Abstract
Background
It is widely recognized that fat will accumulate with ageing and is more prominent centrally. However, there were studies reported that fat might not gain either centrally or generally with ageing.
Methods
The baseline, 2-year and 4-year total body fat mas, trunk fat mass and percentage fat mass, were measured by DXA in 3018 community-living Chinese older than 65 years. The respective 4-year trajectories of adiposity were analyzed by repeated measure ANOVA p-for-trend test.
Results
There was a trend of increase in total body fat mass in men and a decreasing trend in women but neither reached statistical significance. However, there was a significant increase in percent fat mass in both genders. Fat mass was relatively stable in the 2 young-old groups but it declined in the oldest group, aged 75 years or above. (men, p=0.017; women, p<0.001). On the contrary, a corresponding rise of percent fat mass was observed, which was steeper in the 2 younger age groups but did not change in the oldest group. For trunk fat mass, there was a statistically significant decreasing trend in women (p < 0.001) but it remained static in men (p = 0.092). The fat mass in upper limbs of both genders did not change but for the lower limbs, there was a statistically significant increase in both men (p < 0.001) and women (p < 0.02).
Conclusions
Absolute total body fat mass does not accumulate in old age and in the contrary, in the oldest old group (75 years or above), it declined instead. With ageing, fat will redistribute from the central region to the lower limbs.
Key words: Fat mass, prospective, elderly
Background
It is widely recognized that fat will accumulate in older people (1, 2, 3, 4, 5) and the accumulation is more prominent in the central region of the body (5, 6, 7, 8). On the contrary, there were others reporting the opposite about the loss of fat mass in old age (9, 10, 11, 12). Baumgartner et al (10) studied a cohort of 316 men and women across a wide age range from 60 to 95 years, and also had observed that total fat mass and percent body fat decreased with ageing in women but not in men, with no change in body fat distribution. Steen et al (9) followed a cohort of 23 healthy seventy-year-old persons for 10 years and measured their body weight and body composition. This prospective study showed a gradual decline in body weight and part of the reduction was in body fat mass. Buffa et al (12) similarly, in a review, reported that there was a reduction in fat mass with ageing and most of the loss originated from the subcutaneous fat and not visceral fat (13) and that adiposity accumulation stopped after the age of 80 years (14) Moreover, Silver et al (11) observed no difference in percentage body fat with ageing but it was significantly lower in both the over 85-year group and the nursing home residents group.
Recently, Zaslavsky et al have demonstrated that adiposity decline was associated with excess mortality in a cohort of frail older women (15). Therefore examination of the trajectory of fat mass in older persons may carry clinical significance.
We hypothesized that fat mass would decline and redistribute from central to peripheral region with ageing; and therefore followed the changes in absolute total body fat mass, fat mass percentage (%) and regional fat mass for 4 years in a group of community-living older persons.
Methods
This study was generated from a primary survey which was examining the bone mineral density of older Chinese adults. Fat mass data were derived from the parent dual-energy x-ray absorptiometry (DXA) database in the measurement of bone mineral density. In this primary osteoporosis study, 4000 community-dwelling men and women aged 65 years or older were recruited by placing recruitment notices in community centers for the elderly participants and housing estates. Talks were also given at these centers explaining the purpose, procedures, and investigations to be carried out. Written informed consents were obtained. Only ethnic Chinese participants were recruited. We excluded those who (a) were unable to walk without assistance of another person; (b) had had a bilateral hip replacement because that would have affected the bone mineral density measurement; (c) were not competent to give informed consent; and (d) had medical conditions, in the judgment of the study physicians, which made it unlikely that they would survive the duration of the study (3 years). The sample was stratified so that approximately 33% were in each of the age groups: 65–69, 70–74, and 75 years and older. The study was approved by the Clinical Research Ethics Committee of The Chinese University of Hong Kong.
We measured fat mass by DXA, which was originally developed to measure bone mineral density. However, it is also capable of assessing whole body and regional body composition, assuming that human body consists of three components, i.e. bone, fat and lean tissue. Each compartment is distinguishable by their differential X-ray attenuation at two different energy levels. Body composition measurement with DXA has been in vivo validated (16) and has been employed in clinical researches and practice applications (17) The whole body DXA examinations were acquired using Hologic fan beam Delphi densitometer (Hologic, Inc., Bedford, MA), with procedures recommended by the manufacturer. The DXA data were analyzed locally according to Hologic standard methodology with software version 12.1.For the measurement of regional fat mass, we followed the measurement method employed by the National Health and Nutrition Examination Survey (NHANES) (18). A line of delineation was drawn between the head of the humerus and the glenoid fossa of the scapula to separate the upper limb from the trunk, and the leg consisted of the parts of the body between the inferior border of the ischial tuberosity to the most distal tip of the toes.
The maximum coefficient of variation for fat mass measurement was 1.47%, which has been derived from one of our past studies, measuring body composition in the same cohort of participants (19). DXA machine performance was continuously monitored throughout the study period, with both Hologic Spine phantom and Hologic body composition step phantom, scanned on the day that a participant was scanned, and at least 3 days per week when no participants were scanned.
Statistical methods
We calculated the means of total body fat mass and regional body fat mass (truncal, both upper limbs and both lower limbs and examined if there was a trend across time by repeated measure ANOVA, which was done by repeated measure general linear model. Mauchly's test was applied to check sphericity and Greenhouse-Geisser correction was applied for those tests that did not meet the assumption of sphericity. Linear trend and quadratic trend were analyzed across time as well. The procedure was repeated on fat mass %, which was equivalent to fat mass divided by the weight of that region. In the case of total body fat mass %, it was total fat mass divided by total body weight. In comparing the rate of change in different age groups, we divided the cohort into three age stratum as defined in the recruitment criteria, namely, 65–69, 70–74, and 75 years and older. The gradient of the 3 curves, belong to 3 age groups were then compared by calculating the p-value for slope difference (interaction). All tests were two-sided and a p-value < 0.05 was taken as significant. All statistical analyses were performed using the statistical package SAS, version 9.4 (SAS Institute, Inc., Cary, NC).
Results
After 4 years, 3018 (1519 men and 1499 women) completed the three DXA studies. The baseline characteristics of the cohort were tabulated in Table 1. The mean age (standard deviation) of men and women was 71.7 years (4.6 years) and 72.0 years (5.0 years) and their respective body mass indices (BMI) were 23.6 kg/m2 and 24.0 kg/m2. The 4-year trend of total and regional fat mass change was tabulated in Table 2. The comparisons of the rates of change of fat mass and fat mass % among the three age groups were tabulated in Table 3.
Table 1.
Baseline characteristics of the cohort
| Mean (SD)/ Freq. (%) | |||
|---|---|---|---|
| Total (n=3018) | Male (n=1519) | Female (n=1499) | |
| Age | |||
| 65-69 | 1121 (37.14%) | 569 (37.46%) | 552 (36.82%) |
| 70-74 | 1070 (35.45%) | 560 (36.87%) | 510 (34.02%) |
| 75 or above | 827 (27.4%) | 390 (25.67%) | 437 (29.15%) |
| mean | 71.83 (4.82) | 71.72 (4.64) | 71.95 (4.99) |
| Education | |||
| No education | 597 (19.78%) | 71 (4.67%) | 526 (35.09%) |
| Primary or below | 1513 (50.13%) | 816 (53.72%) | 697 (46.5%) |
| Secondary or above | 908 (30.09%) | 632 (41.61%) | 276 (18.41%) |
| Marital status | |||
| Married | 2185 (72.4%) | 1360 (89.53%) | 825 (55.04%) |
| Widowed/Separated | 773 (25.61%) | 135 (8.89%) | 638 (42.56%) |
| Single | 60 (1.99%) | 24 (1.58%) | 36 (2.4%) |
| Current smoker | 188 (6.23%) | 164 (10.8%) | 24 (1.6%) |
| Current drinker | 406 (13.46%) | 367 (24.16%) | 39 (2.6%) |
| Height (cm) | 157.23 (8.20) | 163.27 (5.69) | 151.10 (5.27) |
| Weight (kg) | 58.84 (9.63) | 62.85 (9.12) | 54.77 (8.35) |
| BMI | 23.76 (3.21) | 23.55 (3.04) | 23.97 (3.36) |
| Whole body fat (kg) | 17.28 (5.23) | 15.37 (4.59) | 19.22 (5.13) |
| Whole body % fat | 29.47 (7.16) | 24.36 (4.84) | 34.66 (5.10) |
Table 2.
Longitudinal change of fat mass and fat mass %
| Mean (SD) | p-value of linear trend | p-value of quadratic trend | |||
|---|---|---|---|---|---|
| Baseline | 2Y FU | 4Y FU | |||
| Male (n=1519) | |||||
| Appendicular fat (kg) | 5.73 (1.75) | 5.81 (1.78) | 5.86 (1.87) | <0.001* | 0.160 |
| Appendicular % fat | 21.60 (4.68) | 21.94 (4.77) | 22.19 (5.01) | <0.001* | 0.278 |
| Upper limb fat (kg) | 1.68 (0.56) | 1.67 (0.56) | 1.69 (0.59) | 0.223 | 0.111 |
| Upper limb % fat | 23.41 (5.56) | 23.49 (5.70) | 23.80 (5.98) | <0.001* | 0.045 |
| Lower limb fat (kg) | 4.05 (1.27) | 4.14 (1.29) | 4.17 (1.35) | <0.001* | 0.007 |
| Lower limb % fat | 20.91 (4.69) | 21.34 (4.79) | 21.57 (4.99) | <0.001* | 0.017 |
| Trunk fat (kg) | 8.74 (3.00) | 8.67 (2.99) | 8.69 (3.06) | 0.092 | 0.055 |
| Trunk % fat | 27.46 (6.21) | 27.46 (6.28) | 27.67 (6.45) | 0.018 | 0.068 |
| Whole body fat (kg) | 15.37 (4.59) | 15.37 (4.60) | 15.43 (4.75) | 0.296 | 0.485 |
| Whole body % fat | 24.36 (4.84) | 24.49 (4.90) | 24.70 (5.08) | <0.001* | 0.393 |
| Female (n=1499) | |||||
| Appendicular fat (kg) | 8.44 (2.56) | 8.51 (2.57) | 8.46 (2.63) | 0.236 | <0.001* |
| Appendicular % fat | 35.99 (6.05) | 36.40 (5.99) | 36.46 (6.26) | <0.001* | <0.001* |
| Upper limb fat (kg) | 2.52 (0.83) | 2.53 (0.83) | 2.51 (0.86) | 0.183 | 0.109 |
| Upper limb % fat | 39.98 (7.30) | 40.21 (7.27) | 40.36 (7.69) | <0.001* | 0.585 |
| Lower limb fat (kg) | 5.91 (1.88) | 5.98 (1.90) | 5.95 (1.94) | 0.016 | <0.001* |
| Lower limb % fat | 34.43 (6.29) | 34.90 (6.26) | 34.93 (6.48) | <0.001* | <0.001* |
| Trunk fat (kg) | 9.96 (2.88) | 9.93 (2.93) | 9.83 (3.01) | <0.001* | 0.138 |
| Trunk % fat | 35.72 (5.94) | 35.95 (6.00) | 35.79 (6.31) | 0.445 | <0.001* |
| Whole body fat (kg) | 19.22 (5.13) | 19.25 (5.19) | 19.11 (5.34) | 0.056 | 0.009 |
| Whole body % fat | 34.66 (5.10) | 34.94 (5.10) | 34.88 (5.37) | <0.001* | <0.001* |
p-value <0.05 after Bonferroni adjustment; % fat = fat mass divided by the weight of that region
Table 3.
Comparison of the rates of change in fat mass and fat mass % among three age groups
| Mean (SD) | p-value of linear trend | p-value of quadratic trend | p-value of mean difference among 3 groups | p-value of slope difference among 3 groups (interaction) | ||||
|---|---|---|---|---|---|---|---|---|
| Age | Baseline | 2Y FU | 4Y FU | |||||
| Male (n=1519) | ||||||||
| Appendicular fat (kg) | 65-69 | 5.77 (1.77) | 5.84 (1.82) | 5.87 (1.87) | 0.003 | 0.286 | 0.769 | 0.156 |
| 70-74 | 5.71 (1.71) | 5.82 (1.71) | 5.88 (1.78) | <0.001* | 0.221 | |||
| ≥ 75 | 5.71 (1.79) | 5.74 (1.82) | 5.80 (1.99) | 0.080 | 0.981 | |||
| Appendicular % fat | 65-69 | 21.26 (4.60) | 21.54 (4.67) | 21.70 (4.90) | <0.001* | 0.401 | 0.012 | 0.205 |
| 70-74 | 21.59 (4.58) | 22.02 (4.62) | 22.30 (4.74) | <0.001* | 0.271 | |||
| ≥ 75 | 22.12 (4.88) | 22.39 (5.08) | 22.73 (5.46) | <0.001* | 0.850 | |||
| Upper limb fat (kg) | 65-69 | 1.69 (0.55) | 1.68 (0.56) | 1.70 (0.59) | 0.654 | 0.188 | 0.202 | 0.194 |
| 70-74 | 1.68 (0.55) | 1.70 (0.55) | 1.71 (0.57) | 0.030 | 0.534 | |||
| ≥ 75 | 1.65 (0.57) | 1.62 (0.58) | 1.64 (0.63) | 0.577 | 0.024 | |||
| Upper limb % fat | 65-69 | 23.05 (5.38) | 23.01 (5.43) | 23.26 (5.75) | 0.091 | 0.071 | 0.040 | 0.239 |
| 70-74 | 23.48 (5.47) | 23.83 (5.61) | 24.01 (5.65) | <0.001* | 0.377 | |||
| ≥ 75 | 23.85 (5.90) | 23.72 (6.14) | 24.29 (6.67) | 0.017 | 0.006 | |||
| Lower limb fat (kg) | 65-69 | 4.08 (1.29) | 4.16 (1.33) | 4.17 (1.36) | <0.001* | 0.038 | 0.907 | 0.178 |
| 70-74 | 4.02 (1.23) | 4.12 (1.24) | 4.18 (1.28) | <0.001* | 0.170 | |||
| ≥ 75 | 4.06 (1.29) | 4.12 (1.32) | 4.15 (1.43) | 0.007 | 0.263 | |||
| Lower limb % fat | 65-69 | 20.58 (4.63) | 20.97 (4.75) | 21.10 (4.92) | <0.001* | 0.055 | 0.009 | 0.232 |
| 70-74 | 20.86 (4.60) | 21.33 (4.62) | 21.64 (4.75) | <0.001* | 0.264 | |||
| ≥ 75 | 21.47 (4.84) | 21.89 (5.05) | 22.14 (5.38) | <0.001* | 0.275 | |||
| Trunk fat (kg) | 65-69 | 8.83 (3.00) | 8.76 (2.97) | 8.81 (3.02) | 0.769 | 0.132 | 0.052 | 0.002* |
| 70-74 | 8.79 (2.96) | 8.81 (2.90) | 8.84 (2.96) | 0.454 | 0.837 | |||
| ≥ 75 | 8.55 (3.05) | 8.34 (3.11) | 8.30 (3.20) | <0.001* | 0.101 | |||
| Trunk % fat | 65-69 | 27.37 (6.09) | 27.40 (6.04) | 27.60 (6.24) | 0.090 | 0.343 | 0.275 | 0.004* |
| 70-74 | 27.59 (6.12) | 27.83 (6.14) | 28.08 (6.19) | <0.001* | 0.878 | |||
| ≥ 75 | 27.42 (6.52) | 27.04 (6.78) | 27.19 (7.07) | 0.192 | 0.030 | |||
| Whole body fat (kg) | 65-69 | 15.50 (4.61) | 15.50 (4.62) | 15.57 (4.72) | 0.404 | 0.579 | 0.166 | 0.017 |
| 70-74 | 15.39 (4.49) | 15.52 (4.44) | 15.61 (4.56) | 0.010 | 0.713 | |||
| ≥ 75 | 15.15 (4.69) | 14.96 (4.79) | 14.97 (5.05) | 0.108 | 0.278 | |||
| Whole body % fat | 65-69 | 24.15 (4.74) | 24.28 (4.73) | 24.45 (4.92) | 0.005 | 0.697 | 0.350 | 0.050 |
| 70-74 | 24.42 (4.77) | 24.71 (4.79) | 24.95 (4.85) | <0.001* | 0.751 | |||
| ≥ 75 | 24.56 (5.09) | 24.47 (5.29) | 24.68 (5.59) | 0.453 | 0.121 | |||
| Female (n=1499) | ||||||||
| Appendicular fat (kg) | 65-69 | 8.58 (2.45) | 8.70 (2.48) | 8.69 (2.54) | 0.002 | 0.009 | <0.001* | <0.001* |
| 70-74 | 8.66 (2.53) | 8.77 (2.59) | 8.74 (2.60) | 0.055 | 0.004 | |||
| ≥ 75 | 7.99 (2.67) | 7.95 (2.58) | 7.85 (2.69) | 0.003 | 0.313 | |||
| Appendicular % fat | 65-69 | 36.01 (5.72) | 36.46 (5.75) | 36.61 (5.97) | <0.001* | 0.078 | 0.0045* | 0.013 |
| 70-74 | 36.45 (5.63) | 36.97 (5.64) | 37.04 (5.88) | <0.001* | 0.003 | |||
| ≥ 75 | 35.42 (6.84) | 35.66 (6.60) | 35.59 (6.95) | 0.232 | 0.082 | |||
| Upper limb fat (kg) | 65-69 | 2.55 (0.79) | 2.56 (0.80) | 2.56 (0.81) | 0.211 | 0.453 | <0.001* | <0.001* |
| 70-74 | 2.60 (0.82) | 2.62 (0.84) | 2.61 (0.85) | 0.622 | 0.287 | |||
| ≥ 75 | 2.40 (0.87) | 2.37 (0.85) | 2.32 (0.88) | <0.001* | 0.331 | |||
| Upper limb % fat | 65-69 | 39.88 (6.68) | 40.14 (6.62) | 40.44 (6.87) | <0.001* | 0.734 | 0.006 | 0.020 |
| 70-74 | 40.61 (6.70) | 40.88 (6.74) | 41.16 (7.16) | 0.001* | 0.963 | |||
| ≥ 75 | 39.36 (8.56) | 39.50 (8.49) | 39.30 (9.06) | 0.844 | 0.203 | |||
| Lower limb fat (kg) | 65-69 | 6.03 (1.82) | 6.14 (1.84) | 6.12 (1.89) | <0.001* | 0.001* | <0.001* | <0.001* |
| Age | Baseline | 2Y FU | 4Y FU | |||||
| Lower limb % fat | 65-69 | 34.51 (6.03) | 35.03 (6.12) | 35.12 (6.32) | <0.001* | 0.011 | 0.012 | 0.033 |
| 70-74 | 34.81 (5.95) | 35.42 (5.97) | 35.42 (6.16) | <0.001* | <0.001* | |||
| ≥ 75 | 33.87 (6.95) | 34.14 (6.70) | 34.11 (6.98) | 0.072 | 0.084 | |||
| Trunk fat (kg) | 65-69 | 10.00 (2.80) | 10.07 (2.86) | 10.01 (2.92) | 0.790 | 0.063 | <0.001* | <0.001* |
| 70-74 | 10.25 (2.87) | 10.23 (2.97) | 10.18 (3.04) | 0.194 | 0.700 | |||
| ≥ 75 | 9.58 (2.97) | 9.41 (2.92) | 9.21 (2.99) | <0.001* | 0.807 | |||
| Trunk % fat | 65-69 | 35.60 (5.63) | 36.06 (5.78) | 35.95 (5.95) | 0.003 | 0.002 | 0.022 | <0.001* |
| 70-74 | 36.15 (5.55) | 36.35 (5.70) | 36.32 (6.04) | 0.223 | 0.244 | |||
| ≥ 75 | 35.38 (6.70) | 35.33 (6.56) | 34.96 (6.97) | 0.013 | 0.149 | |||
| Whole body fat (kg) | 65-69 | 19.40 (4.93) | 19.60 (5.01) | 19.52 (5.15) | 0.140 | 0.022 | <0.001* | <0.001* |
| 70-74 | 19.74 (5.10) | 19.82 (5.25) | 19.74 (5.33) | 0.998 | 0.126 | |||
| ≥ 75 | 18.38 (5.33) | 18.16 (5.20) | 17.86 (5.37) | <0.001* | 0.545 | |||
| Whole body % fat | 65-69 | 34.60 (4.78) | 35.02 (4.87) | 35.02 (5.07) | <0.001* | 0.004 | 0.003* | <0.001* |
| 70-74 | 35.09 (4.74) | 35.41 (4.81) | 35.42 (5.05) | 0.004 | 0.040 | |||
| ≥ 75 | 34.24 (5.83) | 34.30 (5.63) | 34.08 (5.99) | 0.248 | 0.116 |
N for age group 65-69, 70-74 and ≥ 75: M (569, 560, 390); F (552, 510, 437); % fat = fat mass divided by the weight of that region; *p-value <0.05 after Bonferroni adjustment
Figure 1a-d illustrated the 4-year trend of total body fat mass and fat mass %. For total body fat mass, there was an increasing trend in men and decreasing trend in women but neither reached statistical significance. However, there was a significant increase in fat mass % in both genders. Among the three age groups, the rate of change in fat mass was different, which was relatively stable in the 2 young-old groups but the fat mass declined in the oldest group, aged 75 years or above. (Figure 1e-f; men, p = 0.017; women, p < 0.001). On the contrary, a corresponding rise of fat mass % was again observed, which was steeper in the 2 younger age groups but did not change much in the oldest group. (Figure 1g-h)
Figure 1.
Longitudinal change of total body fat mass and fat mass %
The trend of the corresponding changes in trunk fat mass was illustrated in Figure 2a-d, which demonstrated a statistically significant decreasing trend in women (p < 0.001) but it remained static in men (p = 0.092). Figure 2e-h illustrated that the fat mass in upper limbs of both genders did not change in 4 years (men, p = 0.223; women, p = 0.183), though there was a corresponding increase in fat mass % (both men and women p < 0.001). For the lower limbs, there was a statistically significant increase in fat mass in both men (p < 0.001) and women (p < 0.02). (Figure 2i-l)
Figure 2.
Longitudinal change of trunk, upper limb and lower limb fat mass and fat mass %
Discussions
In this prospective survey, we have demonstrated that total body fat mass remained relatively stable in old age (Figure 1). However, there was a corresponding increase in fat mass % (Figure 1). This is likely due to a more rapid concurrent decline in fat free mass. This pattern may differ from the past studies which had observed that fat would accumulate with ageing (1, 2, 3, 4, 5) but agrees with other reports (9, 10, 11, 12). We must be cautious in distinguishing fat mass and percent fat mass. In our study, fat mass declined but fat mass % increased instead probably due to a more rapid loss in bone and muscle mass. The increase in fat mass % can easily be mistaken as adiposity accumulation. Although fat mass remained stable in the entire cohort, comparing the three age groups, the oldest old group (75 years or above) demonstrated a decline instead of accumulation in fat mass. Other study has demonstrated that a decline in fat mass was adversely associated with mortality in frail older persons (15). It is difficult to separate the effect of ageing and the decline in adiposity on mortality. Nevertheless unintentional weight loss with or without fat loss is a poor prognostic factor in older population in terms of both function and survival. Therefore the decline in adiposity in the oldest old group we have observed in this survey may worth clinical attention. However, whether maintenance of fat mass by dietary intervention is beneficial for survival warrants further studies. At the same time, we have demonstrated that the central or truncal fat declined in women (Figure 2b; p < 0.001) and a declining trend was observed in men though statistically not significant (Figure 2a). Therefore, in contrary to conventional belief, fat does not accumulate in old age and there is a trend of decline in total and central fat, particularly in the oldest old group, older than 75 years of age. In addition, we have observed that fat mass remained relatively stable in upper limb (Figure 2e and Figure 2f) but would accumulate in lower limbs (Figure 2i and Figure 2j). Taken together with the decline in central adiposity, it seems that there is a redistribution of adiposity from central to periphery accompanying the general decline in fat mass. This “migration of adiposity from the central region may offer metabolic advantages but the gaining of fat in the lower limbs may be functionally disadvantageous as fat infiltration is a hallmark of age-associated sarcopenia. It is uncertain whether this migration is contributing or secondary to muscle degeneration. However, unlike the lower limbs, the fat mass in upper limbs remained unchanged. One possibility was that the fat mass in upper limbs was smaller and our measurement might not be adequately sensitive to detect minute changes. On the other hand, in one study comparing pre- and postmenopausal women, the fat mass was similar in the upper limb but was higher in the lower limb in the post-menopausal group (20). This observation is compatible with our findings.
Our survey has several limitations. The follow-up period of four years may not be sufficiently long to observe a more definite change in the pattern of fat mass and fat redistribution across time. The observed change in fat mass was therefore modest and statistically insignificant (1.28% in men and -0.114% in women; p-for trend both insignificant). However, the decline or redistribution we have observed was modest and we are not certain about its clinical significance. Nevertheless, a robust trend is present and with a longer duration, this trend may result in considerable change and caries clinically significant implications. The number of very old participants (older than 80 years) was few (211 participants or 7.01%) in this cohort and the results might have been diluted by the young-old group. If not, a subgroup analysis of this very old group might have been possible to illustrate a more definite pattern of change of adiposity in old age. All our participants were community living and relatively healthy at baseline. A proportion (982 participants or 24.5%) of the participants could not come back for follow up visits at 2 year or 4 year due to either death or ill health. The difference in baseline characteristics between the participants and non-participants were compared. (Table 4) They were older (p < 0.001) and had a lower BMI (p < 0.05). We postulate that they are a group of older adults that will have more rapid fat loss. If it is true, even if they had have been included, the direction of our results would not be reversed and instead could have had become more robust. Our cohort consisted of all Chinese older adults with a relatively low BMI (Table 1) compared with other ethnicities. In addition, Chinese have higher body fat % (21) and visceral fat (22) at the same levels of BMI or waist circumference compared to the westerners. Whether this peripheral redistribution of central adiposity will apply to other ethnicities with less central adiposity is uncertain. Therefore our results may not be applicable to other ethnicities. Further studies with longer follow up period and recruiting more “oldest old participants, older than 80 years, or in other ethnicities are recommended to examine the longitudinal change in fat mass and fat mass redistribution with ageing.
Table 4.
Comparison between participants and defaulters
| Mean (SD)/ Freq. (%) | |||
|---|---|---|---|
| Participants (N=982) | Defaulters (N=3018) | p-value* | |
| Female | 501 (51.02%) | 1499 (49.67%) | 0.4625 |
| Age | <.0001 | ||
| 65-69 | 212 (21.59%) | 1121 (37.14%) | |
| 70-74 | 303 (30.86%) | 1070 (35.45%) | |
| 75 or above | 467 (47.56%) | 827 (27.4%) | |
| mean | 74.49 (5.73) | 71.83 (4.82) | <.0001 |
| Education | <.0001 | ||
| No education | 259 (26.37%) | 597 (19.78%) | |
| Primary or below | 494 (50.31%) | 1513 (50.13%) | |
| Secondary or above | 229 (23.32%) | 908 (30.09%) | |
| Marital status | <.0001 | ||
| Married | 644 (65.58%) | 2185 (72.4%) | |
| Widowed/Separated | 304 (30.96%) | 773 (25.61%) | |
| Single | 34 (3.46%) | 60 (1.99%) | |
| Current smoker | 87 (8.86%) | 188 (6.23%) | 0.0047 |
| Current drinker | 116 (11.81%) | 406 (13.46%) | 0.184 |
| Height (cm) | 156.30 (8.23) | 157.20 (8.20) | 0.0024 |
| Weight (kg) | 57.39 (10.17) | 58.84 (9.63) | <.0001 |
| BMI | 23.45 (3.55) | 23.76 (3.21) | 0.016 |
| Whole body fat (kg) | 16.94 (5.49) | 17.28 (5.23) | 0.0843 |
| Whole body % fat | 29.49 (7.39) | 29.47 (7.16) | 0.9593 |
p-value of t-test or chi-square test
Conclusions
Absolute total body fat mass does not accumulate in old age and in the contrary, in the oldest old group (75 years or above), there is a decline in fat mass instead. The increase in fat mass % is probably due to a more rapid concomitant decline in fat-free mass but not gaining in fat mass. With ageing, fat will redistribute from the central region to the lower limbs. It is uncertain to what extent, this redistribution of adiposity can account for the fat infiltration of the limb muscles accompanying age-associated sarcopenia.
Conflict of interest
None
Ethical standard
This study was approved by the Clinical Research Ethics Committee of The Chinese University of Hong Kong.
References
- 1.Hughes VA, Frontera WR, Roubenoff R, Evans WJ, Singh MA. Longitudinal changes in body composition in older men and women: role of body weight change and physical activity. The American journal of clinical nutrition. 2002;76(2):473–481. doi: 10.1093/ajcn/76.2.473. 10.1093/ajcn/76.2.473 PubMed PMID: 12145025. [DOI] [PubMed] [Google Scholar]
- 2.Visser M, Pahor M, Tylavsky F, Kritchevsky SB, Cauley JA, Newman AB, Blunt BA, Harris TB. One-and two-year change in body composition as measured by DXA in a population-based cohort of older men and women. Journal of applied physiology. 2003;94(6):2368–2374. doi: 10.1152/japplphysiol.00124.2002. 10.1152/japplphysiol.00124.2002 PubMed PMID: 12598481. [DOI] [PubMed] [Google Scholar]
- 3.Hughes VA, Roubenoff R, Wood M, Frontera WR, Evans WJ, Singh MA. Anthropometric assessment of 10-y changes in body composition in the elderly. The American journal of clinical nutrition. 2004;80(2):475–482. doi: 10.1093/ajcn/80.2.475. 10.1093/ajcn/80.2.475 PubMed PMID: 15277173. [DOI] [PubMed] [Google Scholar]
- 4.Jackson AS, Janssen I, Sui X, Church TS, Blair SN. Longitudinal changes in body composition associated with healthy ageing: men, aged 20–96 years. British Journal of Nutrition. 2012;107(7):1085–1091. doi: 10.1017/S0007114511003886. 10.1017/S0007114511003886 PubMed PMID: 21810289. [DOI] [PubMed] [Google Scholar]
- 5.Bazzocchi A, Diano D, Ponti F, Andreone A, Sassi C, Albisinni U, Marchesini G, Battista G. Health and ageing: a cross-sectional study of body composition. Clinical nutrition. 2013;32(4):569–578. doi: 10.1016/j.clnu.2012.10.004. 10.1016/j.clnu.2012.10.004 PubMed PMID: 23111003. [DOI] [PubMed] [Google Scholar]
- 6.Shimokata H, Tobin JD, Muller DC, Elahi D, Coon PJ, Andres R. Studies in the distribution of body fat: I. Effects of age, sex, and obesity. Journal of gerontology. 1989;44(2):M66–73. doi: 10.1093/geronj/44.2.m66. PubMed PMID: 2921472. [DOI] [PubMed] [Google Scholar]
- 7.Schwartz RS, Shuman WP, Bradbury VL, Cain KC, Fellingham GW, Beard JC, Kahn SE, Stratton JR, Cerqueira MD, Abrass IB. Body fat distribution in healthy young and older men. Journal of Gerontology. 1990;45(6):M181–5. doi: 10.1093/geronj/45.6.m181. 10.1093/geronj/45.6.M181 PubMed PMID: 2229940. [DOI] [PubMed] [Google Scholar]
- 8.Kuk JL, Saunders TJ, Davidson LE, Ross R. Age-related changes in total and regional fat distribution. Ageing research reviews. 2009;8(4):339–348. doi: 10.1016/j.arr.2009.06.001. 10.1016/j.arr.2009.06.001 PubMed PMID: 19576300. [DOI] [PubMed] [Google Scholar]
- 9.Steen B. Body composition and aging. Nutrition Reviews. 1988;46(2):45–51. doi: 10.1111/j.1753-4887.1988.tb05386.x. 10.1111/j.1753-4887.1988.tb05386.x PubMed PMID: 3279343. [DOI] [PubMed] [Google Scholar]
- 10.Baumgartner RN, Stauber PM, McHugh D, Koehler KM, Garry PJ. Cross-sectional age differences in body composition in persons 60+ years of age. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 1995;50(6):M307–16. doi: 10.1093/gerona/50a.6.m307. 10.1093/gerona/50A.6.M307 [DOI] [PubMed] [Google Scholar]
- 11.Silver AJ, Guillen CP, Kahl MJ, Morley JE. Effect of aging on body fat. Journal of the American Geriatrics Society. 1993;41(3):211–213. doi: 10.1111/j.1532-5415.1993.tb06693.x. 10.1111/j.1532-5415.1993.tb06693.x PubMed PMID: 8440839. [DOI] [PubMed] [Google Scholar]
- 12.Buffa R, Floris GU, Putzu PF, Marini E. Body composition variations in ageing. Collegium antropologicum. 2011;35(1):259–265. PubMed PMID: 21667542. [PubMed] [Google Scholar]
- 13.Hughes VA, Roubenoff R, Wood M, Frontera WR, Evans WJ, Singh MA. Anthropometric assessment of 10-y changes in body composition in the elderly. The American journal of clinical nutrition. 2004;80(2):475–482. doi: 10.1093/ajcn/80.2.475. 10.1093/ajcn/80.2.475 PubMed PMID: 15277173. [DOI] [PubMed] [Google Scholar]
- 14.Ding J, Kritchevsky SB, Newman AB, Taaffe DR, Nicklas BJ, Visser M, Lee JS, Nevitt M, Tylavsky FA, Rubin SM, Pahor M. Effects of birth cohort and age on body composition in a sample of community-based elderly. The American journal of clinical nutrition. 2007;85(2):405–410. doi: 10.1093/ajcn/85.2.405. 10.1093/ajcn/85.2.405 PubMed PMID: 17284736. [DOI] [PubMed] [Google Scholar]
- 15.Zaslavsky O, Rillamas-Sun E, Li W, Going S, Datta M, Snetselaar L, Zelber-Sagi S. Association of dynamics in lean and fat mass measures with mortality in frail older women. The journal of nutrition, health & aging. 2017;21(1):112–119. doi: 10.1007/s12603-016-0730-1. 10.1007/s12603-016-0730-1 [DOI] [PubMed] [Google Scholar]
- 16.Prior BM, Cureton KJ, Modlesky CM, Evans EM, Sloniger MA, Saunders M, Lewis RD. In vivo validation of whole body composition estimates from dual-energy X-ray absorptiometry. Journal of applied physiology. 1997;83(2):623–630. doi: 10.1152/jappl.1997.83.2.623. 10.1152/jappl.1997.83.2.623 PubMed PMID: 9262461. [DOI] [PubMed] [Google Scholar]
- 17.Albanese CV, Diessel E, Genant HK. Clinical applications of body composition measurements using DXA. Journal of Clinical Densitometry. 2003;6(2):75–85. doi: 10.1385/jcd:6:2:75. 10.1385/JCD:6:2:75 PubMed PMID: 12794229. [DOI] [PubMed] [Google Scholar]
- 18.Kelly TL, Wilson KE, Heymsfield SB. Dual energy X-Ray absorptiometry body composition reference values from NHANES. PloS one. 2009;4(9):e7038. doi: 10.1371/journal.pone.0007038. 10.1371/journal.pone.0007038 PubMed PMID: 19753111, PMCID 2737140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Woo J, Yu R, You F. Fitness, fatness and survival in elderly populations. Age (Dordr) 2013;35(3):973–984. doi: 10.1007/s11357-012-9398-6. 10.1007/s11357-012-9398-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Douchi T, Yamamoto S, Yoshimitsu N, Andoh T, Matsuo T, Nagata Y. Relative contribution of aging and menopause to changes in lean and fat mass in segmental regions. Maturitas. 2002;42(4):301–306. doi: 10.1016/s0378-5122(02)00161-5. 10.1016/S0378-5122(02)00161-5 PubMed PMID: 12191853. [DOI] [PubMed] [Google Scholar]
- 21.Deurenberg P, Deurenberg-Yap M, Guricci S. Asians are different from Caucasians and from each other in their body mass index/body fat per cent relationship. Obesity reviews. 2002;3(3):141–146. doi: 10.1046/j.1467-789x.2002.00065.x. 10.1046/j.1467-789X.2002.00065.x PubMed PMID: 12164465. [DOI] [PubMed] [Google Scholar]
- 22.Lear SA, Humphries KH, Kohli S, Birmingham CL. The use of BMI and waist circumference as surrogates of body fat differs by ethnicity. Obesity. 2007;15(11):2817–2824. doi: 10.1038/oby.2007.334. 10.1038/oby.2007.334 PubMed PMID: 18070773. [DOI] [PubMed] [Google Scholar]