Severe Acute Pancreatitis — ICU Management

📘 1 · Washington Manual of Critical Care, 4th Ed

Definition, Diagnosis & Aetiology

"Acute pancreatitis is diagnosed when two of three criteria are met: characteristic abdominal pain, serum lipase or amylase ≥3× the upper limit of normal, or characteristic findings on cross-sectional imaging. The great majority of episodes are mild and self-limiting; it is the 15–20% that progress to persistent organ failure and necrosis that consume ICU resources and account for almost all of the mortality. The clinician's task in the first 24 hours is not to confirm the obvious diagnosis, but to identify the patient who is going to deteriorate."

Washington Manual of Critical Care, 4th Ed. Wolters Kluwer 2023. Chapter: Acute Pancreatitis.

Diagnostic Criteria (2 of 3 required)

Characteristic pain — acute, severe, epigastric, radiating to the back
Lipase or amylase ≥3× ULN — lipase is more sensitive and specific and stays elevated longer; the degree of elevation does not correlate with severity
Characteristic imaging — contrast CT, MRI or ultrasound (imaging is not needed to diagnose if the first two are present)

Aetiology (mnemonic: "I GET SMASHED" — the big ones)

Gallstones (~40%) — commonest; always request an ultrasound to look for stones/CBD dilatation on every first presentation
Ethanol (~30%) — usually a background of chronic use
Hypertriglyceridaemia (~2–5%, higher in India) — triglycerides >11.3 mmol/L (1000 mg/dL); often lipase is only mildly raised
ERCP — post-procedure; rectal NSAID (diclofenac/indometacin) reduces the risk
Drugs — azathioprine, valproate, thiazides, GLP-1 agonists, some antiretrovirals
Hypercalcaemia, autoimmune, trauma, scorpion sting, idiopathic (~10–15%)

⚠️ Why severe pancreatitis behaves like septic shock

In severe acute pancreatitis there is a massive sterile inflammatory response with capillary leak, third-space fluid loss and vasodilatation. The patient is tachycardic, hypotensive, oliguric and has a raised lactate and white cell count — without any infection in the first week. Treating this SIRS phase with antibiotics is a classic error: the resuscitation is supportive (fluids, organ support), not antimicrobial.

📗 2 · Marino's The ICU Book, 5th Edition (2025)

Pathophysiology & the Two-Phase Timeline

"Acute pancreatitis is an inflammatory condition that can range from mild self-limited disease to a life-threatening multi-organ failure syndrome. The inciting event triggers premature intrapancreatic activation of digestive enzymes, leading to autodigestion of pancreatic and peripancreatic tissue. The subsequent inflammatory cascade, particularly in severe disease, extends far beyond the pancreas and drives a SIRS-like systemic response indistinguishable from septic shock."

Marino PL. The ICU Book, 5th Ed. GI Failure. Wolters Kluwer; 2025.

🔬 How acute pancreatitis progresses

TriggerGallstone · Alcohol · Hypertriglyceridaemia · ERCP

Premature trypsinogen activation inside the pancreatic acinar cell

Active trypsin activates every other digestive zymogen

Autodigestion of pancreatic parenchyma & peripancreatic fat

Local injury: pancreatic & fat necrosisfree fatty acids chelate Ca²⁺ → hypocalcaemia

Systemic: macrophage & neutrophil activationcytokine storm — TNF-α, IL-1, IL-6

SIRS — capillary leak, vasodilatation, third-space fluid loss

Pathological stateMulti-organ failure — ARDS, AKI, distributive shock, DIC

📗 Marino — the autodigestion cascade

Digestive enzymes are normally synthesised as inactive zymogens and activated only in the duodenum. In pancreatitis this control fails:

Trigger (stone, alcohol, hypertriglyceridaemia, ERCP) → premature trypsinogen activation inside the acinar cell
Active trypsin activates every other zymogen → autodigestion of pancreatic parenchyma and peripancreatic fat
Phospholipase A₂ → membrane disruption → local necrosis; released fatty acids bind calcium → hypocalcaemia
Macrophage/neutrophil activation → cytokine storm (TNF-α, IL-1, IL-6) → SIRS → distant organ injury (ARDS, AKI, cardiovascular collapse, DIC, gut-barrier failure)

📗 Marino — the two phases (this drives every management decision)

Early phase (week 1): the clinical picture is dominated by the systemic inflammatory response and organ failure. Severity is defined by the presence and persistence of organ failure — not by imaging. Deaths in this phase are from multi-organ failure.

Late phase (after week 1, weeks–months): dominated by local complications and infected necrosis. This is when CT findings matter and when infection (and the need for intervention) appears. Deaths in this phase are from infected necrosis and its complications.

📗 Marino — feed the gut, don't "rest" the pancreas

The old doctrine of prolonged "pancreatic rest" (nil-by-mouth until amylase normalises) is wrong and harmful. Early enteral nutrition maintains gut mucosal integrity, prevents bacterial translocation from the gut (the source of later pancreatic infection) and reduces infectious complications. TPN alone increases line sepsis and infectious morbidity. The pancreas tolerates enteral feed — it is the systemic response, not the pancreas, that needs support.

📋 3 · ACG 2013 · IAP–APA · Revised Atlanta 2012

Evidence-Based Guidelines

Severity — Revised Atlanta Classification (2012)

Organ failure is defined by a modified Marshall score ≥2 in any of three systems (respiratory PaO₂/FiO₂, renal creatinine, cardiovascular systolic BP).
• Mild: no organ failure, no local complications — ~80%, mortality <1%
• Moderately severe: transient organ failure (<48h) or local complications — mortality ~5%
• Severe: persistent organ failure (>48h), single or multiple — mortality 20–40% (higher with infected necrosis) Revised Atlanta 2012

Predicting severity — bedside scores

No score beats serial clinical assessment, but useful triage tools:
BISAP (≥3 = high risk) — 1 point each: BUN >25 mg/dL, Impaired mental status, SIRS, Age >60, Pleural effusion.
CTSI = Balthazar grade + necrosis score (≥7 = severe).
Persistent SIRS at 48h and a rising BUN/haematocrit are among the best single predictors of severe disease. Ranson's score requires 48h to complete and is largely of historical interest.

Fluid resuscitation — moderate, not aggressive WATERFALL 2022 · ACG 2013

Lactated Ringer's is preferred over normal saline — less hyperchloraemic acidosis and lower SIRS at 24h.
Moderate resuscitation (WATERFALL 2022): if hypovolaemic, 10 mL/kg bolus, then 1.5 mL/kg/h, reassessed at 12h. The trial's aggressive arm (20 mL/kg bolus → 3 mL/kg/h) was stopped early for fluid overload (20.5% vs 6.3%) with no reduction in progression to severe disease.
Titrate to targets, not a fixed rate: HR <120, MAP 65–85, urine output >0.5 mL/kg/h, falling BUN/haematocrit. Reassess frequently and stop escalating once perfusion is restored.

Nutrition — early enteral STRONG — ACG 2013 / ESPEN 2020

Mild disease: start a low-fat oral diet as soon as pain and nausea allow — do not wait for enzymes to normalise.
Predicted severe disease: enteral nutrition within 24–72h via nasogastric tube. NG feeding is equivalent to nasojejunal (NJ) for tolerance and outcome, and is simpler.
TPN only if the enteral route fails or is contraindicated (e.g. abdominal compartment syndrome, high-output fistula). Target ~25 kcal/kg/day, protein 1.2–1.5 g/kg/day.

Antibiotics — NOT prophylactically STRONG 1A — ACG 2013

No prophylactic antibiotics, even in severe or necrotising pancreatitis (Isenmann 2004, Dellinger 2007 — no benefit; increases MDR and fungal infection).
Antibiotics only for: confirmed infected necrosis (gas in a collection on CT, or clinical deterioration in the late phase), concurrent cholangitis, or a proven extrapancreatic infection.
Choice for infected necrosis: a carbapenem (e.g. meropenem — good pancreatic penetration). Routine FNA to prove infection is no longer recommended; treat on the clinical + CT picture.

ERCP & cholecystectomy in gallstone pancreatitis

Urgent ERCP (<24h): only if there is concurrent acute cholangitis or persistent biliary obstruction — not for severe gallstone pancreatitis by itself (APEC trial: no benefit without cholangitis).
Cholecystectomy: same-admission for mild gallstone pancreatitis (PONCHO trial — fewer recurrences); delay until fluid collections resolve or stabilise (~≥6 weeks) in severe disease.

Managing necrosis — the step-up approach STRONG — PANTER / IAP–APA

Delay intervention for infected necrosis until the collection is walled-off (≥4 weeks) whenever the patient can be supported — mortality is far lower.
Step-up, not open surgery: percutaneous or endoscopic (transluminal) drainage first, escalating to minimally invasive necrosectomy only if needed. The PANTER trial showed the step-up approach reduces major complications and new-onset organ failure versus primary open necrosectomy. Sterile necrosis, however extensive, does not need intervention if the patient is stable.

🇮🇳 Indian Context

Hypertriglyceridaemia-induced pancreatitis is common in Indian diabetics and metabolic syndrome. Triglycerides >1000 mg/dL (11.3 mmol/L); lipase may be only modestly raised. Treat with an insulin infusion (activates lipoprotein lipase) ± plasmapheresis if severe/organ failure, alongside standard pancreatitis care.

Tropical (calcific) pancreatitis — recurrent acute-on-chronic episodes in young patients, historically linked to cassava; present in parts of Kerala and eastern India.

Access: endoscopic step-up necrosectomy and interventional radiology are concentrated in tertiary centres — recognise infected necrosis early and refer, rather than starting open surgery locally.

💊 4 · Drug Doses

Drug Reference — Severe Acute Pancreatitis

Drug / Fluid	Indication	Dose	Notes
Lactated Ringer's	Resuscitation (all)	10 mL/kg bolus if hypovolaemic → 1.5 mL/kg/h; titrate to targets	Preferred over NS; moderate regimen (WATERFALL); reassess at 12h, avoid overload
Fentanyl	Analgesia (first-line IV)	25–50 μg IV boluses titrated, or infusion	Opioids are safe — the old "morphine causes sphincter of Oddi spasm" concern is not clinically significant
Paracetamol	Analgesia (adjunct)	1 g IV/PO QDS (max 4 g/day)	Opioid-sparing; reduce dose if low body weight or liver dysfunction
Ondansetron	Nausea / vomiting	4–8 mg IV TDS PRN	Enables early enteral feeding
Insulin (regular) infusion	Hypertriglyceridaemia pancreatitis	0.1–0.3 U/kg/h with 5% dextrose cover; target TG <5.6 mmol/L (500 mg/dL)	Activates lipoprotein lipase; monitor glucose & K⁺; add plasmapheresis if severe
Meropenem	Infected necrosis / cholangitis	1 g IV every 8h (renal-adjusted)	Good pancreatic penetration; only when infection confirmed/strongly suspected
Enoxaparin	VTE prophylaxis	40 mg SC OD (renal-adjusted)	Pancreatitis is prothrombotic — splanchnic (splenic/portal) vein thrombosis risk
Calcium gluconate	Symptomatic hypocalcaemia	10 mL of 10% IV slowly	Treat symptomatic/ionised hypocalcaemia only; correct magnesium too
Rectal indometacin/diclofenac	Post-ERCP prophylaxis	100 mg PR at time of ERCP	Prevention only — reduces post-ERCP pancreatitis in high-risk cases

🗺 5 · Clinical Flowchart

Management Algorithm

Diagnose & find the cause

2 of 3: pain, lipase ≥3× ULN, imaging
Bloods: FBC, U&E, LFTs, calcium, triglycerides, glucose, lactate, CRP; ABG if unwell
Ultrasound abdomen for gallstones/CBD on every first episode
Do not CT at presentation — it under-stages necrosis in the first 72h

Stratify severity & decide disposition

BISAP, SIRS, BUN/haematocrit trend; modified Marshall for organ failure
Persistent organ failure >48h, or high BISAP/SIRS → HDU/ICU

Resuscitate (moderate) & support organs

Lactated Ringer's — moderate regimen, titrate to HR/MAP/UO/BUN
Analgesia (fentanyl + paracetamol), antiemetics, VTE prophylaxis
Organ support: O₂/ventilation for ARDS, RRT for AKI as needed
No antibiotics unless cholangitis or proven infection

Feed early

Mild: low-fat oral diet when tolerated
Severe: NG enteral feed within 24–72h (NG = NJ); TPN only if EN fails

Watch for complications (late phase)

Contrast CT at 72–96h+ if not improving — defines necrosis/collections
Deterioration in week 2–4 = suspect infected necrosis → carbapenem + plan drainage
Monitor for abdominal compartment syndrome (bladder pressure), splanchnic vein thrombosis
Infected necrosis: delay ≥4 weeks (walled-off) then step-up drainage → minimally invasive necrosectomy

⚠️ 6 · Common Mistakes

Common Mistakes in Acute Pancreatitis

❌ Mistake 1 — Prolonged nil-by-mouth "pancreatic rest"

Amylase/lipase normalisation is irrelevant to feeding — enzymes stay up for days after clinical recovery. Early enteral nutrition (oral in mild disease; NG within 24–72h in severe) reduces infected necrosis and mortality. Prolonged fasting and TPN worsen outcomes.

❌ Mistake 2 — Prophylactic antibiotics for severe/necrotising disease

The early-phase SIRS looks like sepsis but is sterile. RCTs show no benefit of prophylactic carbapenems/quinolones in sterile necrosis, and routine use breeds MDR organisms and fungal infection. Give antibiotics only for cholangitis or confirmed infected necrosis.

❌ Mistake 3 — Aggressive "front-loaded" fluid resuscitation

The WATERFALL trial (NEJM 2022) stopped its aggressive arm early for fluid overload with no reduction in progression to severe disease. Use moderate resuscitation with Lactated Ringer's, titrated to perfusion targets and a falling BUN/haematocrit — and stop escalating once perfused.

❌ Mistake 4 — Normal saline instead of Lactated Ringer's

Large-volume NS causes hyperchloraemic metabolic acidosis and more SIRS at 24h. LR is the preferred resuscitation fluid (avoid LR only in the rare patient with significant hypercalcaemia as the cause).

❌ Mistake 5 — CT scanning too early

Contrast CT in the first 72h under-estimates necrosis and rarely changes early management. Reserve CT for diagnostic doubt or for patients not improving at 72–96h, when it defines necrosis and collections.

❌ Mistake 6 — Operating on necrosis too early

Early open necrosectomy carries very high mortality. Where the patient can be supported, delay intervention until the collection is walled-off (≥4 weeks) and use a step-up (percutaneous/endoscopic drainage first) approach. Sterile necrosis in a stable patient needs no intervention.

❌ Mistake 7 — Missing the treatable cause

Always send triglycerides and calcium and image for gallstones. Hypertriglyceridaemic pancreatitis needs an insulin infusion (± plasmapheresis); gallstone pancreatitis with cholangitis needs urgent ERCP; and mild gallstone pancreatitis needs a same-admission cholecystectomy to prevent recurrence.

🎓 7 · Exam Pearls — DrNB / PDCC / IFCCM / EDIC

Exam Pearls

Q: How is the severity of acute pancreatitis defined (Revised Atlanta 2012)?
By organ failure, not imaging. Mild = no organ failure/no local complications; moderately severe = transient organ failure (<48h) or local complications; severe = persistent organ failure >48h. Organ failure = modified Marshall score ≥2 (respiratory/renal/cardiovascular).

Q: What did the WATERFALL trial (NEJM 2022) change?
Moderate fluid resuscitation (10 mL/kg bolus → 1.5 mL/kg/h) is preferred over aggressive (20 mL/kg → 3 mL/kg/h). The aggressive arm was stopped early for fluid overload (20.5% vs 6.3%) with no reduction in progression to severe disease.

Q: Why Lactated Ringer's over normal saline?
LR reduces SIRS at 24h and avoids the hyperchloraemic acidosis of large-volume NS. (Avoid LR only if hypercalcaemia is the precipitant.)

Q: Components of the BISAP score?
BUN >25 mg/dL, Impaired mental status, SIRS, Age >60, Pleural effusion — 1 point each; ≥3 predicts high mortality. Simple and available within 24h (unlike Ranson, which needs 48h).

Q: Do you give prophylactic antibiotics in necrotising pancreatitis?
No. No benefit in sterile necrosis (Isenmann, Dellinger). Antibiotics (a carbapenem) only for confirmed/strongly suspected infected necrosis, cholangitis, or extrapancreatic infection.

Q: How do you diagnose infected necrosis, and when do you intervene?
Gas within a collection on CT, or clinical deterioration in the late phase — routine FNA is no longer recommended. Intervene by the step-up approach and, where possible, delay until walled-off necrosis (≥4 weeks). PANTER trial: step-up reduces major complications vs open necrosectomy.

Q: When is urgent ERCP indicated in gallstone pancreatitis?
Only with concurrent cholangitis or persistent biliary obstruction — not for severe gallstone pancreatitis alone (APEC trial). Mild gallstone pancreatitis needs same-admission cholecystectomy (PONCHO trial).

Q: How is hypertriglyceridaemia-induced pancreatitis managed?
Insulin infusion (activates lipoprotein lipase) with dextrose cover, targeting TG <500 mg/dL; add plasmapheresis for severe disease/organ failure; then standard pancreatitis care and long-term fibrate/lifestyle management.

Q: Why does hypocalcaemia occur, and should you correct it?
Released free fatty acids chelate calcium (saponification). Correct only symptomatic or significant ionised hypocalcaemia, and replace magnesium — routine correction of an asymptomatic number is unnecessary.

📚 8 · References

References

Tenner S, Baillie J, DeWitt J, Vege SS (ACG). American College of Gastroenterology Guideline: Management of Acute Pancreatitis. Am J Gastroenterol. 2013;108:1400–1415.
Banks PA, Bollen TL, Dervenis C, et al. Classification of Acute Pancreatitis — 2012: Revision of the Atlanta Classification and Definitions. Gut. 2013;62:102–111.
de-Madaria E, Buxbaum JL, Maisonneuve P, et al. (WATERFALL). Aggressive or Moderate Fluid Resuscitation in Acute Pancreatitis. N Engl J Med. 2022;387:989–1000.
Working Group IAP/APA Acute Pancreatitis Guidelines. IAP/APA evidence-based guidelines for the management of acute pancreatitis. Pancreatology. 2013;13(4 Suppl 2):e1–e15.
van Santvoort HC, Besselink MG, Bakker OJ, et al. (PANTER). A Step-up Approach or Open Necrosectomy for Necrotizing Pancreatitis. N Engl J Med. 2010;362:1491–1502.
Schepers NJ, Hallensleben NDL, Besselink MG, et al. (APEC). Urgent endoscopic retrograde cholangiopancreatography with sphincterotomy versus conservative treatment in predicted severe acute gallstone pancreatitis. Lancet. 2020;396:167–176.
da Costa DW, Bouwense SA, Schepers NJ, et al. (PONCHO). Same-admission versus interval cholecystectomy for mild gallstone pancreatitis. Lancet. 2015;386:1261–1268.
Arvanitakis M, Ockenga J, Bezmarevic M, et al. (ESPEN). ESPEN guideline on clinical nutrition in acute and chronic pancreatitis. Clin Nutr. 2020;39:612–631.
Crockett SD, Wani S, Gardner TB, et al. (AGA). American Gastroenterological Association Institute Guideline on Initial Management of Acute Pancreatitis. Gastroenterology. 2018;154:1096–1101.
Marino PL. Marino's The ICU Book, 5th Edition. GI Failure; Nutrition in Critical Illness. Philadelphia, PA: Wolters Kluwer; 2025.
Washington Manual of Critical Care, 4th Edition. Kollef MH, Witt CA (eds). Acute Pancreatitis. Wolters Kluwer; 2023.