Rob Orman MD and Cam Berg MD
Take Home Points
- Pancreatitis is present if two of the following three criteria are met; clinical symptoms consistent with pancreatitis, elevated lipase or radiographic imaging suggesting pancreatitis.
- Lipase greater than three times the upper limit of normal is consistent with pancreatitis.
- Patients should receive an ultrasound to rule out the presence of gallstones as this changes management.
- The mainstay of medical management is IV fluids.
- Prolonged NPO status can lead to bowel atrophy and worse outcomes.
- Recheck the hemoglobin if boarding in the ED as a proxy evaluating the effectiveness of the initial fluid resuscitation.
- How is a patient diagnosed with pancreatitis? There are fairly well validated consensus diagnostic criteria for pancreatitis. You need two of three different domains to meet criteria. The first is clinical symptoms. The patient has abdominal pain, usually epigastric, vomiting, tenderness and suspicion for pancreatitis. The second is laboratory criteria. They likely have an elevated lipase. The third is invoked least frequently; radiographic or imaging that suggests pancreatitis.
- The lipase needs to be elevated by at least three times the upper limit of normal. Lipase is a very good test. However, it only performs at its best when we use the appropriate cutoff. The sweet spot appears to be at three times the upper limit of normal. Changing the cutoff to a lower value increases the sensitivity but is less useful because it makes the specificity so poor. Lipase can be elevated with any inflammatory condition in the abdomen; gastroenteritis, diverticulitis, etc. But very few conditions will elevate the lipase to three times the upper limit of normal. It is possible to have pancreatitis with a lower lipase level but most patients will meet this criterion.
- Is there a role for amylase? You should stop using amylase in undifferentiated abdominal pain. Lipase is a superior test in every regard. Amylase becomes abnormal at a slower rate and will miss early cases. It also clears more rapidly. Overall, it is less sensitive and less specific.
- What about patients with a history of chronic pancreatitis and symptoms consistent with pancreatitis but a normal lipase? This pathway is intended to diagnose acute pancreatitis. However, an episode of acute on chronic pancreatitis should still be diagnosed with the pathway.
- What if the patient has an elevated lipase and a normal CT? There have only been a few papers from the 1980s that have looked at this and the imaging was obtained later in the hospitalization. If you think clinically it is pancreatitis and the lipase agrees but the CT is normal, err on the side of your clinical judgment. CT is a good test but relies on identification of peripancreatic inflammation and fat stranding which may not be present early in the presentation. This is why CT isn’t recommended early in the management of pancreatitis. Also, complications of pancreatitis don’t usually develop early in the disease process.
- What is the trigger for imaging patients? Patients with a new diagnosis of pancreatitis should all receive an ultrasound. This is one of the more impactful clinical steps you can do. Patients may have pancreatitis for a handful of discrete reasons.
- In America, about 60-70% of pancreatitis is due to gallstones. Liver function testing is insufficiently sensitive to identify gallstones as an etiology.
- The second most common cause is alcohol consumption. However, patients who drink alcohol may also have gallstones and the management of gallstone pancreatitis is different as it involves procedures.
- The third most common cause is elevated triglycerides.
- Other causes include idiopathic, congenital, structural, toxin mediated, etc. These account for only about 5-10% of cases.
- Ultrasound of the abdomen is low cost and high yield.
- CT should be reserved for cases where there is diagnostic uncertainty or the patient has a complicated presentation or is toxic appearing. Recurrent episodes of pancreatitis shouldn’t necessarily prompt a CT. We often have pushback from accepting physicians who want CT imaging for diagnostic closure. Early CT imaging is associated with higher cost and longer length of stay. We don’t need to diagnose a pseudocyst in the ER. Advanced imaging is more useful later in the hospital course if the patient is failing to respond with empiric therapy.
- Medical management.
- The mainstay is IV fluids. That’s it. Early hydration is strongly associated with superior outcomes in pancreatitis. There are multiple biochemical and neuroendocrine theories why this occurs. You don’t need to drown them. Berg recommends two liters of lactated Ringers (LR) bolused in the ED followed by roughly another two liters in the subsequent twelve hours.
- Consider pancreatitis as similar to sepsis. Pancreatitis is a tremendously inflammatory process causing cytokine release leading to capillary permeability and third spacing of fluids. This is why many of our early pancreatitis patients will satisfy SIRS criteria.
- Why was lactated Ringers included in the pathway? There is fairly compelling literature that lactated Ringers is superior to normal saline. This mitigates hyperchloremia associated with normal saline resuscitation. When they looked at their internal data, they found patients who received lactated Ringers had improved metrics compared to patients receiving normal saline resuscitation.
- Do you need to adjust the fluid resuscitation if the patient has hypercalcemia? If the patient has significant hypercalcemia, you should probably avoid lactated Ringers as it contains some calcium. Lactated Ringers only has about 2.5 meq of calcium per liter so if you start out your resuscitation using LR and then find out the patient has hypercalcemia, you probably haven’t caused any harm.
- Patients may experience hypocalcemia with pancreatitis. If the calcium has decreased, they are at increased odds of adverse events (i.e. death per Ranson criteria). Severe pancreatitis causes calcium sequestration to the pancreatic tissue.
Most common causes of pancreatitis are gallstone and ETOH (worldwide it is parasitic)
Pulmonary effusion (usually l sided)
Amylase is at least 1.5x normal
Amylase sensitivity (Emerg Med J 20:550. Nov 2003)
Lipase At five times upper limit of normal: 60% sensitive, 100% specific Vissers RJ, et al. J Emerg Med. 1999 Nov-Dec;17(6):1027-37
Table 1. Standard Diagnostic Tests.Test Sensitivity Specificity CommentSerum enzymes high moderate > 3x normal increases specificity Ultrasound moderate high Best for gallstones CT moderate high Detects calcifications, fluid collections CT with pancreatic protocol and IV contrast moderate high Detects necrosis
Adapted from (3), www.gastroslides.org
Table 2B. Causes of Increased Serum Lipase.GI Related Non-GI RelatedAcute cholecystitis Diabetic ketoacidosis Bowel obstruction or infarction HIV Duodenal ulceration Macrolipasemia Pancreatic calculus Idiopathic Pancreatic tumors Drugs Post-ERCP/trauma
- Drug induced
- IBD related
Calculations and Mortality Prediction Using CTSI
This scoring system (Table 5 and Figure 4) uses several CT observations including edema and fluid collections. Importantly, the estimates of the amount of necrosis of the gland, as determined by measuring the portion of the gland that perfuses with contrast injection, is highly predictive of necrosis and poor outcome. Stated simply, greater amounts of estimated necrosis are associated with greater rates of extrapancreatic organ failure and pancreatic infection occurring during the course of the episode.
Table 5. CT Severity Index.CT finding PointsGland enlarged 1 Peripancreatic inflammation 2 One fluid collection 3 Multiple fluid collections 4 Necrosis <30% 2 Necrosis 30 – 50% 4 Necrosis >50% 6
Admit Ranson’s-Age >55, WBC>16, Glucose>200, LDH>350, AST>250
In ICU-Decrease in Crit>10, Increase BUN>5, Ca<8,PaO2<60, Base Deficit>4, Fluid Sequestration>6L
Think hemorrhagic if crit is dropping
If the crit is elevated, portends worse prognosis and possible necrotizing pancreatitis (Am J Gastroenterol 2001; 96: 2081-2085)
Conservative treatment of necrotizing pancreatitis with antibiotics decreases mortality rate. Get C-Reactive protein (values>150, point to necrotizing) All studies were done with Imipenem (1 g TID) J Gastrointest Surg 2001 Mar-Apr;5(2), Ann Surg 2000 Nov;232(5), Arch Surg 1997 May;132(5) Cefuroxime and Norfloxacin have also shown benefit.
Uncomplicated acute pancreatitis is a 1 week disease
First week: Inflammation
pancreas and adjacent structures are inflamed (phlegmon)
beer-like hemorrhagic exudate
Second Week: Necrosis
can be localized or spread to affect the entire retroperitoneum
this may resolve or become walled off as a pseudocyst
may become secondarily infected
Third Week: Infection
colonic bugs or superinfection with candida
Sterile and infected pancreatic necrosis are almost impossible to tell apart clinically
Fourth Week: Late phase
can be abscess formation
Severity of Illness
apache ii is better than ranson
>8 is severe
these folks do not need ct scan in the first week, they do need an abd uts if possibility exists for stone disease
develops in 1/5 of the patients with Apache>8
get a CT at the end of the first week
parenchyma that does not pick up contrast is necrosed
you then must ask is it infected and is an operation indicated
Indications for necrosetomy
1. Infected necrosis
2. clinical deterioration after two weeks of supportive care
3. Suspicion of infection if >50% of the pancreas is necrotic
4. Extensive necrosis and prolonged ileus at least two weeks from onset
many current experts would precede with conservative management as long as possible until forced to operate by infection
Back to top
generally observed 6 weeks for maturation
If infected (FNA) then percutaneous drainage is usually the best choice
Back to top
3rd gen and flagylBack to top
gastrinoma-zollinger EllisonBack to top
diarrhea and hypokalemiaBack to top
flushing and diarrhea
Less common causes pancreatitis: hypertrygliceridemia (at least >1000 unless on estrogen), TTP, sickle cell, incomplete cystic fibrosis, Sjogren’s, Tylenol us
Back to top
Management of crit care pt with Pancreatitis
(Crit Care Med 2004;32(12):2524) maybe antibiotics in necrotizing but this article recommends against empirical
jejunal enteral nutrition as soon as access can be obtained
tight glycemic control
FNA of necrotic pancrea to determine infection
ERCP within 72 hrs of gallstone pancreatitis
A raised serum amylase activity is found in several different clinical contexts. In patients presenting with acute abdominal pain, acute pancreatitis is the most likely diagnosis, especially if amylase activity is markedly raised (> 10 times the upper limit of normal). Causes of more modest elevation include ruptured tubal pregnancy, perforated hollow viscus, small bowel obstruction, generalised peritonitis, diabetic ketoacidosis, and the administration of opiates (which cause spasm of Oddi’s sphincter). A different clinical scenario exists when the salivary glands are inflamed, most often with calculi or mumps; the resultant hyperamylasaemia is not usually a diagnostic problem. Asymptomatic hyperamylasaemia has several causes. Pancreatic pseudocysts may form during or after episodes of acute pancreatic necrosis. The amylase activity within these pseudocysts may be extremely high, and leakage into the blood may be associated with marked hyperamylasaemia. Sometimes amylase molecules combine with plasma proteins (usually immunoglobulins). The resultant complexes are so large that they can be precipitated by certain agents (such as polyethylene glycol) and cannot be cleared by the kidneysin contrast with amylase, which has a relatively low molecular mass, doesn’t precipitate (much), and is readily excreted except in renal failure. Macroamylase is a well recognised phenomenon, at least by laboratory staff, and if suspected can be readily confirmed or excluded.1 In this patient, very high amylase activity was seen in urine (12 218 U/l), and only 9% of amylase activity precipitated (reference interval < 57%), thus excluding macroamylasaemia. Production of amylase (usually the salivary isoenzyme) has been documented with various kinds of ovarian neoplasm. Serous cystadenocarcinoma, mucinous tumour, surface papillary carcinoma, and endometrial adenocarcinoma of the ovary have all been implicated.2-5 Other malignancies reported to secrete amylase include gastric carcinoma, multiple myeloma, lung carcinoma, and phaeochromocytoma.6-9 The basis of amylase production by malignant tissue is unclear, although its secretion by tissues known to contain amylase might be activated through pathological processes such as inflammation or tumour formation.10
Pancreas March 2010 – Volume 39 – Issue 2 – pp 248-251
Conclusion: Total enteral nutrition is better than total parenteral nutrition in the prevention of pancreatic necrotic infection in severe acute pancreatitis.
NJ did not seem to confer benefit over NG feeding (Nasogastric or nasojejunal feeding in predicted severe acute pancreatitis: a meta-analysis. Crit Care 2013, 17:R120.)
From MarylandCC Project:
- Most common causes: Gallstones, alcohol abuse, hypertriglyceridemia (beware of propofol use!), hypercalcemia, Idiopathic will usually be microlithiasis of the gall bladder (not seen on UTS consider cholecystect)
- Acute pancreatitis causes a distributive shock (like sepsis)
- High frequency of end organ & cardiac dysfunction
- A pleural effusion is one of the strongest predictors of severe pancratitis
- ATN is common
- Encephalopathy secondary to phospholipid A2 damage
- Secondary infection is the most significant complication of severe pancreatitis
- Occurs in 40-70% of patients
- Does not occur until Phase II of disease (after week 2)
- Caused by intestinal translocation, possible reperfusion injury, pancreatic necrosis
- Enough with Ranson’s criteria!! CT Severity Index is the way to go when determining pancreatitis severity, mortality risk, & disposition
- Fluids: Patients with acute pancreatitis often need A LOT of IV fluids to address their distributive shock
- Balanced crystalloid solutions are preferred
- Goal directed resuscitation is critical (lactate normalization, echo guided, use of dynamic indicators) because over resuscitation will lead to significant gut edema, intra-abdominal hypertension/abdominal compartment syndrome, renal failure, etc
- Nutrition: An important part of management where we can significantly impact outcomes
- When: Start enteral nutrition EARLY (within 48 hours of admission)
- Jejunal feeds if severe pancreatitis
- Gastric feeds can be started in mild/moderate pancreatitis – just be on the lookout for gastric outlet obstruction!
- TPN: Just don’t do it, it’s the devil’s excrement
- Antibiotics: No clear evidence to support routine prophylactic antibiotic use for acute pancreatitis.
- If you are going to empirically give, make sure to take them away once cultures come back negative
- Intra-abdominal infections often don’t occur until after week 2
- Abdominal Compartment Syndrome: Common with severe pancreatitis
- Decompress everything first (ascites, bladder, effusions)
- Prokinetic GI agents may benefit
- Possible role for early continuous renal replacement therapy
- Mentula P, Leppäniemi A. Position paper: timely interventions in severe acute pancreatitis are crucial for survival. World J Emerg Surg. 2014;9(1):15. [Free Full Text]
- Tenner S, Baillie J, Dewitt J, Vege SS. American College of Gastroenterology guideline: management of acute pancreatitis. Am J Gastroenterol. 2013;108(9):1400-15. [PubMed]
- Jiang K, Huang W, Yang XN, Xia Q. Present and future of prophylactic antibiotics for severe acute pancreatitis. World J Gastroenterol. 2012;18(3):279-84. [Free Full Text]
- Li JY, Yu T, Chen GC, et al. Enteral nutrition within 48 hours of admission improves clinical outcomes of acute pancreatitis by reducing complications: a meta-analysis. PLoS ONE. 2013;8(6):e64926. [Free Full Text]
Back to top