Benzodiazepine pharmacodynamics: utility of eye movement measures

The utility of several measures of saccadic and smooth pursuit eye movements as benzodiazepine pharmacodynamic measures was explored in 24 psychiatrically and medically health control subjects. Measures of sedation and memory impairment were also included. Subjects received four logarithmically increasing doses of intravenous diazepam at 15-min intervals on 1 day resulting in monotonically increasing plasma diazepam levels, and placebo on another day in random order 1 week apart. Measures were collected twice at baseline, once after each dose of diazepam/placebo and twice more, 15 and 30 min after the last dose. Peak saccadic velocity and smooth pursuit gain showed the greatest overall and dose-dependent drug effect among eye movement measures. Although effect sizes at the highest dose for memory impairment and self-rated sedation were comparable to these two measures, reliability (i.e., placebo-day fluctuation) with these measures was considerably poorer. Log-linear pharmacodynamic modeling was used to calculate the effective dose (ED30) or concentration (EC30) required to reduce saccadic velocity or pursuit gain by 30%. Almost all (23/24) subjects had linear and easily interpretable plots for saccadic velocity, while a majority (19/24) of subjects had interpretable plots for smooth pursuit gain. The distribution of ED30 and EC30 values showed a wide range of sensitivities to diazepam. These findings suggest that saccadic velocity and smooth pursuit gain are sensitive, reliable, quantitative benzodiazepine pharmacodynamic measures.