Identifying the Optimal case-volume threshold for pancreatectomy in contemporary practice

Area of Science:

• Surgical Oncology
• Health Services Research

Background:

• The association between surgical volume and patient outcomes has influenced recommendations for centralizing complex procedures like pancreatectomy.
• Despite a general decline in pancreatectomy mortality, established case-volume thresholds have not been updated.

Purpose of the Study:

• To analyze trends in 30-day mortality following pancreatectomy.
• To evaluate the impact of hospital case volume on mortality rates over time.
• To reassess current hospital volume thresholds for pancreatic surgery.

Main Methods:

• Utilized data from the National Cancer Database (NCDB) for patients undergoing pancreatectomy between 2004 and 2013.
• Categorized hospitals into low (LV), medium (MV), and high-volume (HV) groups based on 30-day mortality quartiles and logistic regression with cubic splines.
• Calculated adjusted absolute differences and odds of 30-day mortality across hospital volume strata.

Main Results:

• Hospital volumes for LV, MV, and HV were defined as <4, 4-18, and >18 cases using quartiles, and <6, 6-18, and >18 using cubic splines.
• Observed a downward trend in absolute 30-day mortality, with notable improvements in MV and LV hospitals.
• The benchmark 30-day mortality for high-volume hospitals (>18 cases) was 2.8%, with the associated case-volume threshold decreasing from 31 to 6 between 2004 and 2013.

Conclusions:

• Significant improvements in 30-day mortality were observed in low-volume (LV) and medium-volume (MV) hospitals.
• By 2013, the odds of 30-day mortality became comparable between medium-volume (MV) and high-volume (HV) hospitals.
• These findings suggest a potential need to revise current hospital volume thresholds for pancreatic surgery.