Plasma exchange induces vitamin D deficiency

Abstract

Background: Plasma exchange is used in the treatment of diseases mediated by pathogenic circulating proteins, or for transplant desensitization. Its non-targeted nature results in the depletion of physiologically important molecules, and it is often complicated by hypocalcaemia.

Aim: To determine the effects of plasma exchange on vitamin D binding protein (DBP) and associated vitamin D metabolites.

Design: Single-centre prospective cohort study of 11 patients.

Methods: DBP and vitamin D metabolites were measured before and immediately after five plasma exchanges, and 7 and 28 days after discontinuation of plasma exchange.

Results: Plasma exchange reduced plasma DBP concentration from 196.9 ± 53.2 to 98.5 ± 34 μg/ml (P = 0.0001), 1,25-dihydroxyvitamin D from 103 ± 52 to 42 ± 4 pmol/l (P = 0.003) and 25-hydroxyvitamin D from 49.7 ± 29 to 22 ± 9.4 nmol/l (P = 0.0017), through their removal in effluent. After 7 days, DBP and 1,25-dihydroxyvitamin D were not significantly different from baseline, but 25-hydroxyvitamin D remained significantly lower after 7 days (26.4 ± 9.8 nmol/l, P = 0.02) and 28 days (30.8 ± 15.5 nmol/l, P = 0.048). Corrected calcium decreased from 2.23 ± 0.11 to 1.98 ± 0.08 mmol/l (P = 0.0007) immediately after five treatments. Plasma calcium was significantly associated with 1,25-dihydroxyvitamin D (r(2) = 0.79, P < 0.0001).

Conclusion: Plasma exchange induced an acute reversible decrease in plasma 1,25-dihydroxyvitamin D, DBP, calcium and a sustained decrease in plasma 25-hydroxyvitamin D.

Figure 1.

Effect of plasma exchange on 25-hydroxyvitamin D and DBP. Statistical comparisons are made with baseline values for each variable. Plasma exchange significantly reduced DBP, although this had recovered to baseline after 7 days. In contrast, 25-hydroxyvitamin D concentrations remained significantly lower than baseline after 7 and 28 days.

Figure 2.

Pre-exchange, effluent and post-exchange DBP and 25-hydroxyvitamin D values in a subset of three patients. Bars show mean (standard error) values of (A) DBP and (B) 25-hydroxyvitamin D levels in plasma before and after the first plasma exchange (black), and in the effluent from this exchange (grey).

Figure 3.

Relationship of plasma calcium and 1,25 dihydroxyvitamin D concentrations. Scatterplot and linear regression demonstrating the correlation of plasma corrected calcium and 1,25 dihydroxyvitamin D (r² = 0.79, P = 0.0001, β-coefficient = 0.002). Values represent measurements for all patients from pre-treatment until immediately after the final plasma exchange.