In this paper we elaborate a one-step procedure for the selective extraction of urinary 11-dehydrothromboxane B2 on octylsilyl silica cartridges for reliable determination with radioimmunoassay. The immunoreactivity profile of nonselectively extracted urine after HPLC separation showed that as much as 70% of the total 11-dehydrothromboxane B2 immunoreactivity comigrates with polar interfering material. Its amount could be considerably decreased using acetonitrile:water (18:82, v/v) as wash solvent before elution of 11-dehydrothromboxane B2 from the cartridge. Alternatively, very high immunoreactive purity was achieved without the preceding wash step by selective elution of the analyte with dichloromethane:hexane (70:30). After both optimized steps in the extraction procedure were combined, immunoreactivity was found only in HPLC fractions corresponding to the retention volume of authentic 11-dehydrothromboxane B2. The homogeneity of this immunoreactivity was confirmed by two-step HPLC separation. A significant correlation of values was observed between samples measured after extraction and those measured after subsequent HPLC purification. A high correlation was also found with concentrations determined by radioimmunoassay using four different antisera. The values of 24 h excretion of 11-dehydrothromboxane B2 in 10 male volunteers (595 +/- 114 ng/g creatinine, mean +/- SD) as well as the inhibitory effect of acetylsalicylic acid (80 +/- 13%) closely correspond with those reported in the literature. This selective extraction procedure provides a high validity in radioimmunoassay without requiring any further purification step.