[The effect of a thigh tourniquet on the pharmacokinetics of midazolam].

Although the general pharmacokinetics of midazolam (M) are well documented, little is known about the possible effects of a thigh tourniquet on the distribution and elimination of this drug. METHOD. Institutional approval for the study and individual informed consent were obtained. We studied 30 patients (ASA-I) without premedication who electively underwent a surgical procedure of the lower limb. Patients were divided into three groups of 10. The procedure was done in groups I and II with and in group III without tourniquet use. Anesthesia was induced in groups I and II with 0.1 mg/kg M, fentanyl 5 micrograms/kg, alcuronium-dichloride 0.15 mg/kg and etomidate 0.1-0.2 mg/kg i.v. and maintained with enflurane 0.3-1.0 vol.-%. About 20 min after midazolam injection and after exsanguination the tourniquet was applied on the proximal thigh in group I. In group II anesthesia was induced with etomidate 0.2 mg/kg and alcuronium-dichloride 0.15 mg/kg i.v., and maintained about 20 min with enflurane 1.0-1.5 vol.-% until exsanguination and tourniquet application. After this, these patients also received 0.1 mg/kg M and 5 micrograms/kg fentanyl i.v. Through an indwelling arterial line, blood samples were obtained prior to and 2, 15, 30, 45, 60, 75, 90, 105, 120, 135, 150, 165 and 180 min after M injection. Plasma M levels were measured by high-performance liquid chromatography with UV detection. These concentrations were fitted to a two-compartment open model. Comparison between groups was performed using the Kruskal-Wallis test and p less than 0.05 was considered to indicate significance. RESULTS. The groups were all comparable in age and weight, and groups I and II also in duration of thigh ischemia. Midazolam elimination half-time (t beta 1/2) was significantly shorter in group II than in groups III and I (52 min vs 126 min and 139 min; p less than 0.05). Of the calculated distribution volumes (volume of the central compartment, volume in the steady state and volume in the elimination phase), only the volume in the steady state was significantly smaller in group II than in groups III and I (p less than 0.05). Groups III and I did not differ significantly in the computed parameters. The measured initial midazolam mean concentrations in group II were twice those in groups III and I (655 ng/ml vs 323 ng/ml and 332 ng/ml). Since clearance was not significantly different between any two groups, the shorter t beta 1/2 in group II was probably due to the reduced distribution volume. CONCLUSION. These data demonstrate that in the presence of a thigh tourniquet the timing of the injection - before or after application of the tourniquet is of decisive importance. Injection after the application of a tourniquet leads to an higher plasma level and shortens the elimination half-life.