Controversies in Antibiotic Use for Chronic Wet Cough in Children

Cough is the most common complaint among primary healthcare visits in pediatrics (1, 2). Chronic cough in young children is typically defined as cough lasting 4 weeks (3, 4). Published guidelines for systematic evaluation of children with chronic cough differentiate “dry” vs “wet” cough as part of the evaluation as they suggest distinct clinical entities (5, 6). “Wet” cough suggests secretions in the airways (7). Research varies on how good parents are at noting dry vs wet cough (7, 8). Guidelines recommend treating chronic wet cough of 4 weeks or longer with antibiotics (5, 9).

A common cause of chronic wet cough is protracted bacterial bronchitis (PBB), a bacterial infection of the lower airways leading to a chronic wet cough (10–13). The gold standard for diagnosis of PBB is flexible bronchoscopy with lavage to demonstrate inflammation and bacterial overgrowth (laboratory PBB) (14), but this procedure is invasive and requires travel to specialty centers. Accordingly, guidelines supported by clinical trials from specialty centers in Australia recommend empiric antibiotics without bronchoscopy and lavage if there is no evidence for other causes of cough for treatment (5, 6, 15, 16). Other causes of chronic cough which vary by region around the globe include asthma, sinusitis, upper airway cough syndrome, gastroesophageal reflux disease, infection including tuberculosis, aberrant innominate artery, and pulmonary eosinophilia (17). Clinical improvement after 2 weeks of antibiotics supports a diagnosis of “clinical PBB.” Effective empiric antibiotics obviate the need for referral to a subspecialty center and flexible bronchoscopy with lavage, hastening care and lowering health care costs.

PBB is often unrecognized but has been found to be more common than asthma in some case series (10). PBB is more common in the setting of other conditions, notably airway malacia (18). PBB appears to be especially common in populations with high rates of respiratory infections associated with crowding and indoor air pollution, and in some Indigenous populations (10, 19). Failure to diagnose and treat PBB is associated with long-term impairment of lung function, suppurative lung disease, and bronchiectasis (20). The incidence of bronchiectasis is as high as 20 per 1000 children in some indigenous populations in high-income countries (21).

Conversely, young children, especially those in daycare, commonly develop recurrent respiratory infections (22, 23). Each respiratory infection has the potential to cause a wet cough that lasts for 2–6 weeks, and then the child becomes infected with a new virus prior to resolution of the previous one. This can lead to a chronic cough that lasts for months. Concerns about antibiotic overuse includes effects on the microbiome, increased risk of developing obesity, Clostridium difficile infection, and antibiotic resistance (24–27). Other concerns about antibiotics include hypersensitivity reactions or even more rare side effects such as blood dyscrasias. One study noted that each additional day of antibiotics is associated with a 7% increased odds of an adverse event (28). This risk-benefit assessment has been calculated for the use of penicillin in cases of streptococcal pharyngitis to prevent rhematic heart disease (29). These concerns have been encapsulated in the field of antimicrobial stewardship, which seeks to minimize unnecessary antibiotic exposure by optimizing antibiotic prescribing for infections, including antibiotic agent, dose, and duration for treatment, thereby reducing the risk of promoting antimicrobial resistance or other adverse effects (30, 31).

Empiric treatment of chronic wet cough per the guidelines leads to faster resolution of PBB, prevents long term complications, and saves families the financial and time burden as well as the risks of travel to a specialty center. However, following the guidelines in low-risk populations may lead to antibiotic overuse and serious complications. Adding to the uncertainty, the research supporting the use of antibiotics in chronic wet cough have not been evaluated in primary care settings in North America. Moreover, many antibiotics induce potent immunomodulatory responses (32), and it is uncertain to what extent these immunomodulatory effects may impact the clinical course of chronic wet cough. To identify optimal treatment for chronic wet cough, prospective clinical trials of patients in North America are needed. In addition to controlled trials comparing outcomes of antibiotics versus placebo, research is needed to differentiate children with recurrent viral respiratory infections more accurately from those with cough from other distinct etiologies. Cough frequency, duration, and qualitative features have historically relied on patient/family report. More objective measures of cough duration and characteristics would be useful to determine which patients to treat with antibiotics. Validated cough quality of life reports exist (33, 34), but there is a lack of widely accepted cough sensors that could track children’s cough. Multiple cough sensors currently being evaluated, but they are all still in early stages of development (35–37). Development of technology capable of more reliably differentiating “wet” from “dry” cough would ensure that physicians are applying the most appropriate guidelines to patients. These sensors could potentially provide invaluable data differentiating the specific sound waves of coughs due to PBB, reflux, respiratory infections, and asthma. In addition, to improve knowledge about the clinical cough itself (wet vs dry, duration, prior coughing history, and pneumonia history), non-invasive biomarkers (sputum or serum) for diagnosis of PBB could guide care and reduce unnecessary antibiotic usage.

We originally proposed a placebo-controlled research study within our NIH funded Environmental influences on Child Health Outcomes IDeA States Pediatric Trials Network. Our proposal was to treat with children 6 months to 5 years with wet cough lasting >4 weeks with empiric antibiotics versus placebo. We proposed 2 weeks of amoxicillin-clavulanate versus placebo and then 2 weeks of antibiotics for anyone still coughing as well as a 6 month follow up to determine recurrence rate between the groups. The reviewers within the network felt very strongly about the study, one way or the other. Some felt it was unethical to treat with antibiotics for potential viral illnesses. Some felt it was unethical to NOT treat with antibiotics given the risk of bronchiectasis. Given these strong feelings, the proposal did not move forwards into a placebo-controlled trial. However, we believe that is exactly why this trial needs to be completed to better answer these questions. Duration of antibiotics in PBB is another important question: is 2 weeks necessary? Does four weeks reduce recurrence rate? Better understanding of each individual host could guide these questions and could determine which children do indeed need a longer duration of antibiotics. In cystic fibrosis, a well-studied disease but with completely different pathophysiology, there are no randomized controlled trials evaluating antibiotic duration (38). In non-randomized controlled studies, longer duration antibiotics (13–15 days) are not better than shorter duration (10–12 days), at least in terms of serum biomarkers and lung function (39). Shorter durations of antibiotics for pneumonia have recently demonstrated non-inferiority to longer courses. (40, 41)

Because chronic cough is common, costly, and associated with decreased quality of life, multiple healthcare visits, and missed school (20, 42, 43) priority should be given to conducting trials objectifying cough frequency and quality as well as risk-benefit and cost-effectiveness of evaluations of empiric antibiotic use for chronic wet cough.