Validation of the GLI-2012 spirometry reference equations in Argentinian children.

Abstract

The Global Lung Function Initiative (GLI) has produced spirometry reference equations for use in different ethnic groups. Previous reports have shown that the GLI equations do not adequately describe lung function in all populations and that adopting the new equations into clinical practice can increase the number of patients considered to have abnormal lung function. Therefore, before adopting these equations into local practice it is necessary to establish how well the equations represent the local population. The present study was conducted to determine how well the GLI spirometry reference equations represented the young children in Argentina, a population not included in the GLI dataset. Spirometry was measured in 2072 healthy children (50.9% males) aged 3.0-12.4 years (mean 6.64 ± SD 1.39), with a height range of 93.0-158.5 cm and weight range from 13.1 to 54.7 kg. We used the GLI "Caucasian" and "other/mixed" race equations to create Z-scores. The Z-scores predicted by the Caucasian GLI equations did not differ from zero and fitted the data well. Z-scores calculated using "other/mixed race" fit less well. Using the GLI definition of low lung function (Z-score <1.65) 6.8% of our healthy population had abnormal FVC, 4.9% had abnormal FEV1, 5.9 % had abnormal FEV0.75, and 3.9% had abnormal FEF25-75 when using the "Caucasian" GLI equation. This compares well with the expected 5% below the lower limit of normal. We recommend the use of the GLI-2012 Caucasian equations for spirometry undertaken in Argentinian children.