Retrospective analysis of community hospital red blood cell recovery procedures: improved utilization needed for effectiveness.

BACKGROUND: Perioperative blood recovery (PBR) is an important component of patient blood management. We analyzed our experience providing PBR for community hospitals to determine procedure types and clinical variables associated with efficacy and cost-effectiveness. STUDY DESIGN AND METHODS: PBR cases (>25,000) from January 2008 through December 2012 were analyzed. For each procedure type, the median number of returned red blood cell units (rRBCs) and ratio of cases with at least 1 to less than 1 rRBC unit were calculated. Clinical predictors of rRBC were identified by linear and logistic regression. RESULTS: The overall median rRBC was 0.29 units despite median estimated blood loss (EBL) of 350 mL. Only three of 31 common procedure types had ≥1:<1 rRBC ratios near to or higher than 1. In nine of 31 common procedure types, at least 50% of cases had no rRBC return. Linear regression demonstrated significant association of rRBCs with increased EBL, longer operative duration, surgeon, PBR device type (autoLog vs. CS5), and decreasing age. EBL, autoLog use, high surgeon case volume, vascular procedures, and emergent versus elective procedures associated with higher odds of at least 1 rRBC. CONCLUSION: Discrepancy between rRBC and EBL and high percentages of cases with no rRBC suggests that PBR technique and case selection need optimization. Identification of procedure types and variables associated with PBR efficacy (≥1 rRBC) should improve utilization of PBR. Association of autoLog use with higher rRBC warrants further investigation.

Duke Scholars

Author Beth H. Shaz Pathology