Predicting whether wheeze is due to asthma

Despite growing evidence about different wheezing phenotypes, it remains difficult to exclude or diagnose asthma in young children presenting with their first episode of wheeze.¹ A particular challenge is to distinguish children with transient wheezing from children whose wheezing persists and who later develop asthma.⁷

Airway tissue inflammation leading to airway remodelling occurs at an early age and is fundamental to the development of asthma. While characteristic features of inflammation such as increased eosinophil levels (eosinophilia) can be used to distinguish asthma-related wheezing from wheezing caused by viral infections, these markers can also occur in other allergic conditions not related to asthma.²¹ Until unique markers for distinguishing asthma from other respiratory conditions are identified,²¹ predicting whether wheeze is due to asthma will depend on the integration of different types of information gathered over time.⁸

Evidence from different longitudinal studies has been used to develop indices to assist health care professionals in determining whether episodes of wheeze in young children are due to asthma.^22,23

One such set of indices suggests that nearly 60 per cent of children with any wheeze and close to 80 per cent of children with frequent wheeze are likely to develop asthma if their wheezing is associated with the following risk factors:

• parents with asthma or eczema; and
• two out of the following three features of atopy – eosinophilia, wheezing without colds or allergic rhinitis.

Significantly, 95 per cent of children without these risk factors are not likely to develop asthma.²²

Another predictive model, this time in children at higher risk of asthma due to family history, combined age at presentation and concentration in the blood serum of soluble interleukin-2 receptor (an indicator of eosinophilia) and found that wheeze was more likely to be persistent in older atopic children with two atopic parents. The model was around 70 per cent accurate in predicting clinical outcome and had over 90 per cent chance of predicting children who were destined to become asymptomatic.²³

While the negative predictive value of these indices is much greater than their positive predictive value, they can still be used to indicate subsequent development of asthma with reasonable accuracy and a low risk of including asymptomatic children.^22,23 Further research will continue to improve the clinical usefulness of predictive models.

The box below summarises stages in confirming a diagnosis of asthma that highlights some important predictors of asthma.¹