Secondary Logo

Journal Logo

Titrate to equilibrate and not exsanguinate! Characterization and validation of a novel partial resuscitative endovascular balloon occlusion of the aorta catheter in normal and hemorrhagic shock conditions

Forte, Dominic M. MD; Do, Woo S. MD; Weiss, Jessica B. MD; Sheldon, Rowan R. MD; Kuckelman, John P. DO; Eckert, Matthew J. MD; Martin, Matthew J. MD

Journal of Trauma and Acute Care Surgery: November 2019 - Volume 87 - Issue 5 - p 1015–1025
doi: 10.1097/TA.0000000000002378
Editor's Choice

BACKGROUND Resuscitative endovascular balloon occlusion of the aorta (REBOA) is a significant advancement in the control of noncompressible truncal hemorrhage. However, its ischemic burden and reperfusion injury following balloon deflation limits its utilization. Partial restoration of aortic flow during REBOA has the potential to balance hemorrhage control and ischemia. This study validates the mechanics, physiology, and optimal partial flow rates using a prototype partial REBOA (pREBOA) device.

METHODS Twenty-five swine underwent placement of aortic flow probes and zone 1 pREBOA. Experiment 1 (N = 5) animals were not injured and assessed the tested the catheters ability to titrate and control flow. Experiment 2 (N = 10) added 20% hemorrhage and either solid organ, or abdominal vascular injury to compare flow rate and rebleeding from injuries. Experiment 3 (N = 10) swine were similarly prepared, hemorrhaged, and underwent pREBOA at set partial flow rates for 2 hours followed by complete deflation for 30 minutes.

RESULTS Balloon volume at minimum flow (mean, 0.09 L/min) was 3.5 mL to 6.0 mL. Half maximal flow was achieved with 56.5% of maximum balloon inflation. Partial REBOA allowed very fine titration of flow rates. Rebleeding occurred at 0.45 L/min to 0.83 L/min. Distal flow of 0.7 L/min had 50% survival, 0.5 had 100% survival, and 0.3 L had 50% survival with mean end lactates of 9.6, 12.6, and 13.3, respectively. There was a trend toward hyperkalemia and hypocalcemia in nonsurvivors.

CONCLUSION The pREBOA device demonstrated a high level of titratability for restoration of aortic flow. An optimal partial flow of 0.5 L/min was effective at hemorrhage control while limiting the burden of ischemic injury, and extending the tolerable duration of zone 1 occlusion. Aggressive calcium supplementation prior to and during partial occlusion and reperfusion may be warranted to prevent hyperkalemic arrest.

From the Department of Surgery (D.F., W.S.D., J.B.W., R.R.S., J.P.K., M.J.E.), Madigan Army Medical Center, Tacoma, Washington: and Trauma and Emergency Surgery Service (M.J.M.), Scripps Mercy Medical Center, San Diego, California.

Submitted: January 15, 2019, Revised: May 6, 2019, Accepted: May 9, 2019, Published online: May 21, 2019.

Presented at the 49th Annual Meeting of the Western Trauma Association, March 3 to 8, 2019 in Snowmass, CO.

Address for reprints: Matthew J. Martin, MD, Trauma and Emergency Surgery Service, Scripps Mercy Hospital, Suite 641, 550 Washington Ave, San Diego, CA 92127; email:

Online date: May 23, 2019

Hemorrhage from the thorax, abdomen, and pelvis is the leading cause of preventable death in soldiers with survivable battlefield injuries.1 Although the lethality and complexity in management of these injuries has long been recognized, the term noncompressible torso/truncal hemorrhage (NCTH) only came to common use with reviews of casualties from the wars in Iraq and Afghanistan.2,3 These conflicts saw a return to the use of tourniquets for extremity hemorrhage and motivated the development of a range of effective junctional tourniquets. No comparable advances have been made with regard to NCTH. Application of this definition of NCTH in civilian trauma populations has demonstrated these injuries are equally as life-threatening, with nearly 45% mortality and significant morbidity.3 Furthering the challenge in treating these injuries is the imminent need for rapid and effective interventions, either in the prehospital or very early in-hospital phases of care. Analysis utilizing the National Trauma Databank demonstrated that mortality from NCTH rises steeply after 30 minutes from time of injury, much quicker than the fabled, “golden hour.”4 Taken together, there is a clear and present need for rapid, field-ready means addressing NCTH. Out of this need, resuscitative endovascular balloon occlusion of the aorta (REBOA) has emerged as a promising technology.

It is important to note that the concept of balloon occlusion of the aorta in NCTH is not new. A report details its use with an early balloon tipped catheter in two causalities during the Korean War.5 The development of more widely produced Fogarty catheter saw some advocate its application in aortic occlusion for hemorrhage control in trauma.6,7 There was never widespread use of this technique with these catheters. The requirement for open femoral cutdown, control of the vessel prior to use, and vessel injuries due to the size of the devices or sheaths limited adoption and application. It was not until the current generation of simpler and smaller REBOA catheters and sheaths that enthusiasm increased. Current technology minimizes procedural morbidity by allowing for percutaneous placement through small sheaths providing relatively safe and rapid placement.8,9 Despite these improvements in technology, there remains the substantial threat of ischemia-reperfusion injury implicit in causing complete cessation of aortic flow. This is of particular concern for zone 1 REBOA placement (descending thoracic aorta), which limits the tolerable time of occlusion to 60 minutes or less. This major concern and limitation has resulted in continued uncertainty and debate about the place of REBOA in trauma care.10

Recent work on REBOA has focused on the utility of intermittent and/or partial REBOA (pREBOA) to balance hemorrhage control and ischemic burden. Partial REBOA appears to be a promising technique in large animal studies, although the current literature is limited to a very small number of preclinical studies. Porcine models have shown ultrasound evidence of preservation of mesenteric perfusion with decreased serum measures of ischemia despite prolonged partial occlusion.11 Additionally, pREBOA appears to yield near physiologic carotid blood flow, avoiding the theoretically damaging supraphysiologic proximal pressures observed with complete occlusion.12,13 However, these studies are limited by the lack of detailed flow data and correlations between flow and balloon volumes, or they required a highly complex set of computer-controlled variable aortic cross-clamp and extracorporeal recirculation devices.13–15

The few studies that have attempted pREBOA with a standard catheter have utilized compliant, single balloon devices such as the ER-REBOA (Prytime Medical, Boerne, TX) or Rescue Balloon (Tokai Medical, Tokyo, Japan).11,13 These favorable large animal studies along with intuitive rationale has given rise to attempts at clinical use of pREBOA.16–18 Like the large animal studies, these describe slow partial deflation of single, compliant balloon catheters in an attempt to restore “partial” aortic flow. However, the compliant nature of the balloons in these catheters creates a sudden drop off of occlusion as the balloon is deflated and when it loses apposition to the aortic wall. This results in large changes in flow that are poorly controlled by partial balloon deflation, in essence going rapidly from very low partial flow to full flow. The overall effect is to make careful titration of occlusion difficult if not impossible in a dynamic system, such as an aorta actively responding to hypotension and resuscitation. Although pREBOA is a promising concept that could prove a much needed advance in the treatment of NCTH, new technology and techniques are necessary to make this clinically reliable and relevant.

The objective of this study was to test the function and impact of a new prototype REBOA catheter that was specifically designed to allow for fine and titratable control of aortic flow across the spectrum from balloon inflation to deflation. This new pREBOA device (Fig. 1, PryTime Medical) has a novel and proprietary configuration of multiple balloons that allows fine control of flow in static ex vivo models but has not been evaluated in any in vivo models. We aimed to test this device in uninjured animals and using a previously validated porcine hemorrhagic shock model with predetermined vascular or solid organ injuries.19,20 This report will describe the initial characterization of this catheter's function and ability to finely control partial aortic flow, as well as initial validation in both vascular and solid organ injury models.

Figure 1

Figure 1

Back to Top | Article Outline


All research was performed under a protocol approved by our Institutional Animal Care and Use Committee. Swine were cared for as described by the Institute of Laboratory Animal Research's Guide for the Care and Use of Laboratory Animals. Yorkshire swine (Sus scrofa) ranging in weight from 40 kg to 50 kg were placed under general endotracheal anesthesia with isoflurane.

In all experiments, a midline neck incision was made and a 5-Fr carotid arterial catheter, 7-Fr left external jugular sheath, and a 9.5-Fr Cordis right external jugular sheath was inserted via a Seldinger technique. A Swan-Ganz pulmonary artery catheter was then placed through the Cordis. A left anterolateral thoracotomy was performed through the eighth or ninth intercostal space and the distal thoracic aorta was circumferentially dissected. A perivascular ultrasonic flow probe was then placed posterior to the distal thoracic aorta to accurately measure distal flow during occlusion and then titrated restoration of flow (Transonic, Ithaca, NY). A midline laparotomy was performed and an intra-abdominal Foley catheter was placed into the bladder for decompression and measurement of urine output. A left medial visceral rotation was then performed to expose the pelvic vessels. A 7-Fr sheath was placed through the common iliac artery via a Seldinger technique and secured to the surrounding tissues. A prototype bilobed pREBOA catheter (Prytime Medical) was introduced through this sheath into the distal thoracic aorta (zone 1) until it was palpated to be proximal to the flow probe by at least 3 cm.

In experiment 1 (N = 5), no further preparation was performed. The pREBOA catheter was inflated until distal mean arterial pressure (dMAP) and distal aortic flow was not affected by additional inflation. The balloon was then incrementally deflated in units of 0.1 mL while recording the heart rate, proximal mean arterial pressure (pMAP), dMAP, and distal aortic flow every minute.

In experiment 2 (N = 10), the described neck, thoracic, and abdominal preparations were made. Following this, a controlled 20% blood volume hemorrhage was performed (see below). Animals then underwent a solid organ injury (n = 5) or vascular injury (n = 5) described below. Following injury, the animals were observed until the MAP fell below 40 mm Hg. The pREBOA catheter was inflated for a period of 10 minutes to allow for clot stabilization. The catheter was then incrementally deflated in units of 0.2 mL while the injuries were directly observed for rebleeding.

In experiment 3 (N = 10), the initial setup proceeded as described above. Swine then underwent controlled 20% blood volume hemorrhage (see below) followed by pREBOA at set partial flow rates for 2 hours followed by complete deflation for 30 minutes. Animals were randomized to partial flow goals of 0.3 L/min, 0.5 L/min, and 0.7 L/min as these spanned the range of distal aortic flow rates at which there was observed rebleeding from described injuries in experiment 2. The experiment was determined to be successful if the animal survived to the conclusion of the 30 minutes of deflation of the pREBOA.

Back to Top | Article Outline

Controlled Hemorrhage and Combat Resuscitation

In experiments 2 and 3, following successful placement of all lines, a controlled 20% estimated blood volume hemorrhage was performed via the 7-Fr left external jugular sheath. Hemorrhage was performed as fast as possible without being fatal to the animal and without resuscitation to induce a shock-like state. Shed blood was collected into citrated bags for later use in resuscitation.

Of note, animals did not undergo splenectomies. Prior work in from our group with this model failed to demonstrate a significant influence on hemostasis with splenectomy.18,20,21 As such, this was not a part of the preparation of the animals.

Resuscitation was designed to mimic the prehospital and en-route resuscitation currently utilized in combat and recommended in the Tactical Combat Casualty Care guidelines.22,23 This simulated initial resuscitation at or near the point of injury by medics without blood products available, and then en-route transport with the availability of a limited supply of whole blood. Specifically, a 250-mL synthetic starch-based colloid (Hextend) was administered when pMAP fell below 40 mm Hg following injury. Following completion of the colloid, the shed blood from the 20% blood volume hemorrhage (whole blood) was administered to keep pMAP above 40 mm Hg. If at any point pMAP was above 40 mm Hg, resuscitation was paused.

Back to Top | Article Outline

Solid Organ Injury Creation

The solid organ injury group utilized a previously validated liver injury model with an expected 30% to 50% mortality without early intervention. This consisted of amputation of the left medial lobe of the liver approximately 1 cm from the hilum. The liver was exposed through an upper midline extension of existing midline laparotomy. The lobe of liver lateral to the lobe which the gallbladder is adhered to was grasped and elevated out of the incision. The injury was then created by rapidly and sharply by dividing this lobe of the liver as close to the hilum as possible. This injury was arrived at by model development in which uniform fatality and similar hemorrhage were observed in five consecutive animals.

Back to Top | Article Outline

Vascular Injury Creation

The vascular injury model has also been previously described and utilized by our group,19 and consists of a combined major abdominal arterial and venous injury and resultant uncontrolled hemorrhage. In brief, the right iliac artery and vein were exposed through a low midline extension of existing midline incision. The overlying peritoneum was excised and the common iliac artery and vein were accessed using the Seldinger technique. Through serial dilation a 12-Fr dilator was ultimately placed in the common iliac artery and a 16-Fr dilator placed in the common iliac vein. The injury comprised of the simultaneous removal of these dilators.

Back to Top | Article Outline


The experiment was concluded if the pMAP was persistently <20 mm Hg for 5 minutes, if there was asystole or no detectable blood pressure for at least 1 minute, or when the predetermined time course of the experiment had been completed. For animals that died prior to the timed endpoint, the time from initial to death was recorded. All animals were then euthanized using a barbiturate based euthanasia solution upon completion of experiments. The abdominal cavity was then explored to identify and confirm the injuries, confirm the adequate position and placement of the REBOA catheter, and to collect and quantify the amount of blood loss.

Back to Top | Article Outline

Data Collection

Hemodynamic parameters including pMAP, dMAP, distal aortic flow, and heart rate were monitored continuously and recorded every minute for the duration of the experiment. Pulmonary artery catheter measurements, hemorrhage volume, and urine output were recorded at baseline and every 30 min during the experiment. Laboratory parameters included arterial blood gas, basic metabolic panel, lactate, and calcium were recorded at baseline and every 30 min during the experiment. Rotational thromboelastometry (ROTEM) (Tem International, Munich, Germany) was obtained at baseline and every hour during the experiment.

Back to Top | Article Outline

Statistical Analysis

All data were collected and stored on a secure electronic network. ANOVA was used to compare between flow groups in experiment 3. The Bonferroni test was utilized post hoc to elucidate differences detected by ANOVA. Student's t-test and χ2 were utilized to compare outcomes between surviving and nonsurviving animals in experiment 3. All statistical analyses were preformed using IBM SPSSS 24 (IBM Corp., Armonk, NY).

Back to Top | Article Outline


A total of 25 animals were included in the study. Mean weight was 45.2 kg (standard deviation, 3.4 kg).

Back to Top | Article Outline

Experiment 1

The first experiment sought to demonstrate the pREBOA catheters ability to achieve aortic occlusion and then to allow fine control and titration in restoring partial aortic flow. Table 1 details the heart rate, distal aortic flow, pMAP, and dMAP at four key points through the inflation cycle. Baseline measurements were taken with the pREBOA in place, but prior to inflation. Maximum Balloon inflation was the balloon volume at the nadir of flow. Maximum inflection of flow was the point at which the change in flow with regard to time was maximal. This occurred at an average of 56.5% of maximal balloon inflation (standard deviation 0.007%). Balloon deflation parameters were recorded at the completion of the inflation cycle once the balloon volume had returned to zero.



To achieve maximum occlusion, 3.2 mL to 6.8 mL was needed. To standardize the inflation cycles for purposes of comparison all data points were ordered by the percentage of maximal inflation for that inflation cycle (balloon volume at data point divided by the maximal balloon volume for that inflation cycle). The resulting titration curves are shown in Figures 2A-C. As outlined in these figures, the pREBOA demonstrated a high degree of control and titratability across the full range from minimal distal aortic flow to restoration of complete flow.

Figure 2

Figure 2

Back to Top | Article Outline

Experiment 2

The second experiment aimed to elucidate the rate of distal aortic flow at which rebleeding occurred in the setting of either a major vascular injury or a high-grade solid organ injury as described above. With vascular injury (N = 5), rebleeding was observed at distal aortic flow ranging from 0.45 L/min to 0.83 L/min (average, 0.63 L/min, standard deviation 0.14 L/min). Average pMAP and dMAP at rebleeding were 34 mm Hg and 21 mm Hg, respectively. With solid organ injury (N = 5), rebleeding occurred in three of the five pigs, hemorrhage was never fully controlled in one pig, and in the final pig died of apparent cardiac causes prior to rebleeding being observed. In the three pigs which had observed rebleeding, this occurred at distal aortic flows of 0.24 L/min to 0.50 L/min (average, 0.35 L/min; standard deviation, 0.14 L/min). Average pMAP and dMAP at rebleeding were 45 mm Hg and 12 mm Hg, respectively. Comparing these groups, there was a significantly lower flow and dMAP at observed rebleeding with the solid organ injury animals (p = 0.04 and p = 0.009).

Back to Top | Article Outline

Experiment 3

The third experiment assessed the effects of prolonged partial occlusion at varying distal flow rates. Three goal distal aortic flow rates were utilized spanning the range of observed rebleeding in experiment 2: 0.3 mL/h (n = 2), 0.5 mL/h (n = 4), and 0.7 mL/h (n = 4). Survival outcomes are shown in Figure 3. All animals (4/4) that underwent partial flow at 0.5 L/h survived to completion of the experiment. Half of the animals undergoing partial flow at 0.3 L/h (1/2) and 0.7 L/h (2/4) died prior the completion of the experiment. In two animals that died progressive heart block was observed culminating complete heart block with subsequent death. An additional animal died following deflation of the balloon without observed arrhythmia. Motivated by this observation, serial calcium levels were obtained on the final four animals (0.7 L/h, 0.3 L/h, and two 0.5-L/h).

Figure 3

Figure 3

Physiologic and laboratory parameters over the course of the experiment are shown in Tables 2A and 2B. There were no statistically significant differences between the distal flow goal groups in the laboratory parameters when tested by ANOVA. In regard to the physiologic parameters, significant differences were observed between pMAPs at 90 minutes and dMAPs at 90 minutes and 120 minutes. For each of these data points, they were further analyzed with the Bonferroni Test post hoc to independently assess for differences. The 0.5-L/min group had significantly lower pMAPs than the 0.3-L/min and 0.7-L/min groups. No significant difference was present between 0.3 L/min and 0.7 L/min groups. The 0.7-L/min group had significantly higher dMAP than the 0.3 L/min and 0.5 L/min groups. Distal aortic flow was significantly different between all groups at 30-minute, 60-minute, 90-minute, and 120-minute time points. There was no significant difference between distal aortic flow at baseline or at the final time point (150 minutes). There was no significant difference observed between the flow goal groups in ROTEM parameters including clotting time, clot formation time, maximum clot formation, alpha angle, and max lysis.





Analysis of serum chemistry parameters, and particularly potassium and calcium levels, were then examined. Average values for animals that survived to study endpoint versus those that died are shown in Table 2C and 2D. Although the development of hyperkalemia was near universal in all groups at the later time points, significantly higher potassium levels were observed in animals that did not survive to completion of the experiment at 60 minutes and 90 minutes (p = 0.04 and 0.014) versus animals that survived. Serum calcium levels were only collected for the final four animals following observations of heart block preceding death in earlier animals in the series. As such, calcium levels were obtained in one animal that did not survive to completion of the experiment and three animals which did. Due to these small numbers, testing for significance was not possible. However, a clear trend was present demonstrating higher serum calcium in those animals which survived. On comparison of the interaction of these two laboratory parameters, the combination of significant hyperkalemia and sub-normal serum calcium was associated with early mortality. No other significant differences were observed in serum chemistry values.





Back to Top | Article Outline


The use of balloon occlusion of the Aorta in trauma has transitioned from an innovative, but impractical technique as reported by Hughes in 1954 to a common adjunct in the care for NCTH in 2019.5,10 This progression from the fringes of medicine to routine use has been driven by advances in available technology, as well as the focused effort in military and civilian trauma communities to develop interventions for NCTH. Underscoring this is the current generation of catheters capable of being placed through 7-Fr sheaths, mitigating the risk of ischemic limb complications.8,9,18 These advances in technology and promising results have motivated REBOA utilization in progressively more far forward environments.24,25 The more forward the use of REBOA is brought, the more the concept itself has become the primary limitation. Namely, complete occlusion of the thoracic aorta results in severe and progressive ischemic injury that currently cannot be sustained for more 30 minutes to 60 minutes. Extrapolating from Aortic cross clamp data and preliminary preclinical REBOA studies, Zone 1 deployment carries significant morbidity at times of 30 minutes and is generally fatal at 60 minutes or longer.8,10,21,26 This limits its use to immediate temporization prior to definitive management and relegates its practical use to the in-hospital setting.

Further expansion in the use of REBOA necessitate a practical means of balancing ischemic burden with hemorrhage control to afford longer periods of survivable hemorrhage control. Evidence of preserved mesenteric perfusion during partial occlusion and decreased serum measures of ischemia along with more physiologic cerebral perfusion suggest pREBOA may offer this needed balance.11–13 However, the small existing preclinical literature on true pREBOA cannot be extrapolated to the clinical setting as it has primarily utilized highly complex computer-controlled variable aortic cross-clamp and extracorporeal recirculation devices.13–15 These are not available or practical for clinical use in any current setting, and particularly for deployment in the prehospital or early in-hospital phases of care for patients with NCTH. Clinical calls for pREBOA further highlight the promise of this interesting approach, but also reiterate the need for catheter technology capable of truly titratable partial occlusion.16,18

In this study we first demonstrated the titrability of this catheter. Figures 2A-C illustrate that this catheter avoids the abrupt step off seen with current unilobed compliant balloons. This catheter achieves this titrable partial occlusion by placing a compliant balloon in parallel with a cylindrical noncompliant balloon. The noncompliant balloon quickly fills and prevents early total occlusion of the aortic lumen by providing two channels for blood flow around the balloons. As the compliant balloon reaches capacity it is able to compress the noncompliant balloon and achieve eventual complete occlusion of the aortic lumen. While the smooth titration of blood flow is evident in Figure 2A, aortic blood flow is an impractical measure in clinical use. Figures 4A and B demonstrate that in these inflation cycles, flow had a very linear relationship to both pMAP and dMAP. This is not only crucial in making this a practical technology, but suggests use in conjunction with miniaturized, portable manometers such as the Compass device (Centurion Medical Products, Williamston, MI) as a portable platform for use in austere and far forward environments.

Figure 4

Figure 4

The next major goal of this study was establishing meaningful endpoints of partial occlusion. Prior large animal studies have chosen partial occlusion goals of fluoroscopically visualized partial occlusion, pMAP to dMAP gradients, or fixed pMAPs. These endpoints are divorced from the clinical outcomes pREBOA seeks to balance, namely hemorrhage and ischemia. Experiments 2 and 3 sought to ground our occlusion goals in clinically meaningful endpoints. We utilized models for penetrating intra-abdominal vascular injury and devastating solid organ injury as archetypal patterns of NCTH. By allowing gradually resumption of distal flow we were able to assess tolerance of these injuries to reperfusion. There is a lower tolerance for distal flow with solid organ injury than vascular injury. This may be due the relative poorer ability of the solid organ injury to achieve vasospasm. Postmortem examination of the injuries demonstrated complete arterial thrombosis in many cases of vascular injury, but the corresponding venous injuries rarely exhibited hemostasis. Given that observation, it is unsurprising that the two animals that underwent solid organ injury in whom hemostasis was never gained were observed to have a relatively larger venous contribution on the cut surface of the liver.

With a rough guide to what distal flow would be tolerated without substantial bleeding, we then gauged where in this range the ischemic burden would be tolerable for sustained partial occlusion. Perhaps the most relevant outcome, survival, proved somewhat surprising. Survival was best at the moderate distal flow of 0.5 L/min. Both the higher flow of 0.7 L/min and the lower flow of 0.3 L/min where more poorly tolerated. Examination of the hemodynamic parameters and laboratory values is limited by the low number of animals in each group. As previously stated, there were five animals in the vascular group and five animals in the solid organ injury group. This was underpowered to demonstrate differences between these groups; however, there were clear trends towards progressively higher lactate levels and lower pH in the more restrictive distal flow groups. In regard to hemodynamic parameters, there were some statistically significant but clinically questionable differences. These occurred at single timepoints and did not seem to constitute definite trends or strong associations with a particular set level of partial aortic flow. We believe that this demonstrates the complex and irregular relationship between central hemodynamics, blood loss and bleeding rate, and the ability for physiologic compensation and auto-regulation of major tissue beds. However, we clearly identified that extremely low flow rates were associated with increased mortality that appeared to be due to metabolic abnormalities associated with ischemia-reperfusion, while higher flow rates were associated with earlier deaths due to continued or recurrent bleeding.

While we elected to target distal flow in regards to titrating partial occlusion, the existing literature regarding partial aortic occlusion varies widely not only regarding end points of partial occlusion, but also means of achieving partial occlusion. Russo et al.13 utilized a noncompliant balloon under computerized control to achieve 60% to 70% proximal to distal pressure gradient in a solid organ injury model. It is unclear how this pressure gradient goal was selected. pREBOA administered in this way was found to be inferior to complete REBOA with a trend toward higher hemorrhage rates in the pREBOA group. Comparing this to our data, it would suggest that this was a much more permissive distal perfusion than any of our groups as the most liberal distal flow, 0.7 L/min, allowed a distal pressure 29% of the proximal pressure. Sadeghi et al.11 more closely approximated our distal flow rates of 0.25 L/min to 0.7 L/min and distal pressures of 20 mm Hg, but achieved this through variably inflating a single noncompliant balloon to a goal proximal pressure of 80 mm Hg to 100 mm Hg. This publication is an excellent biochemical illustration of the potential benefits of partial aortic occlusion over complete occlusion, but did not test hemorrhage control capabilities. The lack of injury in this work is likely why they were able to persist in titrating partial flow with a compliant balloon.

A unique take on partial aortic occlusion, or in truth, partial distal perfusion, is endovascular variable aortic control.15 In this publication the authors describe utilizing an automated, computer-controlled extracorporeal circuit in conjunction with complete aortic occlusion. This circuit draws blood from a carotid cannula and gives retrograde distal perfusion through a femoral cannula. Flow is varied through the circuit between 100 mL/min and 300 mL/min based on pMAP. This flow range was chosen based on ability to achieve physiologic pMAPs following solid organ injury in prior testing. It is worth noting these flows are similar to the most restrictive flow tested in our experiments. While the endovascular variable aortic control system is impractical due to its incredible complexity and introduction of the additional stress of extracorporeal circulation on the organism, the concept of computerized autotitration based on proximal and distal pressure sensing is intriguing. Combining auto-titrating technology with a true pREBOA catheter, such as the one tested here, could prove a fruitful avenue for further development. Such a system could achieve complex distal perfusion schedules while allowing the care provider to focus on other lifesaving maneuvers and resuscitation.

One of the primary purposes of this study was to closely examine and characterize the control and ability to “fine tune” the levels of restoration of distal aortic flow after complete occlusion using a prototype device specifically designed for this task. We believe that this is one of the more promising future directions of the many ongoing efforts to mitigate the highly limiting ischemia-reperfusion injury that is a natural consequence of the use of REBOA. Although we have noted frequent clinical references to the use of techniques for “partial” REBOA that are characterized by the user performing slow gradual balloon deflation, there has been no confirmatory data that this approach results in the desired effect of restoring only partial flow. There is also no data that the degree of partial flow that is established has any reliable relationship to the balloon volume, or that this is reliably titratable. This was very nicely demonstrated in a study by Davidson and colleagues utilizing a porcine model of large volume hemorrhage.27 Using a standard commercially available REBOA catheter and small deflation increments of 0.5 mL, they demonstrated highly variable return of flow and an exceedingly steep return of flow with a single 0.5-mL deflation. They also demonstrated that there was no statistical correlation between balloon volume and flow changes or time to restoration of flow. This study confirmed that the current single-balloon catheters should be thought of as “all or none” devices, and that true pREBOA would require alternative catheter and balloon designs.

The prototype pREBOA catheter utilized for this study employs a novel multi-balloon design that was specifically developed to provide more control over the return of partial aortic flow across a wider range of deflation volumes. Although the concept was confirmed to be effective in mechanical trials using a static tube and column of fluid, the ultimate utility in a dynamic system including an elastic aorta and vasculature with inherent contractile and dilatory capacities was needed. This is the first study to our knowledge to examine this new device in vivo and utilizing both uninjured animals as well as animals with shock due to NCTH. Our findings have validated the efficacy of the pREBOA balloon design in allowing for a high degree of fine control over the level of restoration of distal aortic flow ranging from no flow to full flow. We also have provided interesting although preliminary data regarding the efficacy of various levels of partial aortic flow and identified what appears to be an optimal flow level associated with significantly improved survival. We believe that this is the critical question involved in moving true pREBOA from the laboratory to the bedside, and requires identifying the level of flow that strikes the optimal balance between minimizing ischemia-reperfusion injury while simultaneously maximizing hemorrhage control and hemodynamic support.

Another observation from this data was the development of notable metabolic and electrolyte abnormalities associated with aortic occlusion and then restoration of partial flow, and their association with survival or early death. In many of these cases, particularly in the animals that were managed with lower levels of partial aortic reperfusion, we noted that death was immediately preceded by severe cardiac irritability and arrhythmias. Reassessing the hemodynamic parameters and laboratory values by survival demonstrated higher potassium levels in animals that did not survive, routinely well above 8 mg/dL. However, the development of hyperkalemia was seen in all animals that survived longer than 60 minutes, including several survivors with potassium levels above 8 mg/dL. What was also noted was that those animals that tolerated these levels of hyperkalemia had normal or near-normal serum calcium levels, while those that died had relative hypocalcemia. This offers a possible explanation for the observed arrhythmias in animals that did not survive, and points to potential interventions that could enhance survival when restoring either partial or full flow with REBOA. Although there has been much focus on identifying and treating hyperkalemia associated with ischemia-reperfusion injury, we suggest that calcium administration may be as or more important. In patients with hemorrhagic shock, the issue and impact of hypocalcemia may be even more critical as it can be exacerbated by the administration of stored blood products.28,29 Recent guidelines from the Joint Trauma System have mandated early and aggressive calcium supplementation for massive transfusion in the battlefield setting, and this is now tracked as a quality measure.30 We believe that the administration of both calcium and sodium bicarbonate should be utilized prior to reestablishing flow during REBOA utilization, and that this should be considered for inclusion in future guidelines for battlefield and civilian REBOA use.31

Future directions will focus on further validating this pREBOA catheter as capable of balancing ischemia and rebleeding in our validated injury models. We will also pursue validating the use of miniaturized pressure monitoring devices for titration of pREBOA as this would constitute a highly mobile platform that would be advantageous for military applications.

Back to Top | Article Outline


The identified prototype bilobed pREBOA catheter has been demonstrated to be fully capable of smooth, titrable partial aortic occlusion. These findings suggest a distal flow of up to 0.83 L/min can be tolerated before rebleeding occurs from archetypal pelvic vascular injury and solid organ injury. Solid organ injury was found to tolerate relatively less distal aortic flow prior to rebleeding. Prolonged partial occlusion at distal aortic flow less than flows associated with rebleeding was most survivable at a moderate distal flow of 0.5 L/min. These findings suggest that prolonged temporization of NCTH can be effected with purpose built pREBOA catheters capable of smooth titration of flow if a moderate distal flow is targeted. Further refinement of optimal partial occlusion goals and evaluation of miniaturized manometry offers a path toward mobile, deployable technology capable of prolonged prehospital care for NCTH.

Back to Top | Article Outline


All authors conducted and contributed to the literature search. All authors contributed to study design. D.F., W.D., J.W., and R.S. collected the data. D.F., J.K., M.E., and M.M. interpreted the data. D.F., M.E., and M.M. wrote the article. All authors critically revised the final article.

Back to Top | Article Outline


This work was supported by a Department of Defense Medical Research and Development Program (DMRDP) research grant under the DHP 6.7 program: Proposal D6.7 16 C2 I 16 J9 1490, “Intermittent Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) for Severe Noncompressible Truncal Hemorrhage.”

Back to Top | Article Outline


The authors declare no funding or any conflict of interests. Prytime Medical supplied the prototype pREBOA catheters, but provided no financial assistance. The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Army, the Department of Defense, or the U.S. Government.

Back to Top | Article Outline


1. 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. 2013;74(3):830–834.
2. Eastridge BJ, Hardin M, Cantrell J, et al. Died of wounds on the battlefield: causation and implications for improving combat casualty care. J Trauma. 2011;71(Suppl 1):S4–S8.
3. Kisat M, Morrison JJ, Hashmi ZG, Efron DT, Rasmussen TE, Haider AH. Epidemiology and outcomes of non-compressible torso hemorrhage. J Surg Res. 2013;184(1):414–421.
4. Alarhayem AQ, Myers JG, Dent D, et al. Time is the enemy: mortality in trauma patients with hemorrhage from torso injury occurs long before the "golden hour". Am J Surg. 2016;212(6):1101–1105.
5. Hughes CW. Use of an intra-aortic balloon catheter tamponade for controlling intra-abdominal hemorrhage in man. Surgery. 1954;36(1):65–68.
6. Fogarty TJ, Cranley JJ. Catheter technic for arterial embolectomy. Ann Surg. 1965;161:325–330.
7. Ledgerwood AM, Kazmers M, Lucas CE. The role of thoracic aortic occlusion for massive hemoperitoneum. J Trauma. 1976;16(8):610–615.
8. Davidson AJ, Russo RM, Reva VA, et al. The pitfalls of resuscitative endovascular balloon occlusion of the aorta: risk factors and mitigation strategies. J Trauma Acute Care Surg. 2018;84(1):192–202.
9. Taylor JR 3rd, Harvin JA, Martin C, Holcomb JB, Moore LJ. Vascular complications from resuscitative endovascular balloon occlusion of the aorta: life over limb? J Trauma Acute Care Surg. 2017;83(1 Suppl 1):S120–S123.
10. Brenner M, Bulger EM, Perina DG, et al. Joint statement from the American College of Surgeons Committee on Trauma (ACS COT) and the American College of Emergency Physicians (ACEP) regarding the clinical use of resuscitative endovascular balloon occlusion of the aorta (REBOA). Trauma Surg Acute Care Open. 2018;3(1):e000154.
11. Sadeghi M, Horer TM, Forsman D, Dogan EM, Jansson K, Kindler C, Skoog P, Nilsson KF. Blood pressure targeting by partial REBOA is possible in severe hemorrhagic shock in pigs and produces less circulatory, metabolic and inflammatory sequelae than total REBOA. Injury. 2018;49(12):2132–2141.
12. Hoehn MR, Teeter WA, Morrison JJ, Gamble WB, Hu P, Stein DM, Brenner ML, Scalea TM. Aortic branch vessel flow during resuscitative endovascular balloon occlusion of the aorta. J Trauma Acute Care Surg. 2019;86(1):79–85.
13. Russo RM, Williams TK, Grayson JK, Lamb CM, Cannon JW, Clement NF, Galante JM, Neff LP. Extending the golden hour: partial resuscitative endovascular balloon occlusion of the aorta in a highly lethal swine liver injury model. J Trauma Acute Care Surg. 2016;80(3):372–378; discussion 8-80.
14. Johnson MA, Davidson AJ, Russo RM, Ferencz SE, Gotlib O, Rasmussen TE, Neff LP, Williams TK. Small changes, big effects: the hemodynamics of partial and complete aortic occlusion to inform next generation resuscitation techniques and technologies. J Trauma Acute Care Surg. 2017;82(6):1106–1111.
15. Williams TK, Neff LP, Johnson MA, Ferencz SA, Davidson AJ, Russo RM, Rasmussen TE. Extending resuscitative endovascular balloon occlusion of the aorta: endovascular variable aortic control in a lethal model of hemorrhagic shock. J Trauma Acute Care Surg. 2016;81(2):294–301.
16. DuBose JJ. How I do it: partial resuscitative endovascular balloon occlusion of the aorta (P-REBOA). J Trauma Acute Care Surg. 2017;83(1):197–199.
17. Johnson MA, Neff LP, Williams TK, DuBose JJEVAC Study Group. Partial resuscitative balloon occlusion of the aorta (P-REBOA): clinical technique and rationale. J Trauma Acute Care Surg. 2016;81(5 Suppl 2 Proceedings of the 2015 Military Health System Research Symposium):S133–S137.
18. Matsumura Y, Matsumoto J, Kondo H, et al. Fewer REBOA complications with smaller devices and partial occlusion: evidence from a multicentre registry in Japan. Emerg Med J. 2017;34(12):793–799.
19. Kuckelman JP, Barron M, Moe D, Derickson M, Phillips C, Kononchik J, Lallemand M, Marko S, Eckert M, Martin MJ. Extending the golden hour for zone 1 resuscitative endovascular balloon occlusion of the aorta: improved survival and reperfusion injury with intermittent versus continuous resuscitative endovascular balloon occlusion of the aorta of the aorta in a porcine severe truncal hemorrhage model. J Trauma Acute Care Surg. 2018;85(2):318–326.
20. Lallemand MS, Moe DM, McClellan JM, Smith JP, Daab L, Marko S, Tran N, Starnes B, Martin MJ. Resuscitative endovascular balloon occlusion of the aorta for major abdominal venous injury in a porcine hemorrhagic shock model. J Trauma Acute Care Surg. 2017;83(2):230–236.
21. Hatch Q, Debarros M, Eckert M, Satterly S, Nelson D, Porta R, Lesperance R, Long W, Martin M. Acute coagulopathy in a porcine venous hemorrhage and ischemia reperfusion model. Am J Surg. 2014;207(5):637–641; discussion 641.
22. Sheppard FR, Mitchell TA, Macko AR, Fryer DM, Schaub LJ, Ozuna KM, Glaser JJ. Whole blood and Hextend: bookends of modern tactical combat casualty care field resuscitation and starting point for multifunctional resuscitation fluid development. J Trauma Acute Care Surg. 2018;85(1S Suppl 2):S33–S38.
23. Butler FK Jr. Fluid resuscitation in tactical combat casualty care: yesterday and today. Wilderness Environ Med. 2017;28(2S):S74–S81.
24. Rees P, Waller B, Buckley AM, Doran C, Bland S, Scott T, Matthews J. REBOA at role 2 afloat: resuscitative endovascular balloon occlusion of the aorta as a bridge to damage control surgery in the military maritime setting. J R Army Med Corps. 2018;164(2):72–76.
25. Teeter W, Romagnoli A, Glaser J, Fisher AD, Pasley J, Scheele B, Hoehn M, Brenner M. Resuscitative endovascular balloon occlusion of the aorta: pushing care forward. J Spec Oper Med. 2017;17(1):17–21.
26. DuBose JJ, Scalea TM, Brenner M, et al. The AAST prospective aortic Occlusion for Resuscitation in Trauma and Acute Care Surgery (AORTA) registry: data on contemporary utilization and outcomes of aortic occlusion and resuscitative balloon occlusion of the aorta (REBOA). J Trauma Acute Care Surg. 2016;81(3):409–419.
27. Davidson AJ, Russo RM, Ferencz SE, Cannon JW, Rasmussen TE, Neff LP, Johnson MA, Williams TK. 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. J Trauma Acute Care Surg. 2017;83(1):139–143.
28. Lim F, Chen LL, Borski D. Managing hypocalcemia in massive blood transfusion. Nursing. 2017;47(5):26–32.
29. Giancarelli A, Birrer KL, Alban RF, Hobbs BP, Liu-DeRyke X. Hypocalcemia in trauma patients receiving massive transfusion. J Surg Res. 2016;202(1):182–187.
30. Cap AP, Pidcoke HF, Spinella P, et al. Damage control resuscitation. Mil Med. 2018;183(Suppl 2):36–43.
31. Cannon J, Morrison J, Lauer C, Grabo D, Polk T, Blackbourne L, Dubose J, Rasmussen T. Resuscitative endovascular balloon occlusion of the aorta (REBOA) for hemorrhagic shock. Mil Med. 2018;183(Suppl 2):55–59.
Back to Top | Article Outline


Many may not read this paper because it’s about REBOA. Some recent studies have concluded that REBOA has “no advantages in current practice.” I must challenge this. REBOA is a less invasive means of aortic occlusion compared to the relatively “harsh” method of ED thoracotomy and its associated chest wall trauma.

Even if you are not yet a parishioner in the church of clinical REBOA use you must acknowledge the importance of REBOA as a research tool. This is a physiology paper as much as it is a REBOA paper. As the authors report they “fine tune” distal aortic flow to identify the pressure at which bleeding recurs, identify the impact of ischemia on critical electrolytes, and describe the dynamic vasculature changes in shock-induced endovascular environment.

Understanding the physiological response to injury has even more significance when taken from the authors’ perspective of Combat Casualty Care and the resource limited forward combat environment.

Having the ability to modulate the postinjury hypoperfusion states and study the impact of differing pressures and flow rates on the molecular response is invaluable to our profession. Like any good research, more questions are generated than answer. Is a normal SBP, or some other value, the appropriate resuscitation goal? What is the impact of differing degrees of ischemia reperfusion on the metabolic and molecular response to injury? This manuscript does advance the concept of partial REBOA and, perhaps more importantly, furthers our understanding of the physiological response to injury.

Joseph Galante, MD FACS

Sacramento, CA


Resuscitative endovascular balloon occlusion of the aorta; partial REBOA; noncompressible truncal hemorrhage; NCTH; hemorrhage control

© 2019 Lippincott Williams & Wilkins, Inc.