Dietary-Induced Metabolic Acidosis Decreases Bone Mineral Density in Mature Ovariectomized Ewes

Calcif Tissue Int (2004) 75:431–437 DOI: 10.1007/s00223-004-0217-7

Dietary-Induced Metabolic Acidosis Decreases Bone Mineral Density in Mature Ovariectomized Ewes J. M. MacLeay,1 J. D. Olson,3 R. M. Enns,1 C. M. Les,4 C. A. Toth,5 D. L. Wheeler,2 A. S. Turner1 1

Departments of Clinical Sciences, Colorado State University, Fort Collins, CO 80523, USA Department of Engineering, Colorado State University, Fort Collins, CO 80523, USA 3 Independent Consultant, Fort Collins, CO 80523, USA 4 Henry Ford Hospital, Detroit, MI, USA 5 Stryker Biotech, Inc., Hopkinton, MA, USA 2

Received: 28 August 2003 / Accepted: 19 April 2004 / Online publication: 12 August 2004

Abstract. Dietary-induced metabolic acidosis (DIMA) may be a significant confounder in the development of osteoporosis. Diets that are acidifying are typically rich in proteins and grains and relatively poor in fruits and vegetables. Previous studies have not examined whether an interaction between estrogen depletion and DIMA have a compounded affect on bone mineral density loss. Sheep have been used successfully in previous studies to examine the interaction of bone turnover and ovariectomy. Therefore, the goal of this pilot study was to determine if bone mineral density (BMD) loss could be induced using DIMA in skeletally mature ovariectomized (OVX) ewes. Key words: Dima — BMD — OVX ewes

Osteoporosis is a common and severe metabolic disease such that the lifetime risk of having an osteoporoticrelated fracture is close to 40% in postmenopausal women residing within the United States [1]. Osteoporosis is a multifactorial disorder being influenced by lifestyle, life-stage, genetic and dietary factors. While all factors influencing the development of Osteoporosis are important, dietary acid has gained some attention recently. Several authors believe that dietary-induced metabolic acidosis is a contributing factor in the development of Osteoporosis in humans [2–4]. Other investigators have also implicated ethnic differences in heritable titrateable acid excretion as also playing a role in bone mineralization and bone integrity [5]. In humans, metabolic acidosis induces increased calcium excretion without a concomitant increase in calcium absorption from the gut, therefore an overall decrease in total body calcium ensues [6]. This is different from the rat model where there appears to be a concomitant decrease in intestinal calcium excretion to Correspondence to: J. M. MacLeay; E-mail: [email protected]

compensate for increased urinary calcium loss in response to metabolic acidosis [7]. Therefore metabolic acidosis, combined with other factors, appears to have a serious impact on the development of Osteoporosis in susceptible individuals. Metabolic acidosis may arise from endogenous production of pathologic acids (lactic acidosis), decreased ability to excrete acid, and/or absorption of acids from the intestine. Those molecules that act as acids and bases and therefore have the greatest impact on acid-base status according to Stewart’s theories include three independent variables; the strong ions (or strong ion difference), total protein and pCO2, and the partial pressure of carbon dioxide [8, 9]. Decreased ability to excrete acid occurs in patients with respiratory disease (decreased ability to expire pCO2), chronic renal disease or renal tubular acidosis and in elderly patients with an age-related decline in renal capacity to excrete acid (decreased ability to excrete acids and strong ions in the urine) [4, 10]. In the absence of pathology, a relative excess of strong anions compared to strong cations in the diet may also induce metabolic acidosis in the face of normal renal function [4, 11]. Strong ions are those that are completely dissociated at physiologic pH and include Na, K, and Cl. This is because strong ions are absorbed directly from the intestine into the blood and therefore immediately impact acid-base balance, whereas weaker ions are absorbed at lesser efficiencies. Depending on the composition of the diet, the imbalance may be slight or substantial, leading to a corresponding degree of metabolic acidosis [12]. Blood pH is maintained within the normal range because of the combined effect of renal excretion of excess ions and mobilization of calcium from bone to act as a buffer in concert with the relatively stable chemical buffering capacity of the blood and the ability of the body to exchange H+ and HCO3) between the extracellular and intracellular space [10, 13].

432

J. M. MacLeay et al.: Effect of a Diet with Low Cation-Anion Difference in Bone Mineral Density

The influence of diet on acid-base status can be estimated in many ways. Simply, dietary animal protein intake, which is highly correlated with renal net acid excretion, may be recorded [14]. Another method involves knowing the net endogenous noncarbonic acid production capability of typically consumed foods where the renal net acid excretion can be estimated by knowing the foods consumed [14]. In veterinary medicine the study of metabolic acidosis and nutrition has largely centered on the use of total cation-anion difference in the diet to prevent post parturient hypocalcemia in dairy cows [15]. Using this model, total dietary cation-anion difference in the diet may be estimated by knowing the strong ion content of the diet as a whole instead of calculating it from individual multiple feedstuffs. Multiple articles have examined and formulated methods for determining cation-anion balance in the diet. Nevertheless, the most important strong ions in the diet are sodium, potassium, and chloride. Sulfur has been found to have an important impact as it is an anion found in large quantities in protein. Magnesium and calcium play a lesser role because a fraction of what is eaten is absorbed. Therefore, from dietary analysis, the overall dietary cation-anion difference (DCAD) may accurately, albeit roughly, be estimated by the equation: ðNaþ þ Kþ Þ  ðCl þ S Þ ¼ DCAD in mEq=kg DM The ionic balance of the diet may then be expressed in mEq/kg dry matter (DM) through analysis of the total diet or by measuring the 24-hour total intake of ions and then calculating the daily DCAD. Other dietary ions, such as calcium and magnesium, play a role in overall acid-base balance but alter the overall DCAD only slightly such that in many species the above equation is adequate under most circumstances [15]. Unfortunately, the potential acid load of a particular diet has been described differently in the veterinary and human literature. The veterinary literature has focused on analysis of the diet to determine strong ion content of the diet per kg consumed whereas the human literature has focused on urinary acid excretion, calculating the potential renal acid load of foods. This discrepancy in scientific methods has made direct comparison between species difficult. However, in one human study the DCAD could be calculated from the data provided by the authors. In this study a typical omnivore diet of 2700 (±16) kcal per day had a mean DCAD of 27 mEq (range )74 to 128 mEq) per day using the aforementioned equation [16]. In human beings, dietary acid imbalance arises through the consumption of a diet that is relatively low in potassium and high in sulfur-containing amino acids that comprise animal proteins [17]. In response to chronic, dietary acid loads the body seeks to

maintain neutrality by mobilizing calcium, phosphate and carbonate from bone as buffers. The mechanism by which bone is resorbed is both by physicochemical dissolution (in the acidic environment) and cell-mediated resorption, as osteoclasts are stimulated in an acidic micro-environment [10, 18, 19]. Heritable differences in titratable acid excretion may also play a role [5]. An ideal large animal model for osteoporosis has heretofore been elusive to researchers. The ovariectomized (OVX) ewe has been shown to have bone mineral loss similar to but not to the same degree as the postmenopausal woman and has been used successfully in our laboratory to study postmenopausal orthopedic disease [20–22]. The typical diet for sheep is alkalogenic compared to a typical human diet and whether sheep are sensitive to acidogenic diets as a cause for negative calcium balance is unknown. The purpose of this pilot study was to examine the combination of a high acid diet in concert with ovariectomy to determine if we could advance the sheep as an animal model for human postmenopausal osteoporosis.

Materials and Methods Animals and Diet Fifty-two skeletally mature (4–7-year-old) RambouilletColumbia cross ewes were used and all procedures were approved by the Colorado State University institutional animal care and use committee (Protocol # 01-090A-01) and animals were cared for in compliance with the Guiding Principles in the Care and Use of Animals [23]. Thirty-eight sheep underwent ovariectomy (OVX) and 14 sheep underwent sham surgery. Following surgery, sheep were assigned to consume one of two diets. Group one (ND-noOVX) sheep (n = 14) were not ovariectomized and were fed a diet consisting of free choice grass hay. The dietary cation-anion difference ((Na + K) ) (Cl + S) = DCAD mEq) of the hay diet was normal for this species and was +300 mEq/kg dry matter (DM). If it was assumed that the sheep consume approximately 3,000 kcal/day then their daily DCAD was approximately +1,000 mEq/day. Group 2 (ND-OVX) sheep (n = 14) were ovariectomized and were fed the free-choice hay diet. Group 3 (MA-OVX) sheep (n = 24) were ovariectomized and fed a diet to induce metabolic acidosis (MA) that consisted of limit-fed grass hay and a specially formulated pellet that provided adequate amounts of all nutrients, including calcium. The cation-anion difference of this diet was approximately -450 mEq/kg DM or for an intake of approximately 3,000 kcal/day they consumed +160 mEq/ day (Table 1). Table 2 describes the estimated total intake per day for protein and minerals for each experimental diet as well as current recommended intakes per day for the same nutrients. All sheep were weighed on day 0 and on day 90. All sheep underwent dual energy x-ray absorptiometry (DEXA) of the lumbar vertebrae on day 0. At 90 days, all 24 MA-OVX sheep underwent DEXA scans under general anesthesia whereas 7 ND-noOVX and 7 ND-OVX sheep had DEXA scans performed immediately post-euthanasia. Groups within the study were unbalanced as the control sheep (ND-OVX and NDnoOVX) were also part of a separate study. All sheep were humanely euthanized using an intravenous overdose of 8 g of pentobarbital [24].

J. M. MacLeay et al.: Effect of a Diet with Low Cation-Anion Difference in Bone Mineral Density

433

Table 1. Dietary composition Dietary cation-anion kcal Estimated amount offered/day/sheep difference mEq/kg consumed/day/sheep dry matter fed Diet Composition at 3.3% of body weight/day (Total) MA Grain Mix Grass hay ND Grass hay

1.26 kg 1.0 kg 2.7 kg

1,900 1,100 (3,000) 2,970

)765 308 308

Dietary cation-anion difference of total diet consumed/day/sheep if consumed between 2.6-3.3 mcal/day 150–175 mEq 900–1100 mEq

MA = low dietary cation difference, ND = normal dietary cation-anion difference, kcal = kilocalories

Table 2. Major mineral content for each diet assuming sheep on the MA and ND diets consumed the offered amounts listed in Table 1

MA ND Sheep nutrient requirements for 80 kg sheep Sheep nutrient requirements % of diet dry matter

Adjusted crude protein (grams)

Calcium (grams)

Phosphorus (grams)

Magnesium (grams)

Potassium (grams)

Sodium (grams)

Sulfur (grams)

Chloride (grams)

304 208 122

8 11 2.7

5.5 4 2.8

8 4 3.6

31 45 16

6 0.5 3.6

14.5 3.5 5.24

27 6 unknown

0.82%

0.38%

0.18%

0.8%

0.18%

0.26%

unknown

Sheep nutrient requirements are offered for comparison. Sheep nutrient requirements listed are those established by the subcommittee on Sheep Nutrition, Committee on Animal Nutrition, Board of Agriculture, National Research Council and published as Nutrient Requirements of Sheep, 6th, ed 1985, National Academy Press

Ovariectomy

Statistical Analysis

General anesthesia was induced using 7.5 mg of valium and 250 mg of ketamine intravenously and maintained using isoflurane at 1–3% and oxygen at 2 l/min. Ovariectomy was performed via midline laparotomy. Ovaries were exteriorized, the pedicles were ligated and the ovaries were removed. Sham surgeries involved visualization of the ovaries only. After standard midline 3-layer closure the sheep were moved to the dual energy X-ray absorptiometry (DEXA) machine for BMD measurements. Thereafter they recovered and were moved to pens for group housing. No complications with ovariectomy were noted during the course of the study.

The significance of change in bone density between days 0 and 90 was assessed using a restricted maximum likelihood-based mixed-effect model that included the categorical, fixed effects of treatment group (ND-noOVX, ND-OVX and MA-OVX), the continuous, fixed effect of time 0 bone density, and a random animal effect via PROC MIXED in SAS [25]. Kenwardroger’s Approximation was used to estimate denominator degrees of freedom. The same methods were used for percent change in bone density as for change in bone density. The significance and magnitude of treatment group differences in measured bone density were assessed using a restricted maximum likelihood-based mixed effect model, repeated measures analysis via PROC MIXED in SAS [25]. Analysis included the categorical, fixed effects of treatment group, time and treatment by time interaction. A random subject effect was included as animal within treatment group. An unstructured correlations structure was used to model the within-subject errors. Kenwardroger’s approximation was used to estimate denominator degrees of freedom for all analyses.

Bone Densitometry Bone densitometry measurement of the lumbar vertebrae was performed on days 0 and 90. Scans were performed using a Hologic QDR 1000-W dual-energy X-ray absorptiometer (DEXA) (Hologic, Inc. Bedford, MAX The software was supplied by Hologic (version 6.10.01) and all measurements were performed by the same operator/author (A.S.T.). Prior to measurements the machine was calibrated using a standard phantom and the CV was within ±1.5%. The number of lumbar vertebrae in sheep may vary. We identified the last 4 lumbar vertebrae of each animal from the DEXA scan image and have designated these four vertebrae L4, L5, L6 and L7 regardless of the actual number of lumbar vertebrae the sheep possessed [21]. The bone mineral density (BMD g/cm2) of the last four lumbar vertebrae was measured in a standard dorsoventral view.

Results

There were no significant differences in body weight between groups at baseline and at 90 days (Table 3). However, when the mean change in weight from baseline to 90 days was analyzed, sheep in the MA group lost significantly more weight than sheep in the ND groups

434

J. M. MacLeay et al.: Effect of a Diet with Low Cation-Anion Difference in Bone Mineral Density

Table 3. Body weight comparison between groups at baseline and at 90 days DAY 0

DAY 90

Difference

Group

Mean kg

Std dev (±)

Mean kg

Std dev (±)

Mean change (kg)

Std dev (±)

MA/OVX ND/OVX ND/noOVX

77 72 65

10.51 6.30 3.02

72 75 69

8.26 6.21 6.16

-5 3 4

-4.61 6.21 4.22

There was no statistically significant difference between the mean of each group on Days 0 and 90. The change in weight within the MA-OVX group was significantly greater than it was within the other two groups.ND = normal dietary cation-anion difference of approximately +300 mEq/kg dry matter (DM) consumed. MA = low dietary cation-anion difference of approximately -450 mEq/kg DM consumed. OVX = ovariectomy, Std Dev = standard deviation.

Fig. 1. Graph shows the relationship between weight loss and % change in BMD for each sheep in the sstudy. There was no correlation (R2 = 0.13) between weight loss/gain and the percent change in BMD over the course of the study. MA = metabolic acidosis, OVX = ovariectomy, ND = control diet. Diamonds = MA-OVX, squares = ND-OVX, circles = ND-noOVX.

(p < 0.05). The mean change in weight from baseline between the ND groups was not significantly different from each other (b = 0.0518). There was no correlation between change in body weight and change in BMD, R2 = 0.13 (Fig. 1). The treatment group was found to have significant effects on change in bone density (type 3 tests of fixed effects), but baseline bone density was found to have no effect. Tests of differences in least-squares means resulted in no significant differences between the ND groups (ND-NoOVX vs. ND-OVX p = 0.5390) but significant differences between ND-OVX vs. MA (p = 0.0032) and ND-noOVX vs. MA (p = 0.0007). The mean bone mineral density (±standard deviation) of ND sheep at time 0 was 1.03 ± 0.21 g/cm2 and for the MA group it was 0.96 ± 0.11 g/cm2. Mean bone mineral density at 90 days for the ND group was 1.06 ± 0.27 g/cm2 and 0.83 ± 0.10 g/cm2 for the MA group (p < 0.05) (Table 4). The same statistical methods were used to examine percent change in bone density. The group was found to have a significant influence on percent change but baseline bone density was not significant (Type 3 fixed tests by group had a p value of 0.0005 whereas for time 0, p = 0.8105). Mean separation tests were performed

using the DIFFS option in SAS. No significant differences were found between ND-OVX and ND-noOVX. All other differences were found to be significant. p value for ND-noOVX vs. ND-OVX was 0.5741 whereas for ND-OVX vs. MA it was 0.0028 and for ND-noOVX vs. MA it was 0.0007. Mean percent change (± standard deviation) in bone mineral density from baseline at 90 days for the MA group was )10.19 ± 6.23% and )21.95% ± 2.93% for the combined ND groups (Fig. 2). The change in BMD ranged from a 7.38% loss to a gain of 2.99% in the ND sheep, whereas all MA sheep lost BMD ranging from a decrease of 26.11% to a loss of 1.99%. The effect of time and group by time interaction had significant effects on bone density while the main effect of group approached significance (p = 0.0573). Least square means for group by time are demonstrated as a bar graph in Fig 3. Discussion

In humans, DEXA scans are considered to be a good way of screening patients to determine if they are at risk for osteoporotic related fractures [26]. Using this method of screening, patients are considered to be at

J. M. MacLeay et al.: Effect of a Diet with Low Cation-Anion Difference in Bone Mineral Density

435

2

Table 4. Bone mineral density (g/cm ) of the last 4 lumbar vertebrae for each treatment group at baseline and at 90 days Day 0

Day 90

Change

Group

Mean g/cm2

Std dev (±)

Mean g/cm2

Std dev (±)

Mean %

Std dev (±)

MA/OVX

0.96 n = 24 1.03 n = 14 1.00 n = 14

0.11

0.86 n = 24 1.00 n=7 1.12 n=7

0.12

)10.19 n = 24 )2.71 n=7 )1.18 n=7

6.23

ND/OVX ND/noOVX

0.13 1.12

0.13 0.36

2.06 3.61

ND = normal dietary cation-anion difference of approximately +300 mEq/kg dry matter (DM) consumed. MA = low dietary cation-anion difference of approximately -450 mEq/kg DM consumed. OVX = ovariectomy, Std Dev = standard deviation.

Fig. 2. Graph of mean percent decline in BMD (g/cm2) for each group of sheep for 0 versus 90 days. ND = normal dietary cation-anion difference of approximately +300 mEq/kg dry matter (DM) consumed, MA = low dietary cation-anion difference of approximately )450 mEq/kg DM consumed. OVX = ovariectomy. Different letters signify significant differences between groups at p < 0.05. Error bars are equal to 1 standard deviation.

greatest risk when their BMD is at least 2 standard deviations below the population mean. One difficulty in studying osteoporosis in an animal model is achieving bone loss as great as that seen in humans. In addition, achieving such a great loss in BMD in a short time frame would be economically advantageous to the researcher. This would assume that the resulting quality of bone loss is similar to that seen in human osteoporosis. The sheep is a convenient and historically good large animal model for studying orthopedic conditions [20– 22, 27–29]. The sheep is seasonally polyestrous and therefore does experience skeletal changes related to chronic estrogen deficiency post-ovariectomy, but, like most non-primate mammals, the sheep does not experience a natural menopause. However, even in primate studies of menopause-related osteoporosis the ovariectomized model is commonly used. This is based on the fact that women undergoing surgical menopause, while potentially having a more rapid onset of osteoporosis, do not appear to have any differences in the physiologic mechanisms responsible for that bone loss when compared to women undergoing natural menopause [30, 31]. Unfortunately, the use of sheep in studying osteoporosis has been somewhat limited because significant losses in

BMD are difficult to achieve with ovariectomy alone [20]. Methods to achieve additional BMD loss in sheep or goats have heretofore involved either chronic glucocorticoid administration or severe dietary calcium restriction [32, 33]. Chronic glucocorticoid administration in the sheep often leads to septic complications and a high mortality rate due to the immunosuppressive effects of the steroid. In addition, there is evidence that the mechanism of bone loss in animal models treated with chronic glucocorticoids is not identical to that seen in postmenopausal osteoporosis. Therefore, this method of inducing bone loss in sheep is not ideal [34]. Dietary calcium restriction is difficult to achieve in the mature, non-lactating, non-pregnant ewe as her daily requirements are low for Ca and the amount of Ca in grass hay is typically adequate for maintenance. When calcium-restricted diets are used in the sheep model, significant bone mineral loss can be achieved, however, it is unknown if the resulting bone has similar biomechanical properties as human osteoporotic bone. Another problem is that Ca-restricted diets are very difficult to formulate and produce for a ruminant model. They are low in dietary fiber and create complications as ruminants require a basal amount of fiber in their diet for normal gastrointestinal function. In addition, the study of osteoporosis in a calcium-restricted animal model may be of limited use because humans with osteoporosis may have an adequate Ca intake yet still develop significant losses in BMD over time. One theory for loss of bone mineral in humans is that dietary cation-anion difference or dietary-induced metabolic acidosis plays a significant role in bone turnover [4, 14, 35]. The authors felt that the combination of dietary manipulation and ovariectomy presented in this pilot study may more realistically mimic the physiologic processes occurring in postmenopausal women than other experimental methods published in the literature. Based on data from Dwyer et al. [16] we could calculate the daily dietary cation-anion difference (DCAD) of a typical human diet to be approximately +27 ()74 to +128) mEq/day. Sheep typically consume a diet that is

436

J. M. MacLeay et al.: Effect of a Diet with Low Cation-Anion Difference in Bone Mineral Density

Fig. 3. Graph of least square means for BMD (g/cm2) for each group of sheep, showing groups at days 0 and 90. The effect of time and group by time interaction had significant effects on bone density while the main effect of group approached significance (p = 0.0573). ND = normal dietary cation-anion

difference of approximately +300 mEq/kg dry matter (DM) consumed. MA = low dietary cation-anion difference of approximately -450 mEq/kg DM consumed. OVX = ovariectomy. Error bars are equal to the standard error.

approximately +1000 mEq/day. In this study we fed an experimental diet that had a DCAD of approximately +162 mEq/day and this produced a significant decline in bone mineral density. Our experimental diet was relatively more alkalogenic than a typical human diet yet we were able to induce rapid bone mineral loss at a rate not seen in humans consuming a similar diet. Therefore, we can conclude that sheep are well adapted to their typical diet and that when a more acidogenic diet than is typical is consumed, a negative calcium balance results. It has been suggested that the modern diet of humans is somewhat more acidic than that of Paleolithic man which is why the typical diet consumed by humans today contributes to the slow loss of bone mineral over time [17, 36]. Unfortunately, many questions remain and based on this pilot study we cannot make assumptions as to the degree of pressure that any particular diet can induce within any particular species. Nor can we make any assumptions concerning the correlation between a particular degree of acidosis between species consuming diets of similar cation-anion differences. However, because of their apparent sensitivity to dietary manipulation, we can conclude that the findings of this study support the use of sheep as an excellent model in which to study the mechanisms involved in bone loss related to high acid diets. The results presented here demonstrate a mean of approximately 1 standard deviation loss in BMD in 90 days. This decline is not as severe as that seen in humans who have developed osteoporosis over several years, but is considerably greater than other studies in sheep using ovariectomy alone over time spans of up to 12 months. However, a synergistic effect between diet and ovariectomy in this study was not obvious. Certainly, the low dietary cation-anion difference played a more significant

role in bone mineral density loss, and longer-term studies will be necessary to determine if an interaction exists. While no significant differences in body weight were seen between groups at baseline and at 90 days, changes in weight among sheep were significantly different. This may reflect the decreased palatability of the low cationanion diet or the smaller size of the ND group at the 90day time point. Decreased BMD can be a reflection of decreased body weight over time. However, the correlation between weight loss/gain and % change in BMD was not significant in this study. Therefore we can conclude that factors affecting weight loss or gain appeared to be independent of those affecting the change in bone mineral density. The pilot nature of this study had several flaws that will be addressed in future studies, which will include a longer time frame, larger dedicated treatment groups, biomechanical testing and histomorphometry to determine if the bone from these sheep possesses increased fragility similar to that seen in humans and evaluation of biochemical markers of bone turnover. The interaction of a low, moderate or high calcium intake with respect to dietary-induced metabolic acidosis has yet to be explored and longer-term studies combining dietaryinduced metabolic acidosis and ovariectomy will be necessary to determine if any interaction exists between them. Methodology must be developed to relate the physiologic responses to dietary acid-base manipulation between species. In addition, future studies could examine the affect of dietary acidosis on the attainment of peak BMD in adolescents. We conclude that the sheep appears to be very sensitive to dietary-induced metabolic acidosis, as demonstrated by a significant and rapid decline in bone mineral density and therefore may prove to be a valuable animal

J. M. MacLeay et al.: Effect of a Diet with Low Cation-Anion Difference in Bone Mineral Density

model in which to study the interaction between dietary acid-base and bone health. Acknowledgments. Funding was provided by the Colorado Research Council, Stryker Biotech, Inc., National Institutes of Health (R01-AR47434-01). References 1. Melton L, Chrischilles E, Cooper C, et al. (1992) How many women have osteoporosis? J Bone Miner Res 7:1005–1010 2. New S (2002) The role of the skeleton in acid-base homeostasis. Proceedings of the Nutrition Society 61:151– 164 3. Lemann J, Litzow J, Lennon E (1966) The effects of chronic acid loads in normal man: further evidence for the participation of bone mineral in the defense against chronic metabolic acidosis. J Clin Invest 45:1608–1614 4. Sebastian A, Harris S, Ottaway J, et al. (1994) Improved mineral balance and skeletal metabolism in postmenopausal women treated with potassium bicarbonate. N Engl J Med 23:1776–1781 5. Vaitkevicius J, Witt R, Maasadam M, et al. (2002) Ethnic differences in titratable acid excretion and bone mineralization. Med Sci Sports Exerc 34:295–302 6. Fettman MJ (2000) Dietary instead of pharmacological management to counter the adverse effects of physiological adaptations to space flight. Pflugers Arch-Eur J Physiol 44(1suppl:)R15–R202 7. Whiting S, Draper H (1981) Effect of chronic high protein feeding on bone composition in the adult rat. J Nutr 111:178–183 8. Stewart P (1981) How to understand acid base balance. In: Stewart P (ed) A Quantitative Acid-Base Primer for Biology and Medicine. Elsevier, New York 9. Stewart P (1983) Modern quantitative acid-base chemistry. Canad J Physiol Pharmacol 61:1444–1461 10. Arnett T (2003) Regulation of bone cell function by acidbase balance. Proc Nutr Soc 62:511–520 11. Goff JP, Horst RL, Mueller FJ (1991) Addition of chloride to a prepartal diet high in cations increases 1,25-dihydroxyvitamin D response to hypocalcemia preventing milk fever. J Dairy Sci 74:3863–3871 12. Remer T, Manz F (1995) Potential renal acid load of foods and its influence on urine pH. J Am Diet Assoc 95:791–797 13. Ganong W (2001) Regulation of extracellular fluid composition and volume. In: Ganong WF (ed) Review of Medical Physiology. Lange Medical Books-McGraw Hill Medical Publishing Division, New York, pp 704–712 14. Frassetto L, Todd K, Morris RJ, Sebastian A (1998) Estimation of net endogenous noncarbonic acid production in humans from diet potassium and protein contents. Am J Clin Nutr 68:576–583 15. Riond J (2001) Animal nutrition and acid base balance. Eur J Nutr 40:245–254

437

16. Dwyer J, Foulkes E, Evans M, Ausman L (1985) Acid/ Alkaline ash diets: time for assessment and change. J Am Diet Assoc 85:841–845 17. Boyd Eaton S, Cordain L (1997) Evolutionary aspects of diet: old genes, new fuels. World Rev Nutr Diet 81:26–37 18. Bushinsky D (1996) Metabolic alkalosis decreases bone calcium efflux by suppressing osteoclasts and stimulating osteoblasts. Am J Physiol 271:F216–F222 19. Bushinsky D (2001) Chronic acidosis: calcium release. Eur J Nutr 40:240–244 20. Turner AS, Alvis M, Myers W, et al. (1995) Changes in bone mineral density and bone-specific alkaline phosphatase in ovariectomized ewes. Bone 17:395S–402S 21. Turner AS, Alvis M, Mallinckrodt C, Bryant H (1995) Dual-energy x-ray absorptiometry in sheep: experiences with in vivo and in vitro studies. Bone 17(suppl):381S–387S 22. Turner AS, Alvis M, Mallinckrodt C, Bryant H (1995) Dose response effects of estradiol on bone mineral density in ovariectomized ewes. Bone 17:421S–427S 23. Committee on Care and Use of Laboratory Animals of the ILAR (1985) Guide for the Care and Use of Laboratory Animals. US Dept of Health and Human Services, Public Health Service, National Institutes of Health 24. AVMA (1993) Panel on Euthanasia. J Am Vet Med Assoc 202:229–249 25. Littell RC, Ramon C, Milliken GA, Stroup WW (1996) The SAS system for mixed models. SAS Institute Inc., Cary, NC 26. Davey D (1998) Osteoporosis in clinical practice—bone densitometry and fracture risk. S Afr Med J 88:1419–1423 27. Turner AS (2001) Research in orthopedic surgery. In: Surgical Research 28. Turner AS (2001) Animal models of osteoporosis—necessity and limitations. Eur Cells Materials 1:66–81 29. Turner AS (2002) The sheep as a model for osteoporosis in humans. The Vet J 163:1–8 30. Taechakraichana N, Limpaphayom K, Jaisamrarn U, Poshyachinda M (1997) Bone loss in oophorectomized Thai women. J Med Assoc Thai 80:384–390 31. Yildiz A, Sahin I, Gol K, et al. (1996) Bone loss rate in the lumbar spine: a comparison between natural and surgically induced menopause. Int J Gynaecol Obstet 55:153–159 32. Lill CA, Fluegel AK, Schneider E (2000) Sheep model for fracture treatment in osteoporotic bone: a pilot study about different induction regimens. J Orthop Trauma 14:559–565 33. Leung K, Siu W, Cheung N, et al. (2001) Goats as an osteopenic animal model. J Bone Miner Res 16:2348–2355 34. Dalle Carbonare L, Arlot M, Chavassieux P, et al. (2001) Comparison of trabecular bone microarchitecture and remodeling in glucocorticoid-induced and postmenopausal osteoporosis. J Bone Miner Res 16:97–103 35. Barzel U (1995) The skeleton as an ion exchange system: implications for the role of acid-base imbalance in the genesis of osteoporosis. J Bone Miner Res 10:1431–1436 36. Frassetto L, Morris RJ, Sellmeyer D, et al. (2001) Diet, evolution and aging: the pathophysiolgic effects of postagricultural inversion of the potassium-to-sodium and base-to-chloride ratios in the human diet. Eur J Nutr 40:200–213