Article Text

Download PDFPDF

Early rattles, purrs and whistles as predictors of later wheeze
  1. S W Turner1,
  2. L C A Craig2,
  3. P J Harbour1,
  4. S H Forbes1,
  5. G McNeill2,
  6. A Seaton2,
  7. G Devereux2,
  8. G Russell1,
  9. P J Helms1
  1. 1
    Department of Child Health, University of Aberdeen, Aberdeen, Scotland, UK
  2. 2
    Environmental and Occupational Medicine, University of Aberdeen, Aberdeen, Scotland, UK
  1. Dr S W Turner, Department of Child Health, University of Aberdeen, Royal Aberdeen Children’s Hospital, Foresterhill, Aberdeen AB25 2ZG, Scotland, UK; s.w.turner{at}


Background: Asthma is a common condition characterised by wheeze. Many different respiratory sounds are interpreted by parents as “wheeze” in young children.

Aim: To relate different respiratory sounds reported as wheeze in 2-year-olds to asthma outcomes at age 5 years.

Methods: As part of a longitudinal cohort study, parents completed respiratory questionnaires for their children at 2 and 5 years of age. Parents who reported wheeze were given options to describe the sound as rattling, purring or whistling.

Results: Of the 1371 2-year-olds surveyed, 210 had current wheeze, of whom 124 had rattle, 49 purr and 24 whistle. Children with whistle at 2 years were more likely to have mothers with asthma, and children with rattle and purr were more likely to be exposed to tobacco smoke. Wheeze status was ascertained at age 5 years in 162 (77%) children with wheeze at 2 years of age. Whistle persisted in 47% of affected children, rattle in 20%, and purr in 13% (p = 0.023). At 5 years of age, asthma medication was prescribed in 40% with whistle, 11% with rattle, and 18% with purr at 2 years of age (p = 0.017).

Conclusions: This study shows different risk factors and outcomes for different respiratory sounds in 2-year-olds: compared with other respiratory sounds, whistle is likely to persist and require asthma treatment in future.

Statistics from

Wheeze is defined as a high pitched, musical and/or whistling sound1 and is reported in more than 25% of children by the age of 2 years.2 Different respiratory noises made by young children are often interpreted as wheeze by parents,3 and the heterogeneity in outcomes for early wheeze reported in longitudinal studies4 5 might, at least in part, be explained by interobserver interpretation of respiratory sounds.6 Better understanding of the epidemiology and aetiology of different respiratory sounds is likely to lead to a more accurate definition of asthma in younger children, and this will benefit clinicians and researchers.

A large airway rattle is the noise most commonly confused with wheeze,3 and the mechanism is not known but is thought to be independent of asthma.7 The outcome of large airway rattle is not known, but Elphick et al6 have hypothesised that there are different outcomes among those with different respiratory sounds reported as wheeze. In our longitudinal birth cohort study, parental-reported wheeze at 2 years of age was further categorised as whistle, rattle or purr. A rattle is a coarse inspiratory and expiratory noise, whereas purr is a low-pitched inspiratory or expiratory sound, similar to that made by a cat. In this study, we tested the hypothesis that asthma outcomes at 5 years of age will differ among children with whistle, rattle and purr at 2 years of age.


Enrolment and protocol

Study subjects were participants in a longitudinal birth cohort study designed to relate dietary exposure in early life to asthma outcomes in childhood. Two thousand pregnant women were recruited at 12 weeks’ gestation; there was no selection for parental asthma or allergy. Complete details of enrolment are described elsewhere.8 At enrolment, mothers completed a questionnaire from which current maternal asthma status was identified. A modified ISAAC questionnaire9 was completed by parents when the child was 2 and 5 years of age and returned by post. Asthma was defined as affirmative responses to the questions “Has your child ever suffered from asthma?” and “Has this been diagnosed by a doctor?” Current wheeze was defined as an affirmative response to the question “Has your child wheezed in the last 12 months?” Parents were asked to categorise wheeze (if present) as whistle, rattle, purr or other sound. This study was approved by our institutional research ethics committee, and verbal assent was obtained from each child before the hospital assessment.


Student t test, χ2 analysis and analysis of variance (with Bonferroni correction) were used where appropriate. SPSS V14.0 was used for analyses, and significance was assumed at the 5% level.


Study subjects

There were 1924 liveborn infants, and respiratory questionnaires were completed for 1374 aged 2 years; data at age 5 years were available for 1112 of these (81%). Current wheeze was reported in 210 children aged 2 years, and respiratory questionnaire data were available for 162 of these (77%) at age 5 years, 62 of whom had persisting wheeze (fig 1). Reported wheeze at 2 years of age was categorised in 207 children as follows: 24 whistle, 49 purr, 124 rattle and 10 “other noises” (these 10 individuals were excluded from the analysis). We have previously reported that mothers whose children participated in the 5-year assessment were less likely to smoke and have asthma, to be older and of a higher socioeconomic class compared with all mothers at enrolment.10 Table 1 compares outcomes between children assessed at the age of 2 years who did and did not participate at age 5 years.

Figure 1 Consort diagram showing the number of children assessed at 2 years of age, and followed-up at the age of 5 years, and outcomes for reported wheeze at the earlier assessment. *The category of wheeze was not described in eight of these 62 children.
Table 1 Comparison of factors between children assessed at age 2 years and assessed or not assessed aged 5 years

Outcomes at 5 years of age for reported wheeze at age 2 years

Table 2 presents these results, and fig 1 summarises them. At 5 years of age, those children with whistle at 2 years were more likely to have current wheeze (73%) and to be receiving treatment for asthma (40%) compared with children who rattled (34% and 11%, respectively) or purred (39% and 18%, respectively) (p ⩽0.017 for both analyses). The positive predictive values for whistle, purr or rattle at 2 years of age for wheeze at 5 years were 73%, 39% or 34%, respectively; the corresponding negative predictive values were 88%, 89% and 90%.

Table 2 Comparisons of risk factors for wheeze and outcomes at 5 years of age for children with whistle, rattle and purr

Respiratory sounds at 2 years and associated factors

Children with whistle were more likely to have asthmatic mothers (38%) than children with rattle (23%) and purr (12%) (p = 0.047) (table 2). Children with rattle and purr were more likely to be exposed to tobacco smoke at home than a combined group of children with whistle and children with no wheeze (odds ratio 1.8 (95% CI 1.3 to 2.4); p = 0.001). There were no differences in gender between wheezing groups (table 2).

Consistency of whistle, rattle and purr between assessments

The character of the respiratory sound was identified by parents in 54 of the 62 children with wheeze at both 2 and 5 years of age, including 17 (31%) with whistle, 28 (52%) with rattle, and nine (17%) with purr. Whistle persisted at age 5 years in seven out of 15 (47%, including four who were prescribed asthma drugs) children, rattle in 19 out of 97 (20%), and purr in five out of 38 (13%) (χ22 = 7.56, p = 0.023). Three children with whistle aged 2 years were said to rattle at 5 years, none to purr, and five had ceased making any noise. Six children with rattle at 2 years of age were said to whistle aged 5 years, four to purr and 68 had become asymptomatic. Four children with purr at 2 years had whistle at 5 years, six had rattle, and 23 had no respiratory sound. The repeatability analysis (Cronbach’s α, a score of 0.7 or higher indicates good repeatability) for whistle, rattle and purr between the ages of 2 and 5 years was 0.52, and this value was 0.57 when only whistle and rattle were considered.


This is the first study to describe the natural history of different respiratory sounds in young children. The results from this unselected community-based cohort also confirms work based on populations at increased risk of respiratory illness showing that wheeze is commonly regarded by parents of young children as a term describing many different respiratory sounds.3 1113 Whistle persisted in many children, was a good predictor of persisting wheeze, and was associated with later treatment for asthma. Rattle and purr persisted in a sizeable minority of cases, but did not predict future wheeze particularly well. The results of this study may help clinicians to identify asthmatics among young children presenting with parental-reported wheeze.

What is already known on this topic

  • Parents report all respiratory noises as “wheeze”.

What this study adds

  • Most parent-reported “wheeze” has a rattling and not whistling quality.

  • Although rattle at 2 years of age mostly resolves, whistle persists in many and is associated with asthma outcomes.

Rattle and purr resolved in most cases, although by 5 years of age almost half of the children with persistent rattle and purr had at some stage been given a diagnosis of asthma, some being prescribed asthma medication. Asthma remains a clinical diagnosis that can be difficult to confirm in 5-year-old children, and some of the children with rattle and purr may have had asthma, and similarly some with whistle may not have had asthma. Asthma is more easily diagnosed in older children, and extended follow-up of this cohort will allow us to test the hypothesis that whistle at 2 years of age, and not rattle or purr, is symptomatic of asthma.

In addition to maternal diet,7 8 male gender, maternal asthma and tobacco smoke exposure are risk factors for asthma in young children,14 and the distribution of these attributes across the three symptomatic groups was not always consistent. Among the symptomatic children, the highest prevalence of maternal asthma was found for wheezy children with whistle, suggesting that maternal factors are important to whistle, and this is consistent with a previous study that reported maternal asthma to be more closely associated with childhood asthma than paternal asthma.15 In contrast, children with rattle and purr were more likely to be exposed to tobacco smoke at 2 years of age than children with whistle and with no wheeze. Exposure to tobacco smoke is associated with a number of childhood illnesses characterised by increased secretions, such as bronchitis and otitis media,16 and it is possible that rattle and purr may reflect increased airway secretions in the major airways stimulated by irritants in tobacco smoke. Boys were at increased risk of all respiratory noises, suggesting that male gender confers a non-specific increased risk of respiratory illnesses. Most children with wheeze at 2 years of age had non-asthmatic mothers and were not exposed to tobacco smoke; therefore additional factors are important to early wheeze, and these may include infective and non-asthmatic inflammatory conditions.

This study was not designed to validate the use of whistle, rattle and purr, but the findings indicate some consistency in reported whistle and rattle. The overall consistency of respiratory sounds, as evidenced by Cronbach’s α, was relatively poor, and this may be in part due to the 3-year interval between assessments. Although there was attrition in numbers of children with persistent wheeze as respiratory sounds resolved over time, there was a relatively good consistency in reported whistle (70%) and rattle (65%) among those studied. There was relatively poor consistency in reported purr, suggesting that this term was less well understood by parents.

Among parents who described their child as having a wheeze, the rattle to whistle ratio was more than 5:1 in 2-year-olds, but closer to 1:1 at 5 years, and this has implications for epidemiological studies of very young children that asked about wheeze, not otherwise specified. It would appear from this study that most parents seem to think that wheeze is any respiratory noise, of which whistle is one subset. Future questionnaires could be designed with this problem in mind, and should attempt to clarify what the parents of young children mean by “wheeze.”

The positive predictive value for whistle predicting future wheeze is comparable with studies that have used algorithms, including tests that are not routinely available, to predict persistence of wheeze.17 18 Our findings suggest that parents can reliably distinguish between different respiratory sounds made by children, and this readily obtained information has good prognostic potential in a clinical setting. Different outcomes for whistle, rattle and purr could explain some of the apparent differences in results from epidemiological and intervention studies; consideration could be given to including these questions in an epidemiology setting.

There are some potential limitations to this study. Firstly, there was some drop-out between the ages of 2 and 5 years (table 1). However, such drop-out would weaken the associations reported and not strengthen them, and so we do not believe that the outcomes reported here were substantially influenced by drop-out. Secondly, we argue that wheeze is a heterogeneous term, yet have reported wheeze at 5 years as an outcome. This was necessary because of the relatively small numbers involved, but we have shown that, among those with wheeze, whistle was most likely to persist. Finally, by creating subgroups of wheezing categories and with incomplete follow-up, there was an inevitable attrition in study subjects, in particular in the whistle group. Hence the positive predictive values reported were necessarily derived from relatively small numbers of participants and should be interpreted with some caution.

In summary, the results of this study suggest different mechanisms and outcomes for early wheeze depending on the character of the sound, ie, whistle, rattle or purr. The outcome of early wheeze is heterogeneous, with symptoms resolving in most but persisting in others.5 14 An element of this heterogeneity for studies reliant on parent-reported wheeze may be explained by different interpretations of wheeze by the observer. There is also considerable intersubject reporting of respiratory sounds among clinicians,19 and all studies of wheeze in young children are potentially weakened by inconsistency in interpretation of respiratory sounds. The interpretation of wheeze in young children is subjective, and further characterisation by “whistle” or “rattle/ruttle” may be useful in clinical and research practice.


We acknowledge and thank the parents and children who participated with this study.



  • Funding: This study was funded by Asthma UK who was not involved in the study design, the collection, analysis, and interpretation of data, the writing of the report or the decision to submit the paper for publication.

  • 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.