Abstract
Context:
Although obesity is associated with high bone mass, recent reports suggest an increase in the incidence of fractures in obese patients.
Objectives:
The objectives of the study were to evaluate the influence of increasing body fat on bone mineral density (BMD) and to determine the influence of the different adipokines on BMD in frail obese elderly patients.
Design and Setting:
This is a cross-sectional study of baseline characteristics of elderly obese patients participating in a lifestyle therapy with diet with or without exercise and conducted in a university setting.
Patients:
One hundred seventy-three, elderly (≥65 y old), obese (body mass index of ≥30 kg/m2) who were mostly frail participated in the study.
Outcome Measures:
BMD, percentage of total body fat, percentage of fat-free mass, percentage of lean mass, body mass index, adiponectin, leptin, IL-6, bone turnover markers (osteocalcin and C-telopeptide), high-sensitivity C-reactive protein, free estradiol, and 25-hydroxyvitamin D were measured.
Results:
Higher tertiles of percentage body fat and lower lean mass were associated with a lower BMD. High-sensitivity C-reactive protein levels were highest in the highest fat tertile (third, 5.5 ± 5.4 vs first, 1.5 ± 1.3 mg/L, P < .05) for women, whereas IL-6 levels were highest in the highest tertile in men (third, 3.5 ± 3.1 vs first, 1.7 ± 0.8 pg/mL, P < .05). Leptin increased with increasing fat tertiles in both genders (P < .05), whereas adiponectin increased with increasing fat tertiles only in men (P < .05). A multivariate analysis revealed adiponectin as an important mediator of the effect of fat mass on BMD. Osteocalcin levels were highest in the highest fat tertile in women but not in men. Physical function test scores decreased with increasing fat tertiles in women (P < .05) but not in men.
Conclusions:
Increasing adiposity together with decreasing lean mass is associated with lower BMD, higher adipokine levels, and worsening frailty in elderly obese adults.
Traditionally, low body weight is considered as a risk for fractures, whereas obesity is osteoprotective (1). However, this concept has been challenged by recent reports of increased prevalence of vertebral and nonvertebral fractures in obese postmenopausal women and obese older men compared with nonobese age-matched subjects (2–8). Although it is generally accepted that high body weight positively influences the skeleton from increased mechanical loading and higher concentrations of bone-active hormones (9, 10), newer data suggest that increasing adiposity may negatively affect the skeleton (6, 11, 12). A very recent study comparing bone biopsy studies across the different tertiles of trunk fat revealed diminished bone formation among patients in the highest tertile; however, no etiology for the reduced bone formation in these subjects was suggested (13).
Because fat produces a host of substances, termed adipokines, it has been proposed that the increase in the incidence of fractures in obese patients stems from the overproduction of proinflammatory cytokines from the expanded volume of adipose tissues, leading to bone loss (11, 14). Both obesity and aging are associated with a chronic state of low-grade inflammation; thus, obese elderly subjects theoretically represent the subset with the worst inflammatory condition among the population of healthy adults. In this group, it is possible that the heightened state of inflammation from combined aging and obesity results in bone loss and poor bone quality. However, there is little information on the influence of increasing body fat and the levels of the different adipokines on bone mineral density (BMD) in the population of elderly obese subjects who were found to be mostly frail (15). The objective of this study was to determine the influence of body fat and circulating adipokines on BMD among elderly, obese, frail subjects, a group that may be at a high risks for falls and possible fractures.
Materials and Methods
Study design and study population
This study is a cross-sectional analysis of baseline data from subjects who volunteered to participate in two previous lifestyle therapy trials of sedentary, frail, elderly obese patients (16, 17). This study was conducted at Washington University School of Medicine (St Louis, Missouri) in accordance with the guidelines in the Declaration of Helsinki for the ethical treatment of human subjects. The protocol was approved by the Washington University Institutional Review Board. Participant recruitment was through newspaper and radio advertisements. A written informed consent was obtained from each subject. Inclusion/exclusion criteria were as reported previously (16, 17). Briefly, participants were 65 years of age or older, with a body mass index (BMI) of 30 kg/m2 or greater, had a sedentary lifestyle (did not participate in regular exercise more than twice a week), had a stable body weight (±2 kg) over the past year, and were on stable medications for 6 months before enrollment. Those who were treated with bone-acting drugs (eg, bisphosphonates, glucocorticoids, sex steroid compounds) during the previous year were excluded from participation. At enrollment these subjects should not have cardiorespiratory or neuromuscular diseases that would limit their ability to exercise, diabetes mellitus, osteoporosis, hyperparathyroidism, chronic liver disease, uncontrolled or untreated hyperthyroidism and, significant renal impairment.
Frailty was assessed using a modified physical performance test (PPT) as previously described (16, 17). The modified PPT includes seven standardized tasks (walking 50 ft, putting on and removing a coat, picking up a penny, standing up from a chair, lifting a book, climbing one flight of stairs, and performing a progressive Romberg test) plus two additional tasks (climbing up and down four flights of stairs and performing a 360 degree turn). The score for each task ranges from 0 to 4; a perfect score is 36.
BMD and body composition
BMD of the whole body and at the lumbar spine and proximal left femur and body composition (fat mass, lean body mass, and trunk fat) were measured using dual-energy x-ray absorptiometry (DXA) (Delphi 4500/w; Hologic).
Body weight was measured in the morning after the subjects had fasted for 12 hours (16). BMI was calculated as weight in kilograms squared of the patient's height in meters.
Biochemical measurements
Baseline blood samples were obtained in the morning after subjects fasted for at least 12 hours. Serum samples were extracted and stored at −80°C until analysis. ELISA kits were used to measure C-terminal telopeptide of type I collagen (CTX) [Crosslaps; Nordic Bioscience Diagnostics; coefficient of variation (CV) 2.1%] as a marker of bone resorption, osteocalcin (OC) (Metra OC; Quidel Corp; CV 4.4%) as a marker of bone formation, SHBG (Alpco Diagnostics), IL-6 (Quantikine; R&D Systems; CV 7.9%), and adiponectin (Quantikine; R&D Systems; CV 5.5%). RIA kits were used to measure serum 25-hydroxyvitamin D [25(OH)D] (DiaSorin; CV 9.6%), serum estradiol (Ultrasensitive estradiol DSL-4800; Diagnostic Systems Laboratories Inc; CV 7.5%), and leptin (Leptin HL-81K; Linco Research Inc; CV 5.6%). High-sensitive C-reactive protein (hs-CRP) was measured by an immunoturbidimetric assay (Hitachi 917 analyzer; CV 5%). Quality control (QC) procedures for the assays included the use of at least two levels of QC material with each run of the assay. Runs that were not within the established ranges for the QC material were repeated. Pre- and postsamples were batched in the same assay; samples were run in duplicates and those that had CVs of the duplicates greater than 10% were repeated.
Statistical analysis
Results are expressed as means ± SD. A value of P < .05 was considered statistically significant. Normality for outcome variables was verified by Shapiro-Wilks test. Group comparisons for normally distributed variables were analyzed using an ANOVA for continuous variables, whereas categorical variables were compared using χ2 analysis. For variables that failed the Shapiro-Wilks test for normality, nonparametric univariate analyses were done using Kruskal-Wallis test, and multivariate analyses were done with and without outliers as determined by box plots. BMD differences among the different fat tertiles were further analyzed by analysis of covariance adjusted for BMI and for BMI and 25(OH)D separately. Post hoc pair-wise comparisons were performed using the Tukey's honestly significant difference method. Pearson's correlation was used to examine relationships between BMD in all skeletal sites and body composition parameters.
To determine whether circulating adipokines mediate the effect of percentage fat mass on BMD, we performed conditional multivariate regression analyses by adding each adipokine to the relationship between percentage fat mass and BMD. The data were managed using Excel 2010 (Microsoft) and were analyzed using SAS version 9.2 (SAS Institute, Inc).
Results
One hundred seventy three sedentary, elderly, obese adults (92 females, 81 males) participated in the study. The participants had a mean age of 69.5 ± 4.2 years, a BMI of 36.6 ± 25.7 kg/m2, weight of 99.7 ± 15.8 kg, and height of 164.9 ± 14.3 cm. The average modified PPT score to assess frailty available in 134 patients was 28.0 ± 3.2.
As shown is Table 1, there is a positive correlation between BMI and BMD at the femoral neck, intertrochanter, and the whole body, whereas a negative correlation was found between percentage fat mass and BMD in all skeletal sites (Table 1). On the other hand, percentage lean mass correlated positively with BMD in all skeletal sites, whereas no correlation was observed between percentage trunk fat and BMD.
Table 1.
Simple Correlations Between BMD With BMI and Parameters of Body Composition
| BMD | BMI, kg/m2 r | Total Body Fat, % r | Trunk Fat, % r | Total Lean, % R |
|---|---|---|---|---|
| Spine | 0.11 | −0.29a | −0.06 | 0.27a |
| Total hip | 0.10 | −0.40a | 0.10 | 0.38a |
| Femoral neck | 0.21a | −0.22a | 0.05 | 0.20a |
| Trochanter | 0.05 | −0.42a | 0.07 | 0.40a |
| Intertrochanter | 0.16b | −0.31a | 0.09 | 0.29a |
| Whole body | 0.22a | −0.29a | 0.01 | 0.25a |
| Whole-body BMC | 0.13 | −0.48a | 0.07 | 0.51a |
Abbreviation: BMC, bone mineral content.
P < .01.
P < .05.
Table 2 showed analysis according to tertiles of percentage total body fat stratified by gender. Although there was no consistent pattern in body weight across the tertiles, height appears to consistently decrease with increasing tertiles of percentage fat mass in both genders. As a result, BMI increased with increasing percentage fat mass in both female and male participants. An analysis of body composition showed a decreasing percentage fat-free mass and percentage lean mass with increasing percentage total fat mass in both genders (Table 2). These parameters expressed in absolute values (kilogram) showed that both were highest in the lowest compared with the upper two tertiles of percentage fat mass in women, whereas a gradual decrease with increasing tertiles for both parameters were seen in men. The relationship between percentage trunk fat was not consistent. Although there was no association between percentage trunk fat and percentage total fat mass across tertiles in women, surprisingly, percentage trunk fat decreased with increasing percentage total fat mass in men. Furthermore, the absolute trunk fat (kilogram) in women appeared to be highest in the highest tertile of percentage fat mass, whereas no difference was observed for men. The mean modified PPT score was significantly lower in the highest fat tertile in women, whereas there was no difference in the PPT scores across the different fat tertiles in men.
Table 2.
Clinical Characteristics, Body Composition, BMD, and BMC of Female and Male Participants According to Tertiles of Percentage Total Fat
| Clinical Variable | Females (n = 92) |
Males (n = 81) |
||||||
|---|---|---|---|---|---|---|---|---|
| First Tertile (n = 30) | Second Tertile (n = 31) | Third Tertile (n = 31) | P Value | First Tertile (n = 27) | Second Tertile (n = 27) | Third Tertile (n = 27) | P Value | |
| Age, y | 69.6 ± 3.0 | 69.8 ± 3.5 | 69.3 ± 4.7 | .94 | 69.4 ± 3.6 | 69.3 ± 4.5 | 70.4 ± 4.6 | .56 |
| Median (IQR)a | 69.6 (67.0, 71.5) | 68.0 (67.0, 73.0) | 68.0 (66.0, 72.0) | 69.0 (66.0, 71.0) | 67.9 (65.3, 72.0) | 70.0 (66.6, 73.9) | ||
| Weight, kg | 101.0 ± 11.7 | 90.0 ± 13.0 | 102.7 ± 15.6 | <.001 | 107.1 ± 14.3 | 98.1 ± 16.8 | 100.0 ± 19.0 | .12 |
| Height, cm | 172.5 ± 10.5 | 162.4 ± 6.6 | 160.6 ± 7.1 | <.001 | 175.6 ± 8.6 | 159.8 ± 20.8 | 157.6 ± 16.9 | <.001 |
| BMI, kg/m2 | 34.0 ± 3.3 | 35.8 ± 4.4 | 39.0 ± 5.8b,c | <.001 | 35.7 ± 4.0 | 36.0 ± 4.9 | 38.8 ± 5.6 | .04 |
| PPT | 29.4 ± 1.5 | 28.3 ± 3.4 | 26.2 ± 3.9b,c | .001 | 28.5 ± 2.0 | 27.8 ± 3.5 | 29.3 ± 3.9 | .26 |
| Body composition | ||||||||
| Total fat mass, % | 34.2 ± 4.4 | 43.9 ± 2.1a | 49.2 ± 1.8b,c | <.001 | 33.6 ± 3.2 | 42.3 ± 2.2a | 48.6 ± 2.0b,c | <.001 |
| Total Fat mass, kg | 34.59 ± 6.38 | 39.52 ± 6.41a | 50.63 ± 8.42b,c | <.001 | 36.10 ± 6.60 | 41.29 ± 6.41a | 48.68 ± 9.97b,c | <.001 |
| Trunk fat, % | 44.5 ± 11.7 | 47.8 ± 10.2 | 47.5 ± 5.2 | .33 | 57.5 ± 5.9 | 50.2 ± 7.8a | 45.1 ± 9.7b,c | <.001 |
| Trunk fat, kg | 19.86 ± 4.06 | 19.49 ± 3.23 | 23.86 ± 5.22b,c | <.001 | 20.34 ± 3.55 | 22.14 ± 4.48 | 22.12 ± 3.81 | .18 |
| Fat-free mass, % | 63.1 ± 3.6 | 55.9 ± 2.1a | 50.7 ± 1.8b,c | <.001 | 65.6 ± 3.1 | 58.3 ± 2.6a | 50.8 ± 1.6b,c | <.001 |
| Fat-free mass, kg | 66.42 ± 8.28 | 50.45 ± 7.06a | 52.11 ± 7.54b | <.001 | 71.00 ± 9.20 | 56.75 ± 10.93a | 51.36 ± 9.48b,c | <.001 |
| Lean mass, % | 62.8 ± 4.2 | 53.6 ± 2.1a | 48.4 ± 1.7b,c | <.001 | 63.4 ± 3.0 | 57.1 ± 1.7a | 47.8 ± 1.5b,c | <.001 |
| Lean mass, kg | 63.35 ± 7.94 | 48.15 ± 6.85a | 49.63 ± 7.22b | <.001 | 67.84 ± 9.03 | 54.18 ± 10.52a | 48.99 ± 9.11b | <.001 |
| BMD | ||||||||
| Spine | 1.190 ± 0.16 | 1.060 ± 0.14 | 1.063 ± 0.12 | <.001 | 1.228 ± 0.28 | 1.151 ± .13 | 1.152 ± 0.22 | .19 |
| Total femur | 1.049 ± 0.14 | 0.939 ± 0.11 | 0.947 ± 0.14 | <0.001 | 1.082 ± 0.15 | 0.985 ± 0.11 | 0.939 ± 0.11 | <.001 |
| Femoral neck | 0.841 ± 0.11 | 0.815 ± 0.10a | 0.811 ± 0.12b | .03 | 0.876 ± 0.13 | 0.828 ± 0.10a | 0.785 ± 0.13b,c | .002 |
| Trochanter | 0.821 ± 0.14 | 0.699 ± 0.11a | 0.690 ± 0.09b | <.001 | 0.823 ± 0.15 | 0.738 ± 0.10a | 0.711 ± 0.12b | .002 |
| Intertrochanter | 1.224 ± 0.17 | 1.139 ± 0.15a | 1.128 ± 0.13b | .02 | 1.273 ± 0.17 | 1.171 ± 0.13a | 1.141 ± 0.18b,c | .001 |
| Whole body | 1.318 ± 0.15 | 1.161 ± 0.13a | 1.177 ± 0.20b | <.001 | 1.324 ± 0.17 | 1.220 ± 0.13a | 1.152 ± 0.16b | <.001 |
| Whole-body BMC | 3070 ± 523 | 2301 ± 362a | 2485 ± 479b | <.001 | 3165 ± 525 | 2575 ± 503a | 2365 ± 508b | <.001 |
Abbreviation: BMC, bone mineral content; IQR, interquartile range. P value was determined by an ANOVA (BMD adjusted for BMI).
P < .05 first vs second tertile.
P < .05 first vs third tertile.
P < .05 second vs third tertile.
Spine BMD was reduced in the upper two tertiles in both genders, although the difference among the tertiles was not significant in men (Table 2). Femur BMD was significantly lower in the upper two tertiles of percentage total fat mass in all sites (total, neck, trochanter, intertrochanter) relative to the first tertile in both female and male participants (Table 2). There was only one extreme outlier (ie, intertrochanter among males); the analysis remained the same without this outlier. Whole-body BMD and whole-body BMC were also significantly higher in the lowest tertile of percentage total fat compared with the upper two tertiles in both female and male participants.
Biochemical studies are shown in Table 3. Results showed that in women, the hs-CRP level increased with the increasing tertile of percent fat mass and was highest among women in the third tertile, whereas there was no difference in IL-6 level. In men, the IL-6 level increased with increasing fat tertiles and was highest in the third tertile, but there was no difference in hs-CRP. Leptin also increased with the increasing fat tertiles in both genders, whereas adiponectin increased in the upper tertiles of fat mass but only in men. In women, OC was significantly higher in the high fat tertiles, whereas the CTX levels were comparable across the tertiles. In men, there was no difference in the CTX levels across the tertiles, but OC decreased with the increasing tertiles of fat mass, although the differences across the tertiles did not reach statistical significance. Free estradiol was significantly lower in the upper two tertiles of percentage fat mass in women. Although there was no significant difference across tertiles in men, free estradiol tended to be lower in the upper two tertiles. However, compared with the corresponding tertiles in women, these levels still appear slightly higher. There were no significant differences in 25(OH)D among the different tertiles of fat mass in either males or females. Ten percent of the participants had vitamin D deficiency (<10 ng/mL), and 39% were vitamin D insufficient (10–19 ng/mL), whereas 51% are vitamin D sufficient (≥20 ng/mL).
Table 3.
Biochemical Characteristics of Female and Male Participants According to Tertiles of Percentage Total Fat
| Biochemical Variable | Females (n = 92) |
Males (n = 81) |
||||||
|---|---|---|---|---|---|---|---|---|
| First Tertile (n = 30) | Second Tertile (n = 31) | Third Tertile (n = 31) | P Value | First Tertile (n = 27) | Second Tertile (n = 27) | Third Tertile (n = 27) | P Value | |
| IL-6, pg/mL | 2.8 (1.8, 3.6) | 2.2 (1.3, 3.2) | 3.0 (1.5, 4.5) | .62 | 1.7 (1.4, 2.0) | 2.7 (2.2, 3.2) | 3.5 (2.3, 4.8)a | .006 |
| hs-CRP, mg/L | 1.5 (1.0, 2.0) | 4.1 (2.3, 5.8) | 5.5 (3.5, 7.5)a | .002 | 2.8 (1.3, 4.4) | 3.6 (2.0, 5.3) | 3.7 (0.6, 6.7) | .82 |
| Leptin, μg/L | 28.3 (18.8, 37.8) | 38.4 (32.1, 44.6) | 56.8 (48.3, 65.3)a,b | <.001 | 16.4 (12.9, 19.7) | 43.3 (30.4, 56.3)c | 61.7 (50.0, 73.4)a,b | <.001 |
| Adiponectin, ng/mL | 38.6 (22.2, 55.0) | 30.5 (18.1, 42.8) | 29.1 (22.1, 36.1) | .55 | 22.6 (16.4, 28.8) | 50.7 (30.7, 70.7) | 90.5 (47.4,133.6)a,b | <.001 |
| CTX, ng/mL | 0.333 (0.29, 0.37) | 0.359 (0.29, 0.43) | 0.335 (0.28, 0.39) | .86 | 0.362 (0.31, 0.42) | 0.416 (0.32, 0.51) | 0.365 (0.30, 0.43) | .47 |
| OC, ng/L | 16.1 (9.4, 22.8) | 23.5 (19.8, 27.2) | 25.0 (21.2, 28.8)a | .03 | 12.2 (10.6, 13.8) | 11.0 (8.3, 13.6) | 9.2 (6.7, 11.7) | .14 |
| FEI, pmol/nmol | 2.1 (1.7, 2.5) | 1.2 (0.8, 1.5)c | 1.4 (0.9, 2.0) | .02 | 2.0 (1.6, 2.5) | 1.7 (1.0, 2.4) | 1.7 (1.2, 2.2) | .53 |
| 25(OH)D, ng/mL | 22.5 (18.5, 27.5) | 22.1 (19.4, 24.8) | 21.2 (17.5, 24.8) | .86 | 18.5 (16.7, 20.3) | 19.3 (15.8, 22.7) | 17.9 (13.2, 22.6) | .82 |
Abbreviation: FEI, free estradiol index. Cell format was mean (95% confidence interval). P value was determined by an ANOVA.
P < .05 first vs third tertile.
P < .05 second vs third tertile.
P < .05 first vs second tertile.
When adjusted for 25(OH)D levels, spine BMD, femoral neck BMD, and trochanter BMD in the upper tertiles remained significantly lower (P = .045, P = .05, and P = .002, respectively) compared with the lowest tertile in women. However, the significance for the intertertile difference in the total hip became borderline (P = .07), whereas it was no longer significant for the intertrochanter and whole body. In men, BMD remained significantly lower in the upper tertiles compared with the lowest tertile in the total hip, femoral neck, trochanter, intertrochanter, and whole body (P = .004, P = .003, P = .009, P = .002, and P < 0.001, respectively). Intertertile difference in spine BMD remained nonsignificant, even after 25(OH)D adjustment (P = .06).
Adding the different adipokines individually into a multivariate model (Table 4) showed that the negative association (from a simple regression) between fat mass with BMD of the spine, total hip, and intertrochanter in women and with BMD of the femoral neck, trochanter, and intertrochanter in men was lost after adiponectin was added to the model. The addition of leptin to the model also removed the association between fat mass and BMD of the intertrochanter in men. However, the association between fat mass and BMD of the trochanter in women, and of the total femur and whole body in men, were not affected by the addition of any adipokine.
Table 4.
Conditional Analysis on the Role of Adipokines on the Relationship Between Percentage Fat and BMD
| BMD Site | Females |
Males |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Univariate Analysis |
Multivariate |
Univariate |
Multivariate |
|||||||
| STBa | P Value | Adipokine | STB | P Value | STB | P Value | Adipokine | STB | P Value | |
| Spine | −0.40 | <.001 | Percentage fat mass | −0.19 | .098 | −0.20 | .07 | Percentage fat mass | −0.20 | .07 |
| Adiponectin | 0.29 | .02 | ||||||||
| Total hip | −0.35 | <.001 | Percentage fat mass | −0.19 | .10 | −0.45 | <.001 | Percentage fat mass | −0.45 | <.001 |
| Adiponectin | 0.13 | .28 | ||||||||
| Femoral neck | −0.14 | .18 | −0.29 | .009 | Percentage fat mass | −0.19 | .25 | |||
| Adiponectin | −0.16 | .32 | ||||||||
| Trochanter | −0.46 | <.001 | Percentage fat mass | −0.46 | <.001 | −0.37 | <.001 | Percentage fat mass | −0.16 | .12 |
| Adiponectin | −0.25 | .31 | ||||||||
| Intertrochanter | −0.32 | .002 | Percentage fat mass | −0.16 | .18 | −0.28 | .01 | Percentage fat mass | −0.10 | .53 |
| Adiponectin | 0.13 | .27 | Adiponectin or | −0.28 | .10 | |||||
| Percentage fat mass | −0.27 −0.03 | .06 | ||||||||
| Leptin | .84 | |||||||||
| Whole body | −0.19 | .07 | Percentage fat mass | −0.19 | .07 | −0.41 | <.001 | Percentage fat mass | −0.41 | <.001 |
Abbreviation: STB, standardized β. The role of adipokines in the relationship of percentage fat on each BMD outcome was tested by entering each adipokine (adiponectin, leptin, hs-CRP, and IL-6) to the model one at a time. Those adipokines that intervened between a BMD outcome and percentage fat and/or that were independent predictors of BMD outcome are listed in the table under multivariate columns.
STB is the standardized β, the regression effect of a 1 SD change in the predictor variable; the STB for the univariate case is the Pearson correlation coefficient.
Discussion
Our results showed that, although a positive correlation was observed between BMI and BMD for certain skeletal sites (femoral neck and intertrochanter and whole body), overall, a higher fat mass and lower lean mass was associated with lower BMD in both genders in this population of obese older adults. Inflammatory cytokines (hs-CRP for women and IL-6 for men) and leptin increased with increasing fat mass in both genders, whereas adiponectin level increased with increasing fat mass only in men. Regardless, multivariate analysis identified adiponectin as the important mediator of the association between fat mass and BMD. Osteocalcin increased with increasing fat mass in women but not men. The levels of 25OHD appear to attenuate the BMD differences across fat tertiles in women but not in men. Physical function test scores were also lower with increasing fat mass in women, but this observation was not seen in men.
Using BMI, prior studies have concluded that a low body weight is a risk factor for osteoporosis, whereas obesity is osteoprotective because it is associated with higher BMD (1, 18). The higher BMD in overweight/obese subjects is presumed to result from a combination of the positive effects of increased mechanical loading on bone and increased estradiol levels due to increased conversion of androgen precursors to estrogen in the expanded adipose tissue volume (9, 10). Recently data from epidemiological studies challenged this concept because newer evidence suggested that although the prevalence of fractures is reduced as BMI goes from underweight to normal, an increase in BMI from overweight to obese is actually associated with a higher prevalence of fractures (BMD) (2). In the Osteoporotic Fractures in Men study, after controlling for BMD, obese men were 5 times more likely to experience a hip fracture than normal-weight men (2). Moreover, recent reports have indicated that a high body fat is associated with a low BMD (11, 12), osteoporosis, or a high fracture risk (3, 6, 8). In agreement with these reports, our data showed a consistently lower BMD among the patients in the higher fat tertiles in both genders, particularly in the different regions of the femur. Given that whole-body BMC is also lower in the upper two tertiles of the percentage fat mass and that this skeletal phenotype persisted, even with a correction for bone area (ie, whole body BMD), this association may raise the possibility of an increase in systemic bone loss as adiposity increases.
In the elderly obese, aside from varying rates of age-related bone loss, an increase in inflammatory markers and, for women, postmenopausal estrogen loss could be significant contributors to differences in BMD across tertiles. Our data demonstrated high levels of inflammatory cytokines in the higher fat tertiles, even after stratification by sex, hs-CRP for women, and IL-6 for men. Adipose tissues produce inflammatory cytokines, which promote osteoclastogenesis, and increased bone resorption (19, 20). Thus, obesity could contribute to age-related bone loss through an added increase in inflammatory cytokines. High hs-CRP is associated with poorer bone strength and an increased risk of fragility fractures (21, 22). IL-6, which is a potent stimulator of bone resorption (23), is elevated in postmenopausal women and patients with chronic inflammatory disease (eg, rheumatoid arthritis) and is likely responsible for increased prevalence of osteoporosis independent of steroid therapy (24, 25) in these patients. In our analysis, although cytokines were elevated in the higher tertiles in our patients (hs-CRP in women and IL-6 in men), these did not mediate the effect of percentage fat mass on BMD.
On the other hand, adiponectin emerged as a major predictor of adipokine (spine) or a mediator of the apparent negative effect of fat mass on BMD in several skeletal sites in both genders. The role of leptin and adiponectin on the skeleton remains uncertain. Leptin administered centrally into the third ventricle has been reported to decrease bone formation via the sympathetic nervous system in leptin-deficient and -sufficient mice (26, 27). In humans, the subcutaneous administration of recombinant leptin to children with congenital leptin deficiency results in increased bone mass, loss of fat mass, and decreased body weight (28, 29). A meta-analysis of the 931 studies showed that leptin was positively associated with BMD, and high leptin levels were predictive of a low risk of fractures (30). In this same study, adiponectin was negatively associated with BMD. Although some investigators demonstrated that adiponectin stimulates both osteoblastogenesis and the osteoclast receptor activator, nuclear factor-κB ligand pathway (31), others reported that adiponectin increases bone mass by suppressing osteoclastogenesis and by activating osteoblastogenesis (32, 33). Nevertheless, most association studies showed negative correlations between BMD and adiponectin in premenopausal and postmenopausal women and in men (30, 34, 35). In the Osteoporotic Fractures in Men study, the risk of fracture increased in parallel with increasing serum adiponectin (36). Although adiponectin levels were low in obese and diabetic patients (37), in our study it increased with increasing fat mass in men and mediated for the negative effect of fat on BMD as well.
BMD across tertiles seems to be influenced by 25(OH)D levels in women but not men. Although we are not aware of gender differences on the effect of 25(OH)D on BMD, a functional synergy between the active vitamin D metabolite (1,25-dihydroxyvitamin D) and estradiol, mainly in women, has been reported (38, 39). To what degree this interaction influenced our results remains undetermined. Free estradiol is reduced among women in the higher fat tertiles and may partly contribute to the lower BMD in these subjects. Although this was not observed in men, the slightly higher free estradiol level among men in the upper tertiles compared with women in the corresponding tertiles is consistent with the findings from the Mayo Clinic group of lower bioavailable estradiol levels in postmenopausal women relative to age-matched men (40). Finally, although osteocalcin has been regarded as a marker of bone turnover, recent reports suggest that it is a regulator of insulin sensitivity (41). Osteocalcin increased with increasing fat tertiles in women. Given the lack of an associated increase in CTX, it is possible that the glucose homeostasis in our obese women contributed to the observed osteocalcin levels.
Nonetheless, our findings are in agreement with our current understanding of the differentiation of adipose and bone tissues, ie, adipocytes and osteoblasts come from a common progenitor, pluripotent mesenchymal stromal cells. These cells can differentiate into either the adipogenic or osteogenic lineage, depending on the regulatory factors present (42). Activation of peroxisome proliferators activated receptor-γ pathway drives differentiation toward the adipogenic pathway while at the same time suppressing the osteogenic pathway, whereas activation of bone transcription factors (runt-related transcription factor 2 [RUNX2], dishevelled and axin [DIX5], osterix) favors osteogenesis over adipogenesis (43). It is possible that this mechanism in combination with the adipose-derived factors contribute to influence the BMD in our patients. On the other hand, we also recognize the importance of reduction in muscle strain and muscle-derived osteogenic factors from reduced lean or muscle mass on BMD (44, 45). Because the percentage of lean mass decreases as adiposity increases, we believe that the balance between fat and lean mass is a significant contributor to BMD in our patients.
This study has limitations. With this cross-sectional study design, we could not establish a causal mechanism for the association between BMD with fat mass and adipokines. Furthermore, although most subjects were elderly, obese, and frail, we did not have PPT results on 39 subjects. However, it is highly likely that most of these participants were frail because 96% of elderly obese have been reported as frail (15) and thus unlikely to introduce bias. Another limitation is the potential error using DXA on BMD measurement in obese patients. Errors of interpretation in DXA studies have been reported to be greater in individuals in the extremes of body weight (ie, morbidly obese and the very lean) (46). As in our previous studies (47, 48), our approaches to minimize the DXA measurement errors in obese patients included the following: 1) the scans were done at a low speed to minimize the noise and improve the scan image, 2) we wanted to ensure that the bone map could be visualized during the scan or the scan was repeated, and 3) a single expert technician performed all the DXA acquisitions. We believe these measures should make this potential error very minimal or unlikely. Lastly, our biomarker data showed wide variability. Because we followed strict QC procedures in our assays, we speculate that this variability could be related to the characteristics inherent in our population of elderly patients with varying degrees of obesity and functional capacity.
Despite the above limitations, our results have relevant clinical implications. Avoiding obesity during early childhood and into adulthood is not only important in achieving an adequate peak bone mass but necessary in maintaining bone mass with aging. It is possible that long-term weight loss in an otherwise healthy obese elderly adult may actually exert a positive effect on bone metabolism by reducing inflammatory cytokine/adipokine-producing fat cells and shifting cellular differentiation from fat to bone cells. This may allay the concern about weight loss-associated bone loss.
Conclusion
To our knowledge, this is the first study to report the potential role of the different adipokines as mediators for the association between fat mass and BMD in the population of mostly frail elderly obese patients. Our results showed that increasing body fat together with decreasing lean mass negatively influence BMD in these patients. Increasing fat mass was also associated with increasing levels of adipokines, some of which could be inflammatory, ie, hs-CRP in women and Il-6 in men, and may lead to bone loss, whereas others (ie, leptin and adiponectin) affect bone in a complex unclear manner but in combination might result in lower BMD as adiposity increases. Our analysis, however, showed adiponectin and to a certain extent leptin mediate the effect of fat mass on the BMD in most skeletal sites and not the inflammatory cytokines. Furthermore, our results revealed that frailty was worse among the elderly obese subjects in the highest fat group who also had the lowest BMD. Because these subjects have physical functional deficits, it is possible that they are at the highest risk for falls and potential fractures.
Acknowledgments
This work was supported by National Institutes of Health Grants , R01-AG031176, P30-DK56341 (to the Clinical Nutrition Research Unit), UL1-RR024992 (Clinical and Translational Science Award), and DK20579 (to the Diabetes Research and Training Center).
This study was presented, in part, as an oral presentation at 94th Annual Meeting of The Endocrine Society, 2012, Houston, Texas.
Disclosure Summary: The authors declare no conflicts of interest.
Footnotes
- BMD
- bone mineral density
- BMI
- body mass index
- CTX
- C-terminal telopeptide of type I collagen
- CV
- coefficient of variation
- DXA
- dual-energy x-ray absorptiometry
- hs-CRP
- high-sensitive C-reactive protein
- OC
- osteocalcin
- 25(OH)D
- 25-hydroxyvitamin D
- PPT
- physical performance test
- QC
- quality control.
References
- 1. Felson DT, Zhang Y, Hannan MT, Anderson JJ. Effects of weight and body mass index on bone mineral density in men and women: the Framingham study. J Bone Miner Res. 1993;8(5):567–573 [DOI] [PubMed] [Google Scholar]
- 2. Nielson CM, Marshall LM, Adams AL, et al. BMI and fracture risk in older men: the Osteoporotic Fractures in Men Study (MrOS). J Bone Miner Res. 2011;26(3):496–502 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Armstrong ME, Cairns BJ, Banks E, Green J, Reeves GK, Beral V. Different effects of age, adiposity and physical activity on the risk of ankle, wrist and hip fractures in postmenopausal women. Bone. 2012;50(6):1394–1400 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Gnudi S, Sitta E, Lisi L. Relationship of body mass index with main limb fragility fractures in postmenopausal women. J Bone Miner Metab. 2009;27(4):479–484 [DOI] [PubMed] [Google Scholar]
- 5. Premaor MO, Pilbrow L, Tonkin C, Parker RA, Compston J. Obesity and fractures in postmenopausal women. J Bone Miner Res. 2010;25(2):292–297 [DOI] [PubMed] [Google Scholar]
- 6. Compston JE, Watts NB, Chapurlat R, et al. Obesity is not protective against fracture in postmenopausal women: GLOW. Am J Med. 2011;124(11):1043–1050 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Pirro M, Fabbriciani G, Leli C, et al. High weight or body mass index increase the risk of vertebral fractures in postmenopausal osteoporotic women. J Bone Miner Metab. 2010;28(1):88–93 [DOI] [PubMed] [Google Scholar]
- 8. Laslett LL, Just Nee Foley SJ, Quinn SJ, Winzenberg TM, Jones G. Excess body fat is associated with higher risk of vertebral deformities in older women but not in men: a cross-sectional study. Osteoporos Int. 2012;23(1):67–74 [DOI] [PubMed] [Google Scholar]
- 9. Reid IR. Relationships among body mass, its components, and bone. Bone. 2002;31(5):547–555 [DOI] [PubMed] [Google Scholar]
- 10. Kirschner MA, Schneider G, Ertel NH, Worton E. Obesity, androgens, estrogens, and cancer risk. Cancer Res 1982;42(suppl 8):3281s–3285s [PubMed] [Google Scholar]
- 11. Zhao LJ, Liu YJ, Liu PY, Hamilton J, Recker RR, Deng HW. Relationship of obesity with osteoporosis. J Clin Endocrinol Metab. 2007;92(5):1640–1646 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Bhupathiraju SN, Dawson-Hughes B, Hannan MT, Lichtenstein AH, Tucker KL. Centrally located body fat is associated with lower bone mineral density in older Puerto Rican adults. Am J Clin Nutr. 2011;94(4):1063–1070 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Cohen A, Dempster DW, Recker RR, et al. Abdominal fat is associated with lower bone formation and inferior bone quality in healthy premenopausal women: a transiliac bone biopsy study. J Clin Endocrinol Metab. 2013;98(6):2562–2572 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Zhao LJ, Jiang H, Papasian CJ, et al. Correlation of obesity and osteoporosis: effect of fat mass on the determination of osteoporosis. J Bone Miner Res. 2008;23(1):17–29 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Villareal DT, Banks M, Siener C, Sinacore DR, Klein S. Physical frailty and body composition in obese elderly men and women. Obes Res. 2004;12(6):913–920 [DOI] [PubMed] [Google Scholar]
- 16. Villareal DT, Chode S, Parimi N, et al. Weight loss, exercise, or both and physical function in obese older adults. N Engl J Med. 2011;364(13):1218–1229 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Villareal DT, Banks M, Sinacore DR, Siener C, Klein S. Effect of weight loss and exercise on frailty in obese older adults. Arch Intern Med. 2006;166(8):860–866 [DOI] [PubMed] [Google Scholar]
- 18. Albala C, Yanez M, Devoto E, Sostin C, Zeballos L, Santos JL. Obesity as a protective factor for postmenopausal osteoporosis. Int J Obes Relat Metab Disord. 1996;20(11):1027–1032 [PubMed] [Google Scholar]
- 19. Redlich K, Smolen JS. Inflammatory bone loss: pathogenesis and therapeutic intervention. Nat Rev Drug Discov. 2012;11(3):234–250 [DOI] [PubMed] [Google Scholar]
- 20. Mundy GR. Osteoporosis and inflammation. Nutr Rev. 2007;65(12 Pt 2):S147–S151 [DOI] [PubMed] [Google Scholar]
- 21. Pasco JA, Kotowicz MA, Henry MJ, et al. High-sensitivity C-reactive protein and fracture risk in elderly women. JAMA. 2006;296(11):1353–1355 [DOI] [PubMed] [Google Scholar]
- 22. Ishii S, Cauley JA, Greendale GA, et al. C-reactive protein, bone strength, and nine-year fracture risk: data from the Study of Women's Health Across the Nation (SWAN). J Bone Miner Res. 2013;28(7):1688–1698 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. De BF, Rucci N, Del FA, et al. Impaired skeletal development in interleukin-6-transgenic mice: a model for the impact of chronic inflammation on the growing skeletal system. Arthritis Rheum. 2006;54(11):3551–3563 [DOI] [PubMed] [Google Scholar]
- 24. Pfeilschifter J, Koditz R, Pfohl M, Schatz H. Changes in proinflammatory cytokine activity after menopause. Endocr Rev. 2002;23(1):90–119 [DOI] [PubMed] [Google Scholar]
- 25. Edwards CJ, Williams E. The role of interleukin-6 in rheumatoid arthritis-associated osteoporosis. Osteoporos Int. 2010;21(8):1287–1293 [DOI] [PubMed] [Google Scholar]
- 26. Takeda S, Elefteriou F, Levasseur R, et al. Leptin regulates bone formation via the sympathetic nervous system. Cell. 2002;111(3):305–317 [DOI] [PubMed] [Google Scholar]
- 27. Ducy P, Amling M, Takeda S, et al. Leptin inhibits bone formation through a hypothalamic relay: a central control of bone mass. Cell. 2000;100(2):197–207 [DOI] [PubMed] [Google Scholar]
- 28. Farooqi IS, Jebb SA, Langmack G, et al. Effects of recombinant leptin therapy in a child with congenital leptin deficiency. N Engl J Med. 1999;341(12):879–884 [DOI] [PubMed] [Google Scholar]
- 29. Farooqi IS, Matarese G, Lord GM, et al. Beneficial effects of leptin on obesity, T cell hyporesponsiveness, and neuroendocrine/metabolic dysfunction of human congenital leptin deficiency. J Clin Invest. 2002;110(8):1093–1103 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Biver E, Salliot C, Combescure C, et al. Influence of adipokines and ghrelin on bone mineral density and fracture risk: a systematic review and meta-analysis. J Clin Endocrinol Metab. 2011;96(9):2703–2713 [DOI] [PubMed] [Google Scholar]
- 31. Kanazawa I. Adiponectin in metabolic bone disease. Curr Med Chem. 2012;19(32):5481–5492 [DOI] [PubMed] [Google Scholar]
- 32. Oshima K, Nampei A, Matsuda M, et al. Adiponectin increases bone mass by suppressing osteoclast and activating osteoblast. Biochem Biophys Res Commun. 2005;331(2):520–526 [DOI] [PubMed] [Google Scholar]
- 33. Luo XH, Guo LJ, Yuan LQ, et al. Adiponectin stimulates human osteoblasts proliferation and differentiation via the MAPK signaling pathway. Exp Cell Res. 2005;309(1):99–109 [DOI] [PubMed] [Google Scholar]
- 34. Register TC, Divers J, Bowden DW, et al. Relationships between serum adiponectin and bone density, adiposity and calcified atherosclerotic plaque in the African American-Diabetes Heart Study. J Clin Endocrinol Metab. 2013;98(5):1916–1922 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Tohidi M, Akbarzadeh S, Larijani B, et al. Omentin-1, visfatin and adiponectin levels in relation to bone mineral density in Iranian postmenopausal women. Bone. 2012;51(5):876–881 [DOI] [PubMed] [Google Scholar]
- 36. Johansson H, Oden A, Lerner UH, et al. High serum adiponectin predicts incident fractures in elderly men: Osteoporotic Fractures in Men (MrOS) Sweden. J Bone Miner Res. 2012;27(6):1390–1396 [DOI] [PubMed] [Google Scholar]
- 37. Weyer C, Funahashi T, Tanaka S, et al. Hypoadiponectinemia in obesity and type 2 diabetes: close association with insulin resistance and hyperinsulinemia. J Clin Endocrinol Metab. 2001;86(5):1930–1935 [DOI] [PubMed] [Google Scholar]
- 38. Disanto G, Handel AE, Ramagopalan SV. Estrogen-vitamin D interaction in multiple sclerosis. Fertil Steril. 2011;95(1):e3. [DOI] [PubMed] [Google Scholar]
- 39. Correale J, Ysrraelit MC, Gaitan MI. Gender differences in 1,25 dihydroxyvitamin D3 immunomodulatory effects in multiple sclerosis patients and healthy subjects. J Immunol. 2010;185(8):4948–4958 [DOI] [PubMed] [Google Scholar]
- 40. Khosla S, Melton LJ, III, Atkinson EJ, O'Fallon WM. Relationship of serum sex steroid levels to longitudinal changes in bone density in young versus elderly men. J Clin Endocrinol Metab. 2001;86(8):3555–3561 [DOI] [PubMed] [Google Scholar]
- 41. Gower BA, Pollock NK, Casazza K, Clemens TL, Goree LL, Granger WM. Associations of total and undercarboxylated osteocalcin with peripheral and hepatic insulin sensitivity and β-cell function in overweight adults. J Clin Endocrinol Metab. 2013;98(7):E1173–E1180 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Akune T, Ohba S, Kamekura S, et al. PPARγ insufficiency enhances osteogenesis through osteoblast formation from bone marrow progenitors. J Clin Invest. 2004;113(6):846–855 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Pei L, Tontonoz P. Fat's loss is bone's gain. J Clin Invest. 2004;113(6):805–806 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Armamento-Villareal R, Aguirre L, Napoli N, et al. Changes in thigh muscle volume predict bone mineral density response to lifestyle therapy in frail, obese older adults. Osteoporos Int 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Hamrick MW. A role for myokines in muscle-bone interactions. Exerc Sport Sci Rev. 2011;39(1):43–47 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Shapses SA, Sukumar D. Bone metabolism in obesity and weight loss. Annu Rev Nutr. 2012;32:287–309 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Villareal DT, Shah K, Banks MR, Sinacore DR, Klein S. Effect of weight loss and exercise therapy on bone metabolism and mass in obese older adults: a one-year randomized controlled trial. J Clin Endocrinol Metab. 2008;93(6):2181–2187 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Shah K, Armamento-Villareal R, Parimi N, et al. Exercise training in obese older adults prevents increase in bone turnover and attenuates decrease in hip bone mineral density induced by weight loss despite decline in bone-active hormones. J Bone Miner Res. 2011;26(12):2851–2859 [DOI] [PMC free article] [PubMed] [Google Scholar]