Impact of intraoperative fluid management on postoperative complications in patients undergoing minimally invasive esophagectomy for esophageal cancer: a retrospective single-center study

Abstract

Abstract Background Esophagectomy is a high-risk procedure that can involve serious postoperative complications. There has been an increase in the number of minimally invasive esophagectomies (MIEs) being performed. However, the relationship between intraoperative management and postoperative complications in MIE remains unclear. Methods After the institutional review board approval, we enrolled 300 patients who underwent MIE at Tohoku University Hospital between April 2016 and March 2021. The relationships among patient characteristics, intraoperative and perioperative factors, and postoperative complications were retrospectively analyzed. The primary outcome was the relationship between intraoperative fluid volume and anastomotic leakage, and the secondary outcomes included the associations between other perioperative factors and postoperative complications. Results Among 300 patients, 28 were excluded because of missing data; accordingly, 272 patients were included in the final analysis. The median [interquartile range] operative duration was 599 [545–682] minutes; total intraoperative infusion volume was 3,747 [3,038–4,399] mL; total infusion volume per body weight per hour was 5.48 [4.42–6.73] mL/kg/h; and fluid balance was + 2,648 [2,015–3,263] mL. The postoperative complications included anastomotic leakage in 68 (25%) patients, recurrent nerve palsy in 91 (33%) patients, pneumonia in 62 (23%) patients, cardiac arrhythmia in 13 (5%) patients, acute kidney injury in 5 (2%) patients, and heart failure in 5 (2%) patients. The Cochrane-Armitage trend test indicated significantly increased anastomotic leakage among patients with a relatively high total infusion volume (P = 0.0085). Moreover, anastomotic leakage was associated with male sex but not with peak serum lactate levels. Patients with a longer anesthesia duration or recurrent nerve palsy had a significantly higher incidence of postoperative pneumonia than those without. Further, the incidence of postoperative pneumonia was not associated with the operative duration, total infusion volume, or fluid balance. The operative duration and blood loss were related to the total infusion volume. Acute kidney injury was not associated with the total infusion volume or serum lactate levels. Conclusions Among patients who underwent MIE, the total infusion volume was positively correlated with the incidence of anastomotic leakage. Further, postoperative pneumonia was associated with recurrent nerve palsy but not total infusion volume or fluid balance.