Methods

Top Methods Results Discussion Author & Article Info References

Between 1 November 1992 and 1 November 1993, we did a survey of the utility of measuring indices of vitamin D, parathyroid hormone, and thyroid hormone balance in all new patients referred by their primary care providers to the Cedars-Sinai Bone Center (Los Angeles, California) for evaluation of possible osteoporosis or low bone mineral density. Thirty-nine patients, most of whom were from upper-middle-class sections of West Los Angeles, were evaluated. The patients were 34 to 80 years of age; 37 were white, and 32 were women.

Blood samples were collected from all patients for measurement of serum levels of calcium, 25-hydroxyvitamin D, immunoreactive parathyroid hormone, and thyroid-stimulating hormone; a 3-mL aliquot of serum was stored at –70°C for the measurement of 1,25-dihydroxyvitamin D levels. All patients provided a timed (2 hours) fasting urine sample for measurement of levels of calcium and creatinine. Serum levels of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D were measured by using competitive protein-binding assay (Nichols Institute, San Juan Capistrano, California). Serum levels of parathyroid hormone and thyroid-stimulating hormone were measured by using immunoradiometric assays (Nichols Institute). Calcium, albumin, and creatinine levels were determined by spectrophotometry. Bone mineral density of the lumbar spine and the nondominant proximal femur was assessed by using dual-energy x-ray absorptiometry done on the same Lunar DPX machine (Lunar Corp., Madison, Wisconsin).

Four patients were found to have hypercalciuria and elevated serum levels of 25-hydroxyvitamin D. Their serum levels of 1,25-dihydroxyvitamin D were measured, and urine and serum samples were collected repeatedly until the ratio of fasting urinary calcium to creatinine returned to normal (<0.16). These patients were then followed prospectively for a mean (±SD) 3.2 ± 0.2 years; serum 25-hydroxyvitamin D level, ratio of fasting urinary calcium to creatinine, and bone mineral density were measured at least once annually.

All data are expressed as the mean ±SD. Statistical comparisons were made by using the Wilcoxon rank-sum test.

Results

Top Methods Results Discussion Author & Article Info References

Table 1 shows the characteristics of patients categorized according to their serum level of 25-hydroxyvitamin D at presentation. Four patients were found to have elevated 25-hydroxyvitamin D levels. These patients (three women who were 60, 61, and 73 years of age and one man who was 53 years of age) were referred for evaluation of osteoporosis (defined as bone mineral density 2.5 SDs below the peak mean [8]). The 60-year-old woman and the 61-year old woman were receiving conjugated estrogen, 0.625 mg/d. The 61-year-old woman was also receiving levothyroxine sodium, 0.05 mg/d. All gonadal steroid and thyroid hormones were obtained by prescription. During the 2- to 3-year follow-up period, hormonal therapy was not changed. All four patients with vitamin D intoxication were taking 1000 mg of supplemental calcium and a multivitamin preparation daily when they were first evaluated. Multivitamin and calcium salt preparations were obtained either by prescription or over the counter in pharmacies and food stores. No patient knowingly consumed more than 1200 IU of vitamin D daily or took other medications known to alter skeletal metabolism.

The four patients with vitamin D intoxication used several daily dietary supplements in addition to the calcium and multivitamin supplements described above (mean, 4.6 supplements; range, 3 to 8 supplements) for at least 6 months before initial evaluation. Dietary supplements were obtained by mail order, in health food stores, or from personal trainers. None was obtained by prescription. Some of these products were advertised as containing as much as 3600 IU of vitamin D₃ per daily dosage. Two women, 60 and 73 years of age, were taking animal-extract preparations that were not advertised as containing vitamin D. These two patients had the highest levels of 25-hydroxyvitamin D (222 nmol/L and 200 nmol/L) when they were first evaluated (they were evaluated in February and March, a time of year when ultraviolet B irradiation and cutaneous production of vitamin D are at a nadir) [7]. All four patients with elevated 25-hydroxyvitamin D levels were hypercalciuric, but none was hypercalcemic or had an elevated 1,25-dihydroxyvitamin D level. As a group, they had a urinary calcium excretion rate that was threefold greater than the rate in patients with a normal 25-hydroxyvitamin D level. The mean serum levels of calcium and thyroid-stimulating hormone were the same in patients with and patients without hypercalciuria. The mean serum concentration of parathyroid hormone in the four hypercalciuric patients was lower, indicating appropriate suppression of parathyroid hormone secretion.

Figure 1 shows changes in the serum levels of 25-hydroxyvitamin D, the ratio of fasting urinary calcium to creatinine excretion, and the bone mineral density of the lumbar spine in the four hypercalciuric patients after the use of oral calcium, vitamin D, and other dietary supplements was discontinued. Although no correlation was seen between the initial serum level of 25-hydroxyvitamin D and the ratio of urinary calcium to creatinine in individual patients, there was a coordinate decrease in the two values over time and the rate of decline in the serum level of 25-hydroxyvitamin D (10.7 ± 3.0 nmol/L per month) was remarkably similar among the four patients.

View larger version (12K): [in this window] [in a new window]   Figure 1. Serum levels of 25-hydroxyvitamin D (left), ratio of fasting urinary calcium to creatinine (middle), and lumbar vertebral bone density expressed as a percentage of theoretical peak bone density of vertebral bodies L1 to L4 (right) in four osteopenic patients. One patient was male (s) and three were female (closed boxes, closed circles, and open circles). All had hypercalciuria before and after discontinuation of use of dietary supplements that contained vitamin D. A time of 0 years was arbitrarily set at the extrapolated time of resolution of vitamin D intoxication. The solid horizontal lines represent the upper limit of normal values.

The use of dietary calcium supplements, but not other dietary supplements, was reinstituted in all previously intoxicated patients. In response, serum calcium levels, 25-hydroxyvitamin D levels, and ratio of urinary calcium to creatinine remained in the normal range; serial measurement of 1,25-dihydroxyvitamin D and parathyroid hormone levels was not done. After hypercalciuria resolved and serum levels of 25-hydroxyvitamin D returned to normal, all patients had an increase in bone mineral density of the lumbar spine that was independent of any loss in vertebral height. From the initial to the final evaluation, there was a mean increase in lumbar spine bone mineral density of 5.0% ± 1.6% at an annual rate of 1.9% ± 0.6%; a similar but more variable increase (4.3% ± 4.4%) was noted in the bone mineral density of the femoral neck.

Discussion

Top Methods Results Discussion Author & Article Info References

Unintentional vitamin D intoxication has been well documented [9-14]. We report that ingestion of over-the-counter dietary supplements is a nonmedicinal cause of vitamin D intoxication and abnormal calcium balance. Our assertion is supported by the observation that the discontinuation of use of calcium and other dietary supplements resulted in normocalcemia, even after the reintroduction of the original calcium supplement and return of 25-hydroxyvitamin D levels to normal in all patients who had been hypercalciuric. The identity of the vitamin D molecule responsible for hypervitaminosis D in our patients is not known. High-performance liquid chromatography of lipid extracts that were prepared from the supplemental animal extracts ingested in large quantities by the two patients with the highest initial 25-hydroxyvitamin D levels (Figure 1) showed that the supplements were enriched with vitamin D₂, not vitamin D₃. Vitamin D₂ (ergocalciferol) comes from plants; this indicates that the so-called animal extracts were probably supplemented with vitamin D₂.

Given that all of our patients had evidence of hypercalciuria after an overnight fast, we presumed that some of the increase seen in the filtered load of calcium resulted from the mobilization of skeletal calcium. This presumption was substantiated by the increase in bone mineral density (of almost 2% per year) that was seen in all four of the formerly vitamin D-intoxicated patients (Figure 1) after their 25-hydroxyvitamin D levels had returned to normal. The specific cause of intoxication from ingested vitamin D remains uncertain [15]. This intoxication could have resulted from 1) displacement by 25-hydroxyvitamin D of 1,25-dihydroxyvitamin D from the serum vitamin D-binding protein or 2) the binding of 25-hydroxyvitamin D itself to the vitamin D receptor in bone cells.

The unanticipated prevalence of apparent hypervitaminosis from consumption of dietary supplements that were not regulated by the U.S. Food and Drug Administration is of particular concern. Given that 38 to 40 million persons in the United States use dietary supplements daily [16], this form of vitamin D intoxication and occult hypercalciuria may be more widespread than previously thought. We propose that screening osteopenic patients (especially those who are economically advantaged) for hypercalciuria and high levels of 25-hydroxyvitamin D is worthwhile because resolution of vitamin D intoxication results in a substantial increase in bone mineral density.

Requests for-Reprints: John S. Adams, MD, B131, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Los Angeles, CA 94309.