Timing of Red Blood Cell Transfusions and Occurrence of Necrotizing Enterocolitis: A Secondary Analysis of a Randomized Clinical Trial

Abstract

Importance: Observational studies often report that anemia and red blood cell (RBC) transfusions are associated with a higher risk of necrotizing enterocolitis (NEC) among extremely low-birthweight (ELBW) infants.

Objective: To evaluate whether there is a temporal association between 72-hour hazard periods of exposure to RBC transfusions and NEC among ELBW infants randomized to either higher or lower hemoglobin transfusion thresholds.

Design, setting, and participants: This post hoc secondary analysis of 1690 ELBW infants who survived to postnatal day 10 enrolled in the Transfusion of Prematures (TOP) randomized multicenter trial between December 1, 2012, and April 12, 2017, was performed between June 2021 and July 2023.

Exposures: First, the distribution of RBC transfusions and the occurrence of NEC up to postnatal day 60 were examined. Second, 72-hour posttransfusion periods were categorized as hazard periods and the pretransfusion periods of variable duration as control periods. Then, the risk of NEC in posttransfusion hazard periods was compared with that in pretransfusion control periods, stratifying the risk based on randomization group (higher or lower hemoglobin transfusion threshold group).

Main outcomes and measures: The primary outcome was incidence of NEC stage 2 or 3. Secondary outcomes included the incidence rates of NEC within five 10-day intervals, taking into account the number of days at risk.

Results: Of 1824 ELBW infants randomized during the TOP trial, 1690 were included in the present analysis (mean [SD] gestational age, 26.0 [1.5] weeks; 899 infants [53.2%] were female). After categorizing 4947 hazard periods and 5813 control periods, we identified 133 NEC cases. Fifty-nine of these cases (44.4%) occurred during hazard periods. Baseline and clinical characteristics of infants with NEC during hazard periods did not differ from those of infants with NEC during control periods. The risk of NEC was 11.9 per 1000 posttransfusion hazard periods and 12.7 per 1000 control periods (adjusted risk ratio, 0.95; 95% CI, 0.68-1.32; P = .74). This risk did not differ significantly between randomization groups, but the incidence rate of NEC per 1000 days peaked between postnatal days 20 and 29 in the lower hemoglobin transfusion threshold group.

Conclusions and relevance: The findings of this post hoc analysis suggest that, among ELBW infants with the hemoglobin ranges occurring in the TOP trial, exposure to RBC transfusions was not temporally associated with a higher risk of NEC during 72-hour posttransfusion hazard periods. Given that the incidence rate of NEC peaked between postnatal days 20 and 29 among infants with lower hemoglobin values, a more in-depth examination of this at-risk period using larger data sets is warranted.

Trial registration: ClinicalTrials.gov Identifier: NCT01702805.

Figure 1.. Conceptual Model

If an association between red blood cell (RBC) transfusions and necrotizing enterocolitis (NEC) exists, the risk of NEC would be transiently high after a transfusion and low at baseline before a transfusion. The transient, high-risk, posttransfusion periods are considered hazard periods, and the low-risk, pretransfusion periods are considered control periods.

Figure 2.. Incidence Rates of Necrotizing Enterocolitis (NEC) and Distribution of Hazard Days for 5 Time Periods From Postnatal Days 10 to 60 by Randomization Group

A, Incidence rates of NEC per 1000 days within each 10-day interval, categorized by randomization group. B, Proportion of time within each 10-day interval that infants were exposed to red blood cell (RBC) transfusions, also categorized by randomization group. During the 20- to 29-day interval, infants in the higher hemoglobin transfusion threshold group spent an additional 0.6 days under exposure to RBC transfusions compared with infants in the lower hemoglobin transfusion threshold group and did not exhibit a higher incidence rate of NEC per 1000 days (1.8 vs 3.2; P = .09). During the 30- to 39-day interval, infants in the higher hemoglobin transfusion threshold group spent an additional 0.8 days under exposure to RBC transfusions and did not exhibit a higher incidence rate of NEC per 1000 days (2.6 vs 1.8; P = .38).