抄録
Background: We investigated the relationship between peak expiratory flow (PEF), age and standing height in 2785 Japanese volunteers (1047 males), aged 15-84 years, who had never smoked and who satisfied other strict criteria of normality. Many reports of regression equations for PEF by country and ethnicity are prepared using Wright or mini-Wright PEF meters, which have been calibrated by human subjects. Yet, no study has been performed in any country to determine the reference values of PEF on the American Thoracic Society (ATS) scale, which is calibrated by a computer-driven mechanical pump.
Methods: Peak expiratory flow was measured with mini-Wright meters calibrated on the Wright scale. All subjects were taught how to perform a forced expiratory maneuver: the highest of three PEF values was recorded and standing height was measured. The Miles equation was used to convert mini-Wright PEF values (traditional scale) to values using the new 'mechanical' PEF, which the ATS has recommended (ATS scale). In the analysis of the data, a model based on age, age squared and age cubed was used to derive curvilinear regression equations for PEF on age and standing height for each sex.
Results: There was adequate representation of subjects of each sex at all ages to 74 years. Curves plotted from the regression equations rose during adolescence and early adulthood, reached maximum values at 35 years in males and 40 years in females and then declined in an approximately linear manner. For both sexes, standing height fell progressively with increasing age.
Conclusions: From the regression equations, predicted values of PEF can be derived for any Japanese adult aged 15-74 years. We were able to obtain predicted equations for PEF in normal Japanese adults using both the Wright and ATS scales. Direct comparison of our regressions with those reported in other populations was limited by differences in methodology and analysis. In comparable studies of Chinese and Indian populations, the PEF values in those studies were appreciably lower than ours. Our regressions were remarkably similar to those reported in a study of British subjects that used virtually identical entry criteria and methods.
