K.W. Blackburn1, H. Engebretson1, A.W. Wood2, D.I. Cope1, C. Hsu2, E.J. Silberfein2, S.S. Awad2, E.A. Choi2, C.Y. Chai2, E. Camp2, G. Van Buren2, W.E. Fisher2, D.J. Erstad2 1Baylor College of Medicine, School Of Medicine, Houston, TX, USA 2Baylor College of Medicine, Division Of Surgical Oncology, Houston, TX, USA
Introduction: With the advent of ERAS protocols, there has been a trend towards restricted fluid management for pancreatic surgery, which may be beneficial for reducing postoperative complications and shortening recovery time. In this study, we hypothesized that restricted fluid administration would be associated with reduced postoperative complications and shorter hospital length of stay for patients treated with pancreaticoduodenectomy at our institution.
Methods: Retrospective study design evaluating patients treated with pancreaticoduodenectomy at a large, single institution from 2004-2024. Patients were stratified by total intra-operative fluid administration (colloid and crystalloid): <500 mL/hr (low fluid group) or ≥500 mL/hr (high fluid group). Continuous variables were compared using the Wilcoxon rank sum test and categorical variables were compared using the Pearson’s chi-squared test. Propensity score matching (PSM) was performed using a 1:1 nearest neighbor scheme, matching on receipt of neoadjuvant chemotherapy, pathologic diagnosis, estimated blood loss, and technical aspects (minimally invasive and/or pylorus preservation).
Results: Of 933 patients included in the study, 391 (41.9%) received ≥500 mL/hr during their operation. High fluid patients received a median of 4800 mL (3700-6200 IQR) compared to 2600 mL for low fluid patients (2063-3138 IQR). High fluid patients were less likely to have received neoadjuvant chemotherapy (9.0% vs. 16.6%; P=.001) or a minimally invasive operation (10.2% vs. 17.5%; P=.002) but were more likely to receive a pylorus-preserving pancreaticoduodenectomy (65.0% vs. 56.1%; P=.008). After PSM (n=297 per group), there was no difference in overall rate of complication (e.g. pancreatic fistula/leak, ARDS, wound infection, or biliary leak) between high and low fluid administration. However, high fluid administration was associated with delayed return of bowel function, including longer time to tolerate clear liquids [3 (1-4) vs. 2 (1-3); P<.001] and solid food [5 (3-6) vs. 4 (2-5); P=.002].
Conclusion: High fluid administration (≥500 mL/h) during pancreaticoduodenectomy at our institution was associated with delayed return of bowel function without any concomitant increase in complication risk. This finding might support the restricted delivery of fluids in patients receiving a pancreaticoduodenectomy.