Quick Recap
GI & Hepatology System, Protocol 3/13.
1. Definition & Diagnosis
Diagnosis requires 2 of 3: characteristic abdominal pain, serum amylase/lipase >=3x upper limit normal, or characteristic findings on imaging. In patients with characteristic pain + elevated amylase/lipase, imaging is NOT necessary to establish the diagnosis — avoid reflexive CT in straightforward presentations.
2. Initial Imaging
Abdominal ultrasound should be performed to assess for cholelithiasis/choledocholithiasis, especially with suspected gallstone etiology — early gallstone detection dictates ERCP vs cholecystectomy pathway.
Cross-sectional imaging (CECT) is NOT mandatory initially — reserve for diagnostic doubt or complication concern. Contrast-enhanced CT done at initial presentation may UNDERESTIMATE severity, since necrosis can take up to 72 hours to develop — CECT should generally be DELAYED unless there is a specific complication concern or diagnostic uncertainty, not obtained reflexively on admission.
Recurrent pancreatitis (>=2 episodes) workup: IgG4 (autoimmune pancreatitis), triglycerides, calcium, thorough medication review (iatrogenic causes ~2% of cases), endoscopic ultrasound (occult pancreatic cancer if no identified etiology), genetic workup, pancreas divisum evaluation, infectious workup. Consider empiric cholecystectomy for recurrent pancreatitis with no identifiable cause, even with normal LFTs/gallbladder imaging. ~10-20% remain idiopathic despite full workup.
3. Severity Classification (Revised Atlanta Classification, 2012)
Severity | Definition |
Mild | No organ failure, no local/systemic complications — typically recovers with supportive care within a week |
Moderately severe | TRANSIENT organ failure (resolves within 48h) OR local/systemic complication without persistent organ failure |
Severe | PERSISTENT organ failure >48 hours (single or multi-organ) — associated with 33% mortality risk |
The key clinical determinant of severity is persistence of organ failure beyond 48 hours — this single factor matters more than any individual lab value.
4. Severity Prediction Scores
No single reliable severity predictor exists — use multiple tools in combination.
BISAP (Bedside Index for Severity in Acute Pancreatitis) — simple, as accurate as more complex scores, calculated within 24h: 1 point each for BUN >25, impaired mental status, SIRS present, Age >60, Pleural effusion on imaging. Score >=3 = increased complication risk.
Ranson/Glasgow criteria: historical, cumbersome, take up to 48h to complete — largely superseded by BISAP for practical bedside use.
APACHE II >8: usually indicates severe disease, though not pancreatitis-specific.
Other single admission factors associated with severe course: hematocrit >44% (hemoconcentration), obesity, CRP >150 mg/dL, albumin <2.5 mg/dL, calcium <8.5 mg/dL, BUN >20, early hyperglycemia. BUN is a particularly good mortality predictor — BUN >20 OR any RISE in BUN at 24h postadmission both predict increased death risk, making BUN trend (not just baseline) a key monitoring parameter.
CT Severity Index (CTSI): unenhanced CT grade (A-E, 0-4 points) + necrosis score (0% necrosis=0, <33%=2, 33-50%=4, >50%=6) = total score; >=6 indicates severe disease/poor prognosis.
5. ICU Monitoring Indications
All severe acute pancreatitis, PLUS any of: pulse <40 or >150; SBP <80, MAP <60, or DBP >120; RR >35; Na <110 or >170; K <2 or >7; PaO2 <50; pH <7.1 or >7.7; glucose >800; calcium >15; anuria; coma.
6. Immediate Stabilization / Fluid Resuscitation — The Central Intervention
Hypovolemia is the principal driver of pancreatic hypoperfusion and inflammation — rapid intravascular volume restoration is the first and most effective therapy, preventing local ischemia, necrosis progression, SIRS, and multiorgan failure. Early hydration within the first 24 hours decreases morbidity/mortality. Treat ALL suspected acute pancreatitis as if severe until proven otherwise.
Fluid strategy — has genuinely evolved (know this shift):
- Previously, AGGRESSIVE hydration was the cornerstone; recent evidence (de-Madaria et al., NEJM 2022) shows MODERATE resuscitation achieves similar benefit while reducing volume overload risk — aggressive hydration does NOT clearly prevent progression to severe disease and carries its own overload harms
- Minimum starting rate: 250-300 mL/h, targeting urine output >=0.5 mL/kg/h
- Severe volume depletion: initiate 500-1000 mL/h, de-escalate once hypoperfusion signs resolve
- ACG guideline: goal-directed 250-500 mL/h isotonic crystalloid for first 12-24h, reassessed every 6 hours, with BUN decrease from admission value as the therapeutic target
- Lactated Ringer's is PREFERRED over normal saline — RCT evidence shows LR reduces length of stay, ICU admission, and pancreatic necrosis (though no SIRS-score difference at 24h was shown); AVOID LR in hypercalcemic patients (LR contains calcium)
- Routine invasive intravascular monitoring is NOT routinely recommended but may be useful in severe disease
Pain control: IV opiates for the cardinal epigastric pain symptom; PCA pump may be needed for adequate control, since uncontrolled pain itself contributes to hemodynamic instability. Antiemetics (parenteral) for associated nausea/vomiting.
Checklist:
7. Nutrition
Mild pancreatitis: begin oral intake as soon as the patient feels able — no need to withhold feeding pending resolution.
Severe pancreatitis: early enteral nutrition via nasojejunal tube — enteral feeding (not parenteral) is standard, supporting gut barrier integrity and reducing infectious complications.
8. ERCP and Cholecystectomy (Gallstone Pancreatitis)
Early ERCP (<48-72h) for patients with suspected concurrent cholangitis or biliary obstruction.
Cholecystectomy: if a patient with acute pancreatitis has abnormal liver enzymes and an intact gallbladder, cholecystectomy should be performed during the SAME hospital stay or within 30 days of discharge to reduce recurrence risk. ERCP with biliary sphincterotomy for patients who are not surgical candidates.
9. Antibiotics — Explicit Non-Indication (Common Practice Pitfall)
No role for prophylactic antibiotics in preventing infected necrosis, systemic complications, or mortality — multiple trials have failed to show benefit. Fever and leukocytosis EARLY in severe pancreatitis are usually from the inflammatory process ITSELF, not infection — do not reflexively start antibiotics for these signs alone. Infected necrosis is a LATE complication, generally >=2 weeks after symptom onset — antibiotics should NOT be given routinely in necrotizing pancreatitis on this basis. Probiotics have NO role.
Antibiotics ARE appropriate when: documented positive blood cultures, fine-needle aspiration of necrosis confirms infection, OR concurrent cholangitis/other documented infection is present.
Preferred agents for confirmed infected pancreatic necrosis: carbapenems, quinolones, or metronidazole — chosen specifically for necrotic-tissue penetration.
Signs suggesting infected necrosis: gas within the necrotic collection on imaging, persistent fever, bacteremia, clinical deterioration — though these overlap with non-infectious pancreatitis sequelae and can be genuinely hard to distinguish; gas-containing necrosis in a HEMODYNAMICALLY STABLE patient more likely represents a pancreatico-enteric fistula than infection, an important distinguishing consideration before committing to invasive intervention.
10. Local Complications — Classification and Terminology
Entity | Timing | Characteristics |
Acute peripancreatic fluid collection | Within first 4 weeks | Interstitial/edematous pancreatitis, no necrosis, NO wall/encapsulation |
Pancreatic pseudocyst | Matures >4 weeks after onset | Well-circumscribed, well-defined wall, NO nonliquid component |
Acute necrotic collection | Setting of necrotizing pancreatitis | Heterogeneous, nonliquid components, NO definable wall |
Walled-off necrosis | Matures >=4 weeks after necrotizing pancreatitis onset | Heterogeneous, nonliquid components, ENCAPSULATED |
Acute fluid collections occur in up to 40% of severe pancreatitis and MOSTLY RESOLVE SPONTANEOUSLY — little role for drainage unless abdominal compartment syndrome, severe gastric outlet obstruction, or early infection is suspected.
Other complications: peripancreatic artery pseudoaneurysm (gastroduodenal, hepatic, splenic arteries), venous thrombosis (portal, superior mesenteric, splenic veins) — actively screen for these on cross-sectional imaging when obtained.
11. Drainage of Infected Necrosis — The "Step-Up Approach"
Consensus: drainage indicated when necrosis becomes INFECTED, or in STERILE necrosis with a patient who remains clinically unwell (not sterile necrosis alone in an otherwise recovering patient).
Stable patients, even WITH infected necrosis, can be managed CONSERVATIVELY without drainage — clinical stability, not the presence of infection alone, drives the intervention decision.
Current standard: "step-up" approach — endoscopic or percutaneous drainage FIRST, followed by surgery only if necessary — rather than primary open surgical debridement. Delayed drainage (waiting for collections to organize/wall off) is preferred and associated with reduced LOS, pancreatic fistula, and future intervention need — do not rush to early drainage of ill-defined collections unless a specific indication (compartment syndrome, unstable infected necrosis) exists.
Endoscopic approach is now favored over surgical/percutaneous as first-line where technically feasible. Percutaneous drainage preferred when: endoscopic drainage unavailable/not feasible, early-stage necrotizing pancreatitis with ill-defined collections, or collections extending into the paracolic gutters (non-dependent areas poorly served by trans-enteric stents).
Lumen-apposing metal stents: most commonly used for peripancreatic fluid collection drainage — wide diameter enables better drainage and direct endoscopic necrosectomy access; higher technical/clinical success than fully covered metal stents or plastic pigtail stents for walled-off necrosis, but expensive, higher bleeding risk, and optimal indwell time remains unknown.
Unstable patients with infected necrosis or abdominal compartment syndrome usually require drainage — traditionally open surgical debridement (high morbidity/mortality), but compelling evidence now supports endoscopic/percutaneous drainage over open surgery even in this population, with lower morbidity/mortality — minimally invasive approaches (direct endoscopic necrosectomy, percutaneous drainage) are increasingly preferred even for unstable patients where technically feasible.
12. Organ Support
Goal-directed LR resuscitation (Section 6); enteral nutrition (Section 7); pain control; standard ICU supportive care for organ failure (respiratory, renal, cardiovascular) per general critical care principles; RRT if renal failure develops.
13. Consultation Matrix
Consultation | Trigger | Timing |
Gastroenterology | Gallstone pancreatitis (ERCP), endoscopic drainage of collections/necrosis | Per Section 8/11 timing |
Interventional Radiology | Percutaneous drainage indication | As indicated |
Surgery | Cholecystectomy planning, step-up surgical escalation if endoscopic/percutaneous fails | Same admission/within 30 days for cholecystectomy; as needed for necrosis management |
14. Monitoring Framework
Serial BUN trend (key mortality predictor), urine output (fluid resuscitation adequacy), hemodynamic monitoring, serial abdominal exam, repeat imaging only if complication suspected (not routine/reflexive), watch for the 2-week+ window where infected necrosis becomes a realistic concern.
15. Complications
Necrosis (sterile or infected), acute/chronic pseudocyst, peripancreatic pseudoaneurysm, venous thrombosis, abdominal compartment syndrome, pancreatico-enteric fistula, multiorgan failure, ARDS, AKI. Prevention: appropriate goal-directed (not over-aggressive) fluid resuscitation, avoiding unnecessary antibiotics, delayed/step-up drainage approach. Rescue: endoscopic/percutaneous drainage per step-up approach, escalation to surgery only if minimally invasive approaches fail.
16. Escalation & De-escalation
Escalate: persistent organ failure beyond 48h -> severe classification, ICU-level monitoring per Section 5 criteria; suspected infected necrosis in an unstable patient -> drainage per step-up approach.
De-escalate: organ failure resolving within 48h, BUN trending down, tolerating enteral nutrition -> transition to floor-level care, oral diet advancement, outpatient cholecystectomy planning if gallstone-related.
17. ICU Discharge Criteria
Organ failure resolved, hemodynamically stable off aggressive fluid support, tolerating enteral/oral nutrition, pain controlled on oral regimen, gallstone etiology addressed (cholecystectomy performed or planned within 30 days), necrosis/collections stable or appropriately drained.
18. Documentation & Medicolegal Checklist
19. Key Guidelines
ACG (American College of Gastroenterology) Clinical Guideline: Management of Acute Pancreatitis; Revised Atlanta Classification (Banks et al., Gut 2013).
20. Landmark Trials
- de-Madaria E, Buxbaum JL, Maisonneuve P, et al. Aggressive or moderate fluid resuscitation in acute pancreatitis. N Engl J Med. 2022;387:989-1000 — established moderate resuscitation as non-inferior with less volume overload, driving the shift away from maximally aggressive hydration.
- Lee A, Ko C, Buitrago C, et al. Lactated Ringers vs normal saline for mild acute pancreatitis. Gastroenterology. 2021;160:955-957 — supports LR preference.
- Boxhoorn L, van Dijk SM, van Grinsven J, et al. Immediate versus postponed intervention for infected necrotizing pancreatitis. N Engl J Med. 2021;385:1372-1381 — supports delayed/step-up drainage timing.
- Bang JY, Arnoletti JP, Holt BA, et al. Endoscopic transluminal approach reduces complications/costs vs minimally invasive surgery for necrotizing pancreatitis. Gastroenterology. 2019;156:1027-1040.
- Onnekink AM, Boxhoorn L, Timmerhuis HC, et al. (ExTENSION trial): endoscopic vs surgical step-up approach, long-term follow-up. Gastroenterology. 2022;163:712-722.
21. Controversies
Optimal fluid resuscitation rate/type remains an active area even after de-Madaria's trial — "moderate" resuscitation still requires individualization, and the precise rate cutoff between beneficial and harmful remains institution/patient-dependent. Optimal indwell time for lumen-apposing metal stents in walled-off necrosis drainage is not established. The precise threshold for percutaneous vs endoscopic first-line drainage in unstable patients continues to evolve as endoscopic techniques and institutional expertise develop.
22. References
- Fisichella JN, Ahuja N. Acute Pancreatitis. Washington Manual of Critical Care, 4th ed, 2025 (Ch. 53).
- Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis –2012: revision of the Atlanta classification. Gut. 2013;62(1):102-111.
- de-Madaria E, Buxbaum JL, Maisonneuve P, et al. Aggressive or moderate fluid resuscitation in acute pancreatitis. N Engl J Med. 2022;387:989-1000.
- Lee A, Ko C, Buitrago C, et al. Lactated Ringers vs normal saline resuscitation for mild acute pancreatitis. Gastroenterology. 2021;160:955-957.
- Boxhoorn L, van Dijk SM, van Grinsven J, et al. Immediate versus postponed intervention for infected necrotizing pancreatitis. N Engl J Med. 2021;385:1372-1381.
- Bang JY, Arnoletti JP, Holt BA, et al. Endoscopic transluminal approach vs minimally invasive surgery for necrotizing pancreatitis. Gastroenterology. 2019;156:1027-1040.