Publication

• Title: A step-up approach or open necrosectomy for necrotizing pancreatitis
• Acronym: PANTER
• Year: 2010
• Journal published in: New England Journal of Medicine
• Citation: van Santvoort HC, Besselink MG, Bakker OJ, Hofker HS, Boermeester MA, Dejong CH, et al. A step-up approach or open necrosectomy for necrotizing pancreatitis. N Engl J Med. 2010;362(16):1491-1502.

Context & Rationale

• Background

  Infected pancreatic/peripancreatic necrosis was historically treated with open surgical necrosectomy, a strategy associated with substantial procedural morbidity (new organ failure, fistulation, bleeding), prolonged critical care utilisation, and significant late sequelae (abdominal wall complications and pancreatic endocrine/exocrine insufficiency).
  Minimally invasive drainage and retroperitoneal approaches were increasingly used in specialised centres, but comparative evidence against primary open necrosectomy was limited and prone to confounding by indication.
• Research Question/Hypothesis

  Whether a protocolised “step-up” strategy (catheter or endoscopic drainage first, followed by minimally invasive retroperitoneal necrosectomy only if needed) reduces major complications and/or death compared with primary open necrosectomy in patients with (suspected) infected necrotising pancreatitis requiring intervention.
• Why This Matters

  The trial directly tested a shift in therapeutic philosophy: “drain and delay, debride only if necessary”, with implications for procedure selection, timing, ICU burden, late morbidity, and health-economic impact in a high-risk cohort.

Design & Methods

• Research Question: In adults with (suspected) infected necrotising pancreatitis requiring invasive intervention, does a minimally invasive “step-up” approach reduce a composite of major complications or death compared with primary open necrosectomy?
• Study Type: Multicentre, randomised, parallel-group trial conducted across academic and teaching hospitals; unblinded intervention delivery with blinded endpoint adjudication; investigator-initiated.
• Population:
  • Setting: Tertiary centres with interventional radiology and pancreatic surgery expertise.
  • Key inclusion features: Acute pancreatitis with pancreatic/peripancreatic necrosis; confirmed infected necrosis (e.g., positive culture or gas on CT) or suspected infection with clinical deterioration/persistent sepsis; decision that invasive intervention was indicated; necrotic collection considered drainable.
  • Notable exclusions: Prior laparotomy for the current pancreatitis episode; pancreatitis from abdominal surgery; acute intra-abdominal complications requiring urgent laparotomy (e.g., visceral perforation, active bleeding, abdominal compartment syndrome); other contraindications to protocolised drainage/necrosectomy pathways.
  • Baseline severity (illustrative): Mean APACHE II at randomisation 14.6 ± 6.1 (step-up) vs 15.0 ± 5.3 (open); multiple-organ failure present at randomisation 35% vs 29%; symptom onset to randomisation median 30 days (range 11–71) vs 29 days (range 12–155).
• Intervention:
  • Step-up strategy: Initial percutaneous catheter drainage (preferably retroperitoneal) or endoscopic drainage, with reassessment after 72 hours.
  • Escalation rule: If inadequate clinical improvement, additional drainage was prioritised; if persistent sepsis/clinical deterioration despite drainage optimisation, minimally invasive necrosectomy was performed using VARD (videoscopic assisted retroperitoneal debridement) when feasible.
  • Procedural granularity (reported): 41 patients underwent percutaneous drainage and 2 endoscopic drainage; median drain size 14 French (range 12–24); drains upsized in 4 and replaced in 7; multiple drains placed in 7.
  • Adjunct after necrosectomy: Post-necrosectomy lavage used after VARD, targeting at least 10 L per 24 hours by post-operative day 3.
• Comparison:
  • Primary open necrosectomy: Laparotomy with necrosectomy using a standardised technique, with drains placed and the abdomen closed.
  • Adjunct: Continuous post-operative lavage after open necrosectomy (targeting at least 10 L per 24 hours by post-operative day 3).
  • Re-intervention: Additional laparotomies permitted for persistent or recurrent sepsis/complications.
• Blinding: Treating teams and patients were not blinded (procedural strategy); an adjudication committee assessed primary and secondary endpoints blinded to treatment allocation; a radiologist evaluated CT images blinded to allocation.
• Statistics: Power calculation assumed reduction in the primary composite endpoint from 45% to 16% (absolute reduction 29%); 80% power; two-sided alpha 0.05; required sample size 88. Intention-to-treat analysis with risk ratios and 95% confidence intervals; sequential safety monitoring was planned but ultimately the trial proceeded to full accrual with conventional analysis after completion of follow-up because of delay in endpoint confirmation by adjudication. ¹
• Follow-Up Period: Primary endpoint assessed during admission and up to 3 months after discharge; follow-up visits at 3 and 6 months after discharge; resource utilisation and costs captured from admission to 6 months after discharge.

Key Results

This trial was not stopped early. The prespecified sample size of 88 patients was reached (43 step-up; 45 open), and primary endpoint assessment included in-hospital events plus 3 months after discharge.

• The step-up strategy reduced the primary composite (40% vs 69%), driven chiefly by fewer new-onset multiple-organ failure/systemic complications (12% vs 42%).
• Mortality was similar (19% vs 16%), consistent with limited power for survival differences.
• Despite more drainage procedures (total 82 vs 32), the step-up group underwent fewer operations overall (total 53 vs 91) and had lower late morbidity (incisional hernia 7% vs 24%) and lower costs.

Internal Validity

• Randomisation and allocation concealment: Centralised allocation with stratification (centre and feasibility of retroperitoneal access route); balanced baseline severity (APACHE II 14.6 ± 6.1 vs 15.0 ± 5.3; multiple-organ failure at randomisation 35% vs 29%).
• Dropout/exclusions: Primary analysis was intention-to-treat with all randomised patients contributing to the primary endpoint assessment (in-hospital plus 3 months post-discharge).
• Performance/detection bias: Procedural strategies could not be blinded; risk mitigated by blinded adjudication of endpoints and blinded CT review.
• Protocol adherence and treatment fidelity: Step-up arm predominantly used protocol-recommended drainage (40 retroperitoneal percutaneous; 2 endoscopic; 1 transabdominal), with escalation to necrosectomy when required; open arm predominantly underwent laparotomy (44/45), with repeat laparotomy in 19.
• Separation of the variable of interest: Necrosectomy was avoided in 17/43 (40%) in the step-up group vs 0/45 in open; total operations 53 vs 91 (P=0.004); total drainage procedures 82 vs 32 (P<0.001); new ICU admission after first intervention 16% vs 40% (RR 0.41; 95% CI 0.19 to 0.88; P=0.01).
• Timing: Randomisation occurred late in the disease course (median 30 vs 29 days after symptom onset), consistent with a clinical preference to defer invasive intervention where feasible; this strengthens internal coherence with the “delay and drain” paradigm but narrows inference for earlier deterioration trajectories.
• Crossover: Limited; step-up included 2 primary laparotomies due to lack of retroperitoneal access and 1 conversion from VARD to laparotomy; open group included 1 patient receiving primary VARD because of prior oesophagectomy.
• Outcome assessment: Composite endpoint components were protocol-defined and clinically meaningful; adjudication was blinded and consensus-based, improving reliability for complex postoperative events.
• Statistical rigor: Prespecified sample size reached; effect sizes presented as risk ratios with 95% CIs; sequential monitoring was prespecified but did not trigger early stopping; conventional analysis performed after audited follow-up completion.

Conclusion on Internal Validity: Overall, internal validity appears strong to moderate: allocation procedures and blinded adjudication support credibility, but the unblinded nature of procedural care and the complexity of co-interventions in both arms leave residual risk of performance bias (largely buffered by objective outcomes and adjudication).

External Validity

• Population representativeness: Participants were severely ill with infected/suspected infected necrosis in a tertiary-care context (high rates of organ failure and ICU exposure), broadly reflective of the subgroup that prompts invasive intervention in contemporary practice.
• Key exclusions and practice context: Patients requiring immediate laparotomy for acute intra-abdominal catastrophes were excluded; most randomisations occurred around 4 weeks after symptom onset, so applicability to earlier, rapidly progressive infection is constrained.
• System factors: Implementation requires interventional radiology for drain placement and surgical capability for VARD/laparotomy; outcomes may differ in lower-volume settings without established minimally invasive pancreatic necrosis pathways.
• Technique specificity: The comparator used a standardised open necrosectomy with postoperative lavage; centres using different open techniques or endoscopic-first pathways may see different effect sizes, although the core principle (avoiding early extensive debridement) is likely transferable.

Conclusion on External Validity: Generalisability is moderate: findings translate best to systems with multidisciplinary pancreatitis services and capacity for staged escalation, while extrapolation to early intervention scenarios and low-resource environments is limited.

Strengths & Limitations

• Strengths: Pragmatic multicentre design; clinically salient composite endpoint; prespecified escalation algorithm; blinded adjudication and blinded CT review; robust reporting of procedural intensity and downstream morbidity; incorporation of societal cost analysis over 6 months.
• Limitations: Small sample size with limited power for mortality; intervention and comparator were not blinded and involved complex care pathways; primary endpoint is composite (components differ in pathophysiological weight); specialised expertise and late randomisation may limit application to earlier or less selected disease trajectories.

Interpretation & Why It Matters

• Mechanistic inference

  The benefit signal concentrated in reduced new-onset organ failure/systemic complications (12% vs 42%), consistent with the hypothesis that avoiding early extensive debridement and prioritising source control via drainage attenuates the inflammatory hit and secondary deterioration.
• Practical clinical impact

  A structured escalation pathway can avoid necrosectomy entirely in a substantial fraction of patients (40% in the step-up arm), while reducing late morbidity (incisional hernia 7% vs 24%; new diabetes 16% vs 38%) and lowering mean total costs (€90,086 vs €122,268).
• What it does not prove

  Mortality benefit was not demonstrated (19% vs 16%), and the trial does not isolate which component(s) of the step-up bundle (timing, drainage, retroperitoneal access, VARD technique, or re-intervention thresholds) most strongly mediated outcome differences.

Controversies & Subsequent Evidence

• Composite endpoint interpretation: The effect on the primary composite was driven mainly by reduced “new-onset multiple-organ failure or systemic complications”, raising debate about the heterogeneity of clinical importance across composite components and the potential for care-pathway decisions to influence systemic complication ascertainment.
• Technique and setting dependence: An accompanying editorial emphasised that procedural expertise, access route feasibility, and centre experience are likely to modulate outcomes, and that broad adoption should be coupled to multidisciplinary pathways rather than viewed as a single “procedure swap”. ²
• Endoscopic versus surgical step-up: A later multicentre randomised trial compared an endoscopic transluminal step-up pathway with a surgical step-up pathway, supporting endoscopic step-up as a credible alternative strategy in centres with advanced therapeutic endoscopy and potentially different fistula/abdominal wall morbidity profiles. ³
• Timing of intervention: A subsequent randomised trial of immediate versus postponed drainage for infected necrosis reinforced that earlier intervention does not necessarily improve clinical outcomes and increases the procedural burden, conceptually aligning with the PANTER-era principle of deferring invasive escalation where feasible. ⁴
• Direct necrosectomy modality trials: A prior randomised comparison of endoscopic transgastric necrosectomy versus surgical necrosectomy informed the evolution toward less invasive necrosectomy approaches but was smaller and focused on different technical questions than the staged “step-up” concept. ⁵
• Evidence synthesis: Meta-analytic comparisons of endoscopic versus minimally invasive surgical approaches have generally supported fewer external fistulas and shorter recovery signals with endoscopic strategies in selected populations, while highlighting heterogeneity in necrosis maturity, timing, and endpoint definitions across trials. ⁶

Summary

• PANTER randomised 88 patients (43 step-up; 45 open) with (suspected) infected necrotising pancreatitis requiring intervention.
• The step-up strategy reduced major complications or death (40% vs 69%; RR 0.57; 95% CI 0.38 to 0.87; P=0.006), driven mainly by fewer new-onset multiple-organ failure/systemic complications (12% vs 42%).
• Mortality was similar (19% vs 16%), consistent with limited power for survival differences.
• Step-up avoided necrosectomy in 40% of patients and reduced late morbidity (incisional hernia 7% vs 24%; new-onset diabetes 16% vs 38%; pancreatic enzyme use 7% vs 33%).
• Despite more drainage procedures, step-up reduced total operations (53 vs 91) and lowered mean total costs (€90,086 vs €122,268).

Overall Takeaway

PANTER established the “step-up” paradigm as a morbidity-sparing strategy for infected necrotising pancreatitis requiring intervention, demonstrating a substantial reduction in major complications without a detectable mortality difference. Its lasting impact lies in reframing management toward staged, minimally invasive escalation—drain first, debride only if necessary—while foregrounding late morbidity and resource utilisation as core outcomes in this high-risk disease.

Bibliography

• 1Besselink MG, van Santvoort HC, Nieuwenhuijs VB, et al. Minimally invasive 'step-up approach' versus maximal necrosectomy in patients with acute necrotising pancreatitis (PANTER trial): design and rationale of a randomised controlled multicenter trial. BMC Surg. 2006;6:6.
• 2Warshaw AL. Improving the treatment of necrotizing pancreatitis — a step up. N Engl J Med. 2010;362(16):1535-1537.
• 3van 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;391:51-58.
• 4Boxhoorn 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.
• 5Bakker OJ, van Santvoort HC, van Brunschot S, et al. Endoscopic transgastric vs surgical necrosectomy for infected necrotizing pancreatitis: a randomized trial. JAMA. 2012;307(10):1053-1061.
• 6Bang JY, Arnoletti JP, Holt BA, et al. An endoscopic transluminal approach compared with minimally invasive surgery reduces complications and costs for patients with necrotizing pancreatitis. Dig Endosc. 2020;32(2):205-214.