Sage Journals: Discover world-class research

Abstract

Red blood cell transfusion in non-bleeding critically ill patients is traditionally guided by fixed hemoglobin thresholds. Although restrictive strategies are widely recommended based on population-level randomized trials, hemoglobin concentration alone inadequately reflects the complex physiology of oxygen delivery, tissue perfusion, and cellular oxygen utilization in critical illness. Increasing evidence suggests frequent dissociation between hemoglobin increments and meaningful improvements in tissue oxygenation, particularly in the presence of microcirculatory dysfunction, impaired oxygen extraction, or mitochondrial failure. Moreover, recent trials in patients with limited cardiovascular reserve or acute myocardial ischemia challenge the universal safety of restrictive transfusion thresholds, emphasizing interindividual variability in oxygen supply–demand balance. In this mini-review and expert perspective, we synthesize physiological principles, landmark transfusion trials, and emerging monitoring modalities to propose a practical physiology-guided framework for red blood cell transfusion in non-bleeding critically ill patients. We review global, regional, and cellular markers of oxygen balance, including central venous oxygen saturation, oxygen extraction ratio, arterial–venous oxygen content difference, near-infrared spectroscopy–derived tissue oxygenation, microcirculatory flow indices, and mitochondrial oxygen tension, highlighting complementary roles and inherent limitations. We propose a pragmatic bedside approach in which transfusion is considered only after optimization of non-transfusion determinants of oxygen delivery and guided by integrated physiologic evidence of oxygen debt. This paradigm reframes transfusion as targeted therapy rather than numerical correction of anemia. Future research should combine outcome-driven randomized trials with large-scale data-driven and machine learning approaches to validate physiologic transfusion triggers and identify transfusion-responsive phenotypes in critical illness.

Keywords

red blood cell transfusion non-bleeding critically ill patients oxygen delivery central venous oxygen saturation oxygen extraction ratio microcirculation cellular oxygen utilization

Introduction

Red blood cell (RBC) transfusion remains one of the most frequently performed interventions in intensive care units (ICUs) worldwide.^1,2 For decades, transfusion decisions for non-bleeding critically ill patients have been guided primarily by hemoglobin (Hb) thresholds. This ‘number-based’ approach, while simple, overlooks the complexity of oxygen delivery (DO₂), tissue perfusion, and cellular oxygen utilization in critical illness.³ Despite advances in monitoring, Hb alone continues to serve as the dominant transfusion trigger in many ICUs, partly because essential physiologic monitoring tools, such as continuous cardiac output measurement, central venous oxygen saturation, or microcirculatory assessment, remain unavailable in resource-limited environments.⁴ As a result, clinicians often default to Hb thresholds even though they do not reliably reflect tissue-level oxygenation. Importantly, hemoglobin measurement remains objective, reproducible, and widely available across clinical settings, and represents a fundamental determinant of arterial oxygen content. Its clinical utility lies in its stability and accessibility, even if it does not, in isolation, capture dynamic changes in tissue oxygen balance.

Seminal trials demonstrated that a restrictive transfusion strategy in critically ill patients was at least as safe as a liberal one, thereby promoting blood conservation and reshaping transfusion practice.^5,6 However, these studies compared arbitrary hemoglobin thresholds rather than physiological determinants of tissue oxygenation. Recent advances in microcirculatory monitoring and tissue perfusion assessment have reignited debate over whether hemoglobin concentration alone is sufficient to guide transfusion decisions.^7,8 More importantly, recent randomized trials in patients with acute myocardial infarction and limited cardiovascular reserve have suggested that optimal hemoglobin thresholds may differ across specific high-risk subgroups rather than being universally applicable.^9,10 These findings do not represent a fundamental contradiction of hemoglobin-based strategies, but rather reflect progressive clinical stratification, indicating that uniform transfusion triggers may require contextual refinement. Similarly, randomized trials in acute brain injury populations further suggest that ischemia-sensitive tissues may warrant different hemoglobin targets.^11,12 Emerging secondary analyses further suggest that patients with pre-existing heart failure may also exhibit differential responses to transfusion strategies, although optimal Hb thresholds in this population remain incompletely defined.¹³ In this evolving framework, hemoglobin-based strategies are being refined rather than refuted, and physiology-guided assessment should be viewed as complementary, helping to identify patients in whom oxygen supply–demand imbalance persists despite apparently acceptable Hb concentrations. This conceptual tension between uniform thresholds and individualized physiology highlights the intrinsic limitation of a purely number-based transfusion paradigm and provides the biological and clinical rationale for a physiology-guided approach. Transfusion decisions based solely on numerical hemoglobin thresholds have not consistently translated into measurable improvements in physiological oxygen balance or clinical outcomes and may expose patients to unnecessary risks.^1,2 Accordingly, transfusion practice is currently positioned at the intersection of population-level hemoglobin thresholds and individualized physiologic refinement, rather than between mutually exclusive paradigms. In this critical mini-review and expert perspective, we synthesize contemporary physiological principles, key clinical trials, and emerging monitoring technologies, and propose a practical physiology-guided transfusion framework aimed at supporting individualized, precision-based decision making in non-bleeding critically ill patients. This article represents a focused conceptual synthesis informed by published literature and developed in accordance with established quality guidance for narrative reviews.

Hemoglobin-based transfusion thresholds: Evidence and limitations

In the mid-20th century, transfusion practice in surgical patients was commonly guided by a hemoglobin threshold of approximately 10 g/dL, despite lacking any physiological basis and relying largely on expert opinion and perioperative convenience.¹⁴ Although this liberal threshold was never universally codified as a formal rule, it became deeply embedded in routine perioperative practice. Over time, clinical experience and physiological understanding gradually eroded this dogma. Studies of oxygen delivery and consumption demonstrated that most healthy individuals and many critically ill patients can maintain adequate tissue oxygenation at much lower hemoglobin levels. Subsequent shifts toward lower transfusion thresholds, first around 8 g/dL and later near 7 g/dL, reflected a growing recognition that the critical threshold for oxygen delivery varies among individuals and depends on cardiac output, oxygen extraction ratio, and microcirculatory efficiency.¹⁴

Randomized trials subsequently reinforced the apparent safety of restrictive strategies at the population level. The TRICC trial demonstrated that a restrictive transfusion strategy targeting hemoglobin 7–9 g/dL resulted in similar or improved survival compared with a liberal strategy of 10–12 g/dL in critically ill patients.⁵ This landmark study led to global adoption of the restrictive approach. Subsequent investigations across specific clinical contexts reinforced the apparent safety of lower transfusion thresholds. In septic shock, the TRISS trial confirmed that maintaining hemoglobin around 7 g/dL was as safe as 9 g/dL.⁶ Among patients with cardiovascular disease, the FOCUS trial found no survival advantage with a liberal strategy after hip fracture surgery,¹⁵ while the REALITY trial in acute myocardial infarction and anemia reported no increase in major cardiovascular events with a restrictive threshold of 8 g/dL.¹⁶ However, the noninferiority margin applied in REALITY may have been sufficiently wide to permit clinically relevant myocardial injury to be classified as statistically acceptable, underscoring an intrinsic limitation of noninferiority designs in ischemia-sensitive populations.

In contrast, the MINT trial did not demonstrate a clear overall mortality benefit for either strategy but its prespecified subgroup analysis suggested a potential signal of harm associated with restrictive transfusion in patients with type 1 acute myocardial infarction.¹⁷ These findings raise concern that uniform restrictive hemoglobin thresholds may be physiologically inappropriate in the setting of acute coronary ischemia, where oxygen supply–demand mismatch is critically amplified. Finally, in a distinct population of critically ill oncologic patients, the TRICOP trial showed comparable outcomes between restrictive (7 g/dL) and liberal (9 g/dL) transfusion thresholds.¹⁸ Taken together, these trials suggest that restrictive transfusion thresholds are broadly safe at the population level. However, a major limitation shared by most transfusion trials is the reliance on hemoglobin concentration as the primary trigger for transfusion. This approach implicitly assumes physiological homogeneity and does not account for interindividual differences in oxygen transport dynamics. From a physiological perspective, oxygen delivery is defined by the relationships:

{DO}_{2} = cardiac output \times {CaO}_{2}

{CaO}_{2} = 1.34 \times Hb \times {SaO}_{2} + 0.003 \times {PaO}_{2}

This formulation highlights that oxygen delivery depends not only on hemoglobin but also on pulmonary oxygenation and cardiovascular performance. Hemoglobin therefore remains a important component of oxygen transport physiology; however, its interpretation requires integration with the dynamic determinants of flow and extraction, particularly in unstable critically ill patients. Hemoglobin-based thresholds alone cannot determine whether tissue hypoxia is present in an individual patient, nor whether anemia is the dominant contributor to impaired oxygen delivery. This conceptual limitation provides the rationale for a physiology-guided framework that integrates the major components of the oxygen transport pathway rather than relying on hemoglobin concentration alone.

When more is not better and less may be too little

Impaired tissue oxygenation in critically ill patients is rarely determined by hemoglobin concentration alone but instead reflects the interaction of multiple physiological mechanisms influencing oxygen transport and utilization. Clinical responses to red blood cell transfusion therefore vary substantially depending on which component of the oxygen pathway represents the dominant constraint at a given time. Physiologic markers measured at global, regional, and cellular levels have been investigated to better characterize this heterogeneity and to identify patients more likely to exhibit meaningful biological responses to transfusion (Tables 1 and 2).

Table 1.

Physiologic markers and biologic responses to red blood cell transfusion.

Physiologic level	Marker	What it represents	Typical physiologic signal after transfusion	Key message
Global oxygen balance	ScvO₂^19–21	Balance between systemic DO₂ and VO₂	Increases only in patients with impaired baseline ScvO₂	Not all patients increase ScvO₂ after transfusion
Systemic extraction	O₂ER²²	Whole-body oxygen extraction reserve	Decreases when DO₂ dependency is present	Only elevated O₂ER identifies transfusion-responsive patients
Systemic extraction magnitude	A–V O₂diff²³	Amount of oxygen extracted from blood	May decrease if transfusion increases effective DO₂	Reflects supply dependency but influenced by cardiac output
Regional microcirculation	StO₂ (NIRS)^24,25	Regional tissue oxygenation	Global StO₂ often unchanged; responders show improved reperfusion slope	Hb increase does not guarantee microvascular recruitment
Microvascular convection	MFI, PPV²⁶	Capillary red cell flow and distribution	Improves only in patients with baseline microvascular impairment	Transfusion may worsen flow if microcirculation is normal
Cellular oxygen utilization	mitoPO₂, mitoVO₂²⁷	Mitochondrial oxygen tension and consumption	Often unchanged or discordant from Hb and ScvO₂	Cellular hypoxia may persist despite normal systemic oxygenation

A–V O₂diff, arterial–venous oxygen content difference; DO₂, oxygen delivery; Hb, hemoglobin; MFI, microvascular flow index; mitoPO₂, mitochondrial oxygen tension; mitoVO₂, mitochondrial oxygen consumption; NIRS, near-infrared spectroscopy; O₂ER, oxygen extraction ratio; PPV, proportion of perfused vessels; ScvO₂, central venous oxygen saturation; StO₂, tissue oxygen saturation; VO₂, oxygen consumption.

Table 2.

Advantages and limitations of physiologic transfusion markers.

Marker	Main advantage	Major limitation	Principal pitfall
ScvO₂^19–21	Simple global screening of DO₂/VO₂ mismatch	Misses regional ischemia; confounded by impaired VO₂	False reassurance in sepsis
O₂ER²²	Best indicator of real supply dependency	Invalid in hypermetabolic states	Overestimates need during fever or agitation
A–V O₂diff²³	Quantifies extraction magnitude	Strongly influenced by cardiac output	Misinterpreted in low-flow shock
NIRS (StO₂)^24,25	Noninvasive continuous microvascular trend	No validated absolute thresholds	Interpreting absolute values as global hypoxia
MFI, PPV²⁶	Direct visualization of capillary flow	Operator-dependent; organ-specific	Assuming sublingual reflects whole-body perfusion
mitoPO₂, mitoVO₂²⁷	Closest to true cellular oxygen use	Experimental and site-specific	Overinterpretation without outcome validation

A–V O₂diff, arterial–venous oxygen content difference; DO₂, oxygen delivery; MFI, microvascular flow index; mitoPO₂, mitochondrial oxygen tension; mitoVO₂, mitochondrial oxygen consumption; NIRS, near-infrared spectroscopy; O₂ER, oxygen extraction ratio; PPV, proportion of perfused vessels; ScvO₂, central venous oxygen saturation; StO₂, tissue oxygen saturation; VO₂, oxygen consumption.

Pulmonary oxygenation

Adequate arterial oxygenation remains a prerequisite for effective oxygen transport. In conditions such as acute respiratory distress syndrome (ARDS) or severe ventilation–perfusion mismatch, reduced PaO₂ and SaO₂ limit arterial oxygen content irrespective of hemoglobin concentration.²⁸ In this setting, increasing hemoglobin cannot compensate for impaired pulmonary oxygen transfer. Failure to differentiate hypoxaemia due to impaired pulmonary oxygenation from anaemic hypoxia risks attributing tissue hypoxia to anemia when the dominant limitation lies at the level of oxygen diffusion and gas exchange.

Circulatory flow and global oxygen balance

Cardiac output is a major determinant of oxygen delivery. In low-flow states such as cardiogenic shock, reduced forward flow constrains oxygen delivery irrespective of hemoglobin concentration. Conversely, patients with preserved hemodynamics may compensate for moderate anemia through increases in cardiac output.

Importantly, tissue oxygen consumption (VO₂) remains independent of oxygen delivery (DO₂) across a broad physiological range because of compensatory increases in oxygen extraction (O₂ER). Only when a critical DO₂ threshold is reached does supply dependency occur, at which point VO₂ begins to decline.²⁹ This physiological principle explains why reductions in hemoglobin do not automatically mandate transfusion, effective oxygen extraction may preserve tissue oxygenation until extraction reserve is exhausted. Moreover, this compensatory reserve depends on intact vasodilatory and autoregulatory mechanisms. When vascular adaptation is impaired, as in type 1 acute myocardial infarction with fixed coronary stenosis or in acute brain injury with disrupted cerebrovascular autoregulation, the ability to augment blood flow in response to anemia is constrained.^9,12 In such settings, ischemia-sensitive tissues may become supply dependent at higher hemoglobin levels, and overly restrictive strategies may therefore be disadvantageous in selected patients, supporting contextual refinement rather than uniform application of transfusion thresholds. At the systemic level, global shock states further illustrate how distinct hemodynamic patterns modify oxygen supply–demand relationships. The pathophysiology differs between shock phenotypes, cardiogenic shock is characterized by reduced flow, whereas distributive shock such as sepsis involves microvascular heterogeneity and impaired oxygen extraction despite preserved or elevated cardiac output.^30,31 In the latter scenario, increasing hemoglobin may fail to improve oxygen utilization if extraction mechanisms are dysfunctional.

Central venous oxygen saturation (ScvO₂), oxygen extraction (O₂ER), and arterial–venous oxygen content difference (A–V O₂diff) have been proposed to identify supply dependency.^19,22,23 However, observational data demonstrate that only a minority of patients exhibit a meaningful post-transfusion increase in ScvO₂.^19–21 Moreover, normal or elevated ScvO₂ does not exclude tissue hypoxia in distributive shock due to impaired extraction. These considerations emphasize that factors other than anemia frequently contribute to impaired oxygen delivery.

Microcirculation and regional perfusion

Even when global oxygen delivery appears adequate, heterogeneous perfusion at the microcirculatory level may prevent effective oxygen distribution. Microvascular dysfunction is a hallmark of sepsis, where endothelial injury and glycocalyx disruption contribute to flow maldistribution.²⁸

Several investigations have shown that red blood cell transfusion may increase arterial oxygen content without improving microvascular flow, and in some cases may worsen flow distribution.^7,26 Microvascular flow index (MFI) and proportion of perfused vessels (PPV) improve primarily in patients with baseline impairment, whereas patients with preserved microcirculation may experience no benefit or even deterioration.

Importantly, all currently available tools for detecting tissue hypoxia provide regional rather than global information, and organ-specific differences further limit extrapolation from local observations to whole-body oxygenation. No single physiologic parameter captures the entirety of the oxygen transport cascade; each reflects a specific level and must therefore be interpreted within its biological and clinical context.

Cellular oxygen utilization

At the cellular level, mitochondrial oxygen availability represents the final step of the oxygen transport cascade. A recent multicenter observational study evaluated mitochondrial oxygen tension (mitoPO₂) and mitochondrial oxygen consumption (mitoVO₂) in critically ill patients receiving red blood cell transfusion.²⁷ The investigators found that mitoPO₂ and mitoVO₂ did not change consistently before and after transfusion, and neither parameter was strongly associated with hemoglobin concentration or conventional markers of tissue perfusion and oxygenation. mitoPO₂ reflects the partial pressure of oxygen within functioning mitochondria and should be interpreted as an indicator of local cellular oxygen availability, not a direct measure of mitochondrial enzymatic function. In this study, many patients had normal baseline mitoPO₂ values, suggesting that cellular oxygen tension was not critically low at the time of transfusion; in such settings, increases in hemoglobin concentration would not be expected to improve mitochondrial oxygen tension or consumption. These findings suggest that red blood cell transfusion does not consistently increase mitochondrial oxygen tension in unselected critically ill patients with anemia and highlight the importance of identifying true supply dependency before assuming that augmenting hemoglobin will improve cellular oxygen availability.

Clinical heterogeneity and risk–benefit balance

Clinical responses to transfusion are heterogeneous. Some patients exhibit measurable physiologic improvement following transfusion, whereas others show little or no change in oxygen transport variables, reflecting heterogeneity across multiple levels of the oxygen transport cascade.^21–23 This heterogeneity is mirrored in real-world practice patterns, which further illustrate the divergence between physiologic concepts and bedside transfusion decision-making. The TRACE survey reported that tachycardia and hypotension were among the most frequently cited physiological triggers for transfusion, whereas other physiological parameters were reported less consistently.³² Similarly, the InPUT study showed that triggers such as ScvO₂ <65% and electrocardiographic changes accounted for only a small minority of transfusion episodes.¹ These data suggest that transfusion decisions in contemporary practice often rely on hemodynamic surrogates rather than integrated physiologic assessment of oxygen supply dependency. Serum lactate and cardiac biomarkers have also been explored as adjunctive indicators of impaired oxygen balance.^33,34 Lactate may reflect global hypoperfusion but is multifactorial and influenced by adrenergic stimulation, mitochondrial dysfunction, and impaired clearance. High-sensitivity troponin may signal myocardial stress in anemic patients. However, neither biomarker has been validated as a primary transfusion trigger in outcome-based randomized trials.³³

Red blood cell transfusion is associated with potential adverse effects, including transfusion-associated circulatory overload (TACO), transfusion-related acute lung injury (TRALI), immunomodulation, and infection.^35–38 Although uncommon, these risks reinforce the importance of establishing physiological justification before transfusion exposure.

How to implement physiology-guided transfusion

Building on physiologic and clinical evidence, a practical framework can help clinicians apply physiology-guided transfusion at the bedside. Before considering RBC administration, clinicians must ensure that oxygen delivery is maximized through non-transfusion means, including adequate preload, appropriate mean arterial pressure, optimized ventilation and oxygenation, and correction of metabolic and acid-base derangements.³ This step is essential, as anemia is only one of several potentially reversible determinants of impaired oxygen delivery. If objective evidence of tissue hypoxia persists after these measures, transfusion may then be physiologically justified.

Although validated outcome-based thresholds are currently lacking, existing physiologic studies provide pragmatic guidance. These include low central or mixed venous oxygen saturation (ScvO₂ below 65 to 70 percent),^19–21 elevated oxygen extraction ratio (O₂ER above 30 percent),²² increased arterial-to-venous oxygen content difference above 3.5 to 4.0 mL per dL,²³ reduced tissue oxygen saturation measured by near-infrared spectroscopy (StO₂ below 60 percent),^24,25 and impaired microcirculatory indices such as a microvascular flow index below 2.8 or proportion of perfused vessels below 88 percent (Table 3).²⁶ These parameters reflect distinct but complementary levels of the oxygen transport pathway, including global delivery, regional perfusion, and cellular utilization. However, it must be explicitly acknowledged that none of these cut-offs have been validated in randomized trials using hard clinical outcomes such as mortality or organ failure. These suggested limits, derived primarily from physiologic and observational studies, should therefore be interpreted as pragmatic guides rather than definitive transfusion triggers, and always integrated dynamically within the full clinical context.

Table 3.

Pragmatic physiological guides for bedside transfusion decision-making.

Marker	Assessment scope	Pragmatic trigger	When it is most useful
ScvO₂^19–21	Global DO₂/VO₂ balance	<65–70%	Septic shock, low cardiac reserve
O₂ER²²	Global extraction reserve	>30%	Suspected supply dependency
A–V O₂diff²³	Systemic extraction intensity	>3.5–4.0 mL/dL	Low-flow states, high metabolic demand
StO₂ (NIRS)²⁴	Regional tissue perfusion	<60% or blunted reperfusion	Peripheral hypoperfusion, shock
MFI, PPV²⁶	Regional microvascular convection	MFI <2.8 or PPV <88%	Suspected microcirculatory failure

A–V O₂diff, arterial–venous oxygen content difference; DO₂, oxygen delivery; MFI, microvascular flow index; NIRS, near-infrared spectroscopy; O₂ER, oxygen extraction ratio; PPV, proportion of perfused vessels; ScvO₂, central venous oxygen saturation; StO₂, tissue oxygen saturation; VO₂, oxygen consumption.

When transfusion is considered physiologically appropriate, one unit should be administered at a time, followed by immediate reassessment of physiologic response rather than hemoglobin concentration alone.³⁹ The therapeutic target is restoration of adequate oxygen delivery and utilization, not numerical normalization of hemoglobin. Whenever feasible, the use of fresher and leukoreduced blood products may mitigate some adverse effects.⁴⁰ This individualized, response-based strategy is fully aligned with modern patient blood management principles, which emphasize precision use of blood products, early diagnosis and treatment of anemia, minimization of iatrogenic blood loss, and preferential use of pharmacologic or procedural alternatives to transfusion whenever possible.⁴¹

From a research perspective, future investigations should prioritize stratification based on physiologic variables in combination with hemoglobin thresholds, rather than relying on hemoglobin concentration alone. This represents a fundamental conceptual shift from population based thresholds toward phenotype specific oxygen delivery targets. Advances in bedside microcirculatory imaging and mitochondrial function monitoring may ultimately enable near–real-time assessment of oxygen delivery and cellular utilization.⁴² Importantly, the ongoing O₂ER guided randomized trials (NCT06102590) represent the most critical current effort to establish causality between physiology-based transfusion triggers and meaningful clinical outcomes.⁴³ Integrating continuous physiologic data with machine learning approaches could further identify transfusion responsive phenotypes and refine individualized decision making, thereby bridging the prevailing gap between mechanistic physiology and outcome driven clinical practice.⁴⁴

Conclusion

In non-bleeding critically ill patients, transfusion practice is evolving toward a more integrated framework that combines hemoglobin thresholds with physiologic assessment. Hemoglobin remains a fundamental determinant of arterial oxygen content, but its interpretation benefits from contextualization within markers of oxygen delivery, extraction, and microcirculatory function. Integrating parameters such as central venous oxygen saturation, oxygen extraction ratio, and arterial–venous oxygen content difference may help identify patients with true oxygen supply–demand imbalance. Future research should validate physiologic triggers through outcome driven randomized trials and complementary data driven analytic approaches supported by large language models, while incorporating real time physiologic monitoring to advance precision transfusion and improve patient outcomes.

Footnotes

Acknowledgements

During the preparation of this manuscript, the authors used an artificial intelligence–assisted language tool to enhance readability and improve writing quality, including spelling and grammatical accuracy. All outputs were critically reviewed and edited by the authors, who take full responsibility for the integrity and accuracy of the content.

ORCID iD

Thanh Luan Nguyen

Ethical considerations

This article is a narrative mini-review and expert perspective based exclusively on previously published literature. It does not involve any new studies with human participants, human data, or human tissue performed by the authors. Therefore, ethical approval from an Ethics Committee or Institutional Review Board was not required.

Consent to participate

This manuscript does not report primary research involving human participants.

Consent for publication

This manuscript does not include any individual-level data, images, videos, or other potentially identifiable personal information.

Author contributions

Thanh Luan Nguyen conceptualized the study, developed the overall physiology-guided framework, coordinated the project, led the literature synthesis, and drafted the initial version of the manuscript. Quang Dai Le and Van Phieu Duong contributed to study supervision, refinement of the conceptual framework, and critical revision of the manuscript for important intellectual content. Phan Ngoc An Huynh, Phuc Tuong Pham, and Thi Bao Yen Nguyen contributed to the development and interpretation of the physiological rationale, literature analysis, and substantive manuscript revision. Van Hoang Nam Ho and Hoang Ngoc Thao Duong contributed to validation of the conceptual framework, organization tables, and revision of the manuscript for clarity and coherence. Cong Dang Tran and Trang Nguyen Hoai Dinh contributed to literature screening, extraction of relevant evidence from published studies, and drafting and editing of supporting sections of the manuscript. All authors reviewed and approved the final version of the manuscript and agree to be accountable for all aspects of the work.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data sharing is not applicable to this article as no new datasets were generated or analyzed during the current study. All data discussed in this manuscript are derived from previously published studies that are cited in the reference list.*

Appendix

References

Raasveld

de Bruin

Reuland

, et al. Red Blood Cell Transfusion in the Intensive Care Unit. Jama 2023; 330: 1852–1861. https://doi.org/10.1001/jama.2023.20737

Vincent

Jaschinski

Wittebole

, et al. Worldwide audit of blood transfusion practice in critically ill patients. Critical care (London, England) 2018; 22: 102. https://doi.org/10.1186/s13054-018-2018-9

Hébert

Van der Linden

Biro

, et al. Physiologic aspects of anemia. Critical care clinics 2004; 20: 187–212. https://doi.org/10.1016/j.ccc.2004.01.001

Adhikari

Fowler

Bhagwanjee

, et al. Critical care and the global burden of critical illness in adults. Lancet (London, England) 2010; 376: 1339–1346. https://doi.org/10.1016/s0140-6736(10)60446-1

Hébert

Wells

Blajchman

, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. The New England journal of medicine 1999; 340: 409–417. https://doi.org/10.1056/nejm199902113400601

Holst

Haase

Wetterslev

, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. The New England journal of medicine 2014; 371: 1381–1391. https://doi.org/10.1056/NEJMoa1406617

Dilken

Ergin

Ince

. Assessment of sublingual microcirculation in critically ill patients: consensus and debate. Annals of translational medicine 2020; 8: 793. https://doi.org/10.21037/atm.2020.03.222

Donati

Damiani

Domizi

, et al. Near-infrared spectroscopy for assessing tissue oxygenation and microvascular reactivity in critically ill patients: a prospective observational study. Critical Care 2016; 20: 311. https://doi.org/10.1186/s13054-016-1500-5

DeFilippis

Abbott

Herbert

, et al. Restrictive Versus Liberal Transfusion in Patients With Type 1 or Type 2 Myocardial Infarction: A Prespecified Analysis of the MINT Trial. Circulation 2024; 150: 1826–1836. https://doi.org/10.1161/circulationaha.124.071208

10.

Ducrocq

Cachanado

Simon

, et al. Restrictive vs Liberal Blood Transfusions for Patients With Acute Myocardial Infarction and Anemia by Heart Failure Status: An RCT Subgroup Analysis. Canadian Journal of Cardiology 2024; 40: 1705–1714. https://doi.org/10.1016/j.cjca.2024.02.013

11.

Turgeon

Fergusson

Clayton

, et al. Liberal or Restrictive Transfusion Strategy in Patients with Traumatic Brain Injury. The New England journal of medicine 2024; 391: 722–735. https://doi.org/10.1056/NEJMoa2404360

12.

Taccone

Rynkowski

Møller

, et al. Restrictive vs Liberal Transfusion Strategy in Patients With Acute Brain Injury: The TRAIN Randomized Clinical Trial. Jama 2024; 332: 1623–1633. https://doi.org/10.1001/jama.2024.20424

13.

Prochaska

Portela

Brooks

, et al. Transfusion Strategy Effect on Quality of Life in Patients With Myocardial Infarction and Anemia: A Secondary Analysis of the MINT Randomized Clinical Trial. JAMA internal medicine 2025; 185: 947–954. https://doi.org/10.1001/jamainternmed.2025.0654

14.

Greenburg

. A physiologic basis for red blood cell transfusion decisions. American journal of surgery 1995; 170: 44s–48s. https://doi.org/10.1016/s0002-9610(99)80058-0

15.

Carson

Terrin

Noveck

, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. The New England journal of medicine 2011; 365: 2453–2462. https://doi.org/10.1056/NEJMoa1012452

16.

Ducrocq

Gonzalez-Juanatey

Puymirat

, et al. Effect of a Restrictive vs Liberal Blood Transfusion Strategy on Major Cardiovascular Events Among Patients With Acute Myocardial Infarction and Anemia: The REALITY Randomized Clinical Trial. Jama 2021; 325: 552–560. https://doi.org/10.1001/jama.2021.0135

17.

Carson

Brooks

Hébert

, et al. Restrictive or Liberal Transfusion Strategy in Myocardial Infarction and Anemia. The New England journal of medicine 2023; 389: 2446–2456. https://doi.org/10.1056/NEJMoa2307983

18.

Bergamin

Almeida

Landoni

, et al. Liberal Versus Restrictive Transfusion Strategy in Critically Ill Oncologic Patients: The Transfusion Requirements in Critically Ill Oncologic Patients Randomized Controlled Trial. Critical care medicine 2017; 45: 766–773. https://doi.org/10.1097/ccm.0000000000002283

19.

Themelin

Biston

Massart

, et al. Effects of red blood cell transfusion on global oxygenation in anemic critically ill patients. Transfusion 2021; 61: 1071–1079. https://doi.org/10.1111/trf.16284

20.

Fischer

Guinot

Debroczi

, et al. Individualised or liberal red blood cell transfusion after cardiac surgery: a randomised controlled trial. British journal of anaesthesia 2022; 128: 37–44. https://doi.org/10.1016/j.bja.2021.09.037

21.

Zeroual

Samarani

Gallais

, et al. ScvO(2) changes after red-blood-cell transfusion for anaemia in cardiothoracic and vascular ICU patients: an observational study. Vox sanguinis 2018; 113: 136–142. https://doi.org/10.1111/vox.12610

22.

Tüzen

Aksun

Şencan

, et al. Assessment of oxygen extraction rate changes following red blood cell transfusion in the intensive care unit: A prospective observational noninterventional study. Transfusion 2025; 65: 863–875. https://doi.org/10.1111/trf.18164

23.

Fogagnolo

Taccone

Vincent

, et al. Using arterial-venous oxygen difference to guide red blood cell transfusion strategy. Critical Care 2020; 24: 160. https://doi.org/10.1186/s13054-020-2827-5

24.

Creteur

Neves

Vincent

J-L

. Near-infrared spectroscopy technique to evaluate the effects of red blood cell transfusion on tissue oxygenation. Critical Care 2009; 13: S11. https://doi.org/10.1186/cc8009

25.

Leal-Noval

Arellano-Orden

Muñoz-Gómez

, et al. Red Blood Cell Transfusion Guided by Near Infrared Spectroscopy in Neurocritically Ill Patients with Moderate or Severe Anemia: A Randomized, Controlled Trial. Journal of neurotrauma 2017; 34: 2553–2559. https://doi.org/10.1089/neu.2016.4794

26.

Scheuzger

Zehnder

Meier

, et al. Sublingual microcirculation does not reflect red blood cell transfusion thresholds in the intensive care unit-a prospective observational study in the intensive care unit. Critical care (London, England) 2020; 24: 18. https://doi.org/10.1186/s13054-020-2728-7

27.

Baysan

Hilderink

van Manen

, et al. Mitochondrial oxygen tension in critically ill patients receiving red blood cell transfusions: a multicenter observational cohort study. Intensive care medicine experimental 2024; 12: 61. https://doi.org/10.1186/s40635-024-00646-3

28.

Morisaki

Sibbald

. Tissue oxygen delivery and the microcirculation. Critical care clinics 2004; 20: 213–223. https://doi.org/10.1016/j.ccc.2003.12.003

29.

Molnar

Nemeth

. SvO2/ScvO2. In: Pinsky

Teboul

J-L

Vincent

J-L

(eds) Hemodynamic Monitoring. Springer International Publishing, 2019, pp. 157–171.

30.

Merdji

Levy

Jung

, et al. Microcirculatory dysfunction in cardiogenic shock. Annals of intensive care 2023; 13: 38. https://doi.org/10.1186/s13613-023-01130-z

31.

De Backer

Ricottilli

Ospina-Tascón

. Septic shock: a microcirculation disease. Current opinion in anaesthesiology 2021; 34: 85–91. https://doi.org/10.1097/aco.0000000000000957

32.

de Bruin

Scheeren

TWL

Bakker

, et al. Transfusion practice in the non-bleeding critically ill: an international online survey-the TRACE survey. Critical care (London, England) 2019; 23: 309. https://doi.org/10.1186/s13054-019-2591-6

33.

Arynov

Kaidarova

Kabon

. Alternative blood transfusion triggers: a narrative review. BMC Anesthesiology 2024; 24: 71. https://doi.org/10.1186/s12871-024-02447-3

34.

Czempik

Gierczak

Wilczek

, et al. The Impact of Red Blood Cell Transfusion on Blood Lactate in Non-Bleeding Critically Ill Patients-A Retrospective Cohort Study. Journal of clinical medicine 2022; 11: https://doi.org/10.3390/jcm11041037

35.

Nilsson

Bentzer

Andersson

, et al. Mortality and morbidity of low-grade red blood cell transfusions in septic patients: a propensity score-matched observational study of a liberal transfusion strategy. Annals of intensive care 2020; 10: 111. https://doi.org/10.1186/s13613-020-00727-y

36.

Rohde

Dimcheff

Blumberg

, et al. Health care-associated infection after red blood cell transfusion: a systematic review and meta-analysis. Jama 2014; 311: 1317–1326. https://doi.org/10.1001/jama.2014.2726

37.

Semple

Rebetz

Kapur

. Transfusion-associated circulatory overload and transfusion-related acute lung injury. Blood 2019; 133: 1840–1853. https://doi.org/10.1182/blood-2018-10-860809

38.

Youssef

Spitalnik

. Transfusion-related immunomodulation: a reappraisal. Current opinion in hematology 2017; 24: 551–557. https://doi.org/10.1097/moh.0000000000000376

39.

Vlaar

Oczkowski

de Bruin

, et al. Transfusion strategies in non-bleeding critically ill adults: a clinical practice guideline from the European Society of Intensive Care Medicine. Intensive care medicine 2020; 46: 673–696. https://doi.org/10.1007/s00134-019-05884-8

40.

Hod

Spitalnik

. Stored red blood cell transfusions: Iron, inflammation, immunity, and infection. Transfusion clinique et biologique : journal de la Societe francaise de transfusion sanguine 2012; 19: 84–89. https://doi.org/10.1016/j.tracli.2012.04.001

41.

Spahn

Goodnough

. Alternatives to blood transfusion. Lancet (London, England) 2013; 381: 1855–1865. https://doi.org/10.1016/s0140-6736(13)60808-9

42.

Singer

. The role of mitochondrial dysfunction in sepsis-induced multi-organ failure. Virulence 2014; 5: 66–72. https://doi.org/10.4161/viru.26907

43.

Fogagnolo

Azzolina

Taccone

, et al. Oxygen extraction-guided transfusion strategy in critically ill patients: study protocol for a randomised, open-labelled, controlled trial. BMJ open 2024; 14: e089910. https://doi.org/10.1136/bmjopen-2024-089910

44.

Maynard

Farrington

Alimam

, et al. Machine learning in transfusion medicine: A scoping review. Transfusion 2024; 64: 162–184. https://doi.org/10.1111/trf.17582

Beyond hemoglobin thresholds: A physiology-guided framework for red blood cell transfusion in non-bleeding critically ill patients

Abstract

Keywords

Introduction

Hemoglobin-based transfusion thresholds: Evidence and limitations

When more is not better and less may be too little

Pulmonary oxygenation

Circulatory flow and global oxygen balance

Microcirculation and regional perfusion

Cellular oxygen utilization

Clinical heterogeneity and risk–benefit balance

How to implement physiology-guided transfusion

Conclusion

Footnotes

Acknowledgements

ORCID iD

Ethical considerations

Consent to participate

Consent for publication

Author contributions

Funding

Declaration of conflicting interests

Data Availability Statement

Appendix

References