Intraoperative crystalloid utilization variability and association with postoperative outcomes: A post hoc analysis of two multicenter prospective cohort studies.

Abstract

The optimal regimen for intravenous administration of intraoperative fluids remains unclear. Our goal was to analyze intraoperative crystalloid volume administration practices and their association with postoperative outcomes. We extracted clinical data from two multicenter observational studies including adult patients undergoing colorectal surgery and total hip (THA) and knee arthroplasty (TKA). We analyzed the distribution of intraoperative fluid administration. Regression was performed using a general linear model to determine factors predictive of fluid administration. Patient outcomes and intraoperative crystalloid utilization were summarized for each surgical cohort. Regression models were developed to evaluate associations of high or low intraoperative crystalloid with the likelihood of increased postoperative complications, mainly acute kidney injury (AKI) and hospital length of stay (LOS). 7580 patients were included. The average adjusted intraoperative crystalloid infusion rate across all surgeries was to 7.9 (SD 4)mL/kg/h. The regression model strongly favored the type of surgery over other patient predictors. We found that high fluid volume was associated with 40% greater odds ratio (OR 1.40; 95% confidence interval 1.01-1.95, p=0.044) of postoperative complications in patients undergoing THA, while we found no associations for the other types of surgeries, AKI and LOS CONCLUSIONS: A wide variability was observed in intraoperative crystalloid volume administration; however, this did not affect postoperative outcomes.