Introduction: In 2021, the Global Lung Function Initiative (GLI) released reference equations for static lung volumes. We assessed the effect on interpretation of lung volumes in an adult population when changing to GLI from other commonly used reference equations.

Method: 2,353 sets of lung volume results (51% male, mean age 60 years) performed from 2015 to 2019 in an Australian public hospital pulmonary function laboratory were analysed. Mean predicted values and upper and lower limits of normal were calculated using each of the GLI, ECSC and Goldman equations. The proportion of patients with TLC and RV/TLC results outside the normal range were determined.

Results: GLI predicted values for TLC were similar to those of ECSC (mean (SD) difference: 0.07 ± 0.13L), and higher than that of Goldman (0.40 ± 0.24L). Agreement in the proportion of patients with a TLC outside the normal range was superior for GLI compared to ECSC (Cohen?s kappa 0.84 for ECSC vs. 0.76 for Goldman). The predicted GLI RV/TLC was lower than that of both ECSC and Goldman (-6 ± 2, and -5 ±3, respectively), resulting in a higher proportion of results being interpreted as outside the normal range when using GLI equations.

A change in TLC classification to normal occurred more frequently in males moving from Goldman to GLI (males: 10.9%, females 7.3%) and a shift to abnormal occurred more frequently in females moving from ECSC to GLI (females: 7.2%, males: 0.2%).  

Conclusion: When adopting the GLI equations, the most significant alteration in interpretation of lung volumes will occur when changing from the Goldman equations.