Penetrating Cardiac Injuries: A Prospective Study of Variables Predicting Outcomes

Abstract

Background: Penetrating cardiac injuries are one of the leading causes of death from urban violence. Study Design: This is a prospective, 1-year study in a Level I Trauma Center with the objective of analyzing: (1) the parameters measuring the physiologic condition of patients sustaining penetrating cardiac injuries in the field during transport and on arrival, (2) the cardiovascular-respiratory score (CVRS) component of the trauma score, (3) the mechanism and anatomic site of injury, (4) the presence or absence of tamponade, and (5) the cardiac rhythm as a predictor of outcomes. We attempted to correlate cardiac injury grade (AAST-OIS) with mortality. Our main intervention was thoracotomy for resuscitation and definitive repair of cardiac injury. Main outcomes measures were all parameters measuring the physiologic condition of patients, CVRS, mechanism and anatomic site of injury, operative findings and maneuvers, mortality, and grade of injury. Results: The study consisted of 60 patients sustaining penetrating cardiac injuries, 35 gunshot wound (58%) and 25 stab wounds (42%). The injury severity score (ISS) was > 30 in 22 patients; overall survival was 22 of 60 (36.6%); gunshot wound (GSW) survival, 5 of 35 (14%); and stab wound (SW) survival, 17 of 25 (68%). An emergency department thoracotomy was performed in 37 of 60 (61.7%) with 6 of 37 survivors (16%). CVRS: 96% mortality (25 of 26) when CVRS = 0; 67% mortality (6 of 9) when CVRS = 1–3; and 25% mortality (7 of 25) when CVRS > 4 (p < 0.001). Mechanism of injury, and presence of sinus rhythm when pericardium opened predict outcomes (p < 0.001). Anatomic site of injury and tamponade do not predict outcomes (not significant). AAST-OIS injury grade and mortality: grade IV, 31 of 60 (52%); grade V, 20 of 60 (75%), and grade VI, 6 of 60 (100%). Conclusions: Parameters measuring physiologic condition, CVRS, and mechanism of injury plus initial rhythm are significant predictors of outcomes in penetrating cardiac injuries. The need for aortic crossclamping and the inability to restore an organized rhythm or blood pressure after thoracotomy were also predictors of outcomes. The presence of pericardial tamponade was not.