B. S. KARAM1, J. S. Pearl1, A. McCormick1, S. Bou Zein Eddine1, A. Baskaran1, T. W. Carver1, D. J. Milia1, A. Elegbede1, L. Somberg1, C. Trevino1, M. De Moya1 1Medical College Of Wisconsin,Division Of Trauma And Acute Care Surgery,Milwaukee, WI, USA
Introduction: The cardiac box was historically used to identify patients that required a pericardial window in the setting of a penetrating chest injury. Gunshot wounds (GSW) to the chest have an unpredictable trajectory and even stab wounds outside of the traditional landmarks have been associated with cardiac injury. Given the widespread availability of ultrasound we question whether the cardiac box has any clinical relevance.
Methods: This is a retrospective review of the trauma registry and EMR charts of adult patients with a penetrating thoracic injury admitted to a Level 1 urban trauma center between 2013 and 2017. Our primary outcome was penetrating cardiac injury requiring intervention. Descriptive statistics and multivariate regression were performed using SPSS.
Results: We identified 634 patients with penetrating thoracic injuries and 265 (42%) had an injury within the cardiac box. Of those, a pericardial window was performed in 17% (45/265) and 8% (20/265) had a cardiac injury. Eighty four percent (16/19) of stab victims versus only 57% (4/7) of gunshot victims with cardiac injury sustained penetrating trauma within the box (Figure 1). Ultrasound (US) was performed in 32% (208/634) of patients with a penetrating thoracic injury and 51% (135/265) of patients with a cardiac box injury. On multivariate regression, the cardiac box was not a significant predictor for cardiac injury due to stab wounds (OR 23.3 [95% CI 0.58-926.4]) or GSWs (OR 3.14 [95% CI 0.48-20.73]). US had a higher sensitivity (83% vs 80%), specificity (93% vs 59%), and positive predictive value (54% vs 8%) than the cardiac box.
Conclusion: The majority of penetrating injuries to the cardiac box do not result in a cardiac injury and a fourth of our cardiac injuries were outside the box. We recommend eliminating the use of a predefined anatomical cardiac boundary and favor US imaging for any penetrating thoracic injury due to its superior diagnostic performance.
