Inter- and intraindividual variability in the concentration-effect (sedation) relationship of flunitrazepam

1. The relationship between plasma flunitrazepam concentrations and the degree of sedation was evaluated in 20 healthy subjects receiving two single oral doses of 1 mg flunitrazepam on two different occasions (1 week apart). The degree of sedation was rated blindly during the two treatment sessions in parallel with blood sampling (48 h). 2. A strong correlation between the concentrations of flunitrazepam in plasma and the degree of sedation was found according to the sigmoid Emax model. The plasma drug concentration producing 50% of maximal effect (EC50) was found to be 7.0 and 6.5 ng ml-1 on the two occasions, respectively. The variability in EC50 between subjects was larger (C.V. 39%) than the variability within subjects (C.V. 27%). 3. The steepness of the concentration-response curve as reflected in the slope factor(s) showed a virtual 'all or none' response to flunitrazepam with s values ranging from 3 to 30 with a mean of about 14. 4. The results in young healthy subjects suggest that the present dosage recommendations for temporary insomnia (1-2 mg) may be inappropriate; the dose can probably be reduced to 0.5 mg in some patients to achieve moderate sedation.