Dose requirements for thiopental anesthetic induction have significant
age- and gender-related variability. We studied the association of th
e patient characteristics age, gender, weight, lean body mass, and car
diac output with thiopental requirements. Doses of thiopental, infused
at 150 mg/min, required to reach both a clinical end-point and an ele
ctroencephalographic (EEG) end-point were determined in 30 males and 3
0 females, aged 18-83 yr. Univariate least squares linear regression a
nalysis revealed outliers in the relationships of age, weight, lean bo
dy mass, and cardiac output to thiopental dose at clinical and EEG end
points. Differential weighting of data points minimized the effect of
outliers in the construction of a robust multiple linear regression mo
del of the relationship between several selected independent variables
and the dependent variables thiopental dose at clinical and EEG endpo
ints. The multiple linear regression model for thiopental dose at the
clinical end-point selecting the regressor variables age, weight, and
gender (R2 = 0.76) was similar to that for age, lean body mass, and ge
nder (R2 = 0.75). Thiopental dose at the EEG end-point was better desc
ribed by models selecting the variables age, weight, and cardiac outpu
t (R2 = 0.88) or age, lean body mass, and cardiac output (R2 = 0.87).
Although cardiac output varied with age, age always remained a selecte
d variable. Because weight and lean body mass differed with gender, th
eir selection as variables in the model eliminated gender as a selecte
d variable or minimized its importance.