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

The axilla is a vulnerable spot, poorly protected by armoured vests in the military setting. An extremity tourniquet is considered as treatment of first resort in haemorrhage of the upper extremity. If bleeding persists despite an adequately applied tourniquet or the site of bleeding is too proximal to apply a tourniquet, military guidelines advocate to start with a haemostatic gauze in combination with direct pressure [33]. The prehospital use and efficacy of haemostatic gauzes has been reviewed extensively [22‐24] and will not be discussed. After animal-model testing [34], the FDA approved XSTAT-30™ (RevMedx, Wilsonville, Ore), a syringe-like device containing dozens of small compressed chitosan covered cellulose sponges that can be injected in junctional bleedings and secured with regular bandages. Intra-thoracic, −pelvic or –abdominal use are contra-indicated [35]. No clinical use was yet reported at the time of this review.

A downside of applying direct pressure is that a care provider is no longer available to perform other interventions. In the military, due to tactical circumstances and longer retrieval times, this has led to a request for mechanical devices to free the hands of the medical care provider.

Four junctional tourniquets which can be used in case of junctional haemorrhage are the Combat Ready Clamp (CRoC)™, Junctional Emergency Tool (JETT)™, SAM Junctional Tourniquet (SAM-JT)™ and Abdominal Aortic Junctional Tourniquet (AAJT)™ [33]. Simultaneous use of haemostatic dressings and (junctional) tourniquets may work synergistic in controlling haemorrhage [33, 36].

The CRoC™ has been FDA-cleared for axillary haemorrhage and has shown to be effective in a cadaver model [37]. It is a device which consists of a stamp compressing the targeted area underneath. See Fig. 1. It has been reflected that the CRoC™ is heavy and assembling takes too long (1–2 min) [38]. However, in civilian prehospital care, a ready-to-use pre-assembled unit can be stored in an ambulance or helicopter [33].

At the time of this review no studies on the use of JETT™ or SAM-JT™ junctional tourniquet for axillary bleeding were available.

The AAJT™ has been successfully used in the civilian setting in case of a gunshot wound to the axilla [39]. The wedge-formed bladder was positioned in the axilla and the device strapped around the torso. No further bleeding from the axilla was seen after inflating the bladder up to 250 mmHg pressure. Radiographic imaging showed some compression of the left thorax, but ventilator inspiratory pressures were normal [39]. An observational trial in human volunteers showed successful eliminated blood flow in the brachial artery in axillary application of the AAJT™ using a mean pressure of 168 mmHg. Doppler measurement showed complete elimination of flow and spontaneous recovery of flow when the AAJT™ was deflated [40].

When a junctional tourniquet is not available, or will take too much time to apply and/or when maintaining direct pressure is not an option, the use of the iTClamp™ (Innovative Trauma Care Inc., Edmonton, Alberta, Canada) may be considered. The iTClamp™ is a mechanical clamp which looks like a hair clamp with several small needles that seals a wound by approximating the wound edges of the overlying skin firmly, creating a compartment in which the bleeding potentially tamponades. See Fig. 2. Several reports of field use exist [41, 42]. See Table 1. With the iTClamp™ haemorrhage control was gained in 9 out of 10 applications for venous or arterial and venous combined origin in scalp (n = 7), neck, chest wall, and lacerations from open femoral fractures. The clamp failed to control haemorrhage from a combined carotid and vertebral arterial laceration in the neck. By packing the wound with haemostatic gauze, reapplication of the iTClamp™ and direct pressure haemorrhage control was achieved [42].

The JETT™ and SAM-JT™ can be used in bi- or unilateral inguinal or lower extremity haemorrhage. The CRoC™ is a unilateral device. Previously, the AAJT™ had only FDA clearance for umbilical application, which occludes the abdominal aorta and can be used for bilateral inguinal or pelvic haemorrhage [43, 44]. See Fig. 3. The clearance has been expanded to junctional use: placement over the groin provides unilateral occlusion of arterial flow at the common femoral artery [40].

The CRoC™ has been reported to be effective in clinical, cadaver, manikin and animal studies in groin bleeds. See Table 1. Tovmassian et al. reported a case using the CRoC™ successfully in a proximal lower extremity traumatic amputation on the battlefield. Haemorrhage was controlled rapidly after assembling and applying of the CRoC™ which took 90 s [45]. Potential dislodgement during combat situations or transport has been mentioned as a concern of CRoC™ use [46].

The JETT™ is designed as a belt including two pressure pads with threaded T-handles. See Fig. 4. It stabilises the pelvic ring and provides the option of bilateral haemorrhage control by compressing the common femoral artery, thereby occluding arterial blood flow to the lower extremities. Application time to occlude bilateral flow in a cadaveric study was about 10 s with JETT™. In contrast, bilateral CRoC™ application (two devices) took 68 s [47].

The design of the SAM–JT™ is similar to the JETT™: designed like a belt, but instead of threated T-handles, the SAM-JT™ is equipped with two pneumatically inflatable bladders to compress the common femoral artery uni- or bilateral. See Fig. 5. Klotz et al. reported the use of the SAM-JT™ in inguinal bleeding of an Afghan soldier. In combination with haemostatic gauze, haemorrhage was adequately stopped [48].

The Abdominal Aortic and Junctional Tourniquet™, formerly known as Abdominal Aortic Tourniquet™, can be used in several ways. We previously mentioned the use of the AAJT™ in axillary bleeding. The AAJT™ can also be placed over the groin and has a truncal application compressing the distal aorta in the infra-renal zone at the umbilical region. In this way, bilaterally arterial blood flow to the lower extremities and to the pelvic region is blocked. Successful usage of junctional as well as truncal application has been reported in the military setting [49, 50]. Truncal application in animals was found to have minimal side effects when used for 30 min. Pregnancy, abdominal aneurysm and penetrating abdominal trauma are considered to be contraindications for truncal use of the AAJT™ [44].

Kragh et al. compared the four junctional tourniquets in a simulated out-of-hospital situation by army medics. Unfortunately, at the time of the study, the AAJT™ was only FDA-cleared for truncal application. U.S. armed forces medics preferred the CRoC™ and SAM-JT™ over the JETT™ and (truncal) AAJT™ if they could bring one device on a mission, after testing the various designs at each other [43]. Kotwal et al. reported feedback from battlefield use. The AAJT™ was reported “to be easily broken” and the CRoC™ as “bulky, heavy and takes too much time to apply” but evacuation units carry it pre-assembled in their (air) vehicles [33].

The iTClamp™ has been proven to be feasible in swine and perfused cadaver models with lethal femoral artery bleeding [51, 52]. Two case reports in which the iTClamp™ stopped inguinal or proximal lower extremity bleeding were published in respectively non-combatant military and civilian cases [41, 53] See Table 1.

Resuscitative endovascular balloon occlusion of the aorta (REBOA) is an intervention that shows promising results in patients with non-compressible haemorrhage of the torso [54]. Endovascular balloon occlusion of the aorta was first described by Hughes to control an intra-abdominal bleeding in the operating room during the Korean war [55]. REBOA was abandoned after the high complication rate of 35 % reported by Gupta et al. in 1989. Several cases of paraplegia, thrombosis and the catheter exiting the aorta were found [56]. A recent publication of a 60 min lasting thoracic deployment of REBOA in 16 swine resulted in 4 cases of spinal ischaemia, of which 2 were lethal [57]. REBOA requires femoral arterial access for the introduction of a catheter which, after inflation, occludes the aorta. This leads to a decreased blood flow at the site of injury (distally from the balloon), and secondly to an increased central aortic pressure (CAP) [58]. Stannard et al. published an article describing a step-by-step procedure of REBOA use [59]. Zones for balloon deployment are the thoracic Zone I (left subclavian artery to coeliac artery) and zone III (infra renal aorta), for intra-abdominal or pelvic/inguinal haemorrhage respectively [59]. Several series of emergency department (ED) use of REBOA have been reported, but predominantly in cases of pelvic haemorrhage [54, 60‐62]. The first prehospital REBOA intervention was performed by the London Helicopter Emergency Medical Service (HEMS) and will be discussed in the paragraph on pelvic haemorrhage. To facilitate prehospital REBOA, smaller (Fr7) balloon catheters and contrast-enhanced ultrasonic control for adequate positioning have been developed and tested in preclinical animal studies [63, 64]. Usage of REBOA to control junctional haemorrhage from the groin and lower extremities has not been described but may deem feasible.

Resuscitative thoracotomy (RT) is a radical intervention which can be performed by any physician that can handle a scalpel [65, 66]. At the appropriate timing, RT enables relief of cardiac tamponade, aortic cross-clamping, pulmonary hilar clamping or twisting and internal cardiac compression [65‐69]. Recent studies suggest that pre-hospital RT in case of penetrating trauma, more specifically stab wounds, could benefit that subcategory of patients [70]. Chance of survival for patients with gunshot wounds [65, 71] or blunt trauma are poor [65, 70, 72]. The London Helicopter Emergency Medical Service reported prehospital (or field) thoracotomy in 71 patients resulting in thirteen patients (18 %) to survive after additional treatment in a trauma centre. All survivors (n = 13) had a cardiac tamponade and a ventricular (n = 12) or aortic (n = 1) wound [70]. Clamshell incision (CI) may be a superior approach over left anterolateral thoracotomy since it offers better exposure and does not take more time for inexperienced physicians performed in a cadaver model for surgical residents [73]. In the prehospital case series reported by Coats et al. and Davies et al. RT was exclusively done via CI [70, 74]. Morrison et al. stated that pre-hospital thoracotomy in the military setting is futile with an estimated 0.7 % survival for patients with multiple gunshot wounds [75]. A case series of 34 prehospital thoracotomies in blunt trauma by Japanese HEMS showed no survivors [72]. A recent systematic review on emergency department resuscitative thoracotomy (EDT) in Europe showed a far more favourable survival rate of 12.9 % (18 out of 139) in blunt and 41.6 % (37 out of 89) in penetrating injuries [76].

Guidelines by the European Resuscitation Council (ERC) and Eastern Association for the surgery of Trauma (EAST) strongly recommend RT after penetrating trauma with witnessed signs of life (SoL) or ECG activity after short-term, and conditionally recommend RT in penetrating trauma without these features. In blunt trauma without witnessed SoL, RT is considered futile. Patients with blunt trauma with vital SoL and witnessed cardiac arrest RT can be performed. Nevertheless, in this ‘favourable’ group of blunt trauma patients the outcome is poor with an estimated survival rate of 1.6 % [77, 78].

Animal studies showed reduced blood loss by raising abdominal pressure by gas insufflation in case of hepatic [84, 85], splenic [86, 87] and inferior vena cava bleeding [88]. See Table 1. Kasotakis et al. described the possibility to use this technique en-route by using a portable carbon dioxide insufflator [87]. No clinical studies were identified for this review.

Another invention to counter intra-abdominal haemorrhage is to inject biocompatible self-expanding foam intraperitoneally. This foam expands, engulfs the organs and becomes solid, thereby tamponading the bleeding. So far several animal studies have been executed in hepatoportal [82, 83, 89, 90], splenic [91] and iliac artery [92] models of intra-abdominal haemorrhage. See Table 1. Localized enteric pressure necrosis has been reported as frequent complication, requiring seromuscular sutures or resection [91]. Rago et al. stated that in a survey of 3442 injured soldiers in the Operation Iraqi Freedom suffering intra-abdominal bleeding, 34 % of patients required bowel repair and 19 % even resection anyway, so the localized necrosis might be considered an acceptable complication in patients facing imminent exsanguination [91, 93]. In July 2015, Mesar et al., published a study on the optimal human dosage by injecting the foam in recently deceased humans [94]. At the time of this review, no clinical use of ResQFoam™ (Arsenal Medical, Watertown MA) has been published.

The previously described REBOA can also be applied to thoracic zone (Zone I), thereby occluding abdominal aortic flow in case of intra-abdominal haemorrhage. REBOA is considered to be far less invasive compared to a thoracotomy with descendent aortic cross clamping [95]. It has been assessed in various animal models [57, 58, 63, 96]. as well as in the clinical setting [54, 61, 62, 97]. See Table 1. Ogura et al. published a case series of patients (n = 8) with haemoperitoneum after blunt abdominal trauma with a successful in-hospital REBOA in 86 % (n = 7) [97]. Moore et al. reported a lower mortality rate for in-hospital REBOA (n = 24) vs resuscitative thoracotomy (n = 72) in non-compressible truncal haemorrhage: 62.5 % vs 90.3 %. Injury Severity Score did not differ significantly. However, in the RT-group a higher percentage received cardiopulmonary resuscitation (CPR) upon arrival [62]. Analysing patients from the Japanese trauma data bank who received an in-hospital REBOA, Norii et al. found a higher mortality in patients receiving a REBOA compared to those who did not, even after correction for injury pattern and other variables. This might be explained by REBOA being used as a ‘last ditch’ effort. Interestingly, REBOA is part of the Japanese emergency physicians arsenal since trauma surgeons are not always attending 24/7 [98].

The AAJT™ has not been reported for abdominal bleeding, however when no other options are available, its use might be feasible since it has the ability to restrict the intra-abdominal compartment by external pressure, theoretically providing an earlier tamponade of the abdominal cavity. Nevertheless, abdominal application is contra-indicated in the case of penetrating injury [46].

Pelvic fracture, or the so-called ‘the killing fracture’ [99], is potential lethal, especially in severe pelvic disruption with following haemorrhagic shock [100]. Mortality of traumatic pelvic injury is estimated at 28 % by Papakostidis et al. in a recent systematic review [101]. Haemodynamic instability is present in 10 % of pelvic fractures [102], and often the origin of the bleeding is uncertain: it may be caused by (a combination of) bleeding from fracture surfaces, the venous plexus or arteries [103].

A study of 5340 German trauma patients with a pelvic fracture showed a mortality of 4 % (238 patients). In these deceased patients, massive haemorrhage was the cause of death in 34 %, of which 62 % was specified to the pelvic region. Holstein et al. did not include prehospital mortality, so the mortality rate due to exsanguination is likely to be even higher [104].

Several devices providing circumferential pressure are on the market [103], if not equipped with such a device a conventional sheet should be applied [105]. Placing a circumferential binder (CB) or sheet will reduce the volume of the intra-pelvic cavity, compress the fracture surfaces and prevent dislodgement of newly formed clots by preventing second displacement [106].

Tan et al. reported a reduced symphyseal diastasis of 60 % in unstable pelvic fractures after emergency department application of a Trauma Pelvic Orthotic Device™ (T-POD™, Pyng Medical, Richmond Canada) and a rise in mean arterial pressure form 65.3 to 81.2 [107]. Croce et al. stated that the implementation of the T-POD™ significantly reduced transfusion requirement (9.9 vs 21.5 units) and length of stay (16.5 vs 24..4 days) [108]. The SAM-sling™ (SAM Medical Products, Oregon, USA) is an alternative, and was tested clinically by Krieg et al. in 16 patients with a pelvic ring injury upon arrival at the ED resulting in an 9.9 % reduction of the pelvic distension [109]. The PelvicBinder™ (PelvicBinder Inc., Dallas, USA) has been reported to be superior over sheet wrapping for emergency stabilisation of the pelvis [110]. Knops et al. found all of the three commercial temporary pelvic stabilizers providing sufficient reduction in of the pelvic fracture in a cadaveric study, with the T-POD requiring the lowest force [111].

If the patient haemodynamically further deteriorates, despite application of a CB, an arterial origin is suspected [112]. The CB should be left in place, and an AAJT™ can be placed at the umbilical region to externally compresses the abdominal aorta to the spine, prohibiting distal arterial flow to the pelvis and lower extremities. Taylor et al. reported successful aortic occlusion in 15 of 16 voluntary soldiers by Doppler sound examination of the common femoral artery [46].

REBOA in uncontrolled pelvic haemorrhage is of great interest and both preclinical [113‐115] and clinical publications on this topic are vast [54, 60‐62, 116]. See Table 1. As the iliac artery is on the route of the REBOA placement, unilateral femoral pulsation is required. An in-hospital series of 13 cases of uncontrollable haemorrhage due to pelvic injury the use of REBOA led to initial haemorrhage control in 12 patients. Deflation of the balloon in the angiographic suite resulted in recurrence of haemodynamic instability in half of the patients. Ultimately 6 of the initial patients survived to discharge. The Injury Severity Scores (ISS) differed among survivors and non-survivors (ISS 38 versus 58, p = 0.011) but the Revised Trauma Score did not: 4.362 versus 4.779, p = 0.761 [60]. Brenner et al. reported ED deployment of REBOA within 6 min (range 4–6) in a series of 6 patients (5 cases of pelvic fracture and 1 patient with a renal injury). Two patients did not survive due to concomitant brain injury and non-survivable injuries [54]. The London HEMS in 2014 reported the first pre-hospital performed REBOA in a patient with a pelvic fracture with a successful control of the bleeding bridging transport to definitive care [116].