Introduction: PD is a complex operation associated with a marked systemic inflammatory response and significant fluid shifts. Establishing a benchmark for ideal perioperative fluid management is critical to optimising PD patient recovery.
Methods: In this retrospective study, we evaluated perioperative fluid data for patients undergoing PD. We compared an optimal benchmark group who were discharged home by postoperative day five (≤5day) to a group of patients with an in hospital recovery greater than ten days (≥10day).
Results: Seventy-six patients who underwent PD between June 2015 and November 2016 were evaluated. The ≤5day group had a significantly lower intraoperative fluid administration (5.4 vs. 6.6 L, p= 0.012), despite similar operative times (447 mins and 476 mins, respectively). POD1 cumulative fluid balance was lower in the ≤5day group compared to the ≥10day group, 7.8L (97 mL/kg) vs. 9.7L (148 mL/kg) (p= 0.002), respectively. As expected, the postoperative complication rate was reduced in the ≤5day group (5% vs. 95%). Complications included pancreatic fistula (40%), delayed gastric emptying (53%), and intra-abdominal infection (16%). The median weight change from baseline to POD5 was -0.2 Kg for the ≤5day group compared to +2.9 Kg for ≥10day group (p= 0.000006).
Conclusions: Patients in the benchmark PD group received less fluid intraoperatively, had a lower cumulative fluid balance by POD1, and were able to return to their preoperative weight by POD5 when compared to ≥10day group. These data offer insights into optimal fluid administration for PD patients.
Recommended CitationMoskal, David; Walsh, Carrie D.; Duque, Sofia; Yeo, MD, FACS, Charles; and Lavu, MD, FACS, Harish, "Early Perioperative Fluid Benchmarking to Predict Pancreaticoduodenectomy (PD) Outcomes" (2020). Phase 1. Paper 32.