The 300 000-IU bolus of vitamin D2 or D3 was practical, well tolerated, and safe. Vitamin D3 had greater potency than equimolar vitamin D2, with a higher, sustained serum 25(OH)D response and efficacious PTH suppression. To adequately treat vitamin D insufficiency we would recommend administering 300,000 IU oral vitamin D3 approximately three times per year.

There is an increasing interest in the role of vitamin D as a potential treatment for a number of disparate diseases. In addition to its role in calcium homeostasis, vitamin D has a plethora of effects including immunomodulation, pleiotropic effects, modulating propensity to infection and blood pressure regulation. Detection and treatment of vitamin D deficiency in selected patients with RA is relevant as deficiency is common. Vitamin D therapy may modify the increased risk of falls and fracture in this group, and possibly exert additional immunomodulatory effects on disease onset and activity although data are largely epidemiological. Currently, there is no consensus view on vitamin D replacement regimens, nor an agreed optimal level of serum 25-hydroxyvitamin D [25(OH)D] for health. Indeed levels may vary for different organ systems and the concept of 'tissue specific vitamin D deficiency' needs to be considered. Therefore, there is clinical uncertainty regarding both when and how to correct vitamin D deficiency. Older patients, particularly post-menopausal women, and others at high risk of vitamin D deficiency should be preferentially targeted since they are likely to benefit most from supplementation. Clinicians should be aware of the technical difficulties associated with measuring and interpreting 25(OH)D levels. The administration of high-dose vitamin D as an oral weekly bolus is safe and can rapidly correct vitamin D deficiency followed by regular lower doses to maintain adequate levels.

washed, the Vitros Signal Reagent was added to the well, and the luminescence was measured (ALOKA luminometer). The IgG fraction of serum samples from the patient and from five control individuals was purified with a MAbTrap TM Kit (Amersham Biosciences), and the IgG concentration was adjusted to 4.0 g/L. The purified IgG (0.08 mL) was incubated at 4°C for 24 h with 0.08 mL of PBS alone or with T 2 , T 3 , or T 4 (Sigma; at 4570, 42, and 768 nmol/L, respectively) dissolved in PBS. Each sample was mixed vigorously with 1.2 mL of polyethylene glycol (PEG; 125 g/L), centrifuged at 2800g for 30 min, aspirated, and washed with 1.2 mL of PEG (125 g/L). After the precipitates were dissolved in 0.001 mol/L hydrochloric acid (0.04 mL) and neutralized by equal amounts of 0.001 mol/L sodium hydroxide, T 2 and T 3 were measured by the FT 3 assay and T 4 by the FT 4 assay. The concentrations of T 2 were expressed as T 3 concentrations. Each sample was analyzed in duplicate. The ability of the purified IgG to bind T 2 , T 3 , or T 4 was defined as the difference between the FT 3 or FT 4 assay result and the respective blank value and is reported as the ␦T 2 , ␦T 3 , or ␦T 4 value.The ratios of FT 3 and FT 4 concentrations in PEG-treated samples (6 ) to those in untreated samples were significantly lower for the patient than for 37 other patients (Table 1 in the online Data Supplement); therefore, immunoglobulins in the patient's serum interfered with both the FT 3 and FT 4 assays. FT 3 and FT 4 values in the mixtures of serum with sheep IgG, bovine globulin, and gelatin did not differ significantly from those in the mixtures of serum and PBS only, suggesting that heterophilic antibodies and anti-gelatin antibodies did not cause the high FT 3 and FT 4 values.When we examined the patient's IgG binding with Vitros FT3II and FT 4 assay wells, the luminescence generated by the patient's serum was higher than that of the 5 control individuals (Table 1 in the online Data Supplement). This suggested that the patient's IgG bound to T 2 -and T 3 -gelatin.The FT 3 and FT 4 concentrations in purified IgG and in treated samples of purified IgG (6 ) were below the lower detection limits of the Elecsys assays, suggesting an absence of T 3 and T 4 contamination in the IgG fractions. The patient's ␦T 2 and ␦T 3 values were higher than those of the 5 control individuals, but the ␦T 4 value was within 2 SD of the values for the 5 controls (Table 1 in the online Data Supplement). This finding implies that the patient's IgG interacted with T 2 and T 3 but not with T 4 . The crossreactivity of the anti-T 3 antibody with T 2 in the Vitros FT3II assay was very low, whereas the patient's ␦T 2 value was evidently higher than that of 5 control individuals. We conclude that T 2 was bound to the patient's IgG.Because anti-gelatin antibodies in the patient's serum were not recognized, we suggest that the interfering substance were antibodies to T 2 and T 3 . As the interfering antibodies did not interfere with the Elecsys FT 3 assay, the interfering antib...