0021-972X/88/6601-0140$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1988 by The Endocrine Society

Vol. 66, No. 1 Printed in U.S.A.

Relationship of Animal Protein-Rich Diet to Kidney Stone Formation and Calcium Metabolism* NEIL A. BRESLAU, LINDA BRINKLEY, KATHY D. HILL, AND CHARLES Y. C. PAK Center in Mineral Metabolism and Clinical Research, Department of Internal Medicine, and University of Texas Health Science Center, Southwestern Medical School, Dallas, Texas 75235

ABSTRACT. We wished to determine whether different types of dietary protein might have different effects on calcium metabolism and on the propensity for renal stone formation. Fifteen young normal subjects were studied during three 12-day dietary periods during which their diet contained vegetable protein, vegetable and egg protein, or animal protein. While these three diets were constant with respect to Na, K, Ca, P, Mg, and quantity of protein, they had progressively higher sulfur contents. As the fixed acid content of the diets increased, urinary calcium excretion increased from 103 ± 15 (±SEM) mg/day (2.6 ± 0.4 mmol/day) on the vegetarian diet to 150 ± 13 mg/day (3.7 ± 0.3 mmol/day) on the animal protein diet (P < 0.02). Despite the increased urinary calcium excretion, there was a modest reduction of urinary cAMP excretion and serum PTH and 1,25dihydroxyvitamin D levels consistent with acid-induced bone dissolution. There was no change in fractional intestinal 47Ca

absorption. The inability to compensate for the animal proteininduced calciuric response may be a risk factor for the development of osteoporosis. The animal protein-rich diet was associated with the highest excretion of undissociated uric acid due to the reduction in urinary pH. Moreover, citrate excretion was reduced because of the acid load. However, oxalate excretion was lower than during the vegetarian diet [26 ± 1 mg/day (290 ± 10 /xmol/day) vs. 39 ± 2 mg/day (430 ± 20 fimol/day); P < 0.02]. Urinary crystallization studies revealed that the animal protein diet, when its electrolyte composition and quantity of protein were kept the same as for the vegetarian diet, conferred an increased risk for uric acid stones, but, because of opposing factors, not for calcium oxalate or calcium phosphate stones. (J Clin Endocrinol Metab 66:140, 1988)

/CONSIDERABLE evidence has accumulated sugV > gesting that the ingestion of a diet rich in animal protein (meat, fish, poultry, eggs, and milk products) represents a risk for calcium nephrolithiasis. Nephrolithiasis involves principally calcium oxalate stone formation, and particularly affects the affluent in industrialized countries whose animal protein consumption is high (1). Conversely, calcium nephrolithiasis is less common in populations whose dietary protein is primarily vegetable in origin (2). In their analysis of risk factors for calcium stone disease, Robertson et al. (1,2) identified animal protein intake as one of the most important risk factors. They subsequently demonstrated that idiopathic calcium stone formers consumed more animal protein, particularly that derived from meat, fish, and poultry, than did normal subjects (2). This observation was supported by others who ascribed hyperuricosuria in many patients with hyperuricosuric calcium oxalate nephrolithiasis to excessive intake of purine-rich meat products (3, 4). Received June 25,1987. Address requests for reprints to: Dr. Neil A. Breslau, Department of Medicine, University of Texas Health Science Center, 5323 Harry Hines Boulevard, Dallas, Texas 75235. * This work was supported by NIH Grants RO1-AM26253, POlAM20543, MO1-RR-00633, and NASA Grant NAG-9-152

The report upon which this study was largely based was that of Robertson et al. in 1979 (2). In normal subjects, animal protein intake was varied from 1-92 g/ day, while calcium and phosphate intakes were kept constant. Substantial increases in urinary calcium, oxalate, and uric acid occurred. The overall relative probability of forming stones, calculated from risk factors, rose significantly with increasing animal protein ingestion. A number of other studies confirmed that increased dietary protein results in enhanced urinary calcium excretion (5-8). While noteworthy, the above studies left certain issues unresolved. Whether all proteins have a similar effect on urinary calcium excretion or, for example, animal fleshderived proteins have greater effects than vegetarian proteins is not known. In some previous studies, the diets were not constant with respect to fluids, calories, and electrolytes, which may have influenced the results. There is disagreement as to the effect of a high animal protein intake on urinary oxalate excretion (2, 9). Moreover, the greater urinary acidity during such a diet could reduce dissociation of phosphate. A reduction in urinary oxalate or dissociated phosphate may oppose calcium stone formation. The effect of animal protein on stone propensity has not been quantitated by measures of urinary crystallization (e.g. urinary saturation and inhib-

140

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 14:09 For personal use only. No other uses without permission. . All rights reserved.

DIETARY PROTEINS AND MINERAL METABOLISM itor activity), nor has the effect of a meat diet on citrate excretion been examined, although citrate has been recognized as an important inhibitor of calcium stone formation (10). The etiology of the hypercalciuria induced by protein ingestion has not been clearly elucidated; resorptive, renal leak, and hyperabsorptive mechanisms all remain possibilities (7). Simultaneous measurement of serum PTH, 1,25-dihydroxyvitamin D [1,25-(OH)2D], and fractional intestinal 47Ca absorption should distinguish among these causes. If protein-induced hypercalciuria is not accompanied by a compensatory intestinal hyperabsorption of calcium via secondary hyperparathyroidism and stimulation of PTH-dependent 1,25-(OH)2D synthesis, then negative calcium balance may ensue, which could pose a risk for osteoporosis (11, 12). This study was designed to examine these issues.

Materials and methods Experimental subjects Fifteen normal adults (eight women and seven men) were studied. The subjects ranged in age from 23-46 yr [mean, 29.9 ± 5.9 (±SD) yr]. Their weight ranged from 51.0-92.0 kg, averaging 68.5 ± 13.0 kg. Thirteen of the subjects were white, one was black, and one was Latin. All had normal creatinine clearance, ranging from 75-139 mL/min (mean, 101 ± 20 mL/min). None of the subjects was taking any medication. Each had a normal physical examination and routine laboratory screen (systematic multichannel analysis-20 and complete peripheral blood count). The study was approved by the Human Experimentation Committee, and informed consent was obtained from each individual. Study protocol The study was conducted in three consecutive phases, the order of which was randomized by drawing lots. Each phase lasted 12 days. During each phase, the diet was kept constant with respect to calcium (400 mg or 10.0 mmol/day), phosphorus (1000 mg or 32.3 mmol/day), sodium (100 mmol/day), and total protein (75 g/day). However, protein was provided as animal protein during one phase and as soy-based vegetable protein with eggs during a second phase (ovo-vegetarian). Ten of the subjects participated in a third phase in which the diet was

141

similar to the vegetable protein diet except that eggs, which have a high sulfate content, were eliminated (vegetarian). The diets were isocaloric, providing sufficient kilocalories to maintain the subject's weight. For each diet, protein constituted 1520%, fat 35-40%, and carbohydrate 40-45% of the total caloric intake. The compositions of the animal protein and ovovegetarian diets are shown in Table 1. In the completely vegetarian diet, additional soy protein replaced eggs. During the first 6 days of the diet, the subjects prepared and ate the appropriate instructed diet at home. During the last 6 days, they ate metabolic diets provided by the Clinical Research Center. The latter diets were analyzed in the laboratory, and mineral salts were added as necessary to equalize the electrolyte content during each phase (Table 2). The dietary contents of purines, oxalate. sulfate. and fiber were estimated from food tables (Table 2). Fluid intake was kept constant at 3000 mL distilled water daily. Activity was ad libitum, but excessive exercise during the study was discouraged. During the last 4 days of each phase (days 9-12) fasting morning blood and 24-h urine samples were collected. Serum measurements included calcium, phosphorus, magnesium, uric acid, creatinine, total protein, albumin, immunoreactive PTH, 25-hydroxyvitamin D (25OHD), and 1,25-(OH)2D. Urine samples were analyzed for total volume, pH, creatinine, sodium, potassium, calcium, magnesium, phosphate, uric acid, oxalate, ammonium, titratable acidity, carbon dioxide (collected under oil), citrate, and cAMP. Urinary crystallization studies of stoneforming salts were performed, including relative saturation ratio (RSR) of brushite (CaHPO4-2H2O), calcium oxalate, and monosodium urate (13, 14); undissociated uric acid (15); and formation product (FP) of brushite and calcium oxalate (16). On day 12, fractional intestinal calcium absorption was measured by the recovery of 47Ca in the feces after oral administration of the isotope (17). 47Ca was given orally in 100 mg carrier (Calcitest, Doyle Pharmaceutical Co., Minneapolis, MN), which replaced the breakfast meal; food was withheld for 4 h thereafter. Stool was collected until the disappearance of carmine, which was given 24 h after isotope administration. The fecal radioactivity was corrected for the recovery of polyethylene glycol, which was given with the isotope. Analytical procedures Calcium and magnesium were determined by atomic absorption spectrophotometry, and phosphorus by the method of Fiske and Subbarow (18). Serum creatinine, uric acid, total

TABLE 1. Compositions of animal protein and ovo-vegetarian protein diets Food group Milk group (2 or more servings)

Animal protein diet

Meat group (2 or more servings) Vegetable protein group (2 or more servings)

Cheese in amounts necessary to achieve 400 mg Ca/day Beef, chicken, fish None

Fruit and vegetable group: (4 or more servings) Bread and cereals group (4 or more servings) Other foods Eggs (3-4/week) Fats (1 tablespoon/day)

As needed for Cal

Ovo-vegetarian diet

Soybean milk and cheese in amounts necessary to achieve 400 mg Ca/day None Textured vegetable protein (soy protein) Low sodium canned vegetables and "dietetic" canned fruits and fruit juices Low sodium sliced bread, cooked cereals, starches and crackers

As needed for Cal

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 14:09 For personal use only. No other uses without permission. . All rights reserved.

BRESLAU ET AL

142

JCE & M • 1988 Vol 66 • No 1

TABLE 2. Contents of the various protein diets Vegetarian A. Directly measured Ca [mg/day (mmol/day)] Mg [mg/day (mmol/day)] P [mg/day (mmol/day)] Na (mmol/day) K (mmol/day) B. Estimated from food tables Purines (mg/day) Oxalate [mg/day (^mol/day)] Sulfate (mmol/day) Fiber(g/day)

440 ± 19 (11.0 ± 0.5) 370 ± 20 (15.2 ± 0.8) 1017 ± 34 (32.8 ± 1.1) 90 ± 9 80 ± 3 l±0

392 ± 18 (4360 d: 200)6 23 ± 1"

63 ±3*

Animal

Ovo-vegetarian 441 ± 13 (11.0 ± 0.3) 353 ± 16 (14.5 ± 0.7) 1051 ± 23 (33.9 ± 0.7) 93 ± 6 74 ± 3 2±1

326 ± 10 (3620 ± 110)" 29 ± 2 " 54 ±3°

447 ± 11 (11.2 ± 0.3) 318 ± 25 (13.1 ± 1.0) 1016 ± 32 (32.8 ± 1.0) 92 ± 7 72 ± 3 72 ± 6ab 188 ± 14 (2090 ± 160)"'6 33 ± 2°-b 13 ± 2°'fc

Values are the mean ± SEM. Because of the addition of appropriate mineral salts, the electrolyte content of the various diets did not differ. " Significantly different from vegetarian, P < 0.001. 6 Significantly different from ovo-vegetarian, P < 0.001.

protein, and albumin were measured as part of the systematic multichannel analysis (SMA-20). Serum PTH was assayed in duplicate using midmolecule assay kits obtained from Endocrine Metabolic Center (Oakland, CA) (12). The intra- and interassay coefficients of variation were 6% and 13%, respectively. Vitamin D metabolites were extracted with ether and separated by column chromatography, and the column fractions were assayed in triplicate. Serum 25OHD levels were assayed by the rat serum binding protein method (17). Serum 1,25(OH)2D levels were measured by radioreceptor assay, using chick intestinal cytosolic receptor protein (12, 17). The intraand interassay variations were 8% and 15%, respectively. Urinary sodium and potassium were measured by flame photometry. Oxalate was measured by the method of Hodgkinson and Williams (19), and uric acid by the enzymatic method of Liddle et al. (20). Citrate was determined enzymatically with kits obtained from Boehringer-Mannheim Biochemicals (Indianapolis, IN), in which the intra- and interassay variations were 3% and 6%, respectively. Ammonium was measured colorimetrically by a method adopted from that of Chaney and Marbach (21). Urine pH and carbon dioxide tension were determined anaerobically at 37 C (Radiometer BMS-3 pH electrode, Radiometer America Inc., Westlake, OH and Severinghaus PCO2 electrode Instrumentation Laboratory, Lexington, MA). Urinary net acid excretion was determined by subtracting the urinary bicarbonate level from the sum of ammonium plus titratable acidity, each expressed in millimoles per day. Urinary sulfate was determined by the turbidimetric procedure of Ma and Chan (22). Urinary cAMP was measured by the competitive protein-binding assay of Gilman (23).

nuclei in urine. To samples rendered devoid of cellular debris and crystalline material by filtration (Millipore 0.22 m, Millipore Corp, Bedford, MA), a solution of sodium oxalate was added in increasing amounts to cause precipitation of calciun. oxalate. The minimum activity product of calcium oxalate at which spontaneous precipitation of calcium oxalate occurred represented the FP. Thus, a rising FP value reflected increased inhibition of crystallization of calcium oxalate. For brushite, the urine sample was rendered increasingly supersaturated with respect to calcium phosphate by adding a solution of calcium chloride. The FP was determined as for calcium oxalate. Statistical analysis The paired t test was used to test the significance of changes in measured parameters during the various dietary phases. Since there were three comparisons, the Bonferroni correction was applied, and differences were considered significant only at P < 0.02 (25). Correlations were performed by the method of least squares. Data organization and analysis assistance was provided by the CLINFO Project at University of Texas Health Science Center (Dallas, TX).

Results Effects of various protein diets on serum chemistries The serum calcium, phosphorus, magnesium, creatinine, total protein, and albumin concentrations were similar during the various protein diets (Table 3). However, the serum uric acid concentration was significantly higher during the animal protein diet (P < 0.01).

Crystallization studies Methods for the assessment of urinary crystallization were reported previously (13-15). The RSR was calculated using the computer program of Finlayson (24). A value of 1 indicated saturation; more than 1, supersaturation; and less than 1, undersaturation (13). The amount of undissociated uric acid was also calculated using the same computer program (lo). The FP indicated the limit of metastability or minimum supersaturation required for spontaneous nucleation (16). Strictly, it represented heterogeneous nucleation by naturally occurring

Effects of various protein diets on urinary chemistries The urinary chemistry data from each of the dietary phases are shown in Table 4. As the dietary sulfate content increased (from vegetarian to animal protein diet), there was a progressive rise in urinary sulfate excretion and net acid excretion, with a corresponding decrease in urinary pH and rise in urinary ammonium. Concordant with the high acid ash intake during the

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 14:09 For personal use only. No other uses without permission. . All rights reserved.

DIETARY PROTEINS AND MINERAL METABOLISM animal protein diet, urinary calcium excretion increased, whereas urinary citrate excretion declined. Urinary calcium excretion directly correlated with net acid excretion (r = 0.37; P = 0.02). However, there was no correlation between urinary calcium excretion and estimated dietary fiber content. Urinary uric acid excretion during the animal protein diet was significantly higher than that during other phases. Despite the similar phosphate content of each diet, urinary phosphate excretion was highest during the animal protein diet. The fractional excretion of phosphate during the animal protein diet was 0.15 ± 0.01, compared to 0.12 ± 0.01 during the ovovegetarian diet and 0.10 ± 0.01 during the vegetarian diet (P < 0.01). Moreover, the renal threshold for phosphate excretion (26) was reduced during the animal protein diet [3.1 ± 0.2 mg/dL glomerular filtrate (GF)], compared to that during either the ovo-vegetarian diet (3.4 ± 0 . 1 mg/dL GF) or the vegetarian diet (3.7 ± 0.1 mg/dL GF, P < 0.01 for each). Urinary oxalate excretion was highest during the vegetarian diet (39 ± 2 mg/day or 430 ± 20 /xmol/day) compared to only 26 ± 1 mg/day (290 ± 1 0 /imol/day) during the animal protein diet (P < 0.02). Urinary vol-

143

ume, sodium, potassium, and magnesium did not differ during any of the dietary phases. Endogenous creatinine clearance was higher during the animal protein diet (101 ± 5 mL/min) than during the vegetarian diet (93 ± 5 mL/min; P < 0.03). Effects of various protein diets on parathyroid-vitamin D axis and intestinal calcium absorption (Table 5) During the animal protein diet, there was a tendency toward diminished parathyroid function. This was reflected in the lower serum PTH level (P < 0.05) and urinary cAMP excretion (P < 0.01) during the animal protein diet. The serum 25OHD concentration was relatively constant during each of the dietary periods, but the 1,25-(OH)2D concentration was lower during the meat diet (74 ± 1 vs. 98 ± 10 pmol/L on the vegetarian diet; P < 0.01). There were no differences in fractional intestinal calcium absorption during the various diets. Effects of various protein diets on urinary crystallization The urinary crystallization data are shown in Table 6. The RSR (urinary saturation) of calcium oxalate, brushite, and monosodium urate did not change significantly

TABLE 3. Serum chemistry during various protein diets Vegetarian Ca [mg/dL (mmol/L)] P [mg/dL (mmol/L)] Mg [mg/dL (mmol/L)] Uric acid [mg/dL Creatinine [mg/dL (/anol/L)] Total protein [g/dL (g/L)] Albumin [g/dL (g/L)]

9.3 ± 3.8 ± 1.9 ± 5.1 ± 0.93 ± 6.8 ± 4.5 ±

0.1 (2.32 ± 0.02) 0.1 (1.23 ± 0.03) 0.1 (0.78 ± 0.04) 0.3 (300 ± 20) 0.05 (80 ± 4) 0.1 (68 ± 1) 0.1 (45 ± 1)

Ovo-vegetarian 9.6 ± 0.1 (2.40 ± 0.02) 3.6 ± 0.1 (1.16 ± 0.03) 1.9 ± 0 (0.78 ± 0) 5.3 ± 0.2 (320 ± 10) 0.96 ± 0.04 (80 ± 4) 6.8 ± 0.1 (68 ± 1) 4.5 ± 0.1 (45 ± 1)

Animal 9.5 ± 3.5 ± 1.9 ± 5.5 ± 1.00 ± 6.8 ± 4.4 ±

0.1 (2.37 ± 0.02) 0.1 (1.13 ± 0.03) 0 (0.78 ± 0) 0.3 (330 ± 20)" 0.04 (90 ± 4) 0.1 (68 ± 1) 0.1 (44 ± 1)

Values are the mean ± SEM of values in each subject on days 9-12 of the indicated diet. " P < 0.01 compared to ovo-vegetarian. TABLE 4. Effect of type of dietary protein on urinary chemistry

Vol (L/day) Creatinine [g/day (mmol/day)] Sulfate (mmol/day) NAEC (mmol/day) pH Ammonium (mmol/day) Citrate [mg/day (mmol/day)] Ca [mg/day (mmol/day)] Uric acid [mg/day (mmol/day)] P [mg/day (mmol/day)] Oxalate [mg/day (/zmol/day)] Na (mmol/day) K (mmol/day) Mg [mg/day (mmol/day)]

Vegetarian

Ovo-vegetarian

Animal

2.58 ± 0.18 1.33 ±0.13 (11.8 ±1.1) 13.4 ± 1.5 11.9 ± 0.7° 6.55 ± 0.05° 22 ± 1 805 ± 85 (4.3 ± 0.4) 103 ± 15 (2.6 ± 0.4) 479 ± 28 (2.8 ± 0.2) 545 ± 42 (17.6 ± 1.4) 39 ± 2 (430 ± 20) 61 ± 8 59 ± 3 90 ± 6 (3.7 ± 0.2)

2.58 ± 0.11 1.32 ± 0.09 (11.7 ± 0.8) 17.6 ± 1.2 26.0 ± 1.5* 6.32 ± 0.07* 28 ± 2 781 ± 73 (4.1 ± 0.4) 121 ± 12 (3.0 ± 0.3) 490 ± 18 (2.9 ± 0.1) 621 ± 27 (20.1 ± 0.9) 34 ± 2 (380 ± 20) 65 ± 5 55 ± 3 101 ± 5 (4.2 ± 0.2)

2.59 ± 0.12 1.47 ± 0.09 (13.0 ± 0.8)° 20.2 ± 1.3a>* 39.0 ± 1.5°'* 6.17 ± 0.07* 32 ± 3 6 693 ± 66 (3.7 ± 0.3)° 150 ± 13 (3.7 ± 0.3)°'fc 564 ± 26 (3.4 ± 0.2)° 738 ± 35 (23.8 ± 1.1)°'* 26 ± 1 (290 ± 10) °'6 65 ± 1 1 58 ± 5 102 ± 6 (4.2 ± 0.2)

Values are the mean ± SEM of values in each subject on days 9-12 of indicated diet. Significantly different from ovo-vegetarian, P < 0.02. 6 Significantly different from vegetarian, P < 0.02. c Net acid excretion, which is calculated as the sum of urinary ammonium and titratable acidity minus urinary bicarbonate. 0

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 14:09 For personal use only. No other uses without permission. . All rights reserved.

BRESLAU ET AL.

144

JCE & M • 1988 Vol66«Nol

TABLE 5. PTH-vitamin D axis during various protein diets

PTH (ng/L) 24-h Urinary cAMP (nmol/100 mL GF) 25-OHD [ng/mL (nmol/L)] 1,25-(OH)2D [pg/mL (pmol/L)] Intestinal 47Ca absorption fraction

Vegetarian

Ovo-vegetarian

Animal

280 ± 29 3.69 ± 0.18 31 ± 5 (77 ± 12) 41 ± 4 (98 ± 10) 0.48 ± 0.04

287 ± 21 3.97 ± 0.20 33 ± 4 (82 ± 10) 35 ± 4 (84 ± 10)° 0.45 ± 0.03

260 ± 27" 3.28 ± 0.176 34 ± 5 (85 ± 12) 31 ± 3 (74 ± 7)* 0.45 ± 0.03

Values are the mean ± SEM of values in each subject on days 9-12 of the indicated diet (except 47Ca absorption). Significantly different from vegetarian, P < 0.05. 6 Significantly different from vegetarian, P < 0.01; significantly different from ovo-vegetarian, P < 0.01.

0

TABLE 6. Urinary crystallization data during various protein diets Ovo-vegetarian

Vegetarian RSR Calcium oxalate Brushite Monosodium urate Undissociated uric acid [mg/L (junol/L)] FP Calcium oxalate (xlO"8) Brushite (X1O~7)

Animal

3.85 ± 0.77 0.77 ± 0.19 0.63 ± 0.11

3.84 ± 0.47 0.89 ± 0.23 0.64 ± 0.07

3.60 ± 0.42 0.92 ± 0.18 0.74 ± 0.16

13.2 ± 1.9 (80 ± 10)°

23.2 ± 3.9 (140 ± 20)°

35.9 ± 4.6 (210 ± 30)°

3.63 ± 0.25 8.75 ± 0.82

3.41 ± 0.16 8.58 ± 0.85

3.53 ± 0.19 8.83 ± 1.01

Values are the mean ± SEM of values in each subject on days 9-12 of the indicated diet. None of the differences in RSR or FP was significant. ° The urinary undissociated uric acid level varied significantly with each diet (P < 0.02).

during any of the diet periods. The urinary concentration of undissociated uric acid progressively increased as the diet was switched from vegetarian to ovo-vegetarian to animal protein. Compared to the vegetarian diet, urinary undissociated uric acid excretion was nearly 3-fold higher during the animal protein diet. There was no change in the FP of either calcium oxalate or brushite during any of the diets, indicating no change in inhibitor activity.

Discussion We studied the metabolic effects of diets containing identical amounts of protein, but in which the protein was either of animal or vegetable origin. All other aspects of the diet were kept constant, including fluid and caloric intake and calcium, phosphate, magnesium, sodium, and potassium content. As estimated from dietary tables, the animal protein diet had a higher content of purines and sulfate, whereas the vegetarian diets contained more oxalate and fiber. A high protein intake is associated with increased urinary calcium excretion (5-9). In normal subjects, increasing dietary protein by 75 g daily raised urinary calcium excretion by approximately 100 mg/day (6-8). More recently, evidence has emerged that different sources of protein may have varying effects on urinary calcium excretion. Using different types of protein, Whiting and Draper (27) found that in adult rats, the degree of hypercalciuria was proportional to the sulfur content of the diets. In vivo, sulfur is oxidized to sulfate, which generates a fixed acid load that is buffered by bone (28,

29). This participation of bone mineral in defense against chronic metabolic acidosis results in bone dissolution and contributes to protein-induced hypercalciuria. The sulfur in protein is found mainly in the sulfur-containing amino acids cystine and methionine, which are present in greater quantity in animal-derived proteins than in vegetable proteins (9, 30). The content of these amino acids is from 2- to 5-fold higher in meat and eggs than in grains and beans (9, 30). Brockis et a/. (9) compared two normal populations (n = 30), 1 meat-eating, the other lacto-ovo-vegetarian, and found that animal protein was associated with greater urinary calcium excretion. In 42 healthy individuals, Tschope and Ritz (30) found a correlation between urinary sulfate and calcium excretion. They also found that addition of 6 g L-methionine to the diet raised urinary calcium excretion by 80 mg/24 h. In our study, switching from vegetarian to ovovegetarian to animal protein diets caused a progressive increase in urinary sulfate and net acid excretion, which was accompanied by a rise in urinary calcium excretion. The urinary calcium excretion correlated with the net acid excretion. In addition to the buffering action of bone, the increased acid load resulting from a diet rich in animal protein may cause hypercalciuria by other mechanisms. Chronic metabolic acidosis may decrease renal tubular calcium reabsorption (7, 8, 31, 32). Such a diet may increase intestinal calcium absorption, thus contributing to hypercalciuria (7). During the animal protein diet, we found evidence for modest suppression of the parathyroid-vitamin D axis, but no change in intestinal calcium

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 14:09 For personal use only. No other uses without permission. . All rights reserved.

DIETARY PROTEINS AND MINERAL METABOLISM absorption. Chronic acid loading decreases serum 1,25(OH)2D concentrations in rats, in association with a rise in serum ionized calcium (33). Previous human studies have indicated failure of the PTH-vitamin D axis to compensate for protein-induced hypercalciuria (6, 8), but suppression was not found. Our results suggest that the major calciuric factor during animal protein ingestion is bone resorption in response to the acid load. Viewed another way, one might attribute the lower calcium excretion and relative parathyroid stimulation during the vegetarian diet to intestinal binding of calcium by the increased dietary fiber (34, 35). This would not have been detected by the fractional 47Ca absorption test because it was performed after an overnight fast using a standard synthetic meal. However, in this study, urinary calcium excretion did not correlate with dietary fiber intake. Since there is no compensation for the bone-resorptive urinary calcium losses that occur during high animal protein intake, a negative calcium balance ensues (5, 8). This contrasts with the ability of young individuals to adapt to salt-induced urinary calcium losses by raising serum PTH and 1,25-(OH)2D levels, thereby permitting a compensatory rise in intestinal calcium absorption (11). A chronic acid load may result in bone loss. Chronic ammonium chloride ingestion can cause increased bone resorption and osteoporosis in rats (36), and cortical bone density in omnivores is lower than that in age- and sex-matched vegetarians (37, 38). Thus, a predominantly meat diet, with its high acid ash content and associated uncompensated hypercalciuria, may represent a risk for the development of osteoporosis. A diet rich in animal protein also presents several risk factors for the formation of kidney stones. In addition to hypercalciuria, the animal protein diet was associated with the greatest urinary excretion of uric acid. This is not surprising, since animal flesh has a high purine content. In addition, the high acid ash content of the meat diet caused acidification of the urine, which raised the undissociated uric acid concentration. This urinary environment would favor precipitation of uric acid crystals and would represent a particular risk for patients with gouty diathesis or uric acid stones (3, 4, 39). The acid load resulting from the animal protein diet also was associated with a decline in urinary citrate excretion. Reduced urinary citrate excretion is a well known occurrence in renal tubular acidosis or acquired metabolic acidosis (40, 41), probably due to the enhanced renal tubular reabsorption of citrate. A reduction in urinary citrate would be expected to enhance the crystallization of calcium oxalate and calcium phosphate, since citrate is an inhibitor of the crystallization of these salts (10, 42). The animal protein diet was associated with increased urinary phosphate excretion, which has been reported to occur during chronic metabolic acidosis (31). In one respect, the animal protein-rich diet posed less of

145

a risk for stones. In agreement with the finding of Brockis et al. (9), but differing from that of Robertson et al. (2), we found that urinary oxalate excretion was highest during the vegetarian diets, corresponding to the higher estimated oxalate content of the vegetarian diets. Despite the enhanced excretion of certain stoneforming constituents during the animal protein diet, several opposing factors limited the propensity for crystallization of these stone-forming salts. Despite the increased urinary calcium excretion, the reduced oxalate excretion prevented any increase in the RSR of calcium oxalate. Because of greater urinary acidity during the animal protein diet, there was less dissociation of phosphate, and hence, the saturation of brushite was not increased. With greater meat intake, urinary uric acid levels increased, but because of the acidity, more of the uric acid was undissociated, so that the saturation of monosodium urate was not increased. Monosodium urate may serve as a nidus for calcium stones (14). Despite the fall in citrate excretion, the limit of metastability (FP) of calcium oxalate and brushite did not decline during the animal protein diet. Apparently, the reduction in citrate excretion was of insufficient magnitude to facilitate nucleation of the calcium salts. Therefore, under the conditions of this study, a diet rich in animal protein appears to pose a risk for the formation of uric acid stones, but not for calcium stones. Another cause of the increased incidence of calcium stones in recent decades should be sought.

Acknowledgments The following individuals deserve special thanks for helping to assemble this data: Marjorie Rutledge, R.D.; Lalena Farrell, R.N.; Khashayar Sakhaee, M.D.; Alan Stewart, M.B.A.; and Beverley Adams, M.S., CLINFO Specialist. We also appreciate the excellent secretarial support by Betty Bousselot.

References 1. Robertson WG, Peacock M, Hodgkinson A 1979 Dietary changes and the incidence of urinary calculi in the U.K. between 1958 and 1976. J Chron Dis 32:469 2. Robertson WG, Peacock M, Heyburn PJ, Hanes FA, Rutherford A, Clementson E, Swaminathan R, Clark PB 1979 Should recurrent calcium oxalate stone formers become vegetarians? Br J Urol 51:427 3. Coe FL, Moran E, Kavalich A 1976 The contribution of dietary purine over-consumption to hyperuricosuria in calcium oxalate stone-formers. J Chron Dis 29:793 4. Pak CYC, Barilla DE, Holt K, Brinkley L, Tolentino R, Zerwekh JE 1978 Effect of oral purine load and allopurinol on the crystallization of calcium salts in urine of patients with hyperuricosuric calcium urolithiasis. Am J Med 65:593 5. Linkswiler HM, Joyce CL, Anand CR 1974 Calcium retention of young adult males as affected by level of protein and of calcium intake. Trans NY Acad Sci 36:333 6. Adams ND, Gray RW, Lemann J Jr 1979 The calciuria of increased fixed acid production in humans: evidence against a role for parathyroid hormone and l,25-(OH)2-vitamin D. Calcif Tissue Int 28:233 7. Licata AA, Bou E, Bartter FC, Cox J 1979 Effects of dietary protein

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 14:09 For personal use only. No other uses without permission. . All rights reserved.

146

BRESLAU ET AL.

on urinary calcium in normal subjects and in patients with nephrolithiasis. Metabolism 28:895 8. Schuette SA, Zemel MB, Linkswiler HM 1980 Studies on the mechanism of protein-induced hypercalciuria in older men and women. J Nutr 110:305 9. Brockis JG, Levitt AJ, Cruthers SM 1982 The effects of vegetable and animal protein diets on calcium, urate and oxalate excretion. Br J Urol 54:590 10. Nicar MJ, Hill K, Pak CYC 1987 Inhibition by citrate of spontaneous precipitation of calcium oxalate, in vitro. J Bone Mineral Res 2:215 11. Breslau NA, McGuire J, Zerwekh JE, Pak CYC 1982 The role of dietary sodium on renal excretion and intestinal absorption of calcium and vitamin D metabolism. J Clin Endocrinol Metab 55:369 12. Breslau NA, Sakhaee K, Pak CYC 1985 Impaired adaptation to salt-induced urinary calcium loases in postmenopausal osteoporosis. Trans Assoc Am Physicians 98:107 13. Pak CYC, Hayashi Y, Finlayson B, Chu S 1977 Estimation of the state of saturation of brushite and calcium oxalate in urine: a comparison of three methods. J Lab Clin Med 89:891 14. Pak CYC, Waters O, Arnold L, Cox C, Barilla D 1977 Mechanism for calcium urolithiasis among patients with hyperuricosuria: supersaturation of urine with respect to monosodium urate. J Clin Invest 59:426 15. Pak CYC, Sakhaee K, Fuller C 1986 Successful management of uric acid nephrolithiasis with potassium citrate. Kidney Int 30:422 16. Nicar MJ, Hill K, Pak CYC 1983 A simple technique for the assessment of the propensity for the crystallization of calcium oxalate and brushite in urine from the increment in oxalate or calcium necessary to elicit precipitation. Metabolism 32:906 17. Breslau NA, McGuire JL, Zerwekh JE, Frenkel EP, Pak CYC 1984 Hypercalcemia associated with increased serum calcitriol levels in three patients with lymphoma. Ann Intern Med 100:1 18. Fiske CH, Subbarow Y 1925 The colorimetric determination of phosphorus. J Biol Chem 66:375 19. Hodgkinson A, Williams A 1972 An improved colorimetric procedure for urine oxalate. Clin Chim Acta 36:127 20. Liddle L, Seegmiller, Laster L 1959 The enzymatic spectrophotometer method for determination of uric acid. J Lab Clin Med 54:903 21. Chaney AL, Marbach EP 1962 Modified reagents for determinations of urea and ammonia. Clin Chim 8:130 22. Ma A, Chan JCM 1973 Endogenous sulfuric acid production: a method of measurement by extrapolation. Clin Biochem 6:82 23. Gilman AG 1970 A protein binding assay for adenosine 3':5'-cyclic monophosphate. Proc Natl Acad Sci USA 67:305 24. Finlayson B 1977 Calcium stones: some physical aspects. In David DS (ed) Calcium Metabolism in Renal Failure and Nephrolithiasis. Wiley and Sons, New York, p 337

JCE & M • 1988 Vol66»Nol

25. Glantz SA 1980 Biostatistics: how to detect, correct and prevent errors in the medical literature. Circulation 61:1 26. Walton RJ, Bijvoet OLM 1975 Nomogram for derivation of renal threshold phosphate concentration. Lancet 2:309 27. Whiting SJ, Draper HH 1980 The role of sulfate in the calciuria of high protein diets in adult rats. J Nutr 110:212 28. Lemann Jr J, Adams ND, Gray RW1979 Urinary calcium excretion in human beings. N Engl J Med 301:535 29. Lemann Jr J, Litzow JR, Lennon EJ 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 30. Tschope W, Ritz E 1985 Sulfur-containing amino acids are a major determinant of urinary calcium. Mineral Electrolyte Metab 11:137 31. Lemann Jr J, Litzow JR, Lennon EJ 1967 Studies of the mechanism by which chronic metabolic acidosis augments urinary calcium excretion in man. J Clin Invest 46:1318 32. Allen LH, Bartlett RS, Block GD 1979 Reduction of renal calcium reabsorption in man by consumption of dietary protein. J Nutr 109:1345

33. Bushinsky DA, Riera GS, Favus MJ, Coe FL 1985 Response of serum 1,25-(OH)2D3 to variation of ionized calcium during chronic acidosis. Am J Physiol 249:F361 34. Ohkawa T, Ebisuno S, Kitagawa M, Shigeyoshi M, Miyazaki Y 1983 Rice bran treatment for hypercalciuric patients with urinary calculous disease. J Urol 129:1009 35. Dandona P, Mohiuddin J, Weerakoon JW, Freedman DB, Fonseca V, Healey T 1984 Persistence of parathyroid hypersecretion after vitamin D treatment in Asian vegetarians. J Clin Endocrinol Metab 59:535 36. Barzel US 1969 The effect of excessive acid feeding on bone. Calc Tissue Res 4:94 37. Ellis FR, Holesh S, Ellis JW 1972 Incidence of osteoporosis in vegetarians and omnivores. Am J Clin Nutr 25:555 38. Sanchez TV, Mickelsen O, Marsh AG, Gam SM, Mayor GH 1980 Bone mineral in elderly vegetarian and omnivorous females. In Mazess RB (ed) Proceedings of the Fourth International Conference on Bone Measurement. NIH, Bethesda, p 94 39. Smith LH, VanDenBerg CJ, Wilson DM 1978 Current concepts in nutrition—Nutrition and urolithiasis. N Engl J Med 298:87 40. Preminger GM, Sakhaee K, Skurla C, Pak CYC 1985 Prevention of recurrent calcium stone formation with potassium citrate therapy in patients with distal renal tubular acidosis. J Urol 134:20 41. Rudman D, Dedonis JL, Fountain MT, Chandler JB, Gerron GG, Fleming GA, Kutner MH 1980 Hypocitraturia in patients with gastrointestinal malabsorption. N Engl J Med 302:657 42. Pak CYC, Fuller C, Sakhaee K, Preminger G, Britton F1985 Longterm treatment of calcium nephrolithiasis with potassium citrate. J Urol 134:11

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 14:09 For personal use only. No other uses without permission. . All rights reserved.