RESULTS

Section:

ChooseTop of pageAbstractMETHODSRESULTS <<DISCUSSIONReferencesCITING ARTICLES

A total of 583 subjects underwent clinical evaluation; 201 subjects were excluded due to dyspnea (n = 24), cough (n = 17), wheezing (n = 13), or for several previously undetected diseases, such COPD (n = 11), asthma (n = 8), scoliosis (n = 3), or obesity (BMI ≥30 kg/m²) (n = 125). Of the 382 subjects (231 women and 151 men) who entered into the study, we found technically acceptable tests for 321 (189 women and 132 men). Sixty-one subjects (9.8% of men and 13.4% of women) were excluded from analysis because of claustrophobia (n = 2), incapacity to perform satisfactory panting maneuvers (n = 37), or repeated disconnections from the test gas during rebreathing (n = 22) (Figure 1). The elderly subjects who were excluded or were ineligible for the study were similar in age, height, and weight as those who were included.

Figure 1. Flow chart for subject selection.

Download Figure | Download Powerpoint

[More] [Minimize]

The age distribution of the women and men in the analyzed sample demonstrates adequate representation of the study population (see Table E1 in the online supplement). Details of the anthropometric characteristics, spirometric data, lung volumes, and airway resistance in both sexes are shown in Table 1. No significant differences in these parameters were found between excluded subjects and the analyzed sample.

In men and in women, the plethysmographic method provided higher values than the dilutional method for TLC, FRC, or RV (P < 0.001 in all comparisons) (Table 1). The mean differences of lung volumes (plethysmographic − helium dilution) and their 95% limits of agreement were wide, reflecting as great a within-subject variability as that from method-related differences (Figure 2). Because the 95% limits of agreement for static lung volumes include zero, it is not possible to conclude that the helium dilution method systematically underestimates static lung volumes in healthy individuals.

Figure 2. Differences between lung volumes measured by plethysmography and helium dilution plotted against the mean of the two values for TLC, FRC, and residual volume (RV) in healthy elderly women and men. The continuous line represents the mean difference; the dashed lines represent 2 SD around the mean.

Download Figure | Download Powerpoint

[More] [Minimize]

The reference equations for lung volumes from the healthy older European women and men are given in Tables 2 and 3. We did not find that the addition of transformations significantly improved the predictability of the regression equations. No significant interaction between age and height was found. Analysis of residuals showed that homoscedasticity was present in all equations. Regression analysis of these residuals showed neither statistically significant slopes nor correlation coefficients. The residuals corresponding to these models did not differ significantly from a Gaussian distribution in all lung volumes, as determined by the Shapiro-Wilk test. Therefore, one-sided lower 95% prediction intervals were used to determine the lower limit of normal lung functions (5). Because the use of weight or weight-related factors for reference equations would affect to the accuracy of prediction values among morbidly obese subjects, in Table 4 we provide the prediction equations for these lung volumes without weight-related factors.

Tables 5 and 6 show the differences between the observed values found in the subjects of our study and the values calculated from several prediction equations. Aside from our equations, the closest agreements for plethysmographic lung volumes were with Cordero and colleagues (9) and ECSC (5) in women and with Crapo and coworkers (4) in men. Regarding dilutional lung volumes, in women the closest agreements for TLC were with Crapo and colleagues (4) and ECSC (5), whereas the closest for FRC was with Cordero and coworkers (9) and ECSC (5) and for RV with Cordero and colleagues (9) and Matthys and coworkers (7). In men, the closest TLC agreements were with Crapo and colleagues (4) and Cordero and coworkers (9), for FRC with Crapo and coworkers (4) and Cotes and colleagues (10), and for RV with Matthys and colleagues (2) and Crapo and coworkers (4).

To compare our reference equations with other prediction equations, the difference in predicted lung volumes (present study equation – every other equation) by the mean predicted lung volumes were plotted for women and men, respectively. In women, Matthys and colleagues (7) and Roca and coworkers (8) overestimated FRCpleth and FRCHe, whereas Cotes and coworkers (10) underestimated this lung volume (Figure 3). In men, the ratio increased proportionally when the present prediction values for FRCpleth were compared with those from ECSC (5), whereas it decreased proportionally with respect to Cordero and coworkers (9) and Cotes and colleagues (10) (Figure 4). Roca and coworkers (8) and Matthys and colleagues (7) clearly overestimated FRCpleth in men. For women and for men, all the analyzed adult-derived reference equations overpredicted TLCpleth, TLCHe, and RVpleth (Figures E1–E12).

Figure 3. Difference between plethysmographic FRC and mean FRC predicted by the present study versus (A) ECSC and colleagues (5), (B) Crapo and colleagues (4), (C) Cordero and colleagues (9), (D) Roca and colleagues (8), (E) Matthys and colleagues (7), and (F) Cotes and colleagues (10) in women.

Download Figure | Download Powerpoint

[More] [Minimize]

Figure 4. Difference between plethysmographic FRC and mean FRC predicted by the present study versus (A) ECSC and colleagues (5), (B) Crapo and colleagues (4), (C) Cordero and colleagues (9), (D) ROCA and colleagues (8), (E) Matthys and colleagues (7), and (F) Cotes and colleagues (10) in men.

Download Figure | Download Powerpoint

[More] [Minimize]

DISCUSSION

Section:

ChooseTop of pageAbstractMETHODSRESULTSDISCUSSION <<ReferencesCITING ARTICLES

To our knowledge, the current study provides the first data in the international literature on reference values for static lung volumes and airway resistance in healthy older adults. Our results confirm that reference equations should not be extrapolated, in general, for ages or heights beyond those covered by the data that generated them. For patients over 65 years of age, the current study shows that the most commonly used sets of reference equations may lead to inaccurate interpretations.

The exact definition of a “healthy” group is difficult to agree upon. Although criteria to define subjects as healthy have been established in previous ATS statements (22), previous studies have used many different criteria. The ECSC reference equations (5, 6) were not based on original data; rather, they were derived from previously published reference equations. Moreover, the data on TLC, RV, and FRC are derived from populations that include smokers and exsmokers (6); therefore, these equations may not be representative of a healthy nonsmoking population. In the study by Cotes and colleagues (10), some exsmoker women were included in the reference group. In contrast, all of the current study patients were lifelong nonsmokers. Also, notable differences exist in the technique used to determine lung volumes. In some cases, the reference equations were obtained from measurements for single-breath helium dilution (4), in others from multiple breath helium dilution (5, 9, 10), and in others by plethysmography (7, 8).

Although it has been proposed to use the same reference values for plethysmographic and dilutional methods in middle-aged adults (6), the comparison of lung volumes measured by plethysmography and helium dilution shows statistically significant differences despite substantial interindividual variation (see Table 1; Figure 2). As a consequence of the reduction in supporting tissue around the airways, distal airways begin to close at a higher lung volume in elderly people than in younger people (12). Despite the absence of airway obstruction in our subjects, the existence of poorly- or nonventilated airspaces or a premature closure of the airways during tidal breathing (13) could explain the lower lung volumes obtained by helium dilution. In any case, the differences obtained between plethysmography and helium dilution are notable, as demonstrated by the wide limits of agreement for TLC (−0.29 to 0.73 L in women and −0.22 to 0.79 L in men) (see Figure 2). Because the difference obtained between the two procedures is not calculable beforehand, it seems necessary to use specific sets of reference equations for every procedure in elderly subjects.

Usually, lung volumes are related to body size, with standing height being the most important factor (19), but our results support the importance of other factors.

Although many reference equations for FRC only consider age and height (4–7), those of Cordero (9), Roca (8), and Cotes (10) include other body measurements, such as weight, BMI, or BSA. In our case, we verify that body size is an important factor related to FRC, expressed essentially as BSA. This result is consistent with the clinical observation that obesity has the greatest effect on FRC because the effect of body fat on lung volumes is determined by the decrease of total and chest wall compliance (13), which reduces FRC (23). However, the inclusion of BSA in the prediction equations for FRC may lead to misinterpretation of the results. Likewise, it has been suggested that the incorporation of body composition measurements, especially fat-free mass index, increases the accuracy of the reference equations for lung volumes (10). Nevertheless, it is difficult to recommend the incorporation of these measurements in all routine explorations, except in the most sophisticated lung function laboratories. The application of our reference equations to morbidly obese subjects might reduce their accuracy, and this would result in a significant tendency toward misclassification of obesity-associated restriction by undepredicting lung volumes. To avoid this problem, we have generated another set of reference equations without weight-related factors (see Table 4). The use of both prediction equations allows for distinguishing if high weight is the sole cause for a low lung volume or whether there is some other contributing process.

In addition to height dependence, our reference equations confirm the importance of incorporating age in the prediction of TLC in women and in elderly men. The age-related decrease in diaphragmatic strength (24) reduces maximal respiratory pressures in healthy elderly subjects (25) and can consequently reduce inspiratory capacity. A reduction in diaphragmatic strength, by about 25% in healthy elderly subjects compared with young adults (25), must have some effect on TLC.

Similarly, age was directly related to RV in women and in men. In addition to the increased closing volume, this can be justified by the age-related weakness of the expiratory muscles. Previously, it has been demonstrated that the expiratory intercostal muscles undergo atrophy with a decrease of approximately 20% in the mean cross-sectional area after the fifth decade (26). Moreover, with increasing age, the chest wall becomes less compliant, and the lungs become more distensible. As a result of the diminished lung elastic recoil, an increase in RV by approximately 50% between 20 and 70 years of age has been reported (13).

Practically all the estimations made by extrapolation of reference equations for middle-aged adults overpredicted the true lung volumes of our healthy elderly volunteers. An increase in static lung volume has been reported during normal ageing as a consequence of the decrease in the elastic recoil of the lung secondary to changes in the lung connective tissue (12). However, it is possible that these changes mimic those observed in emphysema and are of a lesser magnitude than expected. Biochemical studies suggest that the total lung content of collagen and elastin does not change with ageing and that the collagen becomes more stable because of increased numbers of intermolecular crosslinks (27). Morphometric studies in senescence-accelerated animal models have not shown evidence of cellular infiltration or a decrease in the ratio of lung weight to body weight, suggesting little or no lung destruction, as opposed to what is seen in emphysema (28). As a consequence of changes in the spatial arrangement or crosslinking of the elastic fiber network (29), a homogeneous enlargement of air spaces occurs, followed by discrete lung hyperinflation. The changes observed in senile lungs are morphologically and histologically different from emphysema, and their major consequences seem to be a tendency toward small airway collapse and a decrement in expiratory flow with ageing (13, 17).

An example of the overprediction of lung volumes is provided by the estimation of healthy subjects below or over the limits of normal. Because prediction equations derived from cross-sectional data are primarily used as a screening tool to identify individuals with lung function outside the expected range, the utility of any particular reference equation depends upon its ability to correctly identify individuals with lung function below the lower limit of normal (LLN) or over the upper limit of normal. Some authors have defined the LLN as that value above which the results of 95% of the normal population lie, working under the assumption that larger values have larger variances. However, if skewed distributions are transformed to normalize their shape, the subtraction of 1.645 SD may still be used to estimate the LLN. Middle-aged reference equations classify subjects as being below the plethysmographic TLC LLN between 11.8 and 55.5% of the women and between 14.1 and 25.9% of the men of the current study. In contrast, only 0.6 to 1.7% of the women and 0.1 to 0.3% of the men were above the upper limit of normal of plethysmographic RV.

In conclusion, we have developed reference equations for the prediction of static lung volumes in older adults. Differences among studies in predictions of lung volumes or in the identification of individuals with lung function values outside the limits of normal may be due to differences in the age range of the reference subjects but are also likely to be contributed to by different measurement methods and other differences in the underlying populations. These results underscore the importance of using prediction equations appropriate to the age and height characteristics of the population to whom inferences are to be applied. Our reference equations may be used for the prediction of lung volumes in white patients 65 to 85 years of age with standing heights for women between 144 and 174 cm and men between 152 and 181 cm. The use of our reference equations beyond these age and height limits may lead to inaccuracies.