The aim of this review is to update the management of patients diagnosed with non-trauma HPB emergencies in the context of the ACS service.
General assessment
Similar to trauma, the initial evaluation of patients presenting with a HPB emergency should include simultaneous diagnosis and therapy. This concurrent rapid assessment and treatment is particularly important for patients who present with sepsis. A detailed clinical history of the acute event, including a focused past medical history (i.e. history of gallstones, pancreatitis, duodenal ulcer/NSAID use, and/or cancer) and complete physical examination are crucial. These details may suggest the likely diagnosis, determine the severity of the acute event, and guide both immediate and subsequent treatments. It is important to note that most patients present with an inflammatory and/or septic complication of a previously known disease, as opposed to a completely de novo etiology. Thus, patients presenting with acute cholecystitis or another complication of gallstones typically have a known history of symptomatic cholelithiasis. By contrast, patients suffering from pancreatic diseases generally develop symptoms after an acute new event.
The first step in caring for these patients requires a direct assessment of the severity of the acute event itself. Septic shock and acute bleeding represent the most common causes of hemodynamic compromise and must be addressed immediately. These methodologies include intravenous fluid resuscitation, early initiation of antimicrobial therapy, and blood product transfusion as needed. It is important to highlight that these patients often present with nausea and vomiting, dehydration, acute kidney injury, electrolyte imbalances, anemia, and/or coagulation abnormalities. A recent systematic review of 78 trials showed that resuscitation with colloids does not reduce the risk of death, nor improve survival [
9]. It is also more expensive than resuscitation with crystalloids [
9]. Another prospective randomized trial showed that patients with severe sepsis undergoing resuscitation with Ringer’s solution had a decreased 90-day mortality and renal-replacement therapy rate when they were compared with patients who received Hydroxyethyl starch 130/0.42 [
10]. Thus, it is recommended that patients with HPB emergencies undergo active and early resuscitation with crystalloids [
9].
Numerous scoring models have been created to determine the severity of the acute HPB event [
11]. These include, but are not limited to, the acute and physiology and chronic health evaluation (APACHE) III and IV, the simplified acute physiology score (SAPS) 3, and the mortality probability model (MPM) III. These scores were recently compared in 2596 patients with the aim of determining their value in predicting mortality [
12]. The discriminatory performances of APACHE III and IV were similar, but also superior to both SAPS 3 and MPM III scores. Despite superior calibration and discrimination amongst surgical patients in particular, the APACHE III or IV assessments are not overly simple to determine in the initial resuscitation phase. It is also clearly important to obtain focused liver and pancreatic function, as well as liver enzyme tests during the evaluation to guide resuscitation and narrow the differential diagnosis.
Once effectively resuscitated, patients should undergo diagnostic imaging tests to rapidly determine the etiology and guide further treatment. The type of study required will depend on patient status and clinical suspicion. These may include bedside and/or formal abdominal ultrasound (US), hepatobiliary scintigraphy (HIDA), computed tomography (CT), endoscopic retrograde cholangiography (ERCP), magnetic resonance (MR) cholangiography and/or endoscopic ultrasound (EUS).
Severe acute pancreatitis (SAP)
Acute pancreatitis is a common disease that affects approximately 200,000 people per year in the United States [
47]. It was classically defined by the Atlanta guidelines in 1992 as “an acute inflammatory process of the pancreas with variable involvement of regional or distant organ/systems and high levels of pancreatic enzyme in blood and/or urine” [
48]. Although the differential diagnosis for causes of pancreatitis is broad, choledocholithiasis remains the most common [
49]. SAP is observed in approximately 20% of patients, and depends on the presence of organ failure and local complications (necrosis, pseudocyst, and walled off pancreatic necrosis). The dominant definition of SAP includes subsets of patients with organ failure, necrotizing pancreatitis without organ failure, and infected pancreatic necrosis with organ failure [
2,
50]. Given that this definition considers groups with vastly different prognoses, it has been recently revised by a group of experts [
50]. It is now based on local and systemic determinants of severity. The
local determinants include pancreatic and/or peripancreatic necrosis (P-PN). These 2 entities have been included together and represent all non-viable tissue (solid or semisolid) that does not have a radiologically defined wall. Thus, pancreatic necrosis is defined as any area of non-enhancement on contrast-enhanced CT and peripancreatic necrosis includes every heterogeneous peripancreatic collection on CT until proven otherwise. P-PN may be considered infected when gas bubbles are observed on CT scan, or when a culture obtained by percutaneous or surgical techniques is positive [
50]. It should be noted that the routine use of broad-spectrum prophylactic antibiotics has altered the bacteriology of secondary pancreatic infection in severe acute pancreatitis, from predominantly gram-negative to gram-positive bacteria without changing the rate of beta-lactam resistance or fungal super-infection [
51].
Systemic determinants refer to distant organ failure, which is defined for 3 systems on the basis of the worst measurement in a 24-hour period or when the creatinine is ≥ 2 mg/dl, PaO
2/FiO
2 is ≤ 300, or inotropic agents are required.
Transient failure is defined as failure of the same organ/system for less than 48 hours, while
persistent failure implies more than 48 hours. Based on this new consensus grading, acute pancreatitis is described as [
1]
mild (P-PN and organ failure are absent), [
2]
moderate (sterile P-PN and/or transient organ failure), [
3]
severe (either infected P-PN or persistent organ failure), or [
4]
critical (infected P-PN with persistent organ failure) [
50].
Despite guideline recommendations, every patient should receive personalized treatment based on the severity of their disease and the direction of a multidisciplinary team. The surgical treatment of patients with SAP has evolved dramatically and includes open, laparoscopic, radiologic, and endoscopic techniques of debridement and drainage [
52,
53]. These approaches may be used alone, or in combination [
54]. Despite this long list of potential techniques to remove necrotic tissue, the more dominant issue for the ACS surgeon is one of timing. It is clear that early debridement increases blood loss, morbidity, the number of operative interventions, and mortality across all groups of patients with SAP [
55]. More specifically, almost every patient should be physiologically supported without major intervention until the 28-day mark. Patients with SAP follow a predictable pattern of early SIRS and potentially multi-organ failure as a result. This interval observation is often misinterpreted as sepsis requiring treatment with antimicrobial therapy. Within the first 7 to 10 days, very few of these patients have infected necrosis (and therefore require antibiotics). The most common exceptions to this rule of delayed (28-day) intervention remains patients with concurrent ischemia of the bowel or gallbladder [
56]. As a result, it is ischemia of these 2 organs that must be ruled out if a critically ill patient decompensates, as opposed to focusing on the status of the pancreas itself (infected or not).
Once the patient is stabilized and the pancreatic necrosum is mature, operative therapies may include both minimally invasive (laparoscopic cystgastrostomies and debridements, utilization of percutaneous drains as access guides for rigid scope debridement, step-up procedures, endoscopic transmural debridements) and open (transperitoneal, retroperitoneal-flank) approaches [
47,
57-
59]. The best choice amongst these options is based on patient anatomy and the specific location(s) of the necrosum.
The recent publication popularity in percutaneous techniques as treatment for pancreatic necrosis deserves specific mention. CT-guided drainage followed by repeated irrigation procedures in the context of ever increasingly larger drains placed by involved and committed radiologists may improve the clinical course in up to 75% of patients. It has also been shown to resolve the necrotic collection in 45% of cases. This technique is incredibly labor intensive however, and therefore not available in most centers.
In addition to the role of surgery, significant literature exists with reference to the role of prophylactic antibiotics, early ERCP decompression in persistent biliary pancreatitis [
60,
61], type of nutrition [
62], role of octreotide, and probiotic prophylaxis. The use of prophylactic antibiotics has been evaluated in numerous underpowered prospective randomized trials and meta-analyses [
63-
66]. These 7 RCTs are best summarized by stating that there is no good evidence to support the routine use of prophylactic antibiotics in patients with SAP [
67,
68]. Furthermore, the general recommendation is to stop all antibiotics if they have been previously started in this scenario.
The optimal type of nutrition (enteral vs. parenteral) has also been evaluated in multiple prospective randomized trials. A systematic review that included 348 patients from 8 trials showed that enteral nutrition decreased the risk of death (OR 0.50, 95% CI 0.28 to 0.91), multiple organ failure (OR 0.55, 95% CI 0.37 to 0.81), systemic infection (OR 0.39, 95% CI 0.23 to 0.65), operative interventions (OR 0.44, 95% CI 0.29 to 0.67), local septic complications (OR 0.74, 95% CI 0.40 to 1.35) and length of hospital stay (reduced by 2.4 days) [
69]. More importantly, in patients with SAP, enteral nutrition decreased the risk of death (RR = 0.18, 95% CI 0.06 to 0.58) and multiple organ failure (RR = 0.46, 95% CI 0.16 to 1.29), suggesting that patients should receive enteral over parenteral nutrition [
69]. In many patients, regardless of the extent of necrosis on CT, oral ingestion of a regular diet is well tolerated. If this fails, progression to nasogastric followed by nasojejunal tube feeding as needed is recommended.
The role of early ERCP in the context of choledocholithiasis has also been evaluated in prospective studies [
61,
70-
74]. Amongst 153 patients in a multicenter prospective study, patients were divided into 2 groups (with and without signs of cholestasis) [
75]. Although ERCP was associated with fewer complications than the observation group (25% vs. 54%, p = 0.02) in patients with signs of cholestasis, mortality was not significantly lower (6% vs. 15%, p = 0.2). Additionally, ERCP neither reduced complications (45% vs. 41%, p = 0.8) nor mortality (14% vs. 17%, P = 0.7) in patients without cholestasis, suggesting that ERCP should be indicated only in selected patients with
persistent cholestasis [
61,
74].
Finally, trials evaluating probiotic use reported that prophylaxis did not reduce the risk of infectious complications, but actually increased the risk of mortality in patients with predicted SAP [
76]. Similarly, another trial of 302 patients with moderate to SAP who received either octreotide or placebo had similar rates of mortality, complications, duration of pain, surgical interventions, and length of hospital stay. This finding suggests that octreotide should not be used in SAP [
77].