Acute Pancreatitis

Last updated on Oct 14, 2024

Take home messages

Treatment is supportive
Do not give prophylactic antibiotics
Let them eat

Who designed this?

I'm entirely convinced that if you were to design the pancreas as a concept, or a building, there is absolutely zero chance that you would be granted planning permission.

It is an insanely dangerous organ. A bag of the most lethal digestive enzymes just hanging out right next to all of your important abdominal organs, ready to explode at any moment and liquify everything around it?

Madness.

Either way, somehow evolution has decided it's a fabulous plan and so we're left to deal with the consequences, namely acute pancreatitis.

💡

At any given time in the UK, around 2.5% of ICU bed occupations are due to pancreatitis.

What actually happens?

Inappropriate or excessive release and activation of proteolytic enzymes from acinar cells of pancreas*
Direct damage to pancreas, as well as activation of systemic inflammatory response
Neutrophil invasion of pancreas, with increased release of their own enzymes, free radicals and cytokines
Worsening inflammation and positive feedback leading to progressive pancreatic destruction
Ongoing widespread inflammation then leads to organ dysfunction

*In obstructive causes, the prevention of adequate flow through the pancreatic duct causes premature activation of digestive enzymes, causing direct proteolysis of the pancreatic cells.

Kind of like pulling the pin on a grenade and forgetting to throw it.

What are the causes?

If 'I get smashed' doesn't immediately jump into your mind then that's okay - you're not a third year medical student any more.

The answer is alcohol and gallstones for over 70% of cases.

A more refined way to categorise the causes of acute pancreatitis is by mechanism, rather than a mnemonic that leaves you wondering what that damned 'H' stands for.

What are the common causes of acute pancreatitis?

Idiopathic*

Obstructive

Gallstones
Neoplasm
Cystic fibrosis

Infection

Virus - EBV, CMV, mumps, coxsackie B
Parasites - Clonorchis, Ascaris
Sepsis of any cause

Trauma

Trauma - blunt or penetrating
ERCP

Toxic

Sustained alcohol use
Hypertriglyceridaemia
NSAIDs
Steroids
Azathioprine
Metronidazole
Hypercalcaemia and hyperparathyroidism
Scorpions of course

Ischaemic

Shock and hypoxia
Vasculitis
Cardiopulmonary bypass

Other

Hypothermia
Autoimmune - sclerosing cholangitis
alpha-1-antitrypsin deficiency

Ethanol and Gallstones are the two most common, so if you're asked to list two or more of the most common causes in an OSCE or CRQ question, make sure you mention these two first.

*There is a school of thought that 'idiopathic' just means you haven't looked hard enough for the answer.

Who gets it?

Anyone with a pancreas can get it, but the risk factors are:

Type II diabetes
Smoking
Social deprivation
Obesity

Morbid obesity is unsurprisingly associated with worse outcomes and increased mortality.

How is it diagnosed?

The buzzword here is the revised Atlanta criteria.

The Revised Atlanta Criteria

You need at least two of:

Pancreatitis-sounding abdominal pain
Increase in amylase or lipase to three times normal
Imaging evidence of acute pancreatitis

What investigations should be considered in acute pancreatitis?

VBG with lactate
FBC
Urea + Electrolytes
LFT
Clotting
CRP
Calcium
Glucose
Amylase or lipase
Triglycerides - if you're interested
ECG
Pregnancy test where relevant*

*You can get a raised amylase with ectopic pregnancy.

Imaging

CXR - pneumoperitoneum, major pleural effusions and other significant pulmonary pathology
CT abdomen with contrast - yes contrast even with renal impairment
Ultrasound - looking for gallstones and dilated bile ducts

What are the types?

There are two main categories of acute pancreatitis:

Interstitial oedematous - >80%
Necrotising

As with pretty much every disease process we encounter, the mild version makes up the vast majority of cases, generally self resolving without any major lasting systemic effects.

How bad does it get?

The clever people in Atlanta have clearly been busy, because they've named a severity score as well.

In fact, a lot of people have developed severity scores for acute pancreatitis.

Revised Atlanta Classification

Mild AP - no organ failure
Moderate AP - transient organ failure or complications
Severe AP - organ failure lasting more than 48 hours

Bedside Index for Severity in Acute Pancreatitis (BISAP)

Used to predict severity and prognosis, it is quick and easy to calculate

Age >60
Pleural effusions
Serum urea >25mg/ml
Impaired mental status
2 or more SIRS criteria

Calculation within 24 hours of admission.

Patients with a score of 1 or greater have an increasing risk of mortality.

A score of 2 is suggestive of likely necrosis and organ failure.

A score of 5 had a mortality rate of 22%.

Ranson's Criteria

On admission

White cell count above 16
Age over 55 years
Glucose >10
Raised AST and LDH

After 48 hours

Haematocrit drop greater than 10%
Urea increase by more than 5
Hypocalcaemia*
Hypoxia
Increased base deficit
Massive fluid deficit

A Ranson score of

3 or more is suggestive of severe disease with 15% mortality
5 points = 40% mortality
7 points = 100% mortality

*Hypocalcaemia is a bad sign because it reflects the degree of necrosis. If a whole bunch of lipase is allowed to digest fatty tissues at will, it will produce lots of free fatty acids, which in turn will bind calcium and cause hypocalcaemia.

What are the local complications of pancreatitis?

Pancreatic or peripancreatic collections
Pancreatic pseudocyst
Splenic vein thrombosis
Superior mesenteric vein thrombosis
Portal vein thrombosis
Acute haemorrhage
Pseudoaneurysm formation
Pancreatic fistulae
Abdominal compartment syndrome
Paralytic ileus

💡

Approximately 1 in 5 of patients will get a splenic vein thrombus

How do I look after them?

Mostly you don't, because they're generally admitted to a surgical ward and you as the anaesthetist or ICU person don't get involved.

You just get to see all the really unwell ones.

💡

Treatment of acute pancreatitis is supportive, and that is all.

Nothing new

ABCDE as always
Oxygen if they're hypoxic
Ventilate them if they ain't ventilatin'
Fluids if they're dry - (hint - they're always dry)
Lots of good pain relief - (multimodal but avoid NSAIDs)
Patience

Why are they always dry?

Vomiting
Third spacing
Lack of fluid intake
Fever
Tachypnoea
SIRS-y vasodilatation

As always, you need to adjust your fluid therapy to each patient. Too much fluid is just as bad as not enough, leading to fluid overload, pulmonary oedema and abdominal compartment syndrome.

A lot of studies have been done to try and figure out the correct way to do fluids in acute pancreatitis, and the answer is still we have no idea.

So we should probably stick to what we know:

Boluses with assessment of fluid responsiveness
Titrate to blood pressure, heart rate and urine output
Use vasopressors if inappropriately vasodilated
Replace measured losses
Re-establish enteral intake as early as safely possible

I'm giving co-amox, right?

No. Put the antibiotics down and step away from the drug chart.

💡

Pancreatitis is a sterile inflammatory binfire.

Yes of course it can get infected, and the patient can also pick up a pneumonia or UTI along the way, in which case they would need some antimicrobials.

But antibiotics are not indicated for standard acute pancreatitis - you'll just destroy what little protective gut bacteria they've got left.

You only think about antibiotics if you have good-going evidence of infection, and in that case, you probably want to go straight for the meropenem family, because these appear to penetrate the pancreas best.

If you've reached this point, and have a severely unwell patient, you probably want to throw in an antifungal as well.

Lung stuff

Ah yes, back to an organ we really care about.

Pulmonary complications are understandably common given the diaphragm is the only thing separating the poor delicate lungs from the imploding bleach bag of doom on the other side.

To prevent lung badness:

Proper pain relief, to facilitate adequate tidal volumes and coughing
Avoid excessive fluid administration
Spot and manage ARDS early

Sugar and feeding

It shouldn't come as much of a surprise that acute pancreatitis often triggers a rather sporting hyperglycaemia, for two main reasons:

substantial stress response
loss of islet cells

The evidence suggests you don't need to go crazy with glucose control - less than 10 mmol/Litre is generally fine.

It's more important to avoid hypoglycaemia.

Enteral feeding as soon as possible

Patients with mild pancreatitis should start drinking and eating as soon as they feel well enough to do so.

The pancreas does not require 'rest'
Giving the digestive juices something to play with other than the pancreas itself is probably a good idea

If you've gone five days and not managed to establish adequate intake then it's time to think about supplementing their nutrition.

NG feed brings fewer complications than TPN
NJ is only needed if NG isn't working
TPN is only needed if there is genuine gut dysfunction or fistulae

Should we whip out the gallbladder?

For mild pancreatitis caused by gallstones, it's probably a good idea to cholecystectomise before sending the patient home, but this often isn't logistically feasible for many patients
For severe pancreatitis the risk of cholecystectomy is much higher, so often a long period of recovery is required before surgery is considered
ERCP can be considered for urgent source control if cholecystectomy isn't an option, but clearly this isn't the remit of an anaesthetist

Some other bits

What do I do about necrosis?

Nothing, most of the time
Try and avoid doing any sort of intervention for the first two weeks
If you have to do a laparotomy - for major haemorrhage or gut ischaemia for example - then you try not to disturb the pancreas
Necrosis should be left to 'wall off' over a month or so
If you think necrosis has become infected, you are allowed to start antibiotics
If you think you need source control, consider percutaneous drainage
Endoscopic necrosectomy is not as fun as it sounds

What is a pseudocyst?

A bleb of fluid within the pancreas, with a clear cut wall
Crucially it has little to no evidence of necrosis
Usually start appearing over 4 weeks after the start of symptoms
Can get infected
Usually go away by themselves

What treatments don't work?

Octreotide
Protease inhibitors - Aprotinin, Gabexate mesilate
Prophylactic antibiotics

Example CRQs

The wonderful team over at FRCA-revision.com have put together a series of high-yield and syllabus-mapped CRQs for you to use as part of your revision. They also have an exciting upcoming interactive MCQ quiz which is definitely worth keeping an eye out for.