Hormonal and biochemical parameters correlated with bone densitometric markers in prepubertal Hungarian children

Abstract

Background: The conditions that define bone development in prepuberty profoundly influence bone health later in life. We aimed to reveal important determinants of bone mass in Tanner stage I.

Methods: We studied 84 healthy children (43 girls and 41 boys) aged 7 to 11 years. Serum estradiol (E2), 25-hydroxyvitamin D3-vitamin [25(OH)D3], intact parathyroid hormone (PTHi), osteocalcin (OC) and β-crosslaps (CTXs) were longitudinally analyzed (Roche Diagnostics System). Total and spine bone mineral content (tBMC and LBMC) and density (tBMD and LBMD) were assessed, and total fat body mass index (FBMi) was calculated (DXA Lunar Prodigy).

Results: The serum PTHi, OC and LBMD values were significantly higher in girls than in boys. The mean 25(OH)D3 level was lower but not significantly in girls compared to boys. Significant negative correlation was found between PTHi and 25(OH)D3 levels (r=-0.28; p=0.011) when tested in all subjects, but no correlation was detected when the gender groups were separately tested. There was a trend for higher E2 levels in girls. Significant positive correlation (r=0.32; p=0.042) was detected between FBMi and E2 concentration in girls only. A significant negative correlation was found between E2 and 25(OH)D3 levels (r=-0.37, p<0.05) in girls with elevated (>3.6pmol/l) PTHi and with suboptimal (<75nmol/l) 25(OH)D3 levels. Furthermore, positive correlations were noted between E2 and CTXs and OC (r=0.54, p<0.01 and r=0.39, p<0.03) and a marginally significant positive correlation (r=0.33; p=0.06) was detected between OC and PTHi levels in girls. However, we detected no correlations when these markers were analyzed in boys. There was a significant correlation between E2 and all BMC and LBMD values in both genders. The tBMD, LBMD and tBMC values showed weak, but significant negative associations with 25OHD3 levels (β=-0.44 to -0.55; p<0.001) in girls only. All BMD and BMC values were positively predicted by OC levels, but not by CTXs, in both genders. Among the biochemical markers, E2 was the only factor correlating with all dependent variables (BMCs and BMDs) in both genders. Among all parameters analyzed, FBMi (β=0.64) showed the strongest influence on tBMC characteristically in girls only.

Conclusions: Our results support that 1.) E2 levels play a key role in defining bone turnover and bone mass in both genders already in prepuberty; 2.) high PTHi levels in childhood should be evaluated with caution, because the normal range for serum PTHi in different Tanner stage groups is not well established; and 3.) the negative correlation between 25(OH)D and E2 and the positive correlation between PTHi and OC suggest that estrogens regulate PTHi indirectly and cause lower circulating 25(OH)D3 levels. We propose that the decreased levels of 25(OH)D3 reflect not the real vitamin supply, but may rather be the result of E2 regulation. Therefore, the actual serum 25OHD levels may underestimate the availability of factors supporting bone formation.