Related Entries
Articles Case Presentations
  • There are currently no related case presentations.
  • There are currently no related images.
Blog Posts
  • There are currently no related blog posts.
Site Search
Home > Articles > Management of Exsanguinating Pelvis Injuries


Trauma patients who present with unstable pelvic fractures have sustained a high energy injury that is commonly associated with disruption of arteries and veins resulting in major haemorrhage. Patients with pelvic fractures who present in shock have a mortality of 30-50%. When combined with injuries in other body regions such as the abdomen, the mortality rises even higher, approaching 100% in some series. However a systematic multidisciplinary approach to these injuries, directed initially only at haemorrhage control, can lead to significant improvements in survival.

Key Points

Pelvic Injury

  • An unstable pelvic injury with signs of shock should be treated as a vascular injury.
  • Haemorrhage may be from fractured bone and disrupted veins and arteries.
  • High energy trauma is associated with multi-cavity injury, and there may be haemorrhage in the chest or abdomen as well as the pelvis and long bones.


  • A multidisciplinary approach is essential. All team members must know their roles and key decisions.
  • Appropriate resuscitation manoeuvres are as important as haemorrhage control interventions.
  • Management of massive transfusion, coagulopathy and hypothermia are vital for success.
  • A damage control approach should be adopted for all these patients: do the minimum necessary to save life.

The Algorithm

Decision Nodes

Unstable Patient

This guideline applies to patients who are haemodynamically unstable and have a mechanically unstable pelvic fracture.

Haemodynamic instability is defined here as patients who are non-responders or transient responders to intravenous fluid therapy. Small (250-500ml) boluses of fluid are given and the cardiovascular response assessed. Patients are actively bleeding if there is only temporary, or no improvement in cardiovascular status. Patients who are actively bleeding should be managed with a permissive hypotension strategy until haemorrhage control is achieved. Large volumes of crystalloid should be avoided in favour of blood and clotting factors.

Whether patients are 'stable enough' to go directly to CT scan rather than enter this algorithm is determined not only by the patient's haemodynamics but also by local resource factors such as the location of CT, the speed of the scanner, and the experience of the trauma team in taking such unstable patients to CT scan. It must be recognised that although the scan may be successfully completed, the delay imparted during the scanning period may result in later death from exsanguination or to multiple organ failure due to the prolonged period of shock.

For the most unstable patients, transfer to a place where haemorrhage can be controlled is more important that specifically defining the location of the haemorrhage.

Patients who improve haemodynamically in response to small volume fluid boluses are not actively bleeding. They should proceed to definitive imaging and be fully resuscitated. This algorithm is not applicable to this group of patients 'responders'.

Unstable Pelvis

The algorithm pertains to those patients suffering from severe pelvic fractures leading to mechanical instability and vascular disruption as a consequence of the displacement of tissues at the moment of impact. Although open-book and vertical shear pattern injuries are more likely to be associated with vascular injury than other patterns, it must be remembered that in up to 40% of pelvic fractures requiring angio-embolization, the bleeding is from anterior branches of the internal iliac artery (obturator and pudendal arteries) with a lateral compression pattern injury.

Clinical examination is unhelpful and attempts to test for instability may lead to haemodynamic instability as clot is dislodged and vascular injuries exacerbated. The actual pattern of injury may be difficult to determine from the AP pelvis X-ray. When this is the case and the patient is too unstable for definitive imaging, the pelvis should be assumed to be unstable until proven otherwise.

Once the pelvic binder is on it should be left on until haemorrhage control is achieved, the patient fully resuscitated and coagulopathy corrected. Ideally it should not be removed until a more permanent pelvic fixation is applied. Patient movement, log-rolls etc should be kept to an absolute minimum to avoid displacement of the pelvic fracture and haemostatic clot.

Haemodynamic Response to Pelvic Stabilization

A pelvic binder is applied to splint the pelvis and reduce bleeding from bone ends and provide some fracture stability to allow clot to form at sites of venous haemorrhage. The binder will not control arterial haemorrhage. The haemodynamic response to the binder can aid decision-making as to the nature of the underlying vascular injury. Those patients who remain haemodynamically unstable following application of the binder probably have a significant arterial haemorrhage which will need operative or endovascular control. (This may be in the pelvis or another body cavity).

Conversely those patients whose cardiovascular status becomes normal following binder placement probably do not have a major arterial injury and can proceed to definitive imaging by CT scan. If an arterial blush is identified on the CT, patients can proceed to angio-embolisation in a more controlled manner.

Concurrent Resuscitation

Patients who are in shock, actively bleeding and have not responded to the above simple manoeuvres are in extremis and need expert care while haemorrhage control is being achieved. This includes:

  • Intubation and ventilation
  • Activation of a massive transfusion protocol with aggressive administration of coagulation factors (and minimal crystalloid)
  • Permissive hypotension prior to haemorrhage control
  • No investigations or interventions that delay or compromise early haemorrhage control.

Operating Room or Angiography Suite?

Haemorrhage from branches of the internal iliac artery are best managed by endovascular techniques such as angiographic embolisation or coil placement. Immediate transfer of the patients to an endovascular suite is therefore the ideal for these patients. However in certain situations, patients must be transferred immediately to the operating room prior to angiography:

  1. The patient has co-existant major thoracic or abdominal haemorrhage.

    Thoracic haemorrhage is usually readily identified following chest examination, chest X-ray and chest tube drainage. Major intraperitoneal haemorrhage must be excluded before transfer for angio-embolization. Again, CT is not possible due to the patient's haemodynamic status and so FAST ultrasound or Diagnostic Peritoneal Aspiration (DPA) must be used. In both cases, the aim is to exclude major intraperitoneal haemorrhage of significant volume to account for the degree of cardiovascular derangement. A small amount of intraperitoneal fluid is likely in patients with massive retroperitoneal haematoma. Thus a FAST may be positive, but not 'positive enough' to warrant a laparotomy prior to embolisation. Similarly a weakly blood-stained peritoneal lavage is not significant but easy aspiration of frank blood from the peritoneal cavity is.

    Minimal and significant free fluid on FAST

    Caution should be exercised when interpreting the results of both FAST and DPA. FAST has a high false-negative rate in the presence of pelvic fracture and retroperitoneal haematoma. A repeat scan may be of value. Neither FAST nor DPA can exclude a major abdominal retroperitoneal vascular injury (aorta, common iliacs etc). If there is any doubt, the operating room is a safer bet than the angiography suite for initial haemorrhage control manoeuvres.

  2. The patient is exsanguinating from an open pelvic fracture.

    Patients with open pelvic fractures that are bleeding externally should be taken to the operating room for packing of the external haemorrhage as well as extraperitoneal pelvic packing.

  3. Angio-embolisation is not immediately available.

    Where angioembolization is not available within an appropriate timeframe (30 minutes is suggested), patients should be transferred to the operating room for immediate haemorrhage control manoeuvres.

For all of the above 3 situations, patients will usually be transferred to the angiography suite for embolization after surgery for definitive control of intra-pelvic arterial haemorrhage. Angiographic resources should therefore be mobilised while the patient is in the operating room to avoid further delays after surgery.

Of course if there is 24-hour access to a hybrid endovascular operating room this decision node becomes almost redundant - it remains relevant only for the order in which interventions are carried out.

Operating Room Procedures

The following sequences are suggestions for the management of these injuries. There is no evidence base from which to draw on here. All procedures should be performed on a damage control only basis. The pelvic binder MUST stay on. Any pelvic packing manoeuvres will not work without some form of external stabilization in place.

  1. Pack any open wounds that are exsanguinating.
  2. Extraperitoneal packing through a lower midline incision
  3. Extension of the incision upwards and damage control laparotomy.

Intraperitoneal packing is a less satisfactory procedure as the packs are placed against clot and it is difficult to achieve effective compression of vessels against the bony pelvis.

Where angiography is not available at all, or where the patient is in extremis, attempts to reduce inflow to the pelvic vascular beds may be of value, including distal aortic compression (at the bifurcation) and/or ligation of the internal iliac arteries. This last procedure is difficult and has limited value due to the extensive collaterallisation of the pelvic arterial tree. It also precludes any subsequent endovascular procedures. Nevertheless it is an appropriate maneuvre where the patient is in extremis and angiographic is not readily available.

If the patient becomes haemodynamically normal after pelvic packing and laparotomy, and the pelvic injury is rotationally unstable only (open-book), consideration can be made to place an external fixator or symphyseal plate at this time. If the patient has not yet stabilized and there is continued evidence of active bleeding, angio-embolization is the priority.


Patients whose most likely source of major active haemorrhage is branches of the internal iliac artery (due to the pelvic fracture) should undergo angiography and endovascular control. Where the pelvis is thought to be the only source of significant haemorrhage patients are transferred directly from the emergency department to angiography. Alternatively the patient may proceed to angiography after operative control of other sites of haemorrhage.

Transfer of haemodynamically unstable patients to the angiography suite may not be usual practice for many institutions, and it is important that logistics such as the provision of anaesthesia, blood products etc are worked out in advance.

Patients must be transferred with the full trauma team in attendance, as patients are critically ill and will require complex resuscitation and monitoring during the procedure. The provision of blood and blood products is vital, as successful angioembolisation relies on an adequate haemostatic system to achieve haemorrhage control.

Definitive Imaging (CT)

All patients will ultimately require a full CT scan, to identify other injuries and to accurately define the pelvic fracture prior to definitive pelvic stabilization. Patients who need a head CT to exclude traumatic brain injury should proceed directly to CT following haemorrhage control. Where traumatic brain injury has been excluded clinically prior to intubation, the timing of definitive imaging is a clinical decision based on the patients' current physiological state and the ability to tolerate further movement and a delay to ICU.

Intensive Care

Once on intensive care, the patients should be managed as for any damage control patient. Aggressive resuscitation with blood and blood products should continue to correct coagulopathy and clear persistent signs of tissue hypoperfusion. Patients should be warmed to normothermia.

As the pelvis is still in a pelvic binder and definitive stabilization has not yet occurred, log rolling and other manoeuvres should be kept to a minimum. The degree of mobilisation allowed by the particular injury should be documented by the orthopaedic surgical teams, and a plan for definitive stabilization made when the patients is physiologically able to tolerate the procedure.

The Future

In the future, the availability of CT scanners in the resuscitation room will allow the trauma team to pinpoint the site of active haemorrhage in the unstable exsanguinating patient and allow definitive decision-making about where best to achieve haemorrhage control.

Similarly, the combination of operating room and endovascular intervention into hybrid operating suites will greatly reduce the delays to haemorrhage control and simplify decision making in these difficult patients, where the choice of operating room or angiosuite may essentially be a gamble as to where the site of active haemorrhage is.


Thomas Nau, May 29, 2008

Dear Karim,
congratulation for this algorithm. There is ongoing debate on this topic, a classic sign that we are still not doing optimal here.

In several current guidelines, the major decision is, either OR or angio suite, depending on the FAST/external bleeding.
If FAST is negative, angio is suggested.

I think we need to move on here, as it is not always so clear. We are also facing the argument of radiology to perform CTA prior to angio to see if there is and where there is an arterial bleeding, with some support in the literature. As you mentioned this discussion may be superfluous in the future with CT in the resus room, but at present most centers do not have that.

I think, we need to identify 2 groups of patients after a negative FAST and include them into the diagram:

a. suitable for angio

b. in extremis, not suitable for angio

As you mentioned, time is the important factor, and things may differ from centre to centre.

I think the key to an improved algorithm is, to identify this group b, these patients should not be going to angio they should be going to the OR.


Thomas Nau

medicallstar, June 19, 2008

I am a Paramedic in the United States and I have been reading the articles on Trauma.org for a long time now. The biggest thing I am always the most curious about is what happens once I drop a patient off. I get to b the one who is with them at the scene, cutting them out of the car or what have you. Then I drop the pt. off with the trauma team and try and get out of the way, I never get to follow up because of privacy laws (HIPPA). This article was very fascinating, I would have to say it is one of my favorites. I had a pt. last year involved in a high speed rollover motor vehicle accident in which we experienced a 1 hour and 15 minute extrication time. She had several long bone fractures, La Forte 2 fracture, as well as a severe open pelvic fracture. I was a little stressed for this woman, and I found some closure from that call by reading this article. The mindset of a Trauma Doctor is so different from a Paramedics, and I am glad it is. Thanks for everything you do!

Karim, London, UK, June 22, 2008

@Thomas:  Great comment Thomas.  As you say who is suitable for angioembolization depends not only on patient features but also on hospital resources and available expertise.  At the moment defining those patients who need to go directly to the OR is something that each trauma unit needs to write into their own guidelines, based on their own local situation.

bahtoo, July 24, 2008

Its interesting that protocols are developed for the management of the c-spine but with an injury that also has the potential for catastrophic results we have developed none. To spring or not to spring is a superfluous question now I believe.It should be “under what conditions do I assume a pelvic injury (stable or unstable is secondary at the initial primary assessment) and take steps to quickly stabilise the pelvis until injury is excluded?” Perhaps during the primary assessment we should change ‘C’ to “‘C with pelvic stabilisation”
I shudder when I still see young doctors pressing down firmly on the pelvis to see if it is unstable.


crashlann, March 08, 2009

I appreciate this article…I am a Flight Nurse in the USA and my Paramedic partner and I routinely apply our own makeshift Pelvic compression devices with a cut sheet or blanket on those pts presenting with suspected pelvic injuries. I am currently proposing a Pelvic compression device for our Medical Protocol…in fact we just got back from a call where we used one on a 11 year old. I also agree with the above comment, dont push on the Pelvis once pain or instability has been identified. Thanks.

Emily Mason, June 10, 2010

are there are references for this article??

JS, January 06, 2011

So in a hypothetical patient presenting to a DGH with no out-of-hours interventional radiology facilities, with haemodynamic instabilty, pelvic fracture (lateral compression) and bladder injury (frank haematuria), would the correct management in your view would be something like:

-hypotensive resuscitation
-pelvic binder
-CT if reasonably stable
-if no improvement -laparotomy, packing
-ICU to normalise physiology
-go back later to remove packs / fix the bladder?

I won’t say how it was actually done, but it wasn’t like that.

Second question: can a pelvic binder lead to further injury to pelvic organs, or is this urban myth?


Karim, London, UK, January 06, 2011

Basically you need to transfer this patient to somewhere that has angioembolisation capability.  At which point this transfer should happen depends on how the patient responds to the binder and small-volume resuscitation.  In general the more unstable the patient the earlier and more time-critical the transfer should be.

Extra-peritoneal packing may temporise but the patient is likely to need transfer for embolisation.

Pelvic binders should be applied to stabilise the pelvis.  They are not put on to “close” the pelvis and brute force is not necessary.  Even so I have never heard of binders damaging internal organs.


Submit a Comment

Commenting is not available in this weblog entry.