There is clinical and biological evidence suggesting that patients with anger attacks, sudden spells of anger accompanied by intense autonomic activation, may represent a distinct psychopathological subgroup of patients with depressive disorders. We compared the prevalence of anger attacks in 168 outpatients with atypical depression or primary dysthymia with 38 normal subjects and tested the effect of treatment on anger attacks in a double-blind, placebo-controlled trial of sertraline versus imipramine. Patients were randomly assigned to sertraline (n = 56), imipramine (n = 52), or placebo (n = 60) and were administered the Anger Attacks Questionnaire before and after treatment. Depressed outpatients were significantly more likely to report anger attacks than controls. Anger attacks ceased in 53 percent of the patients receiving sertraline, 57 percent of those receiving imipramine, and 37 percent of those in the placebo group. Our findings support previous studies indicating that anger attacks are more prevalent among depressed outpatients than normals. Our results also suggest that sertraline and imipramine may be more effective than placebo in reducing the number of anger attacks following treatment although the differences were not statistically significant. Larger studies are needed to confirm our findings.