Jaime All • John F. Angle
Embolotherapy for extremity trauma is a well-established and essential tool for the management of hemorrhage and vascular injury with potential for hemorrhage. Familiarity with clinical presentation, pretreatment imaging, and angiographic findings are all essential components to interventionalists providing these services. Embolization technique and the similarities and important differences between pelvic and extremity injury will be discussed.
Within the context of a multispecialty approach to the care of the trauma patient, the transcatheter arterial embolization (TAE) is an essential tool in the treatment of traumatic pelvic arterial injury.1 Many trauma centers have developed algorithms based on the nature of the injury, computed tomography (CT) findings, and other injuries.2–5 Some trauma scenarios are best treated with TAE first, others are supplemented with TAE, and many will not benefit from early pelvic angiography.2–7 Blunt or penetrating pelvic trauma may lead to vascular injury, with blunt injuries being much more common in most geographic areas. Blunt trauma resulting in pelvic fracture is associated with vascular injury in up to 40% of patients, and hemorrhage is still the highest cause of mortality (50% to 74% of patients).8–10 With blunt pelvic injury, the source of hemorrhage may be venous, arterial, or from cancellous bone.7,11TAE is only indicated for the treatment of arterial hemorrhage and is indicated in approximately 3% to 10% of cases.2,8,12,13 In the correct clinical setting, angiography with TAE is effective in controlling pelvic hemorrhage in 85% to 97% of cases.6 Patients embolized within 3 hours of arrival have a statistically improved survival rate as compared to those receiving intervention outside of the initial 3-hour window.14
Indications for TAE in the setting of pelvic trauma include pelvic fractures and hemodynamic instability after exclusion of other sources of hemorrhage outside of the pelvis, active extravasation of contrast on computed tomography angiography (CTA) in the presence of pelvic fracture regardless of hemodynamic status, prior TAE for pelvic trauma with ongoing hemodynamic instability after other sources of hemorrhage have been excluded, and patients older than 60 years of age with a high-risk pelvic fracture pattern.6,12,15,16
Motor vehicle collisions and pedestrians hit by motor vehicles account for most blunt arterial injury.12 Domestic falls are a less common cause, but the need for pelvic vascular injury evaluation in this population is well documented.7,12,16 This is particularly true for the elderly because age is an independent predictor of active bleeding and indication for TAE, regardless of hemodynamic stability.12,16Additionally, osteoporosis in elderly women may result in increased rate of pelvic fracture, even when the mechanism of injury is less dramatic.17 The prevalence of chronic anticoagulation for cerebrovascular and cardiovascular disease in this patient population may also necessitate early intervention.
Types of vascular injury seen with blunt pelvic trauma include complete or partial transection, arteriovenous fistula, pseudoaneurysm, dissection, and traumatic thrombosis. The different angiographic appearances of traumatic vascular injury are listed in Table 24.1.1,18–20 Vascular injury typically occurs in relationship to adjacent osseous fractures or shearing forces against the ligamentous structures of the pelvis.19 It is important to note that disruption of the osseous pelvis significantly enlarges what is normally a confined space. This allows for a larger volume of hemorrhage to accumulate rapidly, and thus early intervention is necessary to prevent exsanguination.7 It is worth repeating that pelvic hemorrhage is usually from venous or fractured cancellous bone, and this type of low-pressure hemorrhage is best controlled with external fixation or compression with pelvic binding. The goal of these methods is to reestablish the normal tamponading effect of the retroperitoneum, although some recent studies call into question the overall efficacy with this approach.6,19,21 Peritoneal pelvic packing has been introduced as an additional method of treating pelvic hemorrhage. Some centers have questioned; whether pelvic packing should replace TAE, most use TAE as an adjunctive treatment, and others suggest TAE first with fixation secondarily.2–7,21 From a practical standpoint, we push for angiography first if active extravasation is seen on CT and suggest deferring the fixation, which can make angiography very difficult to perform. Regardless of the algorithm or combinations of treatment employed, early control of pelvic arterial hemorrhage has a strong association with improved outcome, and TAE continues to occupy a central role in the treatment of traumatic pelvic hemorrhage.1–6,14
As many as 50% of patients with pelvic fracture have additional injuries, so pelvic fracture should be treated as an indicator of polytrauma (Fig. 24.1).21 An algorithm approach to the care of the trauma patient, although institution-dependent, must weigh the severity of multiple injuries and therefore mandates a multidisciplinary approach. Most would prioritize an intra-abdominal source of hemorrhage over intrapelvic bleeding, although a few have argued to the contrary.4,6,22 However, it is accepted that exploratory laparotomy in the setting of unaddressed pelvic hemorrhage can lead to rapid expansion of the pelvic hematoma.9 Additionally, some solid organ injury within the abdomen may be treated at the time of pelvic embolization, such as a splenic or renal laceration. Regardless of when TAE is performed, consideration must be given to the possibility of additional organ injury, and this should guide the goal of the procedure in terms of the degree of vessel selectivity. Failure of hemodynamic status to correct following embolization should prompt search for additional sources of bleeding.
Historically, fracture patterns were felt to be important diagnostic indicators in pelvic trauma.9,19 More recent studies have shown that arterial hemorrhage can occur with any of the fracture patterns, as well as in the less severe subtypes, and may occur in the absence of fracture.22 Those most worrisome (and those which are considered high-grade at our institution) are vertical shear and types II and III lateral compressions and anterior posterior compressions.9,19,23 It is still important for the interventionalist to recognize these patterns because they may suggest which vessel is the culprit, even in the absence of active extravasation at angiography. At the most basic level, anterior fracture patterns are associated with injury to the anterior division of the internal iliac artery, and posterior fractures are associated with injuries to the posterior division.1,23 The importance of fracture pattern recognition has been largely supplanted by CT, which has superior diagnostic sensitivity, and adds a wealth of additional data over conventional radiography.24,25 Advances in technology have allowed for rapid image acquisition and processing, thereby allowing for the imaging of patients whose status may be tenuous and, in some centers, even those who are unstable.
Examples of penetrating pelvic injuries include gunshot wounds, stab wounds, or iatrogenic injury (such as with percutaneous drainage tube placement and placement of orthopedic hardware).26 Penetrating injuries are more likely to result in transection or pseudoaneurysm formation, with or without arteriovenous fistula.27 Traumatic pseudoaneurysms are at high risk for rupture and should all be treated, preferably with endovascular methods when possible.18
Traumatic vascular injury to the extremity may also cause life-threatening hemorrhage, but, more commonly than in the pelvis, it may result in vessel occlusion with signs of acute limb ischemia. Iatrogenic injury accounts for 74% of embolization procedures performed in the extremity.28 Many of these are associated with orthopedic procedures, which are likely to increase as our aging population requires an increasing number of major joint procedures. Arterial injury associated with orthopedic procedures typically involves a branch vessel in the pelvis or lower extremities, rather than a major artery, and is thus ideally suited to treatment via endovascular methods.28
Traumatic arterial injury of the extremity is divided into penetrating or blunt trauma. The distinction between the two has important clinical implications. Penetrating injury (more common) results in pseudoaneurysm, arteriovenous fistula, and partial or complete transection.29 Outside of iatrogenic causes, glass is the most common source of penetrating injury followed by gunshot wounds and stab wounds.30 Blunt injury is often seen with associated orthopedic trauma such as fracture and dislocation. The vascular injury seen with blunt trauma is typically dissection, pseudoaneurysm, or occlusion secondary to intramural hematoma or traumatic thrombosis.31 Diagnosis of arterial injury in the setting of a penetrating mechanism is usually more straightforward as there is obvious external evidence of injury. Hard signs of injury are more likely to be present, such as active external hemorrhage, enlarging hematoma, or obvious indicators of limb ischemia.27 Signs of limb ischemia include pallor, abnormal peripheral pulses, unequivocal neurologic defects, and a thrill or bruit.32
Blunt peripheral arterial injury is more likely to be occult with delayed diagnosis and concomitant traumatic injuries to the head and body, which may take precedence over the extremity.31 Diagnosis of blunt arterial injury requires a higher index of suspicion and recognition of soft signs of vascular injury. These include (1) discrepancies in the ankle–brachial index performed with Doppler ultrasound and (2) adjacent injury, including fracture, nerve injury, or stable hematoma.18 Duplex ultrasonography may be beneficial for characterization of some specific vascular injuries, but CTA provides clinically important additional information about extremity osseous and soft tissue structures.27,33 Additionally, newer CT scanners can acquire arterial phase imaging of the extremities during the concomitant CTA evaluation of the chest, abdomen, and pelvis. Conventional angiography for diagnostic evaluation of acute arterial injury has largely been replaced by CTA, given the excellent sensitivity of CTA.33
Vascular injury is more common in the lower extremity than the upper extremity, and the femoral and popliteal arteries are the most common sites of vascular traumatic injury.34 Knee dislocation is a common cause of occult lower extremity arterial injury.32 In the upper extremity, the brachial artery is the most commonly injured vessel followed by the radial and ulnar arteries.30 The arteries of the forearm typically do not require intervention for occlusive traumatic injury secondary to collateral flow at the palmar arch.30 Blunt brachial artery injury is most commonly seen in children in the setting of elbow dislocation or displaced supracondylar fractures.35 Injury to the axillary artery is less common but can be seen with severe anterior dislocation of the glenohumeral joint.31 Additional damage to the branch vessels of the axillary artery can also be seen with this mechanism of injury.
With respect to extremity treatment, embolotherapy is predominantly indicated for hemorrhage originating from branch vessels that do not provide distal perfusion.18,29 Pseudoaneurysms and arteriovenous fistulas, even those without active extravasation, are often appropriate indications for embolization as well.18,29 Acute injury of the larger vessels is more often treated surgically, although endovascular treatment with stent graft placement is increasing.36 Branches of the profunda femoral artery are an ideal site for endovascular treatment of acute traumatic hemorrhage because these do not provide direct perfusion to the distal extremity and are traditionally difficult to access via a surgical approach.19 The branches surrounding the shoulder and the profunda brachial artery are similarly potential candidates for TAE in the upper extremity.28,37
Familiarity with the trauma history and all available medical information including prior imaging is essential to assisting the trauma team in assessing the appropriateness and timing of angiography. The local algorithm for the treatment of the trauma patient should also be clearly understood. The interventionalist should be familiar with Advanced Trauma Life Support (ATLS) protocol and have an appropriate level of critical care support in the angiography suite.
Embolic materials used in trauma are almost always coils, microcoils, or Gelfoam (Pfizer Inc., New York, New York). Alternative large vessel occlusion devices such as the Amplatzer 4 Vascular Plug (St. Jude Medical, Inc., St. Paul, Minnesota) have seen increasing usage. Some authors have described the use of liquid embolic agents such as Onyx (Covidien, Irvine, California) and N-butyl cyanoacrylate (Trufill; Codman & Shurtleff, Inc., Raynham, Massachusetts), with the potential advantages of achieving vessel occlusion in the setting of coagulopathy and providing distal control when distal catheterization is not possible.38,39 Cost, a steep learning curve, and a potentially narrow safety profile are noteworthy considerations in the use of liquid embolic agents. Catheter-directed thrombin has also been described for the treatment of traumatic pseudoaneurysms of branches of the internal iliac artery.40 We feel that the efficacy and safety profile of autologous clot, Onyx, or glue are better choices when distal control cannot be achieved. Small particles, including Gelfoam particles, have little or no role in the treatment of acute trauma, and the use of alcohol in the trauma patient is contraindicated.18,19 Additional embolic agents that are mostly of historical interest include clipped pieces of silk suture and segments of the outer winding of a 0.035-in Bentson guidewire (Cook Medical, Inc., Bloomington, Indiana).
Periprocedural considerations should include anticipating the impact of CTA contrast within the urinary bladder. If there is suspicion for urethral injury, pelvic angiography should not be delayed and consideration should be given to delaying cystography until after the embolization procedure to avoid extravasation of contrast.19
Examination of the extremity color, temperature, and pedal pulses (radial and ulnar pulses if upper extremity access is planned) should be assessed by the interventionalist performing the procedure. Any postprocedure change should prompt query into a potential complication, such as reflux of embolic agents into the extremity vessels or access site vascular injury.
SITE OF ACCESS AND DIAGNOSTIC IMAGING
Pelvic binders vary among institutions. Some are made of fabric, and a hole can be cut through them for femoral access. Others are leather binders, which may have to be undone before obtaining access. Most pelvic embolizations can be performed via a single common femoral artery access site. If there is known unilateral pelvic or lower extremity injury, contralateral femoral access is preferred. If injuries or external devices preclude the preferred femoral access, the procedure can be performed via a brachial or even popliteal or radial artery access. More recently, dual access has been advocated to allow placement of a temporary occlusion balloon within the infrarenal abdominal aorta while pelvic or lower extremity hemorrhage is treated.7,11 We advocate the use of ultrasound guidance for all arterial punctures. When used routinely, it adds little time to the procedure and avoids the complications encountered with suboptimal access (e.g., retroperitoneal hematoma or arteriovenous fistula). This is especially important when dealing with hemorrhagic shock, as these patients already have little cardiovascular reserve and may not tolerate iatrogenic injury superimposed on traumatic injury. Ultrasound guidance also increases the likelihood that a closure device can be successfully used at the end of the procedure.
Often in the acute setting, renal function has not yet been assessed, and the risk of contrast nephropathy becomes a secondary concern. In a patient with known renal insufficiency, nonselective and selective pelvic and lower extremity angiography may be performed with carbon dioxide (CO2).
The procedure should start with nonselective angiography to quickly localize sites of hemorrhage and serves as a road map for vessel selection. Additionally, the initial imaging is useful in differentiating traumatic injury from guidewire-induced vasospasm. Fractures should prompt selective angiography at potential sites of vascular injury. For instance, the superior gluteal artery is the most commonly injured vessel with a posterior fracture pattern, and this is typically injured as it passes under the sciatic notch. It is commonly injured with open book fractures.19
Appropriate angiographic imaging of the internal iliac artery requires orthogonal selective views. The anterior division is best imaged with a contralateral oblique view of approximately 45 degrees. This can be thought of as looking at the iliac wing en face. An ipsilateral anterior oblique view is important to identify the superior gluteal artery and the internal pudendal artery. It should be stressed that, although initial nonselective pelvic angiogram is an important part of pelvic or extremity evaluation, selective angiography is crucial to achieving appropriate angiographic sensitivity.
Hemorrhage from pelvic trauma is most often associated with the internal iliac artery, but some have advocated that a selective angiogram of the external iliac artery should also be obtained.41 Although the external iliac artery is not usually a source of pelvic hemorrhage, the external pudendal, deep iliac circumflex, inferior epigastric, and the circumflex femoral arteries may be injured.1,3,42 The more inferior lumbar arteries as well as the inferior mesenteric or even the gonadal arteries are other rare sources of pelvic bleeding.
Following completion of contralateral pelvic vessel evaluation, selection of the ipsilateral common iliac artery should be obtained with a reverse curve catheter or a tightly curved catheter such as the Rim catheter (AngioDynamics, Latham, New York). Imaging of the bilateral iliac vessels should always be obtained. This avoids continued hemorrhage resulting from reperfusion of the injured vessel via anastomoses with the contralateral pelvic arteries. It also rules out additional injury, which may have been occult on initial CT imaging.
The decision to embolize should be based on angiographic findings of vessel injury but not limited to active extravasation. This aggressive approach may even include embolization of potential sources of hemorrhage identified on CT. Less selective or nonselective internal iliac embolization is recommended by some if the patient is hemodynamically unstable or if multiple sites of active extravasation are identified. Nonselective embolization of the internal iliac arteries is performed with the injection of Gelfoam, often followed by coils with the tip positioned in the internal iliac artery. Nonselective embolization is probably not appropriate for most pelvic trauma and is not appropriate for the profunda artery given the risk of ischemia. At the very least, CT will usually direct the operator to nonselectively embolize either the anterior or posterior division of the internal iliac artery.
Gelfoam, widely considered the workhorse of trauma embolotherapy, is a temporary agent that allows for potential recanalization within a few weeks. This may allow for reclaimed tissue perfusion once the acute injury has had time to heal. It is also inexpensive and readily available. Gelfoam is typically delivered as slurry. This is mixed in a three-way stopcock with a half-strength mixture of contrast and normal saline. More agitation will create a Gelfoam mixture that is less coarse. If the target of embolization with Gelfoam is distal, the Gelfoam mixture should be kept coarse. This achieves hemostasis through occlusion proximal to the end capillary bed, thereby mitigating against the complication of tissue necrosis.6,43 Gelfoam can also be delivered in the form of 2- to 3-mm pledgets termed torpedoes. This is achieved by front-loading a 2- to 3-mm piece of Gelfoam into the tip of a 1-mL syringe of half-strength contrast. The torpedoes may be soaked in contrast before injection. This can be used to achieve occlusion of larger caliber vessels and may reduce the risk of ischemia. The delivery of Gelfoam should be followed with a contrast injection after every few 0.1- to 0.3-mL injections of slurry or every few torpedoes, with the goal of sluggish flow and not complete stasis of contrast.19,29 This technique helps limit reflux of embolic agents into other branches.
The administration of Gelfoam into internal iliac artery branches is typically followed by selective branch occlusion with coils (Fig. 24.2). The theory is that the coils prevent very early recanalization of the vessel but are proximal enough to not cause ischemia. The problem with coils is that they may block access to bleeding vessels if repeat embolization is needed. If a 4-Fr or 5-Fr catheter has been advanced to the target vessel, then 0.035-in coils such as a Nester or Tornado (Cook Medical, Inc., Bloomington, Indiana) can be used. These 0.035-in coils have Dacron fibers to promote thrombogenesis. The 0.035-in coils should be pushed and not injected. Superselective catheterization of the target vessel is typically accomplished with a 0.021-in lumen 3-Fr microcatheter (e.g., Cantata; Cook Medical, Inc., Bloomington, Indiana). A high-flow microcatheter, which typically has a 0.025-in lumen, such as the Progreat Omega (Terumo Medical Corporation, Somerset, New Jersey), provides the ability to perform rapid power injection. The slightly larger inner lumen introduces concerns with compatibility of 0.018-in microcoils (e.g., Nester or Tornado) with high-flow microcatheters because coils can fold over and become wedged within larger lumen microcatheters. This will potentially make arterial branch access difficult for the operator to obtain.
Microcoils can be deployed conventionally by using a pushing wire such as the TruPush (Cordis Corporation, Bridgewater, New Jersey). Microcoils can also be injected, which may expedite the procedure, but the safety of this technique requires a stable delivery location (i.e., not close to the origin of the vessel) and low risk of refluxing Gelfoam.
The proximity of the microcatheter to the vessel origin during embolization influences the choice of embolization materials. If the catheter is near the origin of the accessed vessel, detachable coils (e.g., the Axiom Concerto coil; Covidien, Irvine, California), although often more expensive, may be considered to help prevent herniation of the coil into the parent vessel.
Attempts should always be made to advance the microcatheter distal to the site of injury. This allows for coils first to be deployed distal to the injury, followed by Gelfoam, and then additional coils proximally (the so-called Gelfoam sandwich technique) (Fig. 24.3). Gelfoam, or in very select circumstances glue or Onyx, and proximal coils without distal coiling can be used if the lesion cannot be crossed. If a traumatic pseudoaneurysm begins expanding subsequent to proximal coiling, endovascular treatment options are limited to percutaneous embolization of the additional branches that are backfilling the pseudoaneurysm. Coils should not be used directly within a traumatic pseudoaneurysm sac due to the potential for rupture as well as reexpansion.18 However, if the anatomy of the pseudoaneurysm allows, coiling can be performed across the neck of a pseudoaneurysm.
Traumatic arteriovenous fistula may occur when a vein and artery are injured simultaneously, either in the pelvis or extremities. Angiographically, this is appreciated as abnormal early venous return concurrent with the opacification of the arteries. Treatment is similar to pseudoaneurysms, with distal and proximal coil embolization of the artery, but rarely, treatment may require access through both the arterial and venous systems. Although coils can be used in this situation, there is a risk of coil migration in the venous system and subsequent migration centrally. Other options include exclusion of the fistulous connection with an arterial stent graft or use of an Amplatzer Vascular Plug.44 The Amplatzer Vascular Plug can be particularly useful for the treatment of AVF because of its low risk of migration. The Amplatzer 4 device has a 5-Fr delivery profile. It is the authors’ opinion that the use of detachable coils with simultaneous balloon occlusion of venous outflow carries a risk of massive coil migration when the balloon is deflated.
Preprocedural assessment with CT may be obtained but is not necessarily commonplace. CTA of the extremity and runoff vessels is reliably sensitive for evaluation of traumatic vascular injury and can be obtained with a single contrast bolus used for concurrent CT imaging of the chest, abdomen, and pelvis.
Although the upper and lower extremities do not differ significantly in terms of the specific methods of treatment, a few differences bear mentioning. First, a double flush technique should be used whenever the catheter is proximal to the cerebral circulation, including the vertebral and thyrocervical trunk. This is accomplished by first aspirating the catheter with a 20-mL syringe hooked up to a stopcock and then using a second 20-mL syringe to make an additional aspiration of a few milliliters. The catheter is then flushed with the second 20-mL syringe, with the stopcock turned off while actively injecting to prevent any blood accumulation in the catheter tip. Second, CO2 should not be used during evaluation of the upper extremity due to the risk of reflux in the cerebral vasculature.
The distal anastomoses of the radial and ulnar arteries via the deep and superficial palmar arches can allow embolization of an upper extremity runoff vessel. Although this is a rare clinical situation where surgical options are usually preferred, perfusion of the hand can be maintained if either the radial artery or ulnar artery is sacrificed so long as the palmar arches are sufficient. Patency of the palmar arches and collateral circulation (with Allen test) should be confirmed on physical exam as well as with angiography before intervention. If the distal anastomoses are satisfactory, embolotherapy, such as with the Gelfoam sandwich technique, can be undertaken within the radial or ulnar artery for the purpose of treating traumatic injury. Coil occlusion distal to site of hemorrhage is critical to prevent Gelfoam embolization to the digits. A single runoff vessel to the lower extremity may be sacrificed in a similar fashion so long as the other two runoff vessels are patent and there are no confounding factors such as multifocal flow-limiting stenoses in the setting of peripheral vascular disease.
Angiographic evaluation should begin with nonselective angiography through a sidehole catheter positioned within the aortic arch for the upper extremity or within the infrarenal abdominal aorta for the lower extremity. Subsequent angiographic evaluation should then be obtained in the proximal vessel, either the axillary artery or external iliac artery. As with the pelvis, initial diagnostic angiographic imaging should be carefully scrutinized for the presence of variant anatomy, vascular injury in areas predicted by preprocedure imaging, and also for potential additional sites of injury, which may have been occult on initial imaging or physical examination. Selective angiography is typically obtained with the use of a glidewire and an angle-tipped catheter such as a Kumpe catheter (AngioDynamics, Latham, New York). A Headhunter catheter (AngioDynamics, Latham, New York) is often used in the selection of the upper extremity.
Trauma is complicated by decreased cardiac output, hypovolemia, and vasospasm, but paradoxically, the injection rate and volume of contrast within the extremity may need to be increased over what is normally used. Slow flow may indicate vascular injury and vasospasm or, in the lower leg, raise suspicion for compartment syndrome.
Large vessel injury, such as the brachial artery, even if diagnosed angiographically, may be preferentially treated surgically depending on the institution. Endovascular treatment within the larger vessels is typically limited to stent graft placement. The use of stent grafts for the purpose of excluding the focal site of traumatic injury is being used more frequently.36 Embolotherapy of side branches near or in the site of injury may be required to prevent endoleak.
The branches of the profunda femoral arteries, muscular branches, and geniculate arteries are an ideal application of embolotherapy for trauma within the extremity.18 Treatment here can often be accomplished without compromising perfusion of the treated extremity. The endovascular treatment of traumatic injury within the extremity is approached in much the same way as treatment within the pelvis, but with careful attention to primary or accidental embolization of the runoff vessels.
Following embolization, the success of the procedure should be assessed with both postembolization angiography and examination of the extremities. Any change in the peripheral vascular exam should be further interrogated, with possible angiography of the involved limb. Routine postprocedure care includes keeping the accessed extremity steady, which can sometimes be challenging in a patient going to the operating room or suffering disorientation.
DEVICE AND MATERIALS
• Nonselective multi-sidehole catheters: Pigtail (AngioDynamics, Latham, New York) or Sos-Omni (AngioDynamics, Latham, New York)
• Selective catheters: Cobra (AngioDynamics, Latham, New York), multipurpose (AngioDynamics, Latham, New York), Bernestein (AngioDynamics, Latham, New York), JB1 (AngioDynamics, Latham, New York), or Headhunter for the upper extremity; Cobra, multipurpose, Bernestein, Rim, Kumpe, or Sos-Omni for the lower extremity
• Microcatheters: Cantata, Renegade (Boston Scientific Corporation, Natick, Massachusetts), high-flow Renegade, Direxion (Boston Scientific Corporation, Natick, Massachusetts), Progreat, Progreat Omega, Prowler (Cordis Corporation, Bridgewater, New Jersey)
• Embolization materials: Gelfoam in slurry or torpedo form (Gelfoam particles are not used), Nester coils and microcoils, Tornado coils and microcoils, Amplatzer occlusion devices
Rebleeding after pelvic TAE should be considered the most important adverse event because it occurs in 15% to 20% of patients and is associated with mortality increasing from 15% to 30%.2 Predictors of recurrent pelvic arterial hemorrhage include hemoglobin less than 7.5 g/dL before the procedure, more than 6 units of packed red blood cells (PRBCs) after the procedure, or a superselective embolization.2This last predictor of rebleeding is contrary to the authors’ opinion on the advantages of superselective embolization.2 Injured vessels may be in vasospasm or otherwise go undetected during angiography. These missed arterial injuries can manifest as delayed hemorrhage after resuscitation restores intravascular volume. Vasospasm or delayed migration/packing of Gelfoam may also contribute to rebleeding. We continue to advocate superselective embolization but also recommend aggressive embolization of occluded vessels, which may be due to traumatic thrombosis or vasospasm. Leaving the vascular sheath in place should also be considered when there is concern the embolization is incomplete or the patient still needs resuscitation.
Complication directly related to embolization is tissue ischemia within the treatment vascular bed or other territories secondary to nontarget embolization. This uncommon complication is manifested in pelvic embolizations as gluteal muscle necrosis, sacral skin breakdown, ischemic necrosis of the bladder wall and rectum, or necrosis of the femoral head.42,43,45–47 Nerve injury following TAE for trauma has been reported, but shear injury from the trauma itself rather than procedure-related ischemic changes may account for many of these reported pelvic TAE complications.42 Impotence in the male patient is likewise usually considered a consequence of the initial traumatic injury rather than nerve injury caused by pelvic TAE.48,49
Because of the possibility of inducing tissue necrosis, selective embolization should be performed if the patient’s hemodynamic status permits, and the patient should be adequately observed for signs of ischemic changes in the days that follow. Long-term follow-up is also required as buttock claudication may be associated with pelvic TAE but not be observed until patient is well into rehabilitation exercises.50
The most important step in reducing the risk of tissue ischemia is the appropriate choice of embolic agent. This should be small enough to effectively occlude the site of bleeding but large enough to avoid the terminal capillary bed. The use of nonpermanent embolic agents (Gelfoam) may potentially aid in reducing the risk of tissue necrosis by allowing for recanalization of the vessel over the course of a few weeks.
Nontarget embolization most often occurs when the rate of embolization is too fast and embolic material refluxes proximal to the catheter tip. Anastomotic vessels not appreciated on angiography can also contribute to nontarget embolization. Nontarget embolization of unintended pelvic vessels is typically well tolerated secondary to pelvic collateral vessels. In contrast, embolic material introduced into the outflow vessels of the extremity, or used during an intervention for the extremity, can jeopardize the perfusion of the limb. Endovascular coils, if placed within a proximal segment of vessel, can herniate into the parent vessel, thereby resulting in unintended occlusion distal to the point of herniation. Coils can also embolize through arteriovenous fistulae. The operator must have familiarity with coil retrieval if providing arterial embolization services.
Knowledge of variant anatomy is easy to forget in the high-stress setting of trauma embolization, but failure to recognize variants can lead to profound ischemia. In the pelvis, the two most important variants are (1) the persistent sciatic artery and (2) the corona mortis.
The persistent sciatic artery (PSA) is a rare congenital variant branch of the internal iliac artery that exits the pelvis through the obturator foramen.51,52 When this is present, the ipsilateral external iliac and common femoral arteries are hypoplastic. The PSA may give rise to the popliteal artery and provide most of the blood flow to the lower extremity or it may supplement the dominant femoral arterial system. In either case, it should be recognized angiographically. This can also be detected on physical exam if normal posterior tibial and dorsalis pedis pulses are present with an absent or diminished ipsilateral femoral pulse.
The corona mortis (or, more ominously translated, “the crown of death”) is an anatomic variant consisting of an obturator artery originating from the external iliac artery or an anastomotic connection between the obturator artery and either the inferior epigastric artery or external iliac artery.53,54 (It takes its name not from its reputation with interventionalists but rather from the propensity for disastrous injury it used to have during orthopedic procedures.) This can also refer to a connection between the veins of the pelvis, which is more common than the arterial variant. Close proximity to the superior pubic ramus predispose this vessel to acute traumatic injury. This vessel is a potential source of hemorrhage fed by the external iliac artery as well as a source for back bleeding subsequent to embolization targeting the internal iliac branches.
Variant anatomy within the extremity is also important to keep in mind. Important variants include a high takeoff of the radial or ulnar arteries in the upper extremity and a takeoff of the anterior or posterior tibial artery in the lower extremity.55
Attention to diagnostic images must also include the recognition of acquired variant anatomy, which is actually far more common. A patient with peripheral vascular disease may have runoff vessels supplied entirely through collateral flow from the profunda femoral artery. Although the branches of the profunda can normally be sacrificed without great consequence, in this setting, their occlusion would be a disaster.
Upper extremity angiography alone introduces a small risk of stroke. Embolization of proximal upper extremity branches raises this risk of stroke. Extreme caution must be taken to know the runoff of any proximal subclavian arteries being considered for embolization and to avoid particles and liquids when working near the vertebral arteries or spinal arteries.
TIPS AND TRICKS
• In pelvic trauma, both internal iliac arteries should be imaged even if the prior CT indicates unilateral injury.
• A compliant occlusion balloon may be placed in the infrarenal aorta at the start of the procedure to limit ongoing hemorrhage during portions of the pelvic embolization procedure.
• Extremity arterial embolization can be safely performed in profunda, muscular, or genicular branches.
• Trauma to the vessels of the lower leg or forearm, although uncommon, can be treated with endovascular occlusion so long as appropriate collateralization at the palmar or plantar arches is confirmed.
• Superselective embolization is almost always preferred, so long as the patient status will allow it, to prevent early rebleeding and reduce risk of ischemia.
• If the microcatheter is in stable position, microcoils can be injected to expedite the procedure.
• The use of high-flow microcatheters allows for power injections to be performed.
• Microcoil compatibility with high-flow microcatheter must be confirmed.
• Detachable coils are particularly helpful as the first coil (to confirm stability and appropriate packing) and as the last coil (to avoid herniation into the parent vessel).
• Renal protection may be aided by the use of CO2 during certain portions of the exam.
• A Gelfoam slurry should be kept coarse, particularly if the target is relatively less selective.
• Gelfoam torpedoes provide a more proximal particle embolization.
• Consider leaving the arterial access if hypothermia or rapid PRBC administration has caused a coagulopathy or if repeat angiography in the next 12–24 h is likely.
• Closure devices can be used in the trauma setting to avoid the time constraint of manual pressure. The use of a closure device does not preclude the use of repeat short interval access at the same site.
1. Hoffer EK. Transcatheter embolization in the treatment of hemorrhage in pelvic trauma. Semin Intervent Radiol. 2008;25(3):281–292.
2. Fang JF, Shih LY, Wong YC, et al. Repeat transcatheter arterial embolization for the management of pelvic arterial hemorrhage. J Trauma. 2009;66(2):429–435.
3. Hagiwara A, Minakawa K, Fukushima H, et al. Predictors of death in patients with life-threatening pelvic hemorrhage after successful transcatheter arterial embolization. J Trauma. 2003;55(4):696–703.
4. Jeske HC, Larndorfer R, Krappinger D, et al. Management of hemorrhage in severe pelvic injuries. J Trauma. 2010;68(2):415–420.
5. Karadimas EJ, Nicolson T, Kakagia DD, et al. Angiographic embolisation of pelvic ring injuries. Treatment algorithm and review of the literature. Int Orthop. 2011;35(9):1381–1390.
6. Cullinane DC, Schiller HJ, Zielinski MD, et al. Eastern association for the surgery of trauma practice management guidelines for hemorrhage in pelvic fracture—update and systematic review. J Trauma. 2011;71(6):1850–1868.
7. Kos S, Gutzeit A, Hoppe H, et al. Diagnosis and therapy of acute hemorrhage in patients with pelvic fractures. Semin Musculoskelet Radiol. 2013;17(4):396–406.
8. Cook RE, Keating JF, Gillespie I. The role of angiography in the management of haemorrhage from major fractures of the pelvis. J Bone Joint Surg Br. 2002;84(2):178–182.
9. Eastridge BJ, Starr A, Minei JP, et al. The importance of fracture pattern in guiding therapeutic decision-making in patients with hemorrhagic shock and pelvic ring disruptions. J Trauma. 2002;53(3):446–450.
10. Smith W, Williams A, Agudelo J, et al. Early predictors of mortality in hemodynamically unstable pelvis fractures. J Orthop Trauma. 2007;21(1):31–37.
11. Frevert S, Dahl B, Lonn L. Update on the roles of angiography and embolisation in pelvic fracture. Injury. 2008;39(11):1290–1294.
12. Starr AJ, Griffin DR, Reinert CM, et al. Pelvic ring disruptions: prediction of associated injuries, transfusion requirement, pelvic arteriography, complications, and mortality. J Orthop Trauma. 2002;16(8):553–561.
13. Miller PR, Moore PS, Mansell E, et al. External fixation or arteriogram in bleeding pelvic fracture: initial therapy guided by markers of arterial hemorrhage. J Trauma. 2003;54(3):437–443.
14. Agolini SF, Shah K, Jaffe J, et al. Arterial embolization is a rapid and effective technique for controlling pelvic fracture hemorrhage. J Trauma. 1997;43(3):395–399.
15. Biffl WL, Smith WR, Moore EE, et al. Evolution of a multidisciplinary clinical pathway for the management of unstable patients with pelvic fractures. Ann Surg. 2001;233(6):843–850.
16. Kimbrell BJ, Velmahos GC, Chan LS, et al. Angiographic embolization for pelvic fractures in older patients. Arch Surg. 2004;139(7):728–732; discussion 732–733.
17. Sarin EL, Moore JB, Moore EE, et al. Pelvic fracture pattern does not always predict the need for urgent embolization. J Trauma. 2005;58(5):973–977.
18. Bauer JR, Ray CE. Transcatheter arterial embolization in the trauma patient: a review. Semin Intervent Radiol. 2004;21(1):11–22.
19. Ben-Menachem Y, Coldwell DM, Young JW, et al. Hemorrhage associated with pelvic fractures: causes, diagnosis, and emergent management. AJR Am J Roentgenol. 1991;157(5):1005–1014.
20. Niola R, Pinto A, Sparano A, et al. Arterial bleeding in pelvic trauma: priorities in angiographic embolization. Curr Probl Diagn Radiol. 2012;41(3):93–101.
21. Suzuki T, Smith WR, Moore EE. Pelvic packing or angiography: competitive or complementary? Injury. 2009;40(4):343–353.
22. Blackmore CC, Cummings P, Jurkovich GJ, et al. Predicting major hemorrhage in patients with pelvic fracture. J Trauma. 2006;61(2):346–352.
23. Metz CM, Hak DJ, Goulet JA, et al. Pelvic fracture patterns and their corresponding angiographic sources of hemorrhage. Orthop Clin North Am. 2004;35(4):431–437, v.
24. Pinto A, Niola R, Tortora G, et al. Role of multidetector-row CT in assessing the source of arterial haemorrhage in patients with pelvic vascular trauma. comparison with angiography. Radiol Med. 2010;115(4):648–667.
25. Roy-Choudhury SH, Gallacher DJ, Pilmer J, et al. Relative threshold of detection of active arterial bleeding: in vitro comparison of MDCT and digital subtraction angiography. AJR Am J Roentgenol. 2007;189(5):W238–W246.
26. Katsanos K, Sabharwal T, Carrell T, et al. Peripheral endografts for the treatment of traumatic arterial injuries. Emerg Radiol. 2009;16(3):175–184.
27. Doody O, Given MF, Lyon SM. Extremities—indications and techniques for treatment of extremity vascular injuries. Injury. 2008;39(11):1295–1303.
28. Maleux G, Herten PJ, Vaninbroukx J, et al. Value of percutaneous embolotherapy for the management of traumatic vascular limb injury. Acta Radiol. 2012;53(2):147–152.
29. Aksoy M, Taviloglu K, Yanar H, et al. Percutaneous transcatheter embolization in arterial injuries of the lower limbs. Acta Radiol. 2005;46(5):471–475.
30. Franz RW, Skytta CK, Shah KJ, et al. A five-year review of management of upper-extremity arterial injuries at an urban level I trauma center. Ann Vasc Surg. 2012;26(5):655–664.
31. Peck MA, Rasmussen TE. Management of blunt peripheral arterial injury. Perspect Vasc Surg Endovasc Ther. 2006;18(2):159–173.
32. Levy BA, Zlowodzki MP, Graves M, et al. Screening for extermity arterial injury with the arterial pressure index. Am J Emerg Med. 2005;23(5):689–695.
33. Pieroni S, Foster BR, Anderson SW, et al. Use of 64-row multidetector CT angiography in blunt and penetrating trauma of the upper and lower extremities. Radiographics. 2009;29(3):863–876.
34. Fox N, Rajani RR, Bokhari F, et al. Evaluation and management of penetrating lower extremity arterial trauma: an eastern association for the surgery of trauma practice management guideline. J Trauma Acute Care Surg. 2012;73(5)(suppl 4):S315–S320.
35. Sidhu MK, Hogan MJ, Shaw DW, et al. Interventional radiology for paediatric trauma. Pediatr Radiol. 2009;39(5):506–515.
36. Piffaretti G, Benedetto F, Menegolo M, et al. Outcomes of endovascular repair for blunt thoracic aortic injury. J Vasc Surg. 2013;58(6):1483–1489.
37. Carrafiello G, Lagana D, Mangini M, et al. Percutaneous treatment of traumatic upper-extremity arterial injuries: a single-center experience. J Vasc Interv Radiol. 2011;22(1):34–39.
38. Mavili E, Donmez H, Ozcan N, et al. Endovascular treatment of lower limb penetrating arterial traumas. Cardiovasc Intervent Radiol. 2007;30(6):1124–1129.
39. Yoo DH, Jae HJ, Kim HC, et al. Transcatheter arterial embolization of intramuscular active hemorrhage with N-butyl cyanoacrylate. Cardiovasc Intervent Radiol. 2012;35(2):292–298.
40. Juszkat R, Zielinski M, Wykretowicz M, et al. Traumatic inferior gluteal artery aneurysm managed with emergency transcatheter thrombin injection. Cardiovasc Intervent Radiol. 2010;33(3):607–609.
41. Johnson GE, Sandstrom CK, Kogut MJ, et al. Frequency of external iliac artery branch injury in blunt trauma: improved detection with selective external iliac angiography. J Vasc Interv Radiol. 2013;24(3):363–369.
42. Auerbach AD, Rehman S, Kleiner MT. Selective transcatheter arterial embolization of the internal iliac artery does not cause gluteal necrosis in pelvic trauma patients. J Orthop Trauma. 2012;26(5):290–295.
43. Yasumura K, Ikegami K, Kamohara T, et al. High incidence of ischemic necrosis of the gluteal muscle after transcatheter angiographic embolization for severe pelvic fracture. J Trauma. 2005;58(5):985–990.
44. Shetty R, Lotun K. Treatment of an iatrogenic femoral artery pseudoaneurysm with concomitant arteriovenous fistula with percutaneous implantation of an Amplatzer Vascular Plug. Catheter Cardiovasc Interv. 2013;81(1):E53–E57.
45. Travis T, Monsky WL, London J, et al. Evaluation of short-term and long-term complications after emergent internal iliac artery embolization in patients with pelvic trauma. J Vasc Interv Radiol. 2008;19(6):840–847.
46. Takahira N, Shindo M, Tanaka K, et al. Gluteal muscle necrosis following transcatheter angiographic embolisation for retroperitoneal haemorrhage associated with pelvic fracture. Injury. 2001;32(1):27–32.
47. Suzuki T, Kataoka Y, Minehara H, et al. Transcatheter arterial embolization for pelvic fractures may potentially cause a triad of sequela: gluteal necrosis, rectal necrosis, and lower limb paresis. J Trauma. 2008;65(6):1547–1550.
48. Ramirez JI, Velmahos GC, Best CR, et al. Male sexual function after bilateral internal iliac artery embolization for pelvic fracture. J Trauma. 2004;56(4):734–739; discussion 739–741.
49. Ellison M, Timberlake GA, Kerstein MD. Impotence following pelvic fracture. J Trauma. 1988;28(5):695–696.
50. Rayt HS, Bown MJ, Lambert KV, et al. Buttock claudication and erectile dysfunction after internal iliac artery embolization in patients prior to endovascular aortic aneurysm repair. Cardiovasc Intervent Radiol. 2008;31(4):728–734.
51. Brantley SK, Rigdon EE, Raju S. Persistent sciatic artery: embryology, pathology, and treatment. J Vasc Surg. 1993;18(2):242–248.
52. Hsu WC, Lim KE, Hsu YY. Inadvertent embolization of a persistent sciatic artery in pelvis trauma. Cardiovasc Intervent Radiol. 2005;28(4):518–520.
53. Smith JC, Gregorius JC, Breazeale BH, et al. The corona mortis, a frequent vascular variant susceptible to blunt pelvic trauma: identification at routine multidetector CT. J Vasc Interv Radiol. 2009;20(4):455–460.
54. Requarth JA, Miller PR. Aberrant obturator artery is a common arterial variant that may be a source of unidentified hemorrhage in pelvic fracture patients. J Trauma. 2011;70(2):366–372.
55. Uflacker R, ed. Atlas of Vascular Anatomy: An Angiographic Approach. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006.