To the Editors:
Recently, an American Thoracic Society (ATS)/European Respiratory Society (ERS) Task Force suggested the need for a new Europe-wide study to derive updated reference equations for lung function 1. Until the results of such a study become available (we have been waiting for new reference values in Europe for ∼25 yrs), the official reference values in Europe are still those implemented by the European Coal and Steel Community (ECSC) 2 and the ERS 3. The ATS/ERS Task Force has not commented on newer reference values for spirometry derived from the Swiss Study on Air Pollution and Lung Disease in Adults (SAPALDIA) 4, 5.
It has previously been pointed out that the main differences between the ECSC and the SAPALDIA reference values comprise the following two factors: 1) a shift to higher reference values; and 2) a nonlinear age dependency derived from newer studies 6. It is highly plausible, and it has been corroborated by newer reference values, that there is a natural growth in values, followed by a steady change over and then a slow but growing decrease with age; the values are near optimal in ∼18-yr-old females and ∼25-yr-old males 4, 7.
It is of great importance, especially for longitudinal epidemiological studies, that age dependency at younger ages is highly different between the ECSC values (where it remains constant between ages 18 and 25 yrs, and is linear thereafter) and the newer reference systems (where it is nonlinear and shows a maximum). For example, the decrease in forced vital capacity (FVC) for a 180-cm tall male aged 21–50 yrs using the ECSC reference values is 22.4 mL·yr−1, but only 9.1 mL·yr−1 using reference values derived from SAPALDIA. Thus, decreases in FVC in longitudinal epidemiological studies will be detected with lower sensitivity if one uses the ECSC reference values.
The ECSC reference values for spirometry are not derived from measured data but, according to 2, from older reference equations that were obviously a combination of those from different researchers: “For each of the regression equations, a set of reference values was computed for each combination of height and age within the ranges given by the author [...] The summary equations as well as the residual standard deviations were calculated without weighting for numbers of subjects [...] In most publications, lung function is assumed to decline linearly with age in adults. However there is evidence that in young adults this is not the case (8 citations), and that the transition between the growth in adolescents and the decay with age in adults occurs at about 18–30 years of age [...]” 2. We want to add that it is not known whether smoking was considered. Approximately 10 yrs ago, Baur et al. 6 and Roca et al. 8 concluded that the present European recommendations on lung function reference values should be reconsidered, because “[...] the errors about the ECSC equations showed the most prominent underestimation of both predicted FVC [...] and predicted FEV1 [forced expiratory volume in one second]” 8.
Newer reference values derived from SALPALDIA are based on ∼3,000 healthy nonsmoking 18–60-yr-old Europeans and meet the methodological, epidemiological and statistical criteria of the ATS for the selection of reference values 9, 10.
We suggest that the European Respiratory Society should withdraw immediately their official statement to use the European Coal and Steel Community reference values for spirometry as these are both incorrect and misleading. Furthermore, we question the need for a new study as an appropriate alternative is already available.
Statement of interest
- © ERS Journals Ltd