Comparison of different reference values for lung function: implications of inconsistent use among centers

Abstract

Abstract Background For interpretation of pulmonary function tests (PFTs), reference values based on sex, age, height and ethnicity are needed. In Norway, the European Coal and Steel Community (ECSC) reference values remain widely used, in spite of recommendations to implement the more recent Global Lung Function Initiative (GLI) reference values. Objective To assess the effects of changing from ECSC to GLI reference values for spirometry, DLCO and static lung volumes, using a clinical cohort of adults with a broad range in age and lung function. Methods PFTs from 577 adults (18–85 years, 45% females) included in recent clinical studies were used to compare ECSC and GLI reference values for FVC, FEV1, DLCO, TLC and RV. Percent predicted and lower limit of normal (LLN) were calculated. Bland-Altman plots were used to assess agreement between GLI and ECSC % predicted values. Results In both sexes, GLI % predicted values were lower for FVC and FEV1, and higher for DLCO and RV, compared to ECSC. The disagreement was most pronounced in females, with mean (SD) difference 15 (5) percent points (pp) for DLCO and 17 (9) pp for RV (p < 0.001). With GLI, DLCO was below LLN in 23% of the females, with ECSC in 49% of the females. Conclusions The observed differences between GLI and ECSC reference values are likely to entail significant consequences with respect to criteria for diagnostics and treatment, health care benefits and inclusion in clinical trials. To ensure equity of care, the same reference values should be consistently implemented across centers nationwide.

Keywords

• Lung function tests
• Reference values
• Interpretation
• Clinical aspects