A Dose-Response Analysis of Crystalloid Administration during Esophageal Resection.

Abstract

The purpose of this retrospective study was to investigate the role of intraoperative crystalloid administration on postoperative hospital length of stay (phLOS) and on the incidence of previously reported adverse events in 100 consecutive patients who underwent esophageal resection. The role of previously reported patient demographics, clinical characteristics, and intraoperative crystalloid administration on the duration of phLOS underwent statistical screening criteria for multivariable analysis, including the use of an instrumental variable to measure the role of unmeasured confounders on phLOS. Tests to assess the likelihood of causality also were performed. When the volumes of intraoperative crystalloids were expressed as dose-response relationships to outcomes, progressive decreases in phLOS, variances in phLOS, and the incidences of unplanned surgical intensive care unit admission, postoperative pneumonia, respiratory failure requiring orotracheal intubation, nonsinus cardiac dysrhythmias, and anastomotic leak were observed. Intraoperative transfusion of packed red blood cells greatly increased the duration of phLOS, which was not associated with estimated blood loss, length of surgical operation, or unplanned surgical intensive care unit admission. Instrumental variable analysis revealed no significant influence on phLOS. Causality tests supported the role of intraoperative crystalloid administration in reducing the duration and variance of phLOS. A dose-response relationship was clinically observed between intraoperative crystalloid administration and the duration and variance of phLOS and with commonly reported postoperative adverse events. Intraoperative transfusion of packed red blood cells greatly increased phLOS that was not associated with the severity of the surgical operation. Instrumental variables and tests for causality further supported the role of intraoperative crystalloid administration in reducing the duration and variance of phLOS.