Perioperative fluid management in esophagectomy for cancer and its relation to postoperative respiratory complications.

The optimal perioperative fluid management during esophagectomy is still not clear. Liberal regimens have been associated with higher morbidity and respiratory complications. Restrictive regimens might raise concerns for kidney function and increase the need to associate vasopressors. The aim of this study was to investigate retrospectively the perioperative fluid administration during esophagectomy and to correlate this with postoperative respiratory outcome. All patients who underwent esophagectomy between January and December 2016 were retrospectively analyzed. Patient characteristics, type of surgery and postoperative course were reviewed. Fluid administration and vasopressor use were calculated intraoperatively and during the postoperative stay at the recovery unit. Fluid overload was defined as a positive fluid balance of more than 125 mL/m2/h during the first 24 hours. Patients were divided in 3 groups: GRP0 (no fluid overload/no vasopressors); GRP1 (no fluid overload/need for vasopressors); GRP2 (fluid overload with/without vasopressors). Postoperative complications were prospectively recorded according to Esophagectomy Complications Consensus Group criteria. A total of 103 patients were analyzed: 35 (34%) GRP0, 50 (49%) GRP1 and 18 (17%) GRP2. No significant differences were found for age, treatment (neoadjuvant vs. primary), type of surgery (open/minimally invasive), histology nor comorbidities. There were significant (P ≤ 0.001) differences in fluid balance/m2/h (75 ± 21 mL; 86 ± 22 mL and 144 ± 20 mL) across GRP0, GRP1 and GRP2, respectively. We found differences in respiratory complications (GRP0 (20%) versus GRP1 (42%; P = 0.034) and GRP0 (20%) versus GRP2 (61%; P = 0.002)) and "Comprehensive Complications Index" (GRP0 (20.5) versus GRP1 (34.6; P = 0.015) and GRP0 (20.5) versus GRP2 (35.1; P = 0.009)). Multivariable analysis (binary logistic regression) for "any respiratory complication" was performed. Patients who received fluid overload (GRP2) had a 10.24 times higher risk to develop postoperative respiratory complications. When patients received vasopressors alone (GRP1), the chances of developing these complications were 3.57 times higher compared to GRP0. Among patients undergoing esophagectomy, there is a wide variety in the administration of fluid during the first 24 hours. There was a higher incidence of respiratory complications when patients received higher amounts of fluid or when vasopressors were used. We believe that a personalized and protocolized fluid administration algorithm should be implemented and that individual risk factors should be identified.