Towards safer and more effective hemotherapy: addressing critical unmet needs in restrictive and liberal blood transfusion strategies

Introduction

Blood transfusion continues to be one of the most significant therapeutic measures in contemporary medicine, supporting the treatment of acute bleeding, severe anemia, major surgeries, trauma, obstetric crises, blood disorders, and critical conditions. From the initial successful human-to-human blood transfusions and the later development of blood banking systems, hemotherapy has progressed into a well-regulated and evidence-based field that has greatly decreased mortality and enhanced clinical results globally. Annually, over 118 million blood donations are gathered worldwide, aiding millions of transfusion events in various healthcare environments. Even with these advancements, guaranteeing that blood transfusions are given safely, efficiently, and solely when clinically necessary remains a significant challenge in modern medical practice. The acknowledgment that blood is a limited biological resource, along with heightened awareness of transfusion-related risks, has fundamentally altered clinical decision-making concerning the timing and methods of blood product administration (1, 2). Historically, liberal transfusion practices prevailed in clinical settings, mainly influenced by the enduring “10/30 rule,” suggesting red blood cell (RBC) transfusions when hemoglobin levels dropped below 10 g/dL or hematocrit fell below 30%. Despite being commonly used for many years, this threshold relied more on expert judgment than solid scientific proof. Recent developments in transfusion medicine and the rise of evidence-based clinical practices have questioned the applicability of liberal transfusion methods. Significant randomized controlled trials carried out over the last thirty years, such as the Transfusion Requirements in Critical Care (TRICC) trial and later multicenter studies, showed that restrictive transfusion strategies with lower hemoglobin thresholds are at least as effective as liberal strategies in numerous patient groups, while also minimizing unnecessary exposure to allogeneic blood products. These results have triggered a shift in practices toward restrictive transfusion methods in critical care, perioperative medicine, trauma management, and various medical fields (3, 4).

The reasoning for limited transfusion strategies goes beyond simply saving blood supplies. While transfusion reinstates the capacity to carry oxygen, receiving allogeneic blood transfusions can lead to a variety of possible adverse effects, such as transfusion-associated circulatory overload (TACO), transfusion-related acute lung injury (TRALI), both acute and delayed hemolytic transfusion reactions, alloimmunization, iron overload after multiple transfusions, transfusion-transmitted infections, and transfusion-related immunomodulation (TRIM). Additionally, growing evidence indicates that extended storage of RBC units can cause biochemical and structural changes referred to as the “storage lesion,” which may impact erythrocyte deformability, nitric oxide availability, oxygen transport, and microvascular circulation. Although the importance of storage lesions in clinical settings is still debated, these issues emphasize that transfusion must be viewed as a therapeutic procedure with specific indications, not just a harmless supportive action (5, 6). On the other hand, overly restricting transfusions could also put patients at significant risk. Severe anemia hinders systemic oxygen transport, diminishes tissue perfusion, worsens myocardial ischemia, elevates cerebral hypoxia, and may lead to multiorgan dysfunction in at-risk patients. The equilibrium between preventing unnecessary transfusions and mitigating the detrimental effects of insufficient oxygen delivery is particularly difficult in groups with reduced physiological capacity, such as elderly individuals, those with heart conditions, patients experiencing active bleeding, persons undergoing intricate surgeries, newborns, and critically ill individuals. Therefore, establishing the best transfusion threshold necessitates thoughtful evaluation of specific patient traits instead of solely depending on hemoglobin levels (7, 8).

The introduction of Patient Blood Management (PBM) has revolutionized contemporary hemotherapy by changing the focus from transfusion as the standard approach for anemia to integrated strategies that enhance patients’ natural blood resources. PBM involves evidence-driven strategies focused on diagnosing and managing anemia prior to surgery, reducing perioperative blood loss, improving coagulation, boosting physiological tolerance to anemia, and decreasing unnecessary transfusions. Global entities such as the World Health Organization (WHO), the AABB (Association for the Advancement of Blood and Biotherapies), the European Society of Anaesthesiology and Intensive Care (ESAIC), and the National Institute for Health and Care Excellence (NICE) currently promote PBM as an essential element of quality healthcare. Effective use of PBM has been linked to fewer transfusions, shorter hospitalization durations, decreased medical expenses, fewer postoperative issues, and enhanced patient results in various clinical environments (9–11).

This narrative review thoroughly investigates current evidence regarding restrictive and liberal blood transfusion approaches, focusing specifically on their physiological bases, clinical uses, comparative efficacy, and safety profiles across key medical and surgical disciplines. It compiles results from pivotal randomized controlled trials, global clinical guidelines, systematic reviews, and new translational research to pinpoint ongoing knowledge deficits and unresolved disputes in transfusion medicine. Additionally, the analysis examines novel methods—such as patient blood management, precision transfusion medicine, biomarker-guided transfusion, and artificial intelligence—that could revolutionize hemotherapy from a threshold-oriented intervention into a personalized, physiology-based treatment approach. By tackling these essential unmet needs, the review seeks to offer clinicians, researchers, and policymakers a thorough framework for promoting safer, more effective, and patient-focused blood transfusion practices in a time of precision medicine and global healthcare sustainability (12–14).

Methods

This study was conducted as a narrative review to critically examine the current evidence comparing restrictive and liberal blood transfusion strategies, identify persistent knowledge gaps, and discuss emerging approaches toward precision hemotherapy. Although this review was not designed as a formal systematic review or meta-analysis, a structured literature search and transparent article selection process were undertaken to enhance methodological rigor and reproducibility. A comprehensive literature search was performed in PubMed/MEDLINE, Embase, Scopus, Web of Science, and Google Scholar for articles published from database inception through June 30, 2026. The search strategy combined Medical Subject Headings (MeSH) and free-text terms, including “blood transfusion,” “red blood cell transfusion,” “restrictive transfusion,” “liberal transfusion,” “transfusion threshold,” “hemotherapy,” “patient blood management,” “anemia,” “critical care,” “acute coronary syndrome,” “traumatic brain injury,” “extracorporeal membrane oxygenation,” “oncology,” “precision medicine,” “artificial intelligence,” and “clinical decision support.” Boolean operators (“AND” and “OR”) were used to refine the search, and the reference lists of relevant reviews, landmark clinical trials, and international transfusion guidelines were manually screened to identify additional eligible publications.

Eligible studies included randomized controlled trials, observational studies, systematic reviews, meta-analyses, consensus statements, and evidence-based clinical practice guidelines addressing red blood cell transfusion thresholds, restrictive and liberal transfusion strategies, patient blood management, or emerging precision transfusion approaches. Editorials, conference abstracts without full-text publications, case reports, duplicate records, and studies not directly relevant to the objectives of the review were excluded. The search identified 962 records. After removing 227 duplicate citations, 735 titles and abstracts were screened. Following initial screening, 128 full-text articles were assessed for eligibility, of which 85 publications met the inclusion criteria and were incorporated into the final narrative synthesis. These included landmark randomized controlled trials, high-quality systematic reviews, meta-analyses, observational studies, and international guideline documents that collectively represent the current evidence base on restrictive and liberal transfusion strategies. A PRISMA-style flow diagram summarizing the literature identification and selection process has been included to improve methodological transparency.

Because this was a narrative review, a formal quantitative risk-of-bias assessment was not performed. However, the methodological quality of the included literature was critically evaluated during evidence synthesis. Particular consideration was given to study design, sample size, patient population, transfusion thresholds, clinical outcomes, methodological strengths and limitations, and the external validity of landmark randomized controlled trials and major guideline recommendations. Greater emphasis was placed on high-quality randomized trials, systematic reviews, meta-analyses, and internationally recognized clinical practice guidelines. The selected evidence was synthesized thematically into key areas, including the physiological basis of anemia tolerance, comparative evidence for restrictive and liberal transfusion strategies across major clinical settings, transfusion-related complications, patient blood management, unresolved controversies, current unmet clinical needs, and future directions in precision hemotherapy. Particular emphasis was placed on critically evaluating areas where evidence remains conflicting or insufficient and identifying opportunities for individualized transfusion decision-making in contemporary clinical practice.

Physiological basis of restrictive and liberal blood transfusion strategies

The physiological basis for blood transfusion primarily focuses on ensuring sufficient tissue oxygenation while preventing the negative effects linked to unnecessary transfusions. Red blood cells (RBCs) serve as the main transporters of oxygen, and blood transfusions have historically been regarded as an efficient method for replenishing oxygen-carrying ability in individuals experiencing anemia or significant blood loss. Nonetheless, developments in cardiovascular physiology, microcirculatory biology, and critical care medicine have shown that hemoglobin levels alone are an inadequate proxy for tissue oxygen supply. Instead, the equilibrium between oxygen delivery (DO₂), oxygen uptake (VO₂), tissue oxygen extraction, and the patient’s physiological reserves decides if transfusion is advantageous. This comprehension establishes the scientific basis for the current discussion surrounding restrictive and liberal transfusion approaches and highlights the significance of personalized hemotherapy (1). The main physiological role of erythrocytes is to carry oxygen from the lungs to distant tissues through hemoglobin. Oxygen supply to tissues relies on arterial oxygen content (CaO₂) and cardiac output, illustrated by the equation: Arterial oxygen content is primarily influenced by hemoglobin concentration, oxygen saturation, and the quantity of oxygen dissolved in plasma. Since dissolved oxygen has a negligible effect on total oxygen content in normal physiological states, hemoglobin concentration is the main factor governing oxygen transport. As a result, decreases in hemoglobin lower oxygen transport ability and might impair tissue oxygenation if compensatory mechanisms are depleted. Nonetheless, sufficient oxygen supply relies not just on hemoglobin levels but also on cardiac function, lung capability, blood vessel health, and microcirculation flow. Consequently, a patient who has a somewhat low hemoglobin level yet retains adequate cardiac output and functional compensatory mechanisms can sustain enough tissue oxygenation without the need for transfusion. On the other hand, individuals with diminished cardiac reserve, active bleeding, or serious lung disease might experience tissue hypoxia even with comparatively elevated hemoglobin levels (2, 3).

The human body has extraordinary adaptive processes that sustain tissue oxygen levels during moderate anemia. These compensatory mechanisms clarify why numerous stable patients endure restrictive transfusion limits without negative clinical effects. A primary compensatory reaction is the rise in cardiac output via heightened heart rate and stroke volume. Amplified sympathetic nervous system activation increases myocardial contractility, whereas peripheral vasodilation lowers systemic vascular resistance, thus promoting enhanced blood flow to essential organs. This flexible rise in cardiac output partially compensates for the decreased oxygen-carrying ability linked to falling hemoglobin levels. At the same time, tissues enhance their oxygen uptake from the blood being circulated. In typical physiological circumstances, peripheral tissues utilize around 25% of the oxygen supplied. In anemia, oxygen extraction can rise significantly, enabling tissues to sustain oxygen usage even with lower oxygen supply. The oxygen extraction ratio acts as a crucial physiological reserve that postpones the development of tissue hypoxia. A further important adaptation includes modifications within the erythrocytes themselves. Lower hemoglobin levels trigger enhanced synthesis of 2,3-bisphosphoglycerate (2,3-BPG), resulting in a decreased affinity of hemoglobin for oxygen by moving the oxygen-hemoglobin dissociation curve to the right. This promotes oxygen release at the tissue level, thus improving oxygen accessibility even with reduced circulating hemoglobin levels (4, 5).

Other compensatory mechanisms involve directing blood flow toward metabolically active organs like the brain and heart, increased erythropoietin production by the kidneys, plasma volume expansion, and improved oxygen diffusion via capillary recruitment. Together, these physiological reactions clarify why numerous otherwise healthy people can comfortably endure hemoglobin levels significantly lower than customary transfusion limits. While compensatory mechanisms maintain tissue oxygen levels in moderate anemia, they prove inadequate when oxygen delivery drops beneath a crucial limit. After this stage, oxygen usage relies on oxygen supply, leading to tissue hypoxia, anaerobic metabolism, lactate buildup, cellular impairment, and eventually organ failure. The idea of critical oxygen delivery (DO₂crit) is fundamental to contemporary transfusion physiology. Above DO₂crit, oxygen uptake stays fairly stable since tissues can enhance oxygen extraction when necessary. Beneath DO₂crit, though, compensatory mechanisms are ineffective, leading to a decrease in oxygen consumption that aligns with the reduction in oxygen delivery. Clinical symptoms can encompass hypotension, tachycardia, changes in mental status, myocardial ischemia, kidney impairment, metabolic acidosis, and increased serum lactate levels (Table 1) (6, 7).

Table 1. Physiological basis of restrictive and liberal blood transfusion strategies.

Current international guidelines on restrictive and liberal blood transfusion strategies

The evolution of transfusion medicine from empiricism to evidence-based practice has been accompanied by the development of comprehensive international guidelines aimed at standardizing transfusion decisions, improving patient outcomes, and promoting the judicious use of blood products. Contemporary guidelines increasingly advocate restrictive red blood cell (RBC) transfusion strategies for most hemodynamically stable patients while emphasizing that transfusion decisions should not rely solely on hemoglobin concentration. Instead, clinical judgment, patient comorbidities, ongoing blood loss, physiological indicators of tissue oxygenation, and individual risk-benefit assessments are integral components of modern transfusion practice. Although broad consensus exists regarding restrictive transfusion thresholds, variations remain for specific patient populations, reflecting differences in available evidence and unresolved clinical controversies (Table 2).

Table 2. Current international guideline recommendations on restrictive and liberal red blood cell transfusion strategies.

ACS, Acute Coronary Syndrome; AABB, Association for the Advancement of Blood & Biotherapies; ASA, American Society of Anesthesiologists; ESC, European Society of Cardiology; ESICM, European Society of Intensive Care Medicine; ESMO, European Society for Medical Oncology; Hb, Hemoglobin; ICU, Intensive Care Unit; NICE, National Institute for Health and Care Excellence; PBM, Patient Blood Management.

Evolution of guideline recommendations

Historically, transfusion protocols were mainly founded on expert opinion instead of robust clinical evidence. For many years, the conventional “10/30 rule” prevailed in clinical practice, advising transfusion when hemoglobin levels fell under 10 g/dL or hematocrit dropped below 30%. Nonetheless, significant randomized controlled trials—such as the Transfusion Requirements in Critical Care (TRICC), Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS), Transfusion Requirements After Cardiac Surgery (TRACS), and Transfusion Indication Threshold Reduction (TITRe2) studies—showed that reduced transfusion thresholds achieved clinical results similar to those of liberal approaches in various patient groups. These results led significant professional groups to update their guidelines to adopt more stringent transfusion methods. Present international guidelines consistently stress that blood transfusion should be seen as a therapeutic measure with specific indications instead of a standard reaction to anemia. Hemoglobin levels are a vital clinical measure; nonetheless, they must be evaluated in relation to overall cardiovascular stability, oxygen transport, anemia symptoms, and the individual’s physiological capacity (15, 16).

Association for the advancement of blood and biotherapies

The AABB, previously referred to as the American Association of Blood Banks, has released several of the most impactful evidence-based transfusion guidelines. The latest guidelines strongly endorse restrictive transfusion approaches for adult patients in the hospital who are hemodynamically stable. The AABB advises evaluating RBC transfusion when hemoglobin levels drop below 7 g/dL in stable hospitalized adults, including critically ill patients who are not experiencing active bleeding. This suggestion is backed by moderate- to high-certainty evidence showing that restrictive transfusion thresholds decrease blood usage without raising mortality, myocardial infarction, stroke, infection, or hospital stays. For patients receiving orthopedic surgery, cardiac surgery, or those with existing cardiovascular disease, the AABB suggests a slightly elevated threshold of about 8 g/dL, acknowledging that these individuals might have diminished physiological reserve and greater vulnerability to ischemic complications. Significantly, the guideline emphasizes that decisions regarding transfusions must consider anemia symptoms, persistent bleeding, hemodynamic status, and clinical judgment instead of solely depending on laboratory results. The AABB also emphasizes multiple clinical scenarios where evidence is inadequate to allow conclusive recommendations, such as individuals with acute coronary syndromes, severe traumatic brain injuries, chronic anemia requiring transfusions, and those on extracorporeal membrane oxygenation (ECMO). In these groups, personalized clinical evaluation is still crucial (17).

World health organization

The World Health Organization (WHO) examines transfusion practices through clinical and public health viewpoints. Acknowledging blood as a finite global healthcare resource, WHO supports the sensible and evidence-driven utilization of blood products while enhancing national blood systems and encouraging universal access to safe transfusion services. WHO strongly supports the adoption of Patient Blood Management (PBM) as a holistic approach to enhance patient results while reducing avoidable transfusions. Instead of suggesting strict universal hemoglobin levels, WHO highlights the importance of early identification and management of anemia, improving hemostasis, minimizing blood loss during surgery, and increasing the body’s tolerance to anemia. Special focus is placed on low- and middle-income countries (LMICs), where blood shortages, insufficient laboratory infrastructure, and weak hemovigilance systems pose major obstacles to safe transfusion practices. WHO advocates for nations to create evidence-driven national transfusion policies, enhance donor recruitment initiatives, bolster blood quality assurance, and adopt standardized clinical transfusion protocols (15).

European society of anaesthesiology and intensive care

The European Society of Anaesthesiology and Intensive Care (ESAIC) has developed comprehensive perioperative transfusion guidelines emphasizing individualized transfusion decisions integrated within multidisciplinary patient blood management programs. ESAIC recommends restrictive transfusion thresholds for most surgical patients, generally supporting transfusion at hemoglobin concentrations below 7–8 g/dL in stable individuals. However, clinicians are advised to consider multiple physiological variables, including cardiovascular function, ongoing blood loss, oxygenation, lactate levels, and signs of tissue hypoperfusion. The guidelines particularly emphasize minimizing perioperative transfusion through preoperative anemia management, meticulous surgical hemostasis, intraoperative cell salvage, antifibrinolytic therapy such as tranexamic acid, and point-of-care coagulation testing using thromboelastography (TEG) or rotational thromboelastometry (ROTEM). These interventions align closely with the principles of patient blood management and have demonstrated significant reductions in perioperative transfusion requirements (16).

National institute for health and care excellence

The National Institute for Health and Care Excellence (NICE) in the United Kingdom released one of the first national guidelines clearly endorsing restrictive transfusion practices. NICE recommends contemplating RBC transfusion when hemoglobin levels drop below 7 g/dL in patients who do not have significant bleeding or acute coronary syndromes, aiming for a post-transfusion hemoglobin target of 7 to 9 g/dL. For individuals with acute coronary syndromes, a target hemoglobin level of 8 to 10 g/dL might be suitable due to worries about oxygen supply to the myocardium. A key aspect of the NICE guideline is its focus on giving single-unit transfusions to non-bleeding adults, with a subsequent clinical reassessment before contemplating more units. This method minimizes unnecessary donor exposure and promotes personalized assessment of transfusion reactions. NICE also firmly endorses comprehensive patient blood management by systematically identifying preoperative anemia, implementing iron replacement therapy, reducing iatrogenic blood loss, and continuously auditing transfusion practices at institutions (14).

American society of anesthesiologists

The American Society of Anesthesiologists (ASA) advises tailored perioperative transfusion approaches instead of standardized hemoglobin thresholds. While transfusion is typically deemed suitable at hemoglobin levels beneath 6–7 g/dL, choices for patients with moderate hemoglobin levels (7–10 g/dL) must take into account factors such as ongoing bleeding, hemodynamic stability, oxygenation, cardiac performance, and expected surgical blood loss. ASA stresses that transfusion must not rely exclusively on laboratory results and promotes a thorough evaluation of clinical indicators of inadequate oxygen delivery, such as ongoing tachycardia, low blood pressure, metabolic acidosis, increased serum lactate, reduced mixed venous oxygen saturation, and signs of myocardial ischemia (15).

Society of critical care medicine

The Society of Critical Care Medicine endorses restrictive transfusion practices for critically ill adults, primarily based on findings from the TRICC trial and later randomized studies. Present SCCM guidelines recommend transfusion when hemoglobin levels fall below 7 g/dL in hemodynamically stable critically ill patients who do not have active bleeding or acute myocardial ischemia. Clinicians are advised to adjust transfusion choices based on disease severity, organ impairment, oxygen usage, and changing clinical conditions. Patients experiencing septic shock, respiratory failure, and multiorgan dysfunction often exhibit intricate physiological disruptions that impact oxygen delivery and its use. As a result, SCCM highlights the ongoing evaluation of tissue perfusion and oxygenation instead of sticking to fixed transfusion limits (16).

Evidence from landmark clinical trials

The shift from liberal to conservative blood transfusion approaches marks a significant change in modern transfusion medicine, primarily influenced by high-quality randomized controlled trials (RCTs). Before the 1990s, transfusion practices mainly relied on expert judgment and the conventional “10/30 rule,” lacking strong evidence to back specific hemoglobin levels. The introduction of extensive multicenter RCTs significantly questioned this approach by showing that restrictive transfusion methods are typically as safe as, and in certain populations better than, liberal methods. These studies have influenced global protocols, advanced patient blood management (PBM), and changed transfusion decision processes across various medical and surgical fields. Nonetheless, although the general evidence supports restrictive transfusion, significant exceptions exist within certain patient groups, emphasizing the necessity for personalized, physiology-driven hemotherapy (18).

The TRICC trial: establishing the foundation for restrictive transfusion

The TRICC trial, conducted by Hébert and team and published in 1999, is considered the pivotal research that revolutionized contemporary transfusion medicine. Implemented in various intensive care units in Canada, the study randomized 838 critically ill adults to either a restrictive transfusion approach (hemoglobin threshold of 7 g/dL, keeping levels between 7 and 9 g/dL) or a liberal approach (threshold of 10 g/dL, keeping levels between 10 and 12 g/dL). The research showed no rise in 30-day mortality for patients treated with restrictive transfusion. Actually, younger patients and individuals with less severe illness showed reduced mortality with the restrictive strategy. Moreover, restrictive transfusion considerably lowered the quantity of red blood cell (RBC) units given, thus reducing patient exposure to allogeneic blood products while maintaining clinical outcomes. Crucially, there was no rise noted in the rates of multiple organ dysfunction, length of stay in the intensive care unit (ICU), or infectious complications. The TRICC trial fundamentally questioned long-held beliefs that elevated hemoglobin levels always enhanced oxygen delivery and clinical results. Rather, it showed that numerous severely ill patients have enough physiological reserve to endure moderate anemia, provided that tissue oxygenation stays adequate. While the research omitted individuals with active bleeding and severe heart ischemia, its results laid the scientific groundwork for restrictive transfusion approaches and prompted many later studies in various clinical environments (19–21).

The FOCUS trial: restrictive transfusion in orthopedic surgery

The FOCUS trial on Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair tackled a significant clinical issue concerning elderly individuals with cardiovascular disease who are receiving orthopedic surgery. Released in 2011, the study included over 2,000 patients with cardiovascular risk factors who experienced postoperative anemia after hip fracture surgery. Participants were randomly assigned to either a liberal approach, where transfusion kept hemoglobin levels above 10 g/dL, or a restrictive approach, in which transfusion was done for symptomatic anemia or hemoglobin below 8 g/dL. The main endpoint included death and the inability to walk unaided after 60 days. The FOCUS trial identified no notable differences in mortality, functional recovery, cardiovascular events, or length of hospital stay between the restrictive and liberal transfusion groups. Restrictive transfusion significantly decreased RBC usage without negatively affecting patient outcomes. These results strengthened the idea that increased hemoglobin levels after surgery do not always result in better functional or cardiovascular results, even in older patients with preexisting cardiovascular conditions. The FOCUS trial greatly impacted perioperative transfusion standards and offered robust evidence in favor of restrictive transfusion practices in orthopedic surgery. However, since patients with acute coronary syndromes were not included, uncertainty continued about the best transfusion thresholds in individuals undergoing active myocardial ischemia (22–25).

The TRACS trial: cardiac surgery

Cardiac surgery poses distinct physiological difficulties as perioperative anemia can occur alongside reduced myocardial oxygen supply. The Transfusion Requirements After Cardiac Surgery (TRACS) trial investigated if restrictive transfusion was safe for this high-risk group. Patients scheduled for elective cardiac surgery were assigned randomly to either a restrictive hematocrit goal of around 24% or a liberal goal of about 30%. Clinical results encompassed death, significant morbidity, and complications after surgery. The TRACS trial showed that restrictive transfusion was not worse than liberal transfusion regarding mortality and significant postoperative complications. While patients undergoing liberal transfusion had elevated hemoglobin levels, there was no significant enhancement in clinical results. Restrictive transfusion notably decreased blood use and donor exposure, endorsing more cautious transfusion methods among numerous cardiac surgery patients. Nonetheless, the somewhat small sample size restricted the capacity to identify rare adverse events, and persistent worries about myocardial ischemia led to further studies in this group of patients (26–30).

TITRe2 trial: reassessing cardiac surgery

The Transfusion Indication Threshold Reduction (TITRe2) trial further examined transfusion thresholds following cardiac surgery by enrolling more than 2,000 patients across multiple United Kingdom hospitals. Patients with postoperative hemoglobin concentrations below 9 g/dL were assigned to either a restrictive threshold of 7.5 g/dL or a liberal threshold of 9 g/dL. The restrictive strategy resulted in significantly fewer transfusions and reduced blood utilization. Although no significant differences were observed in the composite outcome of serious infection or ischemic events, a modest increase in mortality was noted in the restrictive group during secondary analyses. Because mortality was not the primary endpoint and the study was not powered specifically for this outcome, interpretation remains controversial. Consequently, most guideline committees continue to support restrictive transfusion following cardiac surgery while emphasizing individualized assessment in patients with significant cardiovascular disease (31–35).

The Villanueva trial: acute upper gastrointestinal bleeding

The handling of acute gastrointestinal bleeding is another significant area in which transfusion medicine is applied. In a groundbreaking study released in 2013, Villanueva and his team assigned patients with acute upper gastrointestinal bleeding to either restrictive (hemoglobin limit of 7 g/dL) or liberal (limit of 9 g/dL) transfusion approaches. The restrictive approach was linked to notably lower mortality, decreased rebleeding rates, shorter hospital durations, and fewer complications compared to liberal transfusion. These advantages were especially evident in patients suffering from portal hypertension and liver cirrhosis. The suggested physiological rationale pertains to heightened portal venous pressure resulting from excessive transfusion, which could worsen variceal hemorrhaging and hinder hemostasis. This research significantly shaped gastroenterology guidelines and designated restrictive transfusion as the favored strategy for most patients experiencing acute upper gastrointestinal bleeding, excluding those with severe exsanguination or critical hemodynamic instability (36–39).

The REALITY trial: acute myocardial infarction

Patients experiencing acute coronary syndromes have traditionally posed significant uncertainties in transfusion medicine as myocardial oxygen extraction is typically close to maximal under normal physiological circumstances. The REALITY trial aimed to establish if restrictive transfusion was safe for patients with acute myocardial infarction who also had anemia. Patients with hemoglobin levels ranging from 7 to 10 g/dL were assigned to either restrictive or liberal transfusion strategies through randomization. The restrictive method showed non-inferiority in terms of significant cardiovascular incidents, such as mortality, recurrent heart attacks, strokes, and urgent revascularization. While the trial endorsed restrictive transfusion in certain patients with myocardial infarction, various limitations—such as a small sample size and limited follow-up—hindered conclusive findings. As a result, global guidelines persist in endorsing personalized transfusion choices for individuals with acute coronary syndromes while waiting for further evidence (40–43).

MINT trial: ongoing debate in acute coronary syndromes

The Myocardial Ischemia and Transfusion (MINT) trial is among the largest studies exploring transfusion thresholds for patients experiencing acute myocardial infarction. The research examined liberal versus restrictive transfusion approaches in patients suffering from anemia and acute coronary syndromes. While initial results indicated a potential trend for better cardiovascular outcomes with liberal transfusion, the differences were not consistently statistically significant in primary analyses. As a result, MINT has not settled the persistent discussion on ideal transfusion thresholds in myocardial ischemia but has emphasized the intricacies of weighing oxygen delivery against transfusion-associated risks in patients with impaired coronary perfusion. The trial highlights the necessity for upcoming studies to include physiological markers of myocardial oxygenation instead of solely depending on hemoglobin levels (44–46).

TRISS trial: septic shock

The Transfusion Requirements in Septic Shock (TRISS) study analyzed restrictive transfusion in individuals with septic shock, a state marked by significant changes in microcirculation and oxygen usage. Over 1,000 patients were assigned to hemoglobin levels of either 7 g/dL or 9 g/dL at random. No notable differences were noted in mortality, ischemic incidents, use of life-support interventions, or severe adverse reactions between the two groups. Significantly, the restrictive transfusion approach lowered RBC usage by around 50%, showcasing considerable resource savings while maintaining survival rates. The TRISS trial bolstered evidence for restricted transfusion in critically ill sepsis patients and directly influenced guidelines in the Surviving Sepsis Campaign (47–50).

Pediatric and neonatal trials

Support for restrictive transfusion has also surfaced in pediatric critical care. The TRIPICU trial showed that stable critically ill pediatric patients could safely manage restrictive transfusion thresholds without an increase in mortality or organ failure. Likewise, various neonatal studies indicate that restrictive transfusion decreases donor exposure in premature infants, although ideal thresholds remain under investigation due to worries about neurodevelopmental outcomes. These results suggest that age by itself should not rule out restrictive transfusion, but underscore the importance of thoughtfully considering developmental physiology and factors specific to the disease (10, 51).

Precision hemotherapy: from fixed thresholds to individualized transfusion algorithms

The evolution of transfusion medicine is increasingly shifting from the application of uniform hemoglobin thresholds toward precision hemotherapy, in which transfusion decisions are individualized according to each patient’s physiological status, clinical context, and predicted response to therapy. Although restrictive transfusion strategies (hemoglobin thresholds of 7–8 g/dL) remain appropriate for most hemodynamically stable adults and are supported by robust clinical evidence, reliance on hemoglobin concentration alone does not fully capture the complexity of oxygen delivery, tissue perfusion, and organ-specific vulnerability to anemia. Patients with identical hemoglobin levels may have markedly different transfusion requirements depending on age, cardiovascular reserve, inflammatory burden, metabolic demand, comorbid conditions, and the presence of ongoing bleeding or organ ischemia (13, 52). Precision hemotherapy seeks to address these limitations by integrating conventional laboratory parameters with dynamic physiological indicators that more accurately reflect tissue oxygenation and perfusion. In addition to hemoglobin concentration, transfusion decisions may incorporate serum lactate, central or mixed venous oxygen saturation (ScvO₂/SvO₂), arterial oxygen content, hemodynamic stability, near-infrared spectroscopy-derived tissue oxygen saturation, echocardiographic assessment of cardiac function, and biomarkers of endothelial dysfunction and inflammation. Rather than applying a universal transfusion trigger, these complementary parameters provide a more comprehensive assessment of whether anemia is causing clinically significant impairment of oxygen delivery and whether red blood cell transfusion is likely to improve patient outcomes (53, 54).

An emerging component of precision hemotherapy is the development of multivariable clinical decision algorithms that combine laboratory findings with patient-specific clinical characteristics. Such algorithms may integrate factors including patient age, frailty, cardiovascular disease, chronic kidney disease, active hemorrhage, severity of illness, oxygen requirements, vasopressor use, and symptom burden alongside hemoglobin concentration. By simultaneously evaluating multiple variables, these models have the potential to identify patients who are most likely to benefit from transfusion while avoiding unnecessary exposure in those who can safely tolerate lower hemoglobin levels. Although several predictive models have been proposed, prospective validation across diverse clinical settings remains necessary before widespread clinical implementation (55). Recent advances in artificial intelligence (AI) and machine learning have further accelerated the movement toward individualized transfusion practice. Machine learning algorithms can analyze large volumes of multidimensional clinical data—including laboratory values, physiological measurements, medication history, imaging findings, comorbidities, and longitudinal electronic health record information—to estimate transfusion requirements, predict adverse transfusion reactions, and support personalized transfusion thresholds. Unlike traditional rule-based approaches, AI systems continuously refine predictive performance as additional clinical data become available. Nevertheless, these technologies should complement rather than replace clinician judgment, and their adoption requires rigorous external validation, transparency, explainability, and careful evaluation of algorithmic bias to ensure equitable and safe patient care (56, 57).

The integration of digital health technologies offers additional opportunities to enhance precision hemotherapy. Continuous physiological monitoring systems, wearable biosensors, bedside tissue oxygenation devices, and interoperable electronic health records can facilitate real-time reassessment of patient status, allowing transfusion decisions to evolve dynamically rather than relying on isolated laboratory measurements. Clinical decision-support platforms embedded within electronic health systems may automatically synthesize patient-specific information and generate evidence-informed recommendations while preserving clinician oversight. Such approaches have the potential to improve consistency of practice, reduce unnecessary transfusions, and optimize utilization of limited blood resources (58). Patient Blood Management (PBM) provides an ideal foundation for the implementation of precision hemotherapy. By emphasizing optimization of erythropoiesis, minimization of blood loss, and evidence-based transfusion practices, PBM complements individualized decision-making and supports more judicious use of blood components. Integrating PBM principles with physiological monitoring, predictive analytics, and validated clinical decision-support tools may facilitate a transition from threshold-based transfusion toward a truly patient-centered model of hemotherapy (59).

Controversies and emerging directions in restrictive versus liberal blood transfusion strategies

Although restrictive transfusion strategies (hemoglobin thresholds of 7–8 g/dL) are supported by robust evidence and endorsed by major international guidelines for most hemodynamically stable hospitalized adults, important controversies remain regarding their applicability across heterogeneous patient populations. The reliance on uniform hemoglobin thresholds may oversimplify the complex physiological determinants of oxygen delivery, tissue perfusion, and organ-specific vulnerability to anemia. Consequently, individualized transfusion decision-making has emerged as a major priority in contemporary transfusion medicine (60). Patients with acute coronary syndrome (ACS) represent one of the most debated populations. Myocardial oxygen extraction is already near maximal under physiological conditions, limiting the heart’s capacity to compensate for reductions in oxygen-carrying capacity. While restrictive transfusion strategies reduce exposure to transfusion-related complications, several studies have suggested that excessively restrictive thresholds may increase the risk of recurrent ischemia or adverse cardiovascular events in selected patients with active myocardial ischemia. Conversely, liberal transfusion may improve oxygen delivery but also carries risks of transfusion-associated circulatory overload, inflammation, and thrombosis. Recent randomized trials have yielded conflicting results, highlighting the continued need for adequately powered studies to define optimal transfusion thresholds in this high-risk population (61, 62).

Similarly, the optimal transfusion strategy for patients with traumatic brain injury (TBI) remains uncertain. Cerebral tissue is particularly susceptible to hypoxia, yet liberal transfusion has not consistently demonstrated improved neurological outcomes and may increase complications associated with blood product exposure. Heterogeneous study populations, variable injury severity, and differences in neurological outcome measures limit current evidence. Future investigations should incorporate advanced cerebral oxygenation monitoring and neurophysiological biomarkers to better individualize transfusion decisions (4). In oncology, anemia frequently results from chemotherapy, bone marrow infiltration, chronic inflammation, nutritional deficiencies, and repeated phlebotomy. Although restrictive transfusion strategies are generally considered safe in clinically stable patients, transfusion decisions should also consider symptom burden, treatment intent, performance status, and quality of life. Moreover, concerns regarding transfusion-related immunomodulation and its potential influence on infection risk, postoperative recovery, and tumor progression remain incompletely understood. Additional prospective studies are needed to establish disease-specific and treatment-specific transfusion strategies across diverse cancer populations (63).

Patients requiring extracorporeal membrane oxygenation (ECMO) present another area of ongoing controversy. Historically, liberal transfusion practices were adopted to maximize oxygen delivery within the extracorporeal circuit. However, accumulating observational evidence suggests that restrictive transfusion strategies may be feasible in carefully selected ECMO patients without compromising survival, while simultaneously reducing transfusion-related complications and conserving blood resources. Nevertheless, the absence of large randomized controlled trials limits definitive recommendations, particularly in patients with severe hypoxemia or prolonged ECMO support (64). The management of frail older adults also remains challenging. Age-related reductions in physiological reserve, multimorbidity, cardiovascular disease, cognitive impairment, and decreased functional capacity may reduce tolerance to anemia compared with younger individuals. Conversely, older adults are particularly vulnerable to transfusion-associated circulatory overload and other adverse events. Existing randomized trials have included relatively few frail elderly patients, limiting the generalizability of current recommendations. Future research should incorporate measures of frailty, functional recovery, cognitive outcomes, and patient-reported quality of life alongside traditional mortality endpoints (65).

These persistent controversies underscore the limitations of applying universal hemoglobin thresholds across diverse clinical settings. Hemoglobin concentration alone provides an incomplete assessment of tissue oxygenation and does not account for individual variations in cardiovascular reserve, metabolic demand, microcirculatory function, or compensatory physiological responses. Consequently, transfusion medicine is increasingly transitioning toward precision hemotherapy, in which transfusion decisions integrate both laboratory findings and patient-specific physiological characteristics (66). Emerging technologies are expected to accelerate this transition. Artificial intelligence (AI) and machine learning algorithms have the potential to synthesize multidimensional clinical data—including hemoglobin concentration, vital signs, lactate levels, mixed or central venous oxygen saturation, tissue oxygenation measurements, laboratory biomarkers, imaging findings, and comorbid conditions—to estimate individualized transfusion benefit and predict the risk of transfusion-related adverse events. Rather than replacing clinical judgment, these decision-support systems could enhance clinician accuracy by providing dynamic, evidence-based recommendations tailored to individual patients (67). Parallel advances in digital health integration further support personalized transfusion practice. The incorporation of electronic health records, continuous physiological monitoring, wearable sensors, and real-time clinical dashboards may enable automated risk stratification and earlier identification of patients who are most likely to benefit from transfusion. Such systems could facilitate timely reassessment as patients’ clinical conditions evolve, moving beyond static hemoglobin thresholds toward dynamic, physiology-guided transfusion strategies (68).

Current unmet needs and future research priorities

Despite decades of clinical investigation and the widespread adoption of restrictive transfusion strategies in many healthcare settings, important uncertainties remain regarding the optimal use of red blood cell transfusion. While current guidelines provide evidence-based hemoglobin thresholds for most stable hospitalized adults, they cannot fully account for the considerable heterogeneity in patient physiology, disease severity, comorbid conditions, and response to anemia. Consequently, several critical unmet needs continue to limit the implementation of truly individualized transfusion practice. One of the most significant limitations is the continued reliance on hemoglobin concentration as the primary transfusion trigger. Although hemoglobin is a readily available and standardized laboratory parameter, it does not directly reflect tissue oxygen delivery, oxygen extraction, microcirculatory function, or the adequacy of organ perfusion. Patients with identical hemoglobin concentrations may exhibit markedly different physiological responses depending on age, cardiovascular reserve, metabolic demand, inflammatory status, and the presence of underlying organ dysfunction. Future transfusion strategies should therefore incorporate physiological indicators of tissue hypoxia, including serum lactate, central or mixed venous oxygen saturation, tissue oxygenation monitoring, and dynamic hemodynamic parameters, alongside conventional laboratory measurements (69–71).

Another major evidence gap concerns special patient populations, in whom the balance between transfusion benefit and potential harm remains uncertain. Although restrictive transfusion thresholds have demonstrated safety in many clinical settings, high-quality randomized evidence remains limited for patients with acute coronary syndrome, traumatic brain injury, extracorporeal membrane oxygenation (ECMO), hematologic malignancies, chronic transfusion-dependent disorders, and frail older adults. These populations often have unique physiological characteristics that may alter tolerance to anemia and susceptibility to transfusion-related complications. Well-designed multicenter randomized controlled trials with adequate statistical power are needed to establish patient-specific transfusion thresholds and clinically meaningful outcome measures. The development and validation of novel biomarkers represent another important research priority. Biomarkers reflecting tissue hypoxia, endothelial dysfunction, inflammation, oxidative stress, erythropoietic activity, and iron metabolism—including hepcidin, erythroferrone, hypoxia-inducible factor pathways, endothelial activation markers, and metabolomic signatures—may improve assessment of transfusion need beyond hemoglobin concentration alone. However, most remain investigational, and further prospective studies are required to determine their diagnostic accuracy, prognostic value, and clinical utility (72, 73).

Advances in artificial intelligence (AI) and machine learning offer considerable promise for precision transfusion medicine but remain at an early stage of clinical translation. Predictive algorithms capable of integrating laboratory data, physiological parameters, comorbidities, imaging findings, medication profiles, and electronic health record data could support individualized estimates of transfusion benefit and risk. Nevertheless, existing models are frequently developed using retrospective datasets, with limited external validation and uncertain generalizability across diverse patient populations and healthcare systems. Future research should prioritize prospective multicenter validation, transparent algorithm development, bias mitigation, explainable AI methodologies, and regulatory oversight to facilitate safe clinical implementation. The integration of digital health technologies into transfusion medicine also warrants further investigation. Continuous physiological monitoring, wearable biosensors, bedside tissue oxygenation devices, and interoperable electronic health record systems could enable dynamic reassessment of transfusion requirements rather than reliance on isolated laboratory measurements. Research should evaluate whether these technologies improve clinical outcomes, reduce unnecessary transfusions, and enhance clinician decision-making in routine practice (48, 74, 75).

Despite increasing global implementation of Patient Blood Management (PBM) programs, significant disparities remain in their adoption, particularly in low- and middle-income countries. Resource limitations, inadequate infrastructure, shortages of trained personnel, inconsistent availability of blood components, and limited access to diagnostic and monitoring technologies continue to impede implementation. Future implementation science studies should evaluate cost-effective, context-specific PBM strategies that are adaptable to resource-constrained healthcare environments while maintaining patient safety. Another important unmet need is the evaluation of patient-centered outcomes (76–78). Most transfusion trials have focused primarily on mortality, length of hospital stay, and major adverse clinical events. Comparatively little attention has been given to functional recovery, exercise capacity, cognitive performance, health-related quality of life, fatigue, symptom burden, and patient-reported outcomes, which may better reflect the real-world impact of transfusion strategies (79). Incorporating these measures into future clinical trials will provide a more comprehensive assessment of transfusion effectiveness. Finally, there is a growing need to transition from uniform transfusion thresholds toward precision hemotherapy, in which transfusion decisions are individualized according to patient-specific physiological characteristics rather than fixed hemoglobin concentrations alone (80). Such an approach would integrate laboratory findings with measures of tissue oxygenation, hemodynamic status, disease-specific risk factors, biomarker profiles, and validated clinical decision-support algorithms. While this paradigm remains under development, it represents a promising strategy for improving transfusion safety, maximizing therapeutic benefit, minimizing unnecessary blood exposure, and optimizing stewardship of limited blood resources (78).

Future research priorities

Future investigations should focus on several strategic areas that are likely to shape the next generation of transfusion practice:

• Conduct large, adequately powered randomized controlled trials evaluating transfusion thresholds in understudied populations, including patients with acute coronary syndrome, traumatic brain injury, ECMO support, hematologic malignancies, and frail older adults.
• Validate physiological markers of tissue oxygenation and perfusion as adjunctive transfusion triggers beyond hemoglobin concentration (81).
• Develop and prospectively validate AI-enabled clinical decision-support systems using diverse, multicenter datasets with transparent and explainable algorithms.
• Evaluate the clinical effectiveness and cost-effectiveness of integrating wearable monitoring devices, digital health platforms, and real-time electronic health record analytics into transfusion decision-making.
• Expand implementation research on Patient Blood Management to improve adoption across resource-limited healthcare settings.
• Incorporate patient-reported outcomes, functional recovery, cognitive performance, and quality-of-life measures into future transfusion trials (82).
• Explore multi-omics approaches, including genomics, proteomics, metabolomics, and systems biology, to identify novel biomarkers that support individualized transfusion decisions.
• Investigate precision hemotherapy frameworks that combine clinical assessment, physiological monitoring, biomarker data, and predictive analytics to deliver personalized transfusion care.

Conclusion

Restrictive transfusion strategies are supported by strong evidence for most hemodynamically stable patients and have improved transfusion safety, reduced unnecessary blood utilization, and optimized healthcare resources. However, important uncertainties remain in specific populations, including patients with acute coronary syndrome, traumatic brain injury, ECMO support, malignancy, frailty, and ongoing hemorrhage, where individualized clinical assessment is essential. These limitations underscore the need to move beyond fixed hemoglobin thresholds toward a more personalized approach to transfusion decision-making. Future hemotherapy should integrate hemoglobin concentration with physiological markers of tissue oxygenation, patient-specific clinical characteristics, validated biomarkers, and emerging digital technologies, including artificial intelligence–assisted clinical decision support. Combined with the principles of Patient Blood Management, these innovations have the potential to improve transfusion safety, enhance clinical outcomes, and preserve limited blood resources. Continued high-quality research is needed to validate precision transfusion algorithms and establish individualized, evidence-based strategies that address the remaining unmet needs in modern hemotherapy.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Arthur Daniel Rocha Alves, Aggeu Magalhães Institute (IAM), Brazil

Reviewed by: Ajay Gandhi, BLK Super Speciality Hospital, India

Mariagiovanna Cozza, AUSL Bologna, Italy