If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Individual reprints of this article are not available. Address correspondence to Robert P. Heaney, MD, Creighton University Medical Center, 601 N 30th St, Suite 4841, Omaha, NE 68131
Bone mineral consists of calcium phosphate, and phosphorus is as important as calcium in supporting bone augmentation and maintenance. Although typical adult diets contain abundant phosphorus, 10% to 15% of older women have intakes of less than 70% of the recommended daily allowance. When these women take high-dose calcium supplements that consist of the carbonate or citrate salts, all their food phosphorus may be bound and hence unavailable for absorption. Current-generation anabolic agents for treating osteoporosis require positive phosphorus balances of up to 90 mg/d. Attention to the nutritional adequacy of the diets of such patients is essential if they are to realize the full potential of such therapies. A calcium phosphate supplement may be preferable to the usual carbonate or citrate salts because its phosphate serves to spare food phosphorus.
Phosphorus is, in some sense, a pariah among the nutrients. Phosphorus intake is an important problem for patients with end-stage renal disease, and phosphorus retention is the most important factor in extraosseous calcification in patients with renal failure. Although this concern is irrelevant to individuals with adequate kidney function, it has tended to color medicine's approach to the nutrient. Specifically, phosphorus has been indicted in the genesis of osteoporosis,
Interestingly, orange juice has essentially the same phosphorus content as colas, and in countries with calcium fortification of orange juice, the phosphorus content may rise to 5 times that of a typical cola. To my knowledge, there has been no concern raised about the harmful effects of orange juice on the skeleton.
The pertinent facts about phosphorus are that (1) bone mineral is not just calcium but specifically calcium phosphate; (2) adequate quantities of ingested phosphorus (expressed physiologically as a serum phosphate concentration of 1.5-2.0 mmol/L) are essential for bone building during growth
; and (3) hypophosphatemia, from whatever cause, limits mineralization at new bone-forming sites at all ages, impairs osteoblast function, and enhances osteoclastic resorption.
Additionally, ingested phosphorus in the adult years alters certain aspects of the operation of the calcium economy. Phosphorus lowers urinary calcium and elevates endogenous fecal calcium loss.
and phosphorus supplements have been used therapeutically to reduce renal stone formation in patients with hypercalciuria.
Phosphorus is a limiting nutrient in the biosphere. Most phosphorus is tied up in protoplasm or guano deposits. Phosphorus is vital for cellular life and activity, being involved in cell structure, information coding, energy transfer, functional activation of catalytic and signaling proteins, and, in short, virtually every aspect of cell function. Animals get the phosphorus they need by ingesting the protoplasm of other organisms (animal or plant), and plants, by taking up into their roots phosphorus released during decay of other organisms into the superficial soil layers of the biosphere. Phosphorus availability limits biomass in both aquatic and terrestrial habitats.
As would be expected for a limiting nutrient, phosphorus absorption by the human intestine is relatively efficient, with net absorption typically ranging from 55% to 80% (depending on concurrent calcium intake and absorption). Moreover, because phosphorus is distributed widely in many foods, it is not likely to function as a limiting nutrient (as is calcium) in animals at the top of the food chain. Nevertheless, circumstances exist in which phosphorus intake may be insufficient to support the bone rebuilding that today may be possible in patients with osteoporosis. In this brief review, I identify such circumstances and address the concerns sometimes raised about phosphorus intake.
LOW PHOSPHORUS INTAKE
The values that define various percentiles of phosphorus intake for non-Hispanic US white women aged 30 to older than 80 years, as derived from the National Health and Nutrition Examination Survey III, are shown in Table 1.
Dietary Intake of Vitamins, Minerals, and Fiber of Persons Ages 2 Months and Over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988-91. National Center for Health Statistics,
Hyattsville, Md1994
The median intake for all ages is above that level, and substantially so for younger women (Table 1); in fact, 10% to 15% of women of all ages ingest more than twice the RDA. (Such relatively generous intakes, over most age strata, certainly place phosphorus in a category different from that of calcium, in which the bulk of the population has an intake below recommended levels.) However, as can be seen in Table 1, 5% of women older than 30 years, 10% of women older than 60 years, and 15% of women older than 80 years have intakes on any given day that are less than 70% of the RDA. This group of older women, often living alone and eating poorly, is more likely to have osteoporosis than others in their age cohort, and hence their phosphorus intakes should be of concern in the management of the disorder.
Table 1Distribution of Phosphorus Intakes by Age and Percentiles in Non-Hispanic White Women in the United States
Dietary Intake of Vitamins, Minerals, and Fiber of Persons Ages 2 Months and Over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988-91. National Center for Health Statistics,
Hyattsville, Md1994
Even a rudimentary knowledge of food composition will suffice to show that phosphorus intakes below 70% of the RDA will necessarily be low in other key nutrients as well, particularly calcium and protein, both of which are critical for bone maintenance and repair. Hence, the diets of women with low phosphorus intakes are multiply inadequate. The correct solution to this problem is for health professionals first to recognize that the problem exists and then to address it, with either a change in diet or a polyvalent nutritional supplement. Unfortunately, this is not often done in medical practice, and frequently these women, particularly if they have osteoporosis, will be prescribed bone active agents and encouraged to take supplements of calcium and vitamin D. Because of their market dominance, these supplements will usually be the carbonate or citrate salt.
This triad of low phosphorus intake, osteoporosis, and calcium supplementation is shown schematically in Figure 1; also illustrated is the intersection of the 3 sets, where, manifestly, the likelihood is highest that phosphorus intake will be limiting. Bone active therapy is a factor in this context because, precisely to the extent that it may be effective, such therapy increases the phosphorus requirement and aggravates any potential deficiency. Calcium supplement use is a factor because the most commonly used calcium salts will bind food phosphorus in the gut and reduce its absorption.
Figure 1Portion of the elderly osteoporotic population most likely to be susceptible to insufficient phosphorus intake. Domain sizes are not drawn to scale.
Simple maintenance of skeletal mass translates to zero phosphorus balance and places little or no demand on dietary phosphorus intake. This is because urinary phosphorus excretion can be reduced to extremely low levels, and phosphorus released in bone resorption can thus be efficiently reused for mineralization of newly forming bone sites. During any age-related decline in bone or muscle mass, phosphorus released in the process can substitute for phosphorus in the diet. Indeed, it must otherwise be excreted by the kidney. Thus, even intakes well below the RDA, whatever other nutritional effects they may have, should be able to sustain the skeletal phosphorus requirements of the typical elderly individual. Antiresorptive therapies change that picture somewhat. They typically produce a slow increase in bone mineral content of 0.5% to 1.0% yearly at the spine after an initial positive remodeling transient of +4.5% to +5.5% in the first year.
The steadystate increase in bone mineral content after 1 year creates a small tissue-level phosphorus requirement, with net phosphorus uptake by bone in the range of 4.5 to 9.0 mg/d. By contrast, new-generation anabolic therapies, capable of increasing axial bone mineral content by up to 15% yearly, call for a positive phosphorus balance of up to 45 to 90 mg/ d. (The foregoing estimates assume that values measured in the axial skeleton in response to bone active agents apply to the whole skeleton and hence reflect a worst-case, or maximum likely, phosphorus demand.)
Food phosphorus binding by coingested calcium is a well-recognized and well-used approach, particularly in the management of phosphorus absorption in patients with end-stage renal disease. Over a broad range of intakes, net absorption of calcium averages only 10% to 11%, which means that approximately 90% of ingested calcium, whether food or supplement, remains in the gut lumen,
have shown that fecal phosphorus content (and therefore, inversely, absorbed phosphorus) is determined primarily by unabsorbed calcium and to a lesser extent by phosphorus intake. Together, dietary calcium and phosphorus intakes account for nearly three fourths of the variation in the quantity of phosphorus absorbed. Briefly, each 500 mg of ingested calcium binds 166 mg of diet phosphorus (95% confidence interval, 144-188 mg).
Using this relationship, Figure 2 shows the amount of food phosphorus that would be available for absorption in a woman ingesting 70% of the phosphorus RDA (Table 1) under varying conditions of dietary and calcium supplementation. The calculations used in constructing Figure 2 assume that the calcium supplements are coingested with the food phosphorus.
At a calcium supplement intake of between 1000 and 1500 mg/d, all the food phosphorus is complexed by unabsorbed calcium and rendered nonavailable. Above that point, phosphorus absorption becomes negative (ie, phosphorus in digestive secretions and sloughed intestinal mucosa, estimated at as much as 250 mg/d,
will be complexed as well, blocking its reclamation by the gut; this converts the gut into a net excretory organ for phosphorus).
Figure 2Total phosphorus intake and available phosphorus in individuals ingesting 70% of the recommended daily allowance for phosphorus under varying conditions of calcium supplementation. A, All supplemental calcium as carbonate or citrate. B, Same as for A, except for the addition of a single serving of milk to the diet. C, Same as for A, except that the calcium supplemental intake is in the form of tricalcium phosphate.
Adding a single serving of milk to the diet of such women eliminates much of the problem. Alternatively, the use of a calcium phosphate supplement provides the desired calcium, prevents the blocking of food phosphorus absorption, and ensures a generous supply of available phosphorus. Although total phosphorus intake rises nearly 3-fold with the highest of the calcium phosphate supplement scenarios, available phosphorus rises very little. Thus, basically what the calcium phosphate supplement does, in addition to providing the desired calcium, is to spare food phosphorus.
The timing of the calcium supplement intake is important for this relationship. As noted, the foregoing calculations are for simultaneous ingestion of calcium and phosphorus, which is in accord with the usual recommendation. Taking supplements with meals remains a sound stratagem because by slowing gastric emptying, the meal effect improves calcium utilization efficiency, and by spreading the dose throughout the day, calcium absorption is improved further. However, this same stratagem maximizes phosphorus binding and minimizes phosphorus absorption. Schiller et al
have shown, using a single-meal design, that separation of the calcium supplement dose from the time of the phosphorus-containing meal substantially reduces the degree of phosphorus binding. This is, of course, to be expected for a process that involves nothing more complex than simple chemical binding. Figure 3 plots some of the data from the article by Schiller et al, showing both the degree of reduction in phosphorus absorption produced by simultaneous coingestion of calcium and phosphorus and the reduction in this effect produced by giving the calcium supplement 2 hours after the meal.
Figure 3Percentage of phosphorus absorption from single meals (±1 SEM) with and without ingestion of supplemental calcium, showing the relationship to time of supplement ingestion. The phosphorus content of the diet was 370 mg, and the calcium supplement dose was 1000 mg given as the acetate salt. Net phosphorus absorption averaged nearly 80% with no calcium supplement, 31% with simultaneous ingestion, and 59% when the calcium was delayed for 2 hours. Data from Schiller et al.
Having calculated the demand under various treatment and intake scenarios, it is possible to estimate the probable level of risk of phosphorus insufficiency in patients being treated for osteoporosis. Extremely large calcium supplement doses (eg, >1500 mg/d), particularly if taken regularly as 500 to 600 mg with every meal, could create a negative phosphorus balance situation even for women who are not undergoing antiosteoporosis therapies, particularly if their meat and dairy intakes are low. However, it is improbable that a regimen of invariable coingestion would be rigidly followed. Hence, without a bone-building demand for phosphorus, it is unlikely that phosphorus binding by calcium supplements would be sufficiently complete as to constitute a serious problem. However, once bone active agents, and particularly anabolic agents such as teriparatide, are used, the situation changes. Figure 2 shows that, with calcium carbonate or citrate supplement intakes in excess of 1200 mg/d, there is effectively no diet phosphorus available to support bone building. This supplement dose is less than that used in some teriparatide treatment studies.
Two years of parathyroid hormone 1-34 and estrogen produce dramatic bone density increases in postmenopausal osteoporotic women that dissipate only slightly during a third year of treatment with estrogen alone: results from a placebo-controlled randomized trial [abstract].
In brief, because of the likelihood of some irregularity of the coingestion of calcium supplements with foods, it may still be possible for women with low phosphorus intakes and those receiving antiresorptive therapies to realize the small increase in bone mass that these therapies are able to produce. However, it is unlikely that this would be the case with the anabolic agents, which create a phosphorus demand for bone mineralization that is approximately 10 times that of the antiresorptives. Under such circumstances, it would clearly be prudent to adjust the diet of the patient or to use a calcium phosphate salt as the calcium supplement (or both).
effective phosphorus deficiency with respect to bone mineralization must express itself as a low value for serum inorganic phosphorus. As long as the bone-forming site is exposed to normal phosphorus concentrations, it will have sufficient phosphorus to support matrix mineralization. In many older individuals with reduced creatinine clearance, there will commonly be some elevation of serum phosphorus levels, which would be protective of the skeleton. The effect of complete or nearly complete food phosphate binding on serum phosphorus in such individuals is unknown. However, they may be protected to some extent. Greater experience with teriparatide therapy will be needed before their relative vulnerability will become clear.
SAFETY CONSIDERATIONS
The safety of phosphate salts should not be an issue because successful calcium supplementation trials such as that by Chapuy et al
used tricalcium phosphate as the supplement source without evident difficulty (and with considerable benefit), and supplement products based on bone-derived hydroxyapatite have been in worldwide use for more than 30 years, also without known problems. Moreover, Spencer et al
showed no effect of 4-month treatment with 1144 mg/d of phosphorus on either kinetic or histomorphometric indices of bone turnover. Nevertheless, published articles indicate that elevation of serum inorganic phosphate levels increases parathyroid hormone (PTH) secretion,
showed increased parathyroid activity in individuals consuming low-calcium, high-phosphorus diets and attributed the changes to the high-phosphorus component of their regimen. However, Barger-Lux and Heaney
subsequently showed similar changes with a diet low in calcium but normal in phosphorus, indicating that it was the low-calcium characteristic of the diet that was the culprit. Essentially the same conclusion had been reached by 2 expert panels convened by the Life Science Research Office in response to Food and Drug Administration concerns about phosphorus in the food supply.
Effects of dietary factors on skeletal integrity in adults: calcium, phosphorus, vitamin D, and protein. US Dept of Health and Human Services, Food and Drug Administration,
Washington, DC1981
This effect has been assumed to be the result of the direct action of phosphorus on the parathyroid gland, and it has been extrapolated to mean increased bone remodeling with its consequences for bony fragility. However, both of these assumptions are probably incorrect. Studies of bone remodeling with phosphate supplementation show either no effect or decreased resorption. Raisz and Niemann
showed more than 30 years ago that, as ambient phosphate concentration increased, osteoclastic response to standard doses of PTH decreased. Thus, the elevation in serum phosphorus levels produced by large supplement doses would be expected to antagonize PTH action on bone, thereby lowering the serum calcium level and by that means evoking an increase in PTH secretion (rather than directly). However, elevating serum inorganic phosphorus levels does not directly suppress serum calcium because the solubility product constant (KSP) for CaHPO4 is approximately twice the serum ion product that characterizes adults with reasonably normal kidney function. Direct confirmation of that discordance between increased PTH and reduced remodeling with phosphorus supplementation is seen in the work by Bizik et al,
showed increased PTH release in vitro by parathyroid cells into a high calcium medium at phosphorus concentrations of 3 and 4 mmol/L but not at 1.5 mmol/L. The effect on PTH release seen at 3 and 4 mmol/L (equivalent to serum phosphorus values of 9.3 and 12.4 mg/dL) likely reflects complexation of phosphate with medium calcium and lowering of effective calcium ion concentration rather than a direct effect of phosphorus itself because such high phosphorus levels would exceed the KSP for CaHPO4. The 1.5-mmol/L concentration, which produced no effect on PTH release, is above the upper limit of normal for serum phosphorus for adult humans and is not likely to occur with any plausible human diet or supplementation regimen. Hence, this experiment does not support a direct effect of serum inorganic phosphate on parathyroid gland activity.
Similarly, in their experiment with rats, Slatopolsky et al
found effects of phosphorus on serum PTH only at what would be for humans unrealizably high phosphorus intakes (0.8%, equivalent to a human diet containing >5000 mg/d of phosphorus). At a diet phosphorus content of 0.2% (equivalent to a human diet with a phosphorus intake 60% above the RDA), they observed no effect on parathyroid activity. Thus, this experiment also provides no evidence for an effect of phosphorus intake on parathyroid cell activity within plausible intake or serum concentration ranges.
There was a considerable difference between the phosphorus densities of the diets of humans and laboratory animals most often used in nutritional investigations. At the adult RDA for phosphorus and a total energy intake of 2200 kcal, the human diet has a phosphorus density of 32 mg per 100 kcal (or approximately 1 mmol per 100 kcal). Actual human diets range up to 2 to 3 times that figure. By contrast, rations for rodents have phosphorus densities of 4 to 6 mmol per 100 kcal, and rations for dogs and cats are approximately 9 mmol per 100 kcal. (The highest human value in Table 1 translates to a phosphorus density of only 3.2 mmol per 100 kcal.) Essentially all animal experiments that show harmful effects of high-phosphorus diets start from intakes already high by human standards and raise them still further. Failure to note this important difference has been the source of much of the undeservedly bad reputation phosphorus has acquired during the past several decades.
Nevertheless, to help resolve any remaining uncertainties and to add to the still small body of data reporting phosphorus effects on bone remodeling, we addressed this problem in a study involving 24 postmenopausal women, focusing on the effects of various supplement sources on PTH and bone remodeling (R. P. Heaney, MD, K. Rafferty, RD, M. S. Dowell, PhD, unpublished data, 2003). The subjects had low usual intakes of calcium and phosphorus, and we used 3 consecutive 1-week supplementation regimens. All meals were provided, and foods were selected to ensure total food phosphorus intakes of less than 600 mg/d. During the first week, there was no supplemental calcium or phosphorus given; during the second, 1800 mg/d of calcium was given as the carbonate; and during the third, 1800 mg/d of calcium was given as tricalcium phosphate (providing 930 mg of additional phosphorus, ie, increasing phosphorus intake by more than 150%). These intakes are close to or higher than the maximum likely intakes of both calcium and phosphorus from supplement sources. At the end of each week, serum and urine were collected and analyzed for PTH and N-telopeptide, respectively (Figure 4). Relative to the baseline, when no calcium or phosphorus supplements were used, fasting PTH levels were reduced by 15% by both supplement sources (P<.01). The reduction was identical for the phosphate and the carbonate salts. Similarly, 24-hour urinary N-telopeptide levels were reduced by 32% by both sources (P<.01), with no difference between them. The reduction in bone resorption was presumably an effect of the calcium. However, the reassuring aspect is that the large increase in phosphorus intake during the week with tricalcium phosphate did not alter either the PTH suppression or the resorptive decrease produced by the added calcium.
Figure 4Fasting serum parathyroid hormone (PTH) and 24-hour urinary N-telopeptide (NTx) levels (±1 SEM) after 1 week of no supplementation (control), 1 week of CaCO3, and 1 week of Ca3(PO4)2. The 2 calcium preparations provided 1800 mg/d of calcium. The difference from control was statistically significant (P<.01) in each case, whereas there was no difference between the 2 calcium sources for either end point.
In brief, phosphate salts of calcium produce the same degree of PTH suppression and reduction in bone resorption as do carbonate sources. Thus, both relatively largescale clinical use of phosphate-based supplements and metabolic experiments indicate that there is essentially no cause for concern about the use of phosphate-containing calcium salts.
CONCLUSION
Although phosphorus intakes are generally adequate to meet likely skeletal needs in most patients with osteoporosis, some individuals with low intakes of meat and dairy products may not be able to take full advantage of the bonerebuilding potential of modern bone active agents, particularly if they are given calcium carbonate or citrate supplements. Such supplements, taken concurrently with meals, will block absorption of most or all of their food phosphorus. We need to either focus more attention on the quality of diets of women with osteoporosis or consider a calcium phosphate supplement in place of the usual carbonate or citrate salts. The extra phosphorus provided by such salts has been shown to be without adverse effects on either the skeleton or the calcium economy.
Dietary Intake of Vitamins, Minerals, and Fiber of Persons Ages 2 Months and Over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988-91. National Center for Health Statistics,
Hyattsville, Md1994 (Advance Data From Vital and Health Statistics, No. 258.)
Two years of parathyroid hormone 1-34 and estrogen produce dramatic bone density increases in postmenopausal osteoporotic women that dissipate only slightly during a third year of treatment with estrogen alone: results from a placebo-controlled randomized trial [abstract].
Effects of dietary factors on skeletal integrity in adults: calcium, phosphorus, vitamin D, and protein. US Dept of Health and Human Services, Food and Drug Administration,
Washington, DC1981
This study was supported in part by grant AR-07912 from the National Institutes of Health, Bethesda, Md, and by the Health Future Foundation, Creighton University, Omaha, Neb, and Rhodia, Inc, Cranbury, NJ.
63261-8&id=gr1.jpg){kind=link}
63261-8&id=gr2.gif){kind=link}
63261-8&id=gr3.jpg){kind=link}
63261-8&id=gr4.jpg){kind=link}