点击显示 收起
School of Medicine and Pharmacology
University of Western Australia
Perth
Australia
E-mail: adevine{at}cyllene.uwa.edu.au
Dear Sir:
We thank Dr Sebastian for his interest in and comments on our investigation of dietary protein intake and bone health in elderly women.
The first question raised by Sebastian related to the average values of protein intake and within-group distribution of 2 of the protein groups, ie, the low- (<66 g/d) and high- (>87 g/d) protein groups. The low-protein group had an average protein intake of 54 ± 9 g/d (range: 23.05–65.95 g/d), and the high-protein group had an average protein intake of 111 ± 25 g (range: 87.03–258.33 g/d). Sebastian suggested that the low-protein group may have suboptimal protein intakes, a possibility that is supported in part by these data, because approximately one-third of the low-protein group had a protein intake below the Australian recommendation of 0.75 g protein/kg body wt (1), whereas none of the high-protein group had less than the recommended intake.
Sebastian suggested that the subjects in the low-protein group, who consumed suboptimal amounts of protein, may be susceptible to the catabolic effect of the positive dietary net acid load. As suggested, we estimated the dietary net acid load by using the Frassetto algorithm (2). As rightly stated by Sebastian, the net acid load indeed was significantly lower in the low-protein group than in the high-protein group (38 ± 7 and 66 ± 18 mEq/d, respectively). Furthermore, we found that the net acid load in the whole group was positively related to bone density at all hip sites and measurements of quantitative ultrasound (broadband ultrasound attenuation and stiffness), which reflected increasing bone density with increasing protein intake. With the use of multiple regression analysis, protein intake was the only significant determinant of bone density or broadband ultrasound attenuation after we accounted for dietary net acid load in the model. This suggests that, in our population, net acid balance was not a critical factor in measuring bone density and that protein intake was beneficial regardless of the net acid load it induced. This possibility was further borne out by an analysis within each tertile of protein intake (ie, low, medium, and high protein intake), which found, in a linear correlation, that no association existed with any measure of bone density or net acid load.
Sebastian went on to hypothesize that we may see an independent negative effect of dietary net acid load on bone density that is greater in those persons with protein intakes below the median (ie, the low-protein group) than in those with intakes above the median (ie, the high-protein group). Using multivariate analysis, we did not, however, see a negative effect of net acid load on bone density in the low-protein group when protein intake was dichotomized to above or below the median (56.3 g/d), nor did we see a negative effect of net acid load on bone density in either the medium- or high-protein group.
Sebastian discussed the possibility that the anabolic effect is greater in bone when low habitual intakes are increased from 40 to 110 g/d rather than when moderate intakes are increased from 80 to 150 g/d. He speculated that, across this range of protein intakes, bone may respond more readily to protein, which offsets the catabolic effect of the potentially higher net acid loads. As ascertained from our data, we did not see a negative relation between the amount of dietary net acid load and the bone mass in any tertile of intake. Essentially, in our population, net acid balance was not a critical factor in measuring bone density.
ACKNOWLEDGMENTS
The authors had no conflicts of interest.
REFERENCES