POstopeRative Standardization of Care: THe Implementation of Best Practice After Pancreatic Resection

Rationale

Pancreatic resection is a major abdominal operation with 50% chance of postoperative complications. A feared complication is severe pancreatic fistula, in which there is leakage of enzyme rich fluid into the abdominal cavity. Adequate complication management appears to be the most important factor in improving outcomes of patients undergoing pancreatic resection.

Objective

To investigate whether implementation of a best practice algorithm for postoperative care focusing on early detection and step-up management of postoperative pancreatic fistula results in a lower rate of major complications and death after pancreatic resection as compared to current practice

Study design

A nationwide stepped-wedge, cluster randomized, superiority trial. In this design all participating centers cross over from current practice to best practice according to the algorithm, but are randomized to determine the exact order. At the end of the trial, all centers will have implemented the best practice algorithm.

Study population

All centers performing pancreatic surgery in the Netherlands (i.e. the Dutch Pancreatic Cancer Group).

Intervention

Cluster level education on postoperative care according to a best practice algorithm, focusing on early detection and step-up management of postoperative pancreatic fistula. This algorithm is based on findings in Dutch observational cohort studies, systematic literature analyses, an inventory in current protocols on postoperative care and expert opinion. The proposed algorithm is validated in a multicenter cohort and consensus upon this algorithm is reached with pancreatic surgeons from all centers of the Dutch Pancreatic Cancer Group. The final algorithm was reviewed critically by the advisory committee of internationally respected experts in the field of pancreatology before implementation in this trial.

Comparison

Postoperative care according to current practice.

Endpoints

The primary outcome was measured in all patients undergoing pancreatic resection and is a composite of major complications (i.e. postpancreatectomy bleeding, new-onset organ failure and death). Secondary endpoints include the individual components of the primary endpoint and other clinical outcomes, number of patients receiving adjuvant chemotherapy, healthcare resource utilization and costs analysis. Follow-up will be 90 days after pancreatic resection.

• Best practice algorithm for postoperative care (Other): See arm/group description

Primary outcomes

• Composite primary endpoint [90 days after index pancreatic resection]: The primary endpoint of this trial is a composite of the most severe complications associated to postoperative pancreatic fistula. This endpoint will be considered positive if one of the following complications occurs within 90 days after pancreatic resection: late postpancreatectomy bleeding, new-onset organ failure and/or death

Secondary outcomes

• Postoperative mortality [90 days after index pancreatic resection]: Measured as rate of death at 90-day follow-up
• New-onset organ failure [90 days after index pancreatic resection]: Measured as organ failure occuring any time within 90 days after resection, not present at time of index pancreatic resection.
• Late postpancreatectomy bleeding [90 days after index pancreatic resection]: Defined in accordance to the International Study Group on Pancreatic Surgery (ISGPS) definition as bleeding occurring any time after 24 hours after pancreatic resection.
• Postoperative morbidity [90 days after index pancreatic resection]: Including complications according to the Clavien-Dindo system and pancreatectomy specific complications according to the ISGPS definitions (e.g. postpancreatectomy bleeding, postoperative pancreatic fistula, postoperative bile leak, postoperative chyle leak and delayed gastric emptying).
• Adjuvant chemotherapy [90 days after index pancreatic resection]: Measured as number of patients receiving adjuvant chemotherapy at 90-day follow-up
• Success of implementation [90 days after index pancreatic resection]: Measured as number of patients in whom the algorithm was not followed and timing of abdominal CT scans in both strategies.
• Cost-effectiveness [90 days after index pancreatic resection]: Calculated by comparing health effects and medical costs related to both strategies up to 90 days after pancreatic resection.

Sex: All, accepts healthy volunteers

Inclusion Criteria for Clusters:

* All Dutch centers performing pancreatic surgery (i.e. performing at least 20 pancreatoduodenectomies a year)

Exclusion Criteria for Clusters:

* None

Inclusion Criteria for Patients:

* Patients underoging pancreatic resection for any indication

Exclusion Criteria for Patients:

* None (i.e. complete enumeration)

• Results publication: Smits FJ, Molenaar IQ, Besselink MG, Busch OR, van Eijck CH, van Santvoort HC; Dutch Pancreatic Cancer Group. Management of postoperative pancreatic fistula after pancreatoduodenectomy: high mortality after completion pancreatectomy: Reply to: Bressan et al. completion pancreatectomy in the acute management of pancreatic fistula after pancreaticoduodenectomy. HPB (Oxford). 2018 Dec;20(12):1223. doi: 10.1016/j.hpb.2018.05.015. Epub 2018 Jun 22. No abstract available. PMID 29941289

• Quintus Molenaar, MD, PhD, Principal investigator, UMC Utrecht
• Hjalmar C van Santvoort, MD, PhD, Principal investigator, St. Antonius Hospital