Hepato-pancreato-biliary trauma

Introduction

Trauma care in general has been revolutionized by the increased availability and accuracy of cross-sectional imaging and interventional radiology. The ‘trauma CT scan’ is an integral part of most units’ major trauma protocols and specialized trauma centres depend on 24-hour availability of endovascular embolisation. As a result, injuries are diagnosed quickly, often allowing detailed assessment of severity and direction of treatment strategy with traumatic haemorrhage frequently controlled without surgery.

As the diagnosis of hepatopancreaticobiliary (HPB) trauma has become faster and more accurate, the general surgeon will be obliged to manage and determine initial treatment strategies for a growing number of patients with, predominantly, liver trauma.

Accurate diagnosis of such trauma has led to an increased confidence in non-operative management (NOM) of isolated liver trauma, and this has become increasingly tolerated as experience has grown. Furthermore, interventional radiology is a crucial pillar of care for these patients and can augment non-operative as well as operative surgery and has become the platform for HPB trauma care.

Within this chapter the aims are to describe the presentation, assessment and treatment options as well as the outcomes related to these injuries and their treatment. Although some injuries such as liver trauma are common and there is good evidence available for scrutiny, some rarer aspects of HPB trauma are less well reported and the available evidence as well as the authors’ experience are discussed.

Liver trauma

Epidemiology

The liver is a commonly injured abdominal organ. With the increase in motor vehicle use and also the increased use of sensitive imaging modalities, both the occurrence and detection of liver injuries has greatly increased since the 1990s and is reported at levels of up to 36% of all abdominal injuries.

Mechanism of injury

The mechanism of liver injury is dichotomised into blunt and penetrating injury and is critical when determining the treatment strategy. Blunt mechanisms of abdominal trauma most commonly include motor vehicle collisions, falls from height and assaults. Blunt trauma in European centres is responsible for 85–90% of liver injuries and will be the focus of this chapter. In isolated liver injuries, the proportion of blunt liver injuries is less, reported in a major study as 64.5%. ^, Mortality rates of all liver injuries are reported between 14% and 16%. Due to the liver’s size and position as well as its relatively fragile consistency, it is vulnerable to blunt abdominal trauma. The high vascular make-up of the liver makes injury a potentially catastrophic situation and must be carefully considered by an experienced trauma team, often in consultation with the HPB specialist team.

Liver injury grading

The American Association for the Surgery of Trauma Organ Injury Scale (AAST-OIS) grading classification of liver trauma is a validated and commonly used system to quantify liver injuries.

In recognition of the increased reliance of computed tomography (CT) scanning to diagnose liver injury and also its significant influence on management, the AAST 2018 version update ( Table 19.1 ) incorporates radiological findings and evidence of vascular injury into the grading system. This important inclusion therefore identifies and prioritises the presence of an identifiable vascular injury over more significant parenchymal compromise and any evidence of active bleeding detected on CT equates to at least a grade 3 injury. This is in recognition of ongoing bleeding as a predictor of failure of NOM, and mortality.

Table 19.1

Liver injury scale—2018 revision

AAST grade	AIS severity	Imaging criteria (CT findings)	Operative criteria	Pathologic criteria
I	1	– Subcapsular hematoma < 10% surface area – Parenchymal laceration < 1 cm in depth	– Subcapsular hematoma < 10% surface area – Parenchymal laceration < 1 cm in depth – Capsular tear	– Subcapsular hematoma < 10% surface area – Parenchymal laceration < 1 cm – Capsular tear
II	2	– Subcapsular hematoma 10–50% surface area; intraparenchymal hematoma – ≤ 10 cm in diameter − Laceration 1–3 cm in depth and ≤ 10 cm length	– Subcapsular hematoma 10–50% surface area; intraparenchymal hematoma < 10 cm in diameter – Laceration 1–3 cm in depth and ≤ 10 cm length	– Subcapsular hematoma 10–50% surface area; intraparenchymal hematoma – < 10 cm in diameter – Laceration 1–3 cm depth and ≤ 10 cm length
III	3	– Subcapsular hematoma >5 0% surface area: ruptured subcapsular or parenchymal hematoma – Intraparenchymal hematoma > 10 cm – Laceration > 3 cm depth – Any injury in the presence of a liver vascular injury or active bleeding contained within liver parenchyma	– Subcapsular hematoma > 50% surface area or expanding; ruptured subcapsular or parenchymal hematoma – Intraparenchymal hematoma > 10 cm – Laceration > 3 cm in depth	– Subcapsular hematoma > 50% surface area; ruptured subcapsular or intraparenchymal hematoma – Intraparenchymal hematoma > 10 cm – Laceration > 3 cm in depth
IV	4	– Parenchymal disruption involving 25–75% of a hepatic lobe – Active bleeding extending beyond the liver parenchyma into the peritoneum	– Parenchymal disruption involving 25–75% of a hepatic lobe	– Parenchymal disruption involving 25–75% of a hepatic lobe
V	5	– Parenchymal disruption > 75% of hepatic lobe – venous injury to include retrohepatic vena cava and central major hepatic veins	– Parenchymal disruption > 75% of hepatic lobe – Juxtahepatic venous injury to include retrohepatic vena cava and central major hepatic veins	– Parenchymal disruption > 75% of hepatic lobe – Juxtahepatic venous injury to include retrohepatic vena cava and central major hepatic veins

Vascular injury is defined as a pseudoaneurysm or arteriovenous fistula and appeals as a focal collection of vascular contrast that decreases in attenuation with delayed imaging. Active bleeding from a vascular injury presents as vascular contrast, focal or diffuse that increases in size or attenuation in delayed phase. Vascular thrombosis can lead to organ infarction. Grade based on highest grade assessment is made on imaging, at operation or on pathologic specimen. More than one grade of liver injury may be present and should be classified by the higher grade of injury. Advance one grade for multiple injuries up to a grade III.

The World Society of Emergency Surgery (WSES) guidelines for trauma management grades liver injury as follows: grades 1-2 are minor injuries, grade 3 is moderate and grades 4-5 are severe (with the caveat that haemodynamic instability of any grade is classified as severe). This incorporates not only the anatomical and radiological findings but also the overall clinical stability of the patient and therefore the clinical context in which to determine a management strategy.

Minor hepatic injuries:

WSES grade I includes AAST grade I-II hemodynamically stable either blunt or penetrating lesions.

Moderate hepatic injuries:

WSES grade II includes AAST grade III hemodynamically stable either blunt or penetrating lesions.

Severe hepatic injuries:

WSES grade III includes AAST grade IV-VI hemodynamically stable either blunt or penetrating lesions.
WSES grade IV includes AAST grade I-VI hemodynamically unstable either blunt or penetrating lesions.

Assessment of liver injuries

Clinical assessment of liver trauma follows the same assessment protocol as for any trauma evaluation. Patients are evaluated as per the Advanced Trauma Life Support (ATLS) guidelines according to ABCDE approach. Clinical evaluation of circulatory status is carefully performed and response to fluid resuscitation is assessed.

Haemodynamic stability is a critical component of management of liver trauma as this is a crucial factor when determining the grade of liver injury and an appropriate management strategy. This is not a stand-alone measurement and involves serial assessment and evaluation in the context of clinical interventions and response to treatment. A recent review concluded that there is no universal consensus between care providers of the definition of haemodynamic instability. Overall, the most common parameters quoted were systolic blood pressure, heart rate and response to fluid resuscitation, although the precise cut-off of these parameters at which a patient is deemed to be haemodynamically unstable was not universally defined.

This highlights the importance of recognition of difference in opinions regarding the classification of patients as haemodynamically unstable and if this has been achieved with or without intervention. As haemodynamic instability is a critical component of liver trauma assessment and care, what constitutes ‘instability’ needs to be defined, if not universally then between individuals’ care providers. Changes in patient status and interventions performed to achieve stability must be recognized and managed accordingly in the appropriate locations.

Investigation

Ultimately detailed imaging is required to fully examine the abdominal cavity if a liver injury is suspected. Focused Assessment with Sonography for Trauma (FAST) scanning can be performed in the emergency room, although the sensitivity of this to detect intracapsular bleeding, ongoing bleeding or small bleeds is compromised with reports of injury being picked up in less than 40% of liver injuries compared with CT scanning.

The mainstay of imaging and investigation to diagnose and guide management of liver trauma has for around 20 years been CT scanning. ^, Importantly, CT scanning will provide a complete overview of injuries to other abdominal organs or in other compartments. With arterial and portovenous phases, one can identify the extent of injury as well as evidence of active haemorrhage from the liver. This provides the detail to determine if emergency interventional radiology is indicated. Furthermore, evidence of biliary complication can also be estimated from the initial CT pictures. Timely CT scanning is therefore a requirement in order to safely assess and manage liver injuries.

Evidence of laceration and haematoma as well as evidence of active bleeding can all be detected by CT which, along with the clinical interpretation of haemodynamic stability contribute to assign a severity grading to liver injury. Furthermore, injury to major hepatic veins and portal structures can also be identified which can aid surgical planning if appropriate.

CT scanning is reported as having 83.3% positive predictive value (PPV) for liver injuries with a negative predictive value (NPV) of 100%. When compared with FAST scanning in a retrospective study looking at 226 trauma patients with liver or splenic injury, FAST scanning recorded a PPV and NPV overall of 99.5% and 83.3%, respectively. However, a 20% rate of missed injuries in FAST scans was reported compared with CT findings, with these cases having injuries or free fluid that was not detected by FAST. This was particularly evident in lower-grade solid organ injuries. A negative FAST scan is therefore not reassuring in patients suffering major trauma.

Magnetic resonance imaging (MRI) scanning is not normally indicated in the acute stages but will be useful at a later evaluation of suspected biliary complications.

Management

Evolution of practice

The management of liver trauma has developed significantly over the past two decades. Major liver injury has historically been managed by exploratory laparotomy and surgical control of haemorrhage, but the development of selective non-operative management (SNOM) has completely changed the management approaches for these injuries.

A report of liver injuries over a 25-year period from 1975 to 1999 identified significant changes in the management of liver trauma over this time. There has been an evolution of practice away from surgical intervention for liver trauma in the 70s and 80s and increased use of NOM, packing and re-looking as part of damage control surgery and latterly, angiography and embolisation of bleeding vessels. This was also shown alongside a reduction in mortality rates associated with liver trauma overall and a general acceptance and favouring of NOM for the majority of cases. ^,

Selective non-operative management

The mainstay for the majority of blunt liver injury is SNOM. This requires the presence of a high-quality CT scan with quantification of the liver injury and confidence of absence of arterial bleeding that requires embolisation. Additionally, SNOM should be undertaken in haemodynamically stable patients following initial resuscitation without suspicion of further abdominal injuries that require laparotomy A critical component of NOM is the use of embolisation of bleeding liver arterial vasculature. This is an established adjunct to conservative management. Patients undergoing SNOM should be monitored closely in a critical care environment with on-site access to a surgeon experienced in managing liver trauma as well as access to on-site 24-hour interventional radiology capacity.

Definition of non-operative management

The precise NOM protocol will vary according to individual unit. The underlying premise is quantification of all underlying injuries to ensure no other indication for surgery has been missed. Once this has been established, close monitoring in a critical care environment is crucial. ^, Continuous observation of hemodynamic stability is required and also regular assessment of haemoglobin levels and liver function tests. This period of close monitoring will be variable but normally can be reduced after 48 hours if there are no signs of deterioration. The average length of stay of all liver injuries undergoing SNOM in a large trauma registry review of isolated blunt liver injuries was 2 days. The recommendation being that patients with stable haemoglobin levels and normal abdominal examination can be discharged safely. Some units favour repeat imaging by way of further CT scan or contrast-enhanced ultrasound scan, but the benefits of this are uncertain. The critical component of NOM is the capacity to rapidly perform laparotomy if there is deterioration and for immediate access to interventional radiology and angiography with embolisation should active bleeding be encountered.

For patients with minor liver injuries, SNOM is the standard treatment recommendation. For patients with higher grade of injury 4-5, then particular consideration should be made to the potentially rapid deterioration from a secondary bleed. With this in mind, the logistical consideration of transport, availability of IR, operating theatres and the surgical team needs to be considered in case rapid intervention and change from conservative management is required. This will mean transfer of patients to tertiary HPB units for observation and readiness for rapid operative or IR intervention.

Outcomes and failure of SNOM

Failure of NOM is rare in low-grade isolated liver injuries.

The vast majority of such cases are managed with SNOM with studies reporting over 90% of cases of blunt low-grade trauma, without other indications for laparotomy, being managed non-operatively. Failure rates are reported as low as 3% in some series , although prospective data report failure of SNOM at up to 17% . The reasons for failure are predominantly secondary bleeding, but biliary complications and missed injuries can manifest thus mandating laparotomy. SNOM demands close monitoring, and evidence of haemodynamic instability, haemoglobin drop or signs of infection suggests that repeat imaging is required and reconsideration of the continued appropriateness of NOM.

SNOM is more likely to be successful due to the de-selection of the majority of major hepatic vessel/retrohepatic caval injuries not surviving to reach a medical facility. The majority of survivable blunt injuries involves damage to the liver parenchymal vessels which will stop bleeding by tamponade and therefore are likely to be successfully managed with simple observation.

SNOM in severe liver injury

When looking at NOM in minor liver injuries, NOM is accepted as the first management option. When the extent of liver injury is severe, or classed as grades 3–5, then the evidence for SNOM is less well established.

Saqib performed a systematic review of NOM in severe liver injury. No randomized trials were identified but of the included eight studies, five reported outcomes for SNOM in high-grade liver trauma. SNOM in these patients resulted in a success rate of 92.4% of 210 patients and a mortality rate of 3.3%. These figures were superior to the surgically treated patients with mortality figures of 15.8%. The author of this review raises the important issue of lack of randomisation and the inherent bias between the two groups in terms of haemodynamic instability, namely the definitions of haemodynamic stability, and also the almost exclusive inclusion of unstable patients in the surgically managed groups. Nevertheless, it does provide evidence that the extent of injury alone is not an immediate indication for surgery, rather, the overall stability and concern for other injuries. A further study from Italy reported a 100% success rate of SNOM with grade 1 patients and a 90% success rate in patients with grade 5 liver injuries, and of those requiring laparotomy, none of these was for bleeding supporting the feasibility of this approach.

The use of transarterial embolisation (TAE) in NOM in severe hepatic injury is also less well established than for minor injuries. A systematic review by Virdis et al. reported outcomes of 659 patients with severe liver injury undergoing TAE with a 6% failure rate, 28% morbidity rate and 5% mortality rate. Morbidity reported included bile leak (5.7%), abscess (6.8%), ischaemia (8.6%) and biloma (2.8%).

Melloul et al. in 2015 conducted a systematic review of severe blunt hepatic injuries—namely grades 3–5 and identified just under 5000 patients from 12 studies. A success rate of 92% was reported in patients undergoing SNOM from eight studies, with up to 5% requiring TAE and 4% going on to record biliary complications and a 90-day mortality of 5%. This review also comments on a further group of patients with severe liver injury and active bleeding at the time of presentation. Five studies reported TAE performed initially in this group with a success rate of 94% in these patients. However, up to 30% of this group subsequently required laparotomy for abdominal compartment syndrome (ACS) or biliary complications, and the mortality rate was reported at up to 10%.

Similarly Gaski et al. in 2018 addressed this issue with a SNOM protocol that included selective TAE in severe liver trauma with signs of active bleeding only, as opposed to a previous protocol of mandatory embolisation in the severe injury grade group. They reported a 0% failure rate in severely injured patients undergoing this approach, with 12% requiring TAE and a mortality rate of 15%. This was compared with a mortality rate of 46% for severe liver injuries requiring laparotomy. This protocol incorporated damage control resuscitation principles and the major haemorrhage protocol was only initiated in patients with a systolic blood pressure of less than 90 mmHg and when signs of ongoing active bleeding were present, and reported a liver-related death from haemorrhage of 4%.

These trials suggest that with the appropriate use of NOM and selection of patients with haemodynamic stability, as per DCR criteria, even in the most severely injured livers with signs of ongoing bleeding, the use of TAE can not only avoid laparotomy but this approach can also be associated with improved mortality rate and low failure rates of NOM. The caveat to this is that there are no randomised trials supporting this and there is always the risk of an operative comparative group selecting the sickest and most unstable patients with highest risk of death. Viewed together with the available data for SNOM, a randomised controlled trial is not likely to be possible in this population group.

Based on the current evidence, the WSES has published guidelines on the NOM of blunt hepatic injury with findings as described above with the strongest evidence base supporting:

Non-operative approach should not be undertaken for haemodynamically unstable patients, or those with diffuse peritonitis
A laparotomy is not indicated if patient is haemodynamically stable without peritonitis
A contrast CT scan is mandated to establish severity of liver injury
The severity of hepatic injury is not an absolute contraindication to NOM
TAE can be considered in actively bleeding but haemodynamically stable patients
SNOM can only be trialled where there is capacity for close monitoring and emergency surgery.

Non-operative management in penetrating trauma

Penetrating injury, less common in Western Europe than in USA, South Africa and other areas, mainly include gunshot wounds (GSWs) and stabbing incidents. ^, While SNOM is accepted initial management for the majority of blunt hepatic injuries, its role in the management of penetrating liver injuries is less well established. However, evidence is available to suggest that in selected patients NOM can provide satisfactory results and prevent non-therapeutic laparotomies.

The selection criteria for SNOM in penetrating liver injuries requires haemodynamic stability of patients at admission and also the absence of peritonitis due to hollow viscus perforation—more likely following penetrating trauma, and also the ability to perform serial accurate clinical examination without the masking effect of intoxication.

Data from the USA have reported that 33% of all penetrating liver injuries were selected for NOM; 28% of all penetrating liver injuries, including 23.5% of all severe injuries, were managed successfully non-operatively. A recent study identified that 16.2% of all penetrating liver injuries were amenable to being managed successfully with SNOM. Only one patient selected for SNOM failed due to undetected hollow viscus injury. Navsaria et al. reported in 2015 a success rate of 91.4% for patients undergoing SNOM in liver GSWs from a large prospective trauma series from South Africa. More recently, the same group published outcomes of a larger series of liver GSWs, of which 71.7% of the cohort required emergency laparotomy. Of the 54 patients selected for SNOM, 3 patients required surgical intervention due to development of peritonitis and thus representing a 94.4% success rate for SNOM in carefully selected complex liver GSWs.

These figures support the potential success for SNOM in penetrating trauma. However, caution must be exercised in this group as the data reveal high degrees of patient selection to achieve very low failure rates of SNOM in their population. The data are from high-volume trauma centres in the USA and South Africa, and a high level of experience in managing severe penetrating abdominal injuries is required to achieve good outcomes in this challenging group.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here