Prehospital Partial Resuscitative Endovascular Balloon Occlusion of the Aorta for Exsanguinating Subdiaphragmatic Hemorrhage

Key Points Question Is it feasible to deploy prehospital zone 1 (supraceliac) partial resuscitative endovascular balloon occlusion of the aorta (Z1 P-REBOA) in the prehospital resuscitation of adult trauma patients at risk of cardiac arrest and death due to exsanguination? Findings In this cohort study of 16 patients with severe injuries and shock, prehospital Z1 P-REBOA was delivered successfully in 8 of the 11 patients who underwent Z1 REBOA. This strategy was associated with improved proximal blood pressure and early mortality. Meaning The findings indicate that in the prehospital resuscitation of adult trauma patients at risk of cardiac arrest and death due to exsanguination, Z1 P-REBOA is feasible and may enable early survival, but with a significant incidence of late death.


U
7][8] For those that do reach the hospital, mortality is highest in the first 3 to 72 hours, attributable to the consequences of early and profound hemorrhagic shock. 8,9Most deaths are due to noncompressible torso hemorrhage, and yet there are no established solutions for temporizing this bleeding in the prehospital or far-forward environment.
Resuscitative endovascular balloon occlusion of the aorta (REBOA) is a technique for the temporary proximal control of subdiaphragmatic torso hemorrhage. 10Balloon occlusion of the descending thoracic aorta (zone 1 [Z1], supraceliac) reduces distal blood loss while also improving coronary and cerebral perfusion. 11,12This intervention can temporarily delay cardiac arrest from exsanguination, potentially extending the time window to provide definitive treatment. 13However, in the UK, the average time taken to reach a hospital following major injury is 90 minutes, long after the window of potential benefit. 3,14n the prehospital setting, where there is potentially the most benefit from earlier, temporary truncal hemorrhage control, we have previously shown that it is possible to deploy REBOA. 15,16However, this was limited to infrarenal placement (Z3) to avoid prolonged visceral ischemia. 15To further investigate this resuscitation strategy in this setting, it is important to establish whether it can be delivered in patients with more proximal hemorrhage in the abdomen (Z1 REBOA) and to achieve maximal increase in proximal blood pressure (BP) and therefore myocardial perfusion in patients with profound shock and traumatic cardiac arrest (TCA) caused by subdiaphragmatic exsanguinating hemorrhage. 10,13,17However, this comes with an increased risk of visceral ischemic injury.
Partial REBOA (P-REBOA) describes allowing titrated lowvolume aortic flow past the balloon.It can provide a balance by creating a region of distal permissive hypotension, thus reducing hemorrhage, while mitigating distal ischemia or reperfusion injury and simultaneously augmenting proximal hemodynamics. 18,19To our knowledge, Z1 P-REBOA has not previously been deployed in the prehospital environment.
The overall aim of this study was to explore the potential use of prehospital Z1 P-REBOA in patients with immediately lifethreatening noncompressible torso hemorrhage.Our first objective was to determine whether it was feasible to deploy prehospital Z1 REBOA in critically bleeding trauma patients.The second objective was to determine the feasibility of delivering controlled P-REBOA in these patients and assess technical performance.Finally, we wished to explore the potential impact of Z1 P-REBOA on prehospital and in-hospital clinical outcomes and safety.

Study Design
We performed a prospective observational cohort study of prehospital Z1 P-REBOA delivered by a physician-led prehospital team in an advanced urban trauma system including injured patients with immediately life-threatening subdiaphragmatic hemorrhage.The study is a sequentially reported case series, adhering to the Idea, Development, Exploration, Assessment and Long-term follow-up (IDEAL) 2A framework for surgical innovation. 20Patients were followed up until discharge, death, or for 90 days from the date of the final patient's admission to hospital.Due to the emergency context, enrollment proceeded without consent.Consent for continued participation was sought once patients were no longer in a life-threatening condition.Ethics approval was granted by the London-Southeast Research Ethics Committee and the Health Research Authority and sponsored by Barts Heath National Health Service Trust.All cases were reviewed by a multidisciplinary investigative committee, which provided oversight and guidance for the study management group.

Setting
London's Air Ambulance provides a 24-hour dedicated advanced trauma service to Greater London, working in conjunction with the London Ambulance Service.The physicianparamedic team attends approximately 2000 injured patients per year and can deliver advanced prehospital interventions, including prehospital anesthesia, blood transfusion, resuscitative thoracotomy, and REBOA (Z3, since 2014). 15,21Severely injured patients are transported to 1 of London's 4 major trauma centers.

Inclusion and Exclusion Criteria
Eligible participants for enrollment were trauma patients aged 16 years and older and attended by London's Air Ambulance with a clinical diagnosis of exsanguinating subdiaphragmatic hemorrhage, with recent or imminent risk of hypovolemic cardiac arrest thought to be amenable to treatment with Z1 REBOA.Exclusion criteria were patients suspected to be younger than 16 years, unsurvivable injuries, or pregnancy.Patients entered the study at the point an attempt was made at common femoral artery access for Z1 REBOA.

Prehospital REBOA
Z1 REBOA was performed as an adjunct to standard trauma care via an 8F access sheath (Avanti; Cordis) inserted percutaneously into the common femoral artery under

Key Points
Question Is it feasible to deploy prehospital zone 1 (supraceliac) partial resuscitative endovascular balloon occlusion of the aorta (Z1 P-REBOA) in the prehospital resuscitation of adult trauma patients at risk of cardiac arrest and death due to exsanguination?Findings In this cohort study of 16 patients with severe injuries and shock, prehospital Z1 P-REBOA was delivered successfully in 8 of the 11 patients who underwent Z1 REBOA.This strategy was associated with improved proximal blood pressure and early mortality.

Meaning
The findings indicate that in the prehospital resuscitation of adult trauma patients at risk of cardiac arrest and death due to exsanguination, Z1 P-REBOA is feasible and may enable early survival, but with a significant incidence of late death.ultrasound guidance and using the ER-REBOA catheter (Prytime Medical Devices).Arterial BP was measured distal to the balloon (common femoral artery, sidearm access sheath) and proximal to the balloon (thoracic aorta, catheter tip).Catheter insertion depth was determined by measuring external landmarks (cannulation site to mid sternum) targeting approximately 45 cm. 22,23The balloon was inflated using 0.9% saline until the distal pulsatile pressure trace (measured from the access sheath sidearm) became nonpulsatile and proximal pressure improved. 19ressure measured distal to the site of balloon occlusion is directly proportional to flow, exhibiting a linear relationship. 18n increase in the distal mean arterial pressure of 5 to 10 mm Hg above the postinflation baseline is associated with an increase in distal flow of 250 to 500 mL/min. 18,19Therefore, when instituting P-REBOA, distal pressure can be targeted as a surrogate for flow.

Outcomes
Successful prehospital Z1 REBOA was defined as the composite end point of balloon insertion to 35 to 55 cm, proximal arterial BP transduced, and balloon inflation.Successful prehospital P-REBOA was defined as the composite end point of evidence of an increase in distal mean arterial pressure of at least 5 to 10 mm Hg above the postinflation baseline and/or a return of distal pulsatility, or distal pulsatility never absent postnitial balloon inflation.
Clinical end points included systolic BP (SPB) response to Z1 REBOA, incidence of prehospital traumatic cardiac arrest, mortality rate (1 hour, 3 hours, 24 hours, or 30 days postinjury), cause of death, and survival to hospital discharge.All adverse events related to patient injury, resulting critical illness, and treatment as well as femoral cannulation, REBOA catheter insertion, and/or the anticipated effects of aortic occlusion were predefined and recorded.

Data Collection
Temporal, physiological, and clinical data were prospectively collected and recorded using REDCap. 24Physiological data were downloaded directly from the patient monitor (ZOLL X-series; Zoll Medical).Other data were obtained from original sources, including standard prehospital and in-hospital patient records, blood tests, imaging, and postmortem data.Injuries were classified according to the Abbreviated Injury Scale 2005 and Injury Severity Score by certified coders. 25,26

Statistical Analysis
Continuous data are reported as medians with IQRs and categorical data as frequency and percentage.Paired data were compared using the Wilcoxon matched-pairs signed rank test (PRISM version 10; GraphPad).Statistical significance was set as a 2-tailed P value less than .05.

Feasibility
Of the 16 patients, the clinical condition of 2 improved and REBOA was not attempted.In the other 14 patients who had an attempt at Z1 REBOA, all demonstrated ongoing hemodynamic deterioration despite blood product transfusion, with 8 patients in TCA by the time of intervention (57%).Three patients (all in TCA) had failed attempts at femoral arterial access (21%) due to an inability to visualize the common femoral artery with ultrasonography.All 3 underwent immediate resuscitative thoracotomy for open aortic control.The 11 remaining patients (5 of 11 in TCA [46%]) had REBOA catheters inserted in Z1.There were initial, technical arterial BPmonitoring issues in 2 of these patients (both in TCA from the point of clinical contact), leaving 9 patients with a measured pre-REBOA median (IQR) SBP of 47 (33-52) mm Hg and DBP of 28 (14-34) mm Hg.

Z1 Balloon Occlusion
All catheters were inserted to 45 cm except 1 (47 cm).Proximal arterial (aortic) pressure was transduced in 10 of the 11 patients undergoing REBOA at the point of balloon inflation and all balloons were inflated.Therefore, the proportion of patients in whom Z1 REBOA was delivered as per definition was 91% (10 of 11).The median (IQR) time from patient injury (emergency call) to balloon inflation in Z1 was 57 (54-67) minutes.All procedures were performed at the scene of the injury except 1, which was performed during ambulance transfer to the hospital.Balloon occlusion was associated with significant improvement in BP; the 10 patients in whom BP was successfully measured, the median (IQR) post-REBOA SBP was 88 (64-107) mm HG and DBP was 51 (36-73) mm Hg (Figure 1).Median (IQR) group-level improvements in SBP and DBP were 40 mm Hg (95% CI, 5-64; P = .008)and 30 mm Hg (95% CI, 13-45; P = .008),respectively (Figures 1 and 2; Table 2; eFigure 1 and eTable 1 in Supplement 1).
Proximal hemodynamic stability was maintained in 7 of the 8 patients during institution and maintenance of P-REBOA.The 3 patients who underwent Z1 REBOA in whom P-REBOA was not possible had associated severe ongoing proximal hypotension and therefore an inability to generate partial flow either spontaneously or by balloon deflation (Figures 1 and 2; Tables 1 and 2; eFigure 1 and eTable 1 in Supplement 1).

Prehospital Course
The median (IQR) time from injury (emergency call) to emergency department arrival was 89 (82-108) minutes.Continued improvement in proximal BP was observed compared with the pre-REBOA values, with a median (IQR) emergency department arrival SBP (in the 10 patients transported to the hospital) of 101 (77-107) mm Hg and DBP of 67 (51-75) mm Hg.Median group level improvements in SBP and DBP from preinflation values were 52 mm Hg (95% CI, 42-77; P = .004)and 42 mm Hg (95% CI, 20-58; P = .004),respectively.Three patients remained hypotensive (SBP <90 mm Hg), and 1 patient died on scene and therefore was not transported.External cardiac massage was performed post-Z1 REBOA in 3 individuals due to established TCA.This led to a sustained return of spontaneous circulation in 1 individual, with the patient surviving to hospital discharge (Figure 2).Four transient adverse events related to pressure transducer setup and operation and 1 to balloon deflation (5 of 11 cases [46%]) were reported (Figures 1 and  2; Table 2; eTables 1 and 2 in Supplement 1).

In-Hospital Management and Outcome
Definitive hemorrhage control was achieved with immediate damage-control surgery or angioembolization in 9 of the 11 patients undergoing Z1 REBOA.Of the remaining 2 patients, 1 underwent emergent orthopedic surgery and neurosurgery and 1 died on scene and therefore did not access in-hospital treatment.Ten of the 11 patients undergoing Z1 REBOA survived for more than 3 hours postinjury (1-and 3-hour mortality rate, 9%).Eight patients survived for 24 hours (24-hour mortality rate, 27%), and 2 patients survived for 30 days and to hospital discharge (30-day mortality rate, 82%; survival to hospital discharge, 18%).Both survivors underwent early Z1 P-REBOA and therefore survival to hospital discharge in the 8-patient Z1 P-REBOA cohort was 25% (2/8) (Tables 2 and 3; eFigure 2 in Supplement 1).

Discussion
This cohort study reports the initial experience of ZI P-REBOA as a prehospital resuscitation strategy for patients with exsanguinating subdiaphragmatic hemorrhage following injury.Prehospital Z1 REBOA was feasible and associated with significant improvements in proximal BP.Z1 P-REBOA as a means of mitigating distal ischemia was implemented successfully in most cases and tolerated with proximal hemodynamic stability.The cohort of patients who met the inclusion criteria were more severely injured (median Injury Severity Score, 50) than in our previously reported observational study 15 of prehospital Z3 REBOA (median Injury Severity Score, 34) and other in-hospital studies of REBOA. 27,28ll patients displayed a positive hemodynamic response to Z1 REBOA.This response was gradual, with 45% remaining hypotensive (SBP <90 mm Hg) in the immediate post-REBOA phase with general recovery in proximal BP in most by emergency department arrival, and is in contrast to the often immediate BP response seen with REBOA in animal models of hypovolemic shock. 1 7, 1 8 This may reflect profound myocardial hypoperfusion in the immediate postinjury phase and the development of concurrent cardiogenic shock.This cohort had DBP levels below that required to allow effective coronary perfusion (median DBP preintervention, 28 mm Hg) and were therefore at high risk of myocardial ischemia, asystole, and TCA. 29Z1 REBOA was associated with an immediate increase in DBP to levels adequate to restore coronary perfusion. 30There were no instances of proximal hypertension in response to Z1 REBOA, which have been reported in large animal models to lead to hyperemic coronary blood flow, high left ventricular wall tension, and myocardial injury. 12P-REBOA was feasible in all patients who exhibited improved proximal BP post-REBOA, and the transition from C-REBOA to P-REBOA happened in less than 10 minutes.P-REBOA was observed to occur spontaneously in half of the patients.Several factors are likely to be important.First, procedurally, the balloon was inflated until the distal pulsatile trace disappeared, avoiding overinflation.Second, inc reased proximal BP (c ardiac recover y, preload restoration, improved stroke volume and cardiac output) may have contributed.Increased aortic capacitance or diameter may have also been a factor.In individuals in whom a reduction of balloon volume was required to initiate P-REBOA, the volume removed was relatively small and again reflects the procedure (avoiding balloon overinflation).P-REBOA was not feasible in 3 individuals due to severe ongoing proximal hypotension, with 1 remaining in TCA postintervention.The prolonged period of complete aortic occlusion in 2 patients occurred due to an inability to generate a proximal pressure sufficient to deliver or establish P-REBOA or enable balloon deflation.Three-fourths of patients survived for 24 hours postinjury despite more than half experiencing prehospital TCA due to exsanguination.However, survival to 30 days was low (2/11 [18%]), with most patients dying with clinical evidence of multiple organ dysfunction syndrome and cardiogenic shock.Early extracorporeal membrane oxygenation for patients with major injury and cardiogenic shock is now increasingly available. 31,32It is possible that earlier diagnosis of cardiogenic shock and early reperfusion with early extracorporeal membrane oxygenation could have led to survival in some cases.
Both survivors underwent early and sustained P-REBOA (occurring spontaneously in 1 patient), and both had total REBOA times significantly longer than 60 minutes (majority P-REBOA).This is in contrast to previous reports suggesting that there are no survivors beyond 60 minutes of C-REBOA and is likely to reflect the impact of P-REBOA mitigating distal ischemia, albeit this is dependent on the recovery of proximal BP. [33][34][35] The UK-REBOA trial represents the first in-hospital randomized clinical trial of REBOA.It reported that the addition of a REBOA strategy to standard care, incurring a 20-minute delay to definitively control hemorrhage, increased 90-day mortality with high probability. 36This finding is important: the time taken to perform REBOA at the point of injury is likely to increase prehospital time and therefore the time to definitively control hemorrhage, with potential negative impacts on survival. 379]14,38 Further investigation into identifying specific pathophysiological signatures indicative of early death, to potentially improve patient selection, and to allow earlier targeted intervention in those most likely to benefit while avoiding delay in those who may not, is warranted.

Strengths and Limitations
Although the study has a small sample size, it represents most cases of this type performed worldwide.Arterial BP monitoring setup errors occurred in several cases, therefore limiting the number of complete datasets for analysis.Data pertaining to blood product transfusion volume were not collected.This was an early-stage, single-arm observational study of prehospital Z1 P-REBOA focused on technical, short-term clinical and safetyrelated outcomes only and therefore is not designed to evaluate the effectiveness of this strategy in comparison with standard, physician-delivered prehospital care.

Conclusions
In this study, prehospital Z1 P-REBOA using distal pressure as a surrogate for flow was feasible for the resuscitation of exsanguinating trauma patients at risk of imminent prehospital death.This strategy was associated with increased proximal BP, was tolerated with ongoing proximal hemodynamic stability, and may enable early survival, but with a significant incidence of multiple organ dysfunction syndrome and late death.

c
TCA persisted or recurred transiently post-REBOA.d RT indication: predicted intrathoracic pathology.e Location of RT intervention: prehospital.f Noninvasive BP value, proximal transducer failure.g Monitoring issue, proximal intra-aortic balloon pump transduced post-REBOA (transducer cable disconnection), patient in TCA.h External cardiac massage, established TCA.i Died prehospital.j Location of RT intervention: operating theater.k RT indication: failed femoral arterial access, exsanguination, and aortic control.l Died in operating theater.m Died in ED.

Table 1 .
Baseline and Outcome Characteristics Abbreviations: BP, blood pressure; GCS, Glasgow Coma Scale; HR, heart rate; ISS, Injury Severity Score; LAA, London's Air Ambulance; N/R, no REBOA-not attempted due to clinical improvement or not recorded; NIBP, noninvasive BP; PEA, pulseless electrical activity; PLE, pronounced life extinct; P-REBOA, partial REBOA; REBOA, resuscitative endovascular balloon occlusion of the aorta; TCA, traumatic cardiac arrest; U/R, unrecordable; Z, zone.a Includes PEA.b Pulseless electrical activity.c Described immediately postfall from height, responded to initial cardiopulmonary resuscitation, then rearrested pre-REBOA.d Loss of central pulse immediately pre-REBOA.jamasurgery.com(Reprinted) JAMA Surgery September 2024 Volume 159, Number 9 The horizontal lines indicate medians; boxes indicate IQRs, and whiskers show maximum values.BP indicates blood pressure; ED, emergency department.

Table 3 .
In-Hospital Management, Cause, and Pathophysiological Mode of Death and Safety