A porcine model for evaluating the management of noncompressible torso hemorrhage

The Efficacy of Whole Blood Resuscitation During Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) to Mitigate Post-occlusion Circulatory Collapse: A Translational Model in Large Swine

INTRODUCTION

Uncontrolled torso hemorrhage is the primary cause of potentially survivable deaths on the battlefield. Zone 1 Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA), in conjunction with damage control resuscitation, may be an effective management strategy for these patients in the prehospital or austere phase of their care. However, the effect of whole blood (WB) transfusion during REBOA on post-occlusion circulatory collapse is not fully understood.

MATERIALS AND METHODS

Yorkshire male swine (n = 6 per group, 70-90 kg) underwent a 40% volume-controlled hemorrhage. After a 10-minute hemorrhagic shock period, a REBOA balloon was inflated in Zone 1. Fifteen minutes after inflation, 0, 1, or 3 units (450 mL/unit) of autologous WB was infused through the left jugular vein. Thirty minutes after initial balloon inflation, the balloon was deflated slowly over 3 minutes. Following deflation, normal saline was administered (up to 3,000 mL) and swine were observed for 2 hours. Survival (primary outcome), hemodynamics, and blood gas values were compared among groups. Statistical significance was determined by log-rank test, one-way ANOVA, and repeated measures ANOVA.

RESULTS

Survival rates were comparable between groups (P = .345) with 66% of control, 33% of the one-unit animals, and 50% of the 3-unit animals survived until the end of the study. Following WB infusion, both the 1-unit and the 3-unit groups had significantly higher blood pressure (P < .01), pulmonary artery pressure (P < .01), and carotid artery flow (P < .01) compared to the control group.

CONCLUSIONS

WB transfusion during Zone 1 REBOA was not associated with increased short-term survival in this large animal model of severe hemorrhage. We observed no signal that WB transfusion may mitigate post-occlusion circulatory collapse. However, there was evidence of supra-normal blood pressures during WB transfusion.

[Resuscitative endovascular balloon occlusion of the aorta : Bridge to surgery]

BACKGROUND

Severe hemorrhage remains the leading cause of death among trauma patients. Resuscitative balloon occlusion of the aorta (REBOA) is an endovascular alternative to the established emergency room thoracotomy with cross-clamping of the aorta in patients with severe abdominal or pelvic bleeding.

OBJECTIVE

The article reports on initial experiences with REBOA.

METHODS

Based on the literature and own experiences the pathophysiology, indications, contraindications, technical details and first results with REBOA are presented.

RESULTS

The REBOA procedure is indicated in patients with treatment-refractive hemorrhagic shock with severe abdominal or pelvic bleeding. Via a transfemoral approach a balloon catheter is placed in the aorta and inflated. Depending on the indication the aortic occlusion is located in a supradiaphragmatic (zone 1) or infrarenal (zone 3) position. Experimental results proved a significant increase in central perfusion pressure after performance of REBOA. Furthermore, first clinical data indicate an improved patient survival rate after trauma. Improvements of the devices and minimizing the access trauma using small 7 Fr sheaths decreased the perioperative complication rate.

CONCLUSION

The REBOA procedure is a promising endovascular technique for temporary stabilization of the circulation in patients with hemorrhagic shock. Further clinical studies and registries have yet to prove its superiority over emergency room thoracotomy.

Intravenous Hydroxocobalamin Versus Hextend Versus Control for Class III Hemorrhage Resuscitation in a Prehospital Swine Model

Background

Hydroxyethyl starch (Hextend) has been used for hemorrhagic shock resuscitation, however, hydroxyethyl starch may be associated with adverse outcomes.

Objective

To compare systolic blood pressure (sBP) in animals that had 30% of their blood volume removed and treated with intravenous hydroxocobalamin, hydroxyethyl starch, or no fluid.

Methods

Twenty-eight swine (45-55 kg) were anesthetized and instrumented with continuous femoral and pulmonary artery pressure monitoring. Animals were hemorrhaged 20 mL/kg over 20 minutes and then administered 150 mg/kg IV hydroxocobalamin in 180 mL saline, 500 mL hydroxyethyl starch, or no fluid and monitored for 60 minutes. Data were modeled using repeated measures multivariate analysis of variance.

Results

There were no significant differences before treatment. At 20 minutes after hemorrhage, there was no significant difference in mean sBP between treated groups, however, control animals displayed significantly lower mean sBP (p < 0.001). Mean arterial pressure and heart rate improved in the treated groups but not in the control group (p < 0.02). Prothrombin time was longer and platelet counts were lower in the Hextend group (p < 0.05). Moreover, thromboelastography analysis showed longer clotting (K) times (p < 0.05) for the hydroxyethyl starch-treated group.

Conclusion

Hydroxocobalamin restored blood pressure more effectively than no treatment and as effectively as hydroxyethyl starch but did not adversely affect coagulation.

Automated Balloon Control in Resuscitative Endovascular Balloon Occlusion of the Aorta

OBJECTIVE

The goal of this study was to demonstrate the technical feasibility of automated balloon pressure management during resuscitative endovascular balloon occlusion of the aorta (REBOA) in the pre-clinical setting.

METHODS

This paper presents an intelligent balloon management device which automates the balloon inflation process, preventing the possibility of balloon over or under inflation, optimizes inflation pressure, and if indicated, deflates automating partial REBOA to allow the distal organ perfusion. Edwards TruWave pressure transducers are used to monitor the blood pressure proximal and distal to the balloon, as well as the internal balloon pressure. A faux PID controller, implemented on an Arduino platform, is used in a feedback control loop to allow a user-defined mean arterial pressure setpoint to be reached, via a syringe driver which allows intelligent inflation and deflation of the catheter balloon.

RESULTS

Ex vivo testing on a vascular perfusion simulator provided the characteristic behavior of a fully occluded aorta, namely the decrease of distal pressure to zero. In vivo testing on live porcine models indicated that automated partial REBOA is achievable and by enabling partial occlusion may offer improved medical outcomes compared to the manual control.

CONCLUSION

Automated balloon pressure management of endovascular occlusion is feasible and can be successfully implemented without changes on current clinical workflows.

SIGNIFICANCE

With further development, automated balloon management may significantly improve clinical outcomes in REBOA.

Complete and Partial Aortic Occlusion for the Treatment of Hemorrhagic Shock in Swine

Hemorrhage remains the leading cause of preventable deaths in trauma. Endovascular management of non-compressible torso hemorrhage has been at the forefront of trauma care in recent years. Since complete aortic occlusion presents serious concerns, the concept of partial aortic occlusion has gained a growing attention. Here, we present a large animal model of hemorrhagic shock to investigate the effects of a novel partial aortic balloon occlusion catheter and compare it with a catheter that works on the principles of complete aortic occlusion. Swine are anesthetized and instrumented in order to conduct controlled fixed-volume hemorrhage, and hemodynamic and physiological parameters are monitored. Following hemorrhage, aortic balloon occlusion catheters are inserted and inflated in the supraceliac aorta for 60 min, during which the animals receive whole-blood resuscitation as 20% of the total blood volume (TBV). Following balloon deflation, the animals are monitored in a critical care setting for 4 h, during which they receive fluid resuscitation and vasopressors as needed. The partial aortic balloon occlusion demonstrated improved distal mean arterial pressures (MAPs) during the balloon inflation, decreased markers of ischemia, and decreased fluid resuscitation and vasopressor use. As swine physiology and homeostatic responses following hemorrhage have been well-documented and are like those in humans, a swine hemorrhagic shock model can be used to test various treatment strategies. In addition to treating hemorrhage, aortic balloon occlusion catheters have become popular for their role in cardiac arrest, cardiac and vascular surgery, and other high-risk elective surgical procedures.

Biomaterials and Advanced Technologies for Hemostatic Management of Bleeding

Bleeding complications arising from trauma, surgery, and as congenital, disease-associated, or drug-induced blood disorders can cause significant morbidities and mortalities in civilian and military populations. Therefore, stoppage of bleeding (hemostasis) is of paramount clinical significance in prophylactic, surgical, and emergency scenarios. For externally accessible injuries, a variety of natural and synthetic biomaterials have undergone robust research, leading to hemostatic technologies including glues, bandages, tamponades, tourniquets, dressings, and procoagulant powders. In contrast, treatment of internal noncompressible hemorrhage still heavily depends on transfusion of whole blood or blood's hemostatic components (platelets, fibrinogen, and coagulation factors). Transfusion of platelets poses significant challenges of limited availability, high cost, contamination risks, short shelf-life, low portability, performance variability, and immunological side effects, while use of fibrinogen or coagulation factors provides only partial mechanisms for hemostasis. With such considerations, significant interdisciplinary research endeavors have been focused on developing materials and technologies that can be manufactured conveniently, sterilized to minimize contamination and enhance shelf-life, and administered intravenously to mimic, leverage, and amplify physiological hemostatic mechanisms. Here, a comprehensive review regarding the various topical, intracavitary, and intravenous hemostatic technologies in terms of materials, mechanisms, and state-of-art is provided, and challenges and opportunities to help advancement of the field are discussed.

Incremental balloon deflation following complete resuscitative endovascular balloon occlusion of the aorta results in steep inflection of flow and rapid reperfusion in a large animal model of hemorrhagic shock

INTRODUCTION

To avoid potential cardiovascular collapse after resuscitative endovascular balloon occlusion of the aorta (REBOA), current guidelines recommend methodically deflating the balloon for 5 minutes to gradually reperfuse distal tissue beds. However, anecdotal evidence suggests that this approach may still result in unpredictable aortic flow rates and hemodynamic instability. We sought to characterize aortic flow dynamics following REBOA as the balloon is deflated in accordance with current practice guidelines.

METHODS

Eight Yorkshire-cross swine were splenectomized, instrumented, and subjected to rapid 25% total blood volume hemorrhage. After 30 minutes of shock, animals received 60 minutes of Zone 1 REBOA with a low-profile REBOA catheter. During subsequent resuscitation with shed blood, the aortic occlusion balloon was gradually deflated in stepwise fashion at the rate of 0.5 mL every 30 seconds until completely deflated. Aortic flow rate and proximal mean arterial pressure (MAP) were measured continuously over the period of balloon deflation.

RESULTS

Graded balloon deflation resulted in variable initial return of aortic flow (median, 78 seconds; interquartile range [IQR], 68-105 seconds). A rapid increase in aortic flow during a single-balloon deflation step was observed in all animals (median, 819 mL/min; IQR, 664-1241 mL/min) and corresponded with an immediate decrease in proximal MAP (median, 30 mm Hg; IQR, 14.5-37 mm Hg). Total balloon volume and time to return of flow demonstrated no correlation (r = 0.016).

CONCLUSION

This study is the first to characterize aortic flow during balloon deflation following REBOA. A steep inflection point occurs during balloon deflation that results in an abrupt increase in aortic flow and a concomitant decrease in MAP. Furthermore, the onset of distal aortic flow was inconsistent across study animals and did not correlate with initial balloon volume or relative deflation volume. Future studies to define the factors that affect aortic flow during balloon deflation are needed to facilitate controlled reperfusion following REBOA.

A novel model of highly lethal uncontrolled torso hemorrhage in swine

INTRODUCTION

A reproducible, lethal noncompressible torso hemorrhage model is important to civilian and military trauma research. Current large animal models balancing clinical applicability with standardization and internal validity. As such, large animal models of trauma vary widely in the surgical literature, limiting comparisons. Our aim was to create and validate a porcine model of uncontrolled hemorrhage that maximizes reproducibility and standardization.

METHODS

Seven Yorkshire-cross swine were anesthetized, instrumented, and splenectomized. A simple liver tourniquet was applied before injury to prevent unregulated hemorrhage while creating a traumatic amputation of 30% of the liver. Release of the tourniquet and rapid abdominal closure following injury provided a standardized reference point for the onset and duration of uncontrolled hemorrhage. At the moment of death, the liver tourniquet was quickly reapplied to provide accurate quantification of intra-abdominal blood loss. Weight and volume of the resected and residual liver segments were measured. Hemodynamic parameters were recorded continuously throughout each experiment.

RESULTS

This liver injury was rapidly and universally lethal (11.2 ± 4.9 min). The volume of hemorrhage (35.8% ± 6% of total blood volume) and severity of uncontrolled hemorrhage (100% of animals deteriorated to a sustained mean arterial pressure <35 mmHg for 5 min) were consistent across all animals. Use of the tourniquet effectively halted preprocedure and postprocedure blood loss allowing for accurate quantification of amount of hemorrhage over a defined period. In addition, the tourniquet facilitated the creation of a consistent liver resection weight (0.0043 ± 0.0003 liver resection weight: body weight) and as a percentage of total liver resection weight (27% ± 2.2%).

CONCLUSIONS

This novel tourniquet-assisted noncompressible torso hemorrhage model creates a standardized, reproducible, highly lethal, and clinically applicable injury in swine. Use of the tourniquet allowed for consistent liver injury and precise control over hemorrhage. Recorded blood loss was similar across all animals. Improving reproducibility and standardization has the potential to offer improvements in large animal translational models of hemorrhage.

LEVEL OF EVIDENCE

Level I.

Prehospital control of life-threatening truncal and junctional haemorrhage is the ultimate challenge in optimizing trauma care; a review of treatment options and their applicability in the civilian trauma setting

Introduction

Exsanguination following trauma is potentially preventable. Extremity tourniquets have been successfully implemented in military and civilian prehospital care. Prehospital control of bleeding from the torso and junctional area’s remains challenging but offers a great potential to improve survival rates. This review aims to provide an overview of potential treatment options in both clinical as preclinical state of research on truncal and junctional bleeding. Since many options have been developed for application in the military primarily, translation to the civilian situation is discussed.

Methods

Medline (via Pubmed) and Embase were searched to identify known and potential prehospital treatment options. Search terms were|: haemorrhage/hemorrhage, exsanguination, junctional, truncal, intra-abdominal, intrathoracic, intervention, haemostasis/hemostasis, prehospital, en route, junctional tourniquet, REBOA, resuscitative thoracotomy, emergency thoracotomy, pelvic binder, pelvic sheet, circumferential. Treatment options were listed per anatomical site: axilla, groin, thorax, abdomen and pelvis Also, the available evidence was graded in (pre) clinical stadia of research.

Results

Identified treatment options were wound clamps, injectable haemostatic sponges, pelvic circumferential stabilizers, resuscitative thoracotomy, resuscitative endovascular balloon occlusion of the aorta (REBOA), intra-abdominal gas insufflation, intra-abdominal self-expanding foam, junctional and truncal tourniquets. A total of 70 papers on these aforementioned options was retrieved. No clinical reports on injectable haemostatic sponges, intra-abdominal insufflation or self-expanding foam injections and one type of junctional tourniquets were available.

Conclusion

Options to stop truncal and junctional traumatic haemorrhage in the prehospital arena are evolving and may offer a potentially great survival advantage. Because of differences in injury pattern, time to definitive care, different prehospital scenario’s and level of proficiency of care providers; successful translation of various military applications to the civilian situation has to be awaited. Overall, the level of evidence on the retrieved adjuncts is extremely low.

The role of REBOA in the control of exsanguinating torso hemorrhage

The management of patients with exsanguinating torso hemorrhage is challenging. Emergency surgery, with the occasional use of resuscitative thoracotomy for patient in extremis, is the current standard. Recent reports of REBOA (resuscitative endovascular balloon occlusion of the aorta) have led to discussions about changing paradigms in the management of patients in both civilian and military are nas. We submit that broad and liberal application of this technique is premature given the current data and in light of historical experience. We propose an algorithm for the management of patients with exsanguinating torso hemorrhage, as well as a set of research questions that we feel can help clarify the role of REBOA in modern trauma care in a variety of trauma settings.

A prospective, randomized trial of intravenous hydroxocobalamin versus whole blood transfusion compared to no treatment for Class III hemorrhagic shock resuscitation in a prehospital swine model

OBJECTIVES

The objective was to compare systolic blood pressure (sBP) over time in swine that have had 30% of their blood volume removed (Class III shock) and treated with intravenous (IV) whole blood or IV hydroxocobalamin, compared to nontreated control animals.

METHODS

Thirty swine (45 to 55 kg) were anesthetized, intubated, and instrumented with continuous femoral and pulmonary artery pressure monitoring. Animals were hemorrhaged a total of 20 mL/kg over a 20-minute period. Five minutes after hemorrhage, animals were randomly assigned to receive 150 mg/kg IV hydroxocobalamin solubilized in 180 mL of saline, 500 mL of whole blood, or no treatment. Animals were monitored for 60 minutes thereafter. A sample size of 10 animals per group was determined based on a power of 80% and an alpha of 0.05 to detect an effect size of at least a 0.25 difference (>1 standard deviation) in mean sBP between groups. sBP values were analyzed using repeated-measures analysis of variance (RANOVA). Secondary outcome data were analyzed using repeated-measures multivariate analysis of variance (RMANOVA).

RESULTS

There were no significant differences between hemodynamic parameters of IV hydroxocobalamin versus whole blood versus control group at baseline (MANOVA; Wilks' lambda; p = 0.868) or immediately posthemorrhage (mean sBP = 47 mm Hg vs. 41 mm Hg vs. 37 mm Hg; mean arterial pressure = 39 mm Hg vs. 28 mm Hg vs. 34 mm Hg; mean serum lactate = 1.2 mmol/L vs. 1.4 mmol/L vs. 1.4 mmol/L; MANOVA; Wilks' lambda; p = 0.348). The outcome RANOVA model detected a significant difference by time between groups (p < 0.001). Specifically, 10 minutes after treatment, treated animals showed a significant increase in mean sBP compared to nontreated animals (mean sBP = 76.3 mm Hg vs. 85.7 mm Hg vs. 51.1 mm Hg; p < 0.001). RMANOVA modeling of the secondary data detected a significant difference in mean arterial pressure, heart rate, and serum lactate (p < 0.001). Similar to sBP, 10 minutes after treatment, treated animals showed a significant increase in mean arterial pressure compared to nontreated animals (mean arterial pressure = 67.7 mm Hg vs. 61.4 mm Hg vs. 40.5 mm Hg). By 10 minutes, mean heart rate was significantly slower in treated animals compared to nontreated animals (mean heart rate = 97.3 beats/min vs. 95.2 beats/min vs. 129.5 beats/min; p < 0.05). Serum lactate, an early predictor of shock, continued to rise in the control group, whereas it did not in treated animals. Thirty minutes after treatment, serum lactate values of treated animals were significantly lower compared to nontreated animals (p < 0.05). This trend continued throughout the 60-minute observation period such that 60-minute values for lactate were 1.4 mmol/L versus 1.1 mmol/L versus 3.8 mmol/L. IV hydroxocobalamin produced a statistically significant increase in systemic vascular resistance compared to control, but not whole blood, with a concomitant decrease in cardiac output.

CONCLUSIONS

Intravenous hydroxocobalamin was more effective than no treatment and as effective as whole blood transfusion, in reversing hypotension and inhibiting rises in serum lactate in this prehospital, controlled, Class III swine hemorrhage model.

A laparoscopic swine model of noncompressible torso hemorrhage

BACKGROUND

Hemorrhage persists as the leading cause of potentially preventable civilian and military death. Noncompressible torso hemorrhage (NCTH) is a particularly lethal injury complex, with few contemporary prehospital interventions available. Various porcine models of hemorrhage have been developed for civilian and military trauma research. However, the predominant contemporary models lack key physiologic characteristics including the natural tamponade provided by an intact abdominal wall.To improve physiologic and clinical relevance, we developed a laparoscopic model of NCTH. This approach maintains both the integrity of the peritoneum and the natural tamponade effect of an intact abdominal wall while preserving the intrinsic physiologic responses to hemorrhage. Furthermore, we present data quantifying the contribution of the swine contractile spleen in the context of uncontrolled hemorrhage.

METHODS

Anesthetized adult male Yorkshire swine underwent a laparoscopic Grade V liver injury, with or without open preinjury splenectomy. Animals were observed without intervention for a total of 120 minutes after injury to simulate point of injury, transport time, and arrival at hospital.

RESULTS

Shed blood-to-body weight ratio did not differ among groups; however, mortality was higher in splenectomized animals (67% vs. 33%). Cox regression modeling demonstrated a critical time point of 45 minutes and blood pressure as significant predictors of mortality.

CONCLUSION

This study describes a model of NCTH that reflects clinically relevant physiology in trauma and uncontrolled hemorrhage. In addition, it quantitatively assesses the role of the swine contractile spleen in the described model.

Development of a fatal noncompressible truncal hemorrhage model with combined hepatic and portal venous injury in normothermic normovolemic swine

Noncompressible truncal hemorrhage and brain injury currently account for most early mortality of warfighters on the battlefield. There is no effective treatment for noncompressible truncal hemorrhage, other than rapid evacuation to a surgical facility. The availability of an effective field treatment for noncompressible truncal hemorrhage could increase the number of warfighters salvaged from this frequently-lethal scenario. Our intent was to develop a porcine model of noncompressible truncal hemorrhage with a ∼50% one-hour mortality so that we could develop new treatments for this difficult problem. Normovolemic normothermic domestic swine (barrows, 3 months old, 34–36 kg) underwent one of three injury types through a midline incision: 1) central stellate injury (N = 6); 2) excision of a portal vein branch distal to the main PV trunk (N = 6); or 3) hemi-transection of the left lateral lobe of the liver at its base (N = 10). The one-hour mortality of these injuries was 0, 82, and 40%, respectively; the final mean arterial pressure was 65, 24, and 30 mm Hg, respectively; and the final hemoglobin was 8.3, 2.3, and 3.6 g/dL, respectively. Hemi-transection of the left lateral lobe of the liver appeared to target our desired mortality rate better than the other injury mechanisms.

The inflammatory sequelae of aortic balloon occlusion in hemorrhagic shock

BACKGROUND

Resuscitative endovascular balloon occlusion of the aorta (REBOA) is a hemorrhage control and resuscitative adjunct that has been demonstrated to improve central perfusion during hemorrhagic shock. The aim of this study was to characterize the systemic inflammatory response associated and cardiopulmonary sequelae with 30, 60, and 90 min of balloon occlusion and shock on the release of interleukin 6 (IL-6) and tumor necrosis factor alpha.

MATERIALS AND METHODS

Anesthetized female Yorkshire swine (Sus scrofa, weight 70-90 kg) underwent a 35% blood volume-controlled hemorrhage followed by thoracic aortic balloon occlusion of 30 (30-REBOA, n = 6), 60 (60-REBOA, n = 8), and 90 min (90-REBOA, n = 6). This was followed by resuscitation with whole blood and crystalloid over 6 h. Animals then underwent 48 h of critical care with sedation, fluid, and vasopressor support.

RESULTS

All animals were successfully induced into hemorrhagic shock without mortality. All groups responded to aortic occlusion with a rise in blood pressure above baseline values. IL-6, as measured (picogram per milliliter) at 8 h, was significantly elevated from baseline values in the 60-REBOA and 90-REBOA groups: 289 ± 258 versus 10 ± 5; P = 0.018 and 630 ± 348; P = 0.007, respectively. There was a trend toward greater vasopressor use (P = 0.183) and increased incidence of acute respiratory distress syndrome (P = 0.052) across the groups.

CONCLUSIONS

REBOA is a useful adjunct in supporting central perfusion during hemorrhagic shock; however, increasing occlusion time and shock results in a greater IL-6 release. Clinicians must anticipate inflammation-mediated organ failure in post-REBOA use patients.

Modelling for conflict: the legacy of ballistic research and current extremity in vivo modelling

Extremity ballistic injury is unique and the literature intended to guide its management is commonly misinterpreted. In order to care for those injured in conflict and conduct appropriate research, clinicians must be able to identify key in vivo studies, understand their weaknesses and desist the propagation of miscited and misunderstood ballistic dogma. This review provides the only inclusive critical overview of key studies of relevance to military extremity injury. In addition, the non-ballistic studies of limb injury, stabilisation and contamination that will form the basis from which future small animal extremity studies are constructed are presented. With an awareness of the legacy of military wound models and an insight into available generic models of extremity injury and contamination, research teams are well placed to optimise future military extremity injury management.

Aortic balloon occlusion is effective in controlling pelvic hemorrhage

BACKGROUND

The objective of this study was to evaluate the efficacy of resuscitative endovascular aortic balloon occlusion (REBOA) of the distal aorta in a porcine model of pelvic hemorrhage.

METHODS

Swine were entered into three phases of study: injury (iliac artery), hemorrhage (45 s), and intervention (180 min). Three groups were studied: no intervention (NI, n = 7), a kaolin-impregnated gauze (Combat Gauze) (CG, n = 7), or REBOA (n = 7). The protocol was repeated with a dilutional coagulopathy (CG-C, n = 7, and REBOA-C, n = 7). Measures of physiology, rates of hemorrhage, and mortality were recorded.

RESULTS

Rate of hemorrhage was greatest in the NI group, followed by the REBOA and CG groups (822 ± 415 mL/min versus 11 ± 13 and 0.2 ± 0.4 mL/min respectively; P < 0.001). MAP following intervention (at 15 min) was the same in the CG and REBOA groups and higher than in the NI group (70 ± 4 and 70 ± 11 mm Hg versus 5 ± 13 mm Hg respectively; P < 0.001). There was 100% mortality in the NI group, with no deaths in the CG or REBOA group. In the setting of coagulopathy, the rate of bleeding was higher in the CG-C versus the REBOA-C group (229 ± 295 mL/min versus 20 ± 7 mL/min, P = 0.085). MAP following intervention (15 min) was higher in the REBOA-C than the CG-C group (71 ± 12 mm Hg versus 28 ± 31 mm Hg; P = 0.005). There were 5 deaths (71.4%) in the CG-C group, but none in the REBOA-C group (P = 0.010).

CONCLUSION

Balloon occlusion of the aorta is an effective method to control pelvic arterial hemorrhage. This technique should be further developed as an adjunct to manage noncompressible pelvic hemorrhage.