NESS - A Comparative Analysis of Perioperative Outcomes Between Open Distal Pancreatectomy Versus Minimally Invasive Approaches Using the American College of Surgeons National Quality Improvement Program Targeted Pancreatectomy Dataset.

Back to 2016 Annual Meeting Posters

A Comparative Analysis of Perioperative Outcomes Between Open Distal Pancreatectomy Versus Minimally Invasive Approaches Using the American College of Surgeons National Quality Improvement Program Targeted Pancreatectomy Dataset.
*Julius I. Ejiofor¹, *Moska Hamidi¹^,2, *Gaurav Sharma¹, *Heather G. Lyu¹, *Ethan Y. Brovman³, *Richard D Urman³, Thomas E. Clancy¹, Edward E. Whang¹
¹Department of General Surgery, Brigham and Women's Hospital, Boston, MA;²Department of General Surgery, Western University, London Health Sciences Centre, London, ON, Canada³Department of Anesthesiology, Preoperative and Pain Medicine, Brigham and Women's Hospital, Boston, MA

Objective: Compare perioperative outcomes between open distal pancreatectomy and minimally invasive alternatives in a national cohort.
Design: Retrospective cohort study.
Setting: 106 U.S. hospitals participating in the 2014 ACS NSQIP Targeted Pancreatectomy Dataset. This inaugural dataset contains detailed pancreatectomy-specific variables.
Participants: Adults undergoing primary distal pancreatectomy for benign or malignant etiology from Jan 1-Dec 31 2014. We excluded patients with disseminated cancers, emergent indications, misclassification of key variables, concurrent organ resection except splenectomy.
Interventions: Open, laparoscopic(LDP) or robotic distal pancreatectomy(RDP).
Main Outcome Measures: Perioperative morbidity, mortality, unplanned readmission and reoperation rates within 30days.
Results: 1254 patients underwent DP: ODP=599, LDP=498, RDP=157. The cohorts were comparable at baseline except the ODP cohort had more patients with neoadjuvant chemoradiation(p<0.0001), and malignant indications(p<0.0225).RDP had the longest median operative time of 234 minutes(ODP 206 vs LDP204, p<0.0001), and shortest median hospital stay(8.08d ODP vs 5.92d LDP vs 5.67d RDP, p<0.0001). ODP had the highest rate of bleeding requiring transfusions(p<0.0001), unadjusted rates of deep organ infection(p<0.0328), and unadjusted readmission rates(19.7% ODP vs 13.7% vs 17.8%, p<0.0238). There was no difference in adjusted readmission rates, rates of postoperative percutaneous drainage, inpatient mortality or pancreatic fistula(ODP 16.2% vs LDP18.3% vs RDP 21.15%, p=0.3549), unadjusted reoperation rates(4.5% ODP vs 2.8% LDP vs 2.5% RDP, p=0.2437). Logistic regression identified COPD, preoperative weight loss, postoperative DVT/PE, wound infections, and fistulas as predictors of readmission(OR 3.0, p=0.0046, 2.3 p=0.0079, 11.8 p<0.0001, 10.9 p<0.0001, 2.5 p<0.0001, respectively).
Conclusions: Compared with open, RDP and LDP are associated with shorter hospital stay, without an increase in morbidity and mortality. All three had similar 30-day readmission rates. Post-operative DVT/PE and wound infections are the two strongest predictors of 30-day readmission.

Back to 2016 Annual Meeting Posters