living in a remote area [19], or if the cough is nocturnal [20].

Because of the considerations outlined above, the accuracy and generalizability of prevalence studies are limited and it is often impossible to compare results across different studies. In one community-based study in which persistent cough was defined as >3 weeks duration, the prevalence in children aged 5 to 11 years was approximately 10 percent [21]. The study found that reported persistent cough was higher in Australia than Nigeria [21], possibly reflecting a reporting bias. Another questionnaire-based study reported a prevalence of chronic cough of 22 percent among preschoolers without colds [22].

Burden — The burden of chronic cough is substantial for children [7] and their parents or caregivers [4,8]. Their quality of life is impaired [4,7], and in affluent countries, parents often seek recurrent consultations until their child's cough resolves [8]. An Australian multicenter study found that 75 percent of parents sought ≥5 doctor visits (14 percent sought ≥15 consultations) for their child's cough [8]. The study also reported that the children's mean generic pediatric health-related quality-of-life score (using the PedsQL tool) was in the realm of children with other chronic illness (eg, cardiac, diabetes) [8]. Their quality of life significantly improved when the cough resolved [8]. At a societal level, expenses for clinician visits, medication expenses, and absenteeism are substantial. Thus, it is not surprising that various prescription and nonprescription medications are reported to be widely used (sometimes inappropriately) [23,24] in an attempt to ameliorate cough. The use of these medications may result in adverse events (eg, Cushing's disease and accidental poisoning) [24,25].

PATHOPHYSIOLOGY — An intact cough mechanism (afferent and central limb) and apparatus (efferent limb) is essential to maintain respiratory health. Cough also enhances mucociliary clearance in both healthy individuals and those with lung disease [26].

Normal cough mechanisms — The cough pathway, a complex reflex arc, can be simplified into the afferent, central, and efferent pathways (figure 1) [27-30]. The cough pathway may be self-initiated (eg, voluntary cough) or a reflex mechanism. The central pathway involves the nucleus tractus solitarius, the brainstem respiratory network (sometimes known as the "cough center") [31], the subcortex, and cortex. The afferent arm (from cough stimulus to the central pathway) and efferent arms (from central pathway to respiratory muscles, larynx, and pelvic muscles) are likely to be influenced by a multidirectional feedback loop mediated through the cough center [31].

Cough has three mechanical physiologic phases: inspiratory, compressive, and expiratory [27,32], although some include a preceding irritation phase. The duration of inspiratory, compressive, and expiratory phases and magnitude of the respiratory muscles are regulated by the brainstem network [27]. This same network also regulates these respiratory muscles for breathing [27].

The entire respiratory structure and system undergoes a maturation process from infancy to adulthood [33,34]. The cough reflex is weak in premature infants and develops with maturity [35]. Whilst the laryngeal chemoreflex in young infants result in swallowing, apnea, and laryngeal closure, cough becomes increasingly prominent with maturation [35]. Exactly when the cough reflex is fully matured is unknown, but critical windows of exposure in utero and early childhood have long been appreciated [36]. The immune system (which has its own developmental physiology) affects the degree and frequency of respiratory infections and hence presentations to doctors for cough. Other general physiology that influences the pathophysiology and management of chronic cough in children compared with adults includes differences in cognitive function and ability to self-express.

The differences between the pathophysiology of chronic cough in children compared with adults are exemplified in child-specific diagnoses such as protracted bacterial bronchitis (PBB). Thus, it is not surprising that different strategies are required to evaluate and manage chronic cough in children compared with adults, manifested in different evidence-based guidelines [5].

Pathologic cough

●Upregulation of the cough reflex, resulting in increased cough – As the stimulus for cough is multimodal, a variety of stimuli or combination thereof can result in increased coughing. In children, the most common cause of cough is respiratory infections. The infections are associated with inflammatory mediators that stimulate the various nociceptors in the airway walls, thus activating the cough pathway. Studies in animal models suggest that the increased cough can be related to neuromodulation; some viruses likely induce plasticity in the cough neural pathways, sometimes causing the cough to persist after the infection is cleared [37]. While the cough reflex cannot be stimulated from some areas like the nasal mucosa, diseases in the area may sensitize the cough reflex in animals and in adults [38]. Other possible mechanisms include enhancement of cough sensitivity through intrathoracic pressure effects, suggesting that cough may also perpetuate the chronic cough cycle through the intrathoracic pressure effects generated by cough [39]. The understanding of cough receptors and targeting it with therapeutic agents in adults with chronic cough is expanding [38], but there are no pediatric data.

Human neurocellular data describing mechanisms for increased coughing are scarce. The mechanisms underpinning chronic cough have been likened to that of chronic neuropathic pain [40], and increased neurogenic markers have been described in children with increased cough receptor sensitivity [41]. In children, cough sensitivity is increased in conditions where chronic cough occurs and the sensitivity tends to normalize when cough resolves [42-44]. Thus, the concept of "cough hypersensitivity" as a diagnostic entity of chronic cough is not relevant in children and should be avoided. In contrast, this concept is likely beneficial in adults, where idiopathic cough is common and likely involves mechanisms similar to neuropathic pain, in which there are considerable differences between children and adults [45]. Similarly, the concept of "urge to cough" widely described in the adult literature is also not an issue in young children.

●Downregulation of the cough reflex, resulting in ineffectual cough – A variety of abnormalities that interfere with the cough reflex arc result in ineffectual cough (figure 1). An abnormal afferent neural pathway, usually associated with neurodevelopmental abnormalities, can result in decreased airway sensitivity and consequent silent aspiration [46], which can ultimately lead to chronic lung disease.

The efferent limb of the cough pathway may be impaired by:

•Abnormalities in the airway (larynx or trachea), including tracheostomy or vocal cord palsy. This is because closure of the larynx is also required to generate intrathoracic pressure prior to the expulsive phase of cough.

•Abnormalities in the pump mechanism of respiration, ie, muscle (chest and diaphragm), and chest wall abnormalities. This includes weak respiratory muscles (eg, due to neuromuscular disease), which reduce both inspiratory and expiratory volumes and flow.

•Reduction in airway patency (eg, tracheomalacia, trachea stenosis) because this reduces expiratory airflow.

Diagnosis and management of ineffectual cough is an important component of respiratory care in people with the above abnormalities, particularly those with neurodevelopmental or neuromuscular abnormalities. (See "Aspiration due to swallowing dysfunction in children" and "Respiratory muscle weakness due to neuromuscular disease: Management", section on 'Respiratory adjunctive therapy'.)

DIAGNOSTIC APPROACH

General goals — Evaluation and management of chronic cough in a child is informed by the following key questions:

●Does the child have an underlying chronic lung disease that requires further investigations and/or referral?

●Are therapies/medications indicated?

●Are there modifiable factors that exacerbate the cough, such as exposure to tobacco smoke?

●What is the psychosocial impact of the cough on the child and family (eg, quality of life and function), and what are their expectations for treatment and outcomes?

Algorithmic approach — Answering these questions require a systematic approach to evaluation, management decisions, and follow-up. Our approach is outlined in two algorithms (algorithm 1 and algorithm 2) and detailed in the following sections. This algorithm is similar to the guideline from the American College of Chest Physicians [5,47], which was shown to improve clinical outcomes in multicenter randomized trials [48] and is the best validated among several similar algorithmic approaches [49,50]. Remaining uncertainties about this and other algorithms are whether testing or management should depend on the duration and/or severity of chronic cough [47,49] and the age at which adult guidelines should be used.

The approach outlined below is designed for use in all patients up to 18 years of age. However, it is also reasonable to use adult guidelines and algorithms for adolescents 15 years and older, consistent with the American College of Chest Physicians guidelines, since the etiologies of chronic cough in adolescents are fairly similar to those in adults. (See "Causes and epidemiology of subacute and chronic cough in adults" and "Evaluation and treatment of subacute and chronic cough in adults".)

EVALUATION — The systematic evaluation of a child with chronic cough starts with a targeted detailed history, physical examination, chest radiograph, and (if the child is able) spirometry [47,49]. The information obtained is then used to categorize the cough as specific (ie, underlying disease likely present) or nonspecific (no evidence of an underlying disease, so the cough is more likely to resolve spontaneously) [51]. However, this initial evaluation does not definitively distinguish specific cough from nonspecific cough, as depicted by area of overlap in the figure (figure 2). Therefore, a provisional diagnosis can be made, but ongoing evaluation is required, and the final diagnosis depends upon the clinical course (algorithm 1).

History — The history focuses on identifying symptoms or risk factors that predict a specific cause of cough.

●Specific cough pointers – Key clinical features that predict a specific cause of cough during the initial evaluation (warning signs or "pointers") include:

•Wet/productive chronic cough

•Wheezing or dyspnea

•Onset after an episode of choking (even if days or weeks prior)

•Neonatal onset of symptoms

•Any other associated medical conditions (cardiac, neurologic, autoimmune or immunodeficiency, or suspicion thereof)

A full list of specific pointers and their diagnostic implications is in the table (table 2).

●Cough trajectory – Determining whether the cough is progressive (getting worse), subsiding, or static will guide whether further intervention is required. Progressive chronic cough suggests the presence of an underlying illness, while a cough that is subsiding is usually nonspecific cough and ultimately resolves with time [52]. A cough that is initially static (neither progressive nor subsiding) may subsequently follow either of these trajectories.

●Medications – A patient's response to prior therapy may yield diagnostic clues regarding the cause of chronic cough. As examples, previous response to antihistamines suggests a component of rhinitis and postnasal drip, while a response to inhaled bronchodilators suggests possible asthma. However, the history of a prior response to medication needs to be interpreted with great caution since cough symptoms can spontaneously remit. Previous response to antibiotics and asthma therapies should be interpreted with special caution as the antibiotics may have been prescribed for a self-limited viral illness and the cough may have resolved spontaneously regardless of treatment.

Any current medications also should be reviewed carefully; angiotensin-converting enzyme (ACE) inhibitors are well-established causes of chronic cough. A history of exposure to medications should be sought, particularly cytotoxic drugs, which are associated with interstitial lung disease, or immunosuppressive medications such as tumor necrosis factor inhibitors, which are associated with increased risk for pulmonary infections, including with fungal and mycobacterial pathogens. (See "Tumor necrosis factor-alpha inhibitors: An overview of adverse effects", section on 'Infection'.)

●Social history and impact of cough – Exploration of parental expectations, fears, and the impact of the cough on the child and family is valuable when managing a child with a chronic cough [5]. The symptoms of chronic cough may cause stress and concern for patients and their families, which resolve when the cough is managed [4]. Acknowledging the effects of the cough and other parental stress and fears may facilitate communication and appropriate counseling. Clinicians should be cognizant that information available from the internet may provide incorrect advice on the home management of cough in children [53].

The psychobehavioral context and triggers of the cough should be explored. A cough that disappears when the child is distracted and can be voluntarily suppressed suggests possibility of tic cough (habit cough). A child that has disproportionate anxiety about the seriousness of the cough or other symptoms suggests the possibility of somatic cough disorder (psychogenic cough), especially if there is a history of generalized anxiety or a history of other somatic symptoms. (See "Causes of chronic cough in children", section on 'Tic cough (habit cough) and somatic cough disorder (psychogenic cough)'.)

Prescription rates of medications to children with respiratory infections are known to be influenced by the expectations of the parent(s) or caregivers, and by the clinician's perception of those expectations [54], although there are no specific data relating to chronic cough in children. Acknowledging and explicitly discussing parents' and child's expectations can facilitate a conversation about risks and benefits and support objective decision-making.

●Past medical history – Information relevant to chronic cough includes:

•Neonatal history – Low birth weight and/or premature neonates are at risk for asthma-like disorders, bronchiectasis, and recurrent infections. In addition, prematurity and neonatal respiratory distress syndrome are precursors for bronchopulmonary dysplasia, which may cause persistent respiratory symptoms in children and adolescents. (See "Complications and long-term pulmonary outcomes of bronchopulmonary dysplasia" and "Risk factors for asthma", section on 'Pre- and perinatal factors'.)

•Respiratory viral infection – Cough that begins with a respiratory viral infection suggests a postinfectious syndrome, which is a common cause of nonspecific chronic cough. The cough is dry, has no specific cough "pointers" (table 2), may persist for weeks after the other viral symptoms have resolved, and gradually resolves without intervention [51]. (See "Causes of chronic cough in children", section on 'Viral infections'.)

•Previous hospitalizations and pulmonary disease – The history should be reviewed for previous pulmonary infections or other diagnoses, chest radiographs, or other tests. Recurrent pneumonia is a specific cough pointer (table 1). Recurrent or unresolving pneumonia in one lobe or segment of the lung may also be caused by obstruction or anatomic abnormality in that airway and warrants further investigations (usually bronchoscopy and chest scans) (see "Congenital anomalies of the intrathoracic airways and tracheoesophageal fistula"). A history of recurrent pneumonias involving multiple lobes should raise concern for conditions associated with bronchiectasis (eg, cystic fibrosis, primary ciliary dyskinesia and autoimmune disease, immune deficiency, aspiration, tracheoesophageal fistula). Severe infection caused by pertussis or adenovirus has been associated with the subsequent development of bronchiectasis, bronchiolitis obliterans, and chronic lung disease. (See "Approach to the child with recurrent infections" and "Bronchiectasis in children: Clinical manifestations and evaluation".)

•Other medical problems – Children with chronic cough who have a history of underlying chronic illnesses (past or ongoing), such as autoimmune disease, cancers, or immunodeficiency, require further investigation to determine the cause. Those with autoimmune disease are at risk for interstitial lung disease. Those with cancer and/or immunodeficiency are at risk for a chronic or opportunistic pulmonary infection, necessitating lower airway sampling in most situations.

A history of purulent sinusitis and suppurative ear disease raises the likelihood of having primary ciliary dyskinesia, while cardiac abnormalities raises likelihood of coexistent tracheomalacia, cardiac failure, and/or primary ciliary dyskinesia.

●Environmental exposures – Environmental factors that may trigger or exacerbate cough include environmental allergens (dust, pets) or exposure to tobacco smoke, wood smoke, or other airborne contaminants.

The location of the child's home and local epidemiology can inform the diagnostic considerations, especially with respect to endemic fungal and parasitic infections such as parasites (eg, Toxocara infection) [55]. One study found that 40 percent of Chinese children had serologic evidence of exposure to Toxocara [55]. Histoplasmosis is associated with exposure to birds and bats and is particularly common in the Ohio and Mississippi river valleys of the United States. Echinococcosis is associated with exposure to dogs and sheep, typically in endemic rural areas. Residents of the southwestern United States, northern Mexico, and parts of Central and South America are at risk for coccidioidomycosis. (See "Pathogenesis and clinical manifestations of disseminated histoplasmosis" and "Echinococcosis: Clinical manifestations and diagnosis" and "Primary pulmonary coccidioidal infection".)

Physical examination

General examination — The physical examination should evaluate for the following signs of an underlying chronic disease that might explain the chronic cough:

●Appearance

•Poor growth or thinness – Suggests chronic illness.

•Dysmorphism, developmental delay, or other evidence of a genetic syndrome – Suggests possibility of anatomic abnormalities or swallowing dysfunction with aspiration.

●Skin

•Eczema – Suggests atopic disease.

•Other rashes – Evidence of a recent rash raises the possibility of a triggering viral infection, or impetigo, which is more common in patients with an immune deficiency.

●Head and neck

•Allergic "shiners," swollen nasal turbinates, nasal obstruction, nasal polyps, or allergic nasal crease – Suggests allergic disease (a controversial cause of chronic cough).

•Lymphadenopathy – Suggests immunodeficiency, malignancy, or chronic infection.

●Ears

•Tympanic membrane scarring or otorrhea – Suggests the possibility of primary ciliary dyskinesia.

•Objects or disease in the ear canal – Suggests the possibility that cough is triggered by the otogenic reflex (a rare cause of chronic cough).

●Mouth

•Hoarseness – Suggests aspiration or vocal cord dysfunction.

•Tonsillar hypertrophy or pharyngeal cobblestoning – Suggests allergic disease.

•High-arched or cleft palate – Suggests possibility of congenital anomalies that might be associated with swallowing disorders or aspiration.

●Heart

•Abnormal heart location – Dextrocardia is present in 50 percent of patients with primary ciliary dyskinesia.

•Abnormal heart sounds or pulses – Suggests possible congenital cardiovascular anomalies, which can cause chronic cough due to airway compression, pulmonary edema, or arrhythmia.

●Abdomen

•Abnormal liver size or texture – Suggests chronic liver disease, which could be due to cystic fibrosis.

•Splenomegaly – Suggests portal hypertension, including due to cystic fibrosis-related liver disease. Splenomegaly also may be caused by chronic infection, malignancy, storage diseases, or hemoglobinopathies.

•Situs inversus (spleen on the right and liver on the left) – Situs inversus is present in 50 percent of patients with primary ciliary dyskinesia but may be difficult to perceive on abdominal examination.

●Genitourinary

•Rectal prolapse – Suggests cystic fibrosis.

●Extremities

•Edema – Suggests cardiac disease.

•Cyanosis or digital clubbing – Suggests bronchiectasis or interstitial lung disease.

Chest examination — A detailed chest examination includes:

●Cough characteristics – Observe for the characteristics of the cough (sound and pattern) first if it occurs spontaneously during the interview. If not present, ask the child to cough (ie, requested cough), which sometimes requires coaxing and imitation in young children. Also, ask the parent(s) if they have a recording of their child's cough. These cough characteristics are important factors in determining the diagnostic approach. (See 'Diagnostic approach' above.)

●Inspection – Observe for increased work of breathing, retractions, accessory muscle use, and chest wall hyperinflation or deformity. Chest wall abnormalities such as hyperinflation, bell-chest, and presence of Harrison sulci indicate presence of an underlying disease, usually obstructive airways disease. Chronic cough in association with other chest wall anomalies reflective of poor chest wall expansion (eg, kyphoscoliosis, Poland syndrome) may be indicative of poor cough clearance and thus increased risk of bronchiectasis. Tracheomalacia is associated with pectus excavatum.

●Auscultation – Auscultate for abnormal breath sounds, including reduced intensity, asymmetry, wheezing, stridor, or crackles. Even in the absence of a history of wheeze, clinicians should pay close attention to any evidence of this finding on physical examination.

•A unilateral wheeze is always abnormal and may represent an inhaled foreign body or other causes of airway obstruction.

•Polyphonic wheezing (ie, many sounds that start and stop at different times) with cough is typical of asthma; the wheezing occurs on expiration and sometimes also on inspiration. Many children with asthma are also atopic and exhibit signs of eczema and/or allergic rhinitis and conjunctivitis [56]. Other causes of polyphonic wheezing in children with chronic cough include bronchiolitis obliterans, bronchiectasis (cystic fibrosis, allergic bronchopulmonary aspergillosis, primary ciliary dyskinesia), bronchopulmonary dysplasia, heart failure, immunodeficiency, and aspiration.

•Monophonic wheezing (single pitch wheeze) should always raise suspicion of large airway obstruction caused by foreign body aspiration or malacia and/or stenosis of the central airways. The wheeze of large airway obstruction can often be heard without a stethoscope. In addition, vascular rings, lymphadenopathy, and mediastinal tumors can cause extrinsic large airway obstruction. Tuberculosis should always be considered in a child with a monophonic wheeze, particularly in areas where the disease is prevalent. In children with tuberculosis, wheezing may be caused by several mechanisms: extrinsic compression from hilar lymphadenopathy, tuberculoma, or secondary bronchiectasis. (See "Evaluation of wheezing in infants and children" and "Assessment of stridor in children".)

Chest radiograph — A chest radiograph should be obtained as part of an evaluation in children with chronic cough [5]. An abnormal chest radiograph (other than perihilar bronchial thickening (image 1)) is an important pointer for specific cough [47,57]. An abnormal chest radiograph is a very reliable indicator that there is a specific cause of cough; in one study, all patients with an abnormal chest radiograph had a specific cause of cough [51]. However, a normal chest radiograph does not exclude specific cough or an underlying disease process.

Occasionally, but rarely, a chest radiograph provides definitive diagnosis. The following radiographic appearances provide supportive data for specific diagnoses and may be helpful in determining the sequence of subsequent diagnostic tests.

●Normal chest radiograph – Children with nonspecific cough, tic cough (habit cough), or somatic cough disorder have normal radiographs (image 1). A normal radiograph is reassuring but does not exclude a variety of pathologic conditions, including foreign body, asthma, early cystic fibrosis, and bronchiectasis.

●Hyperinflation – Unilateral lung hyperinflation may suggest presence of foreign body aspiration (image 2); both inspiratory and expiratory posterior-anterior chest radiographs should be obtained if foreign body aspiration is suspected. However, a bronchoscopy is still indicated if foreign body aspiration is suspected even if the chest radiograph is normal. Bilateral hyperinflation is most often caused by asthma but can also be found in other chronic respiratory illnesses such as cystic fibrosis, aspiration, primary ciliary dyskinesia, bronchiolitis obliterans.

●Bilateral peribronchial accentuation with one or more focal consolidated infiltrates – These findings are also seen with diffuse airway inflammation, including asthma, protracted bacterial bronchitis (PBB), or disorders of impaired airway clearance (eg, ciliary dyskinesia and cystic fibrosis). Infiltrates are most commonly seen in the right middle lobe.

●Right middle lobe changes – This distribution is frequently seen in patients with obstructive airway disease, eg, asthma (image 3); it reflects atelectasis and appears in the right middle lobe because the collateral ventilation in this lobe is underdeveloped in comparison with the other lobes. Right middle lobe syndrome may also be caused by other causes, eg, immunodeficiency [58].

●Linear atelectasis, dilated, and thickened airways (ie, tram-tracking or parallel lines, ring shadows on cross section) – This appearance is consistent with bronchiectasis. Other findings in bronchiectasis are irregular peripheral opacities that may represent mucopurulent plugs (image 4). (See "Causes of chronic cough in children", section on 'Bronchiectasis'.)

●Lung or lobar collapse – This finding is associated with chronic airway obstruction that may be intraluminal, such as foreign body or mucus plugs (image 5), or extraluminal from a mass causing airway compression.

●Mediastinal widening – This finding in children suggests chronic infection, marked lymphadenopathy, or neoplasm. In an infant or young child, the most common cause of mediastinal widening is a normal thymus (image 6). If the structure is not readily identified on plain film, it can be further characterized by ultrasound, computed tomography (CT), or magnetic resonance imaging (MRI).

●Cardiac abnormalities – A large or abnormally shaped heart suggests a primary cardiac defect. A large pulmonary artery may be caused by primary or secondary pulmonary hypertension. Right-sided arch is associated with vascular sling or ring and tracheomalacia (image 7).

●Abnormalities of the pleura – Chronic pleural effusion and pleural thickening all suggest underlying lung disease as the cause of the chronic cough.

Spirometry — When appropriate for the child's age (>3 years in a pediatric laboratory and >6 years in other laboratories), spirometry should be obtained as part of an evaluation in children with chronic cough [5,57]. Normal spirometry does not exclude disease, but abnormalities in spirometry values indicate specific cough (table 2); in one study, all patients with abnormal spirometry were ultimately found to have a specific cause of cough [51]. An obstructive pattern in the spirometry indicates obstructive airway disease such as asthma or suppurative lung disease, while a restrictive pattern reflects an interstitial or chest wall restrictive processes. However, suboptimal effort on the part of the child will also result in a restrictive picture; thus, spirometry should be conducted by a technician proficient in testing children.

If an obstructive pattern is seen on the expiratory flow-volume loop, airway reversibility should be assessed by measuring forced expiratory volume in one second (FEV₁) before and after inhalation of beta₂-agonist. A positive response to bronchodilators establishes the presence of airway reactivity and is suggestive of asthma but does not rule out other disorders. As an example, children with cystic fibrosis often respond to bronchodilators. Abnormalities on the inspiratory loop correlate with extrathoracic airway obstruction, more often associated with stridor than cough. (See "Overview of pulmonary function testing in children".)

INITIAL CATEGORIZATION

Specific cough pointers present — "Specific cough" is defined as a chronic cough that is ultimately attributable to an underlying physiologic cause (table 1) (see 'Specific cough' above). The presence of specific cough "pointers" (table 2) help to narrow the diagnostic possibilities and signify therapies and/or further specific testing or referral (algorithm 1) [47,49]. However, these pointers have varying degrees of specificity, sensitivity, likelihood ratios, and predictive values [51]. The sequence of evaluation for these disorders is informed by the age and presenting features of the child. Important initial considerations include:

Suspicion of asthma — A provisional diagnosis of cough-dominant is suggested by airflow limitation on spirometry, which reverses to treatment with bronchodilators [5]. Tests for airway hyperresponsiveness may be helpful in older children but are not always diagnostic [59]. Some experts have recommended using exhaled nitric oxide levels (fractional excretion of nitric oxide [F_eNO]) as a biomarker. F_eNO values ≥25 ppb are supportive of asthma [60], but its use and levels of defining abnormality remain controversial [61]. Supportive evidence includes wheezing and a chronic dry and paroxysmal cough that is triggered by exercise, cold air, sleep, or allergens. However, exercise often exacerbates any type of cough even in children without asthma, and thus, this symptom is nonspecific [62]. Although children with poor asthma control may have nocturnal cough, most isolated nocturnal cough in children is not asthma [63]. A history of eczema, recurrent wheezing, or a family history of atopic disease also supports the possibility of asthma. Asthma may also present with recurrent right middle lobe atelectasis, thought to be due to relatively poor collateral ventilation in the right middle lobe (image 3).

Ideally, confirmatory tests for asthma should be undertaken but may not be possible in young children. If clinical evidence supports the possibility of asthma, the next step is a trial of asthma treatments, which is detailed in a separate topic review (see "Asthma in children younger than 12 years: Initial evaluation and diagnosis" and "Asthma in children younger than 12 years: Overview of initiating therapy and monitoring control"). The patient should be reevaluated in two to four weeks and further evaluation performed if the chronic cough does not respond to the treatment for asthma (algorithm 1).

Wet or productive cough — The symptom of a chronic wet cough, with or without productive purulent sputum, is particularly important. This symptom is always pathologic and warrants treatment and/or investigations for a chronic endobronchial infection.

Protracted bacterial bronchitis (PBB) is an important cause of isolated wet-moist cough in a child [57,64]. It is usually diagnosed based on the presence of an isolated chronic wet-moist cough in a child who otherwise appears well and absence of symptoms, signs, or laboratory evidence suggestive of an alternative cause of the cough, with resolution of the cough after antibiotic treatment [57,65]. Children with suspected PBB should be treated with antibiotics, selected based on the likelihood of the lower airway infection, local antibiotic resistance pattern, and the child's tolerance to antibiotic class [5]. Treatment with amoxicillin-clavulanate is usually effective. (See "Causes of chronic cough in children", section on 'Protracted bacterial bronchitis'.)

If the cough does not resolve with antibiotic treatment or if it recurs frequently, other causes of wet/productive cough should be considered and investigated. These include:

●Retained foreign body (see 'Suspicion of retained foreign body' below)

●Recurrent aspiration (in children with neurologic disorders and upper airway abnormalities) (see "Aspiration due to swallowing dysfunction in children")

●Other chronic suppurative lung diseases (eg due to cystic fibrosis, primary ciliary dyskinesia, anatomic abnormalities, or immunodeficiency) (see "Causes of chronic cough in children", section on 'Chronic endobronchial suppurative disease')

Suspicion of retained foreign body — The possibility of a retained foreign body should be considered, particularly if the child is young (age <5 years) and the cough began suddenly after an episode of choking. The choking may have occurred weeks before presentation. Associated symptoms may include wheezing (particularly if monophonic) or halitosis, and the cough may be either wet- or dry-sounding. The chest radiograph may be normal or show signs of unilateral hyperinflation if the location of the foreign body is in the bronchial system, which is the most common location (see 'Chest radiograph' above). Bronchoscopy permits definitive diagnosis and foreign body removal. (See "Airway foreign bodies in children".)

Classically recognized cough sounds — Specific cough pointers also include several classically recognized cough sounds (table 3), including:

●Barking or brassy cough suggests a process in the trachea or more proximal airways, such as airway malacia (with or without vascular and other causes of trachea compression) [66], laryngotracheobronchitis, spasmodic croup, or, rarely, foreign body. Staccato cough in young infants suggests the Chlamydia trachomatis infection.

●Cough that is honking ("Canadian goose-like") and disappears when asleep has traditionally been interpreted as suggesting a tic cough or somatic cough disorder (also known as habit cough and psychogenic cough, respectively), but this is not a reliable distinguishing characteristic [5,67]. For tic cough, key diagnostic features are suppressibility, distractibility, suggestibility, variability, and the presence of a premonitory sensation. For somatic cough disorder, key features are disproportionate thoughts and anxiety about the seriousness of symptoms. (See "Causes of chronic cough in children", section on 'Tic cough (habit cough) and somatic cough disorder (psychogenic cough)'.)

●Presence of an inspiratory "whoop" and paroxysmal cough suggests infection with pertussis or parapertussis; this characteristic cough can be retriggered by subsequent upper respiratory illness. Whoop is often absent in infants with pertussis (such infants often have apnea) and in older children who have had previous pertussis vaccination. Confirmation of the diagnosis in the chronic cough phase can be difficult to ascertain as cultures are typically negative after the fourth week of illness and convalescent-phase serology can be difficult to interpret unless acute-phase titers were also obtained [5]. (See "Pertussis infection in infants and children: Clinical features and diagnosis", section on 'Clinical features'.)

Other specific cough pointers — For patients with other symptoms suggesting a specific cause of cough at the initial evaluation (table 2), further evaluation and management depends upon the suspected disorder. Important considerations include:

●Neonatal onset of symptoms and/or history of recurrent pneumonia – Suggests congenital malformation of the airway, immune dysfunction, or condition predisposing to bronchiectasis (eg, primary ciliary dyskinesia (image 5))

●Neurologic or developmental abnormalities, or feeding difficulties – Suggests recurrent aspiration

No specific cough pointers (nonspecific cough) — A provisional diagnosis of nonspecific cough can be made if no features of specific cough are identified on the initial evaluation (algorithm 1 and table 2). By definition, nonspecific cough is dry and nonproductive, and chest radiograph and spirometry (if performed) are normal. Management is discussed below. (See 'Nonspecific cough' below.)

SUBSEQUENT FOLLOW-UP AND MANAGEMENT — In the management of chronic cough, the key principles are:

●Targeted treatment for the suspected or confirmed cause of the cough, as determined by a systematic evaluation

●Identifying and mitigating exacerbation factors, such as exposure to environmental tobacco smoke

●Defining and discussing the expectations of, and the effect of the cough on, the child and parent(s)

Specific cough — Treatment of children with a specific cough is directed at the suspected cause. In some cases, referral to a specialist for further investigation and management is appropriate. (See 'Additional investigations' below.)

It is important to reassess the child periodically to determine whether treatment has been optimized and to confirm the diagnosis. Some children may have more than one cause of specific cough, such as asthma with a congenital abnormality of the airway.

Nonspecific cough — Management of chronic nonspecific cough involves (algorithm 2):

●Watchful waiting – For most patients with nonspecific cough (dry cough with no specific pointers), the first step is a period of watchful waiting for two to four weeks [5]. Most cases of nonspecific cough are related to a viral illness (postviral cough) [8]; the underlying physiology is thought to be due to increased sensitivity that gradually resolves over time (two to four weeks) [40] (see 'Pathologic cough' above). One study showed that most such children are very unlikely to have underlying disease (negative likelihood ratio 0, 95% CI 0.00-0.03) [51]. The child and family can be reassured and should be followed and reevaluated in two to four weeks if the chronic cough does not resolve or if new symptoms emerge.

●Medication trials – Some cases of nonspecific cough probably represent a cough-dominant asthma; this is more likely if the child has other signs of allergic disease or a family history of asthma. Therefore, if the cough is troublesome and if there is clinical evidence supporting asthma or if other specific causes have been excluded, it is reasonable to perform an empiric trial of bronchodilators (short-acting beta₂-agonists) or low-dose inhaled corticosteroids (<400 micrograms budesonide equivalent daily). High doses of inhaled corticosteroids should be avoided. If empiric treatment is undertaken, the trial should be time-limited (two to four weeks) and the medication should not be given chronically unless the underlying diagnosis is firmly established. (See "Asthma in children younger than 12 years: Overview of initiating therapy and monitoring control".)

Gastroesophageal reflux disease (GERD) is probably an unusual cause of chronic cough in otherwise healthy children. Nonetheless, in selected children with nonspecific cough and other symptoms suggestive of GERD (such as recurrent regurgitation, dystonic neck posturing in infants, and heartburn in older children), it is reasonable to do an empiric trial of treatment for GERD for four weeks, using acid-suppressing medications such as a proton pump inhibitor (PPI) [14]. Longer duration of PPIs should not be used without objective confirmation of GERD [14]. Similarly, a guideline for adults advises a trial of PPIs only in patients with heartburn or regurgitation symptoms [68], although in the past, PPIs were often used as empiric treatment for adults with chronic cough. In children with growth failure, GERD and eosinophilic esophagitis should also be considered as possible associated diseases [69].

We do not use codeine or other antitussive agents (eg, dextromethorphan) for the treatment of cold-related cough in children [5]. They have potential harms with no proven benefit. (See "The common cold in children: Management and prevention", section on 'Unproven therapies'.)

●Assessing the response to therapy – When medications are trialed, reassessment of the child is recommended in two to four weeks, which is the "time to response" for most medications [5]. However, ascribing etiologies for the cough based on treatment trials has an inherent high risk of bias related to the placebo and "period effects" (the natural resolution of cough over time, which is evident in cough-related intervention studies [70]). Thus, when the cough resolves, a trial of ceasing the medications should be undertaken. If the chronic cough relapses, the medication could be recommenced, typically with further investigation to confirm the provisional diagnosis of asthma or GERD.

If formally assessing response to therapy, child-specific assessment tools are required in children [17,71] as the adult assessment tools have limited applicability in children. Several other methods are available to objectively assess cough frequency, cough severity, and its response to therapy.

Other management considerations

●Family expectations and counseling – The significant burden of chronic cough in children for families has been well documented and should always be considered when consulting [4]. A number of studies have documented the frequent consultations for children with chronic cough, with three-quarters of families consulting a doctor more than five times [4,8]. In addition, several studies report reduced quality of life that normalizes as the cough resolves [8,72]. This significant burden and worry for parents/caregivers should be considered, and they should be counseled appropriately about the prognosis and expected course for their child, in particular the excellent prognosis and outcome for children with chronic nonspecific cough [5]. Among other issues, a worried parent/caregiver may administer nonspecific medications to the child, which may not be suitable due to lack of efficacy and safety concerns [73,74].

●Cough suppressants – Over-the-counter cold and cough suppressants are not recommended for the treatment of chronic cough in children [5,11]. There is no evidence that they improve symptoms in this age group, and significant safety concerns have been raised about their use in young children [73,74]. The US Food and Drug Administration has issued a public health advisory against the use of these drugs for young children [75]. In addition, we recommend against the use of opioid drugs such as codeine that may be effective in suppressing cough as evidence is limited, and these drugs are associated with serious side effects and the potential for abuse [5,76,77].

●Avoidance of tobacco smoke – There is evidence that environmental exposure to tobacco smoke increases respiratory infections and thus the risk for chronic cough in children [5,78]. Household members who smoke should be counseled to stop smoking and guided to support for smoking cessation. Until the adult is able to completely stop smoking, they should not smoke in the presence of children, including in a car, even with the window open. The clinician should also ask adolescent patients about the possibility of tobacco use and provide strong support for smoking cessation. (See "Secondhand smoke exposure: Effects in children" and "Overview of smoking cessation management in adults".)

●Alternative therapies – In the absence of specific treatments for the chronic symptom, many families will try alternative therapies, including honey and echinacea. Clinical outcomes of these approaches are described separately. (See "The common cold in children: Management and prevention", section on 'Cough' and "The common cold in children: Management and prevention", section on 'Unproven therapies'.)

ADDITIONAL INVESTIGATIONS — Additional investigations depend upon findings from the history, examination, chest radiograph, and spirometry, which determine the diagnosis (or diagnoses) being considered (table 1 and table 2) [5]. Most of these investigations are usually undertaken by specialists.

●Flexible bronchoscopy – The primary indication for urgent bronchoscopy in children with chronic cough is for suspected foreign body aspiration. This diagnosis should always be considered, even in the absence of obvious findings, especially in younger children and/or if there is a history of a choking episode preceding the onset of symptoms (see "Airway foreign bodies in children"). In addition, bronchoscopy usually is appropriate for children with a provisional diagnosis of PBB whose symptoms do not improve after a four-week course of appropriate antibiotics [5].

Bronchoscopy is also valuable in the evaluation of suspected airway malacia, tracheoesophageal fistula, or stenosis. Patients with presumed infectious etiologies in whom a sputum sample is not obtained or yields negative results can be evaluated with flexible bronchoscopy to perform bronchoalveolar lavage for bacterial (picture 1), fungal, and mycobacterial cultures. Cilia brushings can also be taken for patients with suspected ciliary dyskinesia. (See "Primary ciliary dyskinesia (immotile-cilia syndrome)", section on 'Diagnostic evaluation'.)

●Esophageal pH or multichannel intraluminal impedance (MII) monitoring – It remains controversial whether gastroesophageal reflux disease (GERD; acid or non-acid) is an important cause of isolated chronic cough in children. Most authorities suggest that this is not a common cause [14]. In children with neurologic abnormalities, GERD may be associated with aspiration (see "Causes of chronic cough in children", section on 'Aspiration'). In selected children, esophageal monitoring may be undertaken to determine if the episodes of coughing are associated with reflux events, but results are often inconclusive. (See "Clinical manifestations and diagnosis of gastroesophageal reflux disease in children and adolescents" and "Clinical manifestations and diagnosis of gastroesophageal reflux disease in children and adolescents", section on 'Esophageal pH monitoring or impedance monitoring'.)

●Sinus imaging – Cough is sometimes reported with chronic sinusitis, but most authorities suggest that sinusitis is not a common cause of chronic cough in children [13], except in association with an immune defect predisposing to chronic airway infection. Historical series in which sinusitis was diagnosed on the basis of response to antibiotics may have included cases of protracted bacterial bronchitis (PBB), which would also respond to the same antibiotics recommended for chronic sinusitis, or patients who would have improved even without antibiotics. (See "Causes of chronic cough in children", section on 'Upper airway pathology'.)

For patients with clinical features strongly suggestive of sinusitis (eg, mucopurulent drainage, chronic nasal obstruction, or facial pain or pressure), imaging of the sinuses and/or an empiric trial of treatment for sinusitis may be undertaken. However, the results should be interpreted with caution because there is poor correlation between sinus radiography or CT and clinical disease [5,79]. Imaging reveals some sinus abnormality in 20 to 80 percent of asymptomatic patients [80]. (See "Acute bacterial rhinosinusitis in children: Clinical features and diagnosis", section on 'Radiologic features'.)

●Chest CT scans – Chest CT scans may be required to further evaluate the underlying etiology such as for suppurative lung disease, mass seen on the chest radiograph, or nonresolving pulmonary consolidation [6]. If a vascular abnormality is also suspected (eg, sling or ring), contrast may be indicated concurrently with the CT scan. As this involves radiation, seeking advice prior to undertaking a CT scan in a young child is advocated. MRI may be more appropriate in selected circumstances such as for concurrent cardiac and cardiac vessel evaluation.

●Tests for tuberculosis – In settings where tuberculosis is endemic or has a high prevalence [81], particularly if the child is at high risk for exposure, standard tests for tuberculosis should be considered [82], including tuberculin skin testing or interferon-gamma release assay (IGRA). This may be relevant even in the absence of obvious signs and symptoms in the child because tuberculosis infection can be subtle. (See "Tuberculosis infection (latent tuberculosis) in children" and "Tuberculosis disease in children: Epidemiology, clinical manifestations, and diagnosis".)

●Allergy testing – Tests for atopy include skin prick testing or radioallergosorbent testing (RAST) (see "Overview of skin testing for IgE-mediated allergic disease"). However, this is usually not recommended in the assessment of children with chronic cough [47]. Presence of atopy does not increase cough sensitivity [83]. While atopy increases the probability of having asthma, it is neither specific nor sensitive for asthma [56].

●Other tests – Other tests may be required, depending on the suspected diagnoses. These include evaluations for aspiration and swallowing dysfunction (eg, video fluoroscopic swallowing test, gastrointestinal scintigraphy), further complex lung function tests, or echocardiogram. (See "Aspiration due to swallowing dysfunction in children", section on 'Evaluation'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic cough in children".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Cough in children (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Definitions – Chronic cough is usually defined as a cough lasting more than four weeks. "Specific cough" refers to a chronic cough that is ultimately attributable to an underlying physiologic cause. Conversely, "nonspecific cough" is defined as a chronic cough that does not have an identifiable cause after a reasonable evaluation. (See 'Definitions' above.)

●Initial evaluation – All children with chronic cough should be systematically evaluated including a detailed targeted history, physical examination, chest radiograph, and spirometry (when possible) and/or fractional excretion of nitric oxide (F_eNO) (algorithm 1). (See 'Algorithmic approach' above.)

●Specific cough pointers – Symptoms and signs that are highly predictive of a specific cough are outlined in the table (table 2); these include:

•Chronic wet or productive cough; wheeze; onset after a choking episode; chest pain; hemoptysis; recurrent pneumonia; hypoxia; exertional dyspnea; associated cardiac, immunologic, or neurologic abnormalities; failure to thrive; or feeding difficulties

•Abnormalities on physical examination such as chest wall deformities, digital clubbing, wheezing, rhonchi or crepitations, and hypoxemia

•Abnormal chest radiography and/or spirometry

The presence of these specific cough pointers narrow the diagnostic possibilities and call for further specific testing or referral (algorithm 1). (See 'Specific cough pointers present' above.)

●Causes of specific cough – Causes of chronic specific cough in children are presented in the table (table 1). Particularly important considerations are:

•Protracted bacterial bronchitis (PBB) – The symptom of a chronic wet cough in a young child usually indicates PBB or other suppurative lung disease. Children with suspected PBB should be treated with antibiotics, selected based on the likely underlying bacteria, local antibiotic sensitivity and resistance patterns, and the child's history of antibiotic use. (See 'Wet or productive cough' above and "Causes of chronic cough in children", section on 'Chronic endobronchial suppurative disease'.)

•Foreign body aspiration – A chronic cough that first began after an episode of choking, or that began suddenly while eating or playing (especially in a preschool-aged child), suggests the possibility of foreign body aspiration. (See 'Suspicion of retained foreign body' above and "Airway foreign bodies in children".)

There are many more causes of specific cough, including bronchiectasis, asthma, and chronic respiratory infections (eg, tuberculosis).

●Nonspecific cough

•Diagnosis – A provisional diagnosis of nonspecific cough can be made if no features of specific cough are identified on the initial evaluation (table 2). By definition, nonspecific cough is dry and nonproductive, and the chest radiograph and spirometry (if performed) are normal. (See 'No specific cough pointers (nonspecific cough)' above.)

•Management – For children with nonspecific cough, a "wait, reassess, and review" approach is recommended since medications are generally not efficacious for nonspecific cough and there is a high likelihood that the cough will resolve spontaneously (algorithm 2). If medications for asthma or reflux are trialed, the child should be reassessed in two to four weeks. Cough that has not resolved with treatment trials necessitates further investigation and likely requires referral to a pulmonologist. (See 'Nonspecific cough' above.)

●Other management considerations

•Cough medications – We recommend not treating chronic cough with over-the-counter cough suppressants or other nonprescription cough medicines (Grade 1B). There is no evidence that they are effective, and there are safety concerns. (See 'Other management considerations' above.)

•Family counseling – Exacerbating factors for cough such as environmental tobacco smoke or allergens should be sought and addressed. The clinician should also evaluate the impact of the cough on the child and the family, discuss any concerns, and set realistic expectations. (See 'History' above and 'Other management considerations' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Roni Grad, MD, who contributed to earlier versions of this topic review.