The pharmacokinetics of orally-administered didanosine were evaluated in 6 male and 6 female HIV seropositive patients to determine the effect of pretreatment with metoclopramide, an inducer of gastrointestinal motility, and loperamide, which retards motility. Using a randomized, balanced, crossover design, each patient received the following three treatments under fasting conditions: didanosine as a single agent, didanosine 5 min after a single 10 mg intravenous dose of metoclopramide, and didanosine 1 h after the final of 4 doses, 4 mg each, of loperamide. Serial blood and urine samples were collected for up to 12 h after each dose. Plasma and urine aliquots were analysed for intact didanosine using HPLC with UV detection. Pharmacokinetic parameter values were calculated using noncompartmental methods. The mean Cmax values were significantly greater for the didanosine single agent (2.04 micrograms.ml-1) and didanosine with metoclopramide (2.30 micrograms.ml-1) treatments than for the combination of didanosine with loperamide (1.57 microgram.ml-1). The t1/2 in males was significantly greater than in females for the didanosine (1.75 vs 1.12 h, respectively) and didanosine with metoclopramide treatments (1.74 vs 1.20 h, respectively). No significant treatment or gender effects were observed for AUC or UR (urinary recovery). The pharmacokinetics of didanosine were not altered appreciably by dosing with metoclopramide. Administration with loperamide affected the rate but not the extent of absorption. There were no clinically relevant differences between males and females in the disposition of didanosine.