Controversy about the appropriate evaluation of penetrating abdominal, flank, and back injuries prompted this retrospective review of 311 patients at an urban level I trauma center over 5 years. Seventy-five (24%) patients sustained gunshot wounds to the abdomen. All patients with gunshot wounds underwent exploratory laparotomy; results of 67 laparotomies (89%) were positive. Of 236 patients sustaining stab wounds (140 abdominal wounds, 51 flank wounds, 26 back wounds, and 19 wounds to multiple sites), 147 were treated according to a selective protocol, based on results of physical examination, wound exploration, peritoneal lavage, and ancillary diagnostic studies. No injuries were found at celiotomy in three (2%) of these 147 patients. One false-negative result of evaluation of a flank wound occurred. Significant injuries were found in 13 patients (68%) with stab wounds to multiple sites, 61 patients (44%) with abdominal stab wounds, 15 patients (29%) with flank stab wounds, and four patients (15%) with back stab wounds. Mandatory exploration of gunshot wounds is justified. Physical findings of intra-abdominal injury or positive results of peritoneal lavage identify stab wound victims likely to benefit from surgical exploration. A policy of mandatory observation or routine celiotomy for treatment of stab wounds is not justified.