Reference: Jansen et al. Emergency Department Resuscitative Endovascular Balloon Occlusion of the Aorta in Trauma Patients With Exsanguinating Hemorrhage: The UK-REBOA Randomized Clinical Trial. JAMA. 2023

Date: June 10, 2024

Guest Skeptic: Dr. Rob Leeper is an intensivist, trauma surgeon, and general surgeon at Western University where he also serves as the director of the Royal College Surgical Foundations program. He is particularly enthusiastic about medical education and the use of high-fidelity simulation to identify latent threats to patient safety and improve team-based crisis resource management.

Case: A 24-year-old patient is involved in a high-speed motor vehicle collision.  The patient was the intoxicated driver of a sport utility vehicle that left the road and collided head-on with a traffic light.  The driver was ejected from the vehicle.  On scene, he has a Glasgow Coma Scale (GCS) score of 14, is tachycardic at 130 beats per minute and hypotensive at 85/50 mmHg.  He complains of severe pain in his abdomen and left upper extremity.  He has an obvious angulated deformity of his left wrist.

Pre-hospital personnel placed the patient in a cervical spine collar, obtained peripheral intravenous (IV) access, administered 250cc of 0.9% saline and provided rapid transport to the local trauma center which is less than 15 minutes away.  On arrival at the trauma center, the patient’s vital signs and symptoms remain essentially unchanged from those on scene.

You initiate standard, simultaneous assessment and resuscitation following the Advanced Trauma Life Support (ATLS) principles.  You are debating the addition of resuscitative endovascular balloon occlusion of the aorta (REBOA) as an adjunct to your standard care.  You wonder what the evidence is for this additional approach.

Background: We covered REBOA on the SGEM 5 years ago with Dr. Robert Edmonds (SGEM#258). That was a review of a retrospective observational study of 593,818 trauma patients of which 140 received REBOA compared to 240 matched controls who did not receive this treatment. The overall mortality was worse with REBOA (35.7% vs. 18.9%, p=0.01). The SGEM bottom line in 2019 was that REBOA is currently an intervention of uncertain benefit. Although it has shown promise in some studies, this investigation leaves its therapeutic potential in question and arguably demonstrates harm. There may be substantial benefits in select groups of trauma patients, but these groups are not yet known.

Trauma surgeons are hemorrhage control surgeons.  Bleeding is the enemy and patient survival turns tenuously on our ability to identify and staunch hemorrhage in as expedient a fashion as possible.  Time is our currency and, for patients with massive hemorrhage, minutes matter.  Every strategy that enhances time to hemorrhage control is deserving of examination:

  • Trauma Systems – Enhancing the training and tools for our pre-hospital teams, faster helicopters, more trauma centers, etc.
  • Hemostatic Resuscitation – Permissive hypotension, whole blood and factor ratio resuscitation, warmed products, dynamic coagulation monitoring, etc.
  • Enhanced Operative Strategies – Damage control surgery with truncated interventions, energy devices, hemostatic foams/gels/packs, etc.

These approaches hold the possibility of enhancing our ability to save patient’s lives. One approach, which has seen a revival in its interest and utilization, is proximal balloon occlusion of the aorta.  The rationale is simple, occluding the aorta above the level of bleeding will de facto reduce blood loss at the site of injury and lead to immediately increased blood pressure and, presumably, perfusion to the zone above the occlusion, typically the upper body, the heart, and the brain.  A great variety of medical devices designed for this purpose have come to market over the last 15 years and a relatively heterogeneous collection of retrospective clinical data has been published with conflicting results. There are signals of improvement but also signals of harm with this technique.

Given this equipoise, a robust prospective clinical trial was undertaken by a group of international trauma providers and researchers.  The UK-REBOA was designed and conducted across 16 English trauma centers from 2017 to 2022 to provide the most evidence-based answer to the question of whether the addition of REBOA to standard care improves survival in massively bleeding trauma patients.

Clinical Question: Does the addition of REBOA to standard trauma resuscitation practice improve survival for injured patients presenting to lead trauma centres with massive hemorrhage?

Reference: Jansen et al. Emergency Department Resuscitative Endovascular Balloon Occlusion of the Aorta in Trauma Patients With Exsanguinating Hemorrhage: The UK-REBOA Randomized Clinical Trial. JAMA. 2023

  • Population: Adults aged 16 years or older presenting to 16 major trauma centers in the UK (2017-2022) with confirmed or suspected life-threatening torso hemorrhage amenable to adjunctive treatment with REBOA.
    • Exclusions: Key exclusion criteria were known or suspected pregnancy or injuries that were clearly not survivable.
  • Intervention: REBOA in addition to standard care. This entailed the percutaneous insertion of a balloon catheter into the aorta, which was then inflated to control hemorrhage while maintaining cerebral and myocardial perfusion.
  • Comparison: Standard care alone, which could include intubation, balanced blood product transfusion, tourniquet application, and early operative or endovascular hemorrhage control.
  • Outcomes:
    • Primary Outcome: All-cause mortality at 90 days.
    • Secondary Outcomes: Mortality at various timeframes, the need for definitive hemorrhage control procedures, time to commencement of such procedures, complications, length of stay, blood product use, and cause of death.
  • Trial: This was a pragmatic, Bayesian, multi-centred randomized clinical trial

Authors’ Conclusions: In trauma patients with exsanguinating hemorrhage, a strategy of REBOA and standard care in the emergency department does not reduce, and may increase, mortality compared with standard care alone.”

Quality Checklist for Randomized Clinical Trials:

  1. The study population included or focused on those in the emergency department. Yes
  2. The patients were adequately randomized. Yes
  3. The randomization process was concealed. Yes
  4. The patients were analyzed in the groups to which they were randomized. Yes
  5. The study patients were recruited consecutively (i.e. no selection bias). No
  6. The patients in both groups were similar with respect to prognostic factors. No
  7. All participants (patients, clinicians, and outcome assessors) were unaware of group allocation. No
  8. All groups were treated equally except for the intervention. No
  9. Follow-up was complete (i.e. at least 80% for both groups). Yes
  10. All patient-important outcomes were considered. Yes
  11. The treatment effect was large enough and precise enough to be clinically significant. Yes
  12. Financial conflicts of interest. Several of the biggest names in REBOA technology are investors and shareholders in companies that sell these balloons.  Jansen’s disclosure slide had plenty of industry sponsorships and grants.

Results: The study population (n=90) predominantly comprised males (69%) with a median age of 41 years. Most injuries (97%) were due to blunt trauma, and the median Injury Severity Score was 41, indicating severe injuries.

Key Result: Mortality was higher in those trauma patients allocated to receive REBOA.

  • Primary Outcome: Mortality at 90 days
    • 54% REBOA care group vs 42% in the standard care alone group
    • Odds Ratio (OR)1.58 [95% Credible Interval, 0.72 to 3.52]
    • The posterior probability of an OR >1 (indicating increased odds of death with REBOA) was 86.9%. This means the probability of increased odds of death with REBOA was 86.9%.

  • Secondary Outcomes: All secondary outcomes numerically favoured standard care without REBOA. They are listed in Table 2 of the manuscript

1. Small Sample Size: 90 patients represent a small sample size. It demonstrates how difficult this type of research can be. We should congratulate the authors and all those involved in conducting such a high-quality trial. They asked a good question, used appropriate methods to try to answer the question and in 16 trauma centres over five years found less than 50 patients to allocate to each group. Remember the earlier observational study we mentioned in the background material, those authors had to search ~600,000 trauma patients to identify 140 patients who received REBOA. This treatment is not common and is incredibly hard to conduct an RCT investigating its impact. Again, well done Dr. Jensen and their team.

2. What can we learn from the mortality curve (Kaplan-Meier plot)?

The key thing that jumps out at you visually from the Kaplan-Meier plot is that the separation in the two survival curves happens early with about half of the difference jumping out in the first day and nearly all of it being complete by ten days. After that the curves are very much flat and follow each other out to 90 days in parallel.  To me, this looks like a signal for hemorrhagic death (the first half) and complications of delay in hemorrhage control and protracted resuscitation (in the second half).

3. Can we get any more granular than that on mortality at different time points?

Showing a breakdown of the risks of death at 3 hours, 6 hours, 24 hours, and in hospital you can see that there is an increased hazard of death at every timepoint with REBOA.  Odds Ratios (OR) are higher but the posterior probability of the ORs being > 1 are between 99% and 87% at each time point, once again suggesting that REBOA is associated with more harm in the most critical period of hemorrhage care. the earliest phase.

While this is interesting, we do need to be careful not to over-interpret this data. The trial was not powered for these subgroups and subgroup analyses should be considered hypothesis-generating. We are also dealing with small numbers that make the point estimates less certain. This means there is less confidence that if the trial was repeated the reported effect size would be the same.

4. Is the delay and/or omission of hemorrhage control surgery a possible reason why REBOA appears to increase harm?

There is no such thing as a free lunch in hemorrhage control.  If you are spending time on one thing, you are stealing time from something else.  The UK-REBOA trial makes that point abundantly clear in a very uncomfortable way for trauma providers.

Of the 46 patients in the REBOA plus standard care arm, only a meagre 19 had the balloon inflated.  Why?  Well, if you look at the figure below you can see why:

  • 8 patients had abject failure of arterial access
  • 14 patients had access and/or a catheter inserted but it was never blown up because it took so long to get it in that by that time. They had gotten a bit better
  • 3 got better while the team was still thinking over the arterial access
  • 2 got worse or died before they could even try

Getting a REBOA catheter in on a massively hemorrhaging patient is not easy and takes a long time if it ever does go on.  While we can’t tell exactly how long each procedure took or exactly what that delay meant in terms of how much it was distracting from other resuscitative priorities or preventing transition to other hemorrhage control options, we can extrapolate.

When we look at patients in the REBOA plus Standard care arm, there was a significant delay in the time to get to a hemorrhage control surgery/procedure (83 vs 64 minutes ~ 20 min Absolute Minute Addition) and a reduction in the overall number of hemorrhage control procedures being done, likely due to the competing risk of DEATH.

If you’re thinking that this doesn’t apply to you because you’re really good at doing stuff, the study personnel showed us picture after picture of all of the training courses done to upskill personnel and build competency at the participating hospitals.  I’m sure some people out there are really good at vascular intervention, but the truth is that most people aren’t and this study highlights that unfortunate truth.

The other thing to consider here is that much like the Canadian setting, outside of Winnipeg’s Gold Service, all Canadian trauma surgeons take calls from home as do all trauma surgeons in England.  Even if REBOA was inserted quickly, it may well have increased blood pressure and created a false sense of security that slowed the roll of the hemorrhage control machine long enough to inflict more harm.

Ultimately, the signal here is that time is money, and time spent preferentially on REBOA may cost the patient dearly.

5. REBOA on the first attempt: We talked about subgroups and about difficulties in getting REBOA. What about those patients for whom REBOA was achieved quickly on the first attempt, did they do well?

No.  Even when the subgroup of patients who experienced no technical failure in REBOA insertion was analyzed, harm was still implied.  In economics, the complier average causal effect (CACE), is the effect of a treatment for subjects who comply with the experimental treatment assigned to their sample group.  A CACE analysis therefore compares outcomes for individuals in the intervention condition who complied with treatment with individuals in the control group who would have complied with treatment given the opportunity to do so.  When subjecting the data to a CACE analysis, the OR of death at 90 days (primary outcome) with REBOA plus standard care was still 2.07 times that of standard care alone with a posterior probability of OR >1 of 88%.  A reminder that the basic ITT analysis had an OR of 1.58 and a PP of 86%.  If you are a smooth-brained trauma surgeon like me, you can squint your eyes and almost imagine that the subjects who complied with the treatment (had REBOA placed successfully) did worse than the ones for whom REBOA insertion failed.

Comment on Authors’ Conclusion Compared to SGEM Conclusion: We agree with the authors’ conclusions.

SGEM Bottom Line: REBOA cannot be recommended as a front-line treatment for massively bleeding trauma patients. Rapid access to a hemorrhage control operation is probably the better option and if REBOA is considered, it must not delay operative hemorrhage control.

Case Resolution: This 24-year-old patient received standard ATLS resuscitation which, interestingly, included the US-guided placement of a right common femoral arterial line by the trauma surgeon which was placed simultaneously with a chest tube and central line insertion.  The patient went directly to the operating theatre and had a successful trauma laparotomy, preperitoneal packing, external pelvic fixation and closed reduction of a distal radius fracture.  The patient did well, and the right femoral line was used only as a reliable means of monitoring blood pressure.  However, if the patient had an arrest or peri-arrest on induction, the trauma surgeon would have been happy to be able to weigh the benefits and risks of rapid REBOA insertion via the femoral sheath vs thoracotomy and aortic cross-clamping.

Dr. Rob Leeper

Clinical Application: Am I now much less likely to use REBOA in acutely injured patients.  It has taken the shine off the apple for me in this population and only redoubled my resolve to let nothing stand between me and the operating room for hemorrhagic shock.  We are simply cruising the internet all day and looking to confirm our own biases, and the signal from this trial makes me more keen on things I already believed in like rapid field triage and transport to the hospital rather than stay-and-resuscitate strategies, it makes me more keen to open more operating rooms and close our trauma surgical service to make hemorrhage control procedures more rapidly accessible to injured patients.

Would I never use a REBOA in a trauma patient?  I must admit that I know I still will if my hand is forced.  In my hands, where on one hand I can weigh the availability and rapidity of a hemorrhage control thoracotomy or laparotomy and on the other hand weigh the value of endovascular aortic occlusion, I think I have the best position to decide how to stage and deploy each technique.  I am not biased because I can do either one and enjoy each one quite a lot, so the other home point for me is that if REBOA is going to be used in trauma it should almost certainly be in the hands of vascular and trauma surgeons such that they can equitably weigh the benefits and harms of the delay in their own hemorrhage control interventions.

Finally, we do not have data from this trial about those brave and somewhat incredible HEMS air ambulance teams in London who want to do roadside thoracotomies and REBOAs because they are never more than 10 min by chopper to a lead trauma hospital.  I’ve mentioned that I think anything which delays rapid transit to an operating room is bad news, but this study can’t tell us to take the REBOAs away from those helicopter EMS people yet.  Also, in the setting of non-trauma patients like post-partum hemorrhage and ruptured aneurysms, I think we have no further data from this study.  One of the most dramatic saves of my life happened just a few months ago and it relied heavily on REBOA utilization for postpartum hemorrhage.  There is no question that by the time we had to use REBOA, it was the only way to get that patient back from arrest and so we certainly don’t want to throw out these catheters with the bathwater yet.

What Do I Tell the Patient? Nothing! I would just tell the operating theatre we are coming up, and this paper makes me even more willing to say: “Don’t stop me now”!  We need to operate now to control the hemorrhaging.

Keener Kontest: Last week’s winner was Steven Stelts in NZ. He knew the Germans tried to transplant a uterus into a trans woman back in 1931 to achieve pregnancy. She died of a postoperative infection 3 months later.

Listen to the podcast to hear this week’s keener contest question. Email your answer to with “Keener” in the subject line. The first correct answer will win a cool skeptical prize.

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics Guide to Emergency Medicine.