Emergency Resuscitative Thoracotomy: A Nationwide Analysis of Outcomes and Predictors of Futility

Background: Most studies on emergency resuscitative thoracotomy (ERT) suffer from either small sample size or unclear inclusion criteria. We sought to assess ERT outcomes and predictors of futility using a nationwide database. Methods: Using a novel and comprehensive algorithm of combinations of specific International Classification of Diseases, Ninth Revision and International Classification of Diseases, Tenth Revision procedure codes denoting the multiple steps of an ERT (e.g., thoracotomy, pericardiotomy, cardiac massage) performed within the first 60 min of patient arrival, we identified ERT patients in the 2010-2016 Trauma Quality Improvement Program database. We defined the primary outcome as survival to discharge and the secondary outcomes as hospital length of stay (LOS), intensive care unit LOS, number of complications, and discharge destination. Univariate then backward stepwise multivariable logistic regression analyses were performed to assess independent predictors of mortality. Multiple imputations by chained equations were performed when appropriate, as additional sensitivity analyses. Results: Of 1,403,470 patients, 2012 patients were included. The median age was 32, 84.0% were males, 66.7% had penetrating trauma, the median Injury Severity Score was 26, and 87.5% presented with signs of life (SOL). Of the 1343 patients with penetrating injury, 72.9% had gunshot wounds and 27.1% had stab wounds. The overall survival rate was 19.9%: 26.0% in penetrating trauma (stab wound 45.6% versus gunshot wound 18.7%; P < 0.001) and 7.6% in blunt trauma. Independent predictors of mortality were aged 60 y and older (odds ratio, 2.71; 95% confidence interval [95% CI], 1.26-5.82; P = 0.011), blunt trauma (odds ratio, 4.03; 95% CI, 2.72-5.98; P < 0.001), prehospital pulse <60 bpm (odds ratio, 3.43; 95% CI, 1.73-6.79; P < 0.001), emergency department pulse <60 bpm (odds ratio, 4.70; 95% CI, 2.47-8.94; P < 0.001), and no SOL on emergency department arrival (odds ratio, 3.64; 95% CI, 1.08-12.24; P = 0.037). Blunt trauma was associated with a higher median hospital LOS compared with penetrating trauma (28 d versus 13 d; P < 0.001), higher median intensive care unit LOS (19 d versus 6 d; P < 0.001), higher median number of complications (2 versus 1; P = 0.006), and more likelihood to be discharged to a rehabilitation facility instead of home (72.6% versus 28.7%; P < 0.001). ERT had the highest survival rates in patients younger than 60 y who present with SOL after penetrating trauma. None of the patients with blunt trauma who presented with no SOL survived. Conclusions: The survival rates of patients after ERT in recent years are higher than classically reported, even in the patient with blunt trauma. However, ERT remains futile in patients with a blunt trauma presenting with no SOL. (C) 2020 Elsevier Inc. All rights reserved.