Silicon and bone.

摘要

A main conclusion of a recent paper in Bone [1] that 'rats actively maintain body silicon levels via urinary conservation' is not supported by the primary data. It is clearly stated in the paper that spot urine samples were collected from 6 h-fasted rats from each of the treatment groups at one time point (three weeks prior to sacrifice) only. Problems arise when these one-off urinary silicon concentrations are then compared to serum silicon concentrations taken over a range of times during the entire experimental period and not at the same time as the urine samples. Ratios of urine silicon to serum silicon are computed, presumably from mean values, and are used as the only support for the controversial suggestion that 'rats actively maintain body silicon levels via urinary conservation'. Bearing in mind the extremely limited data set for both urine and serum silicons and that these tissue samples were not taken at equivalent times the authors then proceed to determine that in the Si-supplemented group the silicon urine:serum ratio was approximately 20 whereas it was only approximately 3 in the Si-deprived group. The difference in the ratios is mainly attributable to a ten-fold lower concentration of urinary silicon in the latter which could be expected based upon a feeding regime which included approximately 3500 times more silicon (all available as rapidly absorbed silicic acid) in the Si-supplemented group. Keeping in mind the enormous difference in the amount of silicon fed to each group, their urine:serum silicon ratios could be equivalent either if the serum silicon in the Si-deprived group was ca 7 fold lower (which would, bearing in mind the 10-fold dilution factor used in the analyses, be below the limit of detection by ICP-OES) or if the urine silicon in the Si-supplemented group was ca 7 fold lower. Regarding the latter, there have been several studies in humans which have shown that urinary excretion of silicon peaks several hours after taking a silicon-rich drink [2-4]. Both the height of the peak and the breadth (time-dependency) of the excretion profiles depend upon the concentration of silicon in the drink. However, following the drink urinary silicon invariably falls back to the pre-drink level, a level which approximately reflects serum silicon concentrations. Since only one urine sample was collected for each rat in this study the authors have absolutely no information concerning how each of the samples related to the urinary silicon excretion profile of each rat. For example, did the sample reflect a peak concentration, a background concentration or a concentration in between? The only valid way by which the authors could have computed useful urine:serum silicon ratios would have been from serum and urine samples taken at the same time and for each individual rat. It would be my contention that the ratios as presented are misleading and should not be used to support any discussion of silicon handling in these animals.