Abstract
Background.
Previous cross-sectional studies demonstrate positive associations of fat-free mass and negative associations of centrally distributed fat deposits with respiratory function in older adults. Few studies have evaluated whether greater losses of muscle and increases in fat are associated with more rapid decline in respiratory function in aging.
Methods.
Nine hundred and fifty-seven men and 1,024 women aged, respectively, 73.6 ± 2.8 years and 73.2 ± 2.8 years at baseline were followed for 5 years. Body weight, waist circumference, bone mineral density, fat-free mass, fat mass and fat mass percentage as measured by DXA, abdominal subcutaneous and visceral adipose tissue, thigh muscle area, thigh intermuscular fat by CT and forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) were evaluated at baseline and after 5-years follow-up.
Results.
Cross-sectional analyses showed that height and thigh muscle area were positively and visceral adipose tissue negatively related to FEV1 and FVC. Increase in fat mass over five years was associated with concurrent FEV1 and FVC decline. In analyses stratified by weight-change categories, men and women who gained weight (vs stable/lost weight) had more rapid declines in FEV1 and FVC.
Conclusion.
In this well-functioning cohort, less muscle and greater abdominal fat were each associated with poorer lung spirometry cross-sectionally, whereas increase in fat mass over 5 years was associated with concurrent FEV1 and FVC decline. Weight gain and accompanying fat deposition may accelerate age-related declines in respiratory function.
Keywords: Aging, Lung function, Body composition
LUNG function deteriorates gradually throughout adult life even in the healthy elderly population, with a progressive decline in forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC), independent of smoking or environmental exposure (1,2). Worsening pulmonary function is clinically relevant as associated with increased morbidity and mortality rates (3,4).
Several factors may contribute to age-dependent decline of lung function. With aging, there is loss of muscle mass, a process called sarcopenia (5), and an increase of the amount of fat inside and around muscles (6). Both of these changes have previously been found associated with an increased risk of physical functional decline and disability as well as to an increase in morbidity and mortality (7).
Cross-sectional studies demonstrate, especially in men, associations of lower fat-free mass and greater centrally distributed fat deposits with poorer lung function in older adults (8–10). These observations have been recently confirmed in a 7-year follow-up longitudinal study (11).
No previous study has evaluated the associations of changes in CT-derived muscle and intermuscular fat (IMF) area and abdominal distribution of fat with concurrent deterioration in lung function in aging.
In a large cohort of older adults living in the community, we evaluated associations of body composition evaluated by DXA and CT with lung function, cross-sectionally and over time, to test the hypothesis that body composition changes, particularly abdominal fat deposition, may co-occur with lung function decline in aging.
METHODS
Study Population
The Health Aging and Body Composition Study (Health ABC) was initiated in 1997 to prospectively evaluate the relationship between changes in body composition and functional decline among older adults living in the community (12). Eligibility criteria included age 70–79 years during the recruitment period, no reported difficulty with walking a quarter of a mile, climbing 10 steps without resting, or performing activities of daily living. Exclusion criteria included current treatment for cancer, enrolment in a trial of a lifestyle intervention, plans to move out of the study area in 3 years, difficulty in communicating with study personnel, and cognitive impairment. Of the 3,075 men and women enrolled in Health ABC, 2,863 had lung function measured at the baseline visit. Of the 2,138 of these participants who completed the Year 5 (2001/02) visit, this analysis is restricted to the 1,981 participants who had complete data on change in lung function from baseline to Year 5. Reasons for missing measurements at either year included equipment problems (N = 6), participant unable to understand the examiner's instructions (N = 19) or physically perform the procedure (N = 20), participant refusal (N = 12), and participant medically excluded (N = 291). Medical exclusions included self-reported history of the following conditions or procedures within the past 2 months: surgery on chest or abdomen (N = 35), heart attack (N = 6), hospitalization for other heart problem (13), detached retina or eye surgery (N = 187), or measured blood pressure, systolic blood pressure >199 mmHg or diastolic blood pressure >109 (N = 25).
The study was approved by the Institutional Review Board of the University of California, San Francisco (H5254-12688-14), at the University of Tennessee (95-05531-FB) and at the University of Pittsburgh (#960212), and all participants provided written informed consent.
Demographic Data, Health Conditions, and Body Composition
Age, sex, race, and prevalent health conditions were assessed based on self-report and medication inventory. Height was measured using a wall-mounted stadiometer and body mass index was calculated as weight (kg)/height (m2). Body weight was measured using a calibrated standard balance beam scale to the nearest 0.1 kg.
Body composition at baseline and follow-up were measured by using fan-beam DXA (QDR4500A Hologic, Bedford, MA). The validity and reproducibility of the body composition data in the Health ABC Study were reported previously (14–17).
The cross-sectional areas of muscle, subcutaneous and IMF in both thighs, abdominal visceral adipose tissue (VAT), and subcutaneous adipose tissue were measured at baseline and follow-up by CT scan as described previously (18,19).
Spirometric Tests
Spirometric tests were performed at baseline and 5-year follow-up using a horizontal dry rolling seal spirometer (SensorMedics Corporation, Yorba Linda, CA) as previously described (20).
The spirometers were modified at the National Institute of Occupational Safety and Health (Morgantown, WV) under the direction of Dr John Hankinson by adding an optical shaft encoder to measure piston displacement and installing the spirometry software used in the third National Health and Nutrition Examination survey (19). A minimum of three and a maximum of five tracings were obtained from each subject.
FVC and FEV1 had to meet American Thoracic Society (ATS) criteria for acceptability and reproducibility (21). A quality score on a scale of 0–4 (4 being optimal, 0 being worst) was assigned for FEV1 and FVC.
Statistical Analysis
Chi-square tests and t tests were used to compare baseline characteristics, body composition, and change in respiratory function (Year 5 − Year 1) among black and white participants within each sex. Analysis of variance was used to compare FEV1 and FVC percent change across weight-change categories, defined as: gain (≥3% gain), loss (≥3% loss), stable (within 3% loss or gain) from Year 1 to Year 5 (22,23). Adjusted differences in mean delta FEV1 and FVC percent change across weight-change categories were evaluated using analysis of covariance controlling for age, race, height, smoking, emphysema, and chronic bronchitis.
Separate linear regression models were used to evaluate the cross-sectional association of each body composition measure as the independent variable and each pulmonary function measure as the dependent variable at baseline. Simple models were adjusted for clinic site, age, sex, race, height, and FEV1/FVC quality score. Full models were built using a forward procedure (p-entry = .20) to adjust additionally for smoking, cardiovascular disease, osteoporosis, emphysema, and chronic bronchitis after entering in the body composition measure. This selection procedure was repeated to build a parsimonious model of all body composition measures as independent variables and each pulmonary function measure as the dependent variable. Body composition variables were centered to reduce multicollinearity as evaluated by variance inflation factors and condition indices.
To evaluate whether changes in body composition and respiratory function may co-occur in aging, we used separate linear regression models of change in each body composition measure as the independent variable and change in each pulmonary function measure as the dependent variable. Adjusted models were selected as described above; the forward procedure was then repeated to build a parsimonious model of changes in all body composition measures as independent variables and change in each pulmonary function measure as the dependent variable. Analyses were repeated after exclusion of participants with prevalent emphysema or chronic bronchitis and after additional adjustment for baseline FEV1 or FVC. Adjusted p values (reported in text) were corrected for multiple comparisons using a Bonferroni correction. Analyses were performed in SAS (version 9.1; SAS Institute, Inc., Cary, NC) (24).
RESULTS
Characteristics of the four gender–race groups are summarized in Table 1.
Table 1.
Baseline Characteristics, Body Composition, and Change in Respiratory Function (Y5 − Y1) (M ± SD or n [%]), Stratified by Sex and Race
| Year 1 | Men | Women | ||||
| Total (n = 957) | White (n = 652) | Black (n = 305) | Total (n = 1024) | White (n = 605) | Black (n = 419) | |
| Demographics and health | ||||||
| Age (y) | 73.6 ± 2.8 | 73.7 ± 2.8 | 73.5 ± 2.8 | 73.2 ± 2.8 | 73.4 ± 2.8* | 73.0 ± 2.8* |
| Cardiovascular disease | 284 (30.1) | 204 (31.5) | 80 (27.0) | 191 (19.1) | 98 (16.4)† | 93 (23.0)† |
| Chronic bronchitis | 18 (1.9) | 14 (2.2) | 4 (1.3) | 48 (4.7) | 28 (4.7) | 20 (4.8) |
| Emphysema | 33 (3.5) | 22 (3.4) | 11 (3.6) | 19 (1.9) | 11 (1.8) | 8 (1.9) |
| Osteoporosis§ | 27 (2.9) | 15 (2.3) | 12 (4.0) | 263 (26.1) | 212 (35.8)‡ | 51 (12.3)‡ |
| Current smoker | 85 (8.9) | 29 (4.5) ‡ | 56 (18.4)‡ | 84 (8.2) | 44 (7.3) | 40 (9.6) |
| Former smoker | 561 (58.7) | 412 (63.5)‡ | 149 (48.9)‡ | 335 (32.8) | 195 (32.3) | 140 (33.5) |
| Height (m) | 1.73 ± 0.07 | 1.73 ± 0.06 | 1.73 ± 0.07 | 1.60 ± 0.06 | 1.59 ± 0.06 | 1.60 ± 0.06 |
| Weight (kg) | 81.6 ± 13.1 | 81.5 ± 12.5 | 81.7 ± 14.2 | 70.0 ± 14.5 | 65.8 ± 11.8‡ | 76.1 ± 15.8‡ |
| BMI (kg/m2) | 27.2 ± 3.8 | 27.1 ± 3.7 | 27.3 ± 4.1 | 27.4 ± 5.3 | 25.9 ± 4.4‡ | 29.6 ± 5.8‡ |
| Waist circumference (cm) | 101.0 ± 11.8 | 101.7 ± 11.5† | 99.5 ± 12.3† | 97.6 ± 13.7 | 95.1 ± 12.1‡ | 101.1 ± 15.1‡ |
| DEXA | ||||||
| Total hip bone density (gm/cm2) | 0.97 ± 0.15 | 0.95 ± 0.14‡ | 1.03 ± 0.15‡ | 0.81 ± 0.14 | 0.77 ± 0.13‡ | 0.87 ± 0.15‡ |
| Fat mass (kg) | 24.4 ± 7.1 | 24.8 ± 7.0* | 23.7 ± 7.4* | 28.9 ± 9.2 | 26.7 ± 7.7‡ | 31.9 ± 10.2‡ |
| Fat mass percentage (%) | 29.5 ± 4.8 | 29.9 ± 4.6‡ | 28.4 ± 5.0‡ | 40.4 ± 5.7 | 39.8 ± 5.4‡ | 41.1 ± 6.0‡ |
| Fat-free mass (kg) | 57.1 ± 7.3 | 56.6 ± 6.9† | 58.0 ± 7.9† | 41.1 ± 6.1 | 39.1 ± 5.1‡ | 44.0 ± 6.2‡ |
| CT | ||||||
| Subcutaneous adipose tissue (cm2) | 230.6 ± 87.9 | 226.0 ± 80.7* | 240.8 ± 101.3* | 332.5 ± 119.2 | 303.8 ±100.8‡ | 375.2 ±131.2‡ |
| Visceral adipose tissue (cm2) | 157.3 ± 70.6 | 169.9 ± 70.5‡ | 130.5 ± 62.9‡ | 129.4 ± 58.1 | 131.2 ± 60.8 | 126.7 ± 53.7 |
| Total thigh muscle area (cm2) | 265.3 ± 43.2 | 257.6 ± 38.4‡ | 282.0 ± 47.8‡ | 185.0 ± 34.1 | 170.7 ± 26.9‡ | 206.1 ± 32.8‡ |
| Total thigh IMF area (cm2) | 19.5 ± 12.5 | 18.6 ± 12.3‡ | 21.5 ± 12.7‡ | 20.3 ± 12.0 | 17.2 ± 8.9‡ | 24.8 ± 14.4‡ |
| Lung function | ||||||
| FEV1 baseline (L) | 2.59 ± 0.62 | 2.71 ± 0.60‡ | 2.31 ± 0.55‡ | 1.83 ± 0.41 | 1.91 ± 0.41‡ | 1.72 ± 0.39‡ |
| FEV1 Δ (L) | −0.22 ± 0.28 | −0.24 ± 0.28‡ | −0.16 ± 0.27‡ | −0.17 ± 0.23 | −0.19 ± 0.18* | −0.15 ±0.29* |
| FVC baseline (L) | 3.48 ± 0.71 | 3.66 ± 0.67‡ | 3.10 ± 0.63‡ | 2.40 ± 0.51 | 2.54 ± 0.48‡ | 2.21 ± 0.48‡ |
| FVC Δ (L) | −0.17 ± 0.35 | −0.20 ± 0.34† | −0.12 ± 0.35† | −0.15 ± 0.31 | −0.17 ± 0.26* | −0.13 ± 0.36* |
| FEV1 baseline predicted (L) | 2.80 ± 0.37 | 2.90 ± 0.35‡ | 2.59 ± 0.34‡ | 1.90 ± 0.30 | 2.05 ± 0.25‡ | 1.67 ± 0.23‡ |
| FEV1 Δ predicted (L) | −0.16 ± 0.07 | −0.19 ± 0.06‡ | −0.12 ± 0.07‡ | −0.14 ± 0.06 | −0.16 ± 0.06‡ | −0.11 ± 0.06‡ |
| FVC baseline predicted (L) | 3.84 ± 0.50 | 4.00 ± 0.44‡ | 3.49 ± 0.42‡ | 2.50 ± 0.41 | 2.73 ± 0.31‡ | 2.17 ± 0.29‡ |
| FVC Δ predicted (L) | −0.17 ± 0.09 | −0.20 ± 0.08‡ | −0.11 ± 0.09‡ | −0.18 ± 0.08 | −0.19 ± 0.07‡ | −0.16 ± 0.08‡ |
Notes: BMI = body mass index; FEV1 = forced expiratory volume in 1 second; FVC = forced vital capacity; IMF = intermuscular fat.
Self-report or total hip BMD T-score ≤ −2.5, Δ = Year 5 − Year 1.
*p < .05, †p < .01, ‡p < .001 for t test or chi-square test for racial differences within sex.
White men had on average higher waist, fat mass (FM), fat mass percentage, VAT but lower total hip bone density, fat-free mass, subcutaneous adipose tissue, total thigh muscle area, total thigh IMF area compared with black men. At baseline, white women were older and had on average lower weight, body mass index, waist, FM, fat mass percentage, fat-free mass, total thigh muscle area, total thigh IMF compared with black women. Compared with black men and women, white men and women presented higher FEV1, FVC, FEV1 percent predicted, and FVC percent predicted at baseline and presented greater decrease in both compared with black men and women over 5 years.
Table 2 gives characteristics of the study participants at baseline and 5-year follow-up. Of the 1,981 participants, 51.7% were female and 36.6% were black.
Table 2.
Baseline Characteristics and Change in Body Composition and Respiratory Function (Y5/Y6 − Y1) (M ± SD or n [%])
| n = 1981 | Year 1 | Year 5 | Δ | %Δ | p Value |
| Demographics and health | |||||
| Age | 73.4 ± 2.8 | — | — | — | — |
| Female | 1024 (51.7) | — | — | — | — |
| Black | 724 (36.6) | — | — | — | — |
| Cardiovascular disease | 475 (24.4) | — | — | — | — |
| Chronic bronchitis | 66 (3.4) | — | — | — | — |
| Emphysema | 52 (2.6) | — | — | — | — |
| Osteoporosis* | 290 (14.9) | — | — | — | — |
| Smoking status | |||||
| Current | 169 (8.5) | — | — | — | — |
| Former | 896 (45.3) | — | — | — | — |
| Height (m) | 1.66 ± 0.1 | 1.65 ± 0.1 | −0.01 ± 0.02 | −0.6 | <0.001 |
| Weight (kg) | 75.6 ± 15.0 | 74.3 ± 15.2 | −1.36 ± 4.99 | −1.8 | <0.001 |
| BMI (kg/m2) | 27.3 ± 4.7 | 27.2 ± 4.8 | −0.04 ± 1.89 | −0.1 | 0.39 |
| Waist circumference (cm) | 99.2 ± 12.9 | 98.6 ± 13.4 | −0.59 ± 10.6 | −0.6 | 0.02 |
| DEXA | |||||
| Total hip bone density (gm/cm2) | 0.89 ± 0.17 | 0.88 ± 0.17 | −0.01 ± 0.03 | −1.1 | <0.001 |
| Fat mass (kg) | 26.7 ± 8.5 | 26.8 ± 8.7 | 0.15 ± 3.33 | 0.6 | 0.05 |
| Fat mass percentage (%) | 35.1 ± 7.6 | 35.5 ± 7.5 | 0.41 ± 2.57 | 1.2 | <0.001 |
| Fat-free mass (kg) | 48.8 ± 10.4 | 47.9 ± 10.1 | −0.85 ± 2.03 | −1.7 | <0.001 |
| CT | |||||
| Subcutaneous adipose tissue (cm2) | 282.9 ± 116.8 | 262.1 ± 116.3 | −18.7 ± 46.1 | −6.6 | <0.001 |
| Visceral adipose tissue (cm2) | 142.8 ± 65.9 | 136.3 ± 74.6 | −5.74 ± 40.0 | −4.0 | <0.001 |
| Total thigh muscle area (cm2) | 223.8 ± 55.8 | 214.8 ± 52.9 | −10.1 ± 17.4 | −4.5 | <0.001 |
| Total thigh IMF area (cm2) | 19.9 ± 12.3 | 24.4 ± 11.5 | 4.90 ± 5.77 | 24.6 | <0.001 |
| Lung function | |||||
| FEV1 (L) | 2.20 ± 0.64 | 2.00 ± 0.62 | −0.19 ± 0.26 | −8.6 | <0.001 |
| FVC (L) | 2.92 ± 0.81 | 2.76 ± 0.81 | −0.16 ± 0.33 | −5.5 | <0.001 |
| FEV1 predicted (L) | 2.33 ± 0.57 | 2.18 ± 0.55 | −0.15 ± 0.07 | −6.4 | <0.001 |
| FVC predicted (L) | 3.15 ± 0.81 | 2.97 ± 0.81 | −0.18 ± 0.09 | −5.7 | <0.001 |
| FEV1/FVC predicted (%) | 74.85 ± 1.62 | 74.04 ± 1.63 | −0.82 ± 0.06 | 1.1 | <0.001 |
Notes: BMI = body mass index; IMF = intermuscular fat.
Self-report or total hip BMD T-score ≤ −2.5, Δ = Year 5 − 1 Year.
Participants on average significantly declined in height, weight, waist, total hip bone density, fat-free mass, subcutaneous adipose tissue, VAT, and total thigh muscle area (p < .05 for each) over 5 years. By contrast, participants gained in fat mass percentage (0.41 %, p < .001), total thigh IMF area (4.90 cm2, p < .001), and FM (0.15 kg, p = .05).
Decreases in lung function were observed in both sexes and for all the considered spirometric indexes. FEV1 and FVC significantly decreased over the follow-up period (respectively −190 mL [38 mL/y], −160 mL [32 mL/y], p < .001 for each).
Tables 3 and 4 (Model 3) summarize multivariable linear regression models where all body composition variables evaluated at baseline were considered as independent variables and FEV1 and FVC as the dependent variable. After adjustment for demographics and clinical confounding factors, height and total thigh muscle area remained positively and VAT negatively related with both FEV and FVC after correction for multiple comparisons (adjusted p < .05 for each).
Table 3.
Results of Linear Regression Models of Body Composition Measures (Year 1) as Independent Variables and FEV1 (mL) (Year 1) as the Dependent Variable
| N = 1893 | Model 1 | Model 2 | Model 3 | |||
| Beta (SE) | p Value | Beta (SE) | p Value | Beta (SE) | p Value | |
| Total hip bone mineral density (gm/cm2) | 1246.12 (83.60) | <.001 | 210.75 (78.44) | .007 | — | — |
| Fat mass (kg) | −9.29 (1.72) | <.001 | −2.12 (1.34) | .113 | — | — |
| Subcutaneous adipose tissue (cm2) | −1.50 (0.13) | <.001 | −0.05 (0.11) | .669 | — | — |
| Visceral adipose tissue (cm2) | 0.73 (0.23) | .001 | −1.14 (0.17) | <.001 | −1.66 (0.18) | <.001 |
| Total thigh muscle area (cm2) | 5.37 (0.23) | <.001 | 1.43 (0.31) | <.001 | 1.87 (0.35) | <.001 |
| Total thigh intermuscular fat area (cm2) | −3.58 (1.23) | .004 | −2.17 (0.92) | .018 | — | — |
| Height (cm) | 43.05 (1.24) | <.001 | 29.00 (1.68) | <.001 | 27.77 (1.75) | <.001 |
Notes: Model 1: Body composition measures were entered in separate unadjusted models. Model 2: Model 1 + clinic site, age, sex, race, FEV1 or FVC quality score, and height (CT and DEXA measures only). Model 3: Model 2 + remaining body composition measures, smoking, prevalent cardiovascular disease, osteoporosis, emphysema, and chronic bronchitis. Y1 FEV1 (N = 88) or Y1 FVC (N = 196) of quality score <2 excluded. Model 3 results provided for variables retained in forward selection procedure (p-entry = .20).
Table 4.
Results of Linear Regression Models of Body Composition Measures (Year 1) as Independent Variables and FVC (mL) (Year 1) as the Dependent Variable
| N = 1,785 | Model 1 | Model 2 | Model 3 | |||
| Beta (SE) | p Value | Beta (SE) | p Value | Beta (SE) | p Value | |
| Total hip bone mineral density (gm/cm2) | 1550.99 (108.58) | <.001 | −16.61 (88.04) | .850 | — | — |
| Fat mass (kg) | −19.05 (2.20) | <.001 | −9.74 (1.48) | <.001 | — | — |
| Subcutaneous adipose tissue (cm2) | −2.53 (0.16) | <.001 | −0.57 (0.12) | <.001 | −0.41 (0.14) | .004 |
| Visceral adipose tissue (cm2) | 0.89 (0.29) | .002 | −1.82 (0.18) | <.001 | −2.06 (0.22) | <.001 |
| Total thigh muscle area (cm2) | 6.93 (0.29) | <.001 | 0.46 (0.35) | .187 | 1.76 (0.39) | <.001 |
| Total thigh intermuscular fat area (cm2) | −7.50 (1.57) | <.001 | −5.75 (1.00) | <.001 | — | — |
| Height (cm) | 60.59 (1.45) | <.001 | 40.34 (1.86) | <.001 | 40.57 (1.96) | <.001 |
Notes: Model 1: Body composition measures were entered in separate unadjusted models. Model 2: Model 1 + clinic site, age, sex, race, FEV1 or FVC quality score, and height (CT and DEXA measures only). Model 3: Model 2 + remaining body composition measures, smoking, prevalent cardiovascular disease, osteoporosis, emphysema, and chronic bronchitis. Y1 FEV1 (N = 88) or Y1 FVC (N = 196) of quality score <2 excluded. Model 3 results provided for variables retained in forward selection procedure (p-entry = .20).
In unadjusted analyses of change in body composition measures as independent variables and change in lung function as dependent variable (Tables 5 and 6), increases in FM, VAT, subcutaneous adipose tissue, total thigh muscle area, and total thigh IMF area were associated with worsening FEV1 and FVC over 5 years.
Table 5.
Results of Linear Regression Models of Changes in Body Composition (Y5/Y6 − Y1) as Independent Variables and Change in FEV1 (mL) (Y5 − Y1) as the Dependent Variable
| N = 1,797 | Model 1 | Model 2 | Model 3 | |||
| Beta (SE) | p Value | Beta (SE) | p Value | Beta (SE) | p Value | |
| Δ Total hip bone mineral density (gm/cm2) | −246.12 (147.68) | .096 | −221.75 (147.84) | .134 | — | — |
| Δ Fat mass (kg) | −8.38 (1.49) | <.001 | −7.97 (1.49) | <.001 | −9.66 (2.53) | <.001 |
| Δ Subcutaneous adipose tissue (cm2) | −0.37 (0.13) | .004 | −0.37 (0.13) | .004 | 0.25 (0.18) | .172 |
| Δ Visceral adipose tissue (cm2) | −0.59 (0.14) | <.001 | −0.52 (0.15) | <.001 | — | — |
| Δ Total thigh muscle area (cm2) | −0.87 (0.32) | .007 | −1.01 (0.33) | .002 | — | — |
| Δ Total thigh intermuscular fat area (cm2) | −4.59 (0.97) | <.001 | −3.92 (1.00) | <.001 | −2.15 (1.22) | .079 |
| Δ Height (cm) | 20.70 (39.26) | .598 | −0.55 (39.52) | .989 | — | — |
Notes: Model 1: Body composition change measures were entered in separate unadjusted models. Model 2: Model 1 + clinic site, age, sex, race, height, and FEV1 or FVC quality score. Model 3: Model 2 + remaining body composition change measures, smoking, prevalent cardiovascular disease, osteoporosis, emphysema, and chronic bronchitis. Y1 and Y5 FEV1 (N = 184) or Y1 and Y5 FVC (N = 400) of quality score <2 excluded. Model 3 results provided for variables retained in forward selection procedure (p-entry = .20).
Table 6.
Results of Linear Regression Models of Changes in Body Composition (Y5/Y6 − Y1) as Independent Variables and Change in FVC (mL) (Y5 − Y1) as the Dependent Variable
| N = 1,581 | Model 1 | Model 2 | Model 3 | |||
| Beta (SE) | p Value | Beta (SE) | p Value | Beta (SE) | p Value | |
| Δ Total hip bone mineral density (gm/cm2) | −647.74 (202.33) | .001 | −547.90 (201.60) | .007 | — | — |
| Δ Fat mass (kg) | −13.88 (2.10) | <.001 | −14.21 (2.07) | <.001 | −12.20 (2.95) | <.001 |
| Δ Subcutaneous adipose tissue (cm2) | −0.85 (0.17) | <.001 | −0.82 (0.17) | <.001 | — | — |
| Δ Visceral adipose tissue (cm2) | −1.10 (0.19) | <.001 | −0.86 (0.19) | <.001 | — | — |
| Δ Total thigh muscle area (cm2) | −1.11 (0.44) | .011 | −1.81 (0.45) | <.001 | −1.02 (0.51) | .046 |
| Δ Total thigh intermuscular fat area (cm2) | −8.97 (1.28) | <.001 | −7.42 (1.33) | <.001 | −4.07 (1.59) | .011 |
| Δ Height (cm) | 2.02 (46.31) | .965 | 13.68 (46.18) | .767 | 78.12 (51.53) | .130 |
Notes: Model 1: Body composition change measures were entered in separate unadjusted models. Model 2: Model 1 + clinic site, age, sex, race, height, and FEV1 or FVC quality score. Model 3: Model 2 + remaining body composition change measures, smoking, prevalent cardiovascular disease, osteoporosis, emphysema, and chronic bronchitis. Y1 and Y5 FEV1 (N = 184) or Y1 and Y5 FVC (N = 400) of quality score <2 excluded. Model 3 results provided for variables retained in forward selection procedure (p-entry = .20).
Adjusted for potential confounders, greater FM increase remained associated with greater declines in both FEV1 and FVC after correction for multiple comparisons (adjusted p < .05). Results did not substantially change after exclusion of participants with emphysema or chronic bronchitis or after additional adjustment for baseline FEV1 or FVC (not reported).
Finally, in analyses stratified by weight-change categories, significant differences in delta FEV1 and delta FVC (all p < .01) were identified between weight-change groups, except for delta FEV1 in men (p = .09), as shown in Figures 1 and 2. Men and women who gained weight showed the greatest decrease in FEV1 (men = −11.67%, women = −10.92%) and FVC (men = −8.01%, women = −8.01%).
Figure 1.
Mean forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) percent change (Y5 − Y1) by category of weight change in men. There were significant within-group changes in all weight groups for FEV1 and FVC. Data were analyzed by using analysis of covariance, with adjustment for age, race, height, smoking, emphysema, and chronic bronchitis (*p < .0001).
Figure 2.
Mean forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) percent change (Y5 − Y1) by category of weight change in women. There were significant within group changes in all weight groups for forced expiratory volume in 1 second and forced vital capacity. Data were analyzed by using ANCOVA, with adjustment for age, race, height, smoking, emphysema and chronic bronchitis (*p < .01).
DISCUSSION
This study of community-dwelling older adults shows associations of several body composition measures with pulmonary function, cross-sectionally and over time. In particular, after adjustment for several potential confounders, height and total thigh muscle area are positively and VAT negatively related to FEV1 and FVC. Greater increase in FM over 5 years is further associated with more rapid lung function decline.
Our finding that visceral fat, as evaluated by CT, is negatively associated with lung function complement and expand previous cross-sectional studies performed in elderly patients where fat distribution was evaluated with anthropometry (9,10).
Associations between respiratory function and visceral fat evaluated with MRI technique have been previously investigated in a small sample of young lean (n = 20) and obese (n = 20) men and women (age 32.5 ± 6 years), suggesting that chest and abdominal visceral fat may have a cumulative effect on the reduction of end-expiratory lung volume (25). A consistent negative association between waist circumference and pulmonary function has been demonstrated in normal weight, overweight, and obese subjects (13).
Our findings further suggest that greater amount of fat mass stored in the abdomen and particularly VAT may contribute to worsening pulmonary function, independently of smoking status and comorbidities even in older adults. Several mechanisms may link abdominal obesity to pulmonary function worsening. The mechanical effects of truncal obesity may partly determine the reductions in chest wall compliance, lung elastic recoil, and peripheral airway size that negatively affect FEV1 and FVC. Abdominal fat may also alter the pressure–volume characteristics of the thorax and restrict the descent of the diaphragm, opposing to the downward motion of the contracted diaphragm, thereby limiting lung expansion during spirometric performance. Fat accumulation in the abdominal cavity may finally increase intra-abdominal pressure (26). It has also been shown that the increase of intra-abdominal pressure observed in visceral obesity is able to pump upward the diaphragmatic muscle, compressing the parenchyma of the lung, particularly at the basal regions. Moreover, the over-stretching of the diaphragmatic muscle fibers, caused by the elevation of the diaphragmatic domes produced by visceral fat, can decrease the contractile efficiency of the diaphragmatic muscle (27).
Interestingly, in our study population, we also found that IMF was negatively associated with lung function decline in older subjects, even if this relationship was no longer significant after adjustment for comorbidities. It has been previously reported that IMF, as evaluated at the thigh level by CT, is directly and strongly associated with age, adiposity (28), muscle function impairment (6), and progressive muscle weakness in aging (23). Our results may suggest that age-associated changes in muscle quality, that is, increase in fat infiltration, may in part contribute also to worsening pulmonary function among older adults.
Several assumptions are needed to understand the association between fat content in thigh muscle and respiratory function. As fat infiltration occurs with aging and adiposity in all muscles of the body, to varying extents among different types of muscles, it is reasonable to hypothesize that the observed increase of fat inside thigh muscles with aging in this cohort may be comprehensive even of other muscles and in particular of diaphragm, the abdominal muscles, and chest wall respiratory muscles. Fat deposition in these regions may be expected to account in part for worsening muscle quality and function involved in the expiratory process.
As expected, our findings confirm a 5-year decline in lung function with a 38 mL/y and 32 mL/y decrease in FEV1 and FVC, respectively. Among weight-change groups, men and women who gained weight had the greatest lung function decline. As recently reported in the Health ABC Study population (23), in our study population, weight gainers had the greatest increase in IMF (7.34 ± 5.51 cm2), FM (3.59 ± 2.28 Kg), and VAT (18.33 ± 40.14 cm2) compared with other weight change groups (not shown in table). Our data show that higher amounts of fat mass and VAT and reduced muscle quality are the main predictors of respiratory function decline in elderly men and women and raise the question of whether the maintenance of stable weight even in old age and the prevention of abdominal and IMF deposition may help to slow age-related decline in respiratory function. Subjects who presented unintentional weight loss showed the lower FEV1 and FVC decrease. Our findings of lower decrease in FEV1 and FVC in patients with weight loss should be regarded with caution because it has been shown in several studies that weight loss in older adults is associated with higher mortality (29,30).
Some limitations of our study should be mentioned.
Our results may be sensitive to nonrandom withdrawal of participants who experienced the greatest functional declines throughout the study period. For patients to be assessed for degree of decline in lung function, they must survive and be healthy enough to attend the 5-year follow-up and undergo spirometry. As expected, participants with complete lung function data at 5-year follow-up were younger at baseline (73.4 vs 74.0 years), less likely to have cardiovascular disease (24% vs 34%), emphysema (3% vs 5%), and a history of smoking (54% vs 61%) relative to participants who completed lung function testing at baseline only (p < .01 for each). Lung function data at interim follow-up were not available to evaluate the potential influence of missing Year 5 data on reported associations.
Our results may not generalize to other populations due to the strict inclusion criteria of the study and further exclusion of those with medical conditions that prevented completion of spirometric testing. Rather, this cohort is representative of older men and women with mild levels of functional impairment and relatively low morbidity. Also, the potential influence of systemic inflammation, that could itself negatively affect pulmonary function, was not evaluated in this analysis. Future studies are warranted in order to evaluate the hypotheses that systemic inflammation may play a role in lung function decrease in older adults.
In summary, among this well-functioning cohort, less muscle and greater adipose tissue areas were each independently associated with lower FEV1 and FVC cross-sectionally, whereas greater increase in fat mass over 5 years was associated with concurrent lung function decline. Body composition changes, particularly fat deposition, may in part explain respiratory function declines in aging.
FUNDING
This research was supported by the Intramural Research Program of the National Institutes of Health, National Institute on Aging (N01-AG-6-2101, N01-AG-6-2103, N01-AG-6-2106).
CONFLICT OF INTEREST
There are no conflicts of interest to disclose.
Acknowledgments
The authors’ responsibilities were as follows—A.P.R, N.L.W, B.H.G, and M.Z.: analysis and interpretation of data and preparation of manuscript; T.B.H. and A.B.N.: acquisition of subjects and data, study concept and design, interpretation of data, and preparation of manuscript; S.B.K.: consulted on study design, recruited subjects, and edited the manuscript; D.C.B.: consulted on study design and edited the manuscript; S.S.; acquisition of subjects, collection of data, and review of the manuscript.
References
- 1.Enright PL, Kronmal RA, Higgins M, et al. Spirometry reference values for women and men 65–85 years of age. Am Rev Respir Dis. 1993;147:125–133. doi: 10.1164/ajrccm/147.1.125. [DOI] [PubMed] [Google Scholar]
- 2.Burr ML, Philipp KM, Hurst DN. Lung function in the elderly. Thorax. 1985;40:54–59. doi: 10.1136/thx.40.1.54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Beaty TH, Newill CA, Cohen BH, et al. Effects of pulmonary function on mortality. J Chronic Dis. 1985;8:703–710. doi: 10.1016/0021-9681(85)90024-4. [DOI] [PubMed] [Google Scholar]
- 4.Kannel WB, Hubert H, Lew EA. Vital capacity as predictor of cardiovascular disease: the Framingham Study. Am Heart J. 1983;105:311–315. doi: 10.1016/0002-8703(83)90532-x. [DOI] [PubMed] [Google Scholar]
- 5.Janssen I, Heymsfield SB, Ross R. Low relative skeletal muscle mass (Sarcopenia) in older persons is associated with functional impairment and physical disability. J Am Geriatr Soc. 2002;50:889–896. doi: 10.1046/j.1532-5415.2002.50216.x. [DOI] [PubMed] [Google Scholar]
- 6.Goodpaster BH, Carlson CL, Visser M, et al. Attenuation of skeletal muscle and strength in the elderly: The Health ABC Study. J Appl Physiol. 2001;90:2157–2165. doi: 10.1152/jappl.2001.90.6.2157. [DOI] [PubMed] [Google Scholar]
- 7.Visser M, Harris TB, Langlois J, et al. Body fat and skeletal muscle mass in relation to physical disability in very old men and women of the Framingham Heart Study. J Gerontol A Biol Sci Med Sci. 1998;53A:M214–M221. doi: 10.1093/gerona/53a.3.m214. [DOI] [PubMed] [Google Scholar]
- 8.Cotes JE, Chinn DJ, Reed JW. Body mass, fat percentage, and fat free mass as reference variables for lung function: effects on terms for age and sex. Thorax. 2001;56:839–844. doi: 10.1136/thorax.56.11.839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Santana H, Zoico E, Turcato E, et al. Relation between body composition, fat distribution, and lung function in elderly men. Am J Clin Nutr. 2001;73:827–831. doi: 10.1093/ajcn/73.4.827. [DOI] [PubMed] [Google Scholar]
- 10.Wannamethee SG, Shaper AG, Whincup PH. Body fat distribution, body composition, and respiratory function in elderly men. Am J Clin Nutr. 2005;82:996–1003. doi: 10.1093/ajcn/82.5.996. [DOI] [PubMed] [Google Scholar]
- 11.Rossi A, Fantin F, Di Francesco V, et al. Body composition and pulmonary function in the elderly: a 7-year longitudinal study. Int J Obes. 2008;32:1423–1430. doi: 10.1038/ijo.2008.103. [DOI] [PubMed] [Google Scholar]
- 12.Penninx BW, Kritchevsky SB, Newman AB, et al. Inflammatory markers and incident mobility limitation in the elderly. J Am Geriatr Soc. 2004;52:1105–1113. doi: 10.1111/j.1532-5415.2004.52308.x. [DOI] [PubMed] [Google Scholar]
- 13.Chen Y, Rennie D, Cormier Y, et al. Waist circumference is associated with pulmonary function in normal-weight, overweight, and obese subjects. Am J Clin Nutr. 2007;85:35–39. doi: 10.1093/ajcn/85.1.35. [DOI] [PubMed] [Google Scholar]
- 14.Tylavsky FA, Lohman T, Blunt BA, et al. QDR 4500 DXA overestimates fat-free mass compared with criterion methods. J Appl Physiol. 2003;94:959–965. doi: 10.1152/japplphysiol.00732.2002. [DOI] [PubMed] [Google Scholar]
- 15.Tylavsky FA, Fuerst T, Nevitt M, et al. Measurement of changes in soft tissue mass and fat mass with weight change: pencil- versus fan-beam dual energy X ray absorptiometry. Ann N Y Acad Sci. 2000;904:94–97. doi: 10.1111/j.1749-6632.2000.tb06426.x. [DOI] [PubMed] [Google Scholar]
- 16.Tylavsky FA, Lohman TG, Dockrell M, et al. Comparison of the effectiveness of 2 dual energy X ray absorptiometry with that of total body water and computer tomography in assessing changing in body composition during weight change. Am J Clin Nutr. 2003;77:356–363. doi: 10.1093/ajcn/77.2.356. [DOI] [PubMed] [Google Scholar]
- 17.Visser M, Fuerst T, Lang T, et al. Validity of fan beam dual energy X ray absorptiometry for measuring fat free mass and leg muscle mass. J Appl Physiol. 1999;87:1513–1520. doi: 10.1152/jappl.1999.87.4.1513. [DOI] [PubMed] [Google Scholar]
- 18.Goodpaster BH, Thaete FL, Simoneau JA, et al. Subcutaneous abdominal fat and thigh muscle composition predict insulin sensitivity independently of visceral fat. Diabetes. 1997;46:1579–1585. doi: 10.2337/diacare.46.10.1579. [DOI] [PubMed] [Google Scholar]
- 19.Goodpaster BH, Krishnaswami S, Resnick H, et al. Association between regional adipose tissue distribution and both type 2 diabetes and impaired glucose tolerance in elderly men and women. Diabetes Care. 2003;26:372–379. doi: 10.2337/diacare.26.2.372. [DOI] [PubMed] [Google Scholar]
- 20.Yende, et al. Inflammatory markers are associated with ventilatory limitation and muscle dysfunction in obstructive lung disease in well functioning elderly subjects. Thorax. 2006;61:10–16. doi: 10.1136/thx.2004.034181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.American Thoracic Society. Standardization of spirometry: 1994 update. Am J Respir Crit Care Med. 1995;152:1107–1136. doi: 10.1164/ajrccm.152.3.7663792. [DOI] [PubMed] [Google Scholar]
- 22.Newman AB, Lee JS, Visser M, et al. Weight change and the conservation of lean mass in old age: the Health, Aging and Body Composition Study. Am J Clin Nutr. 2005;82:872–878. doi: 10.1093/ajcn/82.4.872. [DOI] [PubMed] [Google Scholar]
- 23.Delmonico MJ, Harris TB, Visser M, et al. Longitudinal study of muscle strength, quality, and adipose tissue infiltration. Am J Clin Nutr. 2009;90:1579–1585. doi: 10.3945/ajcn.2009.28047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.SAS Institute Inc. SAS/STAT User's Guide, Version 8. Cary, NC: SAS Institute; 2001. [Google Scholar]
- 25.Babb TG, Wyrick BL, DeLorey DS, et al. Fat distribution and end-expiratory lung volume in lean and obese men and women. Chest. 2008;134:704–711. doi: 10.1378/chest.07-1728. [DOI] [PubMed] [Google Scholar]
- 26.Sugerman HJ. Effects of increased intra-abdominal pressure in severe obesity. Surg Clin North Am. 2001;81:1063–1075. doi: 10.1016/s0039-6109(05)70184-5. [DOI] [PubMed] [Google Scholar]
- 27.Sharp JT, Druz WS, Kondragunta VR. Diaphragmatic responses to body position changes in obese patients with obstructive sleep apnea. Am Rev Respir Dis. 1986;133:32–37. doi: 10.1164/arrd.1986.133.1.32. [DOI] [PubMed] [Google Scholar]
- 28.Goodpaster BH, Krishnaswami S, Harris TB, et al. Obesity, regional body fat distribution, and the metabolic syndrome in older men and women. Arch Intern Med. 2005;165(7):777–783. doi: 10.1001/archinte.165.7.777. [DOI] [PubMed] [Google Scholar]
- 29.Payette H, Coulombe C, Boutier V, Gray-Donald KJ. Weight loss and mortality among free-living frail elders: a prospective study. J Gerontol A Biol Sci Med Sci. 1999;54:M440–M445. doi: 10.1093/gerona/54.9.m440. [DOI] [PubMed] [Google Scholar]
- 30.Locher JL, Roth DL, Ritchie CS, Cox K, Sawyer P, Bodner EV, Allman RM. Body mass index, weight loss, and mortality in community-dwelling older adults. J Gerontol A Biol Sci Med Sci. 2007;62:1389–1392. doi: 10.1093/gerona/62.12.1389. [DOI] [PMC free article] [PubMed] [Google Scholar]