A propensity score-matched analysis of laparoscopic versus open pancreaticoduodenectomy: Is there value to a laparoscopic approach?

BACKGROUND: Laparoscopic pancreaticoduodenectomy has been found safe and associated with advantages over open pancreaticoduodenectomy in prior studies. We compared outcomes of laparoscopic pancreaticoduodenectomy versus open pancreaticoduodenectomy at a single institution after applying technical aspects and perioperative care learned from laparoscopic pancreaticoduodenectomy to the open pancreaticoduodenectomy practice. METHODS: From January 2010 to December 2020, all patients undergoing pancreaticoduodenectomy were identified, and information was collected in a prospective fashion. Open pancreaticoduodenectomy (n = 347) and laparoscopic pancreaticoduodenectomy (n = 242) were performed using the same selection criteria, operative technique, and recovery protocols at a single institution. Propensity score matching was performed, and then perioperative data and 90-day outcomes were compared, and statistical analysis was performed. RESULTS: A total of 589 patients underwent pancreaticoduodenectomy, including open pancreaticoduodenectomy (n = 347) and laparoscopic pancreaticoduodenectomy (n = 242). After excluding those undergoing total pancreatectomy or major vascular or concomitant organ resection, there were 497 patients (open pancreaticoduodenectomy = 301 and laparoscopic pancreaticoduodenectomy = 196). Propensity score matching was performed, and 187 open pancreaticoduodenectomy patients were matched to 187 laparoscopic pancreaticoduodenectomy patients. Operative time (475 vs 280 minutes) was longer, and estimated blood loss (150 vs 212 mL) was less for laparoscopic pancreaticoduodenectomy than open pancreaticoduodenectomy, respectively. Pancreatic fistula (18.8% vs 5.4%) and delayed gastric emptying (18.8% vs 9.7%) were higher for laparoscopic pancreaticoduodenectomy than open pancreaticoduodenectomy, respectively. Postpancreatectomy hemorrhage, major morbidity, mortality, hospital stay, and readmissions were nonsignificantly higher for laparoscopic pancreaticoduodenectomy than open pancreaticoduodenectomy. Intensive care use and overall costs were significantly higher for laparoscopic pancreaticoduodenectomy than open pancreaticoduodenectomy. CONCLUSION: In our experience, open pancreaticoduodenectomy offers similar to improved outcomes over laparoscopic pancreaticoduodenectomy, with less use of perioperative resources, thereby offering better value to patients requiring pancreaticoduodenectomy.

Comparison of complications for cirrhotic versus non-cirrhotic patients undergoing pancreaticoduodenectomy.

PURPOSE: To compare the outcomes between patients with cirrhosis and those without who have undergone pancreatoduodenectomy (PD) in our institution. METHODS: A review of patients undergoing PD from the time period of January 2010 to December 2020 was performed. Patients that have undergone open or laparoscopic PD and had liver cirrhosis diagnosed prior to surgery were included and matched on a 1:2 basis with non-cirrhotic patients based on age, gender, Eastern Cooperative Oncology Group (ECOG), and date of surgery. Data was obtained from our medical records and ten major postoperative complications variables were compared to the matched group. RESULTS: Overall, 16 patients with cirrhosis were compared to 32 matched controls. No significant differences were found in pancreatic fistula (18.8% vs. 21.8%; P= 1.000), hemorrhage (6.3% vs. 6.2%; P= 1.000), delayed gastric emptying (6.3% vs. 15.6%; P= 0.648), wound infection (0% vs. 9.3%; P= 0.541), and intraabdominal abscess (31.2% vs 6.2%; 0.4998) for cirrhotic vs. non-cirrhotic respectively. There were no postop ileus, gastric fistula, mesenteric portal thrombosis, biliary fistula, and abdominal ischemic event in either group. The average length of stay for both groups was similar (6.9 vs. 9.3 days; P= 0.4019). There were no mortalities and major morbidity was similar (37.5% vs 34.3%; P=0.3549). One patient required readmission for liver-related decompensation with full recovery. CONCLUSION: PD in patients with cirrhosis can be safe and feasible in well-selected patients. In a high-volume institution, postoperative complications are similar to those patients without cirrhosis of the liver.