Minimally Invasive Versus Open Pancreaticoduodenectomy: A Propensity-matched Study From a National Cohort of Patients.

Abstract

To compare the perioperative outcomes of minimally invasive pancreaticoduodenectomy (MIPD) in comparison with open pancreaticoduodenectomy (OPD) in a national cohort of patients. Limited well-controlled studies exist comparing perioperative outcomes between MIPD and OPD. Patients who underwent MIPD and OPD were abstracted from the 2014 to 2015 pancreas-targeted American College of Surgeons National Surgical Quality Improvement Program. OPD and MIPD patients were matched 3:1 using propensity score, and perioperative outcomes were compared. A total of 4484 patients were identified with 334 (7.4%) undergoing MIPD. MIPD patients were younger, more likely to be White, and had a lower rate of weight loss. They were more likely to undergo classic Whipple and to have a drain placed. After 3:1 matching, 1002 OPD patients were compared with 334 MIPD patients. MIPD was associated with longer mean operative time (426.6 vs 359.6 minutes; P < 0.01), higher readmission rate (19.2% vs 14.3%; P = 0.04) and lower rate of prolonged length of stay >14 days (16.5% vs 21.6%; P = 0.047). The 2 groups had a similar rate of 30-day mortality (MIPD 1.8% vs OPD 1.3%; P = 0.51), overall complications, postoperative pancreatic fistula, and delayed gastric emptying. A secondary analysis comparing MIPD without conversion or open assist with OPD showed that MIPD patients had lower rates of overall surgical site infection (13.4% vs 19.6%; P = 0.04) and transfusion (7.9% vs 14.4%; P = 0.02). MIPD had an equivalent morbidity and mortality rate to OPD, with the benefit of a decreased rate of prolonged length of stay, though this is partially offset by an increased readmission rate.