A modified high-performance liquid chromatography (HPLC) method, based on Davidson and Sadowski [Meth. Enzymol. 282 (1997) 408], with fluorescence detection after zinc postcolumn reduction was developed and validated for the analysis of phylloquinone (vitamin K1) in plasma or serum samples. Compensation for procedural losses of vitamin K1 was made by the method of internal standardization using a proprietary vitamin K derivative. Increased sensitivity of detection by the use of a high-sensitivity Waters 440 fluorescence detector and optimized chromatography conditions increased the sensitivity to 4 fmol vitamin K1. The response was linear and free from interfering peaks and from baseline drift. It is therefore adequately sensitive for 0.25 ml or less of plasma sample. Long-term reproducibility of quality assurance (QA) samples was verified over a period of 4 months. The intraassay precision estimates of the QA samples within-run with mean vitamin K1 concentrations of 0.4, 1.4 and 3.4 nmol/l were 5.2% (n=6), 8.2% (n=6) and 3.0% (n=12), respectively, while interassay precision estimates between runs were 16% (n=22), 12% (n=21) and 8.1% (n=15), respectively. The assay accuracy was validated by comparing the results we obtained for 14 samples from the Vitamin K External Quality Assessment Scheme (KEQAS) with the consensus of the results from the other participating laboratories. Good agreement was obtained, with y=1.06x-0.09, R2=0.99. Validation also included linearity of response, absence of interference and confirmation of vitamin K1 peak purity.