Introduction

In rats and humans, parathyroid hormone accelerates metabolic clearance of vitamin D, either through increased metabolism of 25(OH)D to 1,25(OH)2D, or increased clearance through the liver [1]. However, parathyroid hormone also suppresses 25(OH)D-24-hydroxylase, the enzyme that initiates the catabolic inactivation of 25(OH)D and its metabolites [2, 3].

We describe the case of a woman who had consumed 50,000 IU vitamin D2 per day for over 2 years. She developed severe hypervitaminosis D, and after discontinuation, her elevated 25(OH)D concentration persisted for almost 2½ years. The half-life of vitamin D in the body is thought to be about 2 months, hence the case exhibits an unusual pattern. An underlying primary hyperparathyroidism that was masked initially by the hypercalcemia of vitamin D intoxication may have been responsible for the prolonged high level of 25-hydroxyvitamin D in her blood.

Case report

This 77-year-old woman was referred to the Renal Clinic of the Toronto Western Hospital for the investigation of renal failure in December 2003. For the previous 6 months she had had nausea, had vomited occasionally, and was feeling tired. Her appetite was poor and she had lost 7 kg in weight during this period. For the previous 2 years, she had received diclofenac sodium/misoprostol 75 mg b.i.d, and because of an error 50,000 units of ergocalciferol (Ostoforte, Merck Frosst, Canada) once a day. In 1994 she was found to have colorectal cancer and underwent chemotherapy for 1 year. Colonoscopy, performed 2 and 1½ years before, showed no abnormalities. No previous data on her serum Ca, P, 25(OH)D, and iPTH levels were available.

Presentation and laboratory investigations

On examination, her weight was 54.7 kg and her height was 159 cm, blood pressure was 160/80 mmHg, and her pulse rate 70 b/min. She looked pale but there was no other overt abnormality. Her serum biochemical measures were: creatinine, 168 μmol/l (reference range, 70–100 μmol/l); potassium, 3.3 mmol/l (3.2–5.0 mmol/l); calcium, 3.44 mmol/l (2.2–2.5 mmol/l); ionized calcium, 1.95 mmol/l (1.17–1.29 mmol/l); phosphate, 1.19 mmol/l (0.8–1.4 mmol/l); urea, 13.3 mmol/l (3–7 mmol/l); alkaline phosphatase, 33 U/l (less than 110 U/l), albumin, 41 g/l (38–50 g/l); PTH, 2.7 pmol/l (1.0–5.8 pmol/l); 25(OH)D, >250 nmol/l (36–109 nmol/l winter, 39–208 nmol/l summer; the method used was DiaSorin Radioimmunoassay, which detects 25(OH)D2 and 25(OH)D3 equivalently, but laboratory did not dilute most samples to provide a more specific result). Her hemoglobin was 92 g/l (120–160 g/l); ferritin, 82 mcg (27–220 mcg/l); transferrin saturation, 0.28 (0.25–0.50); serum protein immunoelectrophoresis showed no monoclonal spikes; 24-h-urine calcium, 8.2 mmol/day (less than 7.0 mmol/day in the absence of renal failure), urine phosphorus 19 mmol/day (less than 50 mmol/day); creatinine clearance, 0.57 ml/s (1.33–2.16 ml/s). Whole-body bone scan showed no definite evidence of bone metastases. On ultrasound there was no suspicion of kidney stones or nephrocalcinosis. We reached the diagnosis of vitamin D intoxication because of the presence of hypercalcemia, hypercalciuria, a long history of high doses of vitamin D and very high serum 25(OH)D levels.

Other causes of hypercalcemia were initially excluded based on the laboratory findings and/or clinical presentation. Primary hyperparathyroidism was excluded initially because plasma iPTH was within the reference range (though still inappropriately high in relation to the hypercalcemia), and because serum phosphate was within the reference range. Sarcoidosis was ruled out because there was no adenopathy on clinical examination or X-ray. Plasma 1,25-dihydroxyvitamin-D values were not elevated. We also excluded thyrotoxicosis and familial hypocalciuric hypercalcemia.

Management and clinical course

Ergocalciferol was discontinued, and she was treated with prednisone 30 mg orally once a day, which she took for only 3 days and omeprazol by mouth (20 mg/day). She was given intravenous Pamidronate (60 mg over 6 h). After 3 days of prednisone administration and the injection of intravenous Pamidronate, her serum calcium promptly decreased from 3.44 mmol/l (on admission) to 2.59 mmol/l and iPTH levels unexpectedly increased from 2.7 (at admission) to 16.8 pmol/l (Fig. 1). Following this, she was started on Risedronate 35 mg weekly to protect her bones from the prednisone and previous observations have suggested that the hypercalcaemia of vitamin D intoxication is mediated by increased bone resorption and bisphosphonates have a role in its management [4].

Fig. 1
figure 1

Serum calcium, iPTH, and 25-hydroxyvitamin-D levels during follow-up

Full size image

Ten days after the prednisone was stopped, serum calcium increased to 2.91 mol/l and serum iPTH levels decreased to 3.8 pmol/l. For this reason prednisone 30 mg daily was started again and was then reduced to 15 mg p.o./day at the third week and 10 mg/day at the fourth week, and stopped on the fifth week. As a result, the serum calcium decreased again, though was still hypercalcemic, and PTH increased. Again, after the discontinuation of prednisone the serum calcium increased and PTH decreased; therefore, she was given another course of prednisone for 4 weeks, starting with 30 mg/day by mouth, reduced to 15 mg each day after 2 weeks and discontinued after 4 weeks of treatment. Table 1 shows the biochemical indices during these two courses of prednisone.

Table 1 Biochemical indices during follow-up
Full size table

Five months after her initial visit, we suspected primary hyperparathyroidism because the hypercalcemia had persisted with inappropriate high levels of serum iPTH. A sestamibi parathyroid scan showed foci of increased uptake consistent with multiple parathyroid or thyroid adenomas. Figure 1 shows the PTH, serum Ca, and 25(OH)D levels during the follow-up.

During the follow-up, serum 25(OH)D continued to exceed 250 nmol/l. After ten times dilution of the serum, the 25(OH)D measured 691 nmol/l on April 4th, 2004 and 590 nmol/l on May 17th, 2005 (Table 1). Eventually, after 2½ years, serum 25(OH)D concentration reached reference values, at 66 nmol/l. However, serum calcium concentration remained high at 2.9 mmol/l, and iPTH increased further to 17.6 pmol/l.

When the sestamibi parathyroid scan was repeated, it showed persistent activity in the region of the inferior pole of both lobes of thyroid, a finding that could represent bilateral parathyroid adenomas or hyperplasia. The patient was advised to undergo parathyroidectomy but she refused. Her serum creatinine remained stable during the follow-up period.

Discussion

Vitamin D plays an important role in both healthy individuals and certain disorders, such as in patients with chronic kidney disease [5, 6].

Our patient is interesting because of the unusually long duration of hypervitaminosis D. In previous vitamin D-intoxicated patients, the serum 25(OH)D declined with a half-life of about 2 months after dosage was discontinued [7, 8]. The present patient differed further from previously reported cases of vitamin D intoxication, because she also suffered from primary hyperparathyroidism. Her primary hyperparathyroidism was probably latent (with low PTH levels) because of a high 25(OH)D, severe hypercalcemia, and probably by elevated 1,25(OH)2D [9]. Vitamin D2 treatment precipitated hypercalcemia, thereby unmasking her primary hyperparathyroidism; other patients taking such doses (albeit vitamin D3, not D2) did not become hypercalcemic [8].

Why did it take so long for 25(OH)D to return to a normal concentration in this patient? Vitamin D is stored in both adipose and muscle tissue, and these represent a large volume of distribution within the body, with a relatively long pharmacologic half-life of about 2 months [10, 11]. Our patient was of slim build, so that a large pool of vitamin D in adipose tissue could not account for the slow decline in 25(OH)D. Vitamin D and its metabolites are stored in similar quantities in adipose and muscle tissue [10], and this storage is thought to be a passive, physiochemical distribution of molecules into its body compartments. There is no plausible way by which tissues could exhibit an amplified vitamin D storage behavior in a patient, and such a phenomenon has never been reported in the literature.

A herbal remedy, genistein, slows catabolism of 1,25(OH)2D in laboratory studies [12]. However, our patient did not take any herbal remedy which could affect vitamin D metabolism. We are also confident that the patient did not consume vitamin D surreptitiously.

The most likely explanation for the unusual nature of this case is that the combination of vitamin D excess together with primary hyperparathyroidism distinguishes the effects in our patient. Studies by Clements et al. showed that patients with hyperparathyroidism exhibit rapid metabolic inactivation of vitamin D [1], leading one to expect that primary hyperparathyroid patients should recover rapidly from vitamin D excess. Clements et al. also showed that administration of 1,25(OH)2D to nonhyperparathyroid patients likewise accelerated the metabolic clearance of 25(OH)D [13]. With physiologic concentrations of 25(OH)D, a balance between the signal due to increased 1,25(OH)2D, to remove 25(OH)D, and the signal due to PTH, to downregulate the catabolic enzyme 25(OH)D-24-hydroxylase, was reported by Clements et al. [1]. However, Nikkela et al. [14] evaluated vitamin D metabolites pre- and postoperatively in 49 patients with primary hyperparathyroidism and these authors found that the preoperative serum 24,25-dihydroxyvitamin D levels were significantly lower than postoperative levels. It was reported that PTH downregulates the renal 24-hydroxylase mRNA by affecting its stability and reducing the half-life of 24-hydroxylase mRNA by 4.2-fold in AOK-B50 cells [2].

We believe that, in our patient, the simultaneous presence of hyperparathyroidism resulted in an abnormally low capacity to upregulate 24-hydroxylase activity in the face of vitamin D toxicity. When serum 25(OH)D is unusually high, the presence of hyperparathyroidism creates an overriding limitation on the ability of 24-hydroxylase to upregulate in response to the excessive vitamin D metabolites that should stimulate 24-hydroxylase. The present investigation suffers from the shortcoming that we did not measure serum 24,25-dihydroxyvitamin D. The method was not available to us, and even if results were available, they would be of marginal value because there remains no reference value in vitamin D toxicity against which to compare the result.

It is impossible to address the question of whether the nonphysiologic nature of vitamin D2 could account for the toxicity in this case. There are biological differences between vitamin D2 and vitamin D3 [15], but the present case offers no insight into these. Furthermore, since the high-dose prescription forms of vitamin D in North America remain primarily the vitamin D2 form, the present case is clinically relevant.

In our patient, treatment with prednisone and bisphosphanate lowered the serum calcium concentration. These changes likely triggered an increase in the PTH concentration and enlargement of the parathyroid adenoma, as indicated by the increase in plasma PTH concentration without normalization of the serum calcium.

Prolonged hypercalcemia in primary hyperparathyroidism impairs renal function [16]. In our patient, prolonged hypercalcemia may have caused her chronic renal failure. Parathyroidectomy probably would have been beneficial in this patient. A similar effect on patients with chronic kidney disease/mineral and bone disorders has been described [17].

In conclusion, we believe that primary hyperparathyroidism impaired the capacity to mount a normal vitamin D-catabolic response of 24-hydroxylase activity. Therefore, the pharmacological amounts of vitamin D2 and its metabolites were not efficiently cleared from the body. The consequence was that it took an unusually long time for her serum 25(OH)D to return to normal.