Recommendations
Urgent
Assess haemodynamic status immediately looking for early signs of fluid loss.[18][48]
Early administration of intravenous fluids is of critical importance in all patients with acute pancreatitis, however mild.
Evidence suggests it reduces the risk of organ failure and death.
Beware systemic inflammatory response syndrome (SIRS) and/or multi-organ failure - these are the biggest risk to life in the first week.[18][48][49]
Assess for signs of organ dysfunction, particularly cardiovascular, respiratory or renal.
Consider intensive care unit (ICU) transfer for any patient who has SIRS or early signs of organ failure.
Arrange emergency endoscopic retrograde cholangiopancreatography (ERCP) within 24 hours for any patient with gallstone pancreatitis who has concurrent cholangitis.[18][48][96]
Acute pancreatitis is associated with high rates of morbidity and mortality. The risk of death in patients with infected pancreatic necrosis is as high as 30%.[48]
Key Recommendations
Initial supportive treatment
Initial treatment consists of early goal-directed fluid resuscitation with a crystalloid fluid plus adequate pain control.[8][18][48][97]
Establish oral feeding as soon as the patient can tolerate it; do not keep patients nil by mouth for any longer than is absolutely necessary (e.g., vomiting, ileus).[8][18][96][97]
Beware the traditional dogma of ‘pancreatic rest’. Current evidence shows early oral feeding leads to better outcomes, regardless of disease severity.[97][98]
In patients with severe disease who cannot feed orally, use enteral tube feeding (either nasojejunal or nasogastric) rather than parenteral nutrition.[8][97]
Enteral feeding is associated with lower rates of organ failure, pancreatic necrosis, and death.
Do not give prophylactic antibiotics. There is no evidence to support their use in patients without clear evidence or clinical suspicion of infection. This holds true for patients with sterile necrosis and those with severe disease.[8][18][48][96][97]
Gallstone pancreatitis
Arrange emergency endoscopic retrograde cholangiopancreatography (ERCP) within 24 hours for any patient who has concurrent cholangitis. Suspect this based on Charcot’s triad (jaundice, fever and rigors, right upper quadrant pain).[18][48][96]
[ 
]
ERCP is probably indicated in gallstone pancreatitis with bile duct obstruction. Prior bile duct imaging with endoscopic ultrasound (or MRCP) can prevent unnecessary ERCP procedures.[48]
[ ]
Ensure cholecystectomy is performed during the same admission for any patient with mild gallstone pancreatitis.[49][96][97]
In severe gallstone pancreatitis, cholecystectomy is usually delayed for at least 6 weeks until local inflammation has resolved.[49]
Alcohol-related pancreatitis
Ensure the patient has support from an alcohol abstinence programme, starting with a brief intervention during the initial admission.[97]
Local complications
Do not request contrast-enhanced abdominal computed tomography (CECT) scanning to detect local complications in the first 3 to 4 days after presentation. Pancreatic necrosis often takes >72 hours to develop so early CECT is unreliable.[48][71][93][96]
Most patients with sterile necrosis can be managed conservatively.[48]
Some patients with infected necrosis will recover with solely intravenous antibiotics and supportive care.[18][48][63]
Where further intervention is needed, this should be delayed if at all possible until the necrosis has become walled off (4 weeks).[18][48][49][99]
The step-up approach is the preferred strategy for treating infected pancreatic necrosis. Start with catheter drainage (percutaneous or endoscopic) and only step up to necrosectomy if drainage is unsuccessful.
Minimally invasive necrosectomy techniques are preferred to open surgery although further research is needed on different subgroups of patients.
Assessing severity
Organ failure is of paramount importance in assessing severity[2]
During the first 48 hours, use SIRS and patient risk factors to predict severity.[18][48]
Never classify a patient as having mild disease until at least 48 hours after onset of symptoms. Many patients who go on to develop severe disease present without signs of organ failure or local complications.
Around 50% of patients will have predicted severe pancreatitis but only half of these will go on to develop moderately severe or severe disease.
During admission, assess severity based on persistent SIRS and the absence/presence of local complications.[48]
Severity scoring tools such as APACHE II and Glasgow have limited value and are not generally recommended by evidence-based guidelines.[18][48][51]
Acute pancreatitis has early and late phases and deaths occur in a biphasic pattern:[18]
Early phase (<2 weeks) - early mortality is caused by SIRS/multi-organ failure.
Late phase (>2 weeks) - late mortality is caused by local complications, in particular infected pancreatic necrosis.
The main treatment goals are to:
Provide supportive treatment to reduce the risk of progression to organ failure and/or local complications
Treat the underlying cause
Manage local complications.
Initial treatment for any patient with suspected or confirmed acute pancreatitis is supportive care with fluid resuscitation, oxygen, pain control, an antiemetic, and nutrition support.
Early assessment of predicted severity is important.
Guidelines recommend using SIRS-based criteria rather than more complicated scoring systems such as APACHE II or Glasgow.[18][48][97] For more information, see Risk stratification under Diagnosis Recommendations.
Fluid resuscitation
Early and adequate fluid replacement is the single most important step in initial treatment.
Administer intravenous fluids to all patients with suspected or confirmed acute pancreatitis, even those with mild disease.[18][48]
Check local protocols for specific recommendations on fluid choice. Evidence from critically ill patients in general (not specifically people with acute pancreatitis) suggests that there is no difference in benefit between normal saline and a balanced crystalloid (such as Hartmann's solution [also known as Ringer’s lactate], or Plasma-Lyte®), and therefore either choice of fluid is reasonable.[100][101]
Be aware that large volumes of normal saline as the sole fluid for resuscitation may lead to hyperchloraemic acidosis. This has particular relevance in acute pancreatitis where low pH may trigger premature trypsinogen activation and exacerbate abdominal pain. If the patient has had a large volume of saline infused, it is important to monitor sodium and chloride.
Some guidelines specifically recommend use of Ringer’s lactate (a balanced crystalloid) for patients with acute pancreatitis.[18][48] There is weak evidence of an anti-inflammatory effect with Ringer’s lactate; however, high-quality evidence on choice of fluid specific to patients with acute pancreatitis is lacking.[29]
Start intravenous fluid therapy, titrating to one or more of the following indicators or end-organ perfusion:[48][97]
Patients with acute severe pancreatitis should be catheterised to monitor urine output.
Evidence suggests early fluid resuscitation reduces the risk of organ failure and death.[102][103]
Failure to give enough fluids is a common error, particularly in patients who present with mild disease and may seem otherwise well.[18]
Pancreatic inflammation and the SIRS response can lead to third space fluid loss, resulting in profound hypovolaemia, hypoperfusion and end-organ damage.
Practical tip
Note that use of lactate-containing fluid in a patient with impaired liver metabolism may lead to a spuriously elevated lactate level, so results need to be interpreted with other markers of volume status.
Evidence: Choice of fluid
Evidence from all critically ill patients points to no benefits from using a balanced crystalloid in preference to normal saline. Direct evidence from patients with acute pancreatitis is very limited, although Ringer’s lactate potentially has an additional anti-inflammatory effect in these patients. Clinical practice varies widely, so you should check local protocols.
There has been extensive debate over the choice between normal saline (an unbalanced crystalloid) versus a balanced crystalloid (such as Hartmann’s solution [also known as Ringer’s lactate] or Plasma-Lyte®) for fluid resuscitation in critically ill patients.
In 2021 to 2022 two large double-blind randomised controlled trials (RCTs) were published assessing intravenous fluid resuscitation in intensive care unit (ICU) patients with a balanced crystalloid solution (Plasma-Lyte®) versus normal saline: the Plasma-Lyte® 148 versus Saline (PLUS) trial (53 ICUs in Australia and New Zealand; N=5037) and the Balanced Solutions in Intensive Care Study (BaSICS) trial (75 ICUs in Brazil; N=11,052).[100][101]
In the PLUS study, 45.2% of patients were admitted to the ICU directly from surgery (emergency or elective), 42.3% had sepsis, and 79.0% were receiving mechanical ventilation at the time of randomisation.
In the BaSICS study, almost half the patients (48.4%) were admitted to the ICU after elective surgery and around 68% had some form of fluid resuscitation before being randomised.
Both found no difference in 90-day mortality overall or in prespecified subgroups for patients with acute kidney injury (AKI) or sepsis, or post-surgery. They also found no difference in the risk of AKI.
Neither study included specific/subgroup information on patients with acute pancreatitis.
One meta-analysis of 13 RCTs (including PLUS and BaSICS) confirmed no overall difference, although the authors did highlight a non-significant trend towards a benefit of balanced solutions for risk of death.[104]
Previous evidence has been mixed.
One 2015 double-blind, cluster-randomised, double-crossover trial conducted in four ICUs in New Zealand (N=2278), the 0.9% Saline vs Plasma-Lyte® for ICU fluid Therapy (SPLIT) trial, found no difference for in-hospital mortality, acute kidney injury, or use of renal-replacement therapy.[105]
However, a 2018 US multicentre unblinded cluster-randomised trial - the isotonic Solutions and Major Adverse Renal events Trial (SMART), among 15,802 critically ill adults receiving ICU care - found possible small benefits from balanced crystalloid (Ringer’s lactate or Plasma-Lyte®) compared with normal saline. The 30-day outcomes showed a non-significant reduced mortality in the balanced crystalloid group versus the normal saline group (10.3% vs. 11.1%; odds ratio [OR] 0.90, 95% CI 0.80 to 1.01) and a major adverse kidney event rate of 14.3% versus 15.4%, respectively (OR 0.91, 95% CI 0.84 to 0.99).[106]
One 2019 Cochrane review included 21 RCTs (N=20,213) assessing balanced crystalloids versus normal saline for resuscitation or maintenance in a critical care setting.[107]
The three largest RCTs in the Cochrane review (including SMART and SPLIT) all examined fluid resuscitation in adults and made up 94.2% of participants (N=19,054).
There was no difference in in‐hospital mortality (OR 0.91, 95% CI 0.83 to 1.01; high-quality evidence as assessed by GRADE), acute renal injury (OR 0.92, 95% CI 0.84 to 1.00; GRADE low), or organ system dysfunction (OR 0.80, 95% CI 0.40 to 1.61; GRADE very low).
There are very few high-quality studies on choice of fluid specifically for patients with acute pancreatitis. The 2018 American Gastroenterological Association (AGA) Institute guideline and the 2018 guideline from the UK National Institute for Health and Care Excellence (NICE) both concluded there is insufficient evidence to recommend one crystalloid over another.[8][97] Similarly, the 2019 World Society of Emergency Surgery (WSES) guidelines for the management of severe acute pancreatitis recommend isotonic crystalloids without specifying a preference for type of fluid; however, the WSES also states that Ringer’s lactate may be associated with anti-inflammatory effect, although evidence for superiority of Ringer’s lactate versus normal saline based on randomised trials is weak.[29]
A balanced crystalloid is recommended in preference to normal saline by both the 2013 American College of Gastroenterology (ACG) guideline and the 2013 International Association of Pancreatology/American Pancreatic Association guideline (IAP/APA).[18][48]
This is based on an RCT that found the incidence of SIRS was lower among acute pancreatitis patients treated with Ringer’s lactate compared with normal saline.
The ACG also notes potential benefits linked to the better pH balance of Ringer’s lactate compared with normal saline, which could theoretically result in metabolic acidosis and make the pancreatic acinar cells more susceptible to further injury.
Evidence: Fluid volume and duration
The overall weight of evidence is converging towards a moderate goal-directed fluid replacement strategy for patients with acute pancreatitis.
There has been extensive debate over the relative benefits of an aggressive versus a more conservative approach to fluid resuscitation in acute pancreatitis. While, traditionally, early aggressive hydration was standard practice there was no evidence for this approach. More recent studies suggest that both overly aggressive and overly conservative fluid therapy can cause harm in acute pancreatitis, and that a moderate goal-directed replacement strategy offers the best overall patient outcomes.[96][108]
Overly aggressive fluid replacement can lead to increased mortality due to fluid overload. Two randomised controlled trials (RCTs) from the same research group in 76 and 116 patients with acute severe pancreatitis found that aggressive non-targeted fluid therapy (10-15 mL/kg/hour) was associated with significantly higher rates of sepsis, abdominal compartment syndrome, need for mechanical ventilation, and death compared with more conservative fluid replacement (5-10 mL/kg/hour).[109][110]
Conversely, one RCT in 60 patients with predicted mild acute pancreatitis found that aggressive early fluid resuscitation (20 mL/kg bolus followed by 3 mL/kg/hour) was associated with better clinical outcomes at 36 hours than more conservative hydration (10 mL/kg bolus followed by 1.5 mL/kg/hour). Both groups had a goal-directed approach with haematocrit, blood urea nitrogen (BUN), and creatinine levels assessed every 12 hours.[111]
However, one subsequent larger RCT (249 patients) was halted when interim analysis showed a significant between-group difference in safety outcomes, with the aggressive early fluid resuscitation group (20 mL/kg bolus followed by 3 mL/kg/hour) experiencing a higher incidence of fluid overload without improvement in clinical outcomes compared with the moderate fluid resuscitation group (10 mL/kg bolus followed by 1.5 mL/kg/hour). In both groups, fluids were adjusted based on clinical assessment at 12, 24, 48, and 72 hours.[108]
Guidelines differ in their specific recommendations so you should check local protocols:
The guideline committee for the UK NICE guideline on pancreatitis (first published 2018) was unable to make a recommendation on the speed of fluid resuscitation due to the lack of clear evidence at that time. The committee recommended further research in this area.[8]
The 2018 AGA Institute and 2013 IAP/APA guidelines both recommend goal-directed fluid resuscitation, with the latter specifying a starting volume of 5 to 10 mL/kg/hour.[48][97]
The 2019 WSES guideline for the management of severe acute pancreatitis recommends that fluid volume be adjusted to the patient’s age, weight, and pre-existing renal and/or cardiac conditions, and states that the value of early goal-directed therapy in patients with acute pancreatitis remains unknown.[29]
These guideline recommendations were published before the subsequent, larger RCT (249 patients) findings discussed above.
Electrolyte abnormalities
Monitor electrolytes frequently in the first 48 to 72 hours, especially in predicted moderate/severe pancreatitis or if large volumes of fluid replacement are needed.[18][48]
Calcium. Take steps to correct hypocalcaemia in (predicted) severe acute pancreatitis because it may lead to cardiac arrhythmias.
Magnesium. Magnesium should be replaced if low levels are identified as this can itself cause hypocalcaemia. Low magnesium is common in alcoholic and malnourished patients.
Magnesium is necessary for the secretion of parathyroid hormone (PTH), therefore low levels will reduce the action of PTH on calcium homeostasis.
If the patient has had a large volume of saline infused, also monitor sodium and chloride because of the risk of hyperchloraemic acidosis.
Oxygen
It is important to monitor oxygenation saturations, because patients may be hypoxaemic, requiring supplemental oxygen.
Patients with acute pancreatitis are at high risk of hypoxia because of one or more of: abdominal splinting, atelectasia, pulmonary oedema, acute respiratory distress syndrome.
During initial management, consider the need for blood gas analysis (arterial or venous) to assess both oxygenation and acid-base status if the patient shows signs of deterioration.[5]
Patients with worsening hypoxia may need intensive care and ventilatory support.
PaO2 <60 mmHg is a sign of organ failure.[18]
Analgesia
Pain is the predominant symptom. Ensure it is treated promptly and effectively.[96]
Failure to control pain can compromise breathing and contribute to haemodynamic instability.[49]
In mild cases, use a standard pain ladder approach to select, monitor, and adjust analgesia.[96]
Use a pain score to monitor the response to analgesia and adjust the dose and/or type of analgesic using local pain management protocols.
Opioids may be needed for effective pain control.[49][112]
[ ]
[Evidence C]
A Cochrane review found that opioids are an appropriate option in acute pancreatitis and may decrease the need for supplementary analgesia. However, the quality of evidence for different analgesia approaches was low.[112]
Morphine is considered safe despite a theoretical risk of exacerbating pancreatic inflammation.[49][112]
Despite the theoretical risk that morphine may increase pressure on the sphincter of Oddi and therefore exacerbate pancreatitis, there is no good evidence that this is clinically significant. In practice morphine is considered a safe option.[49][112]
Practical tip
An example of a suitable pain relief regimen would be to start with a non-steroidal anti-inflammatory drug (NSAID) such as ibuprofen. If pain relief is inadequate, use a low-potency opioid (e.g., codeine) alone or in combination with ibuprofen. If pain relief continues to be inadequate, a high-potency opioid (e.g., morphine) with or without an adjunctive treatment (e.g., an NSAID) may be required.
In practice, many patients will require a high-potency opioid immediately rather than using the standard pain ladder approach.
Check your local protocols for specific advice on choice of drugs and doses.
Nutrition
Oral nutrition
Start a normal oral diet as soon as the patient can tolerate it (ideally within 24 hours).
Use a low/normal fat soft or solid diet for early feeding. A liquid diet is unnecessary.[18][96][97]
Oral nutrition should resume as soon as the pain and any nausea/vomiting begins to subside.
Enteral feeding (oral or by enteral tube if oral feeding is not tolerated) is thought to protect the gut-mucosal barrier, reducing bacterial translocation and thereby reducing the risk of infected necrosis and other serious outcomes.[48][96][97]
Undertake repeat trials of oral feeding if it cannot be achieved on the first attempt in patients with mild disease.[97]
Early oral feeding is not successful in all patients, owing to pain, vomiting, or ileus.
Avoid routine requesting of nil by mouth status; it is important to attempt repeat feeding trials in patients with mild disease.[8]
Most patients with mild disease can re-establish oral feeding within 5 days.[18]
Evidence: Oral feeding
Early oral feeding promotes better outcomes.
Latest evidence demonstrates that enteral (ideally oral) feeding reduces the risk of infected pancreatic necrosis and other serious outcomes.[97][98]
Combined results from 11 RCTs on early versus delayed feeding found:[97]
No difference in mortality.
Delayed feeding was associated with a significantly higher risk of needing intervention for pancreatic necrosis (2.47-fold higher risk with delayed vs. early feeding, 95% CI 1.41 to 4.35).
A trend towards a higher risk of other poor outcomes for delayed vs. early feeding:
Infected necrosis (OR 2.69, 95% CI 0.8 to 3.6)
Multiple organ failure (OR 2.00, 95% CI 0.48 to 8.22)
Total necrotising pancreatitis (OR 1.84, 95% CI 0.88 to 3.86).
Enteral nutrition
In patients who are unable to feed orally, use enteral tube feeding rather than parenteral nutrition.
Start this within 72 hours of presentation.[8]
Nasojejunal or nasogastric tube feeding are both safe options.[8][18][48][49][96][97]
Many patients with predicted severe/moderately severe acute pancreatitis or necrotising pancreatitis do end up needing nutrition support. Always offer enteral (rather than parenteral) nutrition to patients in this group unless it is contraindicated.[8][97]
Either a polymeric or elemental enteral nutrition formulation can be used.[48]
Evidence: Benefits of enteral feeding
Enteral feeding is associated with a lower risk of serious complications and death compared with parenteral.
The use of enteral nutrition is associated with a lower likelihood of harm compared with parenteral nutrition.
There is good evidence to support the benefits of enteral compared with parenteral feeding in patients with acute pancreatitis. Enteral nutrition (orally or by tube) was associated with a significantly lower risk of infected pancreatic necrosis (OR 0.28, 95% CI 0.15 to 0.51), single organ failure (OR 0.25, 95% CI 0.10 to 0.62), and multi-organ failure (OR 0.41, 95% CI 0.27 to 0.63) in a meta-analysis of 12 RCTs for the 2018 AGA Institute guideline group.[97]
An earlier Cochrane review of eight RCTs also found that enteral nutrition, compared with total parenteral nutrition, significantly reduced mortality (RR 0.50, 95% CI 0.28 to 0.91), multi-organ failure (RR 0.55, 95% CI 0.37 to 0.81), systemic infections (RR 0.39, 95% CI 0.23 to 0.65), and the need for surgical intervention (OR 0.44, 95% CI 0.29 to 0.67) in patients with (predicted) severe acute pancreatitis. There was also a trend towards a shorter hospital stay.[115]
Debate: Timing of enteral nutrition
The evidence on the optimum timing to start enteral nutrition in (predicted) severe acute pancreatitis is mixed and guidelines differ.
Guidance from the UK National Institute for Health and Care Excellence (NICE) recommends enteral nutrition within 72 hours of presentation for any patient with severe or moderately severe acute pancreatitis.[8]
NICE’s review of the evidence found no benefit from delaying nutrition whereas underfeeding is known to be an issue in clinical practice, with some patients waiting 5 to 7 days for nutritional support.
A 2017 literature review concluded that the latest evidence supports limiting tube feeding to those patients with predicted severe pancreatitis who are unable to achieve sufficient oral calorie intake after 3 to 5 days.[96]
This is based on two RCTs that found no beneficial effects on infection rates or mortality from early nasojejunal feeding.
However, the IAP/APA 2013 guideline cites one small RCT of 60 patients with severe acute pancreatitis that found outcomes were better when enteral nutrition was started within 48 hours rather than delayed for 7 days of fasting.[116]
Evidence: Nasojejunal versus nasogastric tube feeding
There is no clear evidence of any difference in outcomes from nasojejunal versus nasogastric tube feeding.
Three RCTs that have compared nasoenteral (nasojejunal or nasoduodenal) versus nasogastric tube feeding in acute pancreatitis have failed to find any evidence of a significant difference in mortality.[97]
The 2018 AGA Institute guideline, the 2013 ACG guideline, and the 2013 IAP/APA guideline all recommend that either nasojejunal or nasogastric tube feeding can be used.[18][48][97]
Nasogastric tube insertion is a ward-based procedure that is easier and cheaper than nasojejunal tube insertion, which requires access to radiology or endoscopy.[18][49]
In practice, however, safety concerns over the risk of aspiration with a nasogastric tube mean that nasojejunal feeding is sometimes preferred in patients with severe acute pancreatitis, although there is little evidence to support this approach.[8][97] One Cochrane review of five RCTs that compared nasojejunal versus nasogastric tube feeding in severe acute pancreatitis failed to find evidence of a significant difference in mortality.[119]
Nasojejunal feeding may also be more effective in patients with gastric outlet obstruction or other types of digestive intolerance.[8][120]
Gastric feeding may be more suitable for patients with no duodenal stenosis or oedema.[8]
If nasogastric tube feeding is used, the patient should be put in an upright position and placed on aspiration precautions.[18]
Parenteral nutrition
Parenteral nutrition is associated with higher mortality. Only use it for patients in whom enteral feeding is not feasible, not tolerated, or failing to meet calorie requirements.[18][48]
Evidence suggests that total parenteral nutrition is associated with a higher risk of death, organ failure, and pancreatic necrosis.[97][115]
Parenteral nutrition should be reserved for patients who do not tolerate enteral feeding or cannot tolerate the targeted nutritional requirements, or if there are contraindications to enteral feeding. Contraindications to enteral feeding may include bowel obstruction, abdominal compartment syndrome, prolonged paralytic ileus and mesenteric ischaemia.[120]
Antibiotic prophylaxis
Do not give prophylactic antibiotics, regardless of predicted or actual disease severity, as there is no evidence to support their use.
Only give antibiotics if pancreatic or extra-pancreatic infection is proven or clinically suspected.[8][18][48][97][96]
There is no good evidence to support the use of prophylactic antibiotics to prevent infection in patients with acute pancreatitis, even among those with predicted severe disease or with sterile pancreatic necrosis. Major guidelines recommend against routine use of prophylactic antibiotics.[18][48][97]
Suspect infection based on signs and symptoms such as fever, leukocytosis, and signs of organ dysfunction.
Extra-pancreatic infections (e.g., pneumonia, cholangitis, urinary tract infections, sepsis) are a major cause of mortality in the early phase of acute pancreatitis when SIRS is the main cause of death.
Up to 20% of patients with acute pancreatitis will develop an extra-pancreatic infection.[121]
Suspected infection should be treated empirically and aggressively with intravenous antibiotics. The antibiotics should be discontinued if blood cultures are negative and no source of infection is identified.[18]
Evidence: Prophylactic antibiotics
Prophylactic antibiotics have no benefits in reducing the risk of infected necrosis or death.
Three RCTs and two meta-analyses have failed to show any benefit from prophylactic antibiotics compared with placebo in reducing the risk of infection of pancreatic (or peri-pancreatic) necrosis or reducing the risk of death.[122][123][124][125][126]
Prophylactic antibiotics also have no impact on rates of organ failure or length of hospital stay.[97]
This lack of benefit from prophylactic antibiotics holds true for patients with predicted severe disease and those with sterile necrotising pancreatitis as well as those with milder disease, according to a meta-analysis conducted for the AGA Institute 2018 guideline group.[97]
Treatment setting
You should immediately transfer to ICU any patient who meets the original or revised Atlanta Criteria for organ failure or who fulfils one or more of the below criteria:[18][48]
Heart rate <40 bpm or >150 bpm
Systolic arterial pressure <80 mmHg or mean arterial pressure <60 mmHg, or diastolic arterial pressure >120 mmHg
Respiratory rate >35 breaths/minute
Serum sodium <110 mmol/L or >170 mmol/L
Serum potassium <2 mmol/L or >7 mmol/L
pH <7.1 or >7.7
Serum glucose >44.4 mmol/L
Serum calcium >3.75 mmol/L
Anuria
Coma.
Refer to a specialist pancreatic centre any patient:[8][48]
Who has moderately severe or severe acute pancreatitis
Who is being considered for intervention (drainage or necrosectomy/debridement).
Consider referral to a specialist pancreatic centre if the patient has pancreatic ascites and pleural effusion.[8]
Gallstone (biliary) pancreatitis
All patients with acute pancreatitis should have a right upper quadrant abdominal ultrasound on admission to confirm or exclude gallstones as the cause.[18][48]
Gallstones that get stuck in the common bile duct obstruct the main pancreatic duct, causing pancreatic injury.
Cholecystectomy
Arrange cholecystectomy for any patient with gallstones confirmed as the cause.
Cholecystectomy should be done during the initial admission for all patients with mild acute gallstone pancreatitis.[29][48][49][96][97][127]
In patients with more severe disease, it is common practice to delay cholecystectomy until the inflammation has resolved.[29][48][49]
Cholecystectomy is recommended to reduce the risk of recurrence of acute pancreatitis.
The optimum timing of cholecystectomy in different patient groups has been the subject of vigorous debate.[96][97]
Cholecystectomy should be undertaken during the initial hospital admission in any patient with mild acute pancreatitis, as soon as possible after the initial symptoms have resolved and certainly within 2 weeks.[49][96][97]
In severe acute gallstone pancreatitis, cholecystectomy is often delayed until any organ dysfunction and/or intra-abdominal inflammation has resolved.[49][96]
However, this is based on expert consensus rather than solid evidence.
Evidence: Timing of cholecystectomy
Same-admission cholecystectomy improves outcomes in mild disease.
Early cholecystectomy is favoured in mild disease because the risk of recurrent acute pancreatitis is directly related to the delay between the first attack and the timing of cholecystectomy.[128][129][130]
The shorter the interval, the lower the risk. Even a few weeks delay in undertaking cholecystectomy is associated with a high risk (up to 80%) of recurrence and readmission for acute pancreatitis.[128]
One RCT found that same-admission cholecystectomy was associated with significant reductions in a composite outcome of mortality and gallstone-related complications (OR 0.24, 95% CI 0.09 to 0.61), readmission for recurrence (OR 0.25, 95% CI 0.07 to 0.90), and pancreatobiliary complications (OR 0.24, 95% CI 0.09 to 0.61) when compared with later cholecystectomy.[131]
In patients who are unfit for surgery, consider endoscopic biliary sphincterotomy to reduce the risk of recurrence.
Patients with mild acute gallstone pancreatitis who are unable to have cholecystectomy because of frailty and/or comorbid disease may benefit from biliary sphincterotomy.[18]
Evidence suggests this can effectively reduce the risk of recurrent acute pancreatitis, although cholecystitis may still occur.[79]
ERCP
Most gallstones that cause acute pancreatitis pass into the duodenum.
In a minority of patients, persistent choledocholithiasis can obstruct the common bile duct, leading to severe acute pancreatitis and/or cholangitis.[18]
Gallstone pancreatitis with cholangitis
Arrange emergency ERCP within 24 hours for any patient with acute gallstone pancreatitis who has concurrent cholangitis. Suspect cholangitis if Charcot’s triad is present:[18][29][48][96]
[ ]
Jaundice
Fever and rigors
Right upper quadrant pain.
It is challenging to accurately diagnose cholangitis in patients who have SIRS, which is a common early phase complication in patients with (predicted) severe acute gallstone pancreatitis.
Several definitions have been proposed for cholangitis, including Charcot’s triad, expert opinion, and the updated TF13 Tokyo guidelines.[96][132]
These use low thresholds for inflammation and cholestasis, both of which are commonly present in acute pancreatitis.
This can lead to overdiagnosis of cholangitis in patients with acute pancreatitis, exposing patients to unnecessary and potentially risky ERCP procedures.[96]
In practice, cholangitis is usually suspected if Charcot’s triad is present.[96]
Evidence: Early ERCP in patients with concurrent cholangitis
Early routine ERCP reduces the risk of death or complications in patients with gallstone pancreatitis complicated by cholangitis.
There is evidence that ERCP within 24 hours of admission reduces morbidity and mortality in patients with acute gallstone pancreatitis complicated by cholangitis.[133][134]
One systematic review found that early routine ERCP significantly reduced mortality (RR 0.20, 95% CI 0.06 to 0.68) as well as the risk of both local and systemic complications as defined by the Atlanta classification (RR 0.45, 95% CI 0.20 to 0.99; and RR 0.37, 95% CI 0.18 to 0.78, respectively).[134]
Gallstone pancreatitis without cholangitis
ERCP is not indicated in mild acute gallstone pancreatitis without cholangitis.[29][48][96]
The risks of the procedure outweigh any potential benefits in this group.
A meta-analysis of eight RCTs carried out for the AGA Institute 2018 guideline group found that urgent ERCP has no impact on mortality, multi- or single-organ failure, infected pancreatic necrosis, or total rate of pancreatic necrosis when compared with conservative management.[97]
ERCP is probably not indicated in severe acute gallstone pancreatitis without cholangitis or bile duct obstruction.
There is ongoing debate about the balance of risks and benefits of urgent ERCP (within 24 hours) in patients with (predicted) severe acute gallstone pancreatitis without concurrent cholangitis and without evidence of bile duct obstruction.[96]
The 2019 WSES guideline states that ERCP cannot be currently recommended for this patient group.[29]
The 2018 AGA Institute guideline recommends against its use.[97]
The 2013 IAP/APA guideline says it is “probably” not indicated.[48]
The 2013 ACG guideline concluded that current evidence does not support ERCP in this patient group.[18]
Evidence: ERCP in severe gallstone pancreatitis
There is conflicting evidence on the role of ERCP in severe gallstone pancreatitis.
Several RCTs and meta-analyses have reached different conclusions on the role of ERCP in severe gallstone pancreatitis.[96]
The quality of evidence is poor, with some trials failing to exclude patients with cholangitis (in whom the procedure is certainly indicated) and others using varied definitions of gallstone pancreatitis.[96]
Results are awaited from the APEC trial, a large multicentre RCT looking at whether early ERCP with sphincterotomy reduces complications and mortality in (predicted) severe acute gallstone pancreatitis without cholangitis.[135]
ERCP is probably indicated in gallstone pancreatitis with common bile duct obstruction without cholangitis.[48]
[ ]
A meta-analysis of seven RCTs including 757 patients found support for the use of ERCP in patients with gallstone obstruction, although there were too few patients to study hard clinical outcomes such as mortality.[136]
Arrange bile duct imaging with endoscopic ultrasound (EUS) if available (or MRCP if not) if bile duct obstruction is suspected in the absence of cholangitis. This imaging-first approach can help prevent unnecessary ERCP procedures.[18][48][96]
More widespread use of EUS to detect common bile duct stones is emerging as a useful strategy to guide a decision on the need for ERCP in severe acute gallstone pancreatitis without cholangitis.
EUS is superior to magnetic resonance cholangiopancreatography (MRCP) for detecting small stones (<5mm).
Ongoing derangement of LFTs is an important sign of possible bile duct stones.
MRCP is an alternative option for detecting common bile duct stones and is more widely available than EUS.
However, a negative MRCP does not rule out the presence of small common bile duct stones (<5mm), which are known to cause gallstone pancreatitis.[48]
Evidence: EUS-first strategy
Many ERCP procedures can be avoided by prior bile duct imaging.
A meta-analysis comparing an EUS-first strategy with diagnostic ERCP in patients with acute gallstone pancreatitis found that 71% of ERCP procedures could be avoided without any negative impact on clinical outcomes by employing an EUS-first approach.[137]
Where bile duct stones are identified in the absence of cholangitis, it is reasonable to wait 48 hours to see if there is spontaneous improvement of gallstone obstruction before undertaking ERCP.[48]
There is no specific evidence available to guide the optimal timing of ERCP in patients with bile duct obstruction but no cholangitis and this remains a controversial issue.
Evidence: Timing of ERCP
There is only weak evidence to support decisions on the timing of ERCP.
A meta-analysis of seven RCTs including 757 patients with bile duct obstruction found no statistically significant effect from performing ERCP within 24 hours compared with within 72 hours. However, none of the RCTs included was designed specifically to study this.[48]
Alcohol-related acute pancreatitis
Patients with an alcohol-related cause may need alcohol withdrawal prophylaxis.
A benzodiazepine is generally used in this group of patients.[138] Choice of drug and dose regimen depends on various factors including the indication (e.g., alcohol withdrawal seizures, delirium tremens) and patient-specific factors (e.g., hepatic impairment, ability to take oral medication). See Alcohol withdrawal.
Arrange a brief alcohol counselling intervention during the same admission.[97]
There is sparse evidence available to support decisions on the most effective approach to alcohol counselling for patients with alcohol-related acute pancreatitis.
Encourage patients to abstain from alcohol for 6 to 12 months (regardless of whether or not the acute pancreatitis was alcohol-related).[49]
Evidence: Alcohol abstinence counselling
Weak evidence supports some benefits from same-admission counselling.
A literature review conducted for the AGA Institute 2018 guideline group found a single RCT comparing a single intervention during the initial admission with 6-monthly outpatient interventions for 2 years. There was a strong trend towards a reduction in total hospital admission rates for the single intervention during admission and there were no significant differences in risk of a recurrent attack of acute pancreatitis.[97]
Replace thiamine and other water-soluble B vitamins.[139]
Patients with chronic alcoholism need to have their stores of thiamine (vitamin B1), folic acid (vitamin B9), and cyanocobalamin (vitamin B12) replenished.
Continue this supplementation until the patient is able to return to eating a well-balanced diet.
The formulation, dose, and route of administration will depend on various considerations.
See Alcohol withdrawal.
Request contrast-enhanced abdominal computed tomography (CECT) for any patient who fails to improve after 5 to 7 days of initial treatment to check for the presence of local complications.[48]
This is particularly important for any patient with ongoing signs of SIRS or organ failure.[49]
CECT is unreliable during the first 3 to 4 days of acute pancreatitis for detecting any fluid collections and determining the extent of any necrosis.[71][93][96]
Local complications include:
Pancreatic/peri-pancreatic necrosis.
Around 20% of patients with acute pancreatitis develop pancreatic and/or peri-pancreatic necrosis.[8]
May be sterile or infected.
May be diffuse (typically <4 weeks) or walled off (>4 weeks).
Early in the course of the disease, pancreatic necrosis is a diffuse solid/semi-solid inflammatory mass. After approximately 4 weeks, a fibrous wall develops around the necrosis, making it easier to remove with drainage or necrosectomy/debridement.[18]
Pancreatic/peri-pancreatic fluid collections, including pseudocyst s.
Acute fluid collections are those that occur within the first 4 weeks and lack a defined wall.
They are usually asymptomatic and resolve spontaneously within 20 days.
Pseudocysts are completely liquid fluid collections (i.e., without necrosis) that have persisted for at least 2 weeks (usually more than 4 weeks) and developed a defined wall. Pseudocysts never occur in necrotising pancreatitis.
[Figure caption and citation for the preceding image starts]: Walled off pancreatic necrosisGut 2017;66:2024-2032; used with permission [Citation ends].
[Figure caption and citation for the preceding image starts]: Infected necrotic collection, not fully encapsulated, on day 20 after disease onsetGut 2017;66:2024-2032; used with permission [Citation ends].
Pancreatic necrosis
Infected necrosis
The risk of death in patients with infected necrosis is as high as 30%.[48]
Suspect infected necrosis on the basis of clinical and/or imaging signs.[48]
Give intravenous antibiotics if infected necrosis is proven or highly suspected.[48]
Choose an antibiotic that has good pancreatic penetration, such as a carbapenem (e.g., imipenem/cilastatin), a fluoroquinolone (e.g., ciprofloxacin), or metronidazole.[18]
Many patients with infected pancreatic necrosis will need intervention with drainage or necrosectomy/debridement.[96]
Infected pancreatic/peri-pancreatic necrosis is associated with a mortality rate of up to 30%.[18][48][140]
It tends to develop in the late phase of the disease (>14 days) although one review found 27% of cases occurred within the first 14 days.[18][141]
Clinical signs (persistent fever, increasing inflammatory markers, new/persistent organ failure), and/or imaging signs (the presence of gas within the necrosis on the CECT) are accurate indicators that pancreatic necrosis has become infected.[48]
Do not use routine percutaneous fine needle aspiration (FNA) and culture to confirm or exclude the presence of infection.[48]
Debate: FNA
Guidelines differ on the role of FNA to distinguish infected from sterile necrosis.
There is ongoing debate about the appropriate role of FNA to distinguish infected from sterile necrosis and guidelines differ on this.
The IAP/APA 2013 guideline recommends that because of the high false negative rate, FNA should only be undertaken in patients who fail to improve clinically after several weeks of necrotising pancreatitis where there are no clear clinical or imaging signs of infection.[48]
The ACG 2013 guideline recommends considering an FNA whenever infection is suspected, on the basis that patients with sterile as well as infected necrosis can appear similar in terms of leukocytosis, fever, and organ failure markers.[18]
All patients with confirmed or highly suspected infected necrosis should receive immediate intravenous antibiotics.
Imipenem/cilastatin, a carbapenem antibiotic, is usually the first-line choice because it has good pancreatic penetration.
Check local guidance on antibiotic stewardship.
In most cases, antibiotics are used alongside supportive care to delay the need for further intervention with catheter drainage or necrosectomy/debridement until the necrosis has become walled off.[48][96]
However, some studies have had high success rates from treating infected pancreatic necrosis solely with antibiotics.[81][143][144]
One meta-analysis of 8 studies involving 409 patients found 64% could be managed by conservative antibiotic treatment with 12% mortality.[145]
Intervention with drainage or necrosectomy/debridement is indicated in infected necrosis when there is ongoing clinical deterioration or the patient fails to improve despite antibiotics and supportive care.[18][48][96]
Ideally the intervention should be delayed until the necrotic collection is fully walled off.[18][48]
The decision on timing of intervention(s) and choice of approach requires input from a specialist pancreatic centre.[48]
Choice of intervention in infected pancreatic necrosis
The intervention options for pancreatic necrosis that has failed to respond to conservative treatment with intravenous antibiotics are:
Catheter drainage - either percutaneous retroperitoneal or endoscopic transluminal.
Necrosectomy/debridement - may be minimally invasive percutaneous, endoscopic, or open surgery.
Take a step-up approach to intervention, ideally waiting until the necrosis is walled off. Start with catheter drainage and reserve necrosectomy for those patients who fail to respond.[48]
The step-up approach is the best option even for patients with multi-organ failure.[48]
Always consult a specialist pancreatic team about any patient being considered for intervention.[48][49]
Image-guided catheter drainage (without necrosectomy) is the usual first option in patients who need intervention for pancreatic necrosis. Percutaneous catheter drainage is technically feasible in >95% of patients.[48] Latest evidence suggests comparable results from endoscopic transluminal drainage.[146]
Reduction in collection size of ≥75% after the first 10 to 14 days of percutaneous drainage correctly predicts successful treatment with this technique.[48][147]
Evidence: Step-up approach to infected necrosis
A step-up approach reduces the risk of both short- and long-term complications.
A multicentre RCT of 88 patients with infected necrosis showed that a step-up approach starting with percutaneous catheter drainage reduced both short-term complications (multi-organ failure) and long-term complications (endocrine insufficiency) when compared with open necrosectomy.[81]
Approximately 35% of patients with infected necrosis were successfully managed with solely catheter drainage and did not require necrosectomy.
The step-up approach reduced the composite endpoint of death or major complications from 69% to 40% (RR 0.57, 95% CI 0.38 to 0.87).
Subgroup analysis of severity in one multicentre RCT found the step-up approach was beneficial regardless of whether or not patients had multi-organ failure.[81]
Minimally invasive or open necrosectomy is only indicated for patients who fail to respond to catheter drainage.[48] Minimally invasive necrosectomy is probably safer than open surgery.[8][48][148]
There are no head-to-head RCTs comparing minimally invasive versus open techniques for necrosectomy/debridement.[96]
However, retrospective studies suggest a lower risk of complications and death with minimally invasive techniques.[149][150]
Minimally invasive approaches include percutaneous, laparoscopic, endoscopic, and video-assisted retroperitoneal techniques.[151]
For around 60% to 70% of patients with infected pancreatic necrosis, the preferred intervention will depend on whether the necrosis is associated with the posterior wall of the stomach (endoscopic approach) or the postero-lateral abdominal wall (percutaneous approach).[8]
The UK National Institute for Health and Care Excellence (NICE) guideline recommends an endoscopic approach as the preferred first-line option when it is anatomically possible. This is based on evidence suggesting the endoscopic approach is more acceptable to patients than percutaneous catheter drainage and is associated with lower complication rates and shorter hospital stay. However, endoscopic procedures are high-risk and require referral to a specialist pancreatic centre.[8]
Timing of drainage/necrosectomy in infected pancreatic necrosis
Delay the intervention until the necrosis has become walled off, if at all possible.
Current guidelines recommend delaying intervention with catheter drainage or necrosectomy/debridement until the necrosis has become fully walled off (usually 4-5 weeks after the onset of acute pancreatitis).[18][48][49]
The rationale for this is that intervening once the necrosis has become walled off is presumed to carry a lower risk of bleeding and less likelihood of needing a second intervention.[96]
Evidence: Timing of intervention
Early necrosectomy is associated with worse outcomes but the evidence on timing of catheter drainage is less clear cut.
Evidence does suggest that open necrosectomy is associated with worse outcomes (including higher death rates) if performed early.[143][147][152][153][154] However, some commentators have queried the case for a delay in the case of catheter drainage.[155]
Results are awaited from the multicentre POINTER trial, an RCT that is comparing immediate versus postponed drainage in patients with infected pancreatic necrosis.
Intervention must not be delayed in a subgroup of patients who have an intra-abdominal catastrophe (e.g., ongoing acute bleeding, perforation, abdominal compartment syndrome).[48]
In cases where percutaneous catheter drainage cannot be safely delayed until >4 weeks, ideally necrosectomy should still not be undertaken until the collection has become walled off.[48]
Sterile pancreatic necrosis
Closely monitor patients with sterile necrosis as they can develop organ failure.[18]
The vast majority of patients with sterile necrosis can be managed without the need for intervention with catheter drainage/necrosectomy.[48]
Do not give prophylactic antibiotics to patients with necrosis who have no proven or clinically suspected infection.[8][18][48][96][97]
Although the presence of infection in the necrosis increases the risk of organ failure, patients with sterile necrosis can also develop organ failure.[18]
There is no evidence to support prophylactic antibiotics for sterile necrosis.[8][18][48][96][97]
Evidence: Prophylactic antibiotics
Prophylactic antibiotics do not reduce the risk of infected necrosis or death.
Three RCTs and two meta-analyses have failed to show any benefit from prophylactic antibiotics compared with placebo in reducing the risk of infection of pancreatic (or peri-pancreatic) necrosis or reducing the risk of death.[122][123][124][125][126]
Prophylactic antibiotics also have no impact on rates of organ failure or length of hospital stay.[97]
This lack of benefit from prophylactic antibiotics holds true for patients with predicted severe disease and those with sterile necrotising pancreatitis as well as those with milder disease, according to a meta-analysis conducted for the AGA Institute 2018 guideline group.[97]
Asymptomatic sterile necrosis will often resolve over time without catheter drainage/necrosectomy, regardless of size, location, and extension.[18][48][63]
Intervention performed too early carries a high risk of secondary infection of the necrotic collections. Always seek advice from a specialist pancreatitis centre.[8]
Possible indications for carefully timed radiological, surgical, or endoscopic intervention in sterile necrosis include:[18][48]
Ongoing gastric outlet, intestinal, or biliary obstruction due to the mass effect of walled-off necrosis (normally 4-8 weeks after the onset of acute pancreatitis).
Symptoms (e.g., pain, ‘persistent unwellness’) that continue beyond 8 weeks after the onset of acute pancreatitis.
Disconnected duct syndrome with persistent symptomatic collections beyond 8 weeks after the onset of acute pancreatitis.
Pseudocysts
Monitor the patient for the development of pseudocysts, a late complication of acute pancreatitis.
Pseudocysts form outside of the pancreatic parenchyma at least 2 weeks (usually more than 4 weeks) after the onset of acute pancreatitis as encapsulated, non-epithelialised abdominal fluid collections.[19]
Treat infected pseudocysts (best described as abscesses) with immediate antibiotics and catheter drainage.
Choose an antibiotic that has good pancreatic penetration, such as a carbapenem (e.g., imipenem/cilastatin), a fluoroquinolone (e.g., ciprofloxacin), or metronidazole.[18]
Arrange catheter drainage of symptomatic pseudocysts for symptom relief.[19][50]
Pseudocysts can become painful, or cause early satiety and weight loss when their size affects the stomach and bowel.
Manage asymptomatic, sterile pseudocysts conservatively, regardless of size. Many pseudocysts resolve spontaneously.[19][156]
If a pseudocyst needs to be drained, this can be done percutaneously, surgically, or endoscopically. Endoscopic techniques have largely replaced surgical and percutaneous approaches.[19][157]
NICE recommends endoscopic ultrasound (EUS)-guided drainage or endoscopic transpapillary drainage as the preferred intervention for symptomatic pancreatic head pseudocysts.[8]
More info
Early in the course of acute pancreatitis, large volumes of fluid may leave the intravascular compartment and form fluid collections in the abdomen. Most of this fluid is reabsorbed in the first few weeks.[19] However, in almost one third of patients the fluid collections will persist and develop a fibrous non-epithelialised wall, becoming a pseudocyst.
'Pseudocysts' diagnosed at initial presentation are likely to represent alternative pathology: either a patient with chronic pancreatitis, an old walled-off area of pancreatic necrosis, or a pancreatic cystic neoplasm.
Pseudocysts are always located outside the parenchyma of the pancreas or adjacent structures; determining their location is important because pseudocysts cannot be differentiated from walled-off pancreatic necrosis (WOPN) by computed tomography appearance of the lesion alone, and treatment of a symptomatic or infected pseudocyst is different from treatment of pancreatic necrosis (symptomatic and/or infected pseudocysts require early drainage).
Magnetic resonance imaging and EUS (with fine-needle aspiration) can determine the presence of tissue in WOPN but these imaging modalities are not used routinely.
Severity classification
Use the revised Atlanta Criteria to classify a case of acute pancreatitis as mild, moderately severe, or severe.[2]
Approximately 80% of patients have mild disease and will recover within a week.[49][97]
Approximately 20% will go on to develop moderately severe or severe disease.
Mild
Acute pancreatitis is classified as mild if there is:
No organ failure
AND
No local or systemic complications.
Moderately severe
Acute pancreatitis is classified as moderately severe if there is either one or both of:
Transient organ failure (<48 hours)
Local or systemic complications (e.g., pancreatic/peri-pancreatic fluid collections or necrosis - either sterile or infected).
Severe
Acute pancreatitis is classified as severe if there is persistent single- or multi-organ failure (>48 hours).
Thresholds for organ failure
Under the revised Atlanta Criteria, organ failure is defined as a Modified Marshall Score of ≥2 on any one of three organ systems (cardiovascular, renal, respiratory).
In everyday clinical practice, it is more practical to continue using the original Atlanta Criteria for organ failure:[18]
Shock: SBP ≤90 mmHg
Pulmonary insufficiency: PaO2 ≤60%
Renal failure: creatinine ≥2 mg/dL
Gastrointestinal bleeding: >500 mL blood loss in 24 hours
Practical assessment of severity
Clinical assessment of severity should reflect the biphasic pattern of deaths associated with the early and late phases of the condition:[18]
Early phase (<2 weeks): death occurs predominantly as a result of the development and persistence of SIRS-related organ failure.
Late phase (>2 weeks): morbidity and mortality related to local complications (chiefly pancreatic necrosis) predominates.
Predicting severity in the first 48 hours
Make an initial assessment of severity (<48 hours) based on SIRS criteria and patient risk factors.[18][48] Do not confirm a case as mild until at least 48 hours after onset of symptoms - many patients who go on to develop severe disease present without signs of organ failure or local complications.[18]
SIRS criteria. Systemic inflammatory response syndrome (SIRS) is defined by the presence of any two or more of the following four criteria:[55]
Heart rate >90 bpm
Respiratory rate >20 breaths/minute (or PaCO2 <32 mmHg)
Temperature >38°C or <36°C
White blood cell count >12 x 109/L or <4 x 109/L.
Patient characteristics associated with increased risk of severe disease:[79][80][81][82][83]
Age >55 years
Obesity (BMI >30)
Altered mental status
Comorbid disease.
Looking for SIRS markers is critical in determining disease severity.
Early prediction of which patients with acute pancreatitis will go on to develop severe disease is very difficult. Nonetheless, risk stratification should be performed so that treatment setting and management decisions can be tailored to the level of risk.[18][48]
The most significant risk in the early phase of acute pancreatitis (<1 week) is from SIRS-related organ failure. The reversal of early organ failure has been shown to reduce morbidity and mortality.[18][84][85]
The presence of SIRS at admission has a high sensitivity for predicting organ failure and mortality risk in acute pancreatitis (100% sensitivity for predicting mortality was seen in one study).[86]
However, it lacks specificity for predicting severe disease (31% in one study) and mortality because the presence of SIRS is not as important as its persistence.[86] In one study of 759 patients with acute pancreatitis, patients without SIRS had a mortality of 0.7% whereas transient SIRS (<48 hours) was associated with a mortality of 8% and persistent SIRS (>48 hour) with a mortality of 25%.[87]
Guidelines recommend that you should also be aware of patient characteristics that are associated with a higher risk of developing severe disease.[18][48]
The SIRS criteria are well known and easy to use so are preferred for daily clinical practice over specific scoring tools such as APACHE II or Glasgow.[48]
Around 50% of patients will have predicted severe/moderately severe disease in the early period after presentation.[96]
However, only half of these patients will actually go on to develop severe/moderately severe pancreatitis.
Mortality among patients with predicted severe disease is around 10% compared with <1% in those with predicted mild disease.[96]
Assessing severity during admission
Use a three-strand approach to assess ongoing severity during admission:[48]
Persistent SIRS
Patient risk factors (e.g., age, comorbidities, BMI)
Response to initial therapy (e.g., renal function)
Do not request CECT scanning to detect local complications within the first 3 to 4 days after presentation.[29][48]
Pancreatic necrosis often takes >72 hours to develop so early CECT will not accurately predict severity based on presence or absence of local complications.
Evidence suggests early inappropriate CECT can increase length of hospital stay without improving clinical outcomes.[48]
Request CECT to look for local complications in any patient who fails to improve after 5 to 7 days, particularly if there are ongoing signs of organ failure.[49]
Use of risk scores to assess severity
Clinical scoring systems for severity/prognostication are best avoided as they provide little additional information and can delay appropriate management.[18][48][51]
Such scoring systems typically require 48 hours to pass before they are accurate, by which time severe disease will become clinically obvious regardless of the score.[18]
There are several such scoring systems in widespread use (the most common being APACHE II, Ranson, and Glasgow) so check local guidance for the protocol to follow in your place of work.
Scores that can be used in the emergency department or within first 24 hours
APACHE II Score[50]
Used to predict prognosis of patients in ICU.
Can be assessed within 24 hours of admission.
A score of 8 or higher predicts severe disease and increased mortality risk.
Performs no better than a standard early warning score in predicting severe acute pancreatitis.[49]
Harmless Acute Pancreatitis Score[88]
Can be for early identification of patients at low risk of progression to severe disease.
Based on three factors: creatinine level in normal range; haematocrit in normal range; no rebound abdominal tenderness or guarding.
Bedside Index of Severity in Acute Pancreatitis (BISAP)
For use in the first 24 hours.
Five-factor scoring system based on: urea level >25 mg/dL; impaired mental status; presence of SIRS; age >60 years; pleural effusion on chest x-ray. A score of 2 or higher is associated with a significantly higher risk of organ failure and death.[88]
Scores for use at 48 hours or later
Ranson Criteria[158]
Used for prediction of severe acute pancreatitis - not diagnosis.
A meta-analysis of 110 studies found the predictive power of the Ranson Criteria was no better than clinical judgement.[159]
Ranson Criteria: Non-gallstone pancreatitis
Criteria on admission | Criteria after 48 hours of admission | Number of criteria and approximate mortality (%) |
|---|---|---|
|
|
|
Ranson Criteria: Adjusted criteria for gallstone-associated pancreatitis
Criteria on admission | Criteria after 48 hours of admission |
|---|---|
|
|
Glasgow (Imrie) Score
A simple prognostic system that uses age and seven laboratory values collected during the first 48 hours following admission to predict risk of severe pancreatitis.
Applicable to both gallstone and alcoholic pancreatitis.
A point is assigned if a certain threshold is met at any time during that 48-hour period.
The parameters are:
Age >55 years 1 point
Serum albumin <32 g/L (3.2 g/dL) - 1 point
Arterial PO2 on room air <8 kPa (60 mmHg) - 1 point
Serum calcium <2 mmol/L (8 mg/dL) - 1 point
Blood glucose >10 mmol/L (180 mg/dL) - 1 point
Serum LDH >600 units/L - 1 point
Serum urea nitrogen >16.1 mmol/L (45 mg/dL) - 1 point
WBC count >15 x 109/L (15 x 103/microlitre) - 1 point
The points are added to get the Glasgow Score, which can range from 0 to 8:
Score 0 to 2 points - the likelihood of severe pancreatitis developing is low.
Score 3 to 8 points - the likelihood of severe pancreatitis developing is high.
Balthazar CT severity index[53][97]
Used for grading of severity - not diagnosis.
CT features and score:
I Grade
Normal gland = 0
Focal/diffuse enlargement = 1
Peri-pancreatic inflammation = 2
Single pancreatic fluid collection = 3
Two or more fluid collections or abscess = 4
II Necrosis
None = 0
<30% = 2
30% to 50% = 4
>50% = 6
Morbidity and mortality by Balthazar scoring:
Score 0 to 3 = morbidity 8%; mortality 3%
Score 4 to 6 = morbidity 35%; mortality 6%
Score 7 to 10 = morbidity 92%; mortality 17%
Ultrasound-guided insertion of a non-tunnelled central venous catheter (CVC) into the right internal jugular vein using the Seldinger insertion technique.
How to insert a peripheral venous cannula into the dorsum of the hand.
Use of this content is subject to our disclaimer