In pediatric cardiac surgery, the priming volume required for cardiopulmonary bypass (CPB) circuits typically ranges from 300 to 500 mL, often exceeding 50% of the total blood volume in infants and young children—markedly higher than the approximate 10% observed in adult surgeries (1, 2). This substantial difference renders red blood cell (RBC) transfusion an indispensable component in the initiation of CPB (3, 4). Given the limited circulating blood volume in infants (∼80 mL/kg) and their high oxygen consumption rate (6–8 mL/kg/min), both the quality and volume of transfused RBCs are critical determinants for maintaining perioperative oxygen delivery and organ perfusion stability (5, 6).

However, during storage, RBCs undergo progressive “storage lesions,” including decreased membrane stability, depletion of ATP and 2,3-diphosphoglycerate (2,3-DPG), and the release of microparticles and pro-inflammatory cytokines (7–9). These pathological changes may compromise oxygen delivery efficiency and exacerbate intra- and postoperative complications when transfused into pediatric patients (10–12). Although current FDA and European guidelines permit the use of RBCs stored for up to 35–42 days, whether all storage durations within this window are equally safe remains a topic of debate (13, 14).

Several large randomized controlled trials (RCTs), such as RECESS (15) and ABLE (16), have demonstrated that RBC storage duration does not significantly impact clinical outcomes in adults. Similarly, the ABC PICU trial found no significant difference in the incidence of multiple organ dysfunction syndrome (MODS) between fresh (≤7 days) and standard-issue RBCs in critically ill pediatric patients (17). However, caution should be exercised when extrapolating these findings to pediatric cardiac surgical populations due to key physiological differences: (1) the RBC volume used for CPB priming greatly exceeds that of conventional transfusions (18); (2) children with congenital heart disease often have immature circulatory and immune systems (19); and (3) developing organs—particularly the brain—are more susceptible to ischemia-reperfusion injury (20).

In recent years, increasing attention has been paid to the relationship between RBC storage duration and postoperative outcomes in pediatric cardiac surgery. Some observational studies have reported associations between longer storage durations and increased risks of postoperative infections (21, 22), pulmonary and renal complications (22), prolonged mechanical ventilation, and enhanced inflammatory responses (23). However, other studies have found no significant associations between storage duration and major outcomes, resulting in substantial variability in clinical practice.

Although the 2016 AABB guidelines state that RBCs of any storage duration within the permissible shelf life are acceptable for neonatal and pediatric transfusion (24), many institutions still preferentially use fresher RBCs for CPB priming—decisions often based on theoretical concerns or institutional habits rather than robust evidence (23).

Given the absence of specific consensus guidelines regarding RBC storage thresholds in pediatric CPB settings, and the lack of clear recommendations in current transfusion protocols, clinical decisions largely depend on institutional experience. This underscores the urgent need to systematically review the available evidence to determine whether RBC storage duration significantly affects outcomes in pediatric cardiac surgery.

Therefore, this study aims to perform a systematic review and meta-analysis to evaluate the impact of RBC storage duration on major outcomes (including mortality, infection, and MODS) and secondary outcomes [such as mechanical ventilation duration, Intensive Care Unit (ICU) length of stay, and intraoperative lactate levels] in children undergoing cardiac surgery with CPB. The goal is to establish a more evidence-based transfusion window for this vulnerable population and to inform optimization of perioperative blood management strategies.

In this systematic review and meta-analysis, we synthesized evidence from the past two decades regarding the impact of RBC storage duration on postoperative outcomes in pediatric cardiac surgery. Across all major clinical endpoints—including mortality, infection/sepsis, respiratory complications, and MODS—no statistically significant differences were found between fresher and older RBCs. Secondary outcomes, including mechanical ventilation duration and intraoperative lactate levels, similarly showed no meaningful differences. Leave-one-out sensitivity analyses applied to all primary and secondary outcomes demonstrated that no single study materially altered the pooled estimates, supporting the overall robustness of the findings across study designs, outcome types, and analytical approaches.

A key challenge in interpreting the evidence is the substantial variability in definitions of “fresh” RBCs, which ranged from ≤4 to ≤15 days across included studies. Such inconsistency inevitably introduced clinical heterogeneity and limited direct comparability. To address this, we implemented a threshold-standardized subgroup analysis using ≤7 days as a unified definition, selected because it represented a common definitional midpoint and allowed maximal study inclusion without compromising analytical comparability. Harmonizing storage thresholds improved interpretability and confirmed that variations in original definitions did not materially influence effect estimates.

One notable finding was a modest reduction in ICU length of stay (–1.08 days) in the fresher RBC group. However, this result arose predominantly from observational studies and exhibited substantial heterogeneity (I² = 92.8%). To strengthen interpretability under such conditions, we additionally applied a REML random-effects model. The REML-derived estimates were highly consistent with those from the DerSimonian–Laird model, indicating that the observed direction and significance of effect were robust despite heterogeneity. Sensitivity analyses excluding the single study using an extremely strict ≤4-day definition [Ranucci 2009 (22)] further confirmed that threshold outliers did not exert undue influence.

Although stored RBCs undergo progressive “storage lesions”—including membrane instability, ATP and 2,3-DPG depletion, and release of pro-inflammatory mediators (7–12)—our findings did not identify a clinical advantage of fresher blood in pediatric CPB settings. Several mechanistic explanations may account for these neutral results. First, modern blood banking practices—such as universal leukoreduction, improved additive solutions, and optimized storage conditions—have substantially attenuated storage lesions compared with earlier decades, narrowing physiological differences between “fresh” and “older” units (11, 37). Second, transfusion practices often introduce confounding by indication: neonates or high-risk patients, who inherently experience poorer postoperative outcomes, are more likely to receive fresher units, biasing comparisons toward equivalence (38, 39). Third, the high-flow, hemodilutional environment of pediatric CPB rapidly dilutes transfused RBCs within the priming volume, reducing the effective load of storage-derived byproducts and mitigating their physiological impact (40, 41). Collectively, these considerations offer a biologically plausible explanation for the largely neutral findings observed.

While evidence from major adult RCTs such as RECESS and ABLE provides contextual reference (15, 16), direct extrapolation to pediatric CPB is limited by fundamental physiological differences. Neonates and infants possess immature organ systems, higher metabolic rates, and heightened inflammatory responses to CPB. These distinctions underscore the need to interpret adult data cautiously and highlight the importance of pediatric-focused evidence. Interestingly, two recent adult meta-analyses reported a potential increase in nosocomial infection risk with fresher RBC transfusion (42, 43), whereas observational studies in pediatric cardiac surgery have associated longer storage with increased infection risk (21, 22). Our findings fall between these perspectives, suggesting that storage duration alone may not be a dominant driver of infection risk in pediatric CPB, where multiple patient and management-related factors likely play more influential roles.

The neutral findings observed in this review align with current transfusion guidelines, which do not mandate preferential use of fresh RBCs in pediatric cardiac surgery (24). For institutions facing logistical limitations in blood inventory management, these results provide practical reassurance that using RBCs within approved storage periods for CPB priming is safe and feasible.

Although our meta-analysis integrates the most comprehensive available evidence, several limitations should be acknowledged. Most included studies were observational, limiting causal inference. Definitions of postoperative complications—especially respiratory complications—varied substantially across studies, potentially introducing misclassification bias. To address this, we categorized nonspecific descriptions (e.g., “pulmonary complications,” “respiratory events”) as mild and conducted a sensitivity analysis excluding such studies; notably, the effect estimate for severe respiratory complications remained unchanged, indicating that variability in outcome definitions did not drive the results. Furthermore, substantial variation existed across studies in CPB priming strategies, transfusion thresholds, postoperative management, and age distributions. Although we considered performing subgroup analyses based on age strata, CPB priming volume, or study era, the required stratified data were insufficiently reported to permit reliable analysis. These clinical inconsistencies likely contributed to residual heterogeneity and highlight the need for standardized reporting in future research.

To enhance methodological transparency, raw 2 × 2 event counts for all dichotomous outcomes were provided in Supplementary Table S1, as several original studies reported summary statistics without complete event distributions. Publication bias could not be formally assessed because none of the outcomes included the ≥10 studies required for reliable funnel plots or Egger's tests, and thus small-study effects cannot be excluded. Finally, although a meta-regression examining storage duration as a continuous moderator was considered, no outcome met the methodological minimum of ≥10 studies, and such analyses would likely produce unstable estimates.

In summary, despite extreme definitional variability and clinical heterogeneity across studies, threshold-standardized and sensitivity analyses demonstrated that effect estimates were generally stable. These findings suggest that RBC storage duration may exert less clinical influence in pediatric cardiac surgery than previously assumed. Future research—particularly multicenter RCTs with standardized reporting—remains essential to determine whether specific pediatric subgroups may derive benefit from fresher RBC transfusion.

This systematic review and meta-analysis synthesized current evidence and found no significant differences in major or secondary perioperative outcomes between fresher and longer-stored RBCs when used within the approved storage duration in pediatric cardiac surgery. These findings support the current practice guidelines that do not mandate the preferential use of fresher RBCs and provide valuable implications for both clinical decision-making and blood inventory management. Nevertheless, high-quality randomized controlled trials are still needed to further validate these conclusions and to develop more refined, evidence-based transfusion strategies tailored for pediatric populations.