Reinforced stapler versus ultrasonic dissector for pancreatic transection and stump closure for distal pancreatectomy: A propensity matched analysis

Background: Postoperative pancreatic fistula is the primary contributor to morbidity after distal pancreatectomy. To date, no techniques used for the transection and closure of the pancreatic stump have shown clear superiority over the others. This study aimed to compare the rate of postoperative pancreatic fistula after pancreatic transection conducted with a reinforced stapler versus an ultrasonic dissector after a distal pancreatectomy. Method: Prospectively collected data of consecutive patients who underwent distal pancreatectomy from 2014 to 2017 were reviewed retrospectively. We included distal pancreatectomies in which pancreatic transection was performed by reinforced stapler or ultrasonic dissector; we excluded extended distal pancreatectomies. To overcome the absence of randomization, we conducted a propensity matching analysis according to risk factors for postoperative pancreatic fistula. Results: Overall, 200 patients met the inclusion criteria. The reinforced stapler was employed in 108 patients and the ultrasonic dissector in 92 cases. After one-to-one propensity matching, 92 patients were selected from each group. The matched reinforced stapler and ultrasonic dissector cohort had no differences in baselines characteristics except for the mini-invasive approach, which was more common in the ultrasonic dissector group (34% vs 51%, P = .025). Overall, 48 patients (26%) developed a postoperative pancreatic fistula, 46 (25%) a grade B postoperative pancreatic fistula, and 2 (1%) a grade C postoperative pancreatic fistula. In the reinforced stapler group, the rate of postoperative pancreatic fistula was 12% (n = 11) and in the ultrasonic dissector group 40% (n = 37) with a P < .001. Conclusion: The results of this study suggest that the use of reinforced stapler for pancreatic transection decreases the risk of postoperative pancreatic fistula. A randomized trial is required to confirm these preliminary data.