Original Articles

The Influence of Intraoperative Autotransfusion on Postoperative Hematocrit after Cardiac Surgery: A Cross-Sectional Study

Address communications to: Andrew J. Stasko, MS, CCP, SpecialtyCare, Inc., 3100 West End Avenue, Suite 800, Nashville, TN 37203. E-mail: Andy.stasko@specialtycare.net

Received: 27 April 2017
Accepted: 10 July 2017

Abstract

Utilization of intraoperative autotransfusion (IAT) during cardiac surgery with cardiopulmonary bypass (CPB) has been shown to reduce allogeneic red blood cell transfusion. Previous research has emphasized the benefits of using IAT in the intraoperative period. The present study was designed to evaluate the effects of using IAT on overall hematocrit (Hct) drift between initiation of CPB and the immediate postoperative period. We reviewed 3,225 adult cardiac procedures occurring between February 2016 and January 2017 at 84 hospitals throughout the United States. Data were collected prospectively from adult patients undergoing cardiac surgery with CPB, and stored in the SpecialtyCare Operative Procedural rEgistry (SCOPE), a large quality improvement database. Patients receiving allogeneic transfusion and those with missing covariate data were excluded from analysis. The effect of IAT volume returned to patients on the primary endpoint, hematocrit change from CPB initiation to intensive care unit (ICU) entry, was assessed using a multivariable linear mixed effects regression model controlling for patient demographics, operative characteristics, surgeon, and hospital. Descriptive analysis showed greater positive hematocrit change with increasing autotransfusate volume returned. Those patients with no IAT volume returned saw a median hematocrit change of +2.00%, whereas those with more than 380 mL/m² BSA had a median Hct drift of +5.00% (p < .001). After controlling for known confounds, our regression estimate of the effect of IAT volume returned on Hct drift was +.0045% per 1 mL/m² BSA (p < .001). For a patient with the median autotransfusate volume returned (273 mL/m² BSA), and all other covariate values at their respective medians, this translates to a predicted hematocrit change of +3.6% (95% CI +3.1 to +4.1). These findings lend further support to the notion that autotransfusate volume is positively associated with increases in postoperative hematocrit.