Lead Content of Calcium Supplements

The risk of exposure to lead from over-the-counter calcium supplements has been well recognized for nearly 2 decades.¹ This risk concerns calcium carbonate in "natural source" (ie, oyster shell) and refined formulations.² Despite a flurry of publicity in such venues as Consumer Reports,³ a number of health concerns remain. Since the medical community now more vocally advocates increasing calcium intake (typically in the form of supplements) to prevent or treat osteoporosis, we thought it important to re-examine this issue. It is particularly relevant for specific patient groups, such as those with renal failure who use extremely high doses of calcium as phosphate binders. Furthermore, the public in general is consuming an increasing number of unregulated dietary and health food supplements, which makes ongoing risk surveillance valuable.

We obtained multiple widely available commercial calcium preparations and analyzed them for lead content to address these health concerns. First, we were interested in whether manufacturers of natural source calcium carbonate were self-regulating their sources and methods of formulation to maintain acceptably low levels of lead. Second, products not labeled as being "natural" were tested to determine whether there was risk associated with consuming what was thought to be refined calcium carbonate, a category often perceived to be safe. Third, we analyzed a prescription-only and thereby government regulated and much more expensive calcium salt (calcium acetate). Risk was then assessed by calculating lead exposure based on the range of doses recommended for calcium supplementation or binding dietary phosphate.

In March 2000, a variety of calcium supplements (n = 22, Table 1) were selected based on their being marketed nationally as well as being widely and easily available in the community served by our medical center (University of Florida, Gainesville). The supplements were obtained from national chains of pharmacies, a well-known discount merchandiser, or from an existing stock of patient samples directly from the manufacturers. Samples for every preparation were taken from factory-sealed containers of a single lot number. A total of 21 nonprescription calcium-based products were selected, all of which were calcium carbonate: 7 were labeled as natural source or oyster shell, and 14 were described as refined (or lacked the "natural" designation). Interestingly, 2 preparations from a single brand (1 natural source and 1 refined) had the additional labeling of being "essentially lead free." An additional product was the prescription-only calcium acetate, typically marketed as a phosphate binder for patients with renal failure. We also included as a 23rd product, sevelamer hydrochloride, because this noncalcium polymer-based synthetic compound serves as an alternatively prescribed phosphate binder.

Lead assays were contracted to the University of Florida Analytical Toxicology Core Laboratory, Gainesville. Since this laboratory performs forensic analyses, there was the additional benefit of signature-based protocols for chain of custody that ensured proper handling and identification of the samples.

Tablets were homogenized and then subsampled into 1-g aliquots. Capsules were digested intact with the contents. All samples were prepared in triplicate, and all reagents were trace metal grade. Samples were wet ashed using nitric acid and hydrogen peroxide. Lead analysis was then performed by electrothermal atomic absorption spectrometry with Zeeman background correction. The absorbance was measured at 283.3 nm. Each sample replicate was analyzed twice, with the mean quantified against a 5-point calibration curve (R² = 0.9996). Quality control included method blanks and spikes, matrix spikes, and continuing calibration standards and blanks every 5 samples. No method or calibration blank gave a detectable response for lead. The method analytical limit was 2.5 µg/L (250 ng/g of sample).

The results were expressed as micrograms of lead per gram of homogenized specimen (which thereby included filler and capsule weights) and then converted to micrograms per tablet or capsule. Average unit weight for each formulation was determined from the weight of 10 tablets or capsules. Using the manufacturer's quantification of milligrams of elemental calcium per packaged unit, the lead exposure was then calculated for various daily doses (Table 2). For calcium used as a supplement, the total lead was indicated for the elemental calcium children's recommended daily allowance (RDA) of 800 mg/d (assuming the RDA was met by supplements alone) and for 1500 mg/d because that dosage is recommended by many osteoporosis guidelines.⁴ For dialysis patients, the daily calcium intake varies widely because it is adjusted according to dietary phosphate consumption; thus, a range of possible lead exposure was calculated. It is not unusual for patients to take 4 to 6 or more "binders" with each meal, plus a smaller number with snacks. To represent lead intake for patients needing relatively large doses, we selected an arbitrary total of 15 tablets or capsules per day (typically providing between 8 and 9 g of elemental calcium). To represent those taking fewer binders, the results were calculated for a total daily intake of 6 g of elemental calcium, which also overcame the problem of varying amounts of calcium in each unit dose. All of these lead quantities were then compared with the current maximum accepted total dietary lead intake level of 6 µg/d for populations at particular risk, such as children.^{5 - 7}

A number of products and dosing regimens potentially would result in excessive lead intake (Table 2). Of the 7 formulations labeled as being of natural or oyster shell source, 4 had measurable lead content. For doses in the range of the entire childhood calcium RDA (800 mg/d), all 4 of these products yielded approximately 0.9 to 1.0 µg/d of lead. Used in doses typical of those for the treatment or prevention of osteoporosis, 1500 mg/d of elemental calcium,⁴ all 4 exceeded 1 µg/d of lead but none approached 6 µg/d. However, when dosed in the multigram or multitablet ranges common for patients with renal disease, all 4 products imparted more than the 6-µg/d maximal limit of lead. When taken in high doses, the daily exposure could easily be around 10 µg/d.

Of the 14 preparations of refined calcium carbonate, 4 had detectable lead, and the levels were similar to if not greater than those found in the oyster shell products. For the relatively low calcium dose of 800 mg/d, these 4 products imparted between approximately 1.0 and 1.8 µg/d of lead. Taken in osteoporosis doses, the daily intake of lead was estimated to be in the range of approximately 2 to 3 µg/d, representing half of the permissible total daily limit. In high renal doses, the lead exposure could reach or exceed 20 µg/d. One of these over-the-counter formulations was a popular national brand and had the highest lead levels measured of any product tested in this study.

The 2 formulations of nonprescription calcium carbonate that were specifically labeled as being "essentially lead free" had unmeasurable lead content (<0.25 µg/g of lead per unit dose). The prescription-only calcium acetate preparation also had undetectable lead, as did the noncalcium phosphate binder sevelamer hydrochloride.

Ensuring the quality of over-the-counter, health-related products has been an ongoing public concern, especially at a time when there is increasing publicity and claims as to the benefits of health foods and compounds. An estimated 40% of the population in the United States takes a dietary supplement at least monthly.⁸ We believe, however, that unlike "alternative" health substances of unproven benefit, there is a conceptual difference for calcium because its use is widely and recommended by the medical community.

The need for calcium supplementation continues to receive much attention for the treatment and prevention of osteoporosis, for lactating mothers, and for subsets of the pediatric population. It may even be perceived that if the medical community publicizes specific nutritional guidelines, there is an obligation to be equally vocal in advising the safest or lowest risk method to meet these recommendations. Although the general public has been notified about some of these concerns, ongoing discussion and education are warranted. Approximately 5% of the adult population consumes calcium supplements, thereby posing a concern for a large population⁹ and for those individuals who for ethnic or health reasons opt to take larger or megadoses.¹⁰ The potential risk of lead exposure would be greater in some patient groups, such as those with chronic renal failure for whom larger doses of calcium are prescribed. Since many insurance companies do not cover the cost of calcium in the carbonate form, without specific advice, patients may switch between brands based on cost and unwittingly expose themselves to higher risk.

We thought it was relevant to determine whether 2 decades of knowledge of this health concern^{2 ,11 - 12} had led to significant self-regulation and improved formulations on the part of the manufacturers. Bourgoin et al² had clearly shown the risk of lead exposure from a wide variety of brands and types of calcium supplements. Moreover, even if consumers were well informed of the risks of "natural" products, it was not clear whether there was merit in the perception that the hazard would be avoided by using seemingly refined preparations.

The risk associated with the lead levels reported in our study needs to be interpreted in light of uncertainty and controversy as to the acceptable maximum total oral daily intake of lead. One approach to determining that limit has been to use the maximal acceptable blood lead level and an estimate of the percentage of absorption of oral lead. Complicating this analysis is that children may be more sensitive to any given level of lead exposure and that the degree of absorption declines with advancing age. Although it has been proposed that calcium might attenuate lead absorption, this has not been confirmed.¹³ For populations at particular risk (infants, young children, pregnant or lactating women, and fetuses), the absorption calculations yield a suggested (or provisional) dietary limit of 6 µg/d^{5 - 7} of lead from all oral sources, not just the supplements. Some authors,⁶ however, have suggested a rigorous safety margin that would lower the permissible intake to approximately 1 µg/d. Although this target is commendable, it is unfortunately unrealistically low at this time. Estimated dietary intake of lead has decreased in the last decade to less than 5 µg/d,¹⁴ which calls into question prior advice that healthy individuals could have permissible blood levels as high as 1.4 µmol/L (30 µg/dL) and consume as much as 75 µg/d.^{6 - 7}

We concur with the more recent approach questioning whether there can ever be an allowable amount of lead, since no risk-free blood level has been established.^{15 - 16} It is only with the dramatic success of environmental efforts in reducing lead exposure that the small but clinically significant toxicity of supplements becomes a relevant issue. Importantly, in light of our finding that one can readily obtain calcium carbonate with negligible quantities of lead, we believe that patients should not be put in the precarious position of weighing the risks of lead exposure with the benefits of the supplement. A reasonable and achievable goal² is to limit the lead contained in calcium supplementation to less than 1 µg/d, corresponding to a child's total RDA (800 mg/d). That target is more stringent than proposed in some state regulations (ie, California, 1.5 µg of lead per 1 g of calcium). Thus, the widespread availability of safe formulations could be impetus for the United States Pharmacopeia to further lower the allowable level of lead in calcium carbonate preparations, which was decreased in 1990 from 10 to 3 µg/g.¹⁷

The risk of lead exposure would be especially important for lactating mothers, children, and those individuals intolerant of milk products who opt to obtain the calcium nutriture from manufactured supplements. Since the data in Table 1 2 were calculated for individuals achieving arbitrary daily calcium targets (such as 1500 mg/d for osteoporosis⁴ ), there would be additional risk for those following alternative health regimens involving megadosing of vitamins and minerals. Furthermore, patients with impaired renal excretion would theoretically have lower urinary lead losses and, therefore, experience a greater impact from excess oral intake.

The results of our study indicate that many but not all "natural" fossilized marine limestone bed (ie, oyster shell) derivatives when used in the calcium dose range for osteoporosis do carry the risk of greater than a 1-µg/d lead exposure and up to approximately 50% of the total daily limit, although none reached the 6- or 10-µg/d thresholds. When taken as prescribed by patients with renal failure for binding dietary phosphate, however, the daily intake of lead could exceed that limit and may greatly exceed 10 µg/d.

Products not labeled as being of natural or oyster shell origin are typically assumed to be from refined sources of calcium carbonate. Despite the common perception that these compounds would thus have fewer metal contaminants, our results show that they can still contain considerable amounts of lead. The preparation with the highest lead content identified by our study was in this category, thereby potentially exposing patients who take calcium supplements to more than 2 µg/d of lead and patients with renal failure to as much as 20 µg/d of lead. Importantly, the 2 calcium carbonate products (1 "natural" and 1 refined) specifically labeled as having "essentially" no lead indeed had no detectable quantities.

Patients with renal failure present additional concerns because of potential toxic effects from long-term exposure to trace metals, such as aluminum. Surveying bone specimens from dialysis patients for trace metal content has not revealed a consistent problem with respect to lead. Although there is considerable laboratory evidence of the effect of lead on osteoblasts, collagen synthesis, and vitamin D,¹⁸ it is not clear to what extent it causes clinical bone disease. Whole blood levels have not been observed to be elevated in dialysis patients,¹⁹ and other tissues have not been systematically examined for lead content. Interestingly, it was suggested that bone biopsy specimens with evidence of osteomalacia were associated with increased lead²⁰ or ratios of lead to calcium based on weight.¹² It is not clear how high (osteitis fibrosa cystica) and low (adynamic) bone turnover disorders influence lead deposition and mobilization. The small percentage of dialysis patients who have transiliac bone evidence of lead exposure in the range observed in lead workers has been thought to have been exposed before end-stage renal disease.²¹ Whether there are groups of dialysis patients in whom occult lead toxicity causes or contributes to such clinical diseases as adynamic bone, anemia, erythropoietin resistance, or neurological dysfunction warrants further investigation.

We are also confident in the accuracy of our data because of the nature of the assay design²² and by repeated verification using appropriate controls. Since there may be variation in lead content between different segments of a single tablet,²³ our analysis was from homogenates of multiple pills.

Consumers and patients need to interpret our findings with some caution, even though they confirm and show the ongoing nature of the risk first described years ago.² Although many authors generally have advocated avoiding natural oyster shell products, some of these products had undetectable lead content. For those who broadly suggest refined products, many were safe, but a disturbing number had excess lead. For proponents of name-brand formulations, some had much higher lead content than generic brands. For advocates of products made by nationally recognized pharmaceutical companies, unfortunately, some of these had high lead levels. One cannot assume that a given brand is uniformly safe, because some of their products may have high and others low lead levels. Last, since it has already been recognized that there is much lot-to-lot variation in lead content, one cannot necessarily be assured that a particular formulation will remain safe. This pertains to the preparations reported herein as having undetectable lead levels, because we only studied a single manufacturing lot. Only by having the manufacturer be cognizant of the quality of the source of calcium and by rigorous quality testing can the public be assured of product safety.

In summary, after dramatic environmental protection successes, we believe that there needs to be heightened public awareness of the small but potentially clinically important risk of lead exposure from calcium products of natural (oyster shell) sources packaged as dietary supplements and from formulations that are seemingly refined. There needs to be continued public health concern over the health risks of what have previously been popular products, bone meal and dolomite.²⁴ Patients and consumers ought not assume safe lead levels from preparations that have brand names or are supplied by recognized pharmaceutical companies, especially when calcium products are taken in doses beyond the childhood RDA. If these products continue to lie outside the jurisdiction of federal regulatory agencies, then consumers would be well advised to purchase these products only from reputable manufacturers and only when they are specifically labeled as having been tested for lead content. Since there is such a large market for calcium products, it is hoped that this will lead to self-regulation by the industry and ultimately safer compounds. This issue is even more important to patients with renal failure because they are at greater risk, and unless they can be assured of the safety of the inexpensive over-the-counter calcium products, there would be justification for the much more costly prescription-only calcium or polymer phosphate binders.