Vous êtes sur la page 1sur 4

Injury, Int. J.

Care Injured 45 (2014) 738741

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/injury

Time to pelvic embolization for hemodynamically unstable pelvic

fractures may affect the survival for delays up to 60 min
Shinsuke Tanizaki *, Shigenobu Maeda, Hideyuki Matano, Makoto Sera,
Hideya Nagai, Hiroshi Ishida
The Department of Emergency Medicine, Fukui Prefectural Hospital, Fukui, Japan


Article history: Introduction: We evaluated the relationship between survival and time from arrival to angiography for
Received 19 January 2013 hemodynamically unstable patients with pelvic trauma.
Received in revised form 10 September 2013 Methods: A retrospective review of patients admitted to Fukui Prefectural Hospital with pelvic fractures
Accepted 11 November 2013
during a 7.5-year period. Charts were reviewed for age, injury characteristics, injury severity score,
systolic blood pressure and heart rate on arrival, base decit, and the lactate concentration on arrival,
Keywords: transfusion requirement, fracture pattern, the time from hospital arrival to angiography, and the time
Pelvic trauma
spent in the angiography suite.
Results: Of a total of 140 patients, 68 patients underwent pelvic angiography and embolization. Of the
patients, 24 patients were hemodynamically unstable. The average injury severity score was 41.7. Of the
patients, 17 had major ligamentous disruption. The average time from hospital arrival to angiography
suite was 76 min. Of the hemodynamically unstable 24 patients, there were 12 deaths (50%). Patients
who were embolized within 60 min of arrival had a signicantly lower mortality rate (16 vs. 64%;
p = 0.04). There was no embolization-related complication and repeat angiography was not required in
all patients.
Conclusion: Earlier pelvic embolization within 60 min may affect the survival of hemodynamically
unstable patients with pelvic fracture.
2013 Elsevier Ltd. All rights reserved.

Introduction embolization would reduce the mortality of hemodynamically

unstable patients with pelvic fractures. This retrospective study
The management of hemodynamically unstable patients with analyzed our protocol consisting of early embolization in
pelvic fracture is a multidisciplinary challenge with high mortality hemodynamically unstable patients with pelvic fracture, and
and remains a controversial issue [14]. Haemorrhage due to evaluated the relationship between survival and time from
disruption of the surrounding venous and arterial vessels is the arrival to angiography.
leading cause of death in these patients and require prompt
therapy aimed at hemostasis [1,2].
The combination of angiography and embolization of bleeding Methods
arterial vessels has emerged as an excellent management method
[5]. Some have used angiography without invasive external We conducted a review of a consecutive series of 140 trauma
xation if haemodynamic instability persisted despite volume patients with pelvic ring fractures who were admitted to Fukui
resuscitation for 24 h [57]. Angiography within 90 min with Prefectural Hospital (FPH) from April 1, 2005, through September
external xation was reported [8]. It is unclear whether the delay 30, 2012. The patients were transported to the hospital by ground
in controlling bleeding from pelvic trauma is associated with transportation or by helicopter. The patient data were obtained
worsened outcomes. We hypothesized that earlier pelvic from the hospital records. Patients with any pelvic fractures were
included in this study. Patients who died in the emergency
department and patients requiring emergency laparotomy were
excluded. Data examined included age, injury characteristics,
* Corresponding author at: Fukui Prefectural Hospital, The Department of
Emergency Medicine, 2-8-1, Yotsui, Fukui, 910-8526 Japan. Tel.: +81 776 54 5151;
injury severity score (ISS), systolic blood pressure and heart rate on
fax: +81 776 57 2991. arrival, base decit and the lactate concentration on arrival,
E-mail address: sytanizak@yahoo.co.jp (S. Tanizaki). transfusion requirement, fracture pattern, the time from hospital

00201383/$ see front matter 2013 Elsevier Ltd. All rights reserved.
S. Tanizaki et al. / Injury, Int. J. Care Injured 45 (2014) 738741 739

arrival to angiography suite, and the time spent in the angiography gelfoam. Non-selective embolization of the internal iliac artery
suite. The fractures were classied according to the YoungBurgess was performed in patients with multiple bleeding areas. Emboli-
classication [9]. Major ligamentous disruption was dened as zation with steel coils was performed in patients with pseudoa-
anteroposterior compression type II and type III, lateral compres- neurysm, arteriovenous stula, or bleeding that was difcult to
sion type III, combined mechanism, and vertical shear. Abbreviated arrest with gelfoam. Completion angiography was necessary to
injury scale (AIS) 1990update 98 was used to calculate ISS. conrm the cessation of bleeding radiographically.

General management Statistical analysis

The treatment protocol established for all patients included an All data in the tables are presented as mean SD unless
initial evaluation and volume resuscitation in the emergency otherwise specied. Demographic discontinuous variables were
department (Fig. 1). Focused assessment with sonography for compared using the Chi Square test. Continuous variables were
trauma examination (FAST) was performed to detect hemoper- compared using Students t-test. Signicance was dened as p < 0.05.
itoneum, massive haemothorax, or hemopericardium. An ante-
roposterior chest radiograph was obtained to exclude intrathoracic Results
injury. Clinical examinations of the pelvis and anteroposterior
pelvic radiographs were used to determine whether the patient During the 7.5-year period, 140 patients with pelvic fractures
had a pelvic fracture and to assess the fracture pattern. If were treated in the FPH. 68 patients (49%) underwent pelvic
haemodynamic stability permitted radiographic workup, comput- angiography and embolization (Fig. 2). Of the patients, 24 patients
ed tomography (CT) scanning of the whole body was performed. (35%) were hemodynamically unstable. All 44 hemodynamically
Patients with arterial extravasation on the contrast-enhanced stable patients had extravasations on CT. The mean patient age was
pelvis CT underwent angiography and embolization. The CT-room 6  22 years and sex distribution was 13 male and 11 female. The
and the angiography suite in our hospital are next to the average injury severity score (ISS) was 41.7  9.1. Out of the patients,
emergency room. If patients were hemodynamically unstable 17 (70%) had major ligamentous disruption. All injuries were the
with only pelvic haemorrhage, angiography, and embolization result of blunt trauma; the mechanisms included 10 (41%) falls from a
were performed immediately. Haemodynamic instability was height, 10 (41%) pedestrians struck by motor vehicles, and 4 (16%)
dened as systolic arterial pressure <90 mm Hg after an infusion motor vehicles crashes (Table 1). Associated extrapelvic injuries AIS
of 2 l of lactated ringers and the initiation of the transfusion of the >3 were as follows: 13 (54%) intracranial injuries, 12 (50%)
packed red blood cells (PRBCs). SAM Pelvic Sling1 (Seaberg, intrathoracic injuries, 4 (16%) intraabdominal injuries, and 2 (8%)
Wilsonville, OR) was used as pelvic binder in our protocol. extremity injuries (Table 2).
Laparotomy was indicated by concomitant haemodynamic insta- Nonsurvivors of the hemodynamically unstable patients, who
bility and progressive abdominal effusion or by the existence of a underwent embolization, were signicantly associated with lower
pneumoperitoneum. GCS score and higher ISS than survivors (Table 1). There were no
signicant differences between survivors and nonsurvivors in age,
Pelvic angiography gender, injury mechanism, the presence of major ligamentous
disruption, base decit on arrival, the lactate concentration on
Angiography was performed using the femoral puncture arrival, PRBCs requirement in the initial 6 and 24 h, or associated
approach. A pelvis ush was performed at the aortic bifurcation extrapelvic injuries (Tables 1 and 2).
level after an abdominal ush. Thoracic aortography was The time from hospital arrival to angiography suite ranged from
performed in the case of severe chest trauma. Embolization agents 30 to 145 min, with an average of 76 min. The time from lesion to
included gelfoam and steel coils. Gelfoam of about 3 mm in each angiography suite ranged from 70 to 250 min, with an average of
dimension diluted in contrast medium was injected to avoid 110 min. The time spent in the angiography suite ranged from 30
permanent arterial occlusion. The signs of vascular injury, such as to 125 min, with an average of 58 min. The bleeding was difcult to
contrast extravasation, vasospasm, pseudoaneurysm, and arterio- arrest even with steel coils in one case. The catheter was not
venous stula, were reasons for embolization. Bleeding sites were positioned easily in the main trunk of the internal iliac artery
directly identied to be controlled with embolization. Selective because of severe tortuosity in two cases. Survivors had a lower
embolization of discrete bleeding sites was performed with average time from hospital arrival to angiography suite than

Pelvic Fracture Pelvic Fractures


Hemodynamically Hemodynamically
unstable stable
No angiography Angiography
72 (51%) 68 (49%)
FAST CT scan

Positive Negative Extravasation Extravasation Hemodynamically stable

Hemodynamically unstable
positive negative 44 (65%)
24 (35%)

Emergent Observe Positive angiography Positive angiography

laparotomy 24 (100%) 36 (88.2%)

Fig. 1. Treatment protocol for pelvic fractures. Fig. 2. Management algorithm of patients with pelvic fractures.
740 S. Tanizaki et al. / Injury, Int. J. Care Injured 45 (2014) 738741

Table 1
Demographic and clinical characteristics of patients with hemodynamically unstable pelvic fractures.

Characteristic Survivor (n = 12) Non-survivor (n = 12) p Value

Age (yr) 57.2  22.2 65.9  23.2 0.361

Male gender (%) 8 (66.7) 5 (41.6) 0.133
Admission SBP (mm Hg) 95.1  34.0 76.5  16.6 0.101
Admission HR (beats/min) 102.6  15.2 107.5  22.8 0.542
GCS score 13.6  0.98 9.25  4.55 <0.01
ISS 37.9  8.03 45.5  8.84 <0.05
Mechanism (%)
Fall 6 (50.0) 4 (33.3) 0.407
Pedestrians accident 3 (25.0) 7 (58.3) 0.097
MVC 3 (25.0) 1 (8.33) 0.273
Pelvic fracture pattern (%)
Major ligamentous disruption 9 (75.0) 8 (66.7) 0.653
The average time from hospital arrival to angiography (min) 63.1  23.5 89.9  28.6 <0.05
The average time from lesion to angiography (min) 94.7  14.5 125.8  49.3 <0.05
The average time of angiography (min) 55.8  29.0 60.8  25.3 0.658
Base decit on arrival (mmol/l) 3.93  4.44 5.44  4.13 0.454
The lactate concentration on arrival (mmol/l) 4.32  2.49 7.38  4.10 0.057
The PRBCs requirement in the initial 6 h (units) 10.4  6.68 10.5  3.52 0.94
The PRBCs requirement in the initial 24 h (units) 13.7  9.51 18.2  10.1 0.274
ICU length of stay (days) 6.92  3.89 3.75  1.96 <0.05

SBP, systolic blood pressure; HR, heart rate; GCS, Glasgow Coma Scale; ISS, injury severity score; MVC, motor vehicle crash; PRBCs, packed red blood cells.

nonsurvivors (63 vs. 89 min; Table 1). If the patient was in the allowing the tamponade effect of the haematoma to control venous
angiography suite within 60 min, the mortality rate was 16%; bleeding [57,11]. Some found early embolization to be associated
however, after 60 min this rate signicantly increased to 64%. with improved outcome [6]. The probability of death increased
Embolization was performed in all 68 patients who underwent approximately 1% for every 3 min of haemodynamic instability
angiography. Of these 68 patients, 60 patients (88.2%) had active that elapsed without haemorrhage control in the emergency
arterial bleeding. Angiography was positive in all 24 hemodynam- department [12]. Although the sample had small size, those
ically unstable patients received angiography. Of the 44 hemody- patients who had delayed arrival to the angiography suite because
namically stable patients who underwent angiography, 36 patients of prolonged resuscitation or unnecessary radiographic tests had a
(81.8%) had active arterial bleeding. In the other 8 patients, higher mortality than those patients who had arrived within
embolization was done because vasospasm occurred in the 60 min. Six of seven patients who arrived to the angiography suite
internal iliac artery. within 60 min survived, whereas only six of seventeen patients
There was no embolization-related complication, such as the who arrived after 60 min survived. This result emphasized the
gluteal muscle and skin necrosis, necrosis of the vesical wall, or need for emergency physician or trauma surgeon to make decision
urogenital complication. Repeatative angiography was not re- for transportation of the hemodynamically unstable patients
quired in all patients. without nonpelvic bleeding sources as soon as possible to the
Of the 68 patients who underwent angiography, there were 12 angiography suite. And also, this result emphasized that angiogra-
deaths (17%). Of the hemodynamically stable 44 patients, there phy was meaningful only in cases when it could be done as fast as
was no death. Of the hemodynamically unstable 24 patients, there an urgent laparotomy for haemorrhage control.
were 12 deaths (50%). One patient died from acute respiratory Markers for the early identication of patients likely to have
distress syndrome due to haemorrhagic shock. Mortality in three arterial bleeding are important. Fracture pattern and its relation-
patients was the result of multiple organ dysfunction syndrome ship to outcome have been examined. Some have demonstrated
failure. The other eight deaths were related to severe head injury. that patients requiring embolization were signicantly more likely
to have fracture patterns associated with major ligamentous
Discussion disruption [13]. Others showed no clear relationship between
fracture pattern and arterial bleeding [14]. The decision to obtain
Early identication and control of pelvic haemorrhage is pivotal angiography based on fracture pattern alone would lead to a very
to decreasing pelvic fracture-related mortality [5,7,10]. This low yield. We employed angiography in the face of haemodynamic
haemorrhage can originate from injury to arteries, injury to the instability, even if there were no fracture patterns associated with
venous plexus in the pelvis, and fracture bleeding. The most major ligamentous disruption.
common methods for arresting pelvic bleeding are external The diagnostic accuracy of the modern 64-slice multidetector
xation of the pelvic fracture, and angiography for the identica- computed tomography (MDCT) in detecting active pelvic arterial
tion and embolization of arterial pelvic bleeding. bleeding associated with blunt pelvic fractures has also been
Angiographic embolization is considered by most authors to be examined. Some showed that the MDCT provides relatively high
the optimal method for controlling arterial haemorrhage, and diagnostic accuracy in detecting a clinically relevant arterial
haemorrhage [15]. They concluded angiography or operative
intervention, warranted in patients with clinical signs of ongoing
Table 2
haemorrhage, even in the absence of MDCT contrast extravasation.
Number of associated extrapelvic injuries.
Some showed that the negative predictive value of contrast
Associated extrapelvic injuries Survivor Non-survivor P value extravasation on CT scans needing embolization was 98% [16].
Intracranial injuries 5 (41.7%) 8 (66.7%) 0.219 They concluded that the presence of contrast extravasation on CT
Intrathoracic injuries 6 (50.0%) 6 (50.0%) 1.000 scans was an indication for angiography, regardless of haemody-
Intraabdominal injuries 2 (16.7%) 2 (16.7%) 1.000
namic status. Others retrospectively examined CT scans with
Extremity injuries 2 (16.6%) 0 (0%) 0.139
contrast extravasation in pelvic trauma and found that nearly half
S. Tanizaki et al. / Injury, Int. J. Care Injured 45 (2014) 738741 741

of contrast extravasation on CT scans did not require embolization Conict of interest statement
[17]. They concluded that angiography may not be required in
patients with stable haemodynamics. Need for angiography in There are no potential conicts of interest. The authors report
hemodynamically stable patients with contrast extravasation from this study did not receive any outside funding.
pelvic trauma requires further study to determine its usefulness.
The mortality rate was 17% (12/68) in those patients requiring
embolization, and this was lower than that reported in the References
literature of 36% [18]. All of death was hemodynamically unstable
[1] Dalal SA, Burgess AR, Siegel JH, Young JW, Brumback RJ, Poka A, et al. Pelvic
and the mortality rate was 50% in those patients with haemody-
fracture in multiple trauma: classication by mechanism is key to pattern of
namic instability requiring embolization. The higher mortality rate organ injury, resuscitative requirements and outcome. J Trauma 1989;29:
may be attributed to the fact that these patients had more severe 9811002.
injuries. Most patients died from acute respiratory distress [2] Moreno C, Moore EE, Rosenberger A, Cleaveland HC. Haemorrhage associated
with major pelvic fracture. J Trauma 1986;26:9879.
syndrome, multiple organ dysfunction syndrome failure, or severe [3] Rotheberger DA, Fischer RP, Strate RG, Velasco R, Perry Jr JF. The mortality
head trauma, which is similar to the previous report [6]. associated with pelvic fractures. Surgery 1978;84:35661.
Various noninvasive external xation devices are available to [4] Rommens PM, Hofmann A, Hessmann MH. Management of acute hemorrhage
in pelvic trauma: an overview. Eur J Trauma Emerg Surg 2010;36:919.
provide emergent stabilization of pelvic fracture. The efcacy of [5] Papakostidis C, Giannoudis PV. Pelvic ring injuries with haemodynamic insta-
temporary external xation devices in controlling pelvic haemor- bility: efcacy of pelvic packing, a systematic review. Injury 2009;40:S5361.
rhage has not been conrmed. However, some reported that the [6] Agolini SF, Shah K, Jaffe J, Newcomb J, Rhodes M, Reed III JF. Arterial emboli-
zation is a rapid and effective technique for controlling pelvic fracture hem-
use of temporary external xation devices placed in the emergency orrhage. J Trauma 1997;43:3959.
department reduced blood transfusion needs compared with [7] Gruen GS, Leit ME, Gruen RJ, Peitzman AB. The acute management of hemo-
invasive external xation placed in the operation room [19]. dynamically unstable multiple trauma patients with pelvic ring fractures. J
Trauma 1994;36:70613.
Noninvasive external xation device was used in the present [8] Balogh Z, Caldwell E, Heetveld M, DAmours S, Schlaphoff G, Harris I, et al.
report because of the rapidity of the device placement compared Institutional practice guidelines on management of pelvic fracture-related
with invasive external xation. hemodynamic instability: do they make a difference? J Trauma 2005;58:
Preperitoneal pelvic packing (PPP) has been used as a new part
[9] Young JWR, Burgess AR, Brumback RJ, Poka A. Pelvic fractures: value of plain
of clinical pathway in controlling haemorrhage from pelvic radiography in early assessment and management. Radiology 1986;160:445
fracture [5]. Cothren et al. reported that the success rate of PPP 51.
in controlling haemorrhage was 83% in those patients with [10] Mucha P, Farnell MB. Analysis of pelvic fracture management. J Trauma
hemodynamically instability. Another study in their institution [11] Velmahos GC, Toutouzas K, Vassiliu P, Sarkisyan G, Chan LS, Hanks SH, et al. A
reported that the mortality rate was 20% in those patients who prospective study on the safety and efcacy of angiographic embolization for
underwent PPP at a median of 45 min from admission, and this was pelvic and visceral injuries. J Trauma 2002;53:3038.
[12] Clarke JR, Trooskin SZ, Doshi PJ, Greenwald L, Mode CJ. Time to laparotomy for
not similar with that of patients who underwent angiography at a intra-abdominal bleeding from trauma does affect survival for delays up to
median of 130 min (30%) [20]. Most relevant is that nobody in the 90 min. J Trauma 2002;52:4205.
packing group died from haemorrhagic shock, but 2 in the [13] Jeroukhimov I, Ashkenazi I, Kessel B, Gaziants V, Peer A, Altshuler A, et al.
Selection of patients with severe pelvic fracture for early angiography remains
angiography group. Our study showed that the mortality rate was controversial. Scand J Trauma Resusc Emerg Med 2009;17:62.
16% in those patients with haemodynamic instability who [14] Miller PR, Moore PS, Mansell E, Meredith JW, Chang MC. External xation or
underwent angiography within 60 min from arrival. If angiography arteriogram in bleeding pelvic fracture: initial therapy guided by markers of
arterial hemorrhage. J Trauma 2003;54:43743.
is available immediately, earlier angiography may be effective in [15] Mohseni S, Talving P, Kobayashi L, Lam L, Inaba K, Branco BC, et al. The
controlling haemorrhage from pelvic fracture. diagnostic accuracy of 64-slice computed tomography in detecting clinically
Complications of embolization have been reported in limited signicant arterial bleeding after pelvic fractures. Am Surg 2011;77:117682.
[16] Stephen DJG, Kreder HJ, Day AC, McKee MD, Schemitsch EH. Early detection of
case-series studies. Our study showed no embolization-related
arterial bleeding in acute pelvic trauma. J Trauma 1999;47:63842.
complication, even using bilateral internal iliac artery emboliza- [17] Michailidou M, Velmahos GC, van der Wilden GM, Alam HB, de Moya M, Chang
tion, as shown by other previous reports [11]. Gelfoam of about Y. Blush on trauma computed tomography: not as bad as we think! J Trauma
3 mm in each dimension (which is larger than the previously 2013;73:5804.
[18] Fangio P, Ashenoune K, Edouard A, Smail N, Benhamou D. Early embolization
reported 2 mm) was the occluding agent of choice [21,22]. This and vasopressor administration for management of life-threatening hemor-
may help to occlude the larger branches of the internal iliac artery rhage from pelvic fracture. J Trauma 2005;58:97884.
temporarily while leaving the smaller branches open to collateral [19] Croce MA, Magnotti LJ, Savage SA, Wood II GW, Fabian TC. Emergent pelvic
xation in patients with exsanguinating pelvic fractures. J Am Coll Surg
blood ow, as well as to avoid signicant ischaemia. The reasons 2007;204:9359.
why there was no embolization-related complication may be, that [20] Osborn PM, Smith WR, Moore EE, Cothren CC, Morgan SJ, Williams AE, et al.
larger gelfoam was used. Direct retroperitoneal packing versus pelvic angiography: a comparison of two
management protocols for haemodynamiclly unstable pelvic fracture. Injury
This study has limitations, the greatest being the small sample 2009;40:5460.
size. We had difculty in reaching many patients, because our [21] Ben-Menachem Y, Coldwell DM, Young JWR, Burgess AR. Hemorrhage associ-
institution is a rural hospital in Japan. ated with pelvic fractures: causes, diagnosis, and emergent management. AJR
In summary, the present study suggests that earlier pelvic
[22] Kaufman JA, Waltman AC. Angiographic management of hemorrhage in pelvic
embolization within 60 min may affect the survival of hemody- fractures. In: Baum S, Pentecost MJ, editors. Abrams Angiography: Interven-
namically unstable patients with pelvic fracture. tional Radiology. Boston: Little, Brown; 1997. p. 86983.