How to evaluate "spirometric" lung age--what method is approvable?

Estimation of lung age via a spline method and its application in chronic respiratory diseases

Lung age is a simplified concept that makes spirometry data easier to understand, but it is not widely used due to limitations in estimation methods. The aim of this study was to develop new equations to estimate lung age and to explore the application value of lung age in chronic respiratory diseases. Retrospective spirometric data of 18- to 80-year-old healthy subjects were used to develop the lung age estimation equations. Models were respectively built by multiple linear regression, piecewise linear regression, and the natural cubic spline method. Patients with chronic obstructive pulmonary disease (COPD) and asthma were subdivided into stages I–IV according to the severity of airflow limitation under the recommendation of the Global Initiative for Chronic Obstructive Lung Disease. Propensity score matching was performed to balance age, height and sex between healthy subjects and patients. The difference between lung age and chronological age (∆ lung age) of patients with COPD and asthma was analyzed. A total of 3409 healthy subjects, 280 patients with COPD and 285 patients with asthma data were included in the analysis. The lung age estimation equation with the best goodness of fit was built by the spline method and composed of FEV₁, FEF_50%, FEF_75% and height as explanatory variables. ∆ lung age progressively increased with the degree of airflow limitation in patients with COPD or asthma. Lung age estimation equations were developed by a spline modeling method. Lung age may be used in the assessment of chronic respiratory patients.

Novel equations better predict lung age: a retrospective analysis using two cohorts of participants with medical check-up examinations in Japan

Background:

The lung age equations developed by the Japanese Respiratory Society encounter several problems when being applied in a clinical setting.

Aims:

To establish novel spirometry-derived lung age (SDL age) equations using data from a large number of Japanese healthy never-smokers with normal spirometric measurements and normal body mass indices (BMIs).

Methods:

The participants had undergone medical check-ups at the Center for Preventive Medicine of St Luke's International Hospital between 2004 and 2012. A total of 15,238 Japanese participants (5,499 males and 9,739 females) were chosen for the discovery cohort. The other independent 2,079 individuals were selected for the validation cohort. The original method of Morris and Temple was applied to the discovery cohort.

Results:

As a result of the linear regression analysis for forced expiratory volume in 1 s (FEV₁), spirometric variables using forced vital capacity (FVC) improved the adjusted R² values to greater than 0.8. On the basis of the scatter plots between chronological age and SDL age, the best model included the equations using FEV₁ and %FVC in females and males (R²=0.66 and 0.55, respectively), which was confirmed by the validation cohort. The following equations were developed: SDL age (females)=0.84×%FVC+50.2–40×FEV₁ (l) and SDL age (males)=1.00×%FVC+50.7–33.3×FEV₁ (l).

Conclusions:

This study produced novel SDL age equations for Japanese adults using data from a large number of healthy never-smokers with both normal spirometric measurements and BMIs.

Novel regression equations predicting lung age from varied spirometric parameters

Although lung age calculated backward from regression formulas constructed for FEV(1) estimation is widely used, it possesses a couple of faults. We developed novel equations predicting lung age from varied spirometric parameters (spirometry-derived lung age (SDL-age)). Applying multiple regression analysis, equations predicting SDL-age were invented using data from 8015 never-smokers with normal spirometry (group I). Validation was made based on data from 6398 never-smokers with normal spirometry (group II). Equations were further applied for 446 subjects with airflow limitation. FEV(1), FEV(1)/FVC, FEF(50), and PEF were selected as explanatory variables for reference value of SDL-age. Normal limits of difference between SDL-age and chronological-age were ± 13.4 years in the male and ± 15.0 years in the female. Established equations predicted SDL-age of group II. SDL-age was older than chronological-age only in subjects with severe airflow limitation. Novel regression equations allowing prediction of reference value of SDL-age and normal limits of difference between SDL-age and chronological-age were elaborated in both genders.