This is a recommended algorithm of the Western Trauma Association for the management of blunt hepatic injuries. Because there are no published prospective randomized trials, the recommendations are based on available published prospective, observational, and retrospective data and expert opinion of Western Trauma Association members. The algorithm (Fig. 1) and accompanying text represent safe and reasonable treatment strategies that could be followed at most trauma centers. We recognize that there will be variability in decision making and institutional and patient-specific factors that may warrant deviation from the recommended algorithm. We encourage institutions to use this algorithm as a basis to develop institution-specific protocols. The algorithm contains letters, which correspond to the text. Their purpose is to explain the critical factors affecting decisions and to guide the reader through the algorithm.1,2 References to support each step are inserted as appropriate. Specific areas where published data are lacking are mentioned as potential topics for future studies.Figure 1.: Algorithm for the management of nonoperative blunt hepatic trauma.OPERATIVE TO NONOPERATIVE MANAGEMENT OF BLUNT HEPATIC TRAUMA During the past 2 decades, treatment of blunt hepatic injuries has dramatically changed. A shift occurred from operative management emphasizing nonresectional techniques and packing in the 1980s to selective nonoperative management in the 1990s and now to nonoperative management with selective operative management. Decreased mortality associated with nonoperative management can be credited to astute observations made by trauma surgeons in concert with the use of computed tomography (CT) to aid in the diagnosis of hepatic injuries, availability of angioembolization for treatment of bleeding hepatic injuries, and appreciation of the coagulopathy of trauma.3–6Table 1 summarizes the success of nonoperative management and its associated low hepatic-related morbidity and mortality.7–11 Only studies with more than 50 adult patients with blunt hepatic trauma treated by nonoperative management were included. A recent review of the National Trauma Data Base noted that 86.3% of hepatic injuries are now managed without operative intervention,12 an even higher percentage than previous studies reported.7,9 Not surprisingly, more complications related to nonoperative management are being diagnosed. The following algorithm focuses on nonoperative management of blunt hepatic trauma.TABLE 1: Nonoperative Management and Outcome of Blunt Liver InjuriesANNOTATED TEXT FOR ALGORITHM The initial assessment of patients with suspected blunt abdominal trauma should focus on the patient’s abdominal examination, vital signs, and response to resuscitation. General principles of advanced trauma life support should be instituted, and the response to resuscitation closely monitored. Peritonitis remains an indication for exploration after blunt abdominal trauma. Although there is no well-accepted definition of hemodynamic instability, the traditionally accepted value is ≤90 mm Hg. However, recent studies demonstrate that patients are at risk for hemorrhage and death with a blood pressure ≤110 mm Hg and a base deficit of 4.13,14 To assist in early triage decisions, a hemodynamic instability score has been proposed for patients with blunt splenic trauma.2 Initial evaluation should also include an assessment of admission coagulopathy. The early use of a massive transfusion protocol, rather than the excessive use of crystalloids, is encouraged for patients with ongoing transfusion needs and has been shown to reduce mortality.15 Recent data also support the early use of plasma to packed red blood cells in a ratio approaching 1:1, although a prospective randomized trial has not yet been performed.16,17 The association between plasma- and transfusion-related acute lung injury suggests the need for further investigation into this practice.18 Hemodynamically unstable patients should have a focused abdominal sonogram for trauma (FAST) if not performed as part of their initial evaluation and if available and reliable. Hemoperitoneum diagnosed by a positive FAST examination in a persistently unstable patient should prompt operative intervention. If the initial FAST is negative, a second should be repeated as part of the secondary survey.19 Patients with persistent hemodynamic instability and a negative FAST pose a diagnostic dilemma and should not be triaged to the CT scanner, rather resuscitation should continue as the differential diagnosis of refractory shock is pursued. Patients with blunt hepatic injuries are at risk for both associated abdominal and extra-abdominal injuries.8,20 Extra-abdominal sources of exsanguinating hemorrhage include massive hemothorax and severe pelvic fracture, whereas nonhemorrhagic shock from cardiogenic (tension pneumothorax, cardiac tamponade, and myocardial contusion or infarct) or neurogenic (spinal shock) causes may be present either as the sole source or in addition to hemorrhagic sources of instability. Rather than continuing shock resuscitation in the trauma bay, an alternative is to proceed to the operating room for an exploratory laparotomy in patients at risk for imminent cardiac arrest. If hemoperitoneum remains a concern in an unstable patient with a negative FAST, a diagnostic peritoneal aspirate should be considered. A positive diagnostic peritoneal aspirate is aspiration greater than 10 mL of gross blood and warrants operative exploration in an unstable patient.21 A CT scan of the abdomen is the optimal diagnostic modality to aide in both the diagnosis and management of blunt hepatic trauma in hemodynamically stable patients.22 Liver injuries are graded per the American Association for the Surgery of Trauma liver injury scale, which was developed as part of the transition to nonoperative management and remains valid today.23 A recent report using data from the National Trauma Data Bank demonstrated that increasing injury severity was associated with increasing organ injury scale grades.12 In addition, organ-specific operative rates increased with increasing grade, although grade alone did not accurately predict the need for operation. When patients with isolated liver injuries were analyzed, 91.5% of grade I and II injuries, 79% of grade III, 72.8% of grade 4, and 62.6% of grade 5 injuries were successfully managed without operative intervention. Therefore, even high-grade injuries have a high likelihood of successful nonoperative management. The finding of a “blush,” or pooling of intravenous contrast material within the liver parenchyma, on CT scanning is indicative of active hemorrhage. Earlier studies suggested that these patients should undergo operative intervention, regardless of hemodynamic stability, though the availability of angiographic embolization may have successfully managed the hemorrhage.9,24,25 More recently, Fang et al.22,26 reported on the significance of a blush in stable patients with blunt hepatic trauma. Their initial study in 1998 followed up eight hemodynamically stable patients with pooling into the peritoneal cavity.25 Six of these patients rapidly became unstable and underwent emergent laparotomy, and the other two required delayed operations for liver-related complications. In a later study, they attempted to categorize pooling of contrast material into free extravasation with pooling into the peritoneal cavity, intraparenchymal contrast pooling with associated hemoperitoneum, and intraparenchymal contrast pooling without hemoperitoneum.26 Although the sample size was very low, all patients (6/6) with free pooling required laparotomy for hemodynamic deterioration, 66% (4/6) of patients with intraparenchymal pooling and hemoperitoneum required operation, while no patient (3/3) with intraparenchymal pooling alone required surgery or angioembolization. Finally, a larger study by this group confirmed that intraperitoneal extravasation was the most specific sign to predict the need for surgery by both univariate and logistic regression analysis.22 Although data are very limited, it seems logical to suggest that hemodynamically stable patients with free intraperitoneal extravasation undergo immediate angiography if readily available, performed in a monitored setting, and at an institution where blood products and an operative team are immediately available. More controversial is the group of stable patients with intraparenchymal contrast pooling. It is not clear from available data whether immediate angiographic embolization is required. Close observation alone with planned angiographic embolization for signs of ongoing bleeding, such as a drop in hematocrit or need for transfusion, is also an option in appropriate facilities.26–28 Neither the true incidence of pseudoaneurysm or arterio-venous fistula nor their natural history (regression or rupture) are well defined. With the current use of multichannel detector CT scanners, pooling of contrast is an increasingly common finding. A well-performed clinical trial to address the optimal management of hemodynamically stable patients with contrast pooling on CT scanning is needed. Neither the presence nor the absence of active bleeding on CT scanning absolutely predicts the need for angiography. Vasospasm at the time of CT and delayed clot lysis can both contribute to an initial lack of contrast extravasation, whereas active bleeding may be due to hemorrhage from portal or hepatic vein lesions. Bleeding seen during the arterial phase of the scan, however, confirms bleeding from an arterial source. Angioembolization is an important adjunct to management of patients managed both operatively and nonoperatively with high-grade liver injuries. Early angioembolization can decrease the need for transfusions and liver-related operations.29,30 Conflicting data exist as to whether it can improve outcome in patients requiring operative intervention.20,31,32 Unless angiography is immediately available, most would consider preoperative angiography only for stable patients with pooling seen on CT scanning. There are several reports, however, suggesting that angiography be used as an extension of resuscitation in patients with ongoing resuscitative needs.33,34 This practice cannot be advocated except in selected centers. Carillo et al.35 proposed criteria for nonoperative management of hepatic trauma to include: hemodynamically stable (not defined) patients with liver injuries diagnosed on CT scan, hepatic-related transfusion limited to 4 units of blood, and absence of other abdominal injuries that required exploration. The hepatic-related transfusion limit has not been verified. The morbidity of ongoing transfusions in an otherwise hemodynamically stable patient versus the morbidity of hepatic surgery remains to be well defined. Transfusion requirement in the first 24 hours postinjury has been shown to predict the development of liver-related complications.36 Operative intervention is still required for the patients sustaining blunt hepatic trauma, primarily related to hemodynamic instability on presentation. Christmas et al.37 confirmed higher liver-related morbidity and mortality for the patients undergoing operative intervention. When surgery is indicated, management should focus on cessation of bleeding, applying the principles of damage control surgery and hemostatic resuscitation.38 Different from operative intervention for blunt splenic injuries, bleeding from blunt hepatic injuries may actually be exacerbated by operation. In fact, Richardson et al.3 suggested that the primary reason for the decrease in hepatic-related mortality over the past several decades is the shift to nonoperative management. A detailed description of operative management is beyond the scope of this text but will be a future topic for critical decisions. Not surprisingly, because more aggressive nonoperative management is being pursued, more liver-related complications are being diagnosed. Although routine follow-up CT scans are not necessary, persistent systemic inflammatory response syndrome, abdominal pain, jaundice, or an unexplained drop in hemoglobin should prompt an evaluation by CT scanning.39 Complications are primarily related to the grade of liver injury and the need for transfusion.36 Reported complication rates range from 0% to 7% when all grades are considered, but can be as high as 14% when only high-grade injuries are considered. Paramount to the successful management of hepatic complications is a multimodality treatment strategy to include endoscopic retrograde cholangiographic embolization (ERCP) and stenting, transhepatic angioembolization, and image guided percutaneous drainage techniques. Despite these advances, operative intervention still plays a role. When patients not requiring laparotomy within the first 24 hours after injury were examined, complications that required delayed operative intervention included bleeding, abdominal compartment syndrome, and failure of percutaneous drainage techniques.36 Delayed hemorrhage is the most frequent, although still rare, postinjury complication.9,36,38 “Late” bleeds from blunt hepatic injuries generally occur within the first 72 hours postinjury.36 Management principles discussed earlier should be applied and may include angioembolization or operative stabilization. A hepatic or perihepatic abscess appears on CT scan as a focal collection with gas bubbles or a fluid collection with an air fluid level. The incidence is low and can usually be managed by percutaneous catheter drainage, though operative drainage may still be needed for failures. Biliary complications include biloma, biliary fistula, bile leak, and bile peritonitis. Approximately, one-third of liver-related complications are biliary in nature with an overall incidence of approximately 3%, although higher rates have been reported.28,36,40 Typically, biliary complications present in a delayed fashion in patients with grade 4 injuries.36 A biloma results when bile leaks into the hepatic parenchyma, increasing pressure leads to necrosis, and eventual formation of a biloma.41 Management consists of percutaneous catheter drainage. It is not known whether all asymptomatic bilomas require treatment. Bile peritonitis, defined as peritoneal and systemic signs of inflammation, typically presents several days after injury. Patients develop systemic inflammatory response syndrome in response to devitalized liver tissue rather than sepsis.42 Although laparotomy remains an option, drainage can be safely and effectively performed by laparoscopy.28,43–45 Although these signs may prompt consideration of a missed bowel injury, the incidence is very low even in patients with high grade injuries.36 Although most peripheral biliary leaks will seal without treatment, continued high output biliary drainage may warrant adjunctive ERCP to aid in healing. Ponsky and coworkers46 suggest stenting rather than spincterotomy for more effective resolution of biliary leaks. Neither the optimal duration of leak before ERCP nor the duration of leak before operative consideration has been studied. Continued observation is necessary, as multiple and recurrent late complications occur.28 The Pittsburgh group recently reported the safety of hepatic resection in the management of complex liver injuries.47 Although patients in this series underwent primary operative therapy and the majority were not major resections, their excellent results suggest that delayed resection for necrotic or devitalized hepatic tissue or major intraparenchymal bile leaks may be an option in select cases at institutions with the appropriate hepatic expertise.