Early observations and pilot data that first suggested a new direction
Early observational studies in the 1990s suggested that hospitals performing higher volumes of pancreaticoduodenectomy had lower mortality rates, but the magnitude and consistency of this relationship was debated. Smaller studies demonstrated that mortality varied substantially across institutions, raising questions about whether regionalization of complex surgical procedures could improve outcomes.
Landmark RCTs and pivotal trials that established the evidence base
Birkmeyer's 2002 NEJM study provided the definitive evidence for the volume-outcome relationship in pancreaticoduodenectomy. Using Medicare data encompassing 2.5 million procedures across 14 operation types from 1994-1999, the study demonstrated that mortality for pancreatic resection ranged from 16.3% at very-low-volume hospitals (<1/year) to 3.8% at high-volume hospitals (≥5/year) — the largest absolute difference among all procedures studied. This 12-percentage-point mortality gap was far greater than for any other surgical procedure analyzed, providing compelling evidence for regionalization.
Follow-up studies, subgroup analyses, and real-world validation
Subsequent studies extended the volume-outcome analysis to surgeon-level data. Birkmeyer's 2003 study demonstrated that surgeon volume independently predicted mortality even after adjusting for hospital volume, with the surgeon-volume effect most pronounced for pancreatic resection. International studies from the UK, Netherlands, and Japan confirmed the volume-outcome relationship, leading to regionalization policies in multiple countries. In the Netherlands, mandatory centralization of pancreatic surgery from 2005 onward was associated with a reduction in operative mortality from approximately 10% to under 5%.
Integration into clinical practice guidelines and recommendations
Volume-based referral patterns are now embedded in national healthcare policies. The Leapfrog Group established minimum volume standards for pancreatic resection, and multiple European countries implemented centralization mandates.
Leapfrog Group
Minimum volume standards for pancreatic resection to ensure quality outcomes
NCCN
Pancreaticoduodenectomy for pancreatic cancer performed at high-volume centers (≥15-20 cases/year)
Now
Current standard of care and ongoing research directions
The volume-outcome relationship for pancreaticoduodenectomy is one of the most established findings in surgical outcomes research. Regionalization to high-volume centers (typically defined as ≥20 cases/year) is associated with mortality under 3-5%. Current research focuses on defining additional quality metrics beyond volume (textbook outcomes, failure-to-rescue rates), optimizing patient selection and preoperative optimization, and understanding whether telemedicine and hub-and-spoke models can extend expertise to underserved regions.
How much does hospital volume affect mortality for pancreaticoduodenectomy?+
Birkmeyer's 2002 NEJM study of 2.5 million procedures found that operative mortality for pancreatic resection ranged from 16.3% at very-low-volume hospitals (<1 case/year) to 3.8% at high-volume hospitals (≥5 cases/year). This 12.5-percentage-point mortality gap was the largest among all 14 surgical procedures studied and provided the strongest evidence for regionalization of complex surgical care.
What is the recommended minimum volume for a center performing pancreaticoduodenectomy?+
Based on the cumulative evidence, guidelines recommend performing pancreaticoduodenectomy at centers with at least 15-20 cases per year. The Leapfrog Group established minimum volume standards, and several countries (notably the Netherlands) have implemented mandatory centralization with demonstrated mortality reductions from approximately 10% to under 5%.
