Children with cough, in particular chronic cough, are sometimes referred to otolaryngologists for assessment, diagnosis, and management. Although the likely diagnoses encountered by otolaryngologists are rhinosinusitis, foreign body aspiration, and tracheomalacia, otolaryngologists should be cognizant of the many other possible diagnoses and the evidence for and against their association. This article highlights and focuses the discussion on the cough issues relevant to otolaryngologists.
Otolaryngologists are sometimes referred children with cough as cough is the most common symptom that results in medical consultations. In the United States, 29.5 million visits to doctors are for cough. Although most of these consultations are likely for acute cough (cough lasting less than 2 weeks ), a significant number of consults are for chronic cough. Otolaryngologists are unlikely to encounter children with acute cough, however, other than in the context of an aspiration of a foreign body. Hence, this article concentrates on chronic cough defined as cough present for more than 4 weeks. The time-length definition of chronic cough in children differs from that used in adults (more than 8 weeks), primarily based on the natural history of acute respiratory infection–associated cough in children. It also takes into consideration child-specific safety issues, such as aspiration of foreign body, where delayed diagnosis can lead to long-term significant consequences.
In general, although children and adults share some commonalities, there are significant differences. An example pertinent to cough is the marked difference between adults and children in the likelihood of chronic cough being a symptom of bronchial carcinoma. Cough-related issues related to adults are discussed elsewhere in articles in this issue. This article highlights and focuses the discussion on the cough issues relevant to otolaryngologists.
Epidemiology data
Prevalence
There are few accurate data on the prevalence of chronic cough in children as questionnaires for chronic cough assessment have limited validity. Published data describe community prevalence of chronic cough in primary school–aged children (6–12 years) as 5% to 10%. The prevalence is likely higher in preschool-aged children as retrospective and prospective studies have shown that the majority of children with chronic cough seen in clinics were young (median age 2 to 3 years).
Burden of Illness
Chronic cough is considered trivial to some medical practitioners; however, the symptom is associated with significant morbidity in children and their parents. The burden of cough is also reflected in the use of over-the-counter (OTC) cough medications consumed worldwide. Approximately 10% of US children use a cough and cold medication in a given week. The cost of prescribed medicines (eg, those for asthma, gastroesophageal reflux, and so forth) used for cough is unknown. Furthermore, people report a high number of medical visits for cough before they visit a specialist. In one study that assessed the number of prior medical consultations for coughing illness in the 12 months before children first presented to a respiratory pediatrician, more than 80% of children had five or more doctor visits and 53% had more than 10.
Understanding why parents consult is helpful in evaluating children with chronic cough. There are only three studies of children that have examined parental evaluations or concerns of their children’s coughing illness. In the two older studies, parents’ main concerns were related to disturbed sleep, discomfort, and fear that cough would cause permanent chest damage. A recent Australian study involving 190 families described that the most significant concerns and worries expressed by parents were feelings of frustration, being upset, sleepless nights, awakenings at night, helplessness, stress, and feeling sorry for the child. Specific issues that bothered parents most were the cause of cough, cough relating to a serious illness, their child not sleeping well, and cough causing damage. These concerns and worries of parents were significantly reduced when the child’s cough resolved.
Unlike adults with chronic cough, children with chronic cough did not have symptoms of anxiety. Also parents of children with chronic cough did not have symptoms of anxiety or depression but were under stress. This is in contrast to the data in adults with chronic cough, which showed associated anxiety or depression symptoms.
Pathophysiology
There is direct and indirect evidence that age influences the physiologic domains that influence the clinical manifestation of conditions where cough is a dominant feature. These physiologic domains can be simplified:
- 1.
Cough-specific physiology
- 2.
General respiratory physiology
- 3.
Other direct systems, such as the immune system, that influence the respiratory system
- 4.
Other general physiology.
Much of these data are available elsewhere. Examples of cough-specific physiologic differences include age- and gender-related variations in cough sensitivity. In children, gender does not influence cough sensitivity, whereas in postpubertal adolescents and adults, girls and women have significantly increased sensitivity. Also, the cough reflex is weak in premature infants and develops with maturity. Exactly when the cough reflex is fully matured is unknown but it is likely at approximately age 5, which is the cutoff age for risk of accidental nut inhalation; thus, parents are advised not to give nuts to their children prior to age 5. Also, adults easily expectorate when airway secretions are present whereas children do not, even when secretions are abundant. Thus classical adult terms, such as productive cough, cannot be applied to young children.
Examples of general respiratory physiology include differences in caliber of large and small airways and percentage of time spent in rapid eye movement sleep, which influences cough frequency. Smaller airway caliber (which influences air flow exponentially as opposed to linearly) and lack of collateral ventilation in immature lungs influence the likelihood of presence of wheeze and atelectasis in respiratory conditions in which chronic cough is common, such as chronic suppurative lung disease and right middle lobe syndrome.
The immune system, which has its own developmental physiology, affects the degree and frequency of respiratory infections, hence, presentations to doctors for cough. Other general physiology that influences the pathophysiology and management in children compared to adults includes differences in cognitive function, ability to self-express, and so forth.
Thus, there are observed differences in cause, management, and measurement of response between children and adults. Cough quality (eg, brassy cough and wet cough) used in children as a marker of some conditions is less useful in adults. Also, although empirical therapy (for asthma, gastroesophageal reflux disease [GERD], and upper airway disease) for chronic cough is widely advocated in adults, such an approach is not advocated in children. There is some evidence that an empirical approach is potentially harmful, related to the use of medications and the delay in obtaining a correct diagnosis, such as missed foreign body aspiration, that lead to bronchiectasis. The differences between adults and children are also exemplified in child-specific diagnoses, such as protracted bacterial bronchitis and evidence-based cough guidelines.
Pathophysiology
There is direct and indirect evidence that age influences the physiologic domains that influence the clinical manifestation of conditions where cough is a dominant feature. These physiologic domains can be simplified:
- 1.
Cough-specific physiology
- 2.
General respiratory physiology
- 3.
Other direct systems, such as the immune system, that influence the respiratory system
- 4.
Other general physiology.
Much of these data are available elsewhere. Examples of cough-specific physiologic differences include age- and gender-related variations in cough sensitivity. In children, gender does not influence cough sensitivity, whereas in postpubertal adolescents and adults, girls and women have significantly increased sensitivity. Also, the cough reflex is weak in premature infants and develops with maturity. Exactly when the cough reflex is fully matured is unknown but it is likely at approximately age 5, which is the cutoff age for risk of accidental nut inhalation; thus, parents are advised not to give nuts to their children prior to age 5. Also, adults easily expectorate when airway secretions are present whereas children do not, even when secretions are abundant. Thus classical adult terms, such as productive cough, cannot be applied to young children.
Examples of general respiratory physiology include differences in caliber of large and small airways and percentage of time spent in rapid eye movement sleep, which influences cough frequency. Smaller airway caliber (which influences air flow exponentially as opposed to linearly) and lack of collateral ventilation in immature lungs influence the likelihood of presence of wheeze and atelectasis in respiratory conditions in which chronic cough is common, such as chronic suppurative lung disease and right middle lobe syndrome.
The immune system, which has its own developmental physiology, affects the degree and frequency of respiratory infections, hence, presentations to doctors for cough. Other general physiology that influences the pathophysiology and management in children compared to adults includes differences in cognitive function, ability to self-express, and so forth.
Thus, there are observed differences in cause, management, and measurement of response between children and adults. Cough quality (eg, brassy cough and wet cough) used in children as a marker of some conditions is less useful in adults. Also, although empirical therapy (for asthma, gastroesophageal reflux disease [GERD], and upper airway disease) for chronic cough is widely advocated in adults, such an approach is not advocated in children. There is some evidence that an empirical approach is potentially harmful, related to the use of medications and the delay in obtaining a correct diagnosis, such as missed foreign body aspiration, that lead to bronchiectasis. The differences between adults and children are also exemplified in child-specific diagnoses, such as protracted bacterial bronchitis and evidence-based cough guidelines.
Etiologic factors
Common etiologic causes are dependent on the setting or practice and consideration of whether or not coexistent symptoms are taken into account. For example, in Iran, pediatric chronic cough is not uncommonly associated with eosinophilia related to parasitic infestation. In Indigenous settings in rural Australia, the most common cause of chronic cough is bronchiectasis. In affluent urban settings, these diagnoses for chronic cough are rare. Also, common etiology depends on whether or not isolated cough is considered (ie, cough without other symptoms or nonspecific cough) or cough in association with other symptoms (specific cough). In a Turkish study, 25% of the children with chronic cough managed in accordance with American College of Chest Physicians guidelines had asthma. This was not surprising, as the cohort included children without isolated cough; 18.5% of children had airway reversibility at the first assessment. In contrast, most children with isolated cough do not have asthma. Thus, although an otolaryngologist may encounter any underlying cause of a child’s chronic cough, the frequency of the etiologic factors is likely different from that seen by pulmonologists. Table 1 highlights common etiologic factors in children, including differences from adults and level of evidence defining cause and effect. Conditions that otolaryngologists may encounter in the evaluation of children with chronic cough include the conditions listed in Table 1 .
| Type of Cough b | Key Difference Between Children and Adults | Best Method for Confirmation of Diagnosis | Alternative Method for Confirmation of Diagnosis | Highest Level of Evidence c | Limitation | |
|---|---|---|---|---|---|---|
| Asthma | Dry | Nonspecific cough in children is unlikely asthma. In adults, isolated cough is commonly asthma. | Reversibility of FEV 1 demonstrated on spirometry | Indirect challenge for airway hyper-responsiveness^ | RCT for cough associated with other symptoms. | Asthma medications for nonspecific cough not effective. |
| Protracted bronchitis | Wet | Not described in adults | Bronchoscopy and response to antibiotics | Response to antibiotics | Meta-analysis cohort | |
| Tracheobron-chomalacia | Brassy for tracheomalacia | Not known if useful in adults (no studies) | Bronchoscopy | Airway screen (no data on specificity or sensitivity) | Cohort | |
| Tonsillar hypertrophy | Dry | In adults, snoring is associated with bronchitis. In otherwise normal children with OSA, most do not have a cough. | Direct visualization | Case series | ||
| Rhinitis | Dry | Consideration of immune testing, CF screen, etc recommended in children with persistent rhinitis | Defined by AAAAI and ACAAI as “characterised by one or more of following symptoms: nasal congestion, rhinnorhea, sneezing, itching ” | No diagnostic tests but in some skin testing (preferred method), IgE immunoassays or nasal endoscopy are required | RCT but meta-analysis of studies of antihistamines for seasonal allergic rhinitis shows different results to single positive RCT. | Change in cough marginal compared to other symptoms and no difference seen in night-time cough |
| Chronic rhinosinusitis | Wet or dry | Consideration of immune testing, CF screen, etc recommended in children with persistent rhinitis. Different sinus develops at different ages with adult sizes attained usually by 14 years to late adolescents. | Major symptoms (nasal congestion or obstruction, nasal discharge with or without facial pain/pressure, olfactory disturbance) and either endoscopic signs (polyps, mucopurulent discharge, edema, obstruction at meatus) or CT changes | Clinical diagnosis primarily with limited role for radiology as CT scans are abnormal in a third of population | Descriptive studies. Cough inconsistently mentioned in guidelines as a symptom. | Other studies have shown that while there was a significant difference in rhinitis-specific symptoms, there was no significant difference in cough. |
| Aspiration | Wet | In children, aspiration is associated with those with multisystem dysfunction (ie, not only neurology problems ). In adults, aspiration most common in those with stroke | Videofluoroscopy | Speech pathology clinical evaluation | Case series | |
| GERD and laryngophargngeal reflux | Dry unless associated with aspiration | GERD as cause of nonspecific cough is uncommon in children but common in adults | pH monitoring with limitations | Role of barium meal and esophageal impedence monitoring uncertain | Case series Gastroenterology-based guidelines are less definitive about the association between cough and GERD. | Increased respiratory problems in infants with GERD treated with proton pump inhibitors |
| Underlying lung disease, such as bronchiectasis, interstitial lung disease | Depends on cause | High-resolution CT scan | Depends on cause | Depends on cause, meta-analysis for bronchiectasis | ||
| Pulmonary infections (eg mycoplasma, pertussis, chlamydia, tuberculosis) | Depends on cause. In pertussis with or without whoop and vomiting. Generally dry post acute phase. | Adults with pertussis rarely whoop or have post-tussive vomiting | Bronchoalveolar lavage, nasopharyngeal specimen (PCR or culture) | Sputum or bloods when specific tests available | ||
| Habit and psychogenic cough | Dry | Characteristic cough recognized in children but not in adults | No diagnostic test | Response to psychologic based Rx | Cohort studies |
a More rare causes associated with cough are not included (ie, almost all pulmonary and some cardiac disorders can present with chronic cough).
b Type of cough refers to quality of spontaneous cough. When a child has not coughed spontaneously during the consultation, the child’s cough character may be elicited by requesting the child to cough several times. 25 There is sometimes a difference, however, between spontaneous cough and volunteered cough.
c Highest level of evidence is evidence that refers to whether not cough resolves with treatment specific for cause in pediatric studies.
Evaluation of children with chronic cough
Etiologically based diagnosis is the main strategy for treating cough in adults and children; however, although an integrative empirical treatment strategy works well in adults, an etiologically based treatment is best for children. Thus, defining the cause is an important component of the evaluation. In defining the cause, it is helpful to define cough types in accordance with different constructs. Pediatric chronic cough can be classified in several constructs based on (1) likelihood of an identifiable underlying primary cause (specific and nonspecific cough) or (2) characteristic (moist versus dry). Specific cough refers to cough associated with other symptoms and signs that are suggestive of an associated or underlying problem, and nonspecific cough is dry cough in the absence of any identifiable respiratory disease or known cause. These classifications are, however, not mutually exclusive.
Wet/Moist/Productive Cough Versus Dry Cough
Even when airway secretions are present, young children rarely expectorate sputum. Hence, wet/moist cough is a preferable to productive cough. The distinction of dry and wet/moist cough has been shown to be valid and reliable. In a study of 106 children, cough quality (wet/dry) assessed by clinicians had good agreement with parents assessment (κ = 0.75; 95% CI, 0.58–0.93) and had good sensitivity (0.75) and specificity (0.79) when compared to bronchoscopy findings. The use of wet versus dry cough for predicting cause of cough or response to treatment has not been shown in children, except in protracted bronchitis.
Specific and Nonspecific Chronic Cough
Clinical signs and symptoms that are suggestive of an underlying pulmonary or systemic disorder are termed specific cough pointers ( Box 1 ). When any of these pointers are present, the cough is referred to as specific cough. If specific cough pointers are present, further investigations and management of the primary pulmonary pathology are usually warranted. If not, a counsel, watch, wait, and review approach is suggested. This approach is suggested for nonspecific cough (dry cough in the absence of specific cough pointers). Such a cough is more likely to undergo natural resolution ; however, a dry cough may be the early phase of a wet cough, so children should be reviewed and attention to exacerbation factors and parental concerns is warranted (discussed later).
Auscultatory findings (wheeze, crepitations/crackles, differential breath sounds, stridor)
Cough characteristics (eg, cough with choking, cough quality cough starting from birth)
Cardiac abnormalities (including murmurs)
Chest pain
Chest wall deformity
Daily moist or productive cough
Digital clubbing
Dyspnea (exertional or at rest)
Exposure to pertussis, tuberculosis, etc
Failure to thrive
Feeding difficulties or dysphagia (including choking/vomiting)
Hemoptysis
Immune deficiency
Medications or drugs (angiotensin-converting enzyme inhibitor)
Neurodevelopmental abnormality
Recurrent pneumonia
Data from Chang AB, Landau LI, van Asperen PP, et al. The Thoracic Society of Australia and New Zealand. Position statement. Cough in children: definitions and clinical evaluation. Med J Aust 2006;184:398–403.
In selected children with nonspecific cough, diagnoses with simple treatment options (eg, asthma and complications of upper respiratory tract infections, such as sinusitis and bronchitis, may be considered). The evidence for the association between asthma and upper airway disorders and cough, however, is not as straightforward in children as it is in adults. The use of isolated cough as a marker of asthma is controversial, with more recent evidence (clinical and community epidemiologic studies) showing that in most children, isolated cough or nonspecific cough does not represent asthma.
Investigations
At minimum, all children with chronic cough should have a spirometry (if age appropriate) and chest radiograph performed. The validity of these has been shown. When a chest radiograph taken for chronic cough is abnormal, the odds ratio of a specific cause was 3.16 (95% CI, 1.32–7.62). Other tests to identify the cause of nonspecific cough have limited applicability in pediatrics (but are included in Table 1 ). Identification of airway hyper-responsiveness, diagnostic for asthma in adults, is of limited use in children because of interpretation difficulties, and reliable tests can only be performed in older children in whom cough is less common. The three most common causes of chronic cough in adults (GERD, asthma, and upper airways syndrome) are not as common in children. Increased cough sensitivity is found in most conditions causing chronic cough but its utility is limited to research.
Except when classical asthma is the cause, children with specific cough usually require additional tests (chest high-resolution CT scan, bronchoscopy, videofluoroscopy, echocardiograph, sleep polysomnography, nuclear medicine scans, immunologic assessment, and so forth). Brief points on the role of selected investigations for the more common causes of chronic cough are summarized in Table 1 . A comprehensive role of all available tests for evaluation of lung disease is beyond the scope of this article, as it would encompass the entire spectrum of pediatric respiratory illness.
Medications
Treatment of chronic cough should be etiologically based, which for specific cough (which includes almost all respiratory diseases) is beyond the scope of this article. Clinicians should be cognizant that cough is subjected to the period effect (ie, spontaneous resolution of cough); the benefit of placebo treatment of cough has been reported to be as high as 85% and, therefore, nonplacebo-controlled intervention studies have to be interpreted with caution. Evidence (or lack of) for efficacy of treatment trials for nonspecific cough is summarized in Table 2 . Systematic reviews have concluded that cough OTCs have little benefit in the symptomatic control of childhood cough. Significant morbidity and mortality from cough OTCs can occur and OTCs are common unintentional ingestions in children aged under age 5. Cochrane reviews of symptomatic treatment of cough have shown that diphenhyramine is not beneficial for pertussis-related cough; cromones and anticholinergics have little, if any, role in nonspecific chronic childhood cough; and, 10 days of antimicrobials reduces persistent cough in children with chronic nasal discharge but benefits are modest (number needed to treat was eight). Antimicrobials may be indicated for subacute/chronic moist cough; two randomized controlled trials (RCTs) showed predominance of Moraxella catarrhalis . Meta-analysis of antimicrobials for acute bronchitis in older children (age older than 8 years) and adults showed a small benefit of 0.58 days but with significantly more adverse events. Systematic review of uncomplicated sinusitis in children showed that the clinical improvement rate was 88% with antimicrobial and 60% with no antimicrobial. Observational studies show an improvement in cough associated with rhinitis. A single small (n = 20) RCT in children with allergic rhinitis treated with antihistamines showed benefit but larger safety studies (combined n = 793) did not. In comparison, a large difference between groups was found for nasal symptoms and there was no difference in nighttime cough. An earlier study in children with asthma and nasal obstruction showed decreased cough in addition to asthma severity in those randomized to intranasal budesonide. There are only two published RCTs on inhaled corticosteroids for nonspecific cough in children and both groups cautioned against its prolonged use.
