Circulatory collapse is a leading cause of mortality among traumatic major exsanguination and in ruptured aortic aneurysm patients. Approximately 40% of patients die before hemorrhage control is achieved. Resuscitative endovascular balloon occlusion of the aorta (REBOA) is an adjunct designed to sustain the circulation until definitive surgical or endovascular repair. A systematic review was conducted for the current clinical use of REBOA in patients with hemodynamic instability and to discuss its potential role in improving prehospital and in-hospital outcome.
Systematic review and meta-analysis (1900–2017) using MEDLINE, Cochrane, EMBASE, Web of Science and Central and Emcare using the keywords “aortic balloon occlusion”, “aortic balloon tamponade”, “REBOA”, and “Resuscitative Endovascular Balloon Occlusion” in combination with hemorrhage control, hemorrhage, resuscitation, shock, ruptured abdominal or thoracic aorta, endovascular repair, and open repair. Original published studies on human subjects were considered.
A total of 490 studies were identified; 89 met criteria for inclusion. Of the 1436 patients, overall reported mortality was 49.2% (613/1246) with significant differences (p < 0.001) between clinical indications. Hemodynamic shock was evident in 79.3%, values between clinical indications showed significant difference (p < 0.001). REBOA was favored as treatment in trauma patients in terms of mortality. Pooled analysis demonstrated an increase in mean systolic pressure by almost 50 mmHg following REBOA use.
REBOA has been used in trauma patients and ruptured aortic aneurysm patients with improvement of hemodynamic parameters and outcomes for several decades. Formal, prospective study is warranted to clarify the role of this adjunct in all hemodynamic unstable patients.
B. L. S. Borger van der Burg and Thijs T. C. F. van Dongen contributed equally to this manuscript; shared first author.
Controlling catastrophic bleeding is the major life saving skill in trauma and vascular surgery. The revival of the tourniquet for management of extremity bleeding and massive transfusion protocols found their basis on the battlefields of Iraq and Afghanistan , however, difficult anatomical locations such as the neck, truncal and ilio-junctional regions continue to represent challenges for prompt bleeding control. Truncal and junctional hemorrhage are often described within the trauma literature as non-survivable injuries contributing for almost 90% of catastrophic hemorrhage fatalities in the prehospital phase, in contrast to a 10% fatality rate in extremity injuries [2‐4]. Bleeding from other non-compressible sites not related to trauma, such as the gastrointestinal tract, the post-partum uterus, a ruptured abdominal aortic aneurysm (rAAA), and traumatic disruption of thoracic, abdominal, or pelvic viscera can also represent similar challenges for prompt direct bleeding control and have traditionally been characterized as potentially non-survivable sources of hemorrhage. rAAA represents a classic source of non-traumatic major hemorrhage and has been associated with reported in-hospital mortality rates occurring in 30% of patients treated with EVAR and 42% of the patients undergoing open repair . According to the Dutch Surgical Aneurysm Audit (DSAA) the number of patients that were treated in the Netherlands for a ruptured aortic aneurysm in 2013 was 293, with an in-hospital mortality of 35.3% . It has also been demonstrated that approximately 25% of rAAAs undergo complete circulatory collapse before or during the procedure. The natural history of this uncontrolled hemorrhage has been shown to be cardiovascular collapse with consequent cerebral and myocardial hypoperfusion, ultimately leading to death [7, 8].
Endovascular balloon occlusion of the aorta is a technique where a compliant balloon is advanced into the aorta and then inflated, thereby obstructing flow into the distal circulation. This has the effect of increasing cardiac afterload and proximal aortic pressure, resulting in an increase in myocardial and cerebral perfusion . This technique was first described by Hughes in 1954 when an intra-aortic balloon catheter tamponade was utilized in two moribund Korean War casualties with uncontrolled intra-abdominal hemorrhage .
In 1964, Heimbecker reported the first use of an aortic tampon for emergency control of a ruptured abdominal aneurysm .
Recent champions of the endovascular resuscitation and trauma management (EVTM) concept have re-introduced the concept of using resuscitative endovascular balloon occlusion of the aorta (REBOA) to modern clinical practice . Two recent systematic analysis by these investigators have helped to consolidate a disparate evidence base [13, 14]. However, a clear mortality benefit has yet to be demonstrated and consequently, the REBOA concept is still not fully embedded as the standard of care in most hospitals.
Furthermore, the pioneering use of this hemorrhage control adjunct in the prehospital phase, despite early promising results [15, 16], has not been widely adopted. Recent terrorist threats in western countries have contributed to an increased awareness of the value for bleeding control in the earliest phases after injury, suggesting that the potential utilization of this REBOA in this setting warrants examination . To date, however, a complete analysis of all international published literature has not been conducted. In particular, there is a need for an examination of focusing on differences between the main indications; trauma and non-trauma related major hemorrhage (including rAAA).
The primary aim of this systematic review was to examine the use of REBOA and the mortality and morbidity associated with REBOA in patients with hemodynamic instability due to major exsanguination from both traumatic and non-traumatic sources. The secondary aim was to provide an evidence-based rationale for optimal utilization of REBOA in the in-hospital phase and, ultimately contribute to the discussion of the potential for this technology to be more aggressively employed as an on-scene adjunct for control of major hemorrhage in both trauma and non-trauma settings.
The protocol for objectives, literature search strategies, inclusion and exclusion criteria, outcome measurements, and methods of statistical analysis was prepared a priori, according to the Preferred Reporting Items for Systematic Reviews and of Observational Studies in Epidemiology recommendations for study reporting and is described in this section [18, 19].
Search strategy and selection criteria
The MEDLINE, Cochrane, EMBASE, Web of Science and Central and Emcare databases were searched for relevant articles published from January 1900 to September 2017, using the Ovid medical search engine. The keywords used in the search were composed of combinations of “aortic balloon occlusion”, “aortic balloon tamponade”, “REBOA”, “Resuscitative Endovascular Balloon Occlusion” in combination with hemorrhage control, hemorrhage, resuscitation, shock, ruptured abdominal, thoracic aorta, endovascular repair, open repair. The search was limited to original studies on human subjects (abstracts), published in English language journals.
Inclusion and exclusion criteria, data extraction and outcomes of interest
Three reviewers (B.B.B., T.D. and R.H.) independently screened the abstracts for suitability. Publications were excluded where non-ruptured aortic aneurysms were reported or ineligible study types (e.g., letters and reviews) were identified. Following abstract screening, the remaining publications were subjected to a full-text assessment. The full-text assessment consisted of two reviewers independently assessing the publication to determine suitability for inclusion. Level I–V evidence was considered, and particular attention was paid to the reporting of hemodynamic performance, balloon type, balloon deployment technique, complications, and mortality. During the full-text assessment, articles were excluded if little or no clinical data were reported on the subject of balloon occlusion. Where articles undergoing full-text review identified additional relevant studies that had not been previously identified, the additional relevant studies were also reviewed and included if eligible. If disagreement was encountered at any stage of the publication inclusion/exclusion process, the senior reviewer (RH) and a fourth independent reviewer (P.H.J) arbitrated a final decision.
Studies were rated for the level of evidence provided according to criteria by the Centre for Evidence Based Medicine in Oxford. The risk of bias was assessed for each study using the Cochrane Collaboration’s tool for assessing the risk of bias. This tool assesses six domains (selection, performance, detection, attrition, reporting, and other) and rates the risk of bias in each as “high”, “low”, or “unclear” . The key characteristics of the eligible studies in terms of study type, clinical setting, aortic zone of occlusion, mortality, and morbidity were extracted as described earlier and summarized in a tabular format. These studies are the best evidence of mortality and morbidity associated with REBOA published up to 2017.
The categorical variables were analyzed by their absolute and relative frequencies in percentages. For the trauma cohort, we performed a meta-analysis to calculate the effect on systolic blood pressure and mortality. Only studies were REBOA was used as a primary means to control blood loss and contained an adequate control group, were used in the meta-analysis. The association between two categorical variables was calculated by applying the Pearson Chi-square test or I2 test. We used with fixed effect and used Χ2/I2 for heterogeneity. The software package SPSS (24.0, IBM Corporation, Armonk, New York), was used for statistical analysis to achieve a combined outcome. In all cases, p < 0.05 (two-sided) was considered statistically significant.
A total of 440 unique studies were identified and underwent abstract screening (Fig. 1), 107 of these studies were deemed appropriate for and underwent full-text review. Of these, 30 studies reported little or no clinical data, leaving 77 studies for inclusion following full-text review. Reference review of these studies identified an additional 53 eligible studies, which were not identified during the key word search. These studies were predominantly older in publication date but were still eligible for inclusion. Therefore, after final quantitative synthesis, 89 articles were selected, including 28 case reports [11, 20‐46], 25 case series [10, 47‐71], and 36 (retrospective) cohort studies [12, 72‐105]. Also 8 reviews were identified but not included in calculations, due to concern for double counting of unique cases [5, 13, 14, 106‐110]. Most studies were deemed to be at high risk of bias. In total, 1482 patients treated with REBOA were included in analysis (Table 1).
EVAR endovascular aneurysm repair, N number, in-hosp in hospital, AO “Arbeitsgemeinschaft für Osteosynthesefragen”, OXLE Oxford level of evidence, Art.Ins location of arterial incision, tech technique, Perc percutaneous, Occl occlusion, SD standard deviation, Δ P mmHg pressure difference in mmHg, rAAA ruptured abdominal aneurysm, abdpel hem abdominal/pelvic haemorrhage, PPH post partum haemorrhage, (U)GI (upper) gastro intestinal, OR operation room, r. ruptured, iatr/gyn iatrogenic/gynaecological, Y yes, NA not applicable, U unknown, nm not measurable, Ao aortal, Ax axillary, B brachial, C carotis, F femora, (I)AOB (intra) aortic occlusion balloon, bal balloon, cath catheter, pts patients, fem a. femoral artery, def deflation, pt(s) patient(s)
Indications, hemodynamic instability and mortality
REBOA has been used for the management of hemorrhage grouped in three major clinical indications: (1) traumatic abdominopelvic hemorrhage (18 studies), (2) hemorrhage arising from rAAA (50 studies) and, (3) miscellaneous causes such as post-partum, gastro-intestinal bleeding or exsanguination during pelvic surgery (21 studies). The REBOA concept was used in all studies as a hemorrhage control and resuscitation adjunct for prompt hemorrhage control.
Hemodynamic instability (transient or non-responding to fluid therapy) was present in 79.3% (1175/1482) of the total studied population. The studies with patients with established hemorrhagic shock [systolic blood pressure (SBP) below 90 mmHg] due to trauma 90.9% (791/870), rAAA 91.7% (355/387) and other 12.9% (29/225) were respectively described. Overall reported mortality within the group that was hemodynamically stabilized with REBOA was 49.2% (613/1246). Reported mortality among studies concerning trauma was 63.0% (545/865). Meta-analysis revealed a significant difference in mortality (p < 0.001) of REBOA compared with the mortality of patients treated with other means. It showed a risk difference of 0.27 (0.14–0.49) favoring REBOA (Fig. 2). In studies regarding rAAA mortality was 39.1% (61/156) and for miscellaneous causes 3.1% (7/225). There were no episodes of mortality reported among studies on post-partum hemorrhage, gastrointestinal bleeding, or with the use of balloon occlusion during pelvic and sacral surgery. Additional information can be found in supplemental Tables 1, 2 and 3.
The effects of REBOA
The effect on SBP in hemodynamic instability, as seen in Table 2, was a significant increase of SBP by a mean of 78.9 mmHg in trauma (Fig. 3), 56.1 mmHg in rAAA and 52.4 mmHg other types of patients in established hemorrhagic shock, among the 23 studies [10, 12, 22, 25, 29‐31, 34, 36, 40, 41, 44, 45, 47, 48, 52, 55, 58, 59, 62, 66, 71, 97, 104, 105, 111] reporting pre-REBOA and post-REBOA SBP values.
Mortality, occlusion time and rise of systolic blood pressure with application of REBOA
Number of studies
Meta-analysis (n = 240)
Mean occlusion time
Number of studies
11,111/190 = 58.5 min
270/6 = 45.0 min
2175/32 = 68.0 min
90/3 = 30.0 min
30/1 = 30.0 min
31/2 = 15.5 min
205/4 = 51.3 min
10,114/189 = 53.5 min
11,496/197 = 58.4 min
305/7 = 43.6 min
12,320/223 = 55.2 min
REBOA indicates resuscitative endovascular balloon occlusion of the aorta, rAAA ruptured abdominal aneurysm, Art.Ins location of arterial incision, tech technique, min minutes, ΔP rise of systolic blood pressure in mmHg
REBOA balloon occlusion times were adequately reported in 35 studies [10, 12, 23, 25, 26, 29‐34, 36, 38‐41, 43‐45, 47, 48, 55, 58, 62, 66, 68‐71, 82, 83, 97, 101, 104, 105]. Occlusion times within the zones of occlusion were described in 15 studies examining Zone I occlusion times, where 196 patients had a median occlusion time of 58.4 min. Only three studies reported on occlusion in zone 2, with a mean of 43.6 min. The remaining nineteen studies examined 223 patients undergoing Zone III occlusion, with a median occlusion time of 55.2 min.
Of the studies included for review, 81 (92.0%) reported adequate data regarding arterial access and REBOA balloon deployment (Table 1). The femoral artery was the most commonly reported access site (96.8% of the patients) for REBOA, followed by direct aortic access (2.2% of the patients). Interestingly the largest study in our review, Norii et al (n = 452), did not report information on the access sites utilized . Iatrogenic injuries related to REBOA use were 3.7% (32/870), 2.6% (10/387) and 5.3% (12/225) respectively on trauma, rAAA and miscellaneous causes (p = 0.21).
The current review is the largest systematic analysis of the literature on the subject of REBOA. A total of 89 studies, reporting on 1482 patients have been identified in the current published literature. The REBOA concept increases the SBP in hemorrhagic shock and is an adjunct for both endovascular and open repair in hemodynamic instability due to major exsanguination. The described iatrogenic injury rate related to REBOA is below 5%.
Central traumatic major exsanguination
The most important question to answer is whether REBOA imparts a survival benefit in the in-hospital phase of care. Within the trauma literature, both non-survivors and survivors tended to have a higher burden of injury, presenting with gross physiologic derangements. The majority of studies lack a sound control group, with the exception of the studies by Norii et al.  and Moore et al.  In most studies, the study of Norii et al.  included, it was unclear whether REBOA was used in the context of a formal damage-control protocol, with consistent application, or whether it was used as a “last-ditch” attempt to salvage in hospital patients (e.g. ascending aortic ruptures) in civilian or military environment who were anticipated to have impending unfavorable outcome . For this reason the study by Norii et al. , had to be excluded from the meta-analysis.
While our report suggests REBOA utilization has increased in recent years with results revealing a significant rise in blood pressure and a survival rate of approximately 40% with patients in shock (Table 2), the use of this adjunct is not universally embedded in hospital settings, not to mention the pioneering prehospital phase . In the combat environment, major hemorrhage from central vascular injuries endures as the leading cause of death and most of the casualties decease before reaching a hospital [112‐115]. REBOA can, therefore be an adjunct in the resuscitation of these major central/junctional hemorrhages in both the military as well as the civilian environment. Future research should focus on the usage of the REBOA concept in the prehospital phase, including selection and training of the right medical personnel for this important form of bleeding control.
Ruptured acute aortic aneurysm
The treatment of choice for rAAA depends on hemodynamic, morphological characteristics as well as the availability of an endovascular team trained for EVAR procedures. Historically, patients in profound hemodynamic shock have been deemed not suitable for endovascular approach, because of the extra time involved in measuring and preparing for an endovascular procedure. The use of REBOA in these patients may extend the period needed to assemble the necessary staff for endovascular repair. It may even have potential application in the pre- or inter-hospital setting, to bridge the transport time to an endovascular center with sufficient expertise.
Three RCT’s have studied the optimal treatment for ruptured aneurysms, EVAR or open [116‐120]. In the reported results of the ECAR and AJAX trials, EVAR was found to be equal to open surgical repair in terms of 30 day and 1-year mortality. In contrast, the IMPROVE trial concluded that a strategy of endovascular repair was not associated with significant reduction in either 30-day mortality or cost. The 2014 Cochrane review on this subject concluded that, based on available data at the time of analysis, there is no difference in the 30-day mortality outcomes between EVAR and open repair . It is important to note, however, that none of the presently available trials have randomized high-risk patients with ruptured aneurysms, particularly those in shock or with low blood pressure . These patients potentially represent the population most likely to benefit from early aortic balloon occlusion, possibly even in the pre-hospital phase or in the emergency department. This patient subset warrants further examination.
Miscellaneous causes of major hemorrhage
Miscellaneous causes of significant hemorrhage, such as post-partum hemorrhage, gastro-intestinal bleeding or exsanguination during pelvic surgery, are also likely to benefit from early direct bleeding control to prevent the onset of the lethal triad (hypothermia, acidosis and coagulopathy). While the data on this subset of patients is limited, our observed mortality rate of lower than 5% among REBOA patients is suggestive that this technology has potentially important applications for these indications. Since shock was not presented in the studies regarding PPH, GI bleeding or pelvic surgery, additional examination of these patient subgroups is warranted.
While our review represents the largest of such effort in the medical literature, it does possess several limitations that must be addressed. The major limitation of the current review is the quality of the available evidence, which is limited. The risk of population bias in this systematic review is inevitable, to minimize best possible effects of heterogeneity and cohort overlap we provided a narrative description of prevalence and characteristics. A recent published review of Gamberini et al.  contained several papers that are not mentioned in our study [123‐133], most of them based on the Japanese trauma registry. It was not possible to determine how much overlap there was with the studies we included from the Japanese registry. To prevent bias by double patient inclusion, these studies were not included in this analysis. The majority of the evidence identified consists of case reports and case series (Grade IV, V evidence) with only limited cohort studies and reviews identified (Grade II, III evidence). Most studies are at significant risk of bias, underestimating the true mortality and morbidity (including iatrogenic injuries). These findings should prompt efforts to improve on this by stimulating formal prospective evaluation.
REBOA is a potential lifesaving adjunct for resuscitation but prolonged occlusion of the aorta can lead to organ failure due to resulting ischemia–reperfusion injury. Consensus regarding safe occlusion times is not yet reached although animal studies indicate an optimal duration of 30 min. Occlusion time of 60 min or above is associated with an increased mortality [134, 135]. For prolonged field care or surgery, intermittent or partial occlusion could be a solution . Three studies reported on occlusion times in zone 2 [40, 104, 111]. When the exact location of the bleeding is known, control in this zone could be needed in specific circumstances. The use of zone 2 as deployment zone is usually not recommended .
Prospective data collection is underway in the form of an American Association for the Surgery of Trauma sponsored observation study (AORTA), the European registry (ABOT), the Japanese registry (DIRECT-IABO) and the UK-REBOA Randomized Control Trial which should permit the consistent capture and reporting of REBOA-specific data such as indications, hemodynamic performance, outcome, cause of death, and morbidity [137‐140].
Growing experience with the REBOA concept has proven its value in (acute) aortic repair, and expanding experience is gathering in the clinical implementation of this technology in major hemorrhage for trauma indications. Ongoing research should provide a much-needed higher quality of data in the coming years. These maturing experiences should yield improved data on the degree of hemorrhagic shock, specifics of the setting of use (pre- and in-hospital), the influence of use on organ failure, associated blood product requirements, encountered iatrogenic injuries and the specific techniques utilized for deployment (balloon occlusion time; full/partial /intermittent inflation approaches) [12, 104, 141].
The REBOA concept has been used as an effective early hemorrhage control and resuscitation adjunct following traumatic abdominopelvic hemorrhage, hemorrhage arising from rAAA and significant bleeding from miscellaneous causes (post-partum, gastro-intestinal bleeding or exsanguination during pelvic surgery). Once placed in the correct aortic zone it has been shown to effectively raise the SBP in patients in major hemorrhagic shock. Despite these documented hemodynamic improvements, however, there remains a need for improved evidence supporting mortality benefit following REBOA use in the setting of traumatic hemorrhage. Presently available data, however, continues to suggest a potential value for REBOA use in austere environments or for patients in extremis that cannot be ignored. For this specific subset of patients, where the traditional alternative is likely death, the adage “do you need a RCT to prove that a parachute works” may prove worthy of further discussion. A feasibility study to setup a formal training program for obtaining vascular access and REBOA placement by non-vascular medical specialists has recently been conducted by this study group. This systematic review underscores that solid prospective evaluation of the REBOA concept remains a significant requirement for the establishment of optimal guidelines for REBOA employment in the management of hemorrhagic shock in any phase of medical care.
We thank J.W. Schoones (Walaeus Library, Leiden University Medical Center) for his methodological support.
Conflict of interest
All authors declared that they have no conflict of interest.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
Mit e.Med Orthopädie & Unfallchirurgie erhalten Sie Zugang zu CME-Fortbildungen der Fachgebiete, den Premium-Inhalten der dazugehörigen Fachzeitschriften, inklusive einer gedruckten Zeitschrift Ihrer Wahl.