Pharmacokinetics and pharmacodynamics of atracurium obtained with arterial and venous blood samples.

To determine the influence of sampling site on atracurium pharmacokinetic-pharmacodynamic relationships, blood was drawn simultaneously from the radial artery and peripheral vein during a 20-minute period after injection of atracurium, 0.2 mg/kg, in eight patients. Atracurium and laudanosine concentrations were measured by HPLC. Neuromuscular blockade was measured at the adductor pollicis, after stimulation of the ulnar nerve. Venous levels were lower than corresponding arterial values for up to 20 minutes, and this difference was marked for the early samples. Neuromuscular blockade was maximum after 5 to 7 minutes, much later than the peak venous concentration (1 to 3 minutes). Nonparametric analysis yielded (mean +/- SEM) a rate constant, concentration for 50% blockade, and slope of the effect-concentration relationship of 0.092 +/- 0.01 min-1, 379 +/- 27 ng/ml, and 7.3 +/- 1.67, respectively, when based on arterial samples. The values were statistically different (0.135 +/- 0.011 min-1, 235 +/- 42 ng/ml, and 3.41 +/- 0.37, respectively) when venous levels were used (p less than 0.05). It is concluded that forearm venous levels do not correspond to adductor pollicis neuromuscular blockade and the kinetics and kinetic-dynamic relationship for atracurium are heavily dependent on sampling site.