Direct relaxant effects of intravenous anesthetics on airway smooth muscle.

Ketamine, at concentrations achieved with the usual clinical doses, has a direct relaxant effect on airway smooth muscle (ASM). This study investigates the dose-dependent direct relaxation effects of midazolam and propofol on both proximal and distal ASM compared with ketamine. The proximal and distal airways were dissected from eight mongrel dogs and cut into 2-mm rings. The rings were attached to pressure transducers and equilibrated in a Krebs-Ringer bicarbonate bath kept at 37 degrees C, pH 7.4, CO2 37 mm Hg, and PaO2 > 100 mm Hg. Optimal length was determined, a dose-response curve to acetylcholine was established, and the 50% effective dose (ED50) of acetylcholine was calculated. Ketamine, midazolam, or propofol were given in random order to each ring preconstricted with ED50 of acetylcholine in cumulative log incremental doses from 10(-6) to 10(-4) M. Relaxation response was the tension during anesthetic equilibrium, expressed as a percentage of the tension from ED50 of acetylcholine. The drug vehicles were tested for their effects on the ASM. No bronchorelaxation was seen with any of the intravenous anesthetics at 10(-6) M. Ketamine 10(-5) M produced at 17.9% +/- 2.1% relaxation in the distal ASM but had no effect on the proximal ASM. Neither propofol nor midazolam affected the ASM at 10(-5) M. The distal ASM was significantly (P < 0.005) more sensitive to 10(-4) M of all three drugs compared with the proximal ASM. In the proximal ASM, 10(-4) M of ketamine, midazolam and propofol reduced ASM tension by 14.9% +/- 4.4%, 19.0% +/-8.8%, and 14.7% +/- 5.5%, respectively, versus 36.4% +/- 3.2%, 58.6% +/- 6.1%, and 64.4% +/- 9.0% in the distal ASM. The drug vehicles had no effect on the ASM. We conclude that ketamine, midazolam, and propofol have direct relaxant effects on ASM. All three intravenous anesthetics have a greater direct relaxant effect on distal ASM than on proximal ASM. Only ketamine showed significant direct bronchorelaxing effects at concentrations that are likely to be achieved with the usual clinical dosing patterns.