Send letter to journal:
Re: A follow-up study of children hospitalised with community-acquired pneumonia.

In their study Eastham et al.1 concluded that CAP requiring admission to hospital is associated with deterioration of lung function and persistent respiratory symptoms. We would like to raise some issues regarding this study. Firstly authors state that there is increasing evidence that deficits in lung function and asthma are related to CAP in childhood, with two references in support of this statement. In the second and most recent of these, Castro-Rodriguez et al.2 postulate that the diminished lung function in children with pneumonia is probably due to a pre-existing alteration in the regulation of airway muscle tone, rather than to a direct and unique effect of the infection itself. Secondly the assessment of persistent cough is highly debatable 3,4, as the authors also remark, it is difficult to collect and interpret the validity of these data. In the list of signs of chronic respiratory disease authors included pectus excavatum and pectus carinatum which, to the best of our knowledge, are not a consequence of chronic respiratory disease but a possible cause.Moreover there was no indication of how this assessment was obtained in follow-up (i.e. by blinded operators or not), and this could introduce a possible bias. The authors also conclude that children with CAP and atopic parents have a 7.0% deficit in FEV1 predicted and 4.4% deficit in FVC. The extent of this deficit seems to be hardly relevant in real life, expecially considering the suggestion to parents to consult doctor if they are concerned about their children lung function. It is difficult to determine how physicians should interpret and counsel parents about a possible deficit of 7.0% of FEV1 and, furthermore, how this could change therapeutic choices or allow any kind of preventive treatment. We recently reviewed lung function in children admitted to our hospital for a severe pneumonia complicated by empyema which required drainage. We were able to recall 17 patients out of a cohort of 34 with a median follow-up of 7 years (range 3.4-27.8). Of these only one had a FEV 1 below 90% of predicted value for sex, age and heght while all the others are distributed among normal values.

Authors report no potential conflict of interest relevant to this letter.

References

1. Eastham KM, Hammal DM, Parker L, Spencer DA. A follow-up study of children hospitalised with community-acquired pneumonia. Arch Dis Child. 2008 Sep;93(9):755-9

2. Castro-Rodríguez JA, Holberg CJ, Wright AL, Halonen M, Taussig LM, Morgan WJ, Martinez FDAssociation of radiologically ascertained pneumonia before age 3 yr with asthmalike symptoms and pulmonary function during childhood: a prospective study. Am J Respir Crit Care Med. 1999 Jun;159(6):1891-7.

3. Dales RE, White J, Bhumgara C, McMullen E. Parental reporting of childrens' coughing is biased. Eur J Epidemiol. 1997 Jul;13(5):541-5.

Send letter to journal:
Re: Response to eletter by Alessandro Amaddeo et al

We thank Amaddeo et al for their comments regarding our follow-up study of children hospitalised with community-acquired pneumonia (CAP) 1. They are correct in stating that Castro-Rodriguez et al postulated that diminished lung function in children following CAP may be due to a pre- existing alteration in airway tone, however this is only one of the possible explanations offered by the authors 2. The second explanation is that lower respiratory tract infections in children may produce alterations in lung structure and lung function and that these alterations may be, in part, responsible for the lower levels of lung function observed subsequently in children. Either way, CAP has been found to be associated with deficits in lung function, as was the case in our cohort of children, and can therefore be considered to be an important ‘marker’ in bringing such children to the attention of a Physician.

We agree that the assessment of chronic cough by parental reporting is subject to bias, however the questionnaire chosen has been widely used and previously validated in studies investigating respiratory morbidity in Australian children 3,4. Reproducibility has been shown to be good or excellent for the specific symptom questions, with Cohen’s kappa value greater than 0.7 for the question pertaining to persistent cough 5. The alternative tool available at the time of study, the cough score, is a quantitative measure of cough and had been validated using an ambulatory cough meter in children with normal respiratory examinations, chest radiographs and spirometry findings only 6. It would have been an inappropriate tool to apply to our study population, who had quite different characteristics, without a further validation exercise. It is acknowledged that pectus excavatum and pectus carinatum are congenital chest wall abnormalities and are not a consequence of chronic respiratory disease and therefore the definition of ‘signs of chronic respiratory disease in the absence of intercurrent respiratory tract infection’ has been subsequently revised for the purpose of future investigations. Congenital chest wall abnormalities were present in only 2 of the 13 children found to have with abnormal chest shape; both were controls with normal lung function. The clinical assessment was conducted by the Principle Investigator (KE) who was also responsible for recruitment and therefore was not blinded to case or control status. Strategies to minimise such bias were considered in study methodology. These included a categorical approach to the classification of questionnaire outcomes to reduce the potential for interpretation and misclassification bias, with outcomes cross-checked with the individual components of each item. Lung function was measured in accordance with published guidelines and lower limits of normal were clearly defined 7,8. Abnormal respiratory examination findings were confirmed by a Consultant Respiratory Paediatrician (DS) for those children who attended the Tertiary Paediatric Respiratory Clinic.

We acknowledge in our Discussion section that the deficit in lung function observed in our cohort may not be of immediate clinical significance, and it is not on this basis that we would suggest that a parent consult their doctor in the future, particularly in the absence of clinical symptoms. We suggest that those children with problems of new persistent cough, asthma or wheeze following CAP should be reviewed by their doctor as we have shown that a small percentage of children may have significant underlying chronic respiratory disease. The triad of doctor diagnosis of a asthma, persistent cough and abnormal respiratory examination was present in 2 cases both of whom were found to have non-CF bronchiectasis and obliterative bronchiolitis, 1 case of productive cough with peribronchial thickening on HRCT scan, and 1 case of severe asthma with right middle lobe scarring. Further prospective work is required to delineate ‘at risk’ groups and to investigate an appropriate follow-up strategy. Whilst a deficit of 7.0% may not be clinically significant in itself, this may become more so over time especially if there are further respiratory insults. Furthermore, we do not know if this deficit is static or progressive. Longer term follow-up would be required to answer this question.

References 1. Eastham KM, Hammal D, Parker L, Spencer D. A follow-up study of children hospitalised with community-acquired pneumonia. Archives of Disease in Childhood 2008;93:755-759

2. Castro-Rodriguez JA, Holberg CJ, Wright AL et al. Association of Radiologically Ascertained Pneumonia before Age 3 yr with Asthmalike Symptoms and Pulmonary Function during Childhood. American Journal of Respiratory and Critical Care Medicine 1999;159:1891-1897.

3. Gray EJ, Peat JK, Mellis CM et al. Asthma severity and morbidity in a population sample of Sydney school children: Part I--Prevalence and effect of air pollutants in coastal regions. Australian & New Zealand Journal of Medicine 1994;24(2):168-75.

4. Faniran AO, Peat JK, Woolcock AJ. Persistent cough: is it asthma? Archives of Disease in Childhood 1998;79(5):411-4.

5. Faniran AO, Peat JK, Woolcock AJ. Measuring Persistent Cough in Children in Epidemiological Studies; Development of a Questionnaire and Assessment of Prevalence in Two Countries. Chest 1999;115(2):434-439.

6. Chang AB, Newman RG. Subjective scoring of cough in children: parent-completed vs child- completed diary cards vs an objective method. European Respiratory Journal 1998;11:462-466.

7. American Thoracic Society. Standardization of spirometry: 1987 update. American Review of Respiratory Disease 1987;136:1285-1298.

8. Official statement of the European Respiratory Society. Standardized lung function testing. European Respiratory Journal - Supplement. 1993;16:1-100.

Send letter to journal:
Re: follow up of children with community acquired pneumonia

We read with interest your article on follow up of children with community acquired pneumonia and also your eletter response on the subject. I t is interesting to note that children with pre-existing asthma are at significant risk of persistent cough after admission to hospital for community acquired pneumonia. However the risk of subsequent asthma is increased only in children of non-atopic parents. In routine clinical practice the children with pneumonia are not normally followed up in the clinic and at the most seen once in the clinic to ensure resolution of consolidation. The current British thoracic society guidelines make no recommendation for long term follow up (2). We would like to note the following issues in your study.

1. Due to the retrospective nature of your study it was not possible to assess premorbid lung functions. Castro-Rodriguez et al found those with a diagnosis of pneumonia had lower levels of maximal flows at FRC at mean age of 2 mo (albeit not significantly) and at age 6 yr and lower levels of FEV1 and FEF25–75 at age 11 yr. They conclude that children with x-ray confirmed pneumonia have diminished airway function that is probably present shortly after birth. These deficits are at least in part due to alterations in the regulation of airway muscle tone (1). Therefore the question of whether preexisting abnormal lung function predisposes to pneumonia or of pneumonia resulting in lung function abnormality remains unanswered.

2. Bronchial hyperactivity was not determined in your study. Bronchial hyperactivity is far more common than asthma in children with pneumonia. In a study by korppi et al (3) bronchial hyperactivity indicated by methacholine inhalation challenge was present in 45% of the pneumonia group. All 10 asthmatic patients had bronchial hyperactivity, but only 20% of hyperactive children had asthma. In another study (4) bronchial hyperactivity was present in 42% of children with pneumonia at less than 2 years age when followed to a median age of 19 years. It is possible that certain percentage of children in the bronchial hyperactivity group develop asthma over a period of time. It is not very clear which factors predispose to the development of asthma.

3. Does the outcome of the study suggest that we should be following up all children with pneumonia? At present we do not have sufficient evidence to support a change of practice.

References

1. Castro-Rodriguez JA, Holberg CJ, Wright AL, et al. Association of radiologically ascertained pneumonia before age 3 yr with asthmalike symptoms and pulmonary function during childhood. Am J Respir Crit Care Med 1999;159:1891–7.

2.British Thoracic Society Standards of Care Committee. British Thoracic Society Guidelines for the Management of Community Acquired Pneumonia in Childhood. Thorax 2002;57(Suppl 1):i1–24.

Send letter to journal:
Re: Response to eletter by Egware B Odeka et al

We thank Odeka et al for their comments regarding our follow-up study of children admitted to hospital with community-acquired pneumonia (CAP). They are correct in their initial 2 observations that the study did not measure premorbid lung function or bronchial hyperreactivity. These measurements were not an objective of the present study therefore were not included in study design. Inclusion was limited by the retrospective nature of the study. We agree that further prospective work is required to determine the relationship between premorbid lung function, bronchial hyperreactivity and the development of chronic respiratory disease following CAP.

With regard to the question concerning whether we should be following up all children hospitalised with CAP. We do not have sufficient evidence currently to support this, based on our study findings. However, we would advise that parents of chidlren admitted with CAP are alerted, at time of discharge, to the need to consult their doctor should their child develop persistent cough or a new diagnosis of asthma or wheeze in the future. We have shown that a small percentage of such children may have significant chronic respiratory disease (supplemental file 4 of original article)1. There is a need for prospective work to delineate high risk groups further and to investigate an appropriate follow-up strategy.

Competing interests: None

References