An association between bone mineral density and anabolic hormones in middle-aged and elderly men with prediabetes

Figures & data

Table 1. Characteristics of patients with PD and control group.

Parameter	Prediabetes no. 84	Control no. 58	p*
Age (years)	66.5 ± 3.8	65.7 ± 3,8	NS
TT (nmol/l)	11.24 ± 1.81	16.23 ± 1.6	.001
cFT (nmol/l)	0.331 ± 0.08	0.387 ± 0.07	.005
DHEAS (ng/ml)	542 (186-867)	669 (229-943)	.02
IGF-1 (ng/ml)	87.3 ± 43.3	94.3 ± 41.9	NS
LH (IU/l)	6.7 ± 0.9	4.4 ± 1.2	.05
FSH (IU/l)	8.2 ± 1.1	7.9 ± 1.3	NS
Estradiol (pg/ml)	31.9 ± 8.2	28.6 ± 7.8	.05
SHBG (nmol/l)	31.8 ± 3.2	27.7 ± 3.4	.02
BMI (kg/m^2)	28.6 ± 1.2	27.8 ± 0.9	NS
HbA1c (%)	6,2 ± 0.9	5.2 ± 1.2	.02
FPG (mg/dl)	116 ± 6.6	92 ± 4.6	.001
Glucose in OGTT (mg/dl)	132 ± 6.4	117 ± 6.6	.05
Cholesterol (mg/dl)	225 ± 18	202 ± 17	.05
Triglycerides (mg/dl)	162 ± 13	151 ± 14	.05
HDL cholesterol (mg/dl)	36 ± 4,6	39 ± 4.2	NS
LDL cholesterol (mg/dl)	146 ± 12	132 ± 13	.02
Cas (mmol/l)	2.26 ± 0.12	2.31 ± 0.13	NS
Cau (mmol/24h)	3.98 ± 0.89	4.39 ± 0.85	NS
Ps (mmol/l)	0.98 ± 0.28	1.08 ± 0.17	NS
ALP (U/l)	176.3 ± 45.0	175.3 ± 43.5	NS
25-OH D3 (ng/ml)	32.8 ± 12.8	37.8 ± 13.2	NS
PTH (pg/ml)	19.2 ± 8.5	15.1 ± 8.5	NS
BMD (g/cm^2) measured by DXA
Lumbar spine	1.144 ± 0.048	1.342 ± 0.067	.02
Femur neck	0.984 ± 0.055	1.025 ± 0.063	NS
Total body	1.157 ± 0.088	1.234 ± 0.094	.05
Comorbidities
Obesity, % (no.)	71 (60)	57 (33)	.02
MetS, % (no.)	83 (70)	67 (39)	.01
CVD, % (no.)	26 (22)	19 (11)	.05

Data are presented as a mean ± standard deviation of the mean, a median (with lower and upper quartiles) or number (percentage), where appropriate. *p shows differences between men with PD and control group. BMI: body mass index; WC: waist circumference; TT: total testosterone; cFT: calculated free testosterone; LH: luteinizing hormone; DHEAS: dehydroepiandrosterone sulfate; SHBG: sex hormone-binding globulin; E2: estradiol; IGF-1: insulin-like growth factor 1; FPG: fasting plasma glucose; OGTT: oral glucose tolerance test; HbA1c: glycated hemoglobin; Ca_s: calcemia; Ca_u: calciuria; P_s: phosphatemia; ALP: alkaline phosphatase; 25-OH D₃: 25-hydroxyvitamin D; PTH: parathormone; BMD: bone mineral density; DXA: dual energy X-ray absorptiometry; MetS: metabolic syndrome; CVD: cardiovascular.

Table 2. Univariate correlation between BMD and other variables in patients with PD.

Parameter	Lumbar spine		Femur neck		Total body
	r	p	r	p	r	p
Age	0.08	.22	0.09	.29	0.12	.19
TT	0.376	.023*	0.412	.01*	0.382	.014*
cFT	0.235	.038*	0.421	<.001*	0.327	.028*
DHEAS	0.07	.57	0.24	.84	0.15	.92
IGF-1	0.19	.07	0.23	.06	0.09	.23
Estradiol	−0.12	.34	−0.23	.18	−0.19	.65
SHBG	−0.25	.34	−0.08	.32	−0.329	.023*
BMI	−0.07	.21	−0.23	.16	−0.19	.06
HbA1c	−0.368	.045*	−0.07	.067	−0.167	.49
Cholesterol	−0.23	.27	0.09	.34	−0.18	.67
25-OH D3	0.23	.57	0.12	.58	0.08	.34
PTH	−0.25	.06	−0.16	.38	−0.27	.06

* p < .05, r: correlation coefficients; BMI: body mass index; TT: total testosterone; cFT: calculated free testosterone; DHEAS: dehydroepiandrosterone sulfate; SHBG: sex hormone-binding globulin; E2: estradiol; IGF-1: insulin-like growth factor 1; HbA1c: glycated hemoglobin; 25-OH D₃: 25-hydroxyvitamin D; PTH: parathormone.

Table 3. Univariable and multivariable linear regression analysis for putative predictors of BMD in patients with PD.

Parameter	Lumbar spine		Femur neck		Total body
	Univariable models	Multivariablemodelscorrected R^2 = 27%^c	Univariable models	Multivariablemodels correctedR^2 = 27%^c	Univariablemodels	Multivariable modelscorrected R^2 = 32%^c
Age	0.18	–	0.12	–	0.12	–
TT, nmol/l	0.37^a	0.26^a	0.06^b	0.19^b	0.21	0.22^b
cFT, nmol/l	0.23^a	0.23^b	0.34^a	0.20^a	0.23	0.28^a
DHEAS, ng/ml (In)	0.17^b	0.19	0.18^a	0.12	0.02	–
IGF-1, ng/ml	0.02	–	0.08	–	0.08^a	0.31^a
Estradiol, pg/ml	−0.01	–	−0.25	–	−0.21	–
SHBG, nmol/l	−0.25	–	−0.02	–	−0.37^a	−0.21^a
BMI, kg/m^2	−0.18	–	−0.34	–	−0.19	–
HbA1c, %	−0.21^a	−0.20^a	−0.07^a	−0.37^b	−0.16^b	−0.24^a
25-OH D3, ng/ml	0.23	–	0.41	–	0.23	–
PTH, pg/ml	−0.25	–	−0.37	–	−0.27	–

a: p < .05, b: p < .01, c: p < .001; data are presented as standardized regression coefficients β (both in univariable and multivariable models). All variables shown to be significant determinants of bone mineral density (BMD) in the univariable models (p < .05) were included in the multivariable models. ln: natural logarithm; BMI: body mass index; TT: total testosterone; cFT: calculated free testosterone; DHEAS: dehydroepiandrosterone sulfate; SHBG: sex hormone-binding globulin; E2: estradiol; IGF-1: insulin-like growth factor 1; HbA1c: glycated hemoglobin; 25-OH D₃: 25-hydroxyvitamin D; PTH: parathormone.

Figure 1. BMD in patients with PD after dividing according to cFT levels: low cFT: <0.250 nmol/l, high cFT: > 0.250 nmol/l.