鈥� Nineteen patients, each hospitalized with a major depressive episode, were deprived of sleep for one night. Ten patients responded with clear improvement in depressive symptoms; the substantial clinical change was transient, usually lasting one day. Those who responded had significantly higher initial depression ratings (P <.01) and tended to be older than nonresponders who experienced mild increases in irritability, fatigue, and discomfort following sleep deprivation. Amine metabolites, 5-hydroxyindoleacetic acid (5HIAA), and homovanillic acid (HVA) were not substantially affected by sleep deprivation, although there was a significant interaction of clinical response and direction of 3-methoxy-4-hydroxyphenylglycol (MHPG) change. Sleep deprivation thus produces acute, but only transient improvement in a selected group of severely depressed patients; it appears to be an important tool in the study of the affective disorders.