Risk-stratified approach may cut need for drains after pancreatoduodenectomy

Last Updated: 2016-07-18

By Will Boggs MD

NEW YORK (Reuters Health) - A risk-stratified approach to drain management may eliminate the need for drains in about a quarter of patients who undergo pancreatoduodenectomy, researchers report.

"The most important finding is that we can accurately predict those patients who do not need a drain placed because they definitively won't develop a leak," said Dr. Charles M. Vollmer Jr. from the University of Pennsylvania Perelman School of Medicine in Philadelphia.

"The most interesting finding of the study is how surgeons can intuitively suspect leak formation (and therefore need for prolonged drainage), based on drain characteristics and amylase trends," he told Reuters Health by email.

The approach begins with the intraoperative calculation of the Fistula Risk Score (FRS). Patients whose risk of clinically relevant pancreatic fistula (CR-POPF) was negligible or low (FRS 0-2) did not receive intraoperative drain placement, while those with a moderate/high risk of CR-POPF (FRS 3-10) did. Based on postoperative day 1 drain fluid amylase (DFA) levels, drains were removed on postoperative day 3 (DFA<=5,000 U/L) or at the surgeon's discretion (DFA>5000 U/L).

Dr. Vollmer's team compared CR-POPF outcomes among 260 consecutive pancreatoduodenectomies using this protocol versus outcomes among 557 similar historical controls.

More than a quarter of the patients (26.9%) had negligible/low FRS scores, but 22 patients had intraoperative drain placement anyway because of surgeon discretion. None of these patients experienced a CR-POPF.

Among the 73.1% of patients with moderate/high FRS risk, the CR-POPF rate was 15.3%, the researchers report in an article online June 24 in Annals of Surgery.

Patients who had their drains removed on postoperative day 3 according to the protocol had a CR-POPF rate of 1.7%, whereas those who should have had their drains removed on postoperative day 3 but didn't because of surgeon discretion had a CR-POPF rate of 38.2%.

The rest of the patients, whose moderate/high FRS and elevated DFA dictated drain removal at the surgeon's discretion, had a CR-POPF rate of 35.9%.

Overall, CR-POPF rates were significantly lower after protocol implementation than before implementation (11.2% vs. 20.6%).

Implementation of the protocol was also associated with lower rates of severe complications, any complication, reoperation, intensive care unit admission, and percutaneous drainage for complication management. Hospital stays were also significantly shorter.

"We have sustained these same outcomes since the completion of the study," Dr. Vollmer said. "It should be emphasized that all consecutive patients were evaluated in this study. There were no 'exceptions' or 'dropouts.' We specifically designed a 'surgeon discretion' arm to the study."

"This represents how data-driven outcomes assessment can lead to progress in pancreatic surgery," he added. "While we often rely on Level 1 randomized controlled trial (RCT) data as the gold-standard, there are drawbacks to that sort of data that may not make it practical to actually apply to practice. Here we built upon data derived from two separate RCTs and merged them conceptually. This was then prospectively applied to assess actual, real-world impact."

Co-author Matthew T. McMillan told Reuters Health by email, "This protocol should serve as the standard of care in this setting, since it is the first of its kind to use a data-driven approach in order to integrate three major components of drain management - drain placement, amylase analysis, and early drain removal - into a single evidence-based protocol."

Dr. Uirá Fernandes, a GI surgeon from the Federal University of Health Sciences of Porto Alegre, Brazil, who was not involved in the study, said it is too early to say whether the proposed protocol will become the standard of care.

"Paradigm shifts take time to happen and, in medicine, require reproducibility in other institutions and scenarios," he told Reuters Health by email. "Anyway, physicians must internalize the idea that drains are not always necessary in pancreatic surgery and may even be harmful."

SOURCE: http://bit.ly/29GGAxH

Ann Surg 2016.