Menopausal Hormone Therapy Is Associated With Reduced Total and Visceral Adiposity: The OsteoLaus Cohort

Menopausal Hormone Therapy Is Associated With Reduced Total and Visceral Adiposity: The OsteoLaus... Abstract Context After menopause, fat mass (FM) and visceral adipose tissue (VAT) increase and nonbone lean body mass (LBM) decreases. Whether menopausal hormone therapy (MHT) reverses these changes remains controversial. Objective To assess the effect of MHT on FM, VAT, and LBM before and after its withdrawal and evaluate potential confounders. Design Cross-sectional study. Setting General community. Patients or Other Participants Women of the OsteoLaus cohort (50 to 80 years old) who underwent dual-energy X-ray absorptiometry (DXA) with body composition assessment. After we excluded women with estrogen-modifying medications, the 1053 participants were categorized into current users (CUs), past users (PUs), and never users (NUs) of MHT. Intervention None. Main Outcome Measures VAT measured by DXA was the primary outcome. We assessed subtotal and android FM, LBM, muscle strength (hand grip), and confounding factors (caloric intake, physical activity, biomarkers). Results The groups significantly differed in age, NU < CU < PU. Age-adjusted VAT was lower in CUs than NUs (P = 0.03). CUs exhibited lower age-adjusted body mass index (BMI) (−0.9 kg/m2) and a trend for lower FM (−1.3 kg). The 10-year gain of VAT (P < 0.01) and subtotal and android FM (P < 0.05) was prevented in CUs. No difference in LBM or hand grip was detected. No residual effect was detected for PUs, including for early MHT discontinuers. The confounding factors did not significantly differ between groups except for higher caloric intake in PUs compared with NUs. Conclusions MHT is associated with significantly decreased VAT, BMI, and android FM. No benefit is detected for LBM. The benefits are not preserved in PUs, suggesting caution when MHT is discontinued. Menopause is accompanied by changes in bone, fat, and muscular compartments (1, 2). In particular, menopause transition has been linked to increased propensity for weight gain and fat mass (FM) accumulation (3, 4). Whether this association is caused by declining ovarian hormone secretion or aging remains an open question (2). Data are more robust regarding the effect of menopause on regional fat. Several prospective studies have shown a greater increase of abdominal fat after menopause, leading to a shift from a gynoid to an android pattern of fat distribution (5–7). The causal association with estrogen deficiency is supported by preclinical data demonstrating that disruption of estradiol (E2) signaling by estrogen receptor (ER) deletion or ovariectomy (OVX) accelerates fat accumulation (8). It is important to emphasize that excess of central fat, and specifically of visceral adipose tissue (VAT) in humans, is associated with insulin resistance and high prevalence of metabolic syndrome, which are risk factors for atherosclerotic cardiovascular disease (9). A decline in nonbone lean body mass (LBM), also called fat-free or skeletal muscle mass, has been described across menopause (3, 4). It remains unclear whether this finding is a consequence of estrogen deficiency or of indirect factors such as a more sedentary lifestyle (10). Interventional trials assessing the effect of menopausal hormone therapy (MHT) on body composition have yielded mixed results regarding total FM and LBM (8). Those inconsistent findings can reflect differences in the population studied, study design (natural vs induced menopause), type of MHT, and method for assessing body composition. Conversely, most studies evaluating the effect of gonadotropin-releasing hormone agonists (GnRH-Ags), creating an artificial menopause state, have found increased total adiposity and intra-abdominal fat (8). Interestingly, the most recent one (11) showed that this phenotype could be prevented by estrogen administration. Another point that remains unclear is whether the eventual impact of MHT on FM is the result of a direct effect on adipocytes or indirect mechanisms such as altered energy intake or energy expenditure (8) or behavioral effects on mood and anxiety (12), which in turn might affect food intake and physical activity. In addition, insulin and adipokines (leptin, adiponectin) have been suggested as potential modifiers in the crosstalk between the reproductive axis and energy homeostasis both centrally and peripherally (7, 13). In this cross-sectional study, we assessed the effect of MHT on FM, VAT, and LBM before and after its withdrawal and attempted to explore potential confounders as detailed earlier. Materials and Methods Setting We analyzed data from the OsteoLaus study (14). OsteoLaus is a substudy of the CoLaus study, an ongoing prospective study aiming to assess the determinants of cardiovascular disease by using a population-based sample drawn from the city of Lausanne, Switzerland (15). The aims of the OsteoLaus study are to compare different models of fracture risk prediction and to assess the relationship between osteoporosis and cardiovascular diseases. Recruitment of OsteoLaus participants was detailed previously (16). CoLaus data (second visit) were collected within 6 months before the OsteoLaus visit. The study was approved by the Institutional Ethics Committee of the University of Lausanne. All participants signed an informed consent. Participants A total of 1500 postmenopausal women, aged 50 to 80 years, were questioned on current or past MHT use, its type, and duration, if applicable. All participants underwent a spine and hip dual-energy X-ray absorptiometry (DXA) scan on a Discovery DXA System (Hologic, Inc., Marlborough, MA). We included in this study all the women for whom body composition assessment was performed during the DXA scan (n = 1086). Exclusion criteria were intake of medication with estrogen-mediated effects (aromatase inhibitors, tamoxifen, antiandrogens), extreme body mass index (BMI) values (BMI >37 kg/m2), and uninterpretable or incomplete DXA scans (low-quality images). The remaining participants were divided into three groups: current users (CUs), past users (PUs), and never users (NUs) of MHT. CUs were taking MHT at trial entry or discontinued treatment <6 months earlier. PUs discontinued MHT ≥6 months before trial entry (otherwise considered as CUs). MHT use for <6 months, reported in 25 participants (<3 months in 23/25), was considered unlikely to cause considerable changes in body composition, and these subjects were classified as NUs. DXA measurements All body composition measurements were in accordance with published guidelines by the International Society for Clinical Densitometry (17). The subjects were placed in a supine position with palms down and arms at sides, slightly separated from the trunk, and correctly centered on the scanning field. Regions of interest (ROIs) were defined by the analytical program and included total body, trunk, head, pelvis, upper limbs, lower limbs, and android and gynoid regions. The lower boundary of the android region was defined at the pelvis cut line and the upper boundary above the pelvis cut line by 20% of the distance between the pelvis and chin. The upper boundary of the gynoid ROI was defined below the pelvis cut line by 1.5 times the height of the android space, and gynoid ROI height was equal to 2 times the android ROI height. For each region, DXA scanned weight of total mass, FM, and LBM. VAT was measured as the fat tissue located deep in the abdomen around the internal organs, as opposed to subcutaneous adipose tissue. Android LBM and FM, gynoid LBM and FM, and VAT were analyzed in a second step from the initial body composition images. For technical reasons, 87 examinations could not be reanalyzed, rendering analysis of the aforementioned parameters impossible in these participants. Outcomes Body composition Body composition outcomes were VAT; subtotal FM (calculated by extracting head FM from total FM); android and gynoid FM; fat mass index (FMI), calculated as the ratio of total body FM over height squared; subtotal, android, and gynoid LBM, by analogy to FM; lean mass index (LMI), defined as the ratio of total LBM over height squared; and sarcopenia indices (18): appendicular lean mass index (ALMI), calculated as the ratio of appendicular lean mass (ALM) over height squared, and ALM divided by BMI. Grip strength Assessment of muscle strength via handgrip was available for 990 participants. Participants of the CoLaus aged >50 were invited to participate in a substudy on frailty, which included grip strength, assessed with a Baseline® hydraulic hand dynamometer (Fabrication Enterprises, Inc., White Plains, NY). Positioning of the participants was done according to the American Society of Hand Therapists guidelines (19): subject seated, shoulders adducted and neutrally rotated, elbow flexed at 90°, forearm in neutral position, and wrist between 0° and 30° of dorsiflexion. Three measurements were performed consecutively at the dominant hand, and the highest value (expressed in kilograms) was used for the analysis. Potential confounders Energy intake Dietary intake was available for 988 participants. Dietary intake was assessed with the self-administered, semiquantitative Food Frequency Questionnaire (FFQ), which has been validated against 24-hour recalls among 626 volunteers from the Geneva population (20). Briefly, the FFQ assesses dietary intake for the previous 4 weeks and consists of 97 different food items that account for >90% of the intake of calories, proteins, fats, carbohydrates, alcohol, cholesterol, vitamin D, and retinol and 85% of fiber, carotene, and iron. Conversion of FFQ responses into nutrients was based on the French CIQUAL food composition table. Total energy intake was computed, including alcohol consumption. Sedentarity index Physical activity was estimated in 901 participants by a self-administered physical activity frequency questionnaire. The questionnaire lists 70 activities or groups of activities and was validated against measurements of energy expenditure by heart rate monitor with satisfactory correlations (r = 0.76) between the two methods (21). For this study, only sedentary status (yes/no) was used. Sedentary status was defined as spending <10% of total daily energy expenditure in activities with an intensity >4 basal metabolic rate equivalents. Hormonal assays Blood sampling was performed at the second CoLaus visit. Most biological assays were performed by the Lausanne University Hospital Clinical Laboratory on fresh blood samples within 2 hours of blood collection. Glucose was assessed by glucose dehydrogenase, with a maximum interassay and intra-assay coefficient of variation (CV) of 2.1% and 1.0%, respectively. Insulin was assessed by a solid-phase, two-site chemiluminescent immunometric assay (Diagnostic Products Corporation, Los Angeles, CA), with a maximum intra-assay CV of 13.7%. Homeostatic model assessment of insulin resistance (HOMA-IR) was calculated according to the formula (glucose × insulin)/22.5. Adiponectin and leptin levels were measured with a multiplexed particle-based flow cytometric cytokine assay with maximum intra-assay CVs of 8.4% and 9.5%, respectively (22). The analysis was conducted with a conventional flow cytometer (Guava EasyCyte Plus; Millipore, Zug, Switzerland). HOMA-IR and serum adipokine levels were available for 1046 and 977 participants, respectively. Psychiatric assessment Screening for current or past depression was performed with the Diagnostic Interview for Genetic Studies, as described previously (23). Depression was defined as the presence of depressive personality disorder or major depressive disorder (single or recurrent episode). Antidepressant treatment was considered as present for any reported medicine with an Anatomical Therapeutic Chemical code beginning with “N06A” (antidepressants) or “N06CA” (antidepressants in combination with psycholeptics) (https://www.whocc.no/atc_ddd_index/). Statistical analysis Statistical analyses were conducted in Stata version 14.1 (StataCorp, College Station, TX) for Windows. Because of their skewed distributions, leptin and adiponectin concentrations were log transformed before analysis. Descriptive results were expressed as the number of participants (percentage) or as average ± standard deviation. Bivariate analyses were conducted with χ2 for categorical variables and analysis of variance for continuous variables. Multivariable analyses for continuous variables were conducted with analysis of variance or multiple regression; results were expressed either as adjusted average ± standard error (SE) or as slope and 95% confidence interval. Post hoc pairwise comparisons were performed with the Scheffe method. Statistical significance was considered for a two-tailed test with a P value <0.05. Results Study population The flowchart of the study is shown in Fig. 1. After application of exclusion criteria (n = 26), the remaining 1053 women were classified in the three groups: 549 NUs (52.14%), 216 CUs (20.51%), and 288 PUs (27.35%). Android composition, gynoid composition, and VAT were available for 966/1053 participants (91.7%: 510 NUs, 255 PUs, and 201 CUs). Figure 1. View largeDownload slide Flowchart of the study highlighting the inclusion and exclusion criteria. Of the 1500 postmenopausal women of OsteoLaus with DXA scan, body composition assessment was retrievable for 1086 women who were included in the current analysis. Figure 1. View largeDownload slide Flowchart of the study highlighting the inclusion and exclusion criteria. Of the 1500 postmenopausal women of OsteoLaus with DXA scan, body composition assessment was retrievable for 1086 women who were included in the current analysis. Characteristics of participants Almost all participants were white (>98% for each group). The three groups differed significantly in age: 66.8 ± 6.3, 62.6 ± 6.7, and 61.3 ± 7.9 years for PUs, CUs, and NUs, respectively (CUs vs NUs, P = 0.04; PUs vs NUs, P < 0.001). Accordingly, all results were adjusted for age. In the unadjusted analysis, there was a trend for BMI differences with CUs < NUs < PUs: 24.9 ± 4.1, 25.7 ± 4.3, and 25.8.0 ± 4.3 kg/m2, respectively (CUs vs NUs, P = 0.052; CUs vs PUs, P = 0.049). Average MHT duration was 12.2 ± 8.8 years in CUs and 7.9 ± 6.3 years in PUs. The latter had an average of 8.5 ± 5.8 years since MHT withdrawal at study entry. Association between MHT and measures of body fat, muscle mass, and strength The age-adjusted values of body composition parameters according to MHT status are presented in Table 1. CUs exhibited significantly lower VAT values than NUs. Similarly, a consistently significant advantage of CUs over NUs was found for BMI, android FM, percentage of subtotal FM, and FMI (P < 0.05). PUs showed no advantage in comparison with NUs for all FM outcomes. We did not detect any statistical benefit for the MHT groups regarding LBM, sarcopenia indices, and handgrip strength. On the contrary, there was a trend for lower LMI in the CUs (CUs vs NUs, P = 0.05). The ratio ALM/BMI was the only parameter for which CUs clearly exceeded both PUs and NUs without reaching statistical significance. Table 1. Age-Adjusted Values of Body Composition Parameters According to MHT Status NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 Results are expressed as age-adjusted mean ± SE. Between-group comparisons performed with analysis of variance; post hoc pairwise comparisons performed with the Scheffe method. Boldface values correspond to statistical significant differences (P < 0.05) in between-group comparisons. View Large Table 1. Age-Adjusted Values of Body Composition Parameters According to MHT Status NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 Results are expressed as age-adjusted mean ± SE. Between-group comparisons performed with analysis of variance; post hoc pairwise comparisons performed with the Scheffe method. Boldface values correspond to statistical significant differences (P < 0.05) in between-group comparisons. View Large We also performed a regression analysis of different outcomes with age, stratified by MHT group (Table 2). The slopes for 10-year increments were significantly positive in NUs for BMI, subtotal FM, android FM, VAT, and FMI while being flat for both CUs and PUs. Between-group comparison confirmed a significant benefit for both MHT groups (P for interaction < 0.05) for all the aforementioned outcomes and percentage FM. The most prominent difference was seen for VAT (P = 0.01). The associations between BMI, subtotal FM, android FM, and VAT with age are represented in Fig. 2. There was no difference between groups for the slopes of LBM outcomes, with a tendency for loss of muscle mass in all three groups. When we selectively analyzed women aged <60 years, no statistical differences persisted between groups. Table 2. Regression Between the Body Composition Variables and Age at Study Inclusion (10-Year Increments), Stratified by MHT Status NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 Results are expressed as slope (95% confidence interval) for each 10-year increment. Significant (P < 0.05) slopes are indicated in bold. Statistical analysis by linear regression and interaction analysis by analysis of covariance. a P for interaction. View Large Table 2. Regression Between the Body Composition Variables and Age at Study Inclusion (10-Year Increments), Stratified by MHT Status NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 Results are expressed as slope (95% confidence interval) for each 10-year increment. Significant (P < 0.05) slopes are indicated in bold. Statistical analysis by linear regression and interaction analysis by analysis of covariance. a P for interaction. View Large Figure 2. View largeDownload slide Linear association between age at study inclusion and (A) BMI, (B) subtotal FM, (C) android FM, and (D) VAT, according to MHT group. Results are expressed as slope and 95% confidence interval for CUs (light gray), PUs (medium gray), and NUs (dark gray). Figure 2. View largeDownload slide Linear association between age at study inclusion and (A) BMI, (B) subtotal FM, (C) android FM, and (D) VAT, according to MHT group. Results are expressed as slope and 95% confidence interval for CUs (light gray), PUs (medium gray), and NUs (dark gray). Comparison of potential confounders between MHT groups In an attempt to explore potential confounders, age-adjusted results between MHT groups are shown in Table 3. No significant difference was detected for glucose, insulin, and adipokine levels. Insulin resistance tended to decrease in treatment groups: CUs < PUs < NUs. Adiponectin was higher in PUs and CUs, and leptin levels were lower in CUs (not significant for both parameters). Caloric intake differed between groups but in favor of NUs (NUs < CUs < PUs; NUs vs PUs, P = 0.039). There was no difference between groups in sedentary status, prevalence of depression, or use of antidepressant medications at study entry. Table 3. Age-Adjusted Values for Possible Confounders of Body Composition Parameters, Stratified by MHT Status NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 Results are expressed as age-adjusted mean ± SE or as percentages for sedentarity and depression prevalence. Between-group comparisons performed with analysis of variance. a The exact sample size differs according to the parameter analyzed (glucose, n = 1048; insulin, n = 1046; HOMA-IR, n = 1046; leptin, n = 977; adiponectin, n = 977; total caloric intake, n = 988; sedentarity index, n = 901; depression scale, n = 678). b Statistical analysis performed on log-transformed data. View Large Table 3. Age-Adjusted Values for Possible Confounders of Body Composition Parameters, Stratified by MHT Status NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 Results are expressed as age-adjusted mean ± SE or as percentages for sedentarity and depression prevalence. Between-group comparisons performed with analysis of variance. a The exact sample size differs according to the parameter analyzed (glucose, n = 1048; insulin, n = 1046; HOMA-IR, n = 1046; leptin, n = 977; adiponectin, n = 977; total caloric intake, n = 988; sedentarity index, n = 901; depression scale, n = 678). b Statistical analysis performed on log-transformed data. View Large Subgroup analysis according to MHT duration and time since MHT withdrawal Table 4 shows the main outcomes of CUs according to MHT duration and of PUs according to MHT duration and time since MHT withdrawal. Three subgroups were compared: 0 to 2, 2 to 5, and >5 years. There was no difference between subgroups for any of the outcomes studied. Similar results were noted when we repeated the analysis of PUs between two groups of time since MHT discontinuation: <5 years and >5 years. The effect of time since MHT withdrawal was further explored by a hinge analysis, which did not identify a reliable inflection point (data not shown). Table 4. Body Composition Parameters in MHT PUs According to Duration of and Time Since Discontinuation BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 Results are expressed as adjusted mean ± SE. Statistical analysis was performed with an analysis of variance model including age, BMI, duration of MHT, and time since discontinuation. View Large Table 4. Body Composition Parameters in MHT PUs According to Duration of and Time Since Discontinuation BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 Results are expressed as adjusted mean ± SE. Statistical analysis was performed with an analysis of variance model including age, BMI, duration of MHT, and time since discontinuation. View Large Discussion MHT is associated with lower visceral adiposity This cross-sectional analysis of the OsteoLaus cohort demonstrated that active MHT use is associated with significantly lower levels of VAT measured by DXA (Table 1, Supplemental Fig. 1). The significant increase of VAT with age in NUs was completely prevented in CUs, suggesting that MHT slows down the age-associated increase of VAT. These results are in agreement with a recent randomized study in premenopausal women who experienced an increase in VAT under GnRH-Ag (11), a phenotype reversed by estrogen therapy. Menopause is accompanied by changes in body composition (1, 2). Although menopause-associated bone loss is reversed by MHT (16), the evidence for its effect on FM is less consistent. Randomized controlled trials have yielded mixed results, with some showing a slight decrease in BMI and total FM with MHT (24, 25), whereas a subgroup analysis of the Women’s Health Initiative (WHI) trial (26) did not detect a significant advantage. Despite conflicting results about total FM, most studies detected a reduction in central fat with MHT, as indicated by reduced waist circumference (25), decrease in DXA-measured trunk to leg fat ratio (26), lower waist-to-hip ratio (27), reduced trunk FM measured by whole-body computed tomography (28), and reduced DXA-measured android fat (29). Several small studies have assessed the effect of MHT on VAT, as reviewed by Santen et al. (30). The majority showed reduced VAT, except for a randomized placebo-controlled study in nonobese, early postmenopausal women (31) that showed no benefit of MHT for intra-abdominal fat (assessed by computed tomography at L4 to L5 vertebral disk level). This result was potentially attributed to the continuous estrogen/progestin regimen used in this study and an accompanying decrease in insulin sensitivity, even though another prospective nonrandomized study implementing a continuous MHT regimen detected a benefit regarding android shift of fat distribution (27). Current MHT users have lower BMI, FMI, and android fat Our data also pointed out a slight but significant superiority of CUs regarding lower BMI, android fat, and FMI. Interestingly, all studies showing a significant decrease in total or central adiposity recruited early postmenopausal women (25, 26, 28), whereas differences were less pronounced in older populations, as in the WHI trial (average age >63 years). It is possible that the beneficial effect of MHT on FM is more pronounced in the early postmenopausal period and that age-mediated changes overcome the MHT benefits later in life. Of note, even in the studies showing significant benefits, the effect size was small. The only published meta-analysis (32) showed a significant reduction in waist circumference and abdominal fat (measured by dual energy photon or DXA) by 0.8% (5 trials) and 6.8% (4 trials), respectively. MHT prevents the age-associated gain of body fat The benefit of MHT was confirmed in the regression analysis, which highlighted a clear divergence between CUs and NUs regarding the association between age and body fat parameters. Indeed, NUs had significantly larger slopes for increase of BMI, subtotal and android FM, and FMI. MHT prevented significantly the age-associated increase of these parameters. This type of analysis offers the benefit of a projection over time, going beyond the limits of a simple cross-sectional analysis. Potential confounders do not seem to explain the MHT effect on FM It remains controversial whether the beneficial effect of MHT on FM is caused by a direct effect on adipocytes, mediated by other hormones, or by modifying intermediary factors such as nutrition or physical activity. In the current study, CUs tended to be less sedentary (61.4% vs 65.4% and 67.6% for NUs and PUs, respectively) without reaching statistical significance. Caloric intake was significantly higher in PUs than in NUs; CUs did not differ from the other two groups. Despite findings of positive correlations between E2 and leptin independently of body fat in one study of premenopausal women (33), adipokine levels did not differ significantly in our cohort after adjustment for age and subtotal FM (data not shown). Finally, no difference was found regarding the prevalence of depression between groups. Existing evidence on regulation of energy intake and expenditure by estrogens has been recently reviewed by Leeners et al. (34). Strong preclinical data support an important role for estrogen in bioenergetics. Both OVX mice and rats exhibited a marked reduction of spontaneous physical activity and a decrease in resting energy expenditure, whereas OVX rats developed an additional increase in energy intake (8). The latter was not seen in OVX mice, in line with our data in NUs. In menstruating women, resting energy expenditure is higher in the midluteal phase, when E2 is elevated; low in the early follicular phase, when E2 is lower; and further reduced by GnRH-Ag (35). An indirect effect via an increase in sedentarity was postulated by Lovejoy et al. (4), who prospectively followed physical activity annually by accelerometry in women going through menopause and detected a decrease of 50% over 4 years. The benefit of MHT on FM does not seem to persist after its withdrawal Another interesting point of our study is the clear absence of residual effect of MHT in PUs. PUs were classified according to MHT duration and time since MHT discontinuation; this analysis surprisingly showed no residual effect in early discontinuers, unlike our results regarding bone mineral density (16), suggesting a very rapid rebound effect after MHT withdrawal. However, the regression analysis detected significantly less steep slopes in PUs than in NUs for multiple FM outcomes, a result that deserves further exploration by a longitudinal study. To the best of our knowledge, no other study has specifically assessed body composition in PUs. Studies with GnRH-Ag (11, 36) have shown significant increases in total and central adiposity as soon as 4 months after estrogen withdrawal, consistent with our hypothesis of a rapid rebound effect. The rapid response of FM to external stimuli is also illustrated by the early increase in FM (+21.3%) only 8 weeks after training cessation in elite taekwondo athletes (37). The observed increase in caloric intake of PUs in our study provides another possible explanation for the rapid loss of FM benefits after MHT withdrawal. It would be reasonable to suggest confirmation of these results in the setting of a randomized trial to eliminate contribution of a selection bias. MHT does not have any detectable benefit on lean mass We hypothesized that MHT leads to increased LBM, which in turn would contribute to its favorable bone effects via increased mechanical load. Strongly positive correlations between LBM and bone mineral density, previously demonstrated (29, 38), support a potential link. Surprisingly, we did not detect any benefit among MHT users for LBM or muscle strength. These results were confirmed even after we excluded women using osteoporotic drugs other than MHT (n = 82, data not shown), thus arguing against an intermediate role of LBM in the MHT-mediated bone benefits. Our results add to the conflicting evidence of available studies with the only available meta-analysis (33) showing a slight but significant increase (+3.3%) of LBM in MHT users. One possible explanation might be the type of MHT. Certain progestogens, such as the norethisterone acetate used by Arabi et al. (29), have androgenic properties that could have an anabolic effect on LBM. More importantly, the effect of MHT on LBM can be selective for early postmenopausal women, weaning off rapidly under the stronger effect of age. In favor of this hypothesis, the WHI trial revealed that MHT significantly delayed loss of LBM after 3 years (28). Nevertheless, this relation was completely reversed between year 3 and 6 of the study, with a slight decrease in LBM in all groups at the end of year 6 (39), a finding also confirmed in the subset of women with high compliance. In our analysis, no LBM benefit was revealed when we analyzed only data from younger postmenopausal women (<60 years old). It is possible that this time-dependent effect is limited to a much shorter period after menopause (e.g., up to 5 years), as suggested by the studies discussed earlier (28, 39). Strengths and limitations This study has several limitations. The cross-sectional design is inevitably accompanied by a selection bias. Information on the beginning and the end of MHT was self-reported. This was also the case for the route of administration (oral, transdermal, vaginal), the type of MHT (estrogen-alone or estrogen/progestin), and the history of hysterectomy, preventing us from reliably assessing these factors. Furthermore, we were unable to verify participants’ adherence to MHT. Most participants were white, limiting the generalizability of study’s conclusions to other ethnicities. Our evaluation of confounding factors is partial. The physical activity assessment was only rough. We did not measure resting energy expenditure, which is a potential target of estrogen treatment. On the other hand, our study has considerable strengths to be taken into account. The large sample of the OsteoLaus cohort allows adequate statistical power. Body composition assessment was performed with DXA and last-generation software, which allowed reliable measurement of VAT, differentiating it from subcutaneous adipose tissue (40). This large, prospective study of postmenopausal women has explored the effect of MHT on VAT by reliably distinguishing it from other components of fat tissue. In conclusion, current MHT use prevents the increase in visceral adiposity. This finding may have important cardiovascular, metabolic, and bone implications that should be taken into account when assessing the benefit/risk ratio for MHT prescription. Nevertheless, the effect size on BMI and total FM is small, and MHT prescription cannot substitute for other interventions such as physical activity. Physicians should be aware that the benefit of MHT on body composition might rapidly disappear after its withdrawal and strongly encourage women to optimize nutrition and increase physical activity when stopping MHT. Future research via prospective and ideally randomized studies should assess differences depending on type of MHT and route of administration and on the evolution of body composition after MHT withdrawal. It would also be interesting to specifically investigate the effects of MHT on body composition in populations with an ethnically diverse composition and in early postmenopausal women. Abbreviations: Abbreviations: ALM appendicular lean mass ALMI appendicular lean mass index BMI body mass index CU current user CV coefficient of variation DXA dual-energy X-ray absorptiometry E2 estradiol ER estrogen receptor FFQ Food Frequency Questionnaire FM fat mass FMI fat mass index GnRH-Ag gonadotropin-releasing hormone agonist HOMA-IR homeostatic model assessment of insulin resistance LBM lean body mass LMI lean mass index MHT menopausal hormone therapy NU never user OVX ovariectomy PU past user ROI region of interest SAT subcutaneous adipose tissue SE standard error VAT visceral adipose tissue WHI Women’s Health Initiative Acknowledgments The authors thank Marie Almudena Metzger, principal research nurse of the OsteoLaus cohort, who contributed to the data collection for this study. Financial Support: The OsteoLaus study is supported by research grants from Lausanne University Hospital (Centre Hospitalier Universitaire Vaudois strategic plan funds) and the Swiss National Science Foundation (grants 32473B_156978). The CoLaus study is supported by research grants from GlaxoSmithKline, the Faculty of Biology and Medicine of Lausanne, and the Swiss National Science Foundation (Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, grants 33CSCO-122661, 33CS30-139468, and 33CS30-148401). These funding sources had no involvement in the study design, data collection, analysis and interpretation, writing of the report, or decision to submit the article for publication. Disclosure Summary: The authors have nothing to disclose. References 1. Greendale GA , Sowers M , Han W , Huang MH , Finkelstein JS , Crandall CJ , Lee JS , Karlamangla AS . Bone mineral density loss in relation to the final menstrual period in a multiethnic cohort: results from the Study of Women’s Health Across the Nation (SWAN) . J Bone Miner Res . 2012 ; 27 ( 1 ): 111 – 118 . Google Scholar CrossRef Search ADS PubMed 2. Wang Q , Hassager C , Ravn P , Wang S , Christiansen C . Total and regional body-composition changes in early postmenopausal women: age-related or menopause-related ? Am J Clin Nutr . 1994 ; 60 ( 6 ): 843 – 848 . Google Scholar CrossRef Search ADS PubMed 3. Wing RR , Matthews KA , Kuller LH , Meilahn EN , Plantinga PL . Weight gain at the time of menopause . Arch Intern Med . 1991 ; 151 ( 1 ): 97 – 102 . Google Scholar CrossRef Search ADS PubMed 4. Lovejoy JC , Champagne CM , de Jonge L , Xie H , Smith SR . Increased visceral fat and decreased energy expenditure during the menopausal transition . Int J Obes . 2008 ; 32 ( 6 ): 949 – 958 . Google Scholar CrossRef Search ADS 5. Trémollieres FA , Pouilles JM , Ribot CA . Relative influence of age and menopause on total and regional body composition changes in postmenopausal women . Am J Obstet Gynecol . 1996 ; 175 ( 6 ): 1594 – 1600 . Google Scholar CrossRef Search ADS PubMed 6. Abdulnour J , Doucet E , Brochu M , Lavoie JM , Strychar I , Rabasa-Lhoret R , Prud’homme D . The effect of the menopausal transition on body composition and cardiometabolic risk factors: a Montreal-Ottawa New Emerging Team group study . Menopause . 2012 ; 19 ( 7 ): 760 – 767 . Google Scholar CrossRef Search ADS PubMed 7. Lee CG , Carr MC , Murdoch SJ , Mitchell E , Woods NF , Wener MH , Chandler WL , Boyko EJ , Brunzell JD . Adipokines, inflammation, and visceral adiposity across the menopausal transition: a prospective study . J Clin Endocrinol Metab . 2009 ; 94 ( 4 ): 1104 – 1110 . Google Scholar CrossRef Search ADS PubMed 8. Van Pelt RE , Gavin KM , Kohrt WM . Regulation of body composition and bioenergetics by estrogens . Endocrinol Metab Clin North Am . 2015 ; 44 ( 3 ): 663 – 676 . Google Scholar CrossRef Search ADS PubMed 9. Wajchenberg BL . Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome . Endocr Rev . 2000 ; 21 ( 6 ): 697 – 738 . Google Scholar CrossRef Search ADS PubMed 10. Duval K , Prud’homme D , Rabasa-Lhoret R , Strychar I , Brochu M , Lavoie JM , Doucet E . Effects of the menopausal transition on energy expenditure: a MONET Group Study [published correction appears in Eur J Clin Nutr. 2014;68:142] . Eur J Clin Nutr . 2013 ; 67 ( 4 ): 407 – 411 . Google Scholar CrossRef Search ADS PubMed 11. Shea KL , Gavin KM , Melanson EL , Gibbons E , Stavros A , Wolfe P , Kittelson JM , Vondracek SF , Schwartz RS , Wierman ME , Kohrt WM . Body composition and bone mineral density after ovarian hormone suppression with or without estradiol treatment . Menopause . 2015 ; 22 ( 10 ): 1045 – 1052 . Google Scholar CrossRef Search ADS PubMed 12. Schmidt PJ , Ben Dor R , Martinez PE , Guerrieri GM , Harsh VL , Thompson K , Koziol DE , Nieman LK , Rubinow DR . Effects of estradiol withdrawal on mood in women with past perimenopausal depression: a randomized clinical trial . JAMA Psychiatry . 2015 ; 72 ( 7 ): 714 – 726 . Google Scholar CrossRef Search ADS PubMed 13. Nestor CC , Kelly MJ , Rønnekleiv OK . Cross-talk between reproduction and energy homeostasis: central impact of estrogens, leptin and kisspeptin signaling . Horm Mol Biol Clin Investig . 2014 ; 17 ( 3 ): 109 – 128 . Google Scholar PubMed 14. Lamy O , Krieg MA , Stoll D , Aubry-Rozier B , Metzger M , Hans D . The OsteoLaus Cohort Study . Osteologie . 2012 ; 21 ( 2 ): 77 – 82 . 15. Firmann M , Mayor V , Vidal PM , Bochud M , Pécoud A , Hayoz D , Paccaud F , Preisig M , Song KS , Yuan X , Danoff TM , Stirnadel HA , Waterworth D , Mooser V , Waeber G , Vollenweider P . The CoLaus study: a population-based study to investigate the epidemiology and genetic determinants of cardiovascular risk factors and metabolic syndrome . BMC Cardiovasc Disord . 2008 ; 8 ( 1 ): 6 . Google Scholar CrossRef Search ADS PubMed 16. Papadakis G , Hans D , Gonzalez-Rodriguez E , Vollenweider P , Waeber G , Marques-Vidal PM , Lamy O . The benefit of menopausal hormone therapy on bone density and microarchitecture persists after its withdrawal . J Clin Endocrinol Metab . 2016 ; 101 ( 12 ): 5004 – 5011 . Google Scholar CrossRef Search ADS PubMed 17. Petak S , Barbu CG , Yu EW , Fielding R , Mulligan K , Sabowitz B , Wu CH , Shepherd JA . The Official Positions of the International Society for Clinical Densitometry: body composition analysis reporting . J Clin Densitom . 2013 ; 16 ( 4 ): 508 – 519 . Google Scholar CrossRef Search ADS PubMed 18. Cawthon PM , Peters KW , Shardell MD , McLean RR , Dam TT , Kenny AM , Fragala MS , Harris TB , Kiel DP , Guralnik JM , Ferrucci L , Kritchevsky SB , Vassileva MT , Studenski SA , Alley DE . Cutpoints for low appendicular lean mass that identify older adults with clinically significant weakness . J Gerontol A Biol Sci Med Sci . 2014 ; 69 ( 5 ): 567 – 575 . Google Scholar CrossRef Search ADS PubMed 19. Roberts HC , Denison HJ , Martin HJ , Patel HP , Syddall H , Cooper C , Sayer AA . A review of the measurement of grip strength in clinical and epidemiological studies: towards a standardised approach . Age Ageing . 2011 ; 40 ( 4 ): 423 – 429 . Google Scholar CrossRef Search ADS PubMed 20. Bernstein M , Morabia A , Costanza MC , Landis JR , Ross A , Flandre P , Luong BL , Kumanyika S , Sorenson A , Localio R . Nutritional balance of the diet of the adult residents of Geneva [in French] . Soz Praventivmed . 1994 ; 39 ( 6 ): 333 – 344 . Google Scholar CrossRef Search ADS PubMed 21. Bernstein M , Sloutskis D , Kumanyika S , Sparti A , Schutz Y , Morabia A . Data-based approach for developing a physical activity frequency questionnaire . Am J Epidemiol . 1998 ; 147 ( 2 ): 147 – 154 . Google Scholar CrossRef Search ADS PubMed 22. Vignali DA . Multiplexed particle-based flow cytometric assays . J Immunol Methods . 2000 ; 243 ( 1-2 ): 243 – 255 . Google Scholar CrossRef Search ADS PubMed 23. Preisig M , Waeber G , Vollenweider P , Bovet P , Rothen S , Vandeleur C , Guex P , Middleton L , Waterworth D , Mooser V , Tozzi F , Muglia P . The PsyCoLaus study: methodology and characteristics of the sample of a population-based survey on psychiatric disorders and their association with genetic and cardiovascular risk factors . BMC Psychiatry . 2009 ; 9 ( 1 ): 9 . Google Scholar CrossRef Search ADS PubMed 24. Espeland MA , Stefanick ML , Kritz-Silverstein D , Fineberg SE , Waclawiw MA , James MK , Greendale GA ; Postmenopausal Estrogen-Progestin Interventions Study Investigators . Effect of postmenopausal hormone therapy on body weight and waist and hip girths . J Clin Endocrinol Metab . 1997 ; 82 ( 5 ): 1549 – 1556 . Google Scholar PubMed 25. Thorneycroft IH , Lindsay R , Pickar JH . Body composition during treatment with conjugated estrogens with and without medroxyprogesterone acetate: analysis of the women’s Health, Osteoporosis, Progestin, Estrogen (HOPE) trial . Am J Obstet Gynecol . 2007 ; 197 ( 2 ): 137.e1 – 137.e7 . Google Scholar CrossRef Search ADS 26. Chen Z , Bassford T , Green SB , Cauley JA , Jackson RD , LaCroix AZ , Leboff M , Stefanick ML , Margolis KL . Postmenopausal hormone therapy and body composition--a substudy of the estrogen plus progestin trial of the Women’s Health Initiative . Am J Clin Nutr . 2005 ; 82 ( 3 ): 651 – 656 . Google Scholar CrossRef Search ADS PubMed 27. Reubinoff BE , Wurtman J , Rojansky N , Adler D , Stein P , Schenker JG , Brzezinski A . Effects of hormone replacement therapy on weight, body composition, fat distribution, and food intake in early postmenopausal women: a prospective study . Fertil Steril . 1995 ; 64 ( 5 ): 963 – 968 . Google Scholar CrossRef Search ADS PubMed 28. Jensen LB , Vestergaard P , Hermann AP , Gram J , Eiken P , Abrahamsen B , Brot C , Kolthoff N , Sørensen OH , Beck-Nielsen H , Nielsen SP , Charles P , Mosekilde L . Hormone replacement therapy dissociates fat mass and bone mass, and tends to reduce weight gain in early postmenopausal women: a randomized controlled 5-year clinical trial of the Danish Osteoporosis Prevention Study . J Bone Miner Res . 2003 ; 18 ( 2 ): 333 – 342 . Google Scholar CrossRef Search ADS PubMed 29. Arabi A , Garnero P , Porcher R , Pelissier C , Benhamou CL , Roux C . Changes in body composition during post-menopausal hormone therapy: a 2 year prospective study . Hum Reprod . 2003 ; 18 ( 8 ): 1747 – 1752 . Google Scholar CrossRef Search ADS PubMed 30. Santen RJ , Allred DC , Ardoin SP , Archer DF , Boyd N , Braunstein GD , Burger HG , Colditz GA , Davis SR , Gambacciani M , Gower BA , Henderson VW , Jarjour WN , Karas RH , Kleerekoper M , Lobo RA , Manson JE , Marsden J , Martin KA , Martin L , Pinkerton JV , Rubinow DR , Teede H , Thiboutot DM , Utian WH ; Endocrine Society . Postmenopausal hormone therapy: an Endocrine Society scientific statement . J Clin Endocrinol Metab . 2010 ; 95 ( 7 suppl 1 ): s1 – s66 . Google Scholar CrossRef Search ADS PubMed 31. Sites CK , L’Hommedieu GD , Toth MJ , Brochu M , Cooper BC , Fairhurst PA . The effect of hormone replacement therapy on body composition, body fat distribution, and insulin sensitivity in menopausal women: a randomized, double-blind, placebo-controlled trial . J Clin Endocrinol Metab . 2005 ; 90 ( 5 ): 2701 – 2707 . Google Scholar CrossRef Search ADS PubMed 32. Salpeter SR , Walsh JME , Ormiston TM , Greyber E , Buckley NS , Salpeter EE . Meta-analysis: effect of hormone-replacement therapy on components of the metabolic syndrome in postmenopausal women . Diabetes Obes Metab . 2006 ; 8 ( 5 ): 538 – 554 . Google Scholar CrossRef Search ADS PubMed 33. Puder JJ , Monaco SE , Sen Gupta S , Wang J , Ferin M , Warren MP . Estrogen and exercise may be related to body fat distribution and leptin in young women . Fertil Steril . 2006 ; 86 ( 3 ): 694 – 699 . Google Scholar CrossRef Search ADS PubMed 34. Leeners B , Geary N , Tobler PN , Asarian L . Ovarian hormones and obesity . Hum Reprod Update . 2017 ; 23 ( 3 ): 300 – 321 . Google Scholar CrossRef Search ADS PubMed 35. Day DS , Gozansky WS , Van Pelt RE , Schwartz RS , Kohrt WM . Sex hormone suppression reduces resting energy expenditure and beta-adrenergic support of resting energy expenditure . J Clin Endocrinol Metab . 2005 ; 90 ( 6 ): 3312 – 3317 . Google Scholar CrossRef Search ADS PubMed 36. Yamasaki H , Douchi T , Yamamoto S , Oki T , Kuwahata R , Nagata Y . Body fat distribution and body composition during GnRH agonist therapy . Obstet Gynecol . 2001 ; 97 ( 3 ): 338 – 342 . Google Scholar PubMed 37. Liao Y-H , Sung Y-C , Chou C-C , Chen C-Y . Eight-week training cessation suppresses Physiological stress but rapidly impairs health metabolic profiles and aerobic capacity in elite taekwondo athletes . PLoS One . 2016 ; 11 ( 7 ): e0160167 . Google Scholar CrossRef Search ADS PubMed 38. He H , Liu Y , Tian Q , Papasian CJ , Hu T , Deng H-W . Relationship of sarcopenia and body composition with osteoporosis . Osteoporos Int . 2016 ; 27 ( 2 ): 473 – 482 . Google Scholar CrossRef Search ADS PubMed 39. Bea JW , Zhao Q , Cauley JA , LaCroix AZ , Bassford T , Lewis CE , Jackson RD , Tylavsky FA , Chen Z . Effect of hormone therapy on lean body mass, falls, and fractures: 6-year results from the Women’s Health Initiative hormone trials . Menopause . 2011 ; 18 ( 1 ): 44 – 52 . Google Scholar CrossRef Search ADS PubMed 40. Micklesfield LK , Goedecke JH , Punyanitya M , Wilson KE , Kelly TL . Dual-energy X-ray performs as well as clinical computed tomography for the measurement of visceral fat . Obesity (Silver Spring) . 2012 ; 20 ( 5 ): 1109 – 1114 . Google Scholar CrossRef Search ADS PubMed Copyright © 2018 Endocrine Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Clinical Endocrinology and Metabolism Oxford University Press

Menopausal Hormone Therapy Is Associated With Reduced Total and Visceral Adiposity: The OsteoLaus Cohort

Loading next page...
 
/lp/ou_press/menopausal-hormone-therapy-is-associated-with-reduced-total-and-ffpYmMI0tj
Publisher
Oxford University Press
Copyright
Copyright © 2018 Endocrine Society
ISSN
0021-972X
eISSN
1945-7197
D.O.I.
10.1210/jc.2017-02449
Publisher site
See Article on Publisher Site

Abstract

Abstract Context After menopause, fat mass (FM) and visceral adipose tissue (VAT) increase and nonbone lean body mass (LBM) decreases. Whether menopausal hormone therapy (MHT) reverses these changes remains controversial. Objective To assess the effect of MHT on FM, VAT, and LBM before and after its withdrawal and evaluate potential confounders. Design Cross-sectional study. Setting General community. Patients or Other Participants Women of the OsteoLaus cohort (50 to 80 years old) who underwent dual-energy X-ray absorptiometry (DXA) with body composition assessment. After we excluded women with estrogen-modifying medications, the 1053 participants were categorized into current users (CUs), past users (PUs), and never users (NUs) of MHT. Intervention None. Main Outcome Measures VAT measured by DXA was the primary outcome. We assessed subtotal and android FM, LBM, muscle strength (hand grip), and confounding factors (caloric intake, physical activity, biomarkers). Results The groups significantly differed in age, NU < CU < PU. Age-adjusted VAT was lower in CUs than NUs (P = 0.03). CUs exhibited lower age-adjusted body mass index (BMI) (−0.9 kg/m2) and a trend for lower FM (−1.3 kg). The 10-year gain of VAT (P < 0.01) and subtotal and android FM (P < 0.05) was prevented in CUs. No difference in LBM or hand grip was detected. No residual effect was detected for PUs, including for early MHT discontinuers. The confounding factors did not significantly differ between groups except for higher caloric intake in PUs compared with NUs. Conclusions MHT is associated with significantly decreased VAT, BMI, and android FM. No benefit is detected for LBM. The benefits are not preserved in PUs, suggesting caution when MHT is discontinued. Menopause is accompanied by changes in bone, fat, and muscular compartments (1, 2). In particular, menopause transition has been linked to increased propensity for weight gain and fat mass (FM) accumulation (3, 4). Whether this association is caused by declining ovarian hormone secretion or aging remains an open question (2). Data are more robust regarding the effect of menopause on regional fat. Several prospective studies have shown a greater increase of abdominal fat after menopause, leading to a shift from a gynoid to an android pattern of fat distribution (5–7). The causal association with estrogen deficiency is supported by preclinical data demonstrating that disruption of estradiol (E2) signaling by estrogen receptor (ER) deletion or ovariectomy (OVX) accelerates fat accumulation (8). It is important to emphasize that excess of central fat, and specifically of visceral adipose tissue (VAT) in humans, is associated with insulin resistance and high prevalence of metabolic syndrome, which are risk factors for atherosclerotic cardiovascular disease (9). A decline in nonbone lean body mass (LBM), also called fat-free or skeletal muscle mass, has been described across menopause (3, 4). It remains unclear whether this finding is a consequence of estrogen deficiency or of indirect factors such as a more sedentary lifestyle (10). Interventional trials assessing the effect of menopausal hormone therapy (MHT) on body composition have yielded mixed results regarding total FM and LBM (8). Those inconsistent findings can reflect differences in the population studied, study design (natural vs induced menopause), type of MHT, and method for assessing body composition. Conversely, most studies evaluating the effect of gonadotropin-releasing hormone agonists (GnRH-Ags), creating an artificial menopause state, have found increased total adiposity and intra-abdominal fat (8). Interestingly, the most recent one (11) showed that this phenotype could be prevented by estrogen administration. Another point that remains unclear is whether the eventual impact of MHT on FM is the result of a direct effect on adipocytes or indirect mechanisms such as altered energy intake or energy expenditure (8) or behavioral effects on mood and anxiety (12), which in turn might affect food intake and physical activity. In addition, insulin and adipokines (leptin, adiponectin) have been suggested as potential modifiers in the crosstalk between the reproductive axis and energy homeostasis both centrally and peripherally (7, 13). In this cross-sectional study, we assessed the effect of MHT on FM, VAT, and LBM before and after its withdrawal and attempted to explore potential confounders as detailed earlier. Materials and Methods Setting We analyzed data from the OsteoLaus study (14). OsteoLaus is a substudy of the CoLaus study, an ongoing prospective study aiming to assess the determinants of cardiovascular disease by using a population-based sample drawn from the city of Lausanne, Switzerland (15). The aims of the OsteoLaus study are to compare different models of fracture risk prediction and to assess the relationship between osteoporosis and cardiovascular diseases. Recruitment of OsteoLaus participants was detailed previously (16). CoLaus data (second visit) were collected within 6 months before the OsteoLaus visit. The study was approved by the Institutional Ethics Committee of the University of Lausanne. All participants signed an informed consent. Participants A total of 1500 postmenopausal women, aged 50 to 80 years, were questioned on current or past MHT use, its type, and duration, if applicable. All participants underwent a spine and hip dual-energy X-ray absorptiometry (DXA) scan on a Discovery DXA System (Hologic, Inc., Marlborough, MA). We included in this study all the women for whom body composition assessment was performed during the DXA scan (n = 1086). Exclusion criteria were intake of medication with estrogen-mediated effects (aromatase inhibitors, tamoxifen, antiandrogens), extreme body mass index (BMI) values (BMI >37 kg/m2), and uninterpretable or incomplete DXA scans (low-quality images). The remaining participants were divided into three groups: current users (CUs), past users (PUs), and never users (NUs) of MHT. CUs were taking MHT at trial entry or discontinued treatment <6 months earlier. PUs discontinued MHT ≥6 months before trial entry (otherwise considered as CUs). MHT use for <6 months, reported in 25 participants (<3 months in 23/25), was considered unlikely to cause considerable changes in body composition, and these subjects were classified as NUs. DXA measurements All body composition measurements were in accordance with published guidelines by the International Society for Clinical Densitometry (17). The subjects were placed in a supine position with palms down and arms at sides, slightly separated from the trunk, and correctly centered on the scanning field. Regions of interest (ROIs) were defined by the analytical program and included total body, trunk, head, pelvis, upper limbs, lower limbs, and android and gynoid regions. The lower boundary of the android region was defined at the pelvis cut line and the upper boundary above the pelvis cut line by 20% of the distance between the pelvis and chin. The upper boundary of the gynoid ROI was defined below the pelvis cut line by 1.5 times the height of the android space, and gynoid ROI height was equal to 2 times the android ROI height. For each region, DXA scanned weight of total mass, FM, and LBM. VAT was measured as the fat tissue located deep in the abdomen around the internal organs, as opposed to subcutaneous adipose tissue. Android LBM and FM, gynoid LBM and FM, and VAT were analyzed in a second step from the initial body composition images. For technical reasons, 87 examinations could not be reanalyzed, rendering analysis of the aforementioned parameters impossible in these participants. Outcomes Body composition Body composition outcomes were VAT; subtotal FM (calculated by extracting head FM from total FM); android and gynoid FM; fat mass index (FMI), calculated as the ratio of total body FM over height squared; subtotal, android, and gynoid LBM, by analogy to FM; lean mass index (LMI), defined as the ratio of total LBM over height squared; and sarcopenia indices (18): appendicular lean mass index (ALMI), calculated as the ratio of appendicular lean mass (ALM) over height squared, and ALM divided by BMI. Grip strength Assessment of muscle strength via handgrip was available for 990 participants. Participants of the CoLaus aged >50 were invited to participate in a substudy on frailty, which included grip strength, assessed with a Baseline® hydraulic hand dynamometer (Fabrication Enterprises, Inc., White Plains, NY). Positioning of the participants was done according to the American Society of Hand Therapists guidelines (19): subject seated, shoulders adducted and neutrally rotated, elbow flexed at 90°, forearm in neutral position, and wrist between 0° and 30° of dorsiflexion. Three measurements were performed consecutively at the dominant hand, and the highest value (expressed in kilograms) was used for the analysis. Potential confounders Energy intake Dietary intake was available for 988 participants. Dietary intake was assessed with the self-administered, semiquantitative Food Frequency Questionnaire (FFQ), which has been validated against 24-hour recalls among 626 volunteers from the Geneva population (20). Briefly, the FFQ assesses dietary intake for the previous 4 weeks and consists of 97 different food items that account for >90% of the intake of calories, proteins, fats, carbohydrates, alcohol, cholesterol, vitamin D, and retinol and 85% of fiber, carotene, and iron. Conversion of FFQ responses into nutrients was based on the French CIQUAL food composition table. Total energy intake was computed, including alcohol consumption. Sedentarity index Physical activity was estimated in 901 participants by a self-administered physical activity frequency questionnaire. The questionnaire lists 70 activities or groups of activities and was validated against measurements of energy expenditure by heart rate monitor with satisfactory correlations (r = 0.76) between the two methods (21). For this study, only sedentary status (yes/no) was used. Sedentary status was defined as spending <10% of total daily energy expenditure in activities with an intensity >4 basal metabolic rate equivalents. Hormonal assays Blood sampling was performed at the second CoLaus visit. Most biological assays were performed by the Lausanne University Hospital Clinical Laboratory on fresh blood samples within 2 hours of blood collection. Glucose was assessed by glucose dehydrogenase, with a maximum interassay and intra-assay coefficient of variation (CV) of 2.1% and 1.0%, respectively. Insulin was assessed by a solid-phase, two-site chemiluminescent immunometric assay (Diagnostic Products Corporation, Los Angeles, CA), with a maximum intra-assay CV of 13.7%. Homeostatic model assessment of insulin resistance (HOMA-IR) was calculated according to the formula (glucose × insulin)/22.5. Adiponectin and leptin levels were measured with a multiplexed particle-based flow cytometric cytokine assay with maximum intra-assay CVs of 8.4% and 9.5%, respectively (22). The analysis was conducted with a conventional flow cytometer (Guava EasyCyte Plus; Millipore, Zug, Switzerland). HOMA-IR and serum adipokine levels were available for 1046 and 977 participants, respectively. Psychiatric assessment Screening for current or past depression was performed with the Diagnostic Interview for Genetic Studies, as described previously (23). Depression was defined as the presence of depressive personality disorder or major depressive disorder (single or recurrent episode). Antidepressant treatment was considered as present for any reported medicine with an Anatomical Therapeutic Chemical code beginning with “N06A” (antidepressants) or “N06CA” (antidepressants in combination with psycholeptics) (https://www.whocc.no/atc_ddd_index/). Statistical analysis Statistical analyses were conducted in Stata version 14.1 (StataCorp, College Station, TX) for Windows. Because of their skewed distributions, leptin and adiponectin concentrations were log transformed before analysis. Descriptive results were expressed as the number of participants (percentage) or as average ± standard deviation. Bivariate analyses were conducted with χ2 for categorical variables and analysis of variance for continuous variables. Multivariable analyses for continuous variables were conducted with analysis of variance or multiple regression; results were expressed either as adjusted average ± standard error (SE) or as slope and 95% confidence interval. Post hoc pairwise comparisons were performed with the Scheffe method. Statistical significance was considered for a two-tailed test with a P value <0.05. Results Study population The flowchart of the study is shown in Fig. 1. After application of exclusion criteria (n = 26), the remaining 1053 women were classified in the three groups: 549 NUs (52.14%), 216 CUs (20.51%), and 288 PUs (27.35%). Android composition, gynoid composition, and VAT were available for 966/1053 participants (91.7%: 510 NUs, 255 PUs, and 201 CUs). Figure 1. View largeDownload slide Flowchart of the study highlighting the inclusion and exclusion criteria. Of the 1500 postmenopausal women of OsteoLaus with DXA scan, body composition assessment was retrievable for 1086 women who were included in the current analysis. Figure 1. View largeDownload slide Flowchart of the study highlighting the inclusion and exclusion criteria. Of the 1500 postmenopausal women of OsteoLaus with DXA scan, body composition assessment was retrievable for 1086 women who were included in the current analysis. Characteristics of participants Almost all participants were white (>98% for each group). The three groups differed significantly in age: 66.8 ± 6.3, 62.6 ± 6.7, and 61.3 ± 7.9 years for PUs, CUs, and NUs, respectively (CUs vs NUs, P = 0.04; PUs vs NUs, P < 0.001). Accordingly, all results were adjusted for age. In the unadjusted analysis, there was a trend for BMI differences with CUs < NUs < PUs: 24.9 ± 4.1, 25.7 ± 4.3, and 25.8.0 ± 4.3 kg/m2, respectively (CUs vs NUs, P = 0.052; CUs vs PUs, P = 0.049). Average MHT duration was 12.2 ± 8.8 years in CUs and 7.9 ± 6.3 years in PUs. The latter had an average of 8.5 ± 5.8 years since MHT withdrawal at study entry. Association between MHT and measures of body fat, muscle mass, and strength The age-adjusted values of body composition parameters according to MHT status are presented in Table 1. CUs exhibited significantly lower VAT values than NUs. Similarly, a consistently significant advantage of CUs over NUs was found for BMI, android FM, percentage of subtotal FM, and FMI (P < 0.05). PUs showed no advantage in comparison with NUs for all FM outcomes. We did not detect any statistical benefit for the MHT groups regarding LBM, sarcopenia indices, and handgrip strength. On the contrary, there was a trend for lower LMI in the CUs (CUs vs NUs, P = 0.05). The ratio ALM/BMI was the only parameter for which CUs clearly exceeded both PUs and NUs without reaching statistical significance. Table 1. Age-Adjusted Values of Body Composition Parameters According to MHT Status NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 Results are expressed as age-adjusted mean ± SE. Between-group comparisons performed with analysis of variance; post hoc pairwise comparisons performed with the Scheffe method. Boldface values correspond to statistical significant differences (P < 0.05) in between-group comparisons. View Large Table 1. Age-Adjusted Values of Body Composition Parameters According to MHT Status NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 NUs PUs CUs Global P CUs vs NUs CUs vs PUs PUs vs NUs Sample size 549 288 216 BMI, kg/m2 25.8 ± 0.2 25.6 ± 0.3 24.9 ± 0.3 0.03 0.03 0.21 0.78 FM, kg  Subtotal 23.3 ± 0.3 23.3 ± 0.5 22.0 ± 0.5 0.05 0.06 0.14 0.99  Android 2.01 ± 0.04 2.00 ± 0.06 1.83 ± 0.06 0.02 0.03 0.12 0.97  Gynoid 4.64 ± 0.05 4.71 ± 0.08 4.48 ± 0.08 0.13 0.29 0.13 0.74  Visceral 0.48 ± 0.01 0.48 ± 0.02 0.42 ± 0.02 0.01 0.02 0.07 0.98 FM, % total body weight  Subtotal 35.9 ± 0.3 36.2 ± 0.4 34.6 ± 0.4 0.01 0.03 0.03 0.90 Lean mass, kg  Subtotal 40.2 ± 0.2 39.8 ± 0.3 40.1 ± 0.4 0.62 0.95 0.86 0.62  Android 3.20 ± 0.02 3.17 ± 0.03 3.12 ± 0.04 0.24 0.24 0.60 0.86  Gynoid 6.36 ± 0.04 6.34 ± 0.06 6.29 ± 0.06 0.63 0.63 0.85 0.95 FMI, kg/m2 10.1 ± 0.1 10.0 ± 0.2 9.4 ± 0.2 0.01 0.02 0.08 0.95 LMI, kg/m2 15.9 ± 0.1 15.7 ± 0.1 15.5 ± 0.1 0.04 0.05 0.64 0.37 ALMI, kg/m2 6.6 ± 0.04 6.5 ± 0.05 6.5 ± 0.06 0.08 0.12 0.85 0.35 ALM/BMI 6795 ± 47 6815 ± 68 6978 ± 74 0.10 0.11 0.27 0.97 Hand grip strength, kg 24.6 ± 0.2 23.9 ± 0.3 24.5 ± 0.4 0.19 0.97 0.43 0.20 Results are expressed as age-adjusted mean ± SE. Between-group comparisons performed with analysis of variance; post hoc pairwise comparisons performed with the Scheffe method. Boldface values correspond to statistical significant differences (P < 0.05) in between-group comparisons. View Large We also performed a regression analysis of different outcomes with age, stratified by MHT group (Table 2). The slopes for 10-year increments were significantly positive in NUs for BMI, subtotal FM, android FM, VAT, and FMI while being flat for both CUs and PUs. Between-group comparison confirmed a significant benefit for both MHT groups (P for interaction < 0.05) for all the aforementioned outcomes and percentage FM. The most prominent difference was seen for VAT (P = 0.01). The associations between BMI, subtotal FM, android FM, and VAT with age are represented in Fig. 2. There was no difference between groups for the slopes of LBM outcomes, with a tendency for loss of muscle mass in all three groups. When we selectively analyzed women aged <60 years, no statistical differences persisted between groups. Table 2. Regression Between the Body Composition Variables and Age at Study Inclusion (10-Year Increments), Stratified by MHT Status NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 Results are expressed as slope (95% confidence interval) for each 10-year increment. Significant (P < 0.05) slopes are indicated in bold. Statistical analysis by linear regression and interaction analysis by analysis of covariance. a P for interaction. View Large Table 2. Regression Between the Body Composition Variables and Age at Study Inclusion (10-Year Increments), Stratified by MHT Status NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 NUs PUs CUs Pa Sample size 549 288 216 BMI, kg/m2 0.97 (0.52 to 1.41) −0.15 (−0.94 to 0.63) 0.15 (−0.68 to 0.97) 0.025 FM, kg  Subtotal 1.78 (1.00 to 2.57) −0.21 (−1.55 to 1.13) 0.19 (−1.28 to 1.66) 0.018  Android 0.18 (0.08 to 0.27) 0.02 (−0.15 to 0.18) −0.08 (−0.25 to 0.09) 0.023  Gynoid 0.04 (−0.10 to 0.18) −0.15 (−0.37 to 0.08) −0.05 (−0.29 to 0.19) 0.375  Visceral 0.10 (0.07 to 0.12) 0.05 (−0.01 to 0.09) 0.02 (−0.03 to 0.07) 0.014 FM, % total body weight  Subtotal 2.13 (1.48 to 2.79) 0.75 (−0.36 to 1.85) 0.54 (−0.73 to 1.80) 0.022 Lean mass, kg  Subtotal −0.66 (−1.23 to −0.09) −1.55 (−2.44 to −0.65) −0.62 (−1.67 to 0.44) 0.258  Android 0.01 (−0.06 to 0.07) −0.06 (−0.16 to 0.04) −0.08 (−0.19 to 0.03) 0.322  Gynoid −0.17 (-0.27 to −0.06) −0.24 (-0.40 to −0.08) −0.20 (−0.38 to −0.02) 0.771 FMI, kg/m2 0.80 (0.47 to 1.12) 0.15 (−0.42 to 0.71) 0.09 (−0.50 to 0.69) 0.041 LMI, kg/m2 0.13 (−0.07 to 0.34) −0.24 (−0.55 to 0.08) −0.12 (−0.52 to 0.28) 0.143 ALMI, kg/m2 −0.15 (−0.11 to 0.08) −0.17 (−0.31 to −0.02) −0.14 (−0.32 to 0.03) 0.180 Results are expressed as slope (95% confidence interval) for each 10-year increment. Significant (P < 0.05) slopes are indicated in bold. Statistical analysis by linear regression and interaction analysis by analysis of covariance. a P for interaction. View Large Figure 2. View largeDownload slide Linear association between age at study inclusion and (A) BMI, (B) subtotal FM, (C) android FM, and (D) VAT, according to MHT group. Results are expressed as slope and 95% confidence interval for CUs (light gray), PUs (medium gray), and NUs (dark gray). Figure 2. View largeDownload slide Linear association between age at study inclusion and (A) BMI, (B) subtotal FM, (C) android FM, and (D) VAT, according to MHT group. Results are expressed as slope and 95% confidence interval for CUs (light gray), PUs (medium gray), and NUs (dark gray). Comparison of potential confounders between MHT groups In an attempt to explore potential confounders, age-adjusted results between MHT groups are shown in Table 3. No significant difference was detected for glucose, insulin, and adipokine levels. Insulin resistance tended to decrease in treatment groups: CUs < PUs < NUs. Adiponectin was higher in PUs and CUs, and leptin levels were lower in CUs (not significant for both parameters). Caloric intake differed between groups but in favor of NUs (NUs < CUs < PUs; NUs vs PUs, P = 0.039). There was no difference between groups in sedentary status, prevalence of depression, or use of antidepressant medications at study entry. Table 3. Age-Adjusted Values for Possible Confounders of Body Composition Parameters, Stratified by MHT Status NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 Results are expressed as age-adjusted mean ± SE or as percentages for sedentarity and depression prevalence. Between-group comparisons performed with analysis of variance. a The exact sample size differs according to the parameter analyzed (glucose, n = 1048; insulin, n = 1046; HOMA-IR, n = 1046; leptin, n = 977; adiponectin, n = 977; total caloric intake, n = 988; sedentarity index, n = 901; depression scale, n = 678). b Statistical analysis performed on log-transformed data. View Large Table 3. Age-Adjusted Values for Possible Confounders of Body Composition Parameters, Stratified by MHT Status NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 NUs PUs CUs Global P Sample sizea 549 288 216 Glucose, mmol/L 5.76 ± 0.04 5.65 ± 0.05 5.65 ± 0.06 0.18 Insulin, mU/L 7.67 ± 0.23 7.31 ± 0.32 7.06 ± 0.36 0.32 HOMA-IR 2.04 ± 0.08 1.94 ± 0.11 1.88 ± 0.13 0.53 Leptin, pg/mL 6782 ± 276 7414 ± 385 5965 ± 434 0.19b Adiponectin, ng/mL 6406 ± 234 6709 ± 327 6697 ± 369 0.24b Total caloric intake, kcal 1613 ± 31 1751 ± 43 1655 ± 48 0.04 Current smoking, yes, % 20.9 15.5 16.7 0.12 Sedentary (n = 471) (n = 241) (n = 189)  Yes, % 65.4 67.6 61.4 0.40  No, % 34.6 32.4 38.6 Depression prevalence (n = 363) (n = 168) (n = 147)  Yes, % 51.5 54.2 57.8 0.43 Antidepressant medications, yes, % 11.8 14.2 15.3 0.37 Results are expressed as age-adjusted mean ± SE or as percentages for sedentarity and depression prevalence. Between-group comparisons performed with analysis of variance. a The exact sample size differs according to the parameter analyzed (glucose, n = 1048; insulin, n = 1046; HOMA-IR, n = 1046; leptin, n = 977; adiponectin, n = 977; total caloric intake, n = 988; sedentarity index, n = 901; depression scale, n = 678). b Statistical analysis performed on log-transformed data. View Large Subgroup analysis according to MHT duration and time since MHT withdrawal Table 4 shows the main outcomes of CUs according to MHT duration and of PUs according to MHT duration and time since MHT withdrawal. Three subgroups were compared: 0 to 2, 2 to 5, and >5 years. There was no difference between subgroups for any of the outcomes studied. Similar results were noted when we repeated the analysis of PUs between two groups of time since MHT discontinuation: <5 years and >5 years. The effect of time since MHT withdrawal was further explored by a hinge analysis, which did not identify a reliable inflection point (data not shown). Table 4. Body Composition Parameters in MHT PUs According to Duration of and Time Since Discontinuation BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 Results are expressed as adjusted mean ± SE. Statistical analysis was performed with an analysis of variance model including age, BMI, duration of MHT, and time since discontinuation. View Large Table 4. Body Composition Parameters in MHT PUs According to Duration of and Time Since Discontinuation BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 BMI (kg/m2) Subtotal FM (kg) Subtotal FM (%) Android FM (kg) VAT (kg) FMI (kg/m2) CUs  Sample size 215 215 215 200 200 200  Duration of MHT, y   0–2 24.51 ± 0.97 20.34 ± 1.73 33.14 ± 1.49 1.76 ± 0.19 0.39 ± 0.06 9.12 ± 0.67   2–5 24.62 ± 0.69 20.74 ± 1.23 34.52 ± 1.06 1.81 ± 0.14 0.41 ± 0.04 9.43 ± 0.48   5+ 25.02 ± 0.36 22.5 ± 0.65 34.76 ± 0.56 1.84 ± 0.08 0.43 ± 0.02 9.42 ± 0.27  P 0.856 0.389 0.614 0.924 0.827 0.910 PUs  Sample size 274 274 274 242 242 242  Duration of MHT, y   0–2 26.71 ± 0.72 24.18 ± 1.22 36.38 ± 1.01 2.10 ± 0.14 0.54 ± 0.04 10.47 ± 0.51   2–5 25.39 ± 0.62 23.94 ± 1.05 36.70 ± 0.86 2.00 ± 0.13 0.49 ± 0.04 10.04 ± 0.47   5+ 25.67 ± 0.33 23.48 ± 0.57 36.76 ± 0.47 2.03 ± 0.07 0.50 ± 0.02 10.23 ± 0.25  P 0.334 0.850 0.946 0.878 0.588 0.816  Time since discontinuation, y   0–2 25.72 ± 0.82 24.17 ± 1.40 36.40 ± 1.15 2.14 ± 0.17 0.53 ± 0.05 10.32 ± 0.60   2–5 25.69 ± 0.63 23.54 ± 1.08 36.80 ± 0.89 2.03 ± 0.14 0.51 ± 0.04 10.21 ± 0.49   5+ 25.81 ± 0.32 23.63 ± 0.55 36.71 ± 0.45 2.02 ± 0.07 0.50 ± 0.02 10.22 ± 0.24  P 0.985 0.927 0.960 0.807 0.813 0.988 Results are expressed as adjusted mean ± SE. Statistical analysis was performed with an analysis of variance model including age, BMI, duration of MHT, and time since discontinuation. View Large Discussion MHT is associated with lower visceral adiposity This cross-sectional analysis of the OsteoLaus cohort demonstrated that active MHT use is associated with significantly lower levels of VAT measured by DXA (Table 1, Supplemental Fig. 1). The significant increase of VAT with age in NUs was completely prevented in CUs, suggesting that MHT slows down the age-associated increase of VAT. These results are in agreement with a recent randomized study in premenopausal women who experienced an increase in VAT under GnRH-Ag (11), a phenotype reversed by estrogen therapy. Menopause is accompanied by changes in body composition (1, 2). Although menopause-associated bone loss is reversed by MHT (16), the evidence for its effect on FM is less consistent. Randomized controlled trials have yielded mixed results, with some showing a slight decrease in BMI and total FM with MHT (24, 25), whereas a subgroup analysis of the Women’s Health Initiative (WHI) trial (26) did not detect a significant advantage. Despite conflicting results about total FM, most studies detected a reduction in central fat with MHT, as indicated by reduced waist circumference (25), decrease in DXA-measured trunk to leg fat ratio (26), lower waist-to-hip ratio (27), reduced trunk FM measured by whole-body computed tomography (28), and reduced DXA-measured android fat (29). Several small studies have assessed the effect of MHT on VAT, as reviewed by Santen et al. (30). The majority showed reduced VAT, except for a randomized placebo-controlled study in nonobese, early postmenopausal women (31) that showed no benefit of MHT for intra-abdominal fat (assessed by computed tomography at L4 to L5 vertebral disk level). This result was potentially attributed to the continuous estrogen/progestin regimen used in this study and an accompanying decrease in insulin sensitivity, even though another prospective nonrandomized study implementing a continuous MHT regimen detected a benefit regarding android shift of fat distribution (27). Current MHT users have lower BMI, FMI, and android fat Our data also pointed out a slight but significant superiority of CUs regarding lower BMI, android fat, and FMI. Interestingly, all studies showing a significant decrease in total or central adiposity recruited early postmenopausal women (25, 26, 28), whereas differences were less pronounced in older populations, as in the WHI trial (average age >63 years). It is possible that the beneficial effect of MHT on FM is more pronounced in the early postmenopausal period and that age-mediated changes overcome the MHT benefits later in life. Of note, even in the studies showing significant benefits, the effect size was small. The only published meta-analysis (32) showed a significant reduction in waist circumference and abdominal fat (measured by dual energy photon or DXA) by 0.8% (5 trials) and 6.8% (4 trials), respectively. MHT prevents the age-associated gain of body fat The benefit of MHT was confirmed in the regression analysis, which highlighted a clear divergence between CUs and NUs regarding the association between age and body fat parameters. Indeed, NUs had significantly larger slopes for increase of BMI, subtotal and android FM, and FMI. MHT prevented significantly the age-associated increase of these parameters. This type of analysis offers the benefit of a projection over time, going beyond the limits of a simple cross-sectional analysis. Potential confounders do not seem to explain the MHT effect on FM It remains controversial whether the beneficial effect of MHT on FM is caused by a direct effect on adipocytes, mediated by other hormones, or by modifying intermediary factors such as nutrition or physical activity. In the current study, CUs tended to be less sedentary (61.4% vs 65.4% and 67.6% for NUs and PUs, respectively) without reaching statistical significance. Caloric intake was significantly higher in PUs than in NUs; CUs did not differ from the other two groups. Despite findings of positive correlations between E2 and leptin independently of body fat in one study of premenopausal women (33), adipokine levels did not differ significantly in our cohort after adjustment for age and subtotal FM (data not shown). Finally, no difference was found regarding the prevalence of depression between groups. Existing evidence on regulation of energy intake and expenditure by estrogens has been recently reviewed by Leeners et al. (34). Strong preclinical data support an important role for estrogen in bioenergetics. Both OVX mice and rats exhibited a marked reduction of spontaneous physical activity and a decrease in resting energy expenditure, whereas OVX rats developed an additional increase in energy intake (8). The latter was not seen in OVX mice, in line with our data in NUs. In menstruating women, resting energy expenditure is higher in the midluteal phase, when E2 is elevated; low in the early follicular phase, when E2 is lower; and further reduced by GnRH-Ag (35). An indirect effect via an increase in sedentarity was postulated by Lovejoy et al. (4), who prospectively followed physical activity annually by accelerometry in women going through menopause and detected a decrease of 50% over 4 years. The benefit of MHT on FM does not seem to persist after its withdrawal Another interesting point of our study is the clear absence of residual effect of MHT in PUs. PUs were classified according to MHT duration and time since MHT discontinuation; this analysis surprisingly showed no residual effect in early discontinuers, unlike our results regarding bone mineral density (16), suggesting a very rapid rebound effect after MHT withdrawal. However, the regression analysis detected significantly less steep slopes in PUs than in NUs for multiple FM outcomes, a result that deserves further exploration by a longitudinal study. To the best of our knowledge, no other study has specifically assessed body composition in PUs. Studies with GnRH-Ag (11, 36) have shown significant increases in total and central adiposity as soon as 4 months after estrogen withdrawal, consistent with our hypothesis of a rapid rebound effect. The rapid response of FM to external stimuli is also illustrated by the early increase in FM (+21.3%) only 8 weeks after training cessation in elite taekwondo athletes (37). The observed increase in caloric intake of PUs in our study provides another possible explanation for the rapid loss of FM benefits after MHT withdrawal. It would be reasonable to suggest confirmation of these results in the setting of a randomized trial to eliminate contribution of a selection bias. MHT does not have any detectable benefit on lean mass We hypothesized that MHT leads to increased LBM, which in turn would contribute to its favorable bone effects via increased mechanical load. Strongly positive correlations between LBM and bone mineral density, previously demonstrated (29, 38), support a potential link. Surprisingly, we did not detect any benefit among MHT users for LBM or muscle strength. These results were confirmed even after we excluded women using osteoporotic drugs other than MHT (n = 82, data not shown), thus arguing against an intermediate role of LBM in the MHT-mediated bone benefits. Our results add to the conflicting evidence of available studies with the only available meta-analysis (33) showing a slight but significant increase (+3.3%) of LBM in MHT users. One possible explanation might be the type of MHT. Certain progestogens, such as the norethisterone acetate used by Arabi et al. (29), have androgenic properties that could have an anabolic effect on LBM. More importantly, the effect of MHT on LBM can be selective for early postmenopausal women, weaning off rapidly under the stronger effect of age. In favor of this hypothesis, the WHI trial revealed that MHT significantly delayed loss of LBM after 3 years (28). Nevertheless, this relation was completely reversed between year 3 and 6 of the study, with a slight decrease in LBM in all groups at the end of year 6 (39), a finding also confirmed in the subset of women with high compliance. In our analysis, no LBM benefit was revealed when we analyzed only data from younger postmenopausal women (<60 years old). It is possible that this time-dependent effect is limited to a much shorter period after menopause (e.g., up to 5 years), as suggested by the studies discussed earlier (28, 39). Strengths and limitations This study has several limitations. The cross-sectional design is inevitably accompanied by a selection bias. Information on the beginning and the end of MHT was self-reported. This was also the case for the route of administration (oral, transdermal, vaginal), the type of MHT (estrogen-alone or estrogen/progestin), and the history of hysterectomy, preventing us from reliably assessing these factors. Furthermore, we were unable to verify participants’ adherence to MHT. Most participants were white, limiting the generalizability of study’s conclusions to other ethnicities. Our evaluation of confounding factors is partial. The physical activity assessment was only rough. We did not measure resting energy expenditure, which is a potential target of estrogen treatment. On the other hand, our study has considerable strengths to be taken into account. The large sample of the OsteoLaus cohort allows adequate statistical power. Body composition assessment was performed with DXA and last-generation software, which allowed reliable measurement of VAT, differentiating it from subcutaneous adipose tissue (40). This large, prospective study of postmenopausal women has explored the effect of MHT on VAT by reliably distinguishing it from other components of fat tissue. In conclusion, current MHT use prevents the increase in visceral adiposity. This finding may have important cardiovascular, metabolic, and bone implications that should be taken into account when assessing the benefit/risk ratio for MHT prescription. Nevertheless, the effect size on BMI and total FM is small, and MHT prescription cannot substitute for other interventions such as physical activity. Physicians should be aware that the benefit of MHT on body composition might rapidly disappear after its withdrawal and strongly encourage women to optimize nutrition and increase physical activity when stopping MHT. Future research via prospective and ideally randomized studies should assess differences depending on type of MHT and route of administration and on the evolution of body composition after MHT withdrawal. It would also be interesting to specifically investigate the effects of MHT on body composition in populations with an ethnically diverse composition and in early postmenopausal women. Abbreviations: Abbreviations: ALM appendicular lean mass ALMI appendicular lean mass index BMI body mass index CU current user CV coefficient of variation DXA dual-energy X-ray absorptiometry E2 estradiol ER estrogen receptor FFQ Food Frequency Questionnaire FM fat mass FMI fat mass index GnRH-Ag gonadotropin-releasing hormone agonist HOMA-IR homeostatic model assessment of insulin resistance LBM lean body mass LMI lean mass index MHT menopausal hormone therapy NU never user OVX ovariectomy PU past user ROI region of interest SAT subcutaneous adipose tissue SE standard error VAT visceral adipose tissue WHI Women’s Health Initiative Acknowledgments The authors thank Marie Almudena Metzger, principal research nurse of the OsteoLaus cohort, who contributed to the data collection for this study. Financial Support: The OsteoLaus study is supported by research grants from Lausanne University Hospital (Centre Hospitalier Universitaire Vaudois strategic plan funds) and the Swiss National Science Foundation (grants 32473B_156978). The CoLaus study is supported by research grants from GlaxoSmithKline, the Faculty of Biology and Medicine of Lausanne, and the Swiss National Science Foundation (Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, grants 33CSCO-122661, 33CS30-139468, and 33CS30-148401). These funding sources had no involvement in the study design, data collection, analysis and interpretation, writing of the report, or decision to submit the article for publication. Disclosure Summary: The authors have nothing to disclose. References 1. Greendale GA , Sowers M , Han W , Huang MH , Finkelstein JS , Crandall CJ , Lee JS , Karlamangla AS . Bone mineral density loss in relation to the final menstrual period in a multiethnic cohort: results from the Study of Women’s Health Across the Nation (SWAN) . J Bone Miner Res . 2012 ; 27 ( 1 ): 111 – 118 . Google Scholar CrossRef Search ADS PubMed 2. Wang Q , Hassager C , Ravn P , Wang S , Christiansen C . Total and regional body-composition changes in early postmenopausal women: age-related or menopause-related ? Am J Clin Nutr . 1994 ; 60 ( 6 ): 843 – 848 . Google Scholar CrossRef Search ADS PubMed 3. Wing RR , Matthews KA , Kuller LH , Meilahn EN , Plantinga PL . Weight gain at the time of menopause . Arch Intern Med . 1991 ; 151 ( 1 ): 97 – 102 . Google Scholar CrossRef Search ADS PubMed 4. Lovejoy JC , Champagne CM , de Jonge L , Xie H , Smith SR . Increased visceral fat and decreased energy expenditure during the menopausal transition . Int J Obes . 2008 ; 32 ( 6 ): 949 – 958 . Google Scholar CrossRef Search ADS 5. Trémollieres FA , Pouilles JM , Ribot CA . Relative influence of age and menopause on total and regional body composition changes in postmenopausal women . Am J Obstet Gynecol . 1996 ; 175 ( 6 ): 1594 – 1600 . Google Scholar CrossRef Search ADS PubMed 6. Abdulnour J , Doucet E , Brochu M , Lavoie JM , Strychar I , Rabasa-Lhoret R , Prud’homme D . The effect of the menopausal transition on body composition and cardiometabolic risk factors: a Montreal-Ottawa New Emerging Team group study . Menopause . 2012 ; 19 ( 7 ): 760 – 767 . Google Scholar CrossRef Search ADS PubMed 7. Lee CG , Carr MC , Murdoch SJ , Mitchell E , Woods NF , Wener MH , Chandler WL , Boyko EJ , Brunzell JD . Adipokines, inflammation, and visceral adiposity across the menopausal transition: a prospective study . J Clin Endocrinol Metab . 2009 ; 94 ( 4 ): 1104 – 1110 . Google Scholar CrossRef Search ADS PubMed 8. Van Pelt RE , Gavin KM , Kohrt WM . Regulation of body composition and bioenergetics by estrogens . Endocrinol Metab Clin North Am . 2015 ; 44 ( 3 ): 663 – 676 . Google Scholar CrossRef Search ADS PubMed 9. Wajchenberg BL . Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome . Endocr Rev . 2000 ; 21 ( 6 ): 697 – 738 . Google Scholar CrossRef Search ADS PubMed 10. Duval K , Prud’homme D , Rabasa-Lhoret R , Strychar I , Brochu M , Lavoie JM , Doucet E . Effects of the menopausal transition on energy expenditure: a MONET Group Study [published correction appears in Eur J Clin Nutr. 2014;68:142] . Eur J Clin Nutr . 2013 ; 67 ( 4 ): 407 – 411 . Google Scholar CrossRef Search ADS PubMed 11. Shea KL , Gavin KM , Melanson EL , Gibbons E , Stavros A , Wolfe P , Kittelson JM , Vondracek SF , Schwartz RS , Wierman ME , Kohrt WM . Body composition and bone mineral density after ovarian hormone suppression with or without estradiol treatment . Menopause . 2015 ; 22 ( 10 ): 1045 – 1052 . Google Scholar CrossRef Search ADS PubMed 12. Schmidt PJ , Ben Dor R , Martinez PE , Guerrieri GM , Harsh VL , Thompson K , Koziol DE , Nieman LK , Rubinow DR . Effects of estradiol withdrawal on mood in women with past perimenopausal depression: a randomized clinical trial . JAMA Psychiatry . 2015 ; 72 ( 7 ): 714 – 726 . Google Scholar CrossRef Search ADS PubMed 13. Nestor CC , Kelly MJ , Rønnekleiv OK . Cross-talk between reproduction and energy homeostasis: central impact of estrogens, leptin and kisspeptin signaling . Horm Mol Biol Clin Investig . 2014 ; 17 ( 3 ): 109 – 128 . Google Scholar PubMed 14. Lamy O , Krieg MA , Stoll D , Aubry-Rozier B , Metzger M , Hans D . The OsteoLaus Cohort Study . Osteologie . 2012 ; 21 ( 2 ): 77 – 82 . 15. Firmann M , Mayor V , Vidal PM , Bochud M , Pécoud A , Hayoz D , Paccaud F , Preisig M , Song KS , Yuan X , Danoff TM , Stirnadel HA , Waterworth D , Mooser V , Waeber G , Vollenweider P . The CoLaus study: a population-based study to investigate the epidemiology and genetic determinants of cardiovascular risk factors and metabolic syndrome . BMC Cardiovasc Disord . 2008 ; 8 ( 1 ): 6 . Google Scholar CrossRef Search ADS PubMed 16. Papadakis G , Hans D , Gonzalez-Rodriguez E , Vollenweider P , Waeber G , Marques-Vidal PM , Lamy O . The benefit of menopausal hormone therapy on bone density and microarchitecture persists after its withdrawal . J Clin Endocrinol Metab . 2016 ; 101 ( 12 ): 5004 – 5011 . Google Scholar CrossRef Search ADS PubMed 17. Petak S , Barbu CG , Yu EW , Fielding R , Mulligan K , Sabowitz B , Wu CH , Shepherd JA . The Official Positions of the International Society for Clinical Densitometry: body composition analysis reporting . J Clin Densitom . 2013 ; 16 ( 4 ): 508 – 519 . Google Scholar CrossRef Search ADS PubMed 18. Cawthon PM , Peters KW , Shardell MD , McLean RR , Dam TT , Kenny AM , Fragala MS , Harris TB , Kiel DP , Guralnik JM , Ferrucci L , Kritchevsky SB , Vassileva MT , Studenski SA , Alley DE . Cutpoints for low appendicular lean mass that identify older adults with clinically significant weakness . J Gerontol A Biol Sci Med Sci . 2014 ; 69 ( 5 ): 567 – 575 . Google Scholar CrossRef Search ADS PubMed 19. Roberts HC , Denison HJ , Martin HJ , Patel HP , Syddall H , Cooper C , Sayer AA . A review of the measurement of grip strength in clinical and epidemiological studies: towards a standardised approach . Age Ageing . 2011 ; 40 ( 4 ): 423 – 429 . Google Scholar CrossRef Search ADS PubMed 20. Bernstein M , Morabia A , Costanza MC , Landis JR , Ross A , Flandre P , Luong BL , Kumanyika S , Sorenson A , Localio R . Nutritional balance of the diet of the adult residents of Geneva [in French] . Soz Praventivmed . 1994 ; 39 ( 6 ): 333 – 344 . Google Scholar CrossRef Search ADS PubMed 21. Bernstein M , Sloutskis D , Kumanyika S , Sparti A , Schutz Y , Morabia A . Data-based approach for developing a physical activity frequency questionnaire . Am J Epidemiol . 1998 ; 147 ( 2 ): 147 – 154 . Google Scholar CrossRef Search ADS PubMed 22. Vignali DA . Multiplexed particle-based flow cytometric assays . J Immunol Methods . 2000 ; 243 ( 1-2 ): 243 – 255 . Google Scholar CrossRef Search ADS PubMed 23. Preisig M , Waeber G , Vollenweider P , Bovet P , Rothen S , Vandeleur C , Guex P , Middleton L , Waterworth D , Mooser V , Tozzi F , Muglia P . The PsyCoLaus study: methodology and characteristics of the sample of a population-based survey on psychiatric disorders and their association with genetic and cardiovascular risk factors . BMC Psychiatry . 2009 ; 9 ( 1 ): 9 . Google Scholar CrossRef Search ADS PubMed 24. Espeland MA , Stefanick ML , Kritz-Silverstein D , Fineberg SE , Waclawiw MA , James MK , Greendale GA ; Postmenopausal Estrogen-Progestin Interventions Study Investigators . Effect of postmenopausal hormone therapy on body weight and waist and hip girths . J Clin Endocrinol Metab . 1997 ; 82 ( 5 ): 1549 – 1556 . Google Scholar PubMed 25. Thorneycroft IH , Lindsay R , Pickar JH . Body composition during treatment with conjugated estrogens with and without medroxyprogesterone acetate: analysis of the women’s Health, Osteoporosis, Progestin, Estrogen (HOPE) trial . Am J Obstet Gynecol . 2007 ; 197 ( 2 ): 137.e1 – 137.e7 . Google Scholar CrossRef Search ADS 26. Chen Z , Bassford T , Green SB , Cauley JA , Jackson RD , LaCroix AZ , Leboff M , Stefanick ML , Margolis KL . Postmenopausal hormone therapy and body composition--a substudy of the estrogen plus progestin trial of the Women’s Health Initiative . Am J Clin Nutr . 2005 ; 82 ( 3 ): 651 – 656 . Google Scholar CrossRef Search ADS PubMed 27. Reubinoff BE , Wurtman J , Rojansky N , Adler D , Stein P , Schenker JG , Brzezinski A . Effects of hormone replacement therapy on weight, body composition, fat distribution, and food intake in early postmenopausal women: a prospective study . Fertil Steril . 1995 ; 64 ( 5 ): 963 – 968 . Google Scholar CrossRef Search ADS PubMed 28. Jensen LB , Vestergaard P , Hermann AP , Gram J , Eiken P , Abrahamsen B , Brot C , Kolthoff N , Sørensen OH , Beck-Nielsen H , Nielsen SP , Charles P , Mosekilde L . Hormone replacement therapy dissociates fat mass and bone mass, and tends to reduce weight gain in early postmenopausal women: a randomized controlled 5-year clinical trial of the Danish Osteoporosis Prevention Study . J Bone Miner Res . 2003 ; 18 ( 2 ): 333 – 342 . Google Scholar CrossRef Search ADS PubMed 29. Arabi A , Garnero P , Porcher R , Pelissier C , Benhamou CL , Roux C . Changes in body composition during post-menopausal hormone therapy: a 2 year prospective study . Hum Reprod . 2003 ; 18 ( 8 ): 1747 – 1752 . Google Scholar CrossRef Search ADS PubMed 30. Santen RJ , Allred DC , Ardoin SP , Archer DF , Boyd N , Braunstein GD , Burger HG , Colditz GA , Davis SR , Gambacciani M , Gower BA , Henderson VW , Jarjour WN , Karas RH , Kleerekoper M , Lobo RA , Manson JE , Marsden J , Martin KA , Martin L , Pinkerton JV , Rubinow DR , Teede H , Thiboutot DM , Utian WH ; Endocrine Society . Postmenopausal hormone therapy: an Endocrine Society scientific statement . J Clin Endocrinol Metab . 2010 ; 95 ( 7 suppl 1 ): s1 – s66 . Google Scholar CrossRef Search ADS PubMed 31. Sites CK , L’Hommedieu GD , Toth MJ , Brochu M , Cooper BC , Fairhurst PA . The effect of hormone replacement therapy on body composition, body fat distribution, and insulin sensitivity in menopausal women: a randomized, double-blind, placebo-controlled trial . J Clin Endocrinol Metab . 2005 ; 90 ( 5 ): 2701 – 2707 . Google Scholar CrossRef Search ADS PubMed 32. Salpeter SR , Walsh JME , Ormiston TM , Greyber E , Buckley NS , Salpeter EE . Meta-analysis: effect of hormone-replacement therapy on components of the metabolic syndrome in postmenopausal women . Diabetes Obes Metab . 2006 ; 8 ( 5 ): 538 – 554 . Google Scholar CrossRef Search ADS PubMed 33. Puder JJ , Monaco SE , Sen Gupta S , Wang J , Ferin M , Warren MP . Estrogen and exercise may be related to body fat distribution and leptin in young women . Fertil Steril . 2006 ; 86 ( 3 ): 694 – 699 . Google Scholar CrossRef Search ADS PubMed 34. Leeners B , Geary N , Tobler PN , Asarian L . Ovarian hormones and obesity . Hum Reprod Update . 2017 ; 23 ( 3 ): 300 – 321 . Google Scholar CrossRef Search ADS PubMed 35. Day DS , Gozansky WS , Van Pelt RE , Schwartz RS , Kohrt WM . Sex hormone suppression reduces resting energy expenditure and beta-adrenergic support of resting energy expenditure . J Clin Endocrinol Metab . 2005 ; 90 ( 6 ): 3312 – 3317 . Google Scholar CrossRef Search ADS PubMed 36. Yamasaki H , Douchi T , Yamamoto S , Oki T , Kuwahata R , Nagata Y . Body fat distribution and body composition during GnRH agonist therapy . Obstet Gynecol . 2001 ; 97 ( 3 ): 338 – 342 . Google Scholar PubMed 37. Liao Y-H , Sung Y-C , Chou C-C , Chen C-Y . Eight-week training cessation suppresses Physiological stress but rapidly impairs health metabolic profiles and aerobic capacity in elite taekwondo athletes . PLoS One . 2016 ; 11 ( 7 ): e0160167 . Google Scholar CrossRef Search ADS PubMed 38. He H , Liu Y , Tian Q , Papasian CJ , Hu T , Deng H-W . Relationship of sarcopenia and body composition with osteoporosis . Osteoporos Int . 2016 ; 27 ( 2 ): 473 – 482 . Google Scholar CrossRef Search ADS PubMed 39. Bea JW , Zhao Q , Cauley JA , LaCroix AZ , Bassford T , Lewis CE , Jackson RD , Tylavsky FA , Chen Z . Effect of hormone therapy on lean body mass, falls, and fractures: 6-year results from the Women’s Health Initiative hormone trials . Menopause . 2011 ; 18 ( 1 ): 44 – 52 . Google Scholar CrossRef Search ADS PubMed 40. Micklesfield LK , Goedecke JH , Punyanitya M , Wilson KE , Kelly TL . Dual-energy X-ray performs as well as clinical computed tomography for the measurement of visceral fat . Obesity (Silver Spring) . 2012 ; 20 ( 5 ): 1109 – 1114 . Google Scholar CrossRef Search ADS PubMed Copyright © 2018 Endocrine Society

Journal

Journal of Clinical Endocrinology and MetabolismOxford University Press

Published: Mar 27, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off