Oral Abstract
Patient Blood Management
Carly Maucione, MD (she/her/hers)
Resident Physician
WashU in St. Louis
ST LOUIS, Missouri
Disclosure information not submitted.
Patient samples collected directly from intravenous (IV) catheters may be contaminated with IV fluid, leading to false decreases in measured analytes. The frequency of this lab error in BMPs has been previously reported; however, less is known about its frequency in complete blood counts (CBCs). Furthermore, it is unknown whether and how frequently these contamination events lead to unnecessary red blood cell transfusions (RBCt).
Study
Design/Methods:
CBCs and RBCt from Washington University in St. Louis (WashU) and University of Utah (Utah) were extracted from the Laboratory Information System (LIS) and Electronic Medical Record (EMR). Only CBC “trios” where a current CBC was paired with a prior and post available within 48 hours were included. For each site, 1000 CBC trios were randomly selected and reviewed by one reviewer, who was asked to determine whether the current CBC result was contaminated or not, as evidenced by a transient drop and then recovery in analyte values between the prior and current, and then the current and post-analyte CBCs. Reviewers had access to the EMR to evaluate clinical presentation, vital signs, additional lab results, and fluid administration to determine if the analyte changes might be explained by a mechanism other than contamination.
To assess the impact of potential contamination on RBCt practices, reviewers were asked to review a subgroup of 1000 randomly selected cases in which a RBCt was administered between the current and post-CBC. In addition to supporting evidence from chart review of vital signs, fluid administration, and laboratory documentation, experts flagged transfusions as potentially unnecessary if the current CBC was deemed contaminated and the post-hemoglobin was greater than the prior.
Results/Findings:
Reviewers identified contamination in 5.3% of Utah and 3.7% of WashU cases reviewed. Reviewers identified higher rates of contamination in the transfused subgroup, with rates of 11.8% and 12.5% at Utah and WashU, respectively. The reviewers reported that the percentages of potentially unnecessary transfusions were 8.1% and 7.6% at Utah and WashU, respectively.
Conclusions:
As many as 4-5% of CBCs may be affected by IV fluid contamination, suggesting that this is a vastly underreported laboratory error. Furthermore, as many as 8% of RBCt might not have been administered if not for IV fluid contamination. Future work will focus on improving the detection of CBC contamination in real time clinical laboratory workflows to reduce blood product waste and patient harm.