Sixty male veterans with posttraumatic stress disorder (PTSD) participated in an 8-week, randomized trial comparing phenelzine (N = 19), imipramine (N = 23), and placebo (N = 18). Mean treatment retention was better on phenelzine (7.4 weeks) than on imipramine (5.6 weeks) or placebo (5.5 weeks). By week 5, both medications significantly reduced PTSD symptoms, as assessed by the Impact of Events Scale (IES), but the 44% improvement on phenelzine was greater than the 25% improvement on imipramine. The intrusion, but not the avoidance, subscale of the IES showed significant improvement, and the initial mild to moderate depressive symptoms did not significantly improve.