INTRODUCTION
Hemorrhage is the leading cause of potentially preventable death in trauma, accounting for 90% of military (1) and 26% to 40% of civilian (2, 3) deaths. Noncompressible torso hemorrhage is vascular disruption to axial torso vessels, solid organs, pulmonary parenchyma, and/or the bony pelvis, when accompanied by shock (4). This injury complex constitutes the great burden of hemorrhage-related deaths in both military (5, 6) and civilian (7) trauma with mortality rates of 43% and 42%, respectively. Importantly, many patients exsanguinate before definitive hemorrhage control, either dying before hospital admission or in the emergency department (1, 8, 9).
Patients with noncompressible torso hemorrhage, especially those presenting in extremis, require resuscitation and hemorrhage control (10). Thoracic aortic occlusion is a maneuver that addresses both by augmenting cardiac afterload and providing torso inflow control (11). This can be a lifesaving intervention in patients with end-stage hemorrhagic shock; however, it is generally performed as part of a resuscitative thoracotomy (RT) as a reactive maneuver following the loss of a central pulse. As a consequence, the survival rate following circulatory arrest and RT is dismal, with rates in military and civilian practice of between 0.5% and 20% (12–15). A preferable solution is the proactive use of aortic control, expanding the physiological window of salvage, thereby bridging critical physiology to definitive hemorrhage control.
Resuscitative endovascular balloon occlusion of the aorta (REBOA) is a technique that provides proactive circulatory support in a hypotensive patient at risk of cardiovascular collapse (16). The effectiveness of REBOA has been established as a resuscitative adjunct in the setting of ruptured abdominal aortic aneurysm, another pathology characterized by uncontrolled hemorrhage (17). Despite its usefulness in this setting, the use of REBOA in end-stage hemorrhagic shock from trauma has not been well characterized. The aim of this study was to examine the impact of REBOA on mortality as an adjunct to damage control resuscitation (DCR) in the setting of catastrophic torso trauma.
METHODS
Study design and overview
This study was undertaken at an American Association for Laboratory Animal Science–accredited large animal research facility following protocol approval by the Institutional Animal Care and Use Committee. The study used male Yorkshire-Landrace swine (Sus scrofa) weighing between 70 and 90 kg, which were housed at the facility for 7 days, under the supervision of licensed veterinary staff, before experimentation.
At total of 24 animals were divided into three groups, which are named continuous REBOA (cREBOA), intermittent REBOA (iREBOA), and no REBOA (nREBOA). These groups were then entered into a study protocol consisting of five phases: preparation, injury, intervention, damage control surgery, and critical care (Fig. 1).
;)
Fig. 1: Experimental design.
Preparation
Following induction with ketamine, animals were intubated and ventilated with oxygen (FIO2 0.3) and isoflurane (1.5%–4.0%) sufficient to maintain general anesthesia. Large-bore 8.5F sheaths were placed in both external and right internal jugular veins to permit placement of a pulmonary artery catheter and establish central venous access. A transonic flow probe was placed around the left carotid artery (Transonic Systems Inc, Ithaca, NY), and the right carotid artery was cannulated for invasive blood pressure monitoring. Infradiaphragmatic arterial access was also secured in both femoral arteries to permit blood sampling and REBOA deployment.
As swine possess a contractile spleen that can readily autotransfuse in response to hemorrhagic shock, a splenectomy was performed through a small upper midline laparotomy. Before closure of the abdomen in three layers, three 12-mm and one 5-mm laparoscopy ports were placed under direct vision to facilitate subsequent laparoscopy during the injury phase.
Injury
This injury was designed to replicate a lethal grade V liver injury and is based on a previously described model that utilizes a laparoscopic liver resection method (18). Carbon dioxide was insufflated to attain a 12-mmHg pneumoperitoneum; concomitantly, the FIO2 was titrated to 0.21 to simulate atmospheric conditions. The left lobe of the liver was divided using laparoscopic scissors, following a line 2 cm medial to the hilum, to resect approximately 80% of the left lobe. The goal was to accomplish the transection within 2 min, following which the pneumoperitoneum was rapidly evacuated, the ports removed, and the skin wounds closed. The animal then underwent a 10-min free-bleed period in which no treatment was administered.
Intervention
At the conclusion of the 10-min free-bleed period, all animals had a 0.035-in Lunderquist (Cook Medical, Bloomington, Ind) wire placed into the thoracic aorta under fluoroscopic guidance through the sheath in the right superficial femoral artery. Animals in the cREBOA and iREBOA groups had a 14F Coda Balloon (Cook Medical) placed over the wire through this sheath into the thoracic aorta. The balloon was positioned cephalad to the diaphragm and inflated with contrast medium (Fig. 2). Animals were transitioned to an FIO2 of 100% and received 250-mL boluses of intravenous colloid (Hextend) if mean arterial pressure (MAP) decreased to less than 50 mmHg, up to total of 1,500 mL. For the animals in the cREBOA group, the aorta was occluded continuously for 60 min. Animals in the iREBOA group had the balloon gradually deflated over the course of 1 min with the balloon completely deflated for 60 s; early reinflation was performed if the MAP decreased to less than 30 mmHg. The balloon was deflated at both 20 and 40 min after initial inflation of balloon. The animals in the control group (nREBOA) received colloid boluses (Hextend) only during the 60-min intervention phase.
Fig. 2: Fluoroscopy image of REBOA.
Damage control surgery
Following the 60-min intervention phase, the three groups of animals underwent concomitant whole blood (WB) resuscitation and damage control surgery.
Whole blood was banked from nonstudy animals, stored in citrated bags, and refrigerated for up to one week, with a unit comprising approximately 500 mL. Provision was made for the availability of between 6 and 8 units of WB per study animal. A Belmont Rapid Infuser (Belmont Instrument Corporation, Billerica, Mass) was used to warm and administer the WB through a large-bore central venous catheter. Resuscitation was titrated to a MAP of 60 mmHg and a hemoglobin of 10 g/dL. Hypocalcaemia, as measured by arterial blood gas sampling, was treated with 1-g intravenous infusions of calcium chloride solution.
The damage control laparotomy was performed in the three groups via a full midline incision, and the first surgical maneuver was application of the Pringle maneuver (clamping of the hepatic artery and portal vein) and manual control of the cut edge of the liver. The hemoperitoneum was evacuated, and the volume of evacuated blood recorded. Definitive liver hemostasis was then achieved through a combination of selective vessel ligation, diathermy, and multiaxial packing.
Once hepatic hemorrhage control was obtained, and a sufficient volume of WB had been infused to achieve a MAP consistently greater than 70 mmHg, balloon deflation was commenced. This was performed by the incremental withdraw of 1 to 2 mL over 5 to 10 min. The balloon was reinflated in response to hypotension or liver hemorrhage. Once balloon deflation and liver hemostatsis were complete, a Foley catheter was placed in the urinary bladder, and the midline wound closed.
Critical care
Postoperatively, resuscitation was continued in a critical care environment. Further WB and crystalloid were used to maintain a MAP of 60 mmHg or greater. Animals refractory to volume repletion were administered norepinephrine for hemodynamic support. Hyperkalemia was treated with intravenous insulin administration and 50% dextrose. Active rewarming was performed using a forced air heating blanket (Bair Hugger; Arizant Healthcare Inc, Eden Prairie, Minn). The end of study (EOS) was 6 h after injury at which point the animals were killed according to institutional protocol.
Study endpoints
The primary endpoint of this study was mortality, which was defined as the onset of asystole on electrocardiography. Secondary endpoints were divided into measures of hemodynamic performance, metabolic burden, laboratory parameters of organ function, and resuscitation volumes.
Hemodynamic parameters included systemic systolic blood pressure (SBP), pulmonary SBP, cardiac output (CO; in L/min), central venous oxygen saturation (SVO2; in %), and carotid flow rates (in mL/min) were measured throughout the study. Metabolic burden was quantified by pH and lactate measured at 30-min intervals. Laboratory parameters of organ function included potassium (K+), blood urea nitrogen, creatinine, hepatic aminotransferases, and cardiac troponin I (cTnI). Inflammatory burden was measured using interleukin 6 (IL-6), tumor necrosis factor α (TNF-α), and IL-8 assays. Laboratory and cytokine assays were measured at baseline (BL) and EOS. Total volumes and doses of intravenous fluids (blood, crystalloid, and colloid) and drugs (norepinephrine, calcium, and insulin) were collated at the EOS. After the animals were killed, the liver was excised with measurements obtained of total liver mass, resected liver mass, and total mass of the left lobe.
Statistical analysis
Data were analyzed using SPSS v20.0 (IBM, Chicago, Ill). χ2 Tests were used to compare categorical data, analysis of variance, and t tests for continuous variables. Log-rank test in conjunction with Kaplan-Meier survival plots were used for survival analysis.
RESULTS
Baseline characteristics and splenectomy
Baseline characteristics are presented in Tables 1 and 2 and Supplementary Table 1 (https://links.lww.com/SHK/A193). Physiological BL values were statistically similar across the groups, with the exception of heart rate (in beats/min) which was significantly higher in the nREBOA group compared with the cREBOA and iREBOA groups (98 ± 22 vs. 77 ± 16 and 75 ± 12; P = 0.030) (Table 1). There were no differences in laboratory BL values (Supplementary Table 1, https://links.lww.com/SHK/A193).
Table 1: Baseline physiology, splenectomy, liver injury, operative, and resuscitation data
Table 2: End-of-study laboratory parameter analysis
There was no difference in the time taken to perform the splenectomy across the three groups with an overall mean time of 29 ± 14 min (Table 1). Splenic weight was also consistent among the groups, with a mean weight of 535 ± 150 g. There was no difference in postsplenectomy hemodynamic or laboratory parameters (Supplementary Table 1, https://links.lww.com/SHK/A193).
The liver injury was accomplished within 2 min in all animals, resecting a consistent segment of the left lobe across the groups, with a mean overall weight of 286 ± 71 g and percentage resection of 75.0% ± 7.5% (Table 1).
Hemodynamic performance
Following liver injury, all animals underwent a precipitous cardiovascular collapse during the 10-min free-bleed period (Fig. 3). Systolic blood pressure (in mmHg) at the end of free-bleed period for the cREBOA, iREBOA, and nREBOA groups was 31 ± 14, 48 ± 28, and 28 ± 17, respectively; P = 0.125 (Fig. 3A).
Fig. 3: Hemodynamic response to balloon occlusion as measured by (A) systemic SBP, (B) pulmonary SBP, (C) CO, (D) mixed central venous oxygen saturation, (E) carotid flow. Data are plotted as mean values.
The initiation of balloon occlusion in the cREBOA and iREBOA groups during the intervention phase resulted in restoration of SBP to values significantly higher than BL (79 ± 12 vs. 107 ± 19 [P = 0.015] and 85 ± 9 vs. 117 ± 20 [P = 0.023], respectively). However, the SBP in the nREBOA control group continued to decrease with all animals progressing to asystolic cardiac arrest by 25 min of the intervention phase (35 min after injury) (Fig. 3A).
A similar trend was observed for pulmonary SBP, SVO2, and carotid flow among the groups (Fig. 3, B, D, and E). Following a significant decrease in each parameter during the free-bleed period, the deployment of REBOA resulted in the restoration of supranormal values; this was not observed in the nREBOA group.
Throughout the 60-min intervention phase, the cREBOA and iREBOA groups both maintained SBP following the initial elevation after injury. This is in contrast to CO, which decreased during the injury phase in the cREBOA and iREBOA groups to 3.2 ± 1.3 and 5.1 ± 1.2, respectively (Fig. 3C). Cardiac output made little recovery during the intervention phase decreasing to the lowest value of 2.5 ± 0.5 at 30 min in the cREBOA group and 4.6 ± 1.6 at 10 min in the iREBOA group. Following resuscitation during the damage control surgery phase, CO increased to levels higher than BL, with a peak of 7.6 ± 2.5 at 110 min in the cREBOA group and 7.7 ± 1.3 at 150 min in the iREBOA group. All indices of hemodynamic performance were maintained through to the EOS.
Metabolic burden
Throughout the free bleed and intervention phase, there was an increase in lactate and a commensurate decrease in pH (Fig. 4). In the nREBOA group, the lactate peaked at 14.4 mmol/L before the demise of the last animal in the group at 35 min after injury. The lactate trend was similar for both the cREBOA and iREBOA groups with peaks of 13.7 ± 2.2 at 90 min and 13.5 ± 3.3 at 105 min, respectively. Similarly, the lowest pH occurred at 105 min for both the cREBOA and iREBOA groups with measurements of 7.17 ± 0.09 and 7.17 ± 0.10.
Fig. 4: Metabolic changes in response to balloon occlusion as measured by (A) pH and (B) lactate. Data are plotted as mean values.
Lactate measurements continued to decline from their peak measurement during the critical care phase to a lowest measurement at 4.5 h of 7.7 ± 2.4 and 6.6 ± 2.9 for the cREBOA and iREBOA groups. The pH peaked at this time point with values 7.39 ± 0.83 and 7.31 ± 0.13, respectively. The EOS values saw a small increase in lactate to 9.0 ± 4.5 and 7.8 ± 4.5 and a decrease in pH of 7.22 ± 1.45 and 7.28 ± 0.18 for the cREBOA and iREBOA groups.
Laboratory parameters
A comparison of BL and EOS laboratory parameters for the cREBOA and iREBOA groups is presented in Table 2. Measures of hemoglobin, clotting time (prothrombin and partial thromboplastin time) and fibrinogen did not differ between BL and EOS. There was a significant decrease in platelet count observed between BL and EOS samples in the iREBOA group (292 ± 119 vs. 152 ± 98; P = 0.027).
An increase was observed in laboratory measures of end organ damage including renal, hepatic, and cardiac parameters. Blood urea nitrogen increased significantly in the cREBOA and iREBOA groups to 12.0 ± 1.6 and 11 ± 3.2, respectively. A similar trend was observed in creatinine. There was a nonsignificant increase in alanine aminotransferase across both groups and a significant rise in aspartate aminotransferase in the cREBOA group (20 ± 9 vs. 403 ± 133; P = 0.001). Significant cTnI increases were also observed, greatest in the cREBOA group (0 vs. 2.9 ± 1.5; P = 0.005) but also in the iREBOA group (0 vs. 1.8 ± 1.5; P = 0.033).
Inflammatory burden
Only data from the cREBOA and iREBOA groups are presented as no animal in the nREBOA group survived to the EOS (Table 2). There was no difference in BL measurements between the cREBOA and iREBOA groups for IL-6, TNF-α, or IL-8. A significant increase from BL to EOS was noted in IL-6 measurements in both the cREBOA (32 ± 14 vs. 1,188 ± 279; P = 0.001) and iREBOA (36 ± 26 vs. 227 ± 296; P = 0.003) groups.
While an increase in EOS compared with BL TNF-α and IL-8 was observed, it was not significant in either group (Table 2). In terms of EOS values between the groups, cREBOA had incurred a greater IL-6 (1,187 ± 279 vs. 834 ± 287; P = 0.084), TNF-α (205 ± 92 vs. 110 ± 36; P = 0.063), and IL-8 (329 ± 505 vs. 227 ± 296; P = 0.708) release than the nREBOA group, although again, none achieved statistical significance.
Operative intervention and resuscitation
Damage control surgery was performed in the cREBOA and iREBOA groups as no animal in the nREBOA group survived beyond the intervention phase (Table 2). Total Pringle time was similar between the cREBOA and iREBOA groups (18 ± 5 vs. 12 ± 3; P = 0.095) as was the intraoperative REBOA time (in minutes) (20 ± 9 vs. 15 ± 8; P = 0.293). Emergent reinflation was required for two animals in the cREBOA and four in the iREBOA groups, predominantly for control of hypotension. All laparotomies were performed within the allocated 60 min with an overall mean time of 35 ± 10 min.
Both groups received similar volumes of WB and crystalloid. Insulin and 50% dextrose were used in five animals to treat a K+ greater than 5.5 mmol/L, two required 50% dextrose for hypoglycemia, and inotropic support was necessary in five cases. One animal developed a tension pneumothorax, which was treated with tube thoracostomy.
There were two unplanned relaparotomies. One animal in the cREBOA group had a rising lactate despite resuscitation, and re-exploration identified global small bowel ischemia, which improved following withdrawal of the REBOA catheter from the thoracic aorta. The second relaparotomy occurred in an iREBOA animal, in response to hemodynamic instability and falling hemoglobin suggestive of ongoing bleeding. There was evidence of gross coagulopathy with no surgically remedial solution; despite exhaustion of WB reserves, the animal survived to the EOS.
Mortality
There were 11 early deaths from exsanguination that occurred during the intervention phase. All of the nREBOA animals died within 35 min of injury. Deployment of REBOA was unable to salvage two animals in the iREBOA group (11 and 15 min after injury) and one animal in the cREBOA group (12 min after injury).
There was one intraoperative death in the iREBOA group (80 min after injury). This animal tolerated balloon deflation during the intervention phase poorly, requiring emergent reocclusion during both attempts. The animal displayed gross hemodynamic instability 10 min before damage control laparotomy, deteriorating into cardiac arrest with pulseless electrical activity 3 min before damage control surgery. Despite aggressive resuscitation attempts, including internal cardiac massage and rapid WB infusion, there was no return of spontaneous circulation. One death was observed during the critical care phase in the cREBOA group (5 h after injury). This animal developed cardiogenic shock refractory to inotropic support.
The overall survival rates for the cREBOA, iREBOA, and nREBOA groups to EOS were 75.0%, 62.5%, and 0.0% respectively (Fig. 5). A pairwise log-rank (Mantel-Cox) comparison between the groups demonstrated a significant difference in survival when comparing nREBOA to cREBOA (P = 0.001) and nREBOA to iREBOA (P = 0.007). There was no difference comparing cREBOA to iREBOA (P = 0.572).
Fig. 5: Survival following balloon occlusion; log-rank test, P = 0.001.
DISCUSSION
The current study examines the effectiveness of REBOA as a hemorrhage control adjunct in a highly lethal porcine model of noncompressible torso hemorrhage. It is the first study to evaluate this adjunct in conjunction with modern DCR. Resuscitative endovascular balloon occlusion of the aorta was used to successfully salvage 13 (81.3%) of 16 animals from imminent circulatory arrest and to sustain the circulation of 12 (75.0%) of 16 animals until definitive hemorrhage control. This is in contrast to the control animals, all of whom died of rapid exsanguination. Furthermore, the current study also examined outcomes between intermittent and continuous occlusion, in an effort to assess whether transient reperfusion reduced the metabolic or inflammatory burden. No difference was detected in these outcomes between the continuous and iREBOA groups.
This study confirms and extends previous work characterizing the hemodynamic and metabolic sequelae of balloon occlusion in hemorrhagic shock. White et al. (19) used a porcine model of controlled hemorrhage to demonstrated that REBOA had a comparably favorable hemodynamic profile to open clamp occlusion of the thoracic aorta, but resulted in less of a metabolic burden than RT. Markov et al. (20) used a similar model to evaluate the physiologic tolerance of 30 and 90 min of occlusion. A lactic acidosis was incurred with both occlusion times, greatest in the 90-min group, but ultimately survivable with the restoration of normal physiology following balloon deflation.
Most recently, Scott et al. (21) used an occlusion time of 60 min with which to assess the performance of a newly developed, self-centering, low-profile, prototype REBOA catheter. Results from that study demonstrated the reproducibility of blind or fluoroscopy-free placement, with physiological results that complemented the work of Markov et al. Although this body of literature demonstrates the hemodynamic advantages of REBOA and the survivability following balloon deflation, more clinically relevant models of uncontrolled hemorrhage have been required to assess the practical application of REBOA.
Avaro et al. (22) used a porcine model of open splenic trauma, resuscitated with saline, to evaluate a control group with 40 or 60 min of REBOA followed by splenectomy. The control group all exsanguinated within 80 min of injury, and 9 (75.0%) of 12 of the 60-min group died upon balloon deflation due to metabolic derangement. All animals in the 40-min group in the study of Avaro et al. survived, suggesting that 40 min constitutes an optimum physiologic threshold for resuscitative aortic occlusion.
The current study builds on the work of Avaro et al. (22), by incorporating DCR (including WB administration) that successfully extends this physiological threshold to 60 min. This approach results in a greater overall survival (68.8%), despite a longer occlusion time and a more lethal model. The combination of WB resuscitation, prompt correction of electrolyte derangement, active warming, and hemodynamic support can successfully ameliorate the metabolic consequences of balloon deflation and reperfusion.
However, it is important to acknowledge that the current study’s resuscitation was by no means comprehensive. Whole blood reserves were limited, and adjuncts such as tranexamic acid (23) or specialized blood products such as cryoprecipitate (24) were not used. This was reflected by the ongoing acidosis during the critical care phase, which is frequently corrected intraoperatively using contemporary DCR (25).
There are further limitations that are important to discuss. The use of iREBOA was an attempt to permit a degree of ischemic preconditioning that would enhance resilience to the reperfusion injury. The time of 1 min was used, as during the model development phase, it was clear that animals would not tolerate prolonged deflation without resuscitation. It is likely that the period of minute is too brief, and therefore conclusions relating to the iREBOA group should be cautious. Further work examining techniques to ameliorate reperfusion injury should be explored.
The use of REBOA has not been limited to animal studies, with the earliest reported human use during the Korean War in 1953 by Lt. Col. Hughes (26). He described two casualties with exsanguinating truncal injuries, both of whom responded to balloon occlusion, but ultimately died of their wounds. Despite its attractions, balloon occlusion was surpassed by the experience of open clamp occlusion, which became the standard of care (27).
Resuscitative aortic balloon occlusion was explored again in the 1980s; however, success was limited by patient selection and insertion technique (28, 29). The majority of patients were either in established cardiac arrest or moribund, and the method of arterial access was generally by cut-down, frequently unsuccessful. Resuscitative endovascular balloon occlusion of the aorta may be most successful as a proactive intervention (i.e., aortic pressure monitoring with capacity to inflate a balloon) in patients with a spontaneous circulation, at risk of circulatory arrest.
Following the refinement of both catheter technology and Seldinger insertion techniques during the 2000s (30,31), along with the developments in trauma resuscitation, the reevaluation of this technology is a logical step (16). Brenner and colleagues (32) recently report a series of six trauma patients, injured by a mixture of blunt and penetrating trauma treated using REBOA as an adjunct to DCR. The mean admission SBP was 59 ± 27 mmHg, which was increased to 114 ± 20 mmHg following the deployment of REBOA. While one patient died of traumatic brain injury and another of multiorgan failure, there were no REBOA-related complications. Ultimately, this adjunct was used as a hemostatic and resuscitative bridge to definitive hemorrhage control, incurring no hemorrhage-related mortality.
Resuscitative endovascular balloon occlusion of the aorta can temporize exsanguinating hemorrhage and restore life-sustaining perfusion, bridging critical physiology to definitive hemorrhage control (i.e., surgical hemostasis). In the current study, iREBOA compared with cREBOA offered no additional benefit, although different schedules of occlusion should be explored in future studies. Importantly, the physiologic penalty incurred by 60 min of occlusion could be ameliorated with aggressive DCR. Prospective observational studies of REBOA as a hemorrhage control adjunct should be undertaken in appropriate groups of human trauma patients.
ACKNOWLEDGMENTS
The authors thank the laboratory and veterinary staff at the 59th Medical Wing, Clinical Research Division, Lackland Air Force Base, for their work supporting the experiments and laboratory assays performed during this study.
REFERENCES
1. Eastridge BJ, Mabry RL, Seguin P, Cantrell J, Tops T, Uribe P, Mallett O, Zubko T, Oetjen-Gerdes L, Rasmussen TE, et al.: Death on the battlefield (2001–2011): implications for the future of combat casualty care.
J Trauma 73: S431–S437, 2012.
2. Tien HC, Spencer F, Tremblay LN, Rizoli SB, Brenneman FD: Preventable deaths from hemorrhage at a level I Canadian trauma center.
J Trauma 62: 142–146, 2007.
3. Teixeira PGR, Inaba K, Hadjizacharia P, Brown C, Salim A, Rhee P, Browder T, Noguchi TT, Demetriades D: Preventable or potentially preventable mortality at a mature trauma center.
J Trauma 63: 1338–1347, 2007.
4. Morrison JJ, Rasmussen TE:
Noncompressible torso hemorrhage: a review with contemporary definitions and management strategies.
Surg Clin North Am 92: 843–858, 2012.
5. Stannard A, Morrison JJ, Scott DJ, Ivatury RA, Ross JD, Rasmussen TE: The epidemiology of
noncompressible torso hemorrhage in the wars in Iraq and Afghanistan.
J Trauma Acute Care Surg 74: 830–834, 2013.
6. Morrison JJ, Stannard A, Rasmussen TE, Jansen JO, Tai NRM, Midwinter MJ: Injury pattern and mortality of non-compressible torso haemorrhage in UK combat casualties.
J Trauma Acute Care Surg 75: s263–s268, 2013.
7. Kisat M, Morrison J, Hashmi Z, Efron D, Rasmussen T, Haider A: Epidemiology and outcomes of non-compressible torso hemorrhage.
J Surg Res 179: 198, 2013.
8. Chiara O, Scott JD, Cimbanassi S, Marini A, Zoia R, Rodriguez A, Scalea T, Milan Trauma Death Study Group: Trauma deaths in an Italian urban area: an audit of pre-hospital and in-hospital trauma care.
Injury 33: 553–562, 2002.
9. Gedeborg R, Chen L-H, Thiblin I, Byberg L, Melhus H, Michaelsson K, Warner M: Prehospital injury deaths - strengthening the case for prevention: nationwide cohort study.
J Trauma Acute Care Surg 72: 765–772, 2012.
10. Gruen RL, Brohi K, Schreiber M, Balogh ZJ, Pitt V, Narayan M, Maier RV: Haemorrhage control in severely injured patients.
Lancet 380: 1099–1108, 2012.
11. Eide TO, Aasland J, Romundstad P, Stenseth R, Saether OD, Aadahl P, Myhre HO: Changes in hemodynamics and acid-base balance during cross-clamping of the descending thoracic Aorta.
Eur Surg Res 37: 330–334, 2005.
12. Morrison JJ, Poon H, Rasmussen TE, Khan MA, Midwinter MJ, Blackbourne LH, Garner JP: Resuscitative thoracotomy following wartime injury.
J Trauma Acute Care Surg 74: 825–829, 2013.
13. Seamon MJ, Shiroff AM, Franco M, Stawicki SP, Molina EJ, Gaughan JP, Reilly PM, Schwab CW, Pryor JP, Goldberg AJ: Emergency department thoracotomy for penetrating injuries of the heart and great vessels: an appraisal of 283 consecutive cases from two urban trauma centers.
J Trauma 67: 1250–1257, 2009.
14. Edens JW, Beekley AC, Chung KK, Cox ED, Eastridge BJ, Holcomb JB, Blackbourne LH: Longterm outcomes after combat casualty emergency department thoracotomy.
J Am Coll Surg 209: 188–197, 2009.
15. Asensio J, Wall M, Minei J: Practice management guidelines for emergency department thoracotomy.
J Am Coll Surg 193: 303–309, 2001.
16. Stannard A, Eliason JL, Rasmussen TE:
Resuscitative endovascular balloon occlusion of the aorta (
REBOA) as an adjunct for
hemorrhagic shock.
J Trauma 71: 1869–1872, 2011.
17. Mayer D, Pfammatter T, Rancic Z, Hechelhammer L, Wilhelm M, Veith FJ, Lachat M: 10 Years of emergency endovascular aneurysm repair for ruptured abdominal aortoiliac aneurysms: lessons learned.
Ann Surg 249: 510–515, 2009.
18. Ross JD, Burns CJ, Tadaki D: A novel swine model for abdominal liver injury and hemorrhage.
Shock 35: 16, 2011.
19. White JM, Cannon JW, Stannard A, Markov NP, Spencer JR, Rasmussen TE: Endovascular balloon occlusion of the aorta is superior to resuscitative thoracotomy with aortic clamping in a porcine model of
hemorrhagic shock.
Surgery 150: 400–409, 2011.
20. Markov N, Percival TJ, Morrison JJ, Ross JD, Scott DJ, Spencer J, Rasmussen TE: Physiologic tolerance of descending thoracic aortic balloon occlusion in a swine model of
hemorrhagic shock.
Surgery 153: 848–856, 2013.
21. Scott DJ, Eliason JL, Villamaria C, Morrison JJ, Houston IV R, Spencer J, Rasmussen TE: A novel fluoroscopy-free, resuscitative endovascular aortic balloon occlusion system in a model of
hemorrhagic shock.
J Trauma Acute Care Surg 75: 1–24, 2013.
22. Avaro J-P, Mardelle V, Roch A, Gil C, de Biasi C, Oliver M, Fusai T, Thomas P: Forty-minute endovascular aortic occlusion increases survival in an experimental model of uncontrolled
hemorrhagic shock caused by abdominal trauma.
J Trauma 71: 720–725, 2011.
23. Morrison JJ, Dubose JJ, Rasmussen TE, Midwinter MJ: Military Application of Tranexamic Acid in Trauma Emergency
Resuscitation (MATTERs) study.
Arch Surg 147: 113–119, 2012.
24. Morrison JJ, Ross JD, DuBose JJ, Jansen JO, Midwinter MJ, Rasmussen TE: Association of cryoprecipitate and tranexamic acid with improved survival following wartime injury.
JAMA Surg 148: 218–225, 2012.
25. Morrison JJ, Ross JD, Poon H, Midwinter MJ, Jansen JO: Intra-operative correction of acidosis, coagulopathy and hypothermia in combat casualties with severe haemorrhagic shock.
Anaesthesia 68: 846–850, 2013.
26. Hughes C: Use of an intra-aortic balloon catheter tamponade for controlling intra-abdominal hemorrhage in man.
Surgery 36: 65–68, 1954.
27. Ledgerwood A, Kazmers M, Lucas CE: The role of thoracic aortic occlusion for massive hemoperitoneum.
J Trauma 16: 610–615, 1976.
28. Low RB, Longmore W, Rubinstein R, Flores L, Wolvek S: Preliminary report on the use of the Percluder occluding aortic balloon in human beings.
Ann Emerg Med 15: 1466–1469, 1986.
29. Gupta B, Khaneja S, Flores L, Eastlick L, Longmore W, Shaftan G: The role of intra-aortic balloon occlusion in penetrating abdominal trauma.
J Trauma 29: 861–865, 1989.
30. Arthurs Z, Starnes B, See C, Andersen C: Clamp before you cut: proximal control of ruptured abdominal aortic aneurysms using endovascular balloon occlusion: case reports.
Vasc Endovasc Surg 40: 149–155, 2006.
31. Assar A, Zarins C: Endovascular proximal control of ruptured abdominal aortic aneurysms: the internal aortic clamp.
J Cardiovasc Surg 50: 381–385, 2009.
32. Brenner M, Moore L, Dubose J, Tyson G, McNutt M, Albarado R, Holcomb JB, Scalea TM, Rasmussen TE: A clinical series of
resuscitative endovascular balloon occlusion of the aorta for hemorrhage control and
resuscitation.
J Trauma Acute Care Surg 75: 506–511, 2013.
