Early Detection Research Network

Operative drainage following pancreatic resection: analysis of 1122 patients resected over 5 years at a single institution.

The only prospective randomized trial evaluating the use of intraperitoneal drainage following pancreatic resection was published from our institution approximately 10 years ago. The current study sought to evaluate the evolution of practice over the last 5 years.

Between June 2006 and June 2011, there were 1122 resections performed. Six surgeons were evenly grouped and compared by practice pattern: routine drainers (drains placed > 95%), selective drainers, and routine nondrainers (drains placed ∼15%). Prospectively recorded preoperative, operative, and morbidity data were assessed in uni- and multivariate models.

Our operative drainage rate was 49% and decreased over time (62% 2006-2008 vs 37% 2009-2011, P < 0.001). Patients without operative drains had significantly lower grade ≥3 overall morbidity (26% vs 33%; P = 0.01), shorter hospital stays (7 vs 8 days; P < 0.01), fewer readmissions (20% vs 27%; P = 0.01), and lower rates of grade ≥3 pancreatic fistula (16% vs 20%; P = 0.05). Similar reoperation (both <1%), interventional radiology procedures (15% vs 19%; P = 0.1), and mortality rates (2% vs 1%; P = 0.3) were seen in both groups. There were no differences between the routine drainers group (n = 248) and the nondrainers group (n = 478) in grade ≥3 fistula or need for interventional radiology-guided procedures.

In this study, operative drains were used nearly half of the time and were associated with longer hospital stay, and higher grade ≥3 morbidity, fistula, and readmission rates. They did not decrease the need for reintervention or alter mortality rates. Routine prophylactic drainage after pancreatic resection could be safely abandoned.

Allen PJ, Brennan MF, Correa-Gallego C, Dematteo RP, Dʼangelica M, Fong Y, Jarnagin WR, Kingham TP


Ann. Surg., 2013, 258 (6)

Version 5.0.2