We have previously reported an attenuated plasma cortisol stress respo
nse in alcoholics abstinent from alcohol up to 4 weeks. The present st
udy replicates and extends these findings by examining urinary cortiso
l levels in detoxified alcoholics (n = 40) and controls (n = 14) at re
st and following mental arithmetic and isometric handgrip stress. Alth
ough the groups had similar baseline cortisol levels, the alcoholics s
howed an attenuated cortisol response to the combined stressors. This
cortisol response reduction was unrelated to potential confounds such
as smoking, liver function, are, depression, or anxiety. A multivariat
e model showed a trend for an association between severity of withdraw
al and alcoholics' poststress cortisol levels. Although these results
indicate decreased adrenocortical response to biobehavioral stress in
alcoholics abstinent up to four weeks, higher stress cortisol values w
ere seen in the patients with the most severe withdrawal symptoms.