点击显示 收起
School of Biomedical and Molecular Sciences
University of Surrey
Guildford GU2 7XH
United Kingdom
E-mail: s.new{at}surrey.ac.uk
Osteoporosis Research Unit
Department of Medicine and Therapeutics
University of Aberdeen Medical School
Aberdeen AB251ZO
United Kingdom
Dear Sir:
Dr Remer's pertinent response to our recent publication on associations between estimates of net endogenous (noncarbonic) acid production (NEAP) and bone health indexes in the population of the Aberdeen Prospective Osteoporosis Screening Study (1) is most timely. We have reexamined our findings in light of the 3 important points he raises.
First, although we analyzed the association between NEAP and markers of bone health by using potassium intake converted to mEq/d, we reported NEAP estimates by measuring potassium intake in mg/d and not in mEq/d, which, as correctly pointed out by Remer, is required for the Frassetto algorithm (2). The correct estimates (mean ± SD; median and range) for NEAP and the values used for grouping the data set into quartiles are shown in Tables 1 and 2, respectively.
View this table:
TABLE 1. Net endogenous (noncarbonic) acid production (NEAP) and potential renal acid load (PRAL) estimates of the study population1
View this table:
TABLE 2. Net endogenous (noncarbonic) acid production (NEAP) estimates for quartile classification
The second point raised by Remer is a critical one, given the current controversy concerning the relation between dietary protein and bone (3, 4). We have examined our extensive data set to determine the percentage of women with an intake of protein below the reference nutrient intake (RNI), ie, <45 g/d (5), and we examined their respective bone mineral density (BMD) values. Only 24 women (2.3%) were below the RNI for protein and had a mean estimate for NEAP of 3.94 ± 0.95 mEq · d–1 · MJ–1 (range 2.04–5.50 mEq · d–1 · MJ–1) and a mean estimate of potential renal acid load (PRAL) of –7.90 ± 9.44 mEq/d (range –33.6 to 4.58 mEq/d). Of these subjects, 50% had a lumbar spine BMD below the median value for the population. Figures for the other BMD sites were as follows: 67% (n = 16) for femoral neck BMD; 50% (n = 12) for femoral trochanter BMD; and 71% (n = 17) for femoral Ward's BMD. The numbers of subjects with low protein intake are too small to allow us to comment further on the growing body of evidence that long-term low dietary protein consumption may be harmful to skeletal integrity. It is important also to note that subjects with a low dietary protein intake are likely to be deficient in other nutrients that may be of benefit to bone.
In response to the third point, we estimated the PRAL by using Remer's calculation model (6), and these estimates are shown in Table 1. Furthermore, we investigated the association between estimates of PRAL and measurements of bone metabolism and BMD. The correlation between NEAP and PRAL was 0.93 (P < 0.001). Lower estimates of PRAL were associated with a higher peripheral cortical forearm bone mass (P < 0.03) and lower deoxypyridinoline excretion (P < 0.048), with similar nonsignificant trends for peripheral total bone mass and pyridinoline excretion (P < 0.07 and P < 0.1, respectively). However, correlations were weaker than those we reported for NEAP and bone. There was a nonsignificant trend for the lumbar spine and hip BMD to decrease across increasing tertiles of PRAL (P < 0.1), with similar findings for forearm bone mass.
We thank Dr Remer for providing us with this opportunity for extensive discussion of estimates of renal net acid excretion and its subsequent effects on bone health, and we encourage other groups to reanalyze existing dietary intake and bone health data sets to enable further exploration of the effect of dietary acidity and alkalinity on skeletal integrity.
REFERENCES