THERAPEUTIC STRATEGIES AND OUTCOME OF POLYTRAUMATIZED PATIENTS WITH PELVIC INJURIES- A SIX-YEAR EXPERIENCE
Wolfgang Ertel, M.D., FACS, Karim Eid, M.D., Marius Keel,
M.D., Otmar Trentz, M.D.
From the Division of Trauma Surgery, University Hospital Zurich, CH-8091
Zurich, Switzerland.
Legend
Figure 1: Management protocol for multiple injured
patients with pelvic ring injuries
(ORIF - Open reduction and internal fixation)
Correspondence to:
Abstract
The combination of multiple injuries and pelvic ring disruption often represents a devastating injury pattern. This study evaluates therapeutic strategies and the outcome of 174 severely traumatized patients with pelvic ring injury (ISS: 32.1 + 11.1 pts) over a 6-year period. Patients were divided into three groups according to their hemodynamic status at admission. Forty-one patients in extremis (group A; ISS: 40.1 + 11.1 pts) which required the highest resuscitation efforts revealed a mortality rate of 90%. Many of these patients underwent crash laparotomy (44%)/thoracotomy (22%), aortic clamping (22%), and pelvic packing (44%). Patients with persistent hemodynamic instability (group B; n=39; ISS: 34.2 + 9.6 pts) had a mortality rate of 26% (p<.05 versus group A) with 14 patients (36%) undergoing emergency laparotomy. Mortality rate (5%) markedly (p<.05) decreased in patients with stable hemodynamics despite a relatively high ISS (group C; n=94; ISS: 27.6 + 9.4 pts). Hemorrhage could be controlled in all patients of group B and C, while 23 out of 41 patients (56%) in extremis died due to exsanguination during the first 24 hours after injury. Thus, treatment of patients in extremis must be focused on aggressive resuscitation and surgical intervention without extensive diagnostic procedures to effectively control lethal hemorrhage.
Key words: multiple injury, pelvic injury, hemorrhage.
Introduction
Multiple injuries in combination with severe pelvic ring disruption represent a serious life-threatening trauma pattern. Though isolated pelvic injury exhibits a mortality rate between 8% to 50% dependent on the type of pelvic disruption (1-3), associated injuries further aggravate the outcome of these patients. While infection and multiple organ dysfunction syndrome (MODS) are the major complications during the late posttraumatic course, exsanguinating hemorrhage represents the predominant lethal factor during the first 24 hours (4-13).
Most treatment concepts previously reported (14-18) have focused on isolated pelvic ring disruptions. They are widely accepted and include a rapid evaluation, identification, and control of the source of major blood loss, as well as reduction and primary fixation of the pelvis. Because additional injuries require a more variable scheme of diagnostic and therapeutic approaches, the order of emergent treatments, the procedures of bleeding control, and the time point of pelvic fixation in multiple injured patients have been controversially discussed in the past (11, 14-32). Furthermore, the multidisciplinary approach which is used in many centers worldwide, may even increase the disagreements rather than providing a clear strategy to approach those patients.
This article reviews the management of 174 patients with multiple injuries combined with pelvic trauma over a six year period in an institution exclusively run by trauma surgeons. It identifies the major factors which significantly influence the outcome of those patients and provides a simple and clear protocol to effectively control severe hemorrhage during the acute period after injury.
Patients and Methods
From January 1, 1991 until December 31, 1996, 174 patients with multiple trauma including pelvic ring injuries were admitted to the emergency room of the University Hospital Zurich, a Level I Trauma Center. Severe injury was defined as at least two different system injuries one of these being one life-threatening (33), an injury severity score (ISS) > 16 points (34), a systemic trauma response reflecting systemic inflammatory response syndrome (SIRS), and the need of postoperative intensive care. Excluded were multiple injured patients with isolated acetabular fractures and patients who were referred to our hospital more than 3 days after the accident occurred. Patients` charts and ICU-protocols from 1991 up to 1993 were reviewed retrospectively for mechanisms of injury, age, sex, blood loss, transfusion requirements, incidence of shock, SIRS, adult respiratory distress syndrome (ARDS), multiple organ dysfunction syndrome (MODS), and treatment modality. From 1994 up to 1996, these parameters were recorded prospectively. Hemodynamic, respiratory, laboratory, and radiological data were collected over a surveillance time of 30 days on the ICU or until discharge from the ICU.
Definitions. Pelvic fractures were classified according to the scheme of Tile and Pennal (35). Fracture classification was obtained by a supine anterior-posterior radiograph and by inlet and outlet views, respectively. In 91 out of 174 patients (52.0%) CT-scan was additionally used for classification.
Patients were divided into three groups according to their hemodynamic status at admission: Group A included patients in extremis, with either absent vital signs or with severe shock due to torrential hemorrhage which needed mechanical resuscitation or repeatedly catecholamines despite complete blood volume replacement within 120 minutes (>12 blood transfusions/2 hr) (8). Group B included patients not able to maintain a systolic blood pressure of > 90 mmHg, a pulse of < 100 beats/minute, a central venous pressure (CVP) > 5 cmH2O, or a urine output > 30 mL/h despite adequate fluid replacement and blood transfusion over a time period of 2 h. Group C consisted of patients with stable hemodynamics and absence of clinical signs of shock.
The incidence of shock, SIRS, sepsis, ARDS, and MODS were recorded using retrospective analysis of data files. The presence of shock was evaluated upon arrival of patients in the emergency room and was given if patients showed the following clinical signs: 1) hypotension < 90 mmHg systolic blood pressure, 2) tachycardia > 100 beats/minute, and 3) the requirement of catecholamin. SIRS was present if two or more of the following conditions were met: 1) temperature > 38oC or < 36oC, 2) heart rate > 90 beats per minute, 3) respiratory rate > 20 breaths per minute or PaCO2 > 32 mmHg, 4) white blood cell count > 12,000/mm3, < 4,000/mm3, or > 10% immature band forms (36). Sepsis was diagnosed, if all criteria of SIRS were fulfilled in combination with a positive focus or a positive blood culture (37). ARDS was defined according to the criteria of Murray (38) with a score > 2.5 points. The presence of MODS was evaluated according to the criteria of Goris et al. (39) including our modifications during at least three days running (33).
Management. Evaluation of trauma pattern and resuscitation were carried out in the emergency room following Advanced Trauma Life Support (ATLS) guidelines.
Ultrasound of the abdomen and the chest were performed in the emergency room upon arrival by the trauma surgeon. Further assessment of the patient included plain X-ray of skull, chest, pelvis, cervical and thoraco-lumbar spine. Most of the patients who showed unstable (B or C type) pelvic injuries were further evaluated by pelvic CT-scan except those in extremis or with persistent hemorrhagic shock. In most patients pelvic CT-scans were carried out as part of the CT-scan for evaluation of concomitant brain, chest, and/or abdominal injuries. Scans of the pelvis were obtained at 10 mm intervals throughout the abdomen and total pelvis.
The widely variable injury pattern required a flexible therapeutic approach. The operation schedule was individually adjusted to the hemodynamic and pulmonary status of the patient and the constellation of associated injuries as well as the particular fracture pattern. Only unstable displaced (> 10 mm) pelvic ring injuries were stabilized either through external fixation or open reduction and external fixation (ORIF) dependent on the patient`s hemodynamic status. Additionally, fractures of long bones as well as unstable large joints were externally fixed or definitively stabilized as "day one surgery" (33). Indications for emergency angiography were 1) chest X-ray suspicious for traumatic aortic rupture or 2) persistent blood loss after external stabilization of the pelvis in the presence of a negative abdominal ultrasound and massive fluid requirements or a large retroperitoneal hematoma seen on the CT-scan.
Patients in extremis (group A) with exsanguinating hemorrhage due to abdominal and/or pelvic injuries underwent crash laparotomy or even resuscitative thoracotomy without further diagnostic procedures. In some of those patients, hemodynamic stabilization was attempted by subdiaphragmatic clamping of the aorta or balloon occlusion of the aorta followed by packing of the 4 quadrants and the retroperitoneum. Disruption and dislocation of the pelvic ring were reduced and maintained either by external fixation (external fixator, pelvic anti-shock clamp) or were internally reduced and fixed at the end of damage-control laparotomy.
Patients with persistent hemodynamic instability (group B) underwent primary survey and minimal diagnostics. If initial assessment with ultrasound showed no intraabdominal lesions, the unstable displaced (> 10 mm) pelvic fracture was externally reduced and stabilized. In the case of intrabdominal and/or pelvic bleeding the external stabilization of the pelvic ring was followed by emergent laparotomy.
Angiography was only performed in patients with persistent signs of blood loss despite surgical bleeding control and/or external fixation of the pelvic ring.
Patients which were hemodynamically stable after adequate fluid replacement (group C) were treated with primary and definitive ORIF depending on the degree of pelvic dislocation and associated injuries. The presence of severe brain (GCS < 8 points and intracranial lesions in CT-scan) or chest injury (multiple lung contusions) was an exclusion criteria for primary ORIF. Instead, the disrupted and displaced pelvis was closely reduced and stabilized by external fixator.
Statistical analysis. Results are presented as mean + SD. Comparisons between groups were carried out with Mann-Whitney-U-test and additional Bonferroni correction. For evaluation of potentially predictive parameters of outcome multiple regression analysis was used.
Results
The study included 111 men and 63 women with a mean age of 40.3 + 18.2 years (mean + SD; range 13 - 90) in the three patients` groups (Table 1). Fifty-three percent of the injuries resulted from traffic accidents, the majority (25.9%) of those included motor vehicles, motorcycle and bicycle (17.3%), and pedestrians (9.8%). Thirty-eight percent of all accidents resulted from fall of which more than one third were suicide attempts. Motorcycle accidents showed a high incidence of type B1 pelvic ring instabilities (46%), while fall accounted for the highest incidence of pelvic instabilities type C (43%).
Table 1. Demographic data of enrolled patients.
| number | n = 174 |
| age (years) | 40.3 + 18.2 |
| male/female | 111/63 |
| ISS (points) | 32.1 + 11.1 |
| APACHE II (points) | 1 18.3 + 8.6 |
| GCS (points) | 2 11.6 + 4.6 |
| mortality | 52 (29.9%) |
mean + SD1 at admission, 2 at scene
Forty-one (23.6%) of all patients arrived under resuscitation or severe clinical shock (group A). The average blood requirement was 33.9 + 31.7 units (range 11 to 114) (Table 2). Thirty-nine patients (22.4%) revealed persistent hemodynamic instability despite massive fluid replacement (group B) (blood units: 29.3 + 23.7; range 2 - 117) (Table 2). Only 94 patients (54.0%) of all enrolled were hemodynamically stable (group C) with 19.2 + 24.4 units (range 0 - 92) (Table 2). Patients of group A exhibited a significantly higher ISS, APACHE II, and lower GCS compared to group B and C (Table 2). In addition, group B was significantly (p<0.05) different from group C with regard to ISS, APACHE II, and GCS (Table 2).
Table 2. Severity of injury and mortality related to hemodynamics
group A (n=41) |
group B (n=39) |
group C (n=94) | |
ISS |
40.1 + 11.1 |
34.2 + 9.6* |
27.6 + 9.4** # |
APACHE II |
29.3 + 6.4 |
21.7 + 7.2* |
13.4 + 5.1** # |
GCS |
7.7 + 4.9 |
10.5 + 5.0* |
13.4 + 3.2**# |
blood units |
33.9 + 31.7 |
29.3 + 23.7 |
19.2 + 24.4** |
mortality |
37 (90%) |
10 (26%) |
5 (5%) |
mean + SD
According to the classification of Tile and Pennal (35), there were 61 type A, 38 type B1, 24 type B2, and 51 type C pelvic injuries (Table 3). The symphysis was disrupted in 44 patients. Pubic ramus fractures were found in 139 patients. The sacroiliac joint was disrupted in 42 patients, while sacral fractures were noted in 73 patients. Thirteen patients had displaced iliac wing fractures.
Table 3. Type of pelvic injuries related to hemodynamics and mortality.
classification1 |
group A (n=41) |
group B (n=39) |
group C (n=94) |
A |
10 (10)2 |
18 (4) |
33 (2) |
B1 |
9 (8) |
9 (3) |
20 (0) |
B2 |
3 (1) |
3 (2) |
18 (1) |
C |
19 (18) |
9 (1) |
23 (1) |
1 according to the classification of Tile and Pennal (35)
2 ( ) = number of non-survivors
The associated injuries are illustrated in Table 4. The involvement of other organ systems, especially of the head (110/174, 63.2% of all patients, average-AIS = 3.1 pts), chest (115/174, 66.5%, average-AIS = 3.3 pts), and abdomen (69/174, 39.7%, average-AIS = 3.9 pts) was substantial. While the incidence of head and chest injury was comparable in all three groups, patients of group A revealed a higher percentage (25/41; 61.0%) of associated intraabdominal injuries than the other two groups.
Table 4. Numbers of associated injuries related to hemodynamics.
group A (n=41) |
group B (n=39) |
group C (n=94) | |
head |
27 (65.9%) |
24 (61.5%) |
59 (62.8%) |
chest |
30 (73.2%) |
28 (71.8%) |
57 (60.6%) |
abdomen |
25 (61.0%) |
15 (38.5%) |
29 (30.9%) |
spleen |
12 |
8 |
8 |
liver |
11 |
9 |
8 |
kidney |
2 |
2 |
7 |
gut |
8 |
2 |
6 |
pancreas |
1 |
- |
- |
bladder |
4 |
3 |
6 |
ruptured diaphragm |
4 |
1 |
4 |
musculoskeletal system |
36 (87.8%) |
32 (82.1%) |
62 (66.0%) |
spine |
5 (12.2%) |
13 (33.3%) |
31 (33.0%) |
Overall mortality was 29.9% (52/174). The highest mortality was found in patients in extremis (90.0%). Thirty-three out of 52 non-survivors (63.5%) died during the first 24 hours after admission, most of them in the emergency room (Table 5). Exsanguination (n = 23; 70.0%) was the leading cause of death in these patients, while 8 patients (24.0%) died of fatal head injury, and 2 patients (6.0%) because of acute lung failure (Table 5).
Table 5. Mortality in relation to hemodynamics.
group A (n=41) |
group B (n=39) |
group C (n=94) | |
mortality |
37 (90.2%) |
10 (25.6%) |
5 (5.3%) |
died < 24 hr |
28 (75.7%) |
3 (30.0%) |
2 (40.0%) |
hemorrhage |
23 |
- |
- |
brain edema |
4 |
2 |
2 |
ARDS |
1 |
1 |
- |
died > 24 hr |
9 (24.3%) |
7 (70.0%) |
3 (60.0%) |
brain edema |
5 |
3 |
- |
ARDS |
- |
- |
1 |
MODS |
4 |
4 |
- |
pulmonary embolism |
- |
- |
2 |
In 14 out of 23 patients (60.9%) lethal hemorrhage was due to a combination of pelvic ring instability and severe intraabdominal lesions. Most of these patients suffered from severe liver ruptures (Type III or IV, according to Moore et al.(40)) (Table 6). Eight of the 19 patients (42%) who died after 24 hours after admission succumbed to death due to severe head injury, while the others died from isolated or combined organ failure (Table 5). The mean survival time of these patients was 9.3 + 9.6 days (range 2 - 36).
Table 6. Treatment of patients with fatal pelvic hemorrhage.
mechanism |
Fracture type |
ISS |
Associated abdominal injuries |
aortic occlusion |
resuscitation |
blood (units) |
Fall |
C |
34 |
Liver |
yes |
yes |
11 |
Car |
C |
41 |
Liver |
yes |
yes |
25 |
Pedestrian |
C |
25 |
- |
no |
no |
9 |
Crush |
C |
34 |
Rectum, bladder |
yes |
yes |
18 |
Fall |
C |
41 |
Spleen,liver,kidney |
no |
yes |
58 |
Car |
A |
34 |
Spleen |
no |
no |
21 |
Car |
C |
41 |
Spleen |
no |
no |
68 |
Motocycle |
B1 |
41 |
- |
yes |
no |
28 |
Motocycle |
A |
45 |
Spleen,liver,gut, diaphragm |
yes |
no |
22 |
Crush |
C |
45 |
Colon |
yes |
no |
17 |
Crush |
B1 |
34 |
Liver |
no |
no |
21 |
Fall |
B1 |
45 |
- |
yes |
no |
12 |
Motocycle |
C |
50 |
Spleen |
yes |
yes |
15 |
Car |
A |
27 |
Diaphragm |
no |
no |
20 |
Fall |
C |
50 |
Spleen, liver |
no |
yes |
24 |
Pedestrian |
C |
41 |
- |
no |
no |
20 |
Pedestrian |
B2 |
43 |
- |
no |
no |
4 |
Fall |
C |
18 |
- |
no |
no |
26 |
Car |
C |
50 |
Liver |
no |
no |
21 |
Crush |
A |
27 |
- |
no |
no |
4 |
Crush |
C |
18 |
Iliac artery |
yes |
yes |
30 |
Crush |
C |
32 |
- |
no |
yes |
33 |
Pedestrian |
C |
61 |
- |
no |
yes |
4 |
Abdominal ultrasound and the standard a.p.-view was performed in all patients, even in those in extremis (Table 5). In contrast, CT-scan was only performed in 91/174 patients (52.0%), because the other patients were in critical hemodynamic conditions. Pelvic angiography was carried out in one patient in group A (2.4 %), in 2 patients in group B (5.1%), and in 2 patients in group C (2.1%). Active bleeding was due to injury to the gluteal artery (n=2), internal iliac artery (n=1), pudendal artery (n=1), and obturator artery (n=1). Bleeding was successfully terminated in all patients through selective embolization with either gelfoam particles or coils.
Twenty-three out of 41 patients (56.1%) in extremis (group A) underwent crash laparotomy to control excessive hemorrhage (Table 7). Sixteen of these patients demonstrated type B1 (n=4) or type C pelvic (n=12) disruptions. Aortic occlusion was carried out in 9 patients (22.0%), either with aortic clamping (n=7) or balloon occlusion technique (n=2). Hemorrhage could only be controlled in one of these patients.
Reduction and fixation of the pelvic ring was carried out in patients with type B1 and C fractures with significant dislocation (symphysis disruption > 20 mm, posterior ring dislocation > 10 mm). Additionally, two patients with type B2-fractures met criteria for operative treatment, one for sacral nerve root decompression, the other showed a severely dislocated fracture of the ischial tuberosity.
Six patients in extremis (group A) with type C pelvic ring injuries were treated with external fixation devices following crash laparotomy and pelvic packing. In one of those patients the SI-joint was additionally stabilized with plate fixation. Two patients in group A with type B1 fractures were stabilized by ORIF of the symphysis after crash laparotomy and pelvic packing. In three patients with persistent hemodynamic instability (group B) the unstable pelvic fracture was primarily fixed by external devices. While one patient demonstrated stable hemodynamics, two patients needed laparotomy because of continuous bleeding. In five patients of group B primary ORIF of the pelvic fracture was carried out after bleeding control through laparotomy (Table 7).
Table 7. Acute surgical treatment in relation to hemodynamics.
Group A (n=41) |
Group B (n=39) |
Group C (n=94) | |
Crash thoracotomy |
9 |
- |
- |
Crash laparotomy |
23 |
- |
- |
Emergency laparotomy |
- |
14 |
15 |
Aortic clamping |
9 |
- |
- |
Pelvic packing |
12 |
3 |
3 |
External fixator |
5 (5xC) |
2 (1xB1,1xC) |
4 (2xB1,2xC) |
C-clamp |
2 (2xC) |
1 (1xC) |
- |
ORIF |
2 (2xB1) |
5 (4xB1,1xC) |
25 (2xA,10xB1, 2xB2,11xC) |
( ) = type of pelvic injuries according to Tile and Pennal (35).
Definitive open reduction and internal fixation was performed as "day one surgery" in 30 patients (17.2%), whereas 47 patients (27%) were operated for their pelvic fracture postprimarily (Table 8). Most of the patients with type B1 injuries were stabilized with plate fixation of the symphysis through midline incision after emergency laparotomy or using a Pfannenstiel approach. In two patients, the symphysis was fixed with tension banding following median laparotomy, while two patients were definitively treated with an external fixator (Table 8).
Table 8. Timing and technique of definitive pelvic ring fixation
B1 (n=38) |
B2 (n=24) |
C (n=51) | ||||
< 24 hr |
> 24 hr |
< 24 hr |
> 24 hr |
< 24 hr |
> 24 hr | |
anterior pelvic ring |
16 |
6 |
1 |
- |
16 |
8 |
plate fixation symphysis |
12 |
6 |
- |
- |
8 |
8 |
Tension banding |
1 |
- |
- |
- |
- |
- |
External fixation |
2 |
- |
- |
- |
2 |
- |
Posterior pelvic ring |
- |
- |
1 |
- |
12 |
12 |
plate fixation iliac wing |
- |
- |
- |
- |
- |
4 |
plate fixation SI-joint |
- |
- |
- |
- |
1 |
2 |
internal fixateur sacrum |
- |
- |
1* |
- |
7 |
5 |
screw fixation sacrum |
- |
- |
- |
- |
1 |
1 |
* decompression of sacral nerve roots.
Discussion
The combination of pelvic ring injury and multiple trauma represents a life-threatening injury with a high mortality. Because torrential hemorrhage has been identified as a major cause of death, there is widespread consensus that successful treatment of severe hemorrhage represents the hallmark for survival of these patients. Most trauma centers especially in North America developed treatment protocols for the control of hemorrhage in isolated pelvic ring injuries which include the pneumatic antishock garment (PASG), external pelvic fixation, and angiographic embolization as initial therapeutic procedures (41). Though mortality of isolated pelvic injuries has been decreased by applying appropriate combinations of the mentioned techniques (42,43), the management of pelvic injuries in combination with life-threatening organ injuries remains unsatisfactory. There is still a considerable controversy about the sequence and the timing of various emergency techniques in those patients. In this study, the management and outcome of 174 severely injured patients with pelvic ring instabilities were reviewed to better understand the characteristic of this complex injury pattern.
The high overall mortality rate of 29.9% in this series confirms the serious therapeutic problems of patients with multiple injuries combined with pelvic ring injuries. The observed overall mortality in our study is considerably higher than those reported by McMurtry (19%) (8) and by Evers (14%) (7). However, in none of these two studies patients "in extremis" have been included. This may be due to either the exclusion of those patients from the study or the fact that patients "in extremis" may not reach the emergency room in areas with long distance transport. In fact, the mortality rate in this study excluding patients "in extremis" was 11%. Despite significant hemorrhagic shock in 29% of the remaining patients, none of those patients died due hemorrhage. Moreover, the complexity and the severity of associated injuries in our study population was higher in comparison to recent studies (7,8). Forty percent of all non-survivors (21/52) had a combination of at least two associated life-threatening (AIS > 4 points) injuries of the head, the chest, and/or the abdomen. The mortality rate of patients with pelvic injuries type B or C, head, and abdominal injury requiring surgery has been reported to be > 90% (44). In our series, the mortality rate of these trauma pattern was 75%.
Considering parameters predicting mortality, hypovolemic shock caused by severe hemorrhage presents the predominant "killing factor" in multiple traumatized patients with pelvic ring injuries. Using multiple logistic regression analysis, shock (p<0.001, b -estimate = -3.64) and to a lesser degree the severity of injury (p<0.001, b -estimate = 0.25) and the age (p<0.001, b -estimate = 0.09) were independent variables which predicted survival of these patients. In contrast, the APACHE II score (p=0.31), assessed at admission, was not predictive for mortality. It is surprising that similar to the APACHE II score the type of pelvic ring injury did not predict the outcome as an independent variable. This is further supported by the fact that there was no correlation found between mortality and the type of pelvic injury. Despite a comparable distribution of B1 and C type pelvic injuries in the three patient groups, none of the patients in group B or C died due to pelvic injury. These results are in line with previous data by Poole and co-workers (41,45). Though severity of injury correlated with the severity of the pelvic fracture in their study, hospital outcome was determined by associated injuries and not by the pelvic fracture. Furthermore, no patient with isolated type B1 or C pelvic injuries (n=13) had a lethal outcome at our institution (data not shown). However, there is evidence from our statistical analysis that the hemodynamic situation is in part influenced through the type of pelvic fracture, because it was a dependent variable for survival.
The results of this study confirm previous investigations that rapid localization and control of the major source of hemorrhage is pivotal for survival of multiple injured patients with pelvic ring disruption. We advocate to immediately reduce and stabilize significant dislocation of the pelvic ring in a closed manner in the emergency room. This reduces bleeding from cancellous bone and venous sites. In patients in extremis (group A) crash laparotomy for damage control or even resuscitative thoracotomy including aortic clamping often precede external fixation of the pelvic ring (Figure 1). In patients with persistent hemodynamic instability despite adequate fluid replacement (group B) external fixation of the pelvic ring should acutely be carried out. Persistent bleeding and/or presence of intraabdominal lesions require emergency laparotomy and, if necessary, packing of the retroperitoneal space (Figure 1). Laparotomy does not only allow a rapid evaluation of intraabdominal and retroperitoneal lesions, but also their repair. In addition, through midline laparotomy, a rupture of the SI-joint can be adequately addressed through an transperitoneal approach and a symphysis disruption can be reduced and plate-fixated before closure of laparotomy (46).
Legend
Figure 1: Management protocol for multiple injured
patients with pelvic ring injuries
(ORIF - Open reduction and internal fixation)
Angiography and embolization have recently been described as alternative for controlling pelvic fracture hemorrhage. However, in the study of Agolini and coworkers (47) only 1.9% of patients with pelvic ring injury required embolization. Though they could successfully stop bleeding in all of their embolized patients, the mortality rate was 47%. Moreover, if angiography is not available in the emergency room, hemodynamically unstable patients may not be transported to the angiography unit. Due to the high degree of hemodynamic instability transportation to the angiography suite was not possible in any of the patients from group A and most of the patients from group B. Finally, angiography and embolization are time consuming procedures (50 minutes to 190 minutes (47)) even in the hands of perfectly trained experts. Thus, angiography for control of pelvic bleeding is not advisable in hemodynamically unstable patients, particularly in the presence of severe associated injuries.
Multiple injured patients with pelvic ring injury remain a challenge for the trauma and/or orthopedic surgeon. Patients at risk are those with a severe hemodynamic instability and a concomitant head, thoracic, and/or abdominal injury. These patients should be managed by an aggressive treatment protocol including emergency or even crash laparotomy, pelvic and retroperitoneal tamponade, and early fracture stabilization. In these injury pattern, the diagnostic and therapeutic approach by one surgeon ("trauma leader") seems to our opinion more effective than a multidisciplinary approach.
References