It is known that type 2 diabetes is frequently associated with hypogonadism in male subjects (1,2). Free testosterone concentrations are negatively related with BMI in type 2 diabetic patients (1,3). However, the relationship of free testosterone with adipose tissue mass and lean body mass in diabetic patients is not well described. Studies examining this relationship have been limited by the fact that they have not measured free testosterone by a reliable method such as equilibrium dialysis (4).
Therefore, we decided to study the body composition of hypogonadal and eugonadal type 2 diabetic patients by using dual-energy X-ray absorptiometry (DEXA) to measure subcutaneous adipose tissue, lean body mass, bone mineral content (BMC), and bone mineral density (BMD).
RESEARCH DESIGN AND METHODS—
The study was conducted in two endocrinology practices in Midland, Texas, and Buffalo, New York. It is our practice to screen all diabetic patients for hypogonadism, due to the high prevalence of hypogonadism in diabetic patients. We also routinely evaluate body composition of our diabetic patients by DEXA (done at no cost to our patients). Therefore, informed consent was not obtained.
Data from 164 consecutive male diabetic patients who presented to the endocrinology clinic were prospectively collected for the study. Patients with known history of hypogonadism; panhypopituitarism; a chronic debilitating disease such as renal failure, cirrhosis, HIV, back or hip surgery; or treatment with steroids, bisphosphonates, or recombinant parathyroid hormone were excluded. A total of 26 patients were disqualified, based on the study criteria. Therefore, data on 138 patients were included for analysis in this study. Fasting blood samples were then obtained to measure serum total testosterone, free testosterone, sex hormone–binding globulin, and A1C, using assays previously described (1). Free testosterone was measured by equilibrium dialysis (Esoterix laboratories) (normal range 0.174–0.868 nmol/l).
We measured total body mass, lean mass, subcutaneous fat mass, BMC, and BMD by DEXA (Lunar machine; General Electric Medical Systems). Measurements were made in both arms and legs and the trunk region. BMD was measured at both arms and legs, ribs, L1–L4 spine, and both hips. Hip BMD was defined as the mean BMD of both hips. Data are presented as means ± SE.
RESULTS—
Data from 138 male type 2 diabetic subjects were analyzed. The mean age of the patients was 59.29 ± 0.97 years (range 29.8–84.3). The mean BMI was 31.83 ± 0.44 kg/m2 (18.4–44.6). The mean total testosterone, free testosterone, and sex hormone–binding globulin concentrations were 13.29 ± 0.49 nmol/l (5.17−35.97), 0.184 ± 0.007 nmol/l (0.073–0.465), and 56.72 ± 2.45 nmol/l (8.5–156.0), respectively. Free testosterone was inversely related to BMI (r = −0.19, P = 0.04) and to arm (r = −0.18, P = 0.05), leg (r = −0.24, P = 0.03), trunk (r = −0.20, P = 0.04), and total (r = −0.23, P = 0.02) subcutaneous fat mass. Total testosterone was also inversely related to arm, leg, trunk, and total (r = −0.38, −0.38, −0.40, and −0.41, respectively; P < 0.001) fat mass. Free testosterone and total testosterone were positively related to arm lean mass (r = 0.36, P < 0.001 and r = 0.19, P = 0.05) but not to leg or total lean mass. Free testosterone and total testosterone were positively related to arm BMC (r = 0.27 and 0.31, respectively; P < 0.01) but not to leg or total BMC. Free testosterone was positively related with BMD in arms (r = 0.20, P = 0.04) and ribs (r = 0.28, P < 0.01). Total testosterone was positively related to rib (r = 0.18, P = 0.05) but not arm (r = 0.11, P = 0.25) BMD. Free testosterone and total testosterone were not related to leg, hip, spine, or total BMD.
Total BMD was positively related to total lean mass (r = 0.50, P < 0.01), total fat mass (r = 0.34, P < 0.01), and BMI (r = 0.37, P < 0.01). In a multiple regression model with total BMD as the dependent variable, and BMI, total lean mass, and total fat mass as independent variables, only total lean mass was a significant predictor of total BMD. Body composition of hypogonadal (n = 66) with eugonadal (n = 72) patients is presented in Table 1.
CONCLUSIONS—
Our study demonstrates a strong inverse relation among total testosterone, free testosterone, and subcutaneous fat mass in type 2 diabetic patients. Because hypogonadal type 2 diabetic subjects have higher subcutaneous fat mass, they may be more insulin resistant than eugonadal type 2 diabetic subjects. Therefore, it is relevant that type 2 diabetic patients with hypogonadism have markedly elevated C-reactive protein concentrations; thus, this may pose a marked increase in cardiovascular risk (5). The relationship of obesity with testosterone is probably bidirectional (6,7), with hypogonadism possibly being both the cause and consequence of increased adiposity. From our study, it is not possible to determine whether increased subcutaneous fat mass is the cause or consequence of hypogonadism.
Free testosterone was related positively to arm lean mass but not to leg or total lean mass. It is known that diabetic subjects have higher muscle mass (due to higher BMI) but poorer muscle strength (8). We did not measure muscle strength in our study. Because testosterone has a positive effect on muscle strength, it is possible that hypogonadal diabetic patients have poorer muscle strength than eugonadal diabetic patients, despite having a similar lean body mass.
Free testosterone was positively related to BMD in arms and ribs but not at other sites. It is possible that hypogonadism may have affected BMD at various sites differently, causing a more pronounced or rapid bone loss (or slower bone replacement) at arms and ribs compared with legs, hips, or the spine. Hypogonadism has previously been known to be associated with low BMD at the hip and spine (10). However, studies have not specifically been done in obese subjects. From our study, it appears that hypogonadism in obese type 2 diabetic subjects is not a major risk factor for developing osteopenia in the hips or spine. We do not know if this would be true if a more elderly diabetic population or a larger number of patients was studied. It would be important to determine whether replacement therapy with testosterone in hypogonadal diabetic males will reverse adiposity, improve BMD in the arm and ribs, and improve cardiovascular outcomes.
Comparison of body composition of hypogonadal and eugonadal patients
. | Hypogonadal . | Eugonadal . | P . |
---|---|---|---|
n | 66 | 72 | |
Age (years)* | 63.1 ± 1.3 | 56.0 ± 1.3 | <0.001 |
BMI (kg/m2)* | 32.8 ± 0.7 | 30.9 ± 0.6 | 0.024 |
Total testosterone (nmol/l)* | 10.31 ± 0.45 | 16.0 ± 0.71 | <0.001 |
Free testosterone (nmol/l)* | 0.131 ± 0.003 | 0.236 ± 0.009 | <0.001 |
Sex hormone–binding globulin (nmol/l) | 61.2 ± 3.5 | 60.0 ± 3.5 | 0.83 |
A1C (%) | 7.99 ± 0.27 | 7.67 ± 0.25 | 0.62 |
Fat mass (kg) | |||
Arm | 3.33 ± 0.15 | 3.07 ± 0.13 | 0.19 |
Leg* | 9.89 ± 0.58 | 8.41 ± 0.47 | <0.05 |
Trunk* | 21.92 ± 0.90 | 19.65 ± 0.78 | <0.05 |
Total* | 36.27 ± 1.55 | 32.17 ± 1.32 | <0.05 |
Lean mass (kg) | |||
Arm* | 6.38 ± 0.14 | 7.02 ± 0.11 | <0.001 |
Leg | 19.13 ± 0.44 | 19.27 ± 0.39 | 0.8 |
Trunk | 29.31 ± 0.72 | 29.51 ± 0.55 | 0.82 |
Total | 59.64 ± 1.09 | 60.55 ± 0.93 | 0.53 |
BMC (kg) | |||
Arm* | 0.39 ± 0.02 | 0.43 ± 0.01 | 0.035 |
Leg | 1.27 ± 0.03 | 1.27 ± 0.03 | 0.99 |
Total | 3.19 ± 0.06 | 3.31 ± 0.06 | 0.19 |
BMD (g/cm2) | |||
Arm | 0.99 ± 0.02 | 1.02 ± 0.01 | 0.22 |
Rib | 0.78 ± 0.01 | 0.80 ± 0.01 | 0.17 |
Leg | 1.40 ± 0.02 | 1.43 ± 0.02 | 0.25 |
Total | 1.26 ± 0.01 | 1.29 ± 0.02 | 0.42 |
Spine | 1.24 ± 0.02 | 1.23 ± 0.04 | 0.19 |
Hip | 1.06 ± 0.02 | 1.05 ± 0.02 | 0.99 |
Osteopenia in hip | 27.7 | 30.0 | NS |
Osteoporosis in hip | 2.13 | 0 | NS |
Osteopenia in spine | 23.4 | 26.7 | NS |
Osteoporosis in spine | 0 | 0 | NS |
. | Hypogonadal . | Eugonadal . | P . |
---|---|---|---|
n | 66 | 72 | |
Age (years)* | 63.1 ± 1.3 | 56.0 ± 1.3 | <0.001 |
BMI (kg/m2)* | 32.8 ± 0.7 | 30.9 ± 0.6 | 0.024 |
Total testosterone (nmol/l)* | 10.31 ± 0.45 | 16.0 ± 0.71 | <0.001 |
Free testosterone (nmol/l)* | 0.131 ± 0.003 | 0.236 ± 0.009 | <0.001 |
Sex hormone–binding globulin (nmol/l) | 61.2 ± 3.5 | 60.0 ± 3.5 | 0.83 |
A1C (%) | 7.99 ± 0.27 | 7.67 ± 0.25 | 0.62 |
Fat mass (kg) | |||
Arm | 3.33 ± 0.15 | 3.07 ± 0.13 | 0.19 |
Leg* | 9.89 ± 0.58 | 8.41 ± 0.47 | <0.05 |
Trunk* | 21.92 ± 0.90 | 19.65 ± 0.78 | <0.05 |
Total* | 36.27 ± 1.55 | 32.17 ± 1.32 | <0.05 |
Lean mass (kg) | |||
Arm* | 6.38 ± 0.14 | 7.02 ± 0.11 | <0.001 |
Leg | 19.13 ± 0.44 | 19.27 ± 0.39 | 0.8 |
Trunk | 29.31 ± 0.72 | 29.51 ± 0.55 | 0.82 |
Total | 59.64 ± 1.09 | 60.55 ± 0.93 | 0.53 |
BMC (kg) | |||
Arm* | 0.39 ± 0.02 | 0.43 ± 0.01 | 0.035 |
Leg | 1.27 ± 0.03 | 1.27 ± 0.03 | 0.99 |
Total | 3.19 ± 0.06 | 3.31 ± 0.06 | 0.19 |
BMD (g/cm2) | |||
Arm | 0.99 ± 0.02 | 1.02 ± 0.01 | 0.22 |
Rib | 0.78 ± 0.01 | 0.80 ± 0.01 | 0.17 |
Leg | 1.40 ± 0.02 | 1.43 ± 0.02 | 0.25 |
Total | 1.26 ± 0.01 | 1.29 ± 0.02 | 0.42 |
Spine | 1.24 ± 0.02 | 1.23 ± 0.04 | 0.19 |
Hip | 1.06 ± 0.02 | 1.05 ± 0.02 | 0.99 |
Osteopenia in hip | 27.7 | 30.0 | NS |
Osteoporosis in hip | 2.13 | 0 | NS |
Osteopenia in spine | 23.4 | 26.7 | NS |
Osteoporosis in spine | 0 | 0 | NS |
Data are means ± SE or percentage.
Significant relations.
Article Information
We are thankful to Christie Hinton, bone density technician, and physician assistants Harold Hall and Marie Hall for their assistance during the study.
References
Published ahead of print at http://care.diabetesjournals.org on 6 April 2007. DOI: 10.2337/dc07-0337.
S.D. has received lecturing fees from Solvay, Novartis, and Auxilium.
A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C Section 1734 solely to indicate this fact.