The Differences in Spirometry Predictive Equations in Classifying Presence and Degree of Lung Function Impairment: Which Suit Fits the Best?

Abstract

© 2017, Springer Science+Business Media, LLC. Purpose: The aim of this study was to compare the differences between three most commonly used predictive equations (PE): ECCS (European Community of Coal and Steel), the third National Health and Nutrition Examination Survey (NHANES III), and GLI (Global Lung Initiative) in healthy individuals and when grading severity of lung function impairment in patients with obstructive lung diseases. Methods: The study included 200 healthy volunteers and 200 patients with obstructive lung diseases at the Institute for Pulmonary Diseases of Vojvodina. In all subjects, we calculated the absolute and relative (percent) predicted values using ECCS, NHANES III, and GLI reference equations. Results: The mean differences between ECCS and NHANES III predicted values were 5.63% (95% CI 5.29–5.98%, p < 0.001) for FEV1 and 10% (95% CI 9.52–10.79%, p < 0.001) for FVC. Similar differences were observed between ECCS and GLI predicted values. There were minimal differences between NHANES III and GLI predictive values. In healthy subjects, the mean absolute difference between measured FEV1 and FEV1 ECCS predicted was 0.36l (95% CI 0.32l, 0.40l, p < 0.001), FEV1 NHANES predicted was 0.30 l (95% CI 0.27–0.35l, p < 0.001), and FEV1 GLI predicted was 0.31l (95% CI 0.27– 0.35l, p < 0.001). The use of three different PE leads to significant differences in classification of obstruction severity in both asthma and COPD patients. Conclusions: There were significant differences between FEV1 and FVC predicted when using different PE. The absolute difference between actual and predicted FEV1 in healthy individuals was highest when using ECCS. The use of different PE may change the interpretation of severity of airway obstruction.