Ethnic groups vary in their risk of age-related osteoporosis. There are ethnic differences in the metabolism of the bone forming minerals, calcium (Ca) and phosphate (P), with differences in renal Ca excretion, even at similar intakes and rates of absorption. We have previously reported a lower urinary Ca excretion in Black Gambian (G) compared to Caucasian British (B) subjects( Reference Yan, Schoenmakers and Zhou 1 ). These ethnic groups are known to differ in their dietary Ca intake and vitamin D (VD) status. We have investigated ethnic differences in urinary Ca excretion while allowing for other known predictors of renal Ca handling, including dietary Ca.
Subjects were healthy older (60–75 years) G (n=61) and B (n=60) men and women. Blood and timed 2 h urine samples were collected after an overnight fast. Dietary intake was assessed with country specific food diaries and food composition tables. Anthropometry was performed. Blood ionised Ca (iCa), plasma (p) total Ca (pCa), P (pP), 25-hydroxy VD (25VD), 1,25-dihydroxy VD (1,25VD), parathyroid hormone (PTH), and urinary (u) Ca (uCa), creatinine (uCr), and uP were analysed( Reference Yan, Schoenmakers and Zhou 1 ). The fractional excretion of Ca (FECa) and the tubular maximum of Ca (TmCa) were calculated( Reference Kent, Bhagat and Garcia-Webb 2 ).
Multiple regression was performed to investigate the effect of predictors of renal Ca handling. Ethnicity was used as an independent binary variable to test for differences between ethnic groups in a regression model. Data were transformed to natural logarithms to allow for the expression of ethnic differences (Δ) between G and B subjects as a %( Reference Cole 3 ).
G subjects had significantly lower body weight [ΔG–B: −31(−37, −25)%; P<0.001], pCa [ΔG–B: −5(−7, −2)%; P<0.001], and higher 25VD [ΔG–B: +35(+22, +48)%; P<0.001], PTH [ΔG–B: +62(+38, +85)%; P<0.001] and 1,25VD [ΔG–B: +48(+38, +58)%; P<0.001]. Markers of bone turnover were higher in G compared to B subjects (P<0.001).
There were significant ethnic differences in renal Ca handling as measured by uCa (μmol/minute); uCa/uCr (μmol/μmol); FECa (%); and TmCa (mmol/L) (P<0.001). The p-value and the effect size of these ethnic differences were similar after including dietary Ca (mg/day), or dietary Ca:P ratio (mg/mg) in the model (see table).
* Significance of regression coefficient of ethnicity, indicated by Pu (univariate model); Pa (model adjusted for dietary Ca).
In a multiple regression analysis without ethnicity, the significant predictors of uCa/uCr were: dietary Ca, diastolic blood pressure, iCa, pP, P1NP, CTX and uP. However, in a model including ethnicity, dietary Ca was no longer a significant predictor of uCa/uCr; whereas diastolic BP, iCa, pP, P1NP, CTX and uP remained significant predictors of uCa/uCr. Ethnicity was a significant predictor of uCa/uCr in the multivariate model ([ΔG–B: −133(−175, −107)%; P<0.001]).
These data suggest that G subjects have higher renal Ca conservation than B counterparts. The ethnic differences remained in the presence of other known predictors of renal Ca handling.
Funded by the Medical Research Council under Programmes U105960371 and U123261351.
