A treatment protocol for abdomino-pelvic injuries

Background Abdomino-pelvic injuries often present a challenge for the emergency department. Although literature reports several protocols on the treatment of abdomino-pelvic injuries aiming at defining the most advisable treatment line, optimal treatment is still controversial. This paper describes a protocol that has been used to treat abdomino-pelvic injuries in our hospital since 2002. Materials and methods In literature different protocol of abdomino-pelvic injuries are described and comparing them most of the difference are the timing of CT scan, the angiography and the laparotomy when treating a lesion of pelvic ring. If patient is haemodynamically instable and presents a lesion of pelvic ring our protocol suggest the simplest and fastest stabilization (pelvic external fixator) in emergency room and delay exam such as CT scan as second level exam. In the presence of an abdominal injury, with a positive focused assessment with sonography for trauma test, the first step should be a pelvic ring stabilization, as laparotomy decreases the abdominal pressure and reduces the tamponade effect on the retroperitoneum. According to presented protocol the angiography is not be a first choice treatment. This protocol was applied to 58 cases of abdomino-pevic injury with unstable pelvic lesions from October 2002 to December 2005. Mean injury severity score was 27.2 (CI 24.1–30.3). Results Five patients (8%) died, three due to haemorrhagic shock and two due to pulmonary embolization. Four patients (6.9%) had a partial or complete cauda equina syndrome, four patients (6.9%) complained of mild incontinence, whilst 1 (1.7%) complained of urinary retention with multiple cystitis. Two patients (3.4%) with retention and multiple cystitis, had a malunion and a painful non-union of the fracture. Seven patients (12.3%) had neurological impairment: 5 (8.6%) sciatic nerve palsy, 1 (1.7%) lumbosacral root lesions in a C2-type fracture and there was one case (1.7%) of inconstant lumbago with sciatic pain. Twelve patients reported different levels of sexual dysfunction (20.7%). Conclusions Although validation with a larger cohort is required, our preliminary clinical data are similar to, or better than, those reported in the most recent publications on this question, suggesting that this protocol could well reduce both the mortality rate and the long term complications of abdominopelvic injuries.


Introduction
As abdomino-pelvic injuries must be cared for by a multidisciplinary team, made up of an orthopaedic trauma surgeon, a general surgeon, an anaesthesiologist, a radiologist and an angiographist (and, at times, also a urologist and a gynaecologist), this pathology often presents a challenge for the emergency department [4].
Although the presence of all these specialists allows for complete treatment, it may also create confusion and lead to the deferring of proper decisions [5,33]. Such delays are, however, not acceptable with these injuries, as patients are frequently in critical conditions and require a rapid, correct diagnosis and therapy. Statistics have placed these injuries in the third place as cause of death in motorcycle accidents and the mortality rate of unstable pelvic ring fractures is as high as 20% [10]. That is why many authors [3,6,7,14,27,37] are of the opinion that this kind of trauma should be immediately referred to specialized centres where diagnostic protocols and treatment are used.
All the protocols [3,6,7,14,27,37] reported in literature are based on the ''damage control orthopaedics (DCO)'' [18,34]: this means that any intervention should be rapid and minimally traumatic focusing on haemorrhage control and on other life saving measures. The surgical burden on the immune response and that caused by the primary injury, are considered critical factors that have a direct effect on the patient's clinical course. Sub-clinical consequences of the initial trauma and/or those due to surgery may manifest themselves as abnormalities in organ function, leading to Multiple Organ Dysfunction syndrome (MODS). That is why definitive surgical treatment should be kept to a minimum in multiply injured patients i.e. complex reconstructive operations should be postponed until such times as the patient is haemodynamically stable and in a better general condition.
There are several treatment options that should be considered for the emergency haemostasis of pelvic fractures, such as: the pelvic sling, arterial inflow arrest, external fixation devices, internal fixation, direct surgical haemostasis, pelvic packing, pelvic angiography and embolization. Only external devices that can be easily applied can be used effectively [11,12,17,21,24,28,31,37]. These devices create a tamponade effect against ongoing bleeding by reducing the intrapelvic volume and also restore stability, bone contact and favour blood clotting. Should the bleeding continue despite the application of the external fixator, then pelvic packing should be considered, as should tamponade of the areas, or temporary aortic compression, in the presence of multiple massive bleeding points. Complex abdominal procedures should be avoided in the presence of pelvic haemorrhage [2,9]. A major spleen rupture usually necessitates splenectomy. In liver injuries, attention is paid only to major vessels and hepatic tamponade is applied [19]. Bowel injuries are clamped and covered, leaving definitive treatment until such times as the haemodynamic situation has been stabilized. As angiographic embolization is both time consuming and inhibitive to dynamic assessment and further treatment, it is not usually indicated in this population. However, in cases where haemodynamic stability with volume replacement can be achieved, but ongoing pelvic haemorrhage is suspected (expanding haematoma), it is a good practise to add angiography to the treatment protocol [9]. The management of the pelvic fracture should be conceived as part of the resuscitative effort. Immediate external fixation of the unstable pelvis with pelvic packing to control pelvic haemorrhage is a practical approach in both the borderline and ''in extremis'' patient. Angiographic embolization can be recommended in more stable patients. Any benefits of immediate fracture stabilization are obtained at an early stage i.e. a reduction in blood loss and, therefore the need for transfusion/s and/or a reduced risk of developing systemic complications.
The guidelines reported in literature have advocated the principles shown in Table 1 [2,9,19]. Therefore, in an effort to optimise the emergency department resources, reduce the mortality and morbidity rate of this trauma (and, consequentially the high costs to society), a team of orthopaedic surgeons, anaesthesiologists and general surgeons applied these principles to form the basis of a treatment protocol for abdomino-pelvic injuries. This Protocol has been routinely applied at the Orthopaedic and Traumatologic Hospital of Turin (CTO Hospital) since October 2002 to time of writing.

Materials and methods
A review of international literature on abdomino-pelvic injuries and pelvic ring fractures was carried out through Pub Med with the following medical subject heading (MeSH): blunt abdominal trauma, pelvic fracture, pelvic trauma, pelvic injury, pelvic ring disruption terms. Case reports and small series revisions were excluded and only articles in the English language, or published after 1995 were included. The resulting data formed the basis of our protocol.

Clinical management
First step is the ABCDE (airway, breathing, circulation, disability, and exposure/environment control) assessment, according to the advanced trauma life support (ATLS) guidelines [9] and includes a haemodynamic stability evaluation. In the presence of any one, or a combination of, the following criteria a patient is considered haemodynamically unstable: a systolic blood pressure below 90 mmHg, a heart beat rate above 110 bpm, or in the presence of clinical signs of insufficient organ perfusion (oliguria, small vessel vasoconstriction, or mental confusion not due to intoxication or head trauma) [8,13,15,25] even after ATLS resuscitation and infusions.
The ''in extremis'' patient with massive blood loss and hypovolaemic shock is immediately taken to the operating theatre for a ''blitz laparotomy'' (i.e. clamping the aorta below the renal vessels).

The haemodynamically unstable patient
In the case of a haemodynamically unstable patient who is not in shock, only the easiest and fastest exams are performed, without moving the patient from the emergency room (ER) table i.e. an antero-posterior (AP) pelvis X-ray and a focused assessment with sonography for trauma (FAST) test [26]. Depending on the results of these examinations, four scenarios may be defined.
1. Both X-ray and FAST are negative: in this case other sources of bleeding have to be considered and the FAST is repeated within 1 h. 2. The X-ray is positive for an unstable pelvic fracture (Tile B or C) and FAST is negative: in this case the pelvis is quickly stabilized by an external fixator or pelvic clamp, before any other manoeuvre is made. 3. The X-ray is negative but the FAST test is positive: this patient is taken to the operating theatre for a laparotomy, which is performed according to the principles of the damage control surgery. 4. Both the X-ray and FAST are positive: in this scenario, again, the pelvis is quickly stabilized by an external fixator or pelvic clamp: the easiest stabilization, which can be performed directly in the emergency department without fluoroscopy, is to insert two pins into the iliac crest. If the patient is still haemodynamically unstable after pelvic stabilization, or if the FAST is highly suspicious for an abdominal organ rupture, a laparotomy is carried out. This sequence is possible only if a well trained orthopaedic surgeon performs the pelvic stabilization within a few minutes, otherwise a laparotomy, pelvic packing and treatment of any organ injuries must be done first.
Haemodynamic stability must be re-assessed after each step. Should the patient still be unstable, but other sources of bleeding can be excluded, then an angiography must be done. It is advisable to repeat the FAST test in cases 1 and 2, as some parenchymal lesions do not bleed heavily immediately and may give a first negative ultrasonography (US).

The haemodynamically stable patient
If the patient is haemodynamically stable, then a little more time may be spent on other examinations. It is, therefore, essential to monitor the vital signs continuously. Also in this case the first exams are an AP pelvis X-ray and a FAST test. If the US is positive, then it is necessary to ask for a CT of the pelvis and abdomen, as the CT has a higher specificity than FAST and can offer more information as to the pelvic fracture. A laparotomy and/or a pelvic ring stabilization may be done, depending on the results of the CT. Should both be required, then the bone takes priority due to the reasons explained before. Table 1 The guideline principles with a high level of evidence [25,26,27,37] 1 CT is recommended for the evaluation of haemodynamically stable patients with equivocal findings on physical examination, associated neurological injury, or multiple extra-abdominal injuries. Under these circumstances, patients with a negative CT should be admitted for observation. In haemodynamically stable patients the CT is a complementary diagnostic modality 2 Exploratory laparotomy is indicated in haemodynamically unstable patients with a positive FAST. In haemodynamically stable patients with a positive FAST, follow-up CT allows for a non-operative management of select injuries [9] 3 A negative FAST should prompt follow-up CT for patients at high risk for intra-abdominal injuries (e.g., multiple orthopaedic injuries, severe chest wall trauma, and neurological impairment) 4 Patients with a major pelvic fracture with signs of on going bleeding after non-pelvic sources of blood loss have been ruled out should be considered for pelvic angiography and possible embolization A definitive stabilization (internal fixation with sacroiliac joint screws, symphysis plate etc.) [20,22,23,32,35,36] may be performed if: (1) a surgeon well versed in pelvic fracture is available (2) the general conditions of the patient are not critical (3) there is a negative US and a B or C type fracture. The proposed algorithm is summarized in Fig. 1. Figure 2 shows an example case.

Patients
A clinical retrospective study was set up to support the proposed protocol. Fifty-eight cases (37 males/21 females) of abdomino-pelvic injury with unstable pelvic lesions were treated, according to the aforementioned described protocol, from October 2002 to December 2005. Most percentage of the cases was motorcycle accidents (43%), followed by car accidents (27%), falls from a significant height (16%) and other causes (14%). According to the Tile's classification, there were 37 type B lesions (horizontal instability) and 21 type C (global instability). The average age was 43 years (18-76) and the average followup was 15.5 months (range 7-38). Thirty-five (60%) patients were haemodynamically stable and 55 (95%) required blood transfusion. There were 26 associated lesions: six spleen ruptures, three liver ruptures, three diaphragm lacerations with liver herniation, three bladder ruptures, four urethral partial lesions, two urethral disjunctions and five cases of severe brain injury.
Mean injury severity score (ISS) was 27.2 (CI 24.1-30.3). All the pelvic lesions were treated by external fixation; laparotomy was performed in 14 cases and angiographic embolization in four.
All the patients included in the study gave their informed consent. The study was authorized by the local ethical committee and was performed in accordance with the Ethical standards of the 1964 Declaration of Helsinki.

Results
There was an 8.6% mortality rate: five patients died, three due to haemorrhagic shock and two due to pulmonary embolization.
Four patients (6.9%) had a partial or complete cauda equina syndrome, due to multiple damage of the sacral roots due to sacral fracture. Four patients (6.9%) complained of mild incontinence, whilst 1 (1.7%) complained of urinary retention with multiple cystitis.
Two patients (3.4%) with retention and multiple cystitis, had a malunion and a painful non-union of the fracture. Correction of the deformities (osteotomy and internal fixation) partially resolved the urinary problems.
Twelve patients reported different levels of sexual dysfunction (20.7%): in this cohort three female patients complained of dispareunia due to a mal-united fracture of the ileopubic branch. Two wound infections (3.4%) of the Pfannestiel approach were recorded and were resolved by topical medication.
All haemodynamically unstable patients had surgical stabilization of the pelvis within two hours from the arrival in the emergency department. Thirty-six (62%) of the lesions (15 type B and all the type C fractures) underwent further surgery at an average of 4 days from the first stabilization.

Discussion
Several protocols on the treatment of abdomino-pelvic injuries have been published in an effort to define the best sequence of action, depending on the different scenarios. However, these protocols are of difficult implementation in clinical practice as the resources available to deal with this pathology differ from hospital to hospital.
The clinical protocol herein reported was established on the evidence-based knowledge in abdomino-pelvic injury Fig. 2 A 35-year-old man, involved in car accident, reported an unstable pelvic fracture (tile C1.1), femoral fracture, proximal humeral fracture and a mesenterial lesion. Pelvic fixation was performed a X-ray in anteroposterior view, b in outlet view and c in inlet before laparotomy d so as to take advantage of the best resources available in our hospital by co-ordinating them in the emergency department. We are of the opinion that this protocol is easily adaptable to most trauma centres, as the first steps are based on standard diagnostic examinations and therapeutic procedures. Indeed, more complex procedures, such as angiography, are required only in the very last steps. Table 2 shows a comparison between the most commonly used protocols.
The mortality rate, obtained in our study, compared favourably with those reported in literature, and showed a decrease of mortality respect other papers even if correlated with injury severity score. In fact most of reported case [1,8,14] showed an ISS of about 29 points on average, just slightly higher then 27 points obtained in this study but included in the calculated confidence interval.
Higher mortality rate was reported by Agolini [1] but his paper evaluated patients with a mean ISS of about 38 points; while other authors did not indicate the ISS [16,37].
Although a validation with a larger cohort in a prospective study is advisable, these preliminary clinical data are encouraging, as they are similar to, or better than, those reported in updated literature, suggesting that this protocol may be effective in reducing the mortality rate.