Skin conductance responses in patients sedated with midazolam or propofol.

We have measured the skin conductance response under resting conditions and to innocuous auditory stimuli in 45 patients receiving midazolam (group M), propofol (group P) or no sedative drug (group ND) before minor hand surgery under local anaesthesia. Administration of the sedative drugs was titrated to the end-point of slurring of speech and ptosis. The mean dose of midazolam was 0.06 (SD 0.01) mg kg-1 and the mean infusion rate of propofol was 2.2 (0.39) mg kg-1 h-1. Subjective ratings of anxiety and sedation were measured using visual analogue scales. These were similar in groups M and P and significantly different from those reported by group ND (P = 0.001-0.005). However, measures of skin conductance in group M were significantly lower than in group P (P = 0.002-0.013) and group ND (P = 0.004-0.016). These measures were similar in groups P and ND. Skin conductance measures were related significantly to anxiety scores only in groups M and ND (P < 0.05). We conclude that skin conductance is not a non-specific index of sedative-anxiolytic action and therefore is not useful in comparative studies of anxiolytic drugs that exert their effects by different pharmacological mechanisms.