Table 1.
pQCT-derived bone parameters and bone strength potential correlates in men and women InCHIANTI participants
| Men (n = 372) | Women (n = 435) | Pa | |
|---|---|---|---|
| Age (years) | 74.0 ± 6.5 | 75.4 ± 7.0 | <0.0001 |
| Education (years) | 6.3 ± 3.7 | 4.7 ± 2.8 | <0.0001 |
| BMI (kg/m2) | 27.1 ± 3.3 | 27.8 ± 4.7 | 0.0062 |
| WHR | 95.2 ± 8.9 | 90.5 ± 11.0 | <0.0001 |
| Total bone area (tCSA) (mm2) | 413 ± 60 | 364 ± 55 | <0.0001 |
| Cortical bone area (cCSA) (mm2) | 334 ± 52 | 270 ± 58 | <0.0001 |
| vBMDt (mg/cm3) | 281 ± 51 | 236 ± 48 | <0.0001 |
| vBMDc (mg/cm3) | 1018 ± 65 | 985 ± 73 | <0.0001 |
| CSMA (cm2) | 64 ± 13 | 61 ± 13 | <0.0001 |
| Ankle extension strength (kg) | 35.7 ± 10.2 | 25.5 ± 4.7 | <0.0001 |
| Physical activity | <0.0001 | ||
| Sedentary (%) | 7.8 | 21.4 | |
| Light (%) | 82.8 | 76.1 | |
| Moderate/High (%) | 9.4 | 2.5 | |
| Energy intake (kcal/day) | 2205 ± 553 | 1731 ± 473 | <0.0001 |
| Protein intake (g/day) | 84 ± 20 | 70 ± 19 | <0.0001 |
| Vitamin D intake (μg/day) | 2.0 ± 0.80 | 1.8 ± 0.9 | <0.0001 |
| Calcium intake (mg/day) | 866 ± 284 | 801 ± 313 | 0.0026 |
| Creatinine clearance (ml/min) | 70 ± 18 | 60 ± 19 | <0.0001 |
| Serum PTH (pmol/l) (median and IQ range) | 20.7 [15.2–27.4] | 23.8 [16.2–33.1] | 0.0004 |
| Serum 25(OH)-vitamin D (nmol/l) (median and IQ range) | 55.2 [35.9–75.9] | 34.7 [24.5–21.0] | <0.0001 |
| Ca++ (mg/dl) | 4.1 [4.0–4.2] | 4.1 [4.0–4.3] | 0.051 |
| Bio-T (ng/dl) (median and IQ range) | 90.7 [68.7–116.4] | 9.6 [6.1–15.0] | <0.0001 |
| Total IGF-1 (ng/dl) (median and IQ range) | 122.1 [89.3–152.8] | 100.9 [70.0–137.5] | <0.0001 |
| DHEAS (μg/dl) (median and IQ range) | 70.8 [43.6–11.4] | 64.5 [35.7–114.8] | 0.34 |
| IL-6 (pg/ml) (median and IQ range) | 1.5 [0.9–2.4] | 1.3 [0.8–2.0] | <0.0001 |
| IL-1 beta (pg/ml) (median and IQ range) | 0.0 [0.0–0.40] | 0.0 [0.0–0.85] | 0.23 |
| TNF-alpha (pg/ml) (median and IQ range) | 0.0 [1.46–3.44] | 1.93 [1.51–3.07] | 0.31 |
Data are reported as mean value ± standard deviations if not otherwise specified.
a
From age-adjusted linear or multinomial logistic regression models as appropriate. Variables with a skewed distribution are presented as value and interquartile range and were rank-transformed for the analysis.