1 A mefloquine hydrochloride tablet (250 mg base equivalent to 4.8 ± 0.6 mg kg-'; mean ± s.d.) and deuterium labelled mefloquine hydrochloride solution (250 mg base) were given to six adult male Thai patients with acute falciparum malaria and six healthy Swiss adult male volunteers (equivalent to 3.5 ± 0.1 mg kg-').2 The relative bioavailability of the tablet formulation derived from comparison of the areas under the plasma concentration-time curves was similar in both groups; 87 ± 11% and 89 ± 10% (mean ± s.d.). 3 The rate of drug absorption appeared to be similar in the two groups but peak plasma mefloquine concentrations were approximately three times higher in the Thai patients (1004 ± 276 ng ml-' for the tablet and 1085 ± 280 ng ml-' for the suspension) compared with the Swiss volunteers (319 ± 73 ng ml-' for the tablet, and 369 ± 121 ng ml-' for the suspension). 4 Estimates of the oral clearance CLpo of unlabelled mefloquine were significantly lower (17.5 ± 4.4 ml h-1 kg-1) in the Thai patients compared with 28.8 ± 3.5 ml h-1 kg-1 in the Swiss volunteers; P < 0.05). Terminal elimination half-lives were significantly shorter in the patients (10.3 ± 2.5 days) than in the volunteers (16.7 ± 1.9 days; P < 0.005). Differences of a similar magnitude were observed when comparing the pharmacokinetic parameters derived from the deuteromefloquine plasma concentrations. 5 Both genetic and disease related factors are likely to account for the large pharmacokinetic differences between the two groups. These findings have implications for the planning of dose finding studies and the use of mefloquine in Thailand.

The pharmacokinetics of single- and multiple-dose administration of tenoxicam 20 mg were evaluated in 8 healthy males. Maximum plasma concentration (Cmax) after the first dose was 2.76 +/- 0.48 micrograms/ml (mean +/- s.d.) and the time to reach Cmax (Tmax) was 5.0 +/- 3.0 h. The area under the plasma concentration-time curve (AUC0-infinity) after a single administration of tenoxicam was 242.5 +/- 73.5 micrograms x h/ml. The elimination half-life (t1/2) was 66.3 +/- 15.8 h and the plasma concentration at 24 hours after dosing (Cmin) was 1.84 +/- 0.33 micrograms/ml. Steady-state plasma concentrations of tenoxicam were virtually reached after 10 consecutive daily doses. At steady-state, Cmax averaged 13.63 +/- 3.33 micrograms/ml and Tmax remained 5.0 +/- 3.0 hours. AUC within a dosing interval at steady-state was 262.2 +/- 67.0 micrograms x h/ml, Cminss was 9.67 +/- 3.25 micrograms/ml, and t1/2 averaged 74.2 +/- 13.3 h. The average fluctuation during multiple-dose administration was 26.8 +/- 8.0% and the accumulation ratio was 5.82 +/- 0.60. Steady-state pharmacokinetic parameters predicted from the first-dose data slightly underestimated observed values, but the results supported the assumption of linear pharmacokinetics during multiple-dose tenoxicam administration.

Seven healthy male volunteers (2 1-24 y) received by the ventro-gluteal route a single dose of 100 000 I.U. of the vitamin retinyl palmitate (RP) in a water-miscible preparation (w) and 5 weeks later the same dose in an oily solution (S). After administration of W median (range) peak plasma concentrations of 5*6(44-8.7). lo3 pg 1--' were reached after 12 h and high levels persisted for another 50 h. At 144 h levels were still, by a factor 3, higher than baseline. Plasma levels of RP after S remained close to baseline (20-50 pg. 1-l) suggesting negligibly low bioavailability. The plasma level profile of R P after W could well be described by use of a one-compartment model with Weibulltype absorption and Michaelis-Menten elimination. The median (range) absolute bioavailability (estimates of lower limits) was 42 (32-52) per cent.

The pharmacokinetics of glibornuride (25 mg i.v.) and the accompanying insulin and glucose responses were characterized in eight human subjects in the presence and absence of steady‐state tenoxicam (20 mg p.o./day for 2 weeks). Tenoxicam affected neither the pharmacokinetic parameters of glibornuride (systemic clearance, volume of distribution and biological half‐life) nor the responses of plasma insulin and blood glucose to glibornuride. The single i.v. dose of glibornuride had no detectable effect on the kinetics of tenoxicam.

The effect of the synthetic retinoid acitretin (A) on the disposition of blood glucose and on the serum insulin response following the IV infusion of 139 mmol glucose over 10 min (IGTT) has been investigated in six healthy subjects. The IGTT was performed on Days 1, 10 and 24. On Days 3 to 10 A 50 mg/d was administered. Several parameters of glucose disposition and insulin response (K-values, AUC) were assessed. As a methodological variant, the profiles over time of blood glucose and serum insulin were evaluated by model calculations using the 'minimal model'. Acitretin did not influence any parameter of glucose disposition. The area under the insulin-time curve (baseline corrected) was significantly decreased from 1.20 mU.min.l-1 on Day 1 to 0.89 mU.min.l-1 on Day 10, and was 0.91 mU.min.l-1 on Day 24. The model-derived 'insulin sensitivity' increased from 13.10(-4) l.mU-1.min-1 on Day 1 to 20.10(-4) l.mU-1.min-1 on Day 10 and was 18.10(-4) l.mU-1.min-1 on Day 24. The results suggest that A increased sensitivity to endogenous insulin. It supports a recent report showing greater insulin sensitivity in patients treated with the synthetic retinoid etretinate.