Classification of asthma severity: should the international guidelines be changed?

BACKGROUND

International guidelines recommend that, in addition to symptoms and medication requirements, measurements of forced expiratory volume in one second (FEV1) and peak expiratory flow (PEF) are necessary for the objective assessment of asthma severity. The guidelines suggest that parity exists between measurements of FEV1 and PEF when expressed as percentage of predicted normal values, and that asthma severity can be classified as mild, moderate or severe on the basis of FEV and PEF measurements of > 80%, 60-80% and < 60% of predicted values, respectively.

OBJECTIVE

To determine the relationship between measurements of FEV1 and PEF when expressed as percentage predicted values.

METHODS

A total of 1198 paired measurements of FEV1 and PEF were obtained from the medical records of a random sample of 25 adult asthmatic patients attending a hospital-based chest clinic. Measurements of lung function were expressed as a percentage of predicted normal values, using the European Respiratory Society prediction equations for PEF and FEV1. For the individual paired measurements, the mean differences between PEF and FEV percentage predicted were calculated. Measurements of lung function were used to determine asthma severity with <60%, 60-80%, and >80% predicted FEV1 and PEF values representing severe, moderate and mild asthma, respectively. The proportion of paired measurements in which differences in classification resulted from the use of FEV1 or PEF percentage predicted values was then calculated.

RESULTS

In asthma of differing severity, there was considerable variability between measurements of FEV1 and PEF when expressed as percentage predicted values; calculation of the FEV1% predicted resulted in lower values than those of the PEF percentage predicted, with a mean difference of -17.2% (95% CI -16.3%, -18.1%). There was agreement in classification of asthma severity in only 49.9% (598/1198) of paired measurements. Different prediction equations, while variably altering the degree of misclassification, did not correct the basic differences in the assessment of asthma severity dependent on the use of FEV or PEF.

CONCLUSION

FEV1 and PEF values, expressed as percentage predicted, are not equivalent. Pending further evaluation, the authors suggest that published asthma guidelines should avoid the assumption of parity between these two measurements.