We investigated the pharmacokinetic interaction between 3'-azido-3'-deoxythymidine (AZT) and 2',3'-dideoxyinosine (ddl), given to rats by intravenous injection. For both compounds, the clearances, terminal half-lives, and fractions of the dose excreted unchanged in urine were not altered by the other drug (P > 0.05), indicating no pharmacokinetic interaction between the two drugs.2',3'-Dideoxyinosine (ddI) and 3'-azido-3'-deoxythymidine (AZT) are effective agents in the treatment of patients afflicted with the human immunodeficiency virus (4,19,20). Both nucleoside analogs are converted intracellularly to their triphosphates, which are potent inhibitors of the reverse transcriptase needed for viral replication (6,8). Drug combinations are now used to reduce toxicity, explore therapeutic synergy, and reduce the risk of human immunodeficiency virus resistance (6,11,21). For example, AZT is being used in combination with ddI (11), dideoxycytidine, or acyclovir (18,21). In this study, we investigated the pharmacokinetics of ddl and AZT given to rats separately and in combination by the intravenous route, to gain insights into the possible pharmacokinetic interaction of these two drugs.ddl (lot 234-B-1), AZT (lot RK03-222), and ftorafur [N1-(2-tetrahydrofuranyl)-5-fluorouracil] were gifts from the National Cancer Institute (Bethesda, Md.). High-pressure liquid chromatographic (HPLC) analysis showed that these compounds were >98% pure. Permanent catheters were implanted into the right jugular veins of female Fischer rats (ages, 5 to 6 months; pretreatment weights, 200 + 21 g [mean ± standard deviation; n = 18]) at least 16 h before the study. Between 8 and 10 a.m. the rats received a dose of ddI and/or AZT by intravenous injection. In patients, the concentrations in plasma are less than 10 to 20 ,ug/ml for AZT (9) and ddl (10). In this study, we used a dose of 40 mg/kg to give maximal concentrations in plasma of approximately 100 ,ug/ml for both drugs, because pharmacokinetic interactions such as inhibition of drug metabolism are expected to be more pronounced in the presence of higher concentrations in plasma. The intravenous dose (pH 7.0) was administered over 30 s. Serial venous blood samples (0.25 ml) were obtained over 180 min, and the plasma fractions were stored at -20°C until analysis. ddI and AZT were extracted from plasma samples as described elsewhere (15), with a slight modification that ftorafur was used as the internal standard. Both compounds were analyzed by HPLC. The mobile phase was 10 mM sodium phosphate buffer (pH 5.0) and 4% acetonitrile for ddl and 10 mM sodium phosphate buffer (pH 6.9) and 10% acetonitrile for AZT. The detection limit for both compounds in plasma was 0.1 ,ug/ml. The intraday variations were 5.1% for ddI and 7.4% for AZT. The interday variations were about 10% for both drugs. Urine samples were diluted 10-to 100-fold, spiked with the internal * Corresponding author. standard solution, and analyzed by HPLC without extraction.The plasma concentration-time data for ddI were analyzed...