Traumatic cardiac or pulmonary arrest is often associated with a dismal outcome and is considered by many to be an example of medical futility and inappropriate use of resources. This study aimed to identify the predictors of survival in patients experiencing traumatic cardiac arrest. We retrospectively reviewed all trauma patients undergoing cardiopulmonary resuscitation on arrival to the Emergency Department (ED) at an American College of Surgeons-designated Level I trauma center over 4 years. ED survival, hospital survival, and neurologic outcomes on discharge were the primary outcomes. Survival rates were examined in relation to demographics, mechanism of injury, airway management, cardiac electrical rhythm, and pupil size and reactivity. Statistical analyses used chi-square and t tests, P < 0.05 was considered significant. A total of 195 patients arrived in the ED with traumatic cardiac arrest; 34 were pronounced dead on arrival (no signs of life), and no resuscitation efforts were initiated. Of the remaining 161 patients 53 (33%) survived to leave the ED, and only 15 (9%) left the hospital alive. Demographic features were similar in survivors and nonsurvivors. The setting of intubation (prehospital vs ED) did not influence survival (P = 0.36). Penetrating trauma adversely affected survival in the ED (P = 0.01); however, this only approached significance in the final outcome of hospital survival (P = 0.06). The presence of sinus rhythm and nondilated reactive pupils was highly significant in predicting ED and hospital survival (P = 0.001). No patient with agonal rhythm or ventricular fibrillation/tachycardia survived, and 14 of the 15 hospital survivors had reactive pupils on arrival to the ED. We conclude that sinus rhythm and pupil size and reactivity are important physiologic variables that predict potential survival and may be used to guide continuation of resuscitative efforts in patients with traumatic cardiac arrest.