OA5-AM24-MN-07 - Transfusion Threshold Trends in Extremely Low Birthweight Infants

Monday, October 21, 2024

8:15 AM - 9:15 AM

Location: 362

CE: Zero

Presenting Author(s)

Jeanne E. Hendrickson, MD

Professor, Department of Pathology and Laboratory Medicine Emory University School of Medicine
Emory University
Atlanta, Georgia, United States

Disclosure(s): No financial relationships to disclose

PDF

Disclosure(s):

Jeanne E. Hendrickson, MD: No financial relationships to disclose

Background/Case Studies: Extremely-low-birth-weight infants (ELBW, < 1,000g) frequently receive red blood cell (RBCTx) and platelet transfusions (PTx) while in the Neonatal Intensive Care Unit (NICU), but the thresholds at which these transfusions are given are highly variable. In 2020, the two largest RBCTx threshold trials in preterm neonates reported no differences in outcomes between low and high hemoglobin (Hb) thresholds, and the largest PTx trial reported improved outcomes in preterm neonates randomized to the lower threshold (25 x 10³/µL). This current study was designed to examine whether publication of these landmark studies was associated with the use of lower RBC and PTx laboratory thresholds in a multicenter cohort of infants.

Study

Design/Methods:

Data from ELBW infants cared for at 4 hospitals participating in a multicenter study from 4/2019 to 12/2022 were included in this cohort. The primary exposure was birth year and the primary outcome was pre-RBCTx Hb value and pre-PTx platelet count, tested within 24 hours of the transfusion. Seven percent of RBCTx and six percent of PTx did not have a corresponding lab measure within 24 hours of transfusion. Mixed effects multivariable regression was used to estimate the association between year and hospital with mean pre-transfusion values, accounting for intra-infant correlation since some infants received multiple transfusions.

Results/Findings:

On evaluation of 3,004 pre-RBCTx Hb values from 467 ELBW infants with at least one RBCTx, there was a mean decrease of 0.18 g/dL (95% CI 0.14-0.23) per year (P< 0.0001). The model-estimated mean g/dL Hb values (95% CI) each year from 2019-2022 were 9.9 (9.75-10.02), 9.8 (9.69-9.94), 9.7 (9.59-9.87), and 9.3 (9.18-9.44), respectively. Admitting hospital was associated with mean pre-RBCTx Hb (P< 0.0001). On evaluation of 664 pre-PTx platelet counts from 142 ELBW infants with at least one PTx, there were no significant differences observed over the years evaluated (0.62 increase per year; 95% CI -1.31-2.55). The model-based estimated mean (x 10³/µL) pre-transfusion platelet counts (95% CI) each year from 2019-2022 were 50.4 (42.99-57.86), 49.6 (42.19-56.97), 51.6 (43.33-59.93), and 52.2 (44.10-60.21), respectively. Admitting hospital was not associated with pre-transfusion platelet count (P=0.05).

Conclusions: This study demonstrated a small but significant decrease in pre-RBCTx Hb levels from 2019-2022 in ELBW infants, which might reflect incorporation of evidence from recent trials showing no benefit from the use of higher Hb thresholds. Pre-PTx platelet counts, however, did not change over these years and remained higher than suggested by the best available evidence. These findings may reflect the need for additional studies in this patient population and better implementation strategies for evidence-based transfusion thresholds.