Clinical analysis of vitamin B(6): determination of pyridoxal 5'-phosphate and 4-pyridoxic acid in human serum by reversed-phase high-performance liquid chromatography with chlorite postcolumn derivatization.

A reversed-phase high-performance liquid chromatography (HPLC) method with fluorometric detection was developed for the routine determination of pyridoxal 5'-phosphate (PLP) and 4-pyridoxic acid (4-PA) in serum. Chlorite postcolumn derivatization was used to oxidize PLP to a more fluorescent carboxylic acid form. Sensitivity improved fourfold for PLP using chlorite postcolumn derivatization over traditional bisulfite postcolumn derivatization. The HPLC injection cycle was 15 min, facilitating a throughput of 60 patient samples (72 injections that included standards and quality control (QC) samples) in 18.5h. Method precision was evaluated using three serum QC pools with PLP and 4-PA concentrations of 11.5-34.8 nmol/L and 10.4-21.0 nmol/L, respectively. Within-run (n=7) repeatabilities were 0.6-1.2% for PLP and 0.9-1.8% for 4-PA. Run-to-run (n=23) reproducibilities were 3.6-6.7% for PLP and 3.7-5.6% for 4-PA. Relative detection (3sigma(0)) and quantitation (10sigma(0)) limits were 0.3 and 0.9 nmol/L, respectively, for both PLP and 4-PA using a 10-microl sample injection volume. Analytical recoveries ranged from 97 to 102%. Patient-matched serum and plasma specimens (n=25) were analyzed to evaluate specimen-type bias. Of the plasma types evaluated, heparinized plasma introduced the lowest relative bias for PLP (-5.3%) and minimal bias for 4-PA (-2.3%) compared with serum. Ethylenediaminetetraacetic acid (EDTA) plasma showed the lowest bias for 4-PA (0.7%) but a relatively high bias for PLP (13.0%) due to a chromatographic interference. Human serum samples from a non-representative population subset (n=303) were commensurate with values published for other vitamin B(6) HPLC methods. These values gave geometric means of 42.4 nmol/L for PLP and 27.3 nmol/L for 4-PA. Medians for PLP and 4-PA were 40.1 and 21.8 nmol/L, respectively. The high sensitivity, precision, and throughput of this method, combined with its minimal serum specimen (150 microl) and sample injection (10 microl) volume requirements, make it well suited for routine clinical vitamin B(6) analysis.