Acute pancreatitis: interventional management and complications · ICU Topics

Cinematic clinical photograph of a percutaneous drainage catheter beside an endoscopic transluminal necrosectomy setup, ICU setting, clinical-blue lighting, no text, no people — FigureEndoscopic step-up beats surgical step-up — fewer fistulae and a shorter stay (TENSION trial).

Necrotising pancreatitis with walled-off necrosis and gas locules indicating infection — FigureInfected necrosis drives intervention timing — wall-off ideally beyond 4 weeks; gas on CT is highly suggestive of infection.

Step-up pathway from antibiotics and percutaneous drainage to endoscopic necrosectomy or VARD — FigureStep-up (PANTER): antibiotics ± PCD first; escalate to endoscopic or minimally invasive necrosectomy; open surgery last.

The step-up approach: concept and evidence

The modern management of (infected) pancreatic necrosis is the step-up approach — a graduated, least-invasive-first strategy that replaced primary open necrosectomy after the landmark PANTER trial (2010). The principle: many patients improve with drainage alone and never need necrosectomy; escalation is reserved for those who fail the lower step.[1][10]

The ladder has four rungs: [1]

Antibiotics + supportive care (carbapenem for good pancreatic penetration) — sterile necrosis is managed medically; suspected infection triggers carbapenem and reassessment.
Catheter drainage — percutaneous (CT/US-guided) OR endoscopic ultrasound-guided transluminal (cystgastrostomy/cystduodenostomy). This alone resolves infection in roughly a third of patients.
Minimally invasive necrosectomy — endoscopic (through the transluminal tract) or percutaneous retroperitoneal (VARD / endoscopic). Multiple sessions, days to weeks apart.
Open necrosectomy — last resort, for failure of step-up, uncontrolled bleeding, bowel perforation, or abdominal compartment syndrome. [1]

The evidence base is one of the strongest in interventional pancreatology and is directly examinable: three randomised trials define the field. [1]

The three pivotal RCTs

PANTER — step-up vs primary open necrosectomy

Design: multicentre RCT, 88 patients with infected necrotising pancreatitis; step-up (percutaneous/endoscopic drainage → minimally invasive retroperitoneal necrosectomy if failed) vs primary open necrosectomy.[1]

Primary endpoint (major complications/death composite): step-up 35% vs open 69% — step-up superior (RR 0.51, 95% CI 0.33–0.77). [1]

Key findings:

35% of step-up patients resolved with drainage ALONE — never needed necrosectomy.
Fewer new-onset multi-organ failure (12% vs 40%).
No statistically significant difference in mortality (≈11–19% both arms), but step-up trended lower.
Shorter hospital and ICU stay, lower healthcare costs (Drain, Retrospective cost analysis). [1]

Take-home for exam: the step-up approach reduced the composite of major complications or death by half, and a third of patients were spared any necrosectomy. This is now the standard of care and the basis of all guideline recommendations.

[1]

PANTER 11-year follow-up (Hollemans 2019)

Design: long-term follow-up of the PANTER cohort (median ~6 years).[2]

Findings — step-up remained superior over the long term:

Less new-onset diabetes (step-up lower incidence than open).
Less exocrine insufficiency (pancreatic enzyme dependence).
Fewer incisional hernias (a direct consequence of avoiding laparotomy).
No difference in mortality (confirmed over the long term).
No difference in quality of life at final follow-up. [1]

Take-home: the long-term data show the step-up approach does not simply shift morbidity into the future — it genuinely reduces endocrine, exocrine, and abdominal-wall sequelae. This is the trial to cite when asked about endocrine/exocrine insufficiency and incisional hernia after necrosectomy.

[1]

TENSION — endoscopic step-up vs surgical step-up

Design: multicentre RCT, 98 patients with infected necrotising pancreatitis; endoscopic step-up (EUS-guided transluminal drainage → endoscopic necrosectomy if failed) vs surgical step-up (percutaneous drainage → VARD if failed).[3][4]

Primary endpoint (composite: major complications/death): no significant difference in the composite (≈43% endoscopic vs 45% surgical). [1]

Key secondary findings favouring endoscopic:

Fewer pancreaticocutaneous and enterocutaneous fistulae (5% vs 28%).
Shorter length of hospital stay.
Fewer interventions (necrosectomy sessions) per patient.
Trend to fewer new-onset diabetes. [1]

Take-home: when both minimally invasive routes are available, endoscopic transluminal step-up is preferred over the percutaneous/surgical (VARD) step-up — primarily because it avoids fistulae through the abdominal wall. Surgical step-up (VARD) remains valuable when the collection is not apposed to the gastric/duodenal wall (no endoscopic window) or is paracolic/inferior.

[1]

PENGUIN — endoscopic transgastric vs surgical necrosectomy

Design: RCT, 22 patients; endoscopic transgastric necrosectomy vs surgical (open/VARD) necrosectomy for infected necrotising pancreatitis.[5]

Findings: endoscopic group had a lower composite of major complications/death and lower IL-6 (less inflammatory response) after the index procedure. [1]

Caveat: small single-centre Dutch trial, used as supporting evidence and built the foundation for the larger TENSION trial. Often cited together with TENSION to support the endoscopic-first philosophy.

[1]

Step-up approach

Step-up approach for infected necrosis

Delay intervention ~4 weeks

Allow necrotic tissue to demarcate from healthy pancreas (wall-off). During this time: antibiotics (carbapenem — good pancreatic penetration), supportive care, percutaneous catheter drainage if sepsis uncontrolled. Premature surgery (<4 weeks): high mortality (30-40%), poor tissue planes, risk of bleeding.

Step 1: Percutaneous or endoscopic drainage

Percutaneous catheter drainage (CT-guided — radiology). OR endoscopic ultrasound-guided transluminal drainage (via stomach/duodenum into the necrotic collection — gastroenterology). Both decompress and drain infected material. Many patients improve with drainage alone — no need for necrosectomy (PANTER trial: 35% avoided necrosectomy).

Step 2: Add necrosectomy if drainage insufficient

If ongoing sepsis despite adequate drainage: add minimally invasive necrosectomy. Endoscopic: through the drainage tract (stomach/duodenum) — use a gastroscope, remove necrotic tissue with snares/baskets. Percutaneous: through a percutaneous tract (VARD — video-assisted retroperitoneal debridement). Multiple sessions often needed.

Step 3: Open surgery (LAST RESORT)

Reserved for: failed minimally invasive approach, massive bleeding, bowel perforation, abdominal compartment syndrome. Open necrosectomy: high mortality (15-30%), prolonged ICU stay, fistula formation. Avoid if possible — minimally invasive step-up is superior (PANTER/TENSION/PENGUIN trials).

[1] [2] [3] [10]

Timing of intervention — when to act

Default: DELAY ~4 weeks (let it wall off)

Sterile necrosis: medical management only (no antibiotics unless infection proven). Wait for the necrotic collection to organise into walled-off necrosis (WON) — a mature inflammatory wall makes drainage/necrosectomy safer, cleaner, and more effective. The Atlanta-2012 "4-week" threshold is the cornerstone.<Cite id="6" /> Premature (<4 wk) surgery: mortality 30–40% from bleeding, poor planes, and organ failure.

Exception: intervene EARLY only if clinically deteriorating

If a patient with confirmed/suspected infected necrosis is deteriorating (refractory septic shock, rising lactate, worsening organ failure) you cannot wait 4 weeks. Place a percutaneous drain as source control (the least-invasive bridge) — this buys time and may be definitive. Definitive necrosectomy is still deferred until the collection walls off.

How infection is confirmed

(1) Positive Gram stain/culture from FNA of necrotic tissue (gold standard, rarely needed in practice); OR (2) gas in the peripancreatic collection on CT PLUS clinical deterioration. Gas alone is not diagnostic (can be sterile "gas-forming" or enteric communication) — pair imaging with the clinical picture before declaring infection.

Re-evaluate after drainage

After PCD/endoscopic drainage: reassess at 24–72 h and serially. ~⅓ of PANTER patients resolved with drainage alone. If sepsis persists or the collection fails to shrink → escalate to minimally invasive necrosectomy (endoscopic or VARD).

Never the goal: sterilise and resect everything

The objective is SOURCE CONTROL and symptom relief — NOT complete radiological clearance. Leaving some necrotic debris that is drained and sterilised is acceptable and safer than aggressive complete debridement, which causes bleeding and fistulae.

[1] [6] [10]

Drainage and debridement modalities

Percutaneous catheter drainage (PCD) — the first step

CT- or ultrasound-guided placement of a large-bore (12–30 Fr, often upsized over days) drainage catheter into the necrotic collection. It is the least invasive rung of the ladder and the most widely available (any interventional radiology service). [1]

Indication: suspected/confirmed infected necrosis where endoscopic expertise is unavailable, OR the collection is in a location unfavourable for a transluminal route (e.g. paracolic, inferior, retroperitoneal).
Route: ideally retroperitoneal (flank) rather than transperitoneal — a retroperitoneal tract can be dilated later for VARD and avoids contaminating the peritoneal cavity; a transperitoneal tract risks fistula and precludes retroperitoneal necrosectomy.
Strengths: bedside/IR-suite availability; bridges the patient to a walled-off state; ~35% definitive (PANTER).
Limitations: solid necrotic debris blocks the catheter; secondary infection of a sterile collection (≈10%); pancreaticocutaneous fistula along the tract; need for frequent flushes and upsizing. [1]

Endoscopic transluminal drainage ± necrosectomy

EUS-guided creation of a cystgastrostomy (through the posterior gastric wall) or cystduodenostomy into an apposed collection, followed by placement of plastic double-pigtail stents or a lumen-apposing metal stent (LAMS). The tract is balloon-dilated to allow a forward-viewing therapeutic gastroscope to enter the cavity for direct endoscopic necrosectomy (DEN) using snares, nets, and forceps over multiple sessions (often 3–6, days to a week apart).[3][8]

Indication: infected necrosis where the collection is apposed to the gastric or duodenal wall (the endoscopic "window").
Strengths: NO external fistula (TENSION — fewer fistulae than VARD); no abdominal wall incision; avoids general anaesthesia/laparotomy; can be repeated; preferred route when feasible.
LAMS vs plastic stents: LAMS create a wider, more stable fistula (larger flanges appose the gut and cyst walls), facilitate necrosectomy, and reduce repeat procedures — at higher upfront cost. Cost-effectiveness analyses favour LAMS when multiple necrosectomy sessions are anticipated.[8][9] LAMS must be removed/replaced at 3–4 weeks to avoid buried-stent syndrome, bleeding, and perforation.

Video-assisted retroperitoneal debridement (VARD)

A hybrid surgical-endoscopic technique. After a percutaneous drain has been placed (ideally retroperitoneally), the tract is dilated and a laparoscope is inserted through a small flank incision into the retroperitoneal necrotic cavity under direct vision; necrotic tissue is removed with forceps and continuous irrigation. This is the surgical arm of the TENSION step-up.[3]

Indication: infected necrosis not amenable to endoscopic drainage (no gastric/duodenal apposition, paracolic/inferior extension) where percutaneous drainage alone has failed.
Strengths: avoids laparotomy (less fistula/hernia than open); direct visualisation allows effective debridement; single incision.
Limitations: pancreaticocutaneous fistula along the tract (though fewer than open); requires general anaesthesia and a surgical/endoscopic team; narrower indication than endoscopic route. [1]

Endoscopic ultrasound-guided cystgastrostomy with nasocystic drain / LAMS

For walled-off necrosis apposed to the gastric/duodenal wall, EUS confirms a mature wall and an apposition <1 cm with intervening vessels excluded by Doppler. A fistula is created (needle-knife/cystotome), dilated, and stented: [1]

Plastic double-pigtail stents: cheap, multiple placed, but narrow lumen — necrotic debris occludes them, requiring repeat dilation/necrosectomy.
LAMS (e.g. AXIOS, HotAXIOS): electrocautery-enhanced, single-step deployment, 10–15 mm lumen — allows the gastroscope to pass directly into the cavity for necrosectomy and provides ongoing drainage. Resect/replace at 3–4 weeks.
Nasocystic drain: a tube placed through the cystgastrostomy into the cavity and out through the nose, allowing continuous irrigation/flushing to keep solid debris from blocking the tract — used in combination with stents, especially for large WON with substantial solid component. [1]

Walled-off necrosis vs pseudocyst — do not confuse them

The revised Atlanta classification (2012) divides peripancreatic collections by timing (<4 vs >4 weeks) and content (fluid only vs necrosis).[6][7] Mislabelling a WON as a pseudocyst (or vice versa) leads to wrong management: draining a WON like a pseudocyst (stent only) fails because the solid debris blocks the stent and causes recurrent sepsis.

[6] [7]

Drainage modality comparison

[1] [3] [5]

Step-up vs primary open — the PANTER evidence

[1] [2]

Endoscopic vs surgical step-up — the TENSION evidence

[3] [4]

Clinical translation: when the collection is apposed to the gastric/duodenal wall and endoscopic expertise exists, choose endoscopic step-up first. Reserve VARD for collections without an endoscopic window (paracolic, inferior, or not apposed). Both are vastly superior to open necrosectomy. [1]

Indications for intervention

[1] [6] [10]

Exam-style short-answer questions

SAQ — ERCP in severe gallstone pancreatitis with cholangitis

10 minutes · 10 marks

A 58-year-old woman is admitted to ICU with severe gallstone pancreatitis (APACHE II 18). She is febrile (39.2°C), hypotensive (MAP 58 mmHg on noradrenaline 0.2 mcg/kg/min), confused, with scleral icterus and right-upper-quadrant pain. Bloods: bilirubin 78 µmol/L, ALP 410 U/L, GGT 320 U/L, lipase 1,200 U/L, lactate 3.2 mmol/L, INR 1.6. MRCP confirms a 7 mm CBD stone with upstream duct dilation.

[1]

SAQ — Endoscopic necrosectomy for infected walled-off necrosis

10 minutes · 10 marks

A 45-year-old man with alcohol-induced necrotising pancreatitis is day 32 from onset. He has persistent fever, rising lactate (now 3.8 mmol/L), and new noradrenaline requirement despite meropenem for 9 days. CT shows a 12 × 8 cm walled-off pancreatic necrosis with multiple gas locules, apposed to the posterior gastric wall (apposition <1 cm). A retroperitoneal percutaneous drain placed 6 days ago has minimal output and he remains septic.

[1]

Clinical pearls

High-yight pancreatitis intervention points for the CICM/FFICM exam

Delay intervention ~4 weeks — allow demarcation. Premature surgery = high mortality.[2] }
Step-up approach: drain first → necrosectomy if needed → surgery last.[1] }
PANTER trial: step-up (drain ± minimally invasive necrosectomy) was superior to primary open necrosectomy — composite major complications/death 35% vs 69%.[1] }
TENSION trial: endoscopic step-up superior to surgical (VARD) step-up — fewer fistulae, shorter hospital, fewer interventions.[3] }
PENGUIN trial: endoscopic transgastric necrosectomy beat surgical necrosectomy (fewer complications, lower IL-6).[5] }
Carbapenems: drug of choice for infected necrosis (good pancreatic tissue penetration).[10] }
Pseudoaneurysm: splenic/gastroduodenal artery → life-threatening haemorrhage. Urgent angiographic embolisation.[10] }
Walled-off necrosis (>4 weeks): mature wall around necrotic collection (solid debris). Drain if infected/symptomatic; needs necrosectomy (DEN/VARD), stent-alone often fails.[6] }
Pseudocyst: fluid collection with mature wall (NO necrosis). Observe unless >6 cm and persistent, symptomatic, infected. Drain endoscopically.[6] }
ERCP: for gallstone pancreatitis with cholangitis or persistent biliary obstruction. Not routine.[10] }
Cholecystectomy: same admission (mild pancreatitis) or after recovery (severe). Prevents recurrence.[10] }
** splenic vein thrombosis**: left-sided portal hypertension → gastric varices. May need splenectomy.[10] }
Pancreatic fistula: complication of necrosectomy. Octreotide may reduce output. Most close spontaneously.[3] }
Diabetes: endocrine insufficiency from pancreatic necrosis. PANTER long-term: step-up had LESS new-onset diabetes than open. Monitor glucose. May be permanent.[2] }
Exocrine insufficiency: steatorrhoea from pancreatic enzyme deficiency. PANTER long-term: step-up had LESS exocrine insufficiency. Pancreatic enzyme replacement.[2] }

14 additional interventional pancreatitis pearls (examiner-grade)

The single most testable trial in ICU interventional pancreatology is PANTER (2010) — know the composite endpoint (35% vs 69%), the 35%-resolved-with-drainage-alone figure, and the four-step ladder. If you can quote PANTER, you can answer most viva questions.[1]
"Walled-off necrosis" is an Atlanta-2012 term — replaces the old "pancreatic abscess" and "organized pancreatic necrosis". It is defined by (>4 weeks) + (necrosis/solid debris) + (mature wall). If asked to classify a collection, always give timing AND content.[6][7]
Gas in the collection is suggestive, not diagnostic, of infection — confirm with FNA culture OR clinical deterioration. A sterile collection with gas (from enteric communication) does not mandate drainage unless the patient is unwell.[10]
Prophylactic antibiotics for sterile necrosis do NOT prevent infection and cause harm (resistance, fungal colonisation, C. diff). All major guidelines (ACG, IAP/APA, AGA) recommend AGAINST. Sterile necrosis is managed medically.[10]
The first drainage catheter should ideally be placed retroperitoneally (flank) — because it can be dilated later into a VARD tract. A transperitoneal drain contaminates the peritoneum and precludes retroperitoneal necrosectomy.[1]
Endoscopic step-up wins on FISTULAE — the dominant advantage in TENSION was the drop in pancreaticocutaneous/enterocutaneous fistulae from 28% (surgical) to 5% (endoscopic), because there is no external tract through the abdominal wall.[3]
LAMS (lumen-apposing metal stents) outperform plastic stents for WON — wider lumen, single-step (HotAXIOS) deployment, allows direct endoscopic necrosectomy through the stent, fewer repeat procedures. Cost-effective when multiple sessions anticipated. Remove/replace at 3–4 weeks to avoid buried-stent bleeding/perforation.[8][9]
The goal of intervention is SOURCE CONTROL, not radiological clearance — leaving sterile, drained necrotic debris is acceptable and safer than aggressive complete debridement, which causes bleeding and fistulae. Do not "chase" residual collections on CT.[10]
Nasocystic irrigation drain + LAMS is the modern combination for large WON — the nasocystic tube allows continuous saline flushing to keep solid debris from blocking the tract while the LAMS maintains a wide fistula.[8]
Open necrosectomy is now reserved for catastrophes — failure of minimally invasive step-up, massive uncontrolled bleeding not amenable to embolisation, bowel perforation, or abdominal compartment syndrome. Even then, drain first if possible.[1]
Incisional hernia is a long-term-specific outcome of open necrosectomy — the PANTER 11-year follow-up showed fewer incisional hernias with step-up, a direct consequence of avoiding laparotomy. Cite this when asked about long-term surgical morbidity.[2]
VARD = video-assisted retroperitoneal debridement — hybrid technique, surgical/endoscopic, via a dilated retroperitoneal drain tract, single flank incision, direct-vision debridement. The surgical arm of TENSION. Used when no endoscopic window exists.[3]
Direct endoscopic necrosectomy (DEN) requires multiple sessions — typically 3–6 sessions, days to a week apart, using a therapeutic gastroscope through a balloon-dilated cystgastrostomy (or LAMS), removing necrotic tissue with snares, nets, and forceps under direct vision.[3]
The collection must be "apposed" (within ~1 cm) to the gastric or duodenal wall for endoscopic drainage — EUS confirms apposition and excludes intervening vessels by Doppler. Non-apposed (paracolic, inferior) collections require VARD or percutaneous drainage.[4]
Pseudoaneurysm haemorrhage may present days–weeks after pancreatitis onset — splenic artery most common, then gastroduodenal. CT angiography diagnoses; IR embolisation is first-line. Surgery is a last resort with very high mortality.[10]
Splenic vein thrombosis → left-sided (sinistral) portal hypertension → isolated gastric fundal varices — splenectomy is curative if variceal bleeding occurs. Different from cirrhotic portal hypertension (which affects the whole portal bed).[10]
Disconnected duct syndrome — necrosis separates a viable pancreatic tail from the duct; causes recurrent pseudocyst/WON or pancreatic fistula. May require endoscopic stenting, surgical resection, or enteric drainage of the disconnected segment.[6]
Repeated imaging is normal during recovery — do not "treat the CT". Treat the patient. Asymptomatic residual collections in a recovering patient need no intervention.[10]
Cholecystectomy timing — same admission for mild gallstone pancreatitis (index admission or before discharge); for severe/necrotising pancreatitis, delay until inflammation resolves (often weeks). Prevents recurrence.[10]
The mortality of open necrosectomy (15–30%) is largely why it was abandoned as primary therapy — combined with fistulae, hernia, diabetes, and prolonged ICU stay, this drove adoption of the step-up approach.[1]

Red flags

Additional interventional red flags and exam traps

Never drain a sterile, asymptomatic collection — observation is correct. Prophylactic antibiotics are harmful.[10]
Gas in the collection is NOT diagnostic of infection in isolation — confirm with FNA or clinical deterioration before declaring infected necrosis.[10]
Draining a walled-off necrosis like a pseudocyst (stent only) FAILS — solid debris blocks the stent → recurrent sepsis. WON needs necrosectomy (DEN/VARD).[6]
LAMS left >4 weeks → buried stent, catastrophic bleeding, perforation — schedule removal/replacement.[8]
Place the first drain retroperitoneally if possible — preserves the option to dilate for VARD; transperitoneal drains risk fistula and preclude retroperitoneal access.[1]
Endoscopic drainage needs apposition (<1 cm) to gastric/duodenal wall — non-apposed collections cannot be drained endoscopically; use VARD or PCD.[4]
No endoscopic window + failed PCD = VARD, not open necrosectomy. Open is for failure of the whole step-up or catastrophe (bleeding, perforation, ACS).[3]
Pseudoaneurysm can rupture days–weeks into the illness — sentinel bleed or sudden Hb drop in recovering pancreatitis = urgent CT angiography and embolisation.[10]
Goal is source control, not radiological clearance — do not chase residual collections on CT in a recovering patient.[10]
Recurrent/persistent collection after drainage → consider disconnected duct syndrome — viable pancreatic tail separated from main duct; needs stenting, resection, or enteric drainage.[6]

Common viva pitfalls and model answers

Anticipated examiner questions — model answers

Q: "A patient has necrotising pancreatitis. When and how would you intervene?" — A: "If sterile and stable, I would NOT intervene — medical management with no prophylactic antibiotics. If infected (gas on CT plus deterioration, or positive FNA) I would start a carbapenem and follow the step-up approach: delay definitive necrosectomy ~4 weeks to allow walling-off, place a percutaneous drain as bridge/definitive source control, and add minimally invasive necrosectomy (endoscopic if the collection apposes the stomach, otherwise VARD) only if drainage fails. Open necrosectomy is reserved for failure of step-up or catastrophe. This is supported by PANTER and TENSION."[1][3]
Q: "What did the PANTER trial show?" — A: "PANTER (van Santvoort, NEJM 2010) randomised infected necrotising pancreatitis to a step-up approach versus primary open necrosectomy. The step-up arm halved the composite of major complications or death (35% vs 69%), and a third resolved with drainage alone — never needing necrosectomy. There was no significant mortality difference. The 11-year follow-up confirmed less new-onset diabetes, less exocrine insufficiency, and fewer incisional hernias with step-up."[1][2]
Q: "Endoscopic or surgical drainage — which first?" — A: "Endoscopic transluminal step-up first when the collection apposes the gastric or duodenal wall and expertise is available. TENSION (Lancet 2018) showed no difference in the composite primary endpoint but endoscopic had far fewer fistulae (5% vs 28%), shorter hospital stay, and fewer interventions. VARD is reserved for collections without an endoscopic window."[3]
Q: "How do you distinguish a pseudocyst from walled-off necrosis?" — A: "Both occur >4 weeks after onset and both have a mature wall — the difference is content. A pseudocyst contains fluid only; walled-off necrosis contains solid necrotic debris. This matters because WON requires necrosectomy (a stent alone blocks with debris), whereas a pseudocyst drains well through a stent alone. I would confirm with contrast CT/MRI looking for non-enhancing solid component."[6][7]
Q: "What is a LAMS and when do you use it?" — A: "A lumen-apposing metal stent — a dumbbell-shaped metal stent deployed under EUS guidance to create a wide, stable fistula between the gut lumen (stomach/duodenum) and a walled-off necrosis. Its wide lumen allows direct endoscopic necrosectomy through the stent and reduces repeat procedures versus plastic stents. It must be removed or replaced at 3–4 weeks to prevent buried-stent bleeding or perforation."[8][9]
Q: "The patient deteriorates at day 10 with septic shock and gas in the peripancreatic collection." — A: "This is infected necrosis with clinical deterioration — I cannot wait for wall-off. I would start a carbapenem and place a percutaneous drain immediately as source-control bridge. Definitive necrosectomy is deferred until the collection walls off (~4 weeks). The drain itself may be definitive in a third of patients."[1][10]

Quick-revision summary

Strategy: STEP-UP (drain → minimally invasive necrosectomy → open last). Timing: DELAY ~4 weeks unless deteriorating. Route: endoscopic > surgical (VARD) > open.
PANTER (2010, NEJM): step-up beat primary open — composite 35% vs 69%; 35% resolved with drainage alone.[1]
PANTER long-term (2019): less diabetes, less exocrine insufficiency, fewer incisional hernias; no mortality difference.[2]
TENSION (2018, Lancet): endoscopic step-up beat surgical step-up on fistulae (5% vs 28%), hospital stay, number of interventions.[3]
PENGUIN (2012, JAMA): endoscopic transgastric necrosectomy beat surgical (fewer complications, lower IL-6).[5]
WON vs pseudocyst: both >4 wk + mature wall; WON has solid debris (needs necrosectomy), pseudocyst is fluid only (stent drains well).[6]
LAMS > plastic stents for WON; remove at 3–4 weeks.[8]
Pseudoaneurysm: IR embolisation, not surgery.[10]

References

[1]van Santvoort HC, Besselink MG, Bakker OJ, et al. A step-up approach or open necrosectomy for necrotizing pancreatitis N Engl J Med, 2010.PMID 20410514
[2]Hollemans RA, Bakker OJ, Boermeester MA, et al. Superiority of Step-up Approach vs Open Necrosectomy in Long-term Follow-up of Patients With Necrotizing Pancreatitis Gastroenterology, 2019.PMID 30391468
[3]van Brunschot S, van Grinsven J, van Santvoort HC, et al. Endoscopic or surgical step-up approach for infected necrotising pancreatitis: a multicentre randomised trial Lancet, 2018.PMID 29108721
[4]van Brunschot S, van Grinsven J, Voermans RP, et al. Transluminal endoscopic step-up approach versus minimally invasive surgical step-up approach in patients with infected necrotising pancreatitis (TENSION trial): design and rationale of a randomised controlled multicenter trial [ISRCTN09186711] BMC Gastroenterol, 2013.PMID 24274589
[5]Bakker OJ, van Santvoort HC, van Brunschot S, et al. Endoscopic transgastric vs surgical necrosectomy for infected necrotizing pancreatitis: a randomized trial JAMA, 2012.PMID 22416101
[6]Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis--2012: revision of the Atlanta classification and definitions by international consensus Gut, 2013.PMID 23100216
[7]Sarr MG, Banks PA, Bollen TL, et al. The new revised classification of acute pancreatitis 2012 Surg Clin North Am, 2013.PMID 23632143
[8]Zhu HY, Xie P, Song YX, et al. Lumen-apposing metal stents (LAMS) versus plastic stents for EUS-guided drainage of walled-off necrosis (WON) (LVPWON): study protocol for a multicenter randomized controlled trial Trials, 2018.PMID 30305160
[9]Chen YI, Barkun AN, Adam V, et al. Cost-effectiveness analysis comparing lumen-apposing metal stents with plastic stents in the management of pancreatic walled-off necrosis Gastrointest Endosc, 2018.PMID 29614262
[10]Tenner S, Vege SS, Sheth SG, et al. American College of Gastroenterology Guidelines: Management of Acute Pancreatitis Am J Gastroenterol, 2024.PMID 38857482

[1]

The step-up approach: concept and evidence

The ladder has four rungs: [1]

Antibiotics + supportive care (carbapenem for good pancreatic penetration) — sterile necrosis is managed medically; suspected infection triggers carbapenem and reassessment.
Catheter drainage — percutaneous (CT/US-guided) OR endoscopic ultrasound-guided transluminal (cystgastrostomy/cystduodenostomy). This alone resolves infection in roughly a third of patients.
Minimally invasive necrosectomy — endoscopic (through the transluminal tract) or percutaneous retroperitoneal (VARD / endoscopic). Multiple sessions, days to weeks apart.
Open necrosectomy — last resort, for failure of step-up, uncontrolled bleeding, bowel perforation, or abdominal compartment syndrome. [1]

The evidence base is one of the strongest in interventional pancreatology and is directly examinable: three randomised trials define the field. [1]

The three pivotal RCTs

PANTER — step-up vs primary open necrosectomy

Primary endpoint (major complications/death composite): step-up 35% vs open 69% — step-up superior (RR 0.51, 95% CI 0.33–0.77). [1]

Key findings:

35% of step-up patients resolved with drainage ALONE — never needed necrosectomy.
Fewer new-onset multi-organ failure (12% vs 40%).
No statistically significant difference in mortality (≈11–19% both arms), but step-up trended lower.
Shorter hospital and ICU stay, lower healthcare costs (Drain, Retrospective cost analysis). [1]

[1]

PANTER 11-year follow-up (Hollemans 2019)

Design: long-term follow-up of the PANTER cohort (median ~6 years).[2]

Findings — step-up remained superior over the long term:

Less new-onset diabetes (step-up lower incidence than open).
Less exocrine insufficiency (pancreatic enzyme dependence).
Fewer incisional hernias (a direct consequence of avoiding laparotomy).
No difference in mortality (confirmed over the long term).
No difference in quality of life at final follow-up. [1]

[1]

TENSION — endoscopic step-up vs surgical step-up

Primary endpoint (composite: major complications/death): no significant difference in the composite (≈43% endoscopic vs 45% surgical). [1]

Key secondary findings favouring endoscopic:

Fewer pancreaticocutaneous and enterocutaneous fistulae (5% vs 28%).
Shorter length of hospital stay.
Fewer interventions (necrosectomy sessions) per patient.
Trend to fewer new-onset diabetes. [1]

[1]

PENGUIN — endoscopic transgastric vs surgical necrosectomy

Design: RCT, 22 patients; endoscopic transgastric necrosectomy vs surgical (open/VARD) necrosectomy for infected necrotising pancreatitis.[5]

Findings: endoscopic group had a lower composite of major complications/death and lower IL-6 (less inflammatory response) after the index procedure. [1]

[1]

Step-up approach

Step-up approach for infected necrosis

Delay intervention ~4 weeks

Step 1: Percutaneous or endoscopic drainage

Step 2: Add necrosectomy if drainage insufficient

Step 3: Open surgery (LAST RESORT)

[1] [2] [3] [10]

Timing of intervention — when to act

Default: DELAY ~4 weeks (let it wall off)

Exception: intervene EARLY only if clinically deteriorating

How infection is confirmed

Re-evaluate after drainage

Never the goal: sterilise and resect everything

[1] [6] [10]

Drainage and debridement modalities

Percutaneous catheter drainage (PCD) — the first step

Indication: suspected/confirmed infected necrosis where endoscopic expertise is unavailable, OR the collection is in a location unfavourable for a transluminal route (e.g. paracolic, inferior, retroperitoneal).
Route: ideally retroperitoneal (flank) rather than transperitoneal — a retroperitoneal tract can be dilated later for VARD and avoids contaminating the peritoneal cavity; a transperitoneal tract risks fistula and precludes retroperitoneal necrosectomy.
Strengths: bedside/IR-suite availability; bridges the patient to a walled-off state; ~35% definitive (PANTER).
Limitations: solid necrotic debris blocks the catheter; secondary infection of a sterile collection (≈10%); pancreaticocutaneous fistula along the tract; need for frequent flushes and upsizing. [1]

Endoscopic transluminal drainage ± necrosectomy

Indication: infected necrosis where the collection is apposed to the gastric or duodenal wall (the endoscopic "window").
Strengths: NO external fistula (TENSION — fewer fistulae than VARD); no abdominal wall incision; avoids general anaesthesia/laparotomy; can be repeated; preferred route when feasible.
LAMS vs plastic stents: LAMS create a wider, more stable fistula (larger flanges appose the gut and cyst walls), facilitate necrosectomy, and reduce repeat procedures — at higher upfront cost. Cost-effectiveness analyses favour LAMS when multiple necrosectomy sessions are anticipated.[8][9] LAMS must be removed/replaced at 3–4 weeks to avoid buried-stent syndrome, bleeding, and perforation.

Video-assisted retroperitoneal debridement (VARD)

Indication: infected necrosis not amenable to endoscopic drainage (no gastric/duodenal apposition, paracolic/inferior extension) where percutaneous drainage alone has failed.
Strengths: avoids laparotomy (less fistula/hernia than open); direct visualisation allows effective debridement; single incision.
Limitations: pancreaticocutaneous fistula along the tract (though fewer than open); requires general anaesthesia and a surgical/endoscopic team; narrower indication than endoscopic route. [1]

Endoscopic ultrasound-guided cystgastrostomy with nasocystic drain / LAMS

Plastic double-pigtail stents: cheap, multiple placed, but narrow lumen — necrotic debris occludes them, requiring repeat dilation/necrosectomy.
LAMS (e.g. AXIOS, HotAXIOS): electrocautery-enhanced, single-step deployment, 10–15 mm lumen — allows the gastroscope to pass directly into the cavity for necrosectomy and provides ongoing drainage. Resect/replace at 3–4 weeks.
Nasocystic drain: a tube placed through the cystgastrostomy into the cavity and out through the nose, allowing continuous irrigation/flushing to keep solid debris from blocking the tract — used in combination with stents, especially for large WON with substantial solid component. [1]

Walled-off necrosis vs pseudocyst — do not confuse them

[6] [7]

Drainage modality comparison

[1] [3] [5]

Step-up vs primary open — the PANTER evidence

[1] [2]

Endoscopic vs surgical step-up — the TENSION evidence

[3] [4]

Indications for intervention

[1] [6] [10]

Exam-style short-answer questions

SAQ — ERCP in severe gallstone pancreatitis with cholangitis

10 minutes · 10 marks

[1]

SAQ — Endoscopic necrosectomy for infected walled-off necrosis

10 minutes · 10 marks

[1]

Clinical pearls

High-yight pancreatitis intervention points for the CICM/FFICM exam

Delay intervention ~4 weeks — allow demarcation. Premature surgery = high mortality.[2] }
Step-up approach: drain first → necrosectomy if needed → surgery last.[1] }
PANTER trial: step-up (drain ± minimally invasive necrosectomy) was superior to primary open necrosectomy — composite major complications/death 35% vs 69%.[1] }
TENSION trial: endoscopic step-up superior to surgical (VARD) step-up — fewer fistulae, shorter hospital, fewer interventions.[3] }
PENGUIN trial: endoscopic transgastric necrosectomy beat surgical necrosectomy (fewer complications, lower IL-6).[5] }
Carbapenems: drug of choice for infected necrosis (good pancreatic tissue penetration).[10] }
Pseudoaneurysm: splenic/gastroduodenal artery → life-threatening haemorrhage. Urgent angiographic embolisation.[10] }
Walled-off necrosis (>4 weeks): mature wall around necrotic collection (solid debris). Drain if infected/symptomatic; needs necrosectomy (DEN/VARD), stent-alone often fails.[6] }
Pseudocyst: fluid collection with mature wall (NO necrosis). Observe unless >6 cm and persistent, symptomatic, infected. Drain endoscopically.[6] }
ERCP: for gallstone pancreatitis with cholangitis or persistent biliary obstruction. Not routine.[10] }
Cholecystectomy: same admission (mild pancreatitis) or after recovery (severe). Prevents recurrence.[10] }
** splenic vein thrombosis**: left-sided portal hypertension → gastric varices. May need splenectomy.[10] }
Pancreatic fistula: complication of necrosectomy. Octreotide may reduce output. Most close spontaneously.[3] }
Diabetes: endocrine insufficiency from pancreatic necrosis. PANTER long-term: step-up had LESS new-onset diabetes than open. Monitor glucose. May be permanent.[2] }
Exocrine insufficiency: steatorrhoea from pancreatic enzyme deficiency. PANTER long-term: step-up had LESS exocrine insufficiency. Pancreatic enzyme replacement.[2] }

14 additional interventional pancreatitis pearls (examiner-grade)

The single most testable trial in ICU interventional pancreatology is PANTER (2010) — know the composite endpoint (35% vs 69%), the 35%-resolved-with-drainage-alone figure, and the four-step ladder. If you can quote PANTER, you can answer most viva questions.[1]
"Walled-off necrosis" is an Atlanta-2012 term — replaces the old "pancreatic abscess" and "organized pancreatic necrosis". It is defined by (>4 weeks) + (necrosis/solid debris) + (mature wall). If asked to classify a collection, always give timing AND content.[6][7]
Gas in the collection is suggestive, not diagnostic, of infection — confirm with FNA culture OR clinical deterioration. A sterile collection with gas (from enteric communication) does not mandate drainage unless the patient is unwell.[10]
Prophylactic antibiotics for sterile necrosis do NOT prevent infection and cause harm (resistance, fungal colonisation, C. diff). All major guidelines (ACG, IAP/APA, AGA) recommend AGAINST. Sterile necrosis is managed medically.[10]
The first drainage catheter should ideally be placed retroperitoneally (flank) — because it can be dilated later into a VARD tract. A transperitoneal drain contaminates the peritoneum and precludes retroperitoneal necrosectomy.[1]
Endoscopic step-up wins on FISTULAE — the dominant advantage in TENSION was the drop in pancreaticocutaneous/enterocutaneous fistulae from 28% (surgical) to 5% (endoscopic), because there is no external tract through the abdominal wall.[3]
LAMS (lumen-apposing metal stents) outperform plastic stents for WON — wider lumen, single-step (HotAXIOS) deployment, allows direct endoscopic necrosectomy through the stent, fewer repeat procedures. Cost-effective when multiple sessions anticipated. Remove/replace at 3–4 weeks to avoid buried-stent bleeding/perforation.[8][9]
The goal of intervention is SOURCE CONTROL, not radiological clearance — leaving sterile, drained necrotic debris is acceptable and safer than aggressive complete debridement, which causes bleeding and fistulae. Do not "chase" residual collections on CT.[10]
Nasocystic irrigation drain + LAMS is the modern combination for large WON — the nasocystic tube allows continuous saline flushing to keep solid debris from blocking the tract while the LAMS maintains a wide fistula.[8]
Open necrosectomy is now reserved for catastrophes — failure of minimally invasive step-up, massive uncontrolled bleeding not amenable to embolisation, bowel perforation, or abdominal compartment syndrome. Even then, drain first if possible.[1]
Incisional hernia is a long-term-specific outcome of open necrosectomy — the PANTER 11-year follow-up showed fewer incisional hernias with step-up, a direct consequence of avoiding laparotomy. Cite this when asked about long-term surgical morbidity.[2]
VARD = video-assisted retroperitoneal debridement — hybrid technique, surgical/endoscopic, via a dilated retroperitoneal drain tract, single flank incision, direct-vision debridement. The surgical arm of TENSION. Used when no endoscopic window exists.[3]
Direct endoscopic necrosectomy (DEN) requires multiple sessions — typically 3–6 sessions, days to a week apart, using a therapeutic gastroscope through a balloon-dilated cystgastrostomy (or LAMS), removing necrotic tissue with snares, nets, and forceps under direct vision.[3]
The collection must be "apposed" (within ~1 cm) to the gastric or duodenal wall for endoscopic drainage — EUS confirms apposition and excludes intervening vessels by Doppler. Non-apposed (paracolic, inferior) collections require VARD or percutaneous drainage.[4]
Pseudoaneurysm haemorrhage may present days–weeks after pancreatitis onset — splenic artery most common, then gastroduodenal. CT angiography diagnoses; IR embolisation is first-line. Surgery is a last resort with very high mortality.[10]
Splenic vein thrombosis → left-sided (sinistral) portal hypertension → isolated gastric fundal varices — splenectomy is curative if variceal bleeding occurs. Different from cirrhotic portal hypertension (which affects the whole portal bed).[10]
Disconnected duct syndrome — necrosis separates a viable pancreatic tail from the duct; causes recurrent pseudocyst/WON or pancreatic fistula. May require endoscopic stenting, surgical resection, or enteric drainage of the disconnected segment.[6]
Repeated imaging is normal during recovery — do not "treat the CT". Treat the patient. Asymptomatic residual collections in a recovering patient need no intervention.[10]
Cholecystectomy timing — same admission for mild gallstone pancreatitis (index admission or before discharge); for severe/necrotising pancreatitis, delay until inflammation resolves (often weeks). Prevents recurrence.[10]
The mortality of open necrosectomy (15–30%) is largely why it was abandoned as primary therapy — combined with fistulae, hernia, diabetes, and prolonged ICU stay, this drove adoption of the step-up approach.[1]

Red flags

Additional interventional red flags and exam traps

Never drain a sterile, asymptomatic collection — observation is correct. Prophylactic antibiotics are harmful.[10]
Gas in the collection is NOT diagnostic of infection in isolation — confirm with FNA or clinical deterioration before declaring infected necrosis.[10]
Draining a walled-off necrosis like a pseudocyst (stent only) FAILS — solid debris blocks the stent → recurrent sepsis. WON needs necrosectomy (DEN/VARD).[6]
LAMS left >4 weeks → buried stent, catastrophic bleeding, perforation — schedule removal/replacement.[8]
Place the first drain retroperitoneally if possible — preserves the option to dilate for VARD; transperitoneal drains risk fistula and preclude retroperitoneal access.[1]
Endoscopic drainage needs apposition (<1 cm) to gastric/duodenal wall — non-apposed collections cannot be drained endoscopically; use VARD or PCD.[4]
No endoscopic window + failed PCD = VARD, not open necrosectomy. Open is for failure of the whole step-up or catastrophe (bleeding, perforation, ACS).[3]
Pseudoaneurysm can rupture days–weeks into the illness — sentinel bleed or sudden Hb drop in recovering pancreatitis = urgent CT angiography and embolisation.[10]
Goal is source control, not radiological clearance — do not chase residual collections on CT in a recovering patient.[10]
Recurrent/persistent collection after drainage → consider disconnected duct syndrome — viable pancreatic tail separated from main duct; needs stenting, resection, or enteric drainage.[6]

Common viva pitfalls and model answers

Anticipated examiner questions — model answers

Q: "A patient has necrotising pancreatitis. When and how would you intervene?" — A: "If sterile and stable, I would NOT intervene — medical management with no prophylactic antibiotics. If infected (gas on CT plus deterioration, or positive FNA) I would start a carbapenem and follow the step-up approach: delay definitive necrosectomy ~4 weeks to allow walling-off, place a percutaneous drain as bridge/definitive source control, and add minimally invasive necrosectomy (endoscopic if the collection apposes the stomach, otherwise VARD) only if drainage fails. Open necrosectomy is reserved for failure of step-up or catastrophe. This is supported by PANTER and TENSION."[1][3]
Q: "What did the PANTER trial show?" — A: "PANTER (van Santvoort, NEJM 2010) randomised infected necrotising pancreatitis to a step-up approach versus primary open necrosectomy. The step-up arm halved the composite of major complications or death (35% vs 69%), and a third resolved with drainage alone — never needing necrosectomy. There was no significant mortality difference. The 11-year follow-up confirmed less new-onset diabetes, less exocrine insufficiency, and fewer incisional hernias with step-up."[1][2]
Q: "Endoscopic or surgical drainage — which first?" — A: "Endoscopic transluminal step-up first when the collection apposes the gastric or duodenal wall and expertise is available. TENSION (Lancet 2018) showed no difference in the composite primary endpoint but endoscopic had far fewer fistulae (5% vs 28%), shorter hospital stay, and fewer interventions. VARD is reserved for collections without an endoscopic window."[3]
Q: "How do you distinguish a pseudocyst from walled-off necrosis?" — A: "Both occur >4 weeks after onset and both have a mature wall — the difference is content. A pseudocyst contains fluid only; walled-off necrosis contains solid necrotic debris. This matters because WON requires necrosectomy (a stent alone blocks with debris), whereas a pseudocyst drains well through a stent alone. I would confirm with contrast CT/MRI looking for non-enhancing solid component."[6][7]
Q: "What is a LAMS and when do you use it?" — A: "A lumen-apposing metal stent — a dumbbell-shaped metal stent deployed under EUS guidance to create a wide, stable fistula between the gut lumen (stomach/duodenum) and a walled-off necrosis. Its wide lumen allows direct endoscopic necrosectomy through the stent and reduces repeat procedures versus plastic stents. It must be removed or replaced at 3–4 weeks to prevent buried-stent bleeding or perforation."[8][9]
Q: "The patient deteriorates at day 10 with septic shock and gas in the peripancreatic collection." — A: "This is infected necrosis with clinical deterioration — I cannot wait for wall-off. I would start a carbapenem and place a percutaneous drain immediately as source-control bridge. Definitive necrosectomy is deferred until the collection walls off (~4 weeks). The drain itself may be definitive in a third of patients."[1][10]

Quick-revision summary

Strategy: STEP-UP (drain → minimally invasive necrosectomy → open last). Timing: DELAY ~4 weeks unless deteriorating. Route: endoscopic > surgical (VARD) > open.
PANTER (2010, NEJM): step-up beat primary open — composite 35% vs 69%; 35% resolved with drainage alone.[1]
PANTER long-term (2019): less diabetes, less exocrine insufficiency, fewer incisional hernias; no mortality difference.[2]
TENSION (2018, Lancet): endoscopic step-up beat surgical step-up on fistulae (5% vs 28%), hospital stay, number of interventions.[3]
PENGUIN (2012, JAMA): endoscopic transgastric necrosectomy beat surgical (fewer complications, lower IL-6).[5]
WON vs pseudocyst: both >4 wk + mature wall; WON has solid debris (needs necrosectomy), pseudocyst is fluid only (stent drains well).[6]
LAMS > plastic stents for WON; remove at 3–4 weeks.[8]
Pseudoaneurysm: IR embolisation, not surgery.[10]

References

[1]van Santvoort HC, Besselink MG, Bakker OJ, et al. A step-up approach or open necrosectomy for necrotizing pancreatitis N Engl J Med, 2010.PMID 20410514
[2]Hollemans RA, Bakker OJ, Boermeester MA, et al. Superiority of Step-up Approach vs Open Necrosectomy in Long-term Follow-up of Patients With Necrotizing Pancreatitis Gastroenterology, 2019.PMID 30391468
[3]van Brunschot S, van Grinsven J, van Santvoort HC, et al. Endoscopic or surgical step-up approach for infected necrotising pancreatitis: a multicentre randomised trial Lancet, 2018.PMID 29108721
[4]van Brunschot S, van Grinsven J, Voermans RP, et al. Transluminal endoscopic step-up approach versus minimally invasive surgical step-up approach in patients with infected necrotising pancreatitis (TENSION trial): design and rationale of a randomised controlled multicenter trial [ISRCTN09186711] BMC Gastroenterol, 2013.PMID 24274589
[5]Bakker OJ, van Santvoort HC, van Brunschot S, et al. Endoscopic transgastric vs surgical necrosectomy for infected necrotizing pancreatitis: a randomized trial JAMA, 2012.PMID 22416101
[6]Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis--2012: revision of the Atlanta classification and definitions by international consensus Gut, 2013.PMID 23100216
[7]Sarr MG, Banks PA, Bollen TL, et al. The new revised classification of acute pancreatitis 2012 Surg Clin North Am, 2013.PMID 23632143
[8]Zhu HY, Xie P, Song YX, et al. Lumen-apposing metal stents (LAMS) versus plastic stents for EUS-guided drainage of walled-off necrosis (WON) (LVPWON): study protocol for a multicenter randomized controlled trial Trials, 2018.PMID 30305160
[9]Chen YI, Barkun AN, Adam V, et al. Cost-effectiveness analysis comparing lumen-apposing metal stents with plastic stents in the management of pancreatic walled-off necrosis Gastrointest Endosc, 2018.PMID 29614262
[10]Tenner S, Vege SS, Sheth SG, et al. American College of Gastroenterology Guidelines: Management of Acute Pancreatitis Am J Gastroenterol, 2024.PMID 38857482

Your progress

The step-up approach: concept and evidence

The three pivotal RCTs

PANTER — step-up vs primary open necrosectomy

PANTER 11-year follow-up (Hollemans 2019)

TENSION — endoscopic step-up vs surgical step-up

PENGUIN — endoscopic transgastric vs surgical necrosectomy

Step-up approach

Delay intervention ~4 weeks

Step 1: Percutaneous or endoscopic drainage

Step 2: Add necrosectomy if drainage insufficient

Step 3: Open surgery (LAST RESORT)

Default: DELAY ~4 weeks (let it wall off)

Exception: intervene EARLY only if clinically deteriorating

How infection is confirmed

Re-evaluate after drainage

Never the goal: sterilise and resect everything

Drainage and debridement modalities

Percutaneous catheter drainage (PCD) — the first step

Endoscopic transluminal drainage ± necrosectomy

Video-assisted retroperitoneal debridement (VARD)

Endoscopic ultrasound-guided cystgastrostomy with nasocystic drain / LAMS

Walled-off necrosis vs pseudocyst — do not confuse them

Drainage modality comparison

Step-up vs primary open — the PANTER evidence

Endoscopic vs surgical step-up — the TENSION evidence

Indications for intervention

Exam-style short-answer questions

SAQ — ERCP in severe gallstone pancreatitis with cholangitis

SAQ — Endoscopic necrosectomy for infected walled-off necrosis

Clinical pearls

Red flags

Common viva pitfalls and model answers

Quick-revision summary

References

Your progress

The step-up approach: concept and evidence

The three pivotal RCTs

PANTER — step-up vs primary open necrosectomy

PANTER 11-year follow-up (Hollemans 2019)

TENSION — endoscopic step-up vs surgical step-up

PENGUIN — endoscopic transgastric vs surgical necrosectomy

Step-up approach

Delay intervention ~4 weeks

Step 1: Percutaneous or endoscopic drainage

Step 2: Add necrosectomy if drainage insufficient

Step 3: Open surgery (LAST RESORT)

Default: DELAY ~4 weeks (let it wall off)

Exception: intervene EARLY only if clinically deteriorating

How infection is confirmed

Re-evaluate after drainage

Never the goal: sterilise and resect everything

Drainage and debridement modalities

Percutaneous catheter drainage (PCD) — the first step

Endoscopic transluminal drainage ± necrosectomy

Video-assisted retroperitoneal debridement (VARD)

Endoscopic ultrasound-guided cystgastrostomy with nasocystic drain / LAMS

Walled-off necrosis vs pseudocyst — do not confuse them

Drainage modality comparison

Step-up vs primary open — the PANTER evidence

Endoscopic vs surgical step-up — the TENSION evidence

Indications for intervention

Exam-style short-answer questions

SAQ — ERCP in severe gallstone pancreatitis with cholangitis

SAQ — Endoscopic necrosectomy for infected walled-off necrosis

Clinical pearls

Red flags

Common viva pitfalls and model answers

Quick-revision summary

References