Does hemopericardium after chest trauma mandate sternotomy?

BACKGROUND: Recently, three patients with hemopericardium after severe chest trauma were successfully managed nonoperatively at our institution. This prompted the question whether these were rare or common events. Therefore, we reviewed our experience with similar injuries to test the hypothesis that trauma-induced hemopericardium mandates sternotomy.
METHOD: Records were retrospectively reviewed for all patients at a Level I trauma center (December 1996 to November 2011) who sustained chest trauma with pericardial window (PCW, n = 377) and/or median sternotomy (n = 110).
RESULTS: Fifty-five (15%) patients with positive PCW proceeded to sternotomy. Penetrating injury was the dominant mechanism (n = 49, 89%). Nineteen (35%) were hypotensive on arrival or during initial resuscitation. Most received surgeon-performed focused cardiac ultrasound examinations (n = 43, 78%) with positive results (n = 25, 58%). Ventricular injuries were most common, with equivalent numbers occurring on the right (n = 16, 29%) and left (n = 15, 27%). Six (11%) with positive PCW had isolated pericardial lacerations, but 21 (38%) had no repairable cardiac or great vessel injury. Those with therapeutic versus nontherapeutic sternotomies were similar with respect to age, mechanisms of injury, injury severity scores, presenting laboratory values, resuscitation fluids, and vital signs. Multiple logistic regression revealed that penetrating trauma (odds ratio: 13.3) and hemodynamic instability (odds ratio: 7.8) were independent predictors of therapeutic sternotomy.
CONCLUSION: Hemopericardium per se may be overly sensitive for diagnosing cardiac or great vessel injuries after chest trauma. Some stable blunt or penetrating trauma patients without continuing intrapericardial bleeding had nontherapeutic sternotomies, suggesting that this intervention could be avoided in selected cases. LEVEL OF EVIDENCE: Therapeutic study, level III.