The objective of this study was to compare the precision and accuracy of fluorescence polarization immunoassay (FPIA) with high-performance liquid chromatography (HPLC) for measurement of procainamide (PA) and N-acetylprocainamide (NAPA) concentrations in urine. To determine the correlation between FPIA and HPLC, urine PA and NAPA concentrations were assayed using both techniques in samples obtained from study patients receiving PA and in spiked samples. In samples from patients, FPIA-determined PA and NAPA concentrations were 19 +/- 9% lower and 28 +/- 31% higher, respectively, than those determined by HPLC. The slope of the FPIA-HPLC regression lines for PA and NAPA differed significantly from that of the line of unity (the slope that would result if FPIA and HPLC yielded identical concentrations). In spiked samples, FPIA-determined PA and NAPA concentrations were 15 +/- 2% and 11 +/- 2% lower than HPLC-determined concentrations, respectively, and the slopes of the FPIA-HPLC regression lines differed significantly from the line of unity. Therefore, FPIA cannot be recommended as a urine assay method when quantitative assessment of urine PA or NAPA excretion is needed for pharmacokinetic studies.