Article Text

Download PDFPDF

Which children with persistent cough should be regarded as persistent bacterial bronchitis?
  1. Arzu Bakirtas1,
  2. Ipek Turktas1
  1. 1Department of Pediatric Allergy and Asthma, Gazi University Faculty of Medicine, Ankara, Turkey
  1. Correspondence to:
    Dr Arzu Bakirtas
    Gazi Üniversitesi Tip Fakültesi Pediatrik Allerji ve Astim, Bilim Dali 06500 Besevler, Ankara, Turkey; arzubakirtas{at}
  1. D Donnelly2,
  2. A Critchlow2,
  3. M L Everard2
  1. 2Paediatric Respiratory Unit, Sheffield Children’s Hospital, Sheffield, UK

Statistics from

We read with interest the article by Donnelly et al1 on outcomes treated for persistent bacterial bronchitis. They reported on 81 children with persistent cough for a period of 1 month to 14 years with a retrospective chart review. The authors defined persistent bacterial bronchitis as a wet cough present for >1 month that resolved with appropriate antibiotic treatment. They concluded that persistent bacterial bronchitis is often misdiagnosed as asthma, and that the number of patients is increasing because of the reduction in prescribing of antibiotics in the community. Prolonged antibiotic prescription is an appropriate and safe treatment in children with persistent wet cough as advocated in this paper. Paediatricians reading the paper may think that all children with persistent wet cough lasting >1 month should be labelled first as persistent bacterial bronchitis and that antibiotics should be used for at least 6–8 weeks to prevent progression to bronchiectasis in such children.

There are a number of issues in this study on which we would like to comment. First, the authors defined persistent cough as a cough lasting >1 month, which means that patients with post-infectious cough might have been included in their analysis. However, this type of persistent cough may last >1 month but resolves progressively without any specific treatment. In fact, this was defined as “period effect” in reports defining and evaluating cough in children.2,3 Second, almost 60% of their patients were symptomatic before 2 years of age and almost half also had wheezing and shortness of breath at referral for >1 year, which is very similar to a classic asthma history. These children were reported to be symptomatic despite anti-asthmatic treatment. However, the authors did not mention whether they had checked that these children had been treated effectively with asthma medications (ie, with an appropriate dose and for an appropriate duration). They also did not state whether they checked for problems with compliance with asthma treatment before they started to treat them with an antibiotic. Third, 11% of children diagnosed as having persistent bacterial bronchitis, which was expected to resolve with antibiotics, were still symptomatic at referral although they were on antibiotic treatment. In addition, almost half of the patients required at least 6 weeks of antibiotic treatment and 13% needed treatment for ⩾3 months, which cannot be considered a minority. Most importantly, this approach may mislead general practitioners and paediatricians both to overprescribe and to the routine use of prolonged antibiotics for all children with persistent cough lasting >1 month. Most of these issues have also been previously commented on by Rubin in an editorial.4

On the other hand, chronic productive cough can be a manifestation of asthma5 as well as so-called bacterial bronchitis which may be present coincidentally with asthma in children. Moreover, Donnelly et al1 also showed that one-third of their patients had asthma concomitant with persistent bacterial bronchitis. In fact, this group of children is a considerable diagnostic challenge. As clinicians, which children with persistent cough should we not overtreat with either inhaled steroids or antibiotics for a long time?

We believe that the conclusions of the study by Donnelly et al should be interpreted with caution until they are investigated in detail in well designed, randomised, placebo-controlled, prospective studies. This approach would put such studies on a more scientific and convincing basis.


Authors’ reply

We thank Drs Bakirtas and Turktas for their interest in our paper,1 although it is not entirely clear from their letter whether they doubt the condition exists or whether they feel the approach to management we adopt is incorrect.

Persistent bacterial bronchitis/persistent endobronchial infection without cystic fibrosis or significant immunodeficiency now appears as an important cause of persistent cough in paediatric cough guidelines in the UK (in press), Australia2 and North America.3 What is less clear is how patients should be managed at different stages of the condition. However, their letter does permit us the opportunity to discuss the difficulties in ensuring children with a persistent wet cough (as opposed to the typical dry “post-viral/non-specific” cough2 are correctly managed. Unfortunately, the lungs have a very limited repertoire of responses to an external insult. Inflammation of the conducting airways provoked by a stimulus such as an acute viral infection, persistent endobronchial infection, asthma or recurrent aspiration will induce a cough, frequently lead to mucus hypersecretion and may induce noisy breathing. Young children with frequent viral infections, mild asthma or persistent endobronchial infection look very similar on superficial assessment, with parents reporting a persistent cough particularly at nights and/or exercise and symptoms getting worse with colds. Trying to correctly identify the cause of the symptoms when cough is the most prominent symptom generally requires taking a very careful history of the cough. Examination is frequently unremarkable until the child is asked to cough, although in those with more troublesome endobronchial infection it is not uncommon to hear coarse inspiratory and expiratory sounds.

Ultimately, diagnosis in most cases rests on a clear and unequivocal response to a trail of treatment. One of the most useful questions when assessing a cough is to ask when the child last did not have a cough. Parents of children with frequent viral infections will often state that their child always coughs but, when asked when the child was last well, will report that the child had a good spell for a couple of weeks last month but then got a cold and the cough returned. While persistent endobronchial infection can improve spontaneously, particularly in the summer when it often becomes infrequent and dry or indeed disappears temporarily, it is generally persistent through most of the year. A typical history will include a report that antibiotics do not help or help only transiently, and that the cough quickly returns at the end of the course of treatment. On closer questioning, those who report no response will often report that the child seemed to improve but the cough rapidly deteriorated when the antibiotics were discontinued. Our experience is that, with persistent endobronchial infection, the first thing to improve is the general well being of the child and the cough can take 10–14 days to resolve (and longer in a few cases). The timing of the cough is also typical with a “60-a-day smokers’ cough” on waking and a wet cough with exercise. During viral induced exacerbations or in those with more troublesome symptoms, the cough may be present during sleep. Unfortunately, untreated asthma or exacerbations due to viruses can also produce a wet productive sounding cough since the “wetness” only signifies the presence of excess airway secretions.

As noted in our paper, the two most common referrals we receive are persistent cough and “difficult asthma”. We remain to be convinced that true severe asthma occurs in children if accurate assessment is made, and have never used long-term oral steroids for any patient with asthma in our clinic. Children with “difficult asthma” generally fall into one of three categories: (1) incorrect diagnosis; (2) asthma and co-morbidity such as persistent endobronchial infection; or (3) failure to take their medication correctly either due to regimen compliance or device compliance (competence and contrivance).4 The implication that failure to assess children correctly makes our report invalid is a little disappointing. As a group we have undertaken a considerable amount of research into factors affecting drug delivery (both device and, more importantly, patient factors)4 and in trying to improve our characterisation of phenotypes of respiratory disease.5 We take issues of non-compliance in all its guises very seriously.

One reviewer of the original paper commented that it was easy to make the diagnosis of persistent bacterial bronchitis/persistent endobronchial infection. This is indeed the case in an older child producing sputum at will or giving a clear history of expectorating yellow/green sputum but, in younger children who do not expectorate or who are at an early stage of the disease progression, diagnosis rests largely on the history. The only reliable way to make the diagnosis unequivocally is to undertake bronchoscopic examination, but the number of patients is such that this is not currently our first line investigation, particularly as many will resolve completely with one or two courses of antibiotics. Common things are common, and we do try hard to determine whether asthma is the sole diagnosis or a contributory factor through appropriate trials of treatment.

Fortunately our primary care physicians are very good and it is uncommon for a child to be referred who has not been assessed as possibly having asthma; overdiagnosis or failure to recognise a co-morbidity is much more common. If inhaled therapy and/or oral steroids fail to lead to a resolution and we have dealt with obvious confounders such as compliance, we then start children on a trial of treatment for 2 weeks and contact the parents at the end of the course before planning future interventions.

We would entirely agree that the inappropriate use of antibiotics should be discouraged but, equally, inappropriate treatment due to failure of an accurate diagnosis leading to unnecessary morbidity is also unacceptable. Trying to ensure that a correct diagnosis is made and that appropriate treatment is provided does take care and regular review. The lag phase between being a “chesty” child and an adult with bronchiectasis will mean that adult physicians will only slowly start seeing numbers increase over the next decade unless their paediatric colleagues and, indeed, primary care physicians recognise the potential size of this problem.


View Abstract


  • Competing interests: None.


  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles