0021-972X/07/$15.00/0 Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 92(3):1049 –1057 Copyright © 2007 by The Endocrine Society doi: 10.1210/jc.2006-2060
Effects of Testosterone Supplementation on Whole Body and Regional Fat Mass and Distribution in Human Immunodeficiency Virus-Infected Men with Abdominal Obesity Shalender Bhasin, Robert A. Parker, Fred Sattler, Richard Haubrich, Beverly Alston, Triin Umbleja, and Cecilia M. Shikuma, for the AIDS Clinical Trials Group Protocol A5079 Study Team* Boston University (S.B.), Boston, Massachusetts 02118; Harvard School of Public Health (R.A.P., T.U.), Boston, Massachusetts 02115; University of Southern California (F.S.), Los Angeles, California 90033; University of California (R.H.), San Diego, California 92093; Division of AIDS (B.A.), National Institute of Allergy and Infectious Diseases, Bethesda, Maryland 20892; and University of Hawaii (C.M.S.), Honolulu, Hawaii 96817 Background: Whole body and abdominal obesity are associated with increased risk of diabetes mellitus and heart disease. The effects of testosterone therapy on whole body and visceral fat mass in HIVinfected men with abdominal obesity are unknown. Objective: The objective of this study was to determine the effects of testosterone therapy on intraabdominal fat mass and whole body fat distribution in HIV-infected men with abdominal obesity. Methods: In this multicenter, randomized, placebo-controlled, double-blind trial, 88 HIV-positive men with abdominal obesity (waistto-hip ratio ⬎ 0.95 or mid-waist circumference ⬎ 100 cm) and total testosterone 125– 400 ng/dl, or bioavailable testosterone less than 115 ng/dl, or free testosterone less than 50 pg/ml on stable antiretroviral regimen, and HIV RNA less than 10,000 copies per milliliter were randomized to receive 10 g testosterone gel or placebo daily for 24 wk. Fat mass and distribution were determined by abdominal computerized tomography and dual energy x-ray absorptiometry during wk 0, 12, and 24. We used an intention-to-treat approach and nonparametric statistical methods. Results: Baseline characteristics were balanced between groups. In 75 subjects evaluated, median percent change from baseline to wk 24 in visceral fat did not differ significantly between groups (testosterone
T
HE ACCUMULATION OF visceral fat is strongly associated with increased risk for cardiovascular disease and type 2 diabetes mellitus (1– 8). Even among individuals with nearly normal weight, those with abdominal obesity have a higher risk of diabetes, atherogenic dyslipidemia, hypertension, coronary artery disease, and in women, breast cancer (6). Abdominal adiposity is an important component of the metabolic syndrome (8 –10), a condition associated First Published Online December 12, 2006 * See Acknowledgments for members of the AIDS Clinical Trials Group Protocol A5079 Study Team. Abbreviations: ACTG, AIDS Clinical Trials Group; CT, computerized tomography; DEXA, dual energy x-ray absorptiometry; HDL, highdensity lipoprotein; HOMA, homeostatic model assessment; LDL, lowdensity lipoprotein; PSA, prostate-specific antigen; QUICKI, quantitative insulin sensitivity check index; ULN, upper limit of normal. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.
0.3%, placebo 3.1%, P ⫽ 0.75). Total (testosterone ⫺1.5%, placebo 4.3%, P ⫽ 0.04) and sc (testosterone ⫺7.2%, placebo 8.1%, P ⬍ 0.001) abdominal fat mass decreased in testosterone-treated men, but increased in placebo group. Testosterone therapy was associated with significant decrease in whole body, trunk, and appendicular fat mass by dual energy x-ray absorptiometry (all P ⬍ 0.001), whereas whole body and trunk fat increased significantly in the placebo group. The percent of individuals reporting a decrease in abdomen (P ⫽ 0.01), neck (P ⫽ 0.08), and breast size (P ⫽ 0.01) at wk 24 was significantly greater in testosterone-treated than placebo-treated men. Testosterone-treated men had greater increase in lean body mass than placebo (testosterone 1.3%, placebo ⫺0.3, P ⫽ 0.02). Plasma insulin, fasting glucose, and total high-density lipoprotein and low-density lipoprotein cholesterol levels did not change significantly. Testosterone therapy was well tolerated. Conclusions: Testosterone therapy in HIV-positive men with abdominal obesity and low testosterone was associated with greater decrease in whole body, total, and sc abdominal fat mass and a greater increase in lean mass compared to placebo. However, changes in visceral fat mass were not significantly different between groups. Further studies are needed to determine testosterone effects on insulin sensitivity and cardiovascular risk. (J Clin Endocrinol Metab 92: 1049 –1057, 2007)
with increased risk of atherosclerotic heart disease. The waist-to-hip ratio, a measure of abdominal adiposity, is a better predictor of coronary artery disease than other measures of obesity such as body mass index. Although population-based studies of the prevalence of abdominal obesity in HIV-infected men and women have been scarce, there is agreement that many HIV-infected subjects on antiretroviral therapy experience an increase in abdominal girth (11–15). Widely varying estimates of abdominal adiposity in HIV-infected individuals have been published, reflecting the lack of agreed upon case definitions, and differences in methods for assessing fat distribution (13–17). However, there is agreement that fat redistribution syndromes are being observed with increasing frequency, and there is concern about the long-term cardiovascular impact of abdominal obesity (15, 17). Fat distribution syndromes stigmatize the subject and affect an individual’s selfimage and adherence to antiretroviral therapy.
1049
1050
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
The mechanisms of visceral fat accumulation in HIV-infected subjects are not known, but are undoubtedly multifactorial. Androgen deficiency, increased glucocorticoid sensitivity and cytokine burden, and decreased activity of the retinoid X receptor-peroxisome proliferator-activated receptor ␥ regulatory complex and other unknown mechanisms collectively favor accumulation of visceral fat (12). This study focused only on correcting low testosterone levels in a subset of HIV-infected men with abdominal obesity. Testosterone is a major determinant of regional fat metabolism (18 –20) and body composition (19 –22). Testosterone inhibits uptake of triglycerides and enhances lipid mobilization from the visceral fat (23). Testosterone levels are associated inversely with visceral fat mass (24 –29). Testosterone therapy has been reported to decrease visceral fat, glucose, and insulin levels in middle-aged men with abdominal obesity (19, 20) and type 2 diabetes (30). Although no randomized trials of testosterone therapy in HIV-associated abdominal obesity have been conducted, there is considerable off-label use of testosterone in HIV-infected men. Therefore, we evaluated the efficacy of testosterone therapy in HIV-infected men with abdominal obesity and low or lownormal testosterone levels. We defined abdominal obesity as a waist-to-hip ratio greater than 0.95 or an abdominal circumference greater than 100 cm, because values above these thresholds in men have been associated with increased risk of cardiovascular disease and type 2 diabetes (6, 10). Subjects and Methods The study protocol, AIDS Clinical Trials Group (ACTG) 5079, was approved by institutional review boards for human subjects at all ACTG participating sites. All participants provided written informed consent.
Study design This was a multisite, placebo-controlled, randomized, parallel group trial in which eligible subjects were randomly assigned in a 1:1 ratio to receive either testosterone or placebo gel daily for a 24-wk, double-blind period. At the end of the double-blind period, subjects were given the option of continuing open-label testosterone therapy for additional 24 wk. We report efficacy results for the blinded phase, but include safety data from the entire study.
Subjects The eligible participants were HIV-infected men, 18 –70 yr of age with abdominal obesity, defined as waist-to-hip ratio greater than 0.95 or mid-waist circumference greater than 100 cm, and serum total testosterone between 125– 400 ng/dl. Because HIV-infected men have higher SHBG concentrations than healthy men (31), measurement of total testosterone alone may underestimate the degree of androgen deficiency. Therefore, we also included men who had free testosterone less than 50 pg/ml by equilibrium dialysis or bioavailable testosterone less than 115 ng/dl by ammonium sulfate precipitation method. The participants were required to have been on stable potent antiretroviral regimens for at least 3 months, planning to remain on this regimen and not to change diet and exercise behavior for at least 24 wk after randomization, and have plasma HIV RNA less than 10,000 copies per milliliter. HIV infection was documented by any licensed ELISA test kit and confirmed by a second method. Men with prostate or breast cancer, benign prostatic hypertrophy with American Urological Association symptom score greater than 7, prostate-specific antigen (PSA) equal to or greater than 4 ng/ml, diabetes mellitus, hemoglobin less than or equal to 91 g/liter or greater than the upper limit of normal (ULN), aspartate aminotransferase or alanine aminotransferase more than five times ULN, or creatinine two or more times ULN were excluded. We also excluded those
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
who had received any androgen, recombinant human GH, appetite stimulant, or glucocorticoids within 12 wk of study entry. Opportunistic infection within 12 wk, active malignancy, decompensated congestive heart failure, and untreated severe sleep apnea were additional causes for exclusion.
Randomization, stratification, and blinding Randomization was stratified by RNA copy number (detectable or undetectable, defined as HIV RNA copy number ⬍ 200 copies/ml by Roche Amplicor HIV-1 Monitor UltraSensitive assay) with approximate balance within each site. Investigators, subjects, and staff were unaware of treatment assignment during the double-blind phase.
Treatment During the double-blind phase, participants received either 10 g transdermal testosterone gel (AndroGel; Solvay Pharmaceuticals, Marietta, GA) or matching placebo gel. Testosterone gel is a clear, hydroalcoholic gel that contains 1% testosterone, ethanol, purified water, sodium hydroxide, carbomer 940, and isopropyl myristate that provides a continuous transdermal delivery of testosterone for 24 h. A daily application of 10 g of gel delivers 100 mg testosterone to the surface of the skin. Approximately 10% of applied testosterone is absorbed across the skin during the 24-h period, providing a nominal daily delivery of 10 mg testosterone. This dose was selected because in pivotal trials it raised testosterone levels in androgen-deficient men into the mid-normal range (32). Subjects were advised to apply the gel daily on the abdominal skin, wash their hands after application, and keep the area covered with clothing to minimize the risk of transfer.
Outcome measures At baseline, evidence for HIV-infection was documented, and medical evaluation and blood counts and chemistries were performed to confirm conformity with eligibility criteria. Single-slice abdominal computerized tomography (CT) and body composition analysis by dual energy x-ray absorptiometry (DEXA) were performed at baseline and during wk 12 and 24. Anthropometric and quality of life measurements were made at entry and during wk 12 and 24. Perceptions of weight change and body composition changes assessed by questionnaires, and the American Urological Association prostate symptom score, PSA, and digital prostate examinations were performed at baseline and during wk 12 and 24. Blood counts and chemistries were obtained every 6 wk. Serum total and free testosterone levels, LH, estradiol, and SHBG levels were measured at baseline and during wk 24.
Statistical considerations Sample size. The sample size was based on the percent change in visceral fat cross-sectional area, the primary outcome during the blinded phase. A priori, we determined that a sample size of 86 subjects would provide 80% power to detect an 18.1% decrease in visceral fat area, with sd of 30.4 as previously reported in HIV-seronegative men with abdominal obesity (33), using a significance level of 0.10 (two-sided). This allowed for 20% of subjects not to be unable to be evaluated. An 18% decrease in visceral fat mass was deemed clinically important as in a previous study of recombinant human GH; this change was associated with the perceptions of the patient of change in abdominal girth (33). Data analysis. We used an intention-to-treat strategy with last value carried forward for the blinded phase. Thus, we used wk 12 data if wk 24 data were unavailable for a measurement. If baseline data were unavailable, or if there was no follow-up data available for either wk 12 and 24, the participant was considered unable to be evaluated. Missing data were treated as missing completely at random. For continuous variables, comparisons between groups used a Wilcoxon rank sum test because there was evidence that the overall data were not consistent with normality. For continuous variables, the significance of changes over time within a treatment group was tested using a Wilcoxon signed rank test. Fisher’s exact test was used for categorical data. For paired binary data (e.g. HIV viral loads above or below the lower limit of detection over time), exact McNemar’s test was used to
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
assess changes over time. The Mantel-Haenszel test for trend was used to compare ordered categorical data between groups for participant self-perceptions. The log rank test was used to compare time to dropout and time to drug discontinuation between groups. Spearman rank correlation was used to assess the relationship between continuous variables. As specified in the protocol, results are considered statistically significant if P ⬍ 0.10 (two-sided).
Hormone measurements Hormone levels were measured by Quest Diagnostics-Nichols Institute laboratory (San Juan Capistrano, CA). Screening testosterone level was measured by an automated chemiluminescent assay; at the completion of the study, total testosterone levels in all samples were assayed, using liquid chromatography, tandem mass spectrometry. Free testosterone level was measured by an equilibrium dialysis method. Intraassay and interassay CVs were as follows: total testosterone 7.1 and 9.8%; free testosterone 11.8 and 11.6%. Serum estradiol was measured by the ADVIA Centaur immunoassay system (Bayer HealthCare, Tarrytown, NY), a competitive chemiluminescent immunoassay in which estradiol in the sample competes with acridinium ester-labeled estradiol for rabbit anti-estradiol antibody. LH was measured by a two-site chemiluminescent immunoassay (ADVIA Centaur system; Bayer HealthCare). SHBG was assayed using the IMMULITE 2000 immunoassay (Diagnostic Products Corp., Los Angeles, CA), which uses a solid-phase, two-site chemiluminescent immunometric assay.
FIG. 1. Flow of subjects through different phases of the study.
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
1051
Measurement of abdominal fat by CT scanning Transverse CT scans of abdomen (L4 –L5) were performed using a standardized protocol; this measure of abdominal fat correlates highly with multiple scans through the abdomen (34). All scans were read centrally at Tufts University using a standardized protocol using Slice-o-Matic software (Tomovision, Montreal, Canada). The display field was used to scale the image pixels for analysis and demarcate regions of interest.
Body composition analysis by DEXA Whole body and regional body composition were evaluated by DEXA scanning. The scanners were calibrated by using a soft tissue phantom. At each site, the same scanner was used for all evaluations on individual subjects. Regional body composition analysis was performed centrally at Tufts University using a standardized protocol. The extremities and trunk regions were demarcated manually, and appendicular lean and fat masses were calculated by adding the lean and fat mass of the arms and legs.
Results Flow of subjects
Eighty eight subjects were randomized between July 2001 and July 2004 at 19 sites, divided equally between treatment arms; 80 subjects completed the double-blind phase and 75 completed all phases of the study (Fig. 1). There were eight dropouts during the double-blind phase: five in the testoster-
1052
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
one group (one subject relocated, one incarcerated, one unable to come, one noncompliant with medications, and one refused contact) and three in the placebo group (one started using prescription testosterone, one started using prescription medication that was not allowed, and one withdrew consent). There were 12 premature drug discontinuations before wk 24: eight in the testosterone group (including one inadvertently starting open-label treatment) and four in the placebo group. Baseline characteristics
The two groups did not differ in terms of age, body mass index, visceral fat mass, testosterone levels, CD4⫹ T lymphocyte counts, or HIV copy number (Table 1). The median age was 47 yr; 72% of participants were Caucasian, and 11% previously used iv drugs. CT adipose tissue mass
By intention to treat analysis, neither median changes in visceral fat area from baseline to wk 24 in testosterone (0.3%) and placebo (3.1%) groups (both P ⬎ 0.80), nor the difference between groups was statistically significant (90% confidence
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
interval of difference between groups ⫺12.8%, 11.2%, P ⫽ 0.75) (Fig. 2). Per protocol analysis of subjects who completed both baseline and wk-24 CT scans revealed similar results (median change, 0.9 and 3.2% in testosterone and placebo groups, respectively; P ⫽ 0.83). Although changes in visceral fat area within groups were not statistically significant, total abdominal fat by CT decreased from baseline to wk 24 in testosterone-treated men (⫺1.5%) and increased in placebo-treated men (4.3%); this difference was significantly different between groups (P ⫽ 0.03). Subcutaneous abdominal fat decreased significantly from baseline to wk 24 in testosterone group (⫺7.2%, P ⬍ 0.001), but increased in placebo group (8.1%, P ⫽ 0.03); between-group difference for change in sc fat was significant (P ⬍ 0.001). Qualitatively similar results were found for betweengroup comparisons for changes from baseline to wk 12 in visceral, total abdominal, and abdominal sc fat (not shown). Body composition analyses by DEXA
Whole body fat mass decreased significantly from baseline in testosterone-treated group (⫺1.6 kg, ⫺7.9%), and increased
TABLE 1. Baseline characteristics of the study subjects by treatment assignment Characteristic
Age (yr) Race, no. (%) Black Caucasian Hispanic Other CD4⫹ T lymphocyte count (106/liter) HIV RNA (% detectable)b Total testosterone (ng/dl) Free testosterone (pg/ml) Body weight (kg) BMI (kg/m2) Abdominal circumference (cm) Waist-to-hip ratio CT variables Visceral fat (cm2) Subcutaneous fat (cm2) Total abdominal fat (cm2) DEXA variables Total body fat (kg) Trunk fat (kg) Extremity fat (kg) Total body lean (kg) Trunk lean (kg) Extremity lean (kg) Lipid profile Total cholesterol (mg/dl)c HDL cholesterol (mg/dl)c LDL cholesterol (mg/dl)c Triglycerides (mg/dl)c Non-HDL cholesterol (mg/dl)c Fasting glucose (mg/dl)d Insulin (IU/ml)c
Total (n ⫽ 88)
47 (40, 51)
Testosterone (n ⫽ 44)
48 (41, 51)
Placebo (n ⫽ 44)
43 (40, 52)
P value
0.37 0.13a
6 (7) 63 (72) 13 (15) 6 (6) 574 (438, 750) 4 (5) 405 (324, 502) 75 (58, 97) 87.1 (78.5, 94.9) 26.6 (24.3, 30.2) 99.1 (93.1, 105.1) 1.02 (0.99, 1.05)
3 (7) 27 (61) 10 (23) 4 (9) 544 (402, 729) 2 (5) 427 (309, 521) 76 (62, 99) 90.2 (79.2, 94.9) 27.7 (24.5, 30.8) 99.8 (94.7, 104.5) 1.02 (1.00, 1.05)
3 (7) 36 (82) 3 (7) 2 (4) 586 (460, 765) 2 (5) 395 (331, 475) 75 (58, 94) 83.9 (76.4, 94.5) 26.1 (24.3, 30.1) 98.3 (92.4, 105.3) 1.01 (0.98, 1.05)
0.52 1.00a 0.56 0.88 0.32 0.29 0.69 0.31
151 (110, 210) 159 (99, 238) 328 (239, 433)
153 (95, 191) 179 (88, 238) 314 (215, 446)
150 (116, 230) 148 (109, 232) 345 (254, 403)
0.33 0.96 0.61
18.9 (13.3, 23.8) 11.5 (8.7, 15.0) 5.7 (3.4, 8.3) 63.6 (56.9, 68.9) 32.3 (28.8, 35.4) 27.1 (24.8, 30.3) 207 (181, 241) 33 (27, 40) 125 (101, 146) 296 (196, 461) 177 (146, 205) 91 (83, 98) 12 (7, 18)
20.1 (13.3, 23.8) 11.5 (9.0, 16.3) 5.7 (3.4, 9.5) 63.2 (58.3, 69.2) 32.7 (28.9, 34.3) 27.6 (24.9, 31.1) 206 (178, 246) 32 (26, 39) 121 (100, 143) 333 (226, 491) 177 (142, 213) 93 (83, 100) 12 (7, 20)
17.9 (13.4, 23.2) 11.6 (8.6, 14.5) 4.8 (3.3, 7.7) 63.8 (56.8, 68.8) 32.2 (28.7, 35.6) 26.6 (24.7, 29.1) 211 (185, 235) 35 (29, 43) 125 (104, 157) 257 (165, 398) 171 (149, 203) 87 (84, 96) 11 (7, 17)
0.69 0.82 0.47 0.58 0.64 0.27 0.77 0.17 0.27 0.09 0.93 0.21 0.78
Data are median values with interquartile ranges (Q1, Q3) unless otherwise noted. P values are from Wilcoxon rank sum test unless otherwise noted. To convert total testosterone to SI units (nanomoles per liter), multiply testosterone concentrations in nanograms per deciliter by 0.0347. To convert free testosterone to SI units (picomoles per liter), multiply free testosterone concentration in picograms per milliliter by 3.47. BMI, Body mass index. a Fisher’s exact test. b HIV RNA detectable is defined as equal to or greater than 200 copies per milliliter. c n ⫽ 87/43/44. d n ⫽ 75/34/41.
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
1053
with an increase in lean body mass in the active treatment group (1.3 kg) and a decrease (⫺0.3 kg) in the placebo group. The changes from baseline to wk 24 in trunk (P ⫽ 0.02) and extremity (P ⫽ 0.01) lean mass were significantly greater in testosterone-treated men than in placebo-treated men. Anthropometrics and body weight
Waist circumference and waist-to-hip ratio decreased from baseline to wk 24 in testosterone-treated men (Fig. 3); the changes in waist circumference and waist-to-hip ratio were significantly greater in the testosterone group than placebo group (P ⫽ 0.03 for both). Body weight and body mass index decreased significantly from baseline to wk 24 in the testosterone group (P ⬍ 0.10 for both), but between-group differences in change in either measure were not significant (both P ⬎ 0.20). Perception of body composition change and overall health
In comparison to placebo, a significantly greater percent of individuals in the testosterone group reported a decrease in the abdomen (P ⫽ 0.01), neck (P ⫽ 0.08), and breast size (P ⫽ 0.01) at wk 24. The two groups did not differ significantly in overall health perceptions at wk 24. Hormone levels
FIG. 2. Changes in fat and lean mass and distribution, as assessed by CT scan and DEXA. Data are median percent changes from baseline. Values for active group (testosterone) are shown in darkly shaded boxes, and values for the placebo group are shown in lightly shaded boxes. P values for the comparison of change in placebo and testosterone groups are shown. Statistically significant within group changes are marked with asterisks above or below the bar. A, Changes in visceral adipose tissue (VAT) mass, abdominal sc adipose tissue (SAT) mass, and total abdominal adipose tissue mass, as assessed by CT scan. A single-slice CT scan was obtained at L4 –L5 level, and the mass of visceral, sc, and total abdominal fat was calculated by using Slice-o-Matic software (TomoVision) at the Central Reading Center. B, Changes in trunk fat, extremity fat, and total body fat, as assessed by DEXA. C, Changes in trunk, extremity, and whole body lean mass by DEXA.
in placebo-treated group (⫹0.7 kg, ⫹4.5%); the change from baseline to wk 24 was significantly different between the two groups (P ⬍ 0.001) (Fig. 2). Trunk fat measured by DEXA also decreased in testosterone-treated men (⫺1.2 kg, ⫺9.9%), and increased in placebo-treated men from baseline to wk 24 (0.3 kg, ⫹4.6%); the change in trunk fat was significantly different between the two groups (P ⬍ 0.001). Similarly, extremity fat decreased from baseline to wk 24 to a significantly greater extent in testosterone-treated men (⫺0.5 kg, ⫺10.1%) than in placebo-treated men (⫹0.2 kg, ⫹3.1%) (P ⬍ 0.001 for betweengroup difference). Lean body mass
There were significant differences in change from baseline to wk 24 in whole body lean mass between groups (P ⫽ 0.02),
As expected, the increments in nadir total and free testosterone levels from baseline to wk 24 were significantly greater in testosterone group than placebo (P ⫽ 0.07 for total testosterone, and 0.01 for free testosterone; Fig. 4). Serum LH levels decreased significantly in the testosterone group from baseline to wk 24 (P ⬍ 0.001); the change from baseline to wk 24 was greater in the testosterone group than the placebo group (P ⬍ 0.001). Serum SHBG levels did not change significantly in either group. Serum estradiol levels increased in the testosterone group, but the differences in the change from baseline were not significantly different between groups. Plasma lipids and insulin levels
The changes from baseline in plasma glucose and insulin levels, and homeostatic model assessment (HOMA) and quantitative insulin sensitivity check (QUICKI) indices of insulin sensitivity did not differ significantly between groups (Table 2). Plasma total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, non-HDL cholesterol, and triglyceride levels did not change significantly from baseline in either group (Table 2). Safety analysis
The increase in PSA from baseline to wk 24 was greater in the testosterone (0.1 ng/ml) than placebo group (0 ng/ml; P ⫽ 0.06). PSA level in two men in the placebo group and in one man in the testosterone group exceeded 4 ng/ml or had an increase of more than 1 ng/ml during the blinded phase. During the open-label phase, two additional men originally assigned to placebo group had PSA level above 4 ng/ml or had an increase of more than 1 ng/ml. Hematocrit increased significantly in both groups; the change in hematocrit from baseline to wk 24 was not sig-
1054
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
FIG. 3. Changes in waist circumference and waist-to-hip ratio. Data are median absolute changes from baseline. Values for active group (testosterone) are shown in darkly shaded boxes, and values for the placebo group are shown in lightly shaded boxes. P values for the comparison of change in placebo and testosterone groups are shown. Statistically significant within group changes are marked with asterisks below the bar.
nificantly different between groups. During the blinded phase, three testosterone-treated men developed hematocrit values above 54%. Hematocrit rose above 54% in three men during the open-label phase, one originally assigned to the testosterone group and two in the placebo group. There were no significant differences between groups for changes from baseline to wk 24 in CD4⫹ or CD8⫹ T lymphocyte counts, HIV RNA copy number, aspartate aminotransferase, alanine aminotransferase, or alkaline phosphatase. During the blinded phase, there was one grade 4 elevation in creatine phosphokinase in the testosterone group and one grade 4 elevation in fasting glucose in the placebo group. During the open-label phase, there was one grade 4 event (backache) in the placebo group; one person in the testosterone group developed grade 4 triglyceride level. Relationship of testosterone levels with fat measures
Although total testosterone levels were correlated significantly inversely with visceral fat area at baseline and at wk
24, there was no significant correlation between baseline testosterone levels and changes in visceral or sc fat mass (the greatest magnitude of correlation was ⫺0.20, P ⫽ 0.23). Also, there was no significant correlation between change in testosterone levels and changes in visceral or sc fat mass. Discussion
In comparison to placebo, testosterone therapy of HIVinfected men with abdominal obesity was not associated with a significantly greater reduction in visceral fat mass. However, total body, trunk, abdominal sc, and appendicular fat mass decreased to a greater extent in testosterone group than in the placebo group. These changes in abdominal fat mass in the testosterone-treated men were associated with greater reductions in waist circumference and waist-to-hip ratio and in perceptions of change in abdomen size than those observed in the placebo group. Testosterone administration was also associated with a greater accretion of lean body mass than placebo. The treatment was associated with
FIG. 4. Changes in circulating hormone levels during the 24-wk double-blind phase of study. Data are median absolute changes from baseline. Values for active group (testosterone) are shown in darkly shaded boxes, and values for the placebo group are shown in lightly shaded boxes. P values for the comparison of change in placebo and testosterone groups are shown. Statistically significant within group changes are marked with asterisks above or below the bar. To convert total testosterone to SI units (nanomoles per liter), multiply testosterone concentrations in nanograms per deciliter by 0.0347. To convert free testosterone to SI units (picomoles per liter), multiply free testosterone concentration in picograms per milliliter by 3.47. To convert estradiol to SI units (picomoles per liter), multiply values in picograms per milliliter by 3.671. To convert LH concentrations from milli-international units per milliliter to SI units (units per liter), multiply values in milli-international units per milliliter by 1.
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
1055
TABLE 2. Absolute changes in safety measures and lipid profile from baseline to wk 24 Testosterone
PSA (ng/ml) Hemoglobin (g/dl) Hematocrit (pct) Creatinine (mg/dl) CD4⫹ count (106/liter) HIV RNA (% detectable)a Total cholesterol (mg/dl) HDL cholesterol (mg/dl) LDL cholesterol (mg/dl) Triglycerides (mg/dl) Non-HDL cholesterol (mg/dl) Fasting glucose (mg/dl) Insulin (IU/ml) HOMA-IR indexc QUICKId
Placebo
n
Median
(IQR)
n
Median
(IQR)
35 40 40 32 38 22 39 39 39 39 39 26 39 25 25
0.1 0.3 1.6 0 25 1 ⫺7 1 0 ⫺5 ⫺9 ⫺0.5 0 ⫺0.1 0
(0, 0.2) (0, 1.1) (⫺0.1, 3.0) (⫺0.10, 0.05) (⫺52, 106) (5%) (⫺26, 15) (⫺3, 5) (⫺21, 8) (⫺148, 84) (⫺24, 16) (⫺7, 8) (⫺4, 4) (⫺0.8, 1.2) (⫺0.02, 0.03)
39 41 41 31 41 20 41 41 41 41 41 36 41 36 36
0 0.3 0.9 0 36 1 1 0 ⫺4 ⫺12 0 1.5 1 0.2 0
(⫺0.2, 0.1) (⫺0.5, 0.8) (⫺1.1, 2.7) (⫺0.10, 0.10) (⫺31, 71) (5%) (⫺20, 25) (⫺4, 3) (⫺16, 9) (⫺71, 85) (⫺22, 29) (⫺4, 8) (⫺3, 5) (⫺0.6, 1.0) (⫺0.03, 0.01)
P value
0.06 0.14 0.15 0.60 0.79 1.00b 0.35 0.37 0.75 0.76 0.29 0.74 0.45 0.64 0.31
P values are from Wilcoxon rank sum test unless otherwise noted. The number of patients with HIV RNA undetectable at baseline and HIV RNA detectable at wk 24. There were no changes from HIV RNA detectable to HIV RNA undetectable. b Fisher’s exact test. c HOMA-IR (HOMA of insulin resistance) was calculated as fasting plasma glucose (milligrams per deciliter) ⫻ fasting plasma insulin (micro-international units per milliliter) / 405. d QUICKI index was calculated as 1/关log (fasting plasma insulin [micro-international units per milliliter]) ⫹ log (fasting plasma glucose [milligrams per deciliter])兴. a
a low frequency of adverse events. Thus, testosterone therapy was more effective than placebo in inducing a reduction in whole body and abdominal fat, although it did not significantly reduce visceral fat mass. There was concordance among various measures of fat distribution and between objective and self-reported measures of fat distribution. Thus, whole body and trunk fat mass assessed by using DEXA decreased with testosterone administration. Similarly, CT assessment demonstrated a significant reduction in total abdominal and abdominal sc fat during testosterone administration. These changes in CT and DEXA measures of abdominal fat during testosterone administration were associated with decreases in waist circumference and waist-to-hip ratio. Importantly, changes in abdominal fat, assessed objectively, were accompanied by the perception of the subject of decreased abdomen size. Testosterone treatment was associated with low frequency of adverse effects, in line with that reported in testosterone trials in non-HIV-infected men (35–37) and HIV-infected men (38 – 41). The skin tolerability of the testosterone gel was excellent. The frequency of increases in hematocrit and PSA in testosterone-treated men were similar to those reported in testosterone trials in androgen-deficient men (35, 37, 42). Plasma lipids, including HDL cholesterol, CD4⫹ and CD8⫹ T lymphocyte counts, and HIV RNA copy number did not change significantly during testosterone administration. The study was not powered to determine the effects of testosterone administration on prostate or cardiovascular event rates. Furthermore, the short treatment duration does not permit meaningful inferences about long-term safety of testosterone administration. Our data differ from those of Marin et al. (19, 20) who had reported that testosterone supplementation of seronegative middle-aged men with truncal obesity decreased visceral fat volume and glucose concentration and improved insulin sensitivity. In epidemiological studies, serum testosterone
levels are correlated inversely with fat mass (25, 26, 43). Testosterone replacement of young and older hypogonadal men (36, 37, 44, 45) is associated with reduction in overall fat mass. Although there is considerable evidence that testosterone administration decreases whole body fat mass (37, 45), we do not know whether testosterone therapy can reduce visceral fat mass. In our study, whole body fat mass and trunk fat decreased during testosterone administration, but the changes in visceral fat mass were not significant. In a previous study in healthy, young men, graded doses of testosterone enanthate in GnRH-agonist-treated men were associated with dose-dependent reduction in abdominal sc and visceral fat (46). However, significant reductions in visceral fat were observed only at supraphysiological doses of testosterone enanthate (46). In this trial, the increments in total and free testosterone level with therapy were small, reflecting the nadir levels measured 24 h after gel administration. Smaller-than-expected increments in testosterone levels during treatment could also be the result of alterations in metabolic clearance of testosterone induced by anti-retroviral drugs and other chemotherapeutic agents used in HIV-infected men (47, 48). Some of our participants had baseline testosterone levels in the low-normal range and did not meet the definition of androgen deficiency proposed by the Endocrine Society Expert Panel (37). It is possible that testosterone, when administered in doses that increase serum testosterone levels to a higher level than those achieved in this study, might reduce visceral fat mass in HIV-infected men whose testosterone levels are clearly in the hypogonadal range. We do not know what magnitude of change in abdominal or whole body fat is clinically significant. Whole body fat mass decreased by about 8% in testosterone-treated men and increased by almost 5% in placebo-treated men. Similarly, the reduction in trunk fat was about 10% in testosteronetreated men. This loss of abdominal fat mass was perceived
1056
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
by subjects as a reduction in abdomen size, although we do not know whether this decrease in abdominal fat mass without a significant change in visceral fat mass reduces the risk of diabetes and cardiovascular disease. In epidemiological studies, trunk fat (49) and abdominal sc fat (50) have been associated with insulin resistance. A decrease in whole body fat mass has been associated with decreased risk of diabetes mellitus (51); however, we do not know whether a decrease in peripheral sc fat mass has a beneficial effect on insulin sensitivity, and the risk of diabetes mellitus and heart disease. In patients with lipoatrophy, further loss of peripheral sc fat may possibly be deleterious. This study was limited to men because testosterone administration in the doses that were used in this study would induce virilization in women. The effects of testosterone administration in women might be different from those in men (52, 53). Unlike men, serum testosterone levels are associated with insulin resistance and increased cardiovascular risk in women (43, 52–54). Therefore, these data should not be extrapolated to women. This initial study in HIV-infected men with visceral obesity was not designed to determine the effects of testosterone therapy on cardiovascular event rates; those trials would require a much larger sample size and a longer treatment duration than this study. We do not know whether increases in lean body mass and decreases in whole body fat mass would translate into improvements in insulin sensitivity, which was not assessed rigorously. Also, these data are not applicable to HIV-infected patients with lipoatrophy. Although testosterone therapy decreased whole body, abdominal, and appendicular fat mass, it did not significantly reduce visceral fat mass. Therefore, our data do not justify the widespread off-label use of testosterone in HIV-infected men with fat redistribution syndromes. Further studies are needed to determine the effects of testosterone therapy on insulin sensitivity and cardiovascular outcomes. Acknowledgments This study is a tribute to the tireless energy and vision of Robert Zackin, Sc.D., without whose leadership and single-minded commitment to ACTG, this study would not have been possible. We thank the staff and members of the ACTG TEAM 5079 for their astute stewardship of this study. We thank the investigators and their staff members of the participating ACTG sites for their participation in this study. Received September 20, 2006. Accepted December 6, 2006. Address all correspondence and requests for reprints to: Shalender Bhasin, M.D., Professor of Medicine, Boston University School of Medicine, Chief, Section of Endocrinology, Diabetes, and Nutrition, Boston Medical Center, 670 Albany Street, Second Floor, Boston, Massachusetts 02118. E-mail:
[email protected]. The study was supported in part by the ACTG funded by the National Institute of Allergy and Infectious Diseases (AI38858). Solvay Pharmaceuticals provided the placebo and testosterone gel. List of investigators at the participating ACTG sites and grant support: Linda Meixner and Gary Dyak, University of California, San Diego (A0701) Grant Nos. AI27670, AI064086, and AI 36214; Winston Cavert, M.D., and Kathy Fox, R.N., University of Minnesota (A1501) Grant No. AI27661; Debra OgataArakaki, R.N., and Scott Souza, Pharm.D., University of Hawaii (A5201) Grant No. AI34853; C. Bradley Hare, M.D., and Mary Payne, R.N., University of California, San Francisco/San Francisco General Hospital (A0801); Michael Dube, M.D., and Helen Rominger, N.P., Indiana University (A2601) AIDS Clinical Trials Unit Grant No. UO1 AI025859, General Clinical Research Center Grant No. MO1 RR000750; James
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
Bruce, R.N. (2701), and Alan Tenorio, M.D. (2702), Northwestern University (2701), Rush Medical Center (2702) Grant No. AI 25915; Jorge L. Santana, M.D., and Santiago Marrero, M.D., University of Puerto Rico (A5401) Grant No. 3UOI AI034832; Mussolini Africano, P.A.-C., and Luis M. Mendez, B.S., University of Southern California (A1201) Grant No. AI 27673; Deborah Novak and Karen L. Herbst, M.D., Charles Drew Medical Center (A0608) Grant No. U01 AI27660; University of Maryland, Institute of Human Virology (A4651); Karen Cavanagh, R.N., and Charles Gonzalez, M.D., New York University, New York, NY, HHC at Bellevue (A0401) ACTG Grant No. AI027665, General Clinical Research Center Grant No. M01-RR00096; Sharon Riddler, M.D., and Nancy Mantz, M.S.N., C.R.N.P., University of Pittsburgh (A1001); Pablo Tebas, M.D., and John Stoneman, R.N., B.S.N., Washington University in St. Louis (A2101) Grant No. A125903; Carl J. Fichtenbaum, M.D., and C. Franetta Hyc, S.C., R.N., B.S.N., University of Cincinnati (A2401) Grant No. AI-25897; M. Graham Ray and Steven C. Johnson, Colorado AIDS Clinical Trials Unit (A6101) Public Health Service Grant AI32770; Rob Roy Mac Gregor, M.D., and Keith Mickelberg, R.N., University of Pennsylvania (A6201) ACTG Grant No. U01-AI 032783-13, Centers for AIDS Research Grant No. 5-P30-AI-045008-07; and Karen Herbst, M.D., and Shalender Bhasin, M.D., Charles Drew University, Los Angeles, CA (1RO1DK49296 and 1RO1DK59627). Disclosure Statement: Dr. Bhasin has received research materials and grant support from Solvay Pharmaceuticals and Auxilium Pharmaceuticals. Dr. Sattler has received grant support from Solvay Pharmaceuticals. Other authors did not report any conflicts of interest.
References 1. Kissebah AH, Vydelingum N, Murray R, Evans DJ, Hartz AJ, Kalkhoff RK, Adams PW 1982 Relation of body fat distribution to metabolic complications of obesity. J Clin Endocrinol Metab 54:254 –260 2. Bjorntorp P 1990 Classification of obese patients and complications related to the distribution of surplus fat. Nutrition 6:131–137 3. Bjorntorp P 1990 “Portal” adipose tissue as a generator of risk factors for cardiovascular disease and diabetes. Arteriosclerosis 10:493– 496 4. Krotkiewski M, Bjorntorp P, Sjostrom L, Smith U 1983 Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution. J Clin Invest 72:1150 –1162 5. Larsson B, Bengtsson C, Bjorntorp P, Lapidus L, Sjostrom L, Svardsudd K, Tibblin G, Wedel H, Welin L, Wilhelmsen L 1992 Is abdominal body fat distribution a major explanation for the sex difference in the incidence of myocardial infarction? The study of men born in 1913 and the study of women, Goteborg, Sweden. Am J Epidemiol 135:266 –273 6. AACE/ACE Obesity Task Force 1998 AACE/ACE position statement on the prevention, diagnosis, and treatment of obesity. Endocr Pract 4:297–330 7. Vague J 1999 The degree of masculine differentiation of obesities: a factor determining predisposition to diabetes, atherosclerosis, gout, and uric calculous disease. 1956. Nutrition 15:89 –90; discussion 91 8. Reaven GM 1988 Banting lecture 1988. Role of insulin resistance in human disease. Diabetes 37:1595–1607 9. Eckel RH, Grundy SM, Zimmet PZ 2005 The metabolic syndrome. Lancet 365:1415–1428 10. Grundy SM 2005 Metabolic syndrome scientific statement by the American Heart Association and the National Heart, Lung, and Blood Institute. Arterioscler Thromb Vasc Biol 25:2243–2244 11. Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm DJ, Cooper DA 1998 A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 12:F51–F58 12. Carr A, Samaras K, Chisholm DJ, Cooper DA 1998 Pathogenesis of HIV-1protease inhibitor-associated peripheral lipodystrophy, hyperlipidaemia, and insulin resistance. Lancet 351:1881–1883 13. Miller KD, Jones E, Yanovski JA, Shankar R, Feuerstein I, Falloon J 1998 Visceral abdominal-fat accumulation associated with use of indinavir. Lancet 351:871– 875 14. Tien PC, Grunfeld C 2004 What is HIV-associated lipodystrophy? Defining fat distribution changes in HIV infection. Curr Opin Infect Dis 17:27–32 15. Miller J, Carr A, Emery S, Law M, Mallal S, Baker D, Smith D, Kaldor J, Cooper DA 2003 HIV lipodystrophy: prevalence, severity and correlates of risk in Australia. HIV Med 4:293–301 16. Safrin S, Grunfeld C 1999 Fat distribution and metabolic changes in patients with HIV infection. AIDS 13:2493–2505 17. Bacchetti P, Gripshover B, Grunfeld C, Heymsfield S, McCreath H, Osmond D, Saag M, Scherzer R, Shlipak M, Tien P 2005 Fat distribution in men with HIV infection. J Acquir Immune Defic Syndr 40:121–131 18. Xu XF, De Pergola G, Bjorntorp P 1991 Testosterone increases lipolysis and the number of -adrenoceptors in male rat adipocytes. Endocrinology 128: 379 –382
Bhasin et al. • Testosterone in HIV⫹ Men with Abdominal Obesity
19. Marin P, Holmang S, Jonsson L, Sjostrom L, Kvist H, Holm G, Lindstedt G, Bjorntorp P 1992 The effects of testosterone treatment on body composition and metabolism in middle-aged obese men. Int J Obes Relat Metab Disord 16:991–997 20. Marin P, Krotkiewski M, Bjorntorp P 1992 Androgen treatment of middleaged, obese men: effects on metabolism, muscle and adipose tissues. Eur J Med 1:329 –336 21. Bhasin S 2003 Effects of testosterone administration on fat distribution, insulin sensitivity, and atherosclerosis progression. Clin Infect Dis 37(Suppl 2):S142– S149 22. Bhasin S, Woodhouse L, Storer TW 2003 Androgen effects on body composition. Growth Horm IGF Res 13(Suppl):S63–S71. 23. Marin P, Lonn L, Andersson B, Oden B, Olbe L, Bengtsson BA, Bjorntorp P 1996 Assimilation of triglycerides in subcutaneous and intraabdominal adipose tissues in vivo in men: effects of testosterone. J Clin Endocrinol Metab 81:1018 –1022 24. Barrett-Connor EL 1995 Testosterone and risk factors for cardiovascular disease in men. Diabetes Metab 21:156 –161 25. Seidell JC, Bjorntorp P, Sjostrom L, Kvist H, Sannerstedt R 1990 Visceral fat accumulation in men is positively associated with insulin, glucose, and Cpeptide levels, but negatively with testosterone levels. Metabolism 39:897–901 26. Khaw KT, Barrett-Connor E 1992 Lower endogenous androgens predict central adiposity in men. Ann Epidemiol 2:675– 682 27. Endre T, Mattiasson I, Berglund G, Hulthen UL 1996 Low testosterone and insulin resistance in hypertension-prone men. J Hum Hypertens 10:755–761 28. Pitteloud N, Hardin M, Dwyer AA, Valassi E, Yialamas M, Elahi D, Hayes FJ 2005 Increasing insulin resistance is associated with a decrease in Leydig cell testosterone secretion in men. J Clin Endocrinol Metab 90:2636 –2641 29. Pitteloud N, Mootha VK, Dwyer AA, Hardin M, Lee H, Eriksson KF, Tripathy D, Yialamas M, Groop L, Elahi D, Hayes FJ 2005 Relationship between testosterone levels, insulin sensitivity, and mitochondrial function in men. Diabetes Care 28:1636 –1642 30. Kappor D, Goodwin E, Channer KS, Jones TH 2006 Testosterone replacement therapy improves insulin resistance, glycemic control, visceral adiposity, and hypercholesterolemia in hypogonadal men with type 2 diabetes. Eur J Endocrinol 154:899 –906 31. Arver S, Sinha-Hikim I, Beall G, Guerrero M, Shen R, Bhasin S 1999 Serum dihydrotestosterone and testosterone concentrations in human immunodeficiency virus-infected men with and without weight loss. J Androl 20:611– 618 32. Wang C, Berman N, Longstreth JA, Chuapoco B, Hull L, Steiner B, Faulkner S, Dudley RE, Swerdloff RS 2000 Pharmacokinetics of transdermal testosterone gel in hypogonadal men: application of gel at one site versus four sites: a General Clinical Research Center Study. J Clin Endocrinol Metab 85:964 –969 33. Johannsson G, Marin P, Lonn L, Ottosson M, Stenlof K, Bjorntorp P, Sjostrom L, Bengtsson BA 1997 Growth hormone treatment of abdominally obese men reduces abdominal fat mass, improves glucose and lipoprotein metabolism, and reduces diastolic blood pressure. J Clin Endocrinol Metab 82:727– 734 34. Smith SR, Zachwieja JJ 1999 Visceral adipose tissue: a critical review of intervention strategies. Int J Obes Relat Metab Disord 23:329 –335 35. Calof OM, Singh AB, Lee ML, Kenny AM, Urban RJ, Tenover JL, Bhasin S 2005 Adverse events associated with testosterone replacement in middle-aged and older men: a meta-analysis of randomized, placebo-controlled trials. J Gerontol A Biol Sci Med Sci 60:1451–1457 36. Snyder PJ, Peachey H, Berlin JA, Hannoush P, Haddad G, Dlewati A, Santanna J, Loh L, Lenrow DA, Holmes JH, Kapoor SC, Atkinson LE, Strom BL 2000 Effects of testosterone replacement in hypogonadal men. J Clin Endocrinol Metab 85:2670 –2677 37. Bhasin S, Cunningham GR, Hayes FJ, Matsumoto AM, Snyder PJ, Swerdloff RS, Montori VM 2006 Testosterone therapy in adult men with androgen
J Clin Endocrinol Metab, March 2007, 92(3):1049 –1057
38.
39.
40.
41. 42. 43. 44. 45.
46. 47.
48.
49. 50. 51.
52. 53. 54.
1057
deficiency syndromes: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 91:1995–2010 Bhasin S, Storer TW, Javanbakht M, Berman N, Yarasheski KE, Phillips J, Dike M, Sinha-Hikim I, Shen R, Hays RD, Beall G 2000 Testosterone replacement and resistance exercise in HIV-infected men with weight loss and low testosterone levels. JAMA 283:763–770 Grinspoon S, Corcoran C, Askari H, Schoenfeld D, Wolf L, Burrows B, Walsh M, Hayden D, Parlman K, Anderson E, Basgoz N, Klibanski A 1998 Effects of androgen administration in men with the AIDS wasting syndrome. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 129:18 –26 Grinspoon S, Corcoran C, Parlman K, Costello M, Rosenthal D, Anderson E, Stanley T, Schoenfeld D, Burrows B, Hayden D, Basgoz N, Klibanski A 2000 Effects of testosterone and progressive resistance training in eugonadal men with AIDS wasting. A randomized, controlled trial. Ann Intern Med 133:348 –355 Dobs A, Brown T 2002 Metabolic abnormalities in HIV disease and injection drug use. J Acquir Immune Defic Syndr 31(Suppl 2):S70 –S77 Calof O, Singh AB, Lee ML, Urban RJ, Kenny AM, Tenover JL, Bhasin S 2005 Adverse events associated with testosterone supplementation of older men. J Greontol Med Sci, in press Liu PY, Death AK, Handelsman DJ 2003 Androgens and cardiovascular disease. Endocr Rev 24:313–340 Katznelson L, Rosenthal DI, Rosol MS, Anderson EJ, Hayden DL, Schoenfeld DA, Klibanski A 1998 Using quantitative CT to assess adipose distribution in adult men with acquired hypogonadism. Am J Roentgenol 170:423– 427 Bhasin S, Calof OM, Storer TW, Singh AB, Jasuja R, Gao W, Dalton JT 2006 Ananolic applications of testosterone and selective androgen receptor modulators for physical dysfunction associated with aging and chronic illness. Nature 2:73– 80 Woodhouse LJ, Gupta N, Bhasin M, Singh AB, Ross R, Phillips J, Bhasin S 2004 Dose-dependent effects of testosterone on regional adipose tissue distribution in healthy young men. J Clin Endocrinol Metab 89:718 –726 Javanbakht M, Singh AB, Mazer NA, Beall G, Sinha-Hikim I, Shen R, Bhasin S 2000 Pharmacokinetics of a novel testosterone matrix transdermal system in healthy, premenopausal women and women infected with the human immunodeficiency virus. J Clin Endocrinol Metab 85:2395–2401 Choi HH, Gray PB, Storer TW, Calof OM, Woodhouse L, Singh AB, Padero C, Mac RP, Sinha-Hikim I, Shen R, Dzekov J, Dzekov C, Kushnir MM, Rockwood AL, Meikle AW, Lee ML, Hays RD, Bhasin S 2005 Effects of testosterone replacement in human immunodeficiency virus-infected women with weight loss. J Clin Endocrinol Metab 90:1531–1541 Van Pelt RE, Evans EM, Schechtman KB, Ehsani AA, Kohrt WM 2002 Contributions of total and regional fat mass to risk for cardiovascular disease in older women. Am J Physiol Endocrinol Metab 282:E1023–E1028 Goodpaster BH, Thaete FL, Simoneau JA, Kelley DE 1997 Subcutaneous abdominal fat and thigh muscle composition predict insulin sensitivity independently of visceral fat. Diabetes 46:1579 –1585 Ross R, Dagnone D, Jones PJ, Smith H, Paddags A, Hudson R, Janssen I 2000 Reduction in obesity and related comorbid conditions after diet-induced weight loss or exercise-induced weight loss in men. A randomized, controlled trial. Ann Intern Med 133:92–103 Phillips GB, Pinkernell BH, Jing T-Y 1997 Relationship between serum sex hormones and coronary artery disease in postmenopausal women. Arterioscler Thromb Vasc Biol 17:695–701 Phillips GB, Pinkernell BH, Jing TY 1994 The association of hypotestosteronemia with coronary artery disease in men. Arterioscler Thromb 14:701–706 Sutton-Tyrell K, Wildman RP, Mattews KA, Chae C, Lasley BL, Pasternak RC, Lloyd-Jones D, Sowers MF, Torrens JI, Investigators S 2005 Sex hormone binding globulin and free androgen index are related to cardiovascular risk factors in multiethnic perimenopausal and postmenopausal women enrolled in the Study of Women Across the Nation. Circulation 111:1242–1249
JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.
