Abstract
Background:
Resuscitative thoracotomy is a dramatic operation performed in otherwise unsalvageable trauma patients. Analysis of its efficacy is based mostly on institutional series compiling the experience of multiple surgeons. This study aimed to report more consistent information by describing the resuscitative thoracotomy practice of a single surgeon and its evolution during more than two decades.
Methods:
A retrospective review of consecutive patients who underwent resuscitative thoracotomy in July 1990 to December 2012. Demographics, mechanism of injury, signs of life, injuries, and outcomes were analyzed. Comparisons were made between penetrating and blunt trauma patients and between pre- and post-introduction of a selective resuscitative thoracotomy protocol.
Results:
Sixty-seven resuscitative thoracotomies were performed. Most patients were males (84%), and mean age was 38 years. Mechanism of injury was stab wounds (54%, 36), blunt force (25%, 17), and gunshot wounds (21%, 14). Survival was 22% (8/36), 0% (0/17), and 7% (1/14), respectively. All nine survivors had signs of life upon admission, and survival in patients with signs of life on admission was 25% (8/32) in the stab wounds group and 8% (1/12) in the gunshot wounds group. Seven of the nine survivors (78%) were discharged neurologically intact. The most common injury in survivors was cardiac laceration with tamponade (6/9) and lung injury (3/9). Three survivors had a cardiac and lung injury, one had a lung hilum injury, and one had an abdominal inferior vena cava laceration. The switch to resuscitative thoracotomy protocol (2002) improved overall (31 vs 8%,
Conclusion:
This single-surgeon series supports that resuscitative thoracotomy can be lifesaving in selected penetrating trauma patients in extremis. A switch to a selective evidence-based protocol increased overall and penetrating resuscitative thoracotomy survival and limited resuscitative thoracotomy performance to patients arriving with signs of life.
Introduction
Resuscitative thoracotomy (RT) is a bold procedure, which may present the only hope for some moribund trauma patients (1). Since its introduction about five decades ago, the approach to performing RT has repeatedly alternated between enthusiasm and skepticism and between liberalism and conservatism. This pendulous reality has been fueled by studies with ambiguous nomenclature, poor designs, and variable sizes, as well as ethical and logistical difficulties in performing prospective controlled studies. Consequently, in spite of significant efforts to identify the group(s) of patients who may benefit from RT, no consensus guidelines are available to assist the surgeon in caring for the trauma patient in extremis. For example, some of the American College of Surgeons Committee on Trauma (ACS-COT) guidelines published in 2001 in an attempt to clarify the confusion surrounding RT (2), including the approach to the pre-hospital pulseless patient and to the patient with abdominal vascular injury, have been recently challenged by several large retrospective studies (3,4).
Regardless of the differences among the various studies, they all reported the experience of groups of surgeons in a single or multiple institutions. This study aimed to provide more consistent information regarding RT by describing the practice and results of a single trauma surgeon during more than two decades. Also, given this long time perspective, the study aimed to describe the evolution of the approach to the trauma patient in extremis. Specifically, the study compared management and outcomes before and after the implementation of a selective resuscitative thoracotomy protocol (SRTP).
Methods
A retrospective analysis of a prospectively maintained database was performed on 67 patients in extremis who underwent consecutive RT by the author between January 1990 and December 2012. Extracted data included demographics, mechanism of injury, signs of life (SOL), defined as the presence of organized cardiac electrical activity with or without a palpable pulse on admission, injuries, and outcomes. Comparisons were made between penetrating and blunt trauma patients and between pre- and post-introduction of the SRTP in February 2001 (Fig. 1). This SRTP replaced a protocol indicating RT on all trauma patients with less than 10 min pre-hospital loss of SOL and on those who arrested or acutely deteriorated in the trauma room. Statistical evaluation was done using a one-tail chi-square analysis (GraphPad Software, La Jolla, CA, USA) with significance determined at <0.05. The study was approved by the institutional review board.
An evidence-based protocol for a selective approach to the performance of RT.
Results
There was male dominance (84%) and mean age was 38 ± 6 years, with age range of 8–91 years. The most common mechanism of injury was stab wounds (SW, 54%) followed by blunt trauma (25%) and gunshot wounds (GSW, 21%) (Table 1). Of the 17 patients with blunt trauma, 14 (82%) were injured in a car crash and 3 (18%) from a fall from height.
Survival of patients who underwent RT.
SW: stab wounds; GSW: gunshot wounds.
Overall survival of patients undergoing RT was 13% (Table 1). Survival of patients with SW, especially to the chest, was higher (Table 1). All blunt trauma patients expired, and only 1 out of 17 patients (14%) with extra-thoracic injury survived (Table 1). None of the patients who presented to the trauma room without SOL survived, regardless of the mechanism of injury (Table 2). In contrast, 16% of penetrating trauma (mostly SW) patients who arrived with or lost SOL in the trauma room survived (Table 2).
Survival by the presence or absence of SOL upon admission in patients who underwent RT.
SOL: signs of life; SW: stab wounds; GSW: gunshot wounds; ED: emergency department.
There were nine survivors, all with SOL upon admission (Table 3). The most common injury in survivors was cardiac laceration with tamponade and lung injury (Table 3). Three survivors had cardiac and lung injury, one had lung hilum injury, and one had abdominal inferior vena cava (IVC) laceration. Seven of the nine survivors (78%) were discharged neurologically intact.
Profile of survivors of RT.
MOI: mechanism of injury; SOL: signs of life; ED: emergency department; M: male; F: female; SW: stab wounds; GSW: gunshot wounds; IVC: inferior vena cava.
The implementation of an evidence-based protocol, which advocated a selective use of RT, increased overall survival, improved survival in patients with penetrating trauma, and eliminated the performance of RT in patients with pre-hospital loss of SOL (Table 4). There was also a reduction in the performance of RT in blunt trauma patients (Table 4). However, this difference did not reach statistical significance.
The effect of introducing a protocol for selective performance of RT.
B/P: blunt/penetrating; SOL: signs of life; ED: emergency department; SRT: selective resuscitative thoracotomy.
None of the RT patients became an organ donor.
Discussion
This study describes for the first time the approach to and results of RT performed by a single surgeon. Such a review has several inherent advantages over previous studies, which, without exception, described a collective experience of several surgeons in a single or multiple institutions. These include persistent and clear definition of RT and of SOL, adherence to the same protocols, and consistency in performing the procedure. Furthermore, the study presents the practice dynamics over a period of more than two decades and thus was able to demonstrate for the first time that introduction of an evidence-based selective paradigm for performing RT may reduce mortality. Specifically, the SRTP, introduced in 2001, decreased both overall and penetrating trauma-associated mortality. As the RT technique, surgical infrastructure, and critical care and blood bank capabilities did not change during the study period, it is conceivable that the narrowed indications for RT, which reduced the performance of RT in non-survivable patients, account for this improvement. One might argue that the application of damage control resuscitation may have also contributed to the reported improvement. Nevertheless, as this resuscitation strategy was introduced to the author’s practice only during the last decade, this practice change could have had only a partial effect, if any.
The survival rates in this review for both blunt and penetrating mechanism of injury are comparable to those reported in larger institutional and multi-institutional series (2,3,5). Also, in agreement with the literature (2,5), survival was the highest in patients who sustained penetrating thoracic trauma, mainly chest SW with cardiac tamponade secondary to a cardiac laceration, who presented to the emergency department (ED) with SOL. Two factors likely account for this observation. First, the physiological deterioration and arrest may occur in these patients before blood volume depletion from exsanguination occurs (6). Second, both the pathophysiological consequences of cardiac tamponade and the cardiac injury per se can be reversed precipitously by opening the pericardium and repairing the laceration, respectively.
All survivors had SOL upon admission, and no patient without SOL survived. This observation matches the guidelines recommended by both the ACS-COT (2,7) and the National Association of Emergency Medical Services (EMS) Physicians Standards and Clinical Practice Committee (8), which support the futility of RT in patients with pre-hospital asystole. In contrast, recent recommendations by the Western Trauma Association (WTA) (3), based on its multi-center study (9), suggest performing RT on a selected group of blunt and penetrating trauma patients with pre-hospital arrest based on the duration of their pre-hospital cardiopulmonary resuscitation (CPR). However, as stated in the Editorial Critique to that publication, the WTA recommendations are based on a minimal number of patients, and the data lack clarity on the physiology of the patients upon arrival and the accuracy of the documentation of CPR time.
None of the 67 patients in this series became organ donors. Due to the severe shortage of organs for transplantations, some investigators examined the potential for organ donation from patients undergoing RT. A study of 263 RT patients in an urban Level I trauma center in the United States (10) reported that 11 out of 48 (22.9%) patients who arrived alive to the intensive care unit became potential organ donors. Of these, only three patients were harvested (1.1% of the total). The proponents of organ donation from RT patients argue that although only a minute fraction of the 263 RT patients donated organs, the number of recipients whose survival was improved was greater than the number of RT survivors. Thus, additional studies are required to determine the feasibility of this approach.
This study has several limitations. First, although it presents the largest series of RT performed by a single surgeon, the overall number of patients (N = 67) is somewhat limited. Second, it is unknown how many patients would have survived if the liberal protocol had been kept. This information could have been obtained only by comparing the previous protocol to the new more selective one in a prospective, blinded and controlled fashion, which could not have been ethically and practically performed. In spite of the above limitations, the inherent advantages of the study, as outlined above, make this study useful to the trauma surgeons caring for patients in extremis.