Bad Blood: Is the Debate Over Stored Red Blood Cells Still Relevant?

With age comes wisdom, the saying goes, unless we are talking about red blood cells (RBCs). Then with age comes the theoretical risk of a lack of ability to deliver oxygen, a potential failure to traverse microcirculation, and a slew of biomechanical differences that may affect outcomes after transfusion.

Still, each year, thousands of people in the United States receive a blood transfusion. It would be impossible for each one of these patients to get the freshest RBCs available.

The U.S. Food and Drug Administration regulations state that RBCs with additive solutions can be banked for up to 42 days and still be considered usable. That’s good news for U.S. patients who require about 36,000 units of RBC daily, either because they are undergoing an operation or an emergency procedure, receiving a new organ, or being treated for cancer and other diseases that may compromise bone marrow function.

To maximize RBC supplies, blood banks generally dispense on a first-in, first-out basis, the way milk is rotated in convenience store refrigerators, working off the assumption that blood stored up to 42 days will degrade to some extent, but that blood stored for this long is not inherently harmful.

“We do know that a number of changes occur in red cells while they are stored in added solution over the 42-day limit – these are morphological, biochemical, and physiologic changes,” explained John Roback, MD, PhD, of the Emory University Blood Bank in Atlanta, Georgia. “The biggest concern that patients and physicians have is whether older red cells are as efficacious as fresh red cells, or if they may actually be deleterious.”

But if old blood loses its efficacy and raises safety concerns, and managing supplies means storing RBCs and using these banked supplies in transfusions, he asked, “Where does that leave clinicians who need to waste as little of this precious commodity as possible?”

In a 2013 ASH Hematology paper, James Zimring, MD, PhD, of Puget Sound Blood Center Research Institute in Seattle, Washington, laid out the central point in the fresh versus stored blood for transfusion debate: “The answer to the question of whether there is a difference between fresh and old blood seems to be an unequivocal ‘yes.’ … On the contrary, the question of ‘does it matter’ remains unclear and is likely to remain so for some time.”1

So, does age of transfused RBCs matter? ASH Clinical News turned to some experts in the field for a review of the current data on fresh versus stored blood, and what those data mean for clinical practice.

With RBCs, Age Is Just a Number

As Dr. Zimring noted, RBCs undergo a number of changes during storage. Free hemoglobin (from in-bag RBC breakdown) can scavenge nitrous oxide (NO), which may deprive blood vessels of necessary NO. Studies have also shown that stored RBCs acquire procoagulant activities and have increased advanced glycation end (AGE) products, which play a role in inflammatory pathologies post-transfusion, such as transfusion-related lung injury.1 “The bottom line is, you can look at red cells that have been stored up to the limit of 42 days, and you can clearly document the biochemical differences in those red cells versus those that have been in storage for just a few days,” Steven Kleinman, MD, of the Centre for Blood Research at the University of British Columbia in Vancouver, Canada, and senior medical advisor to AABB, told ASH Clinical News.

In terms of patient outcomes, the data supporting the hypothesis that older blood is inferior to fresher blood come mostly from retrospective and observational studies, as well as animal models.

For instance, in a 2008 retrospective analysis of outcomes among patients undergoing coronary artery bypass surgery who received transfusions of RBCs stored for either <2 weeks or ≥2 weeks, researchers found that 1.7 percent of the 2,000-plus patients who received fresher RBCs died in the hospital, compared with 2.8 percent of patients who received older blood (p=0.004). One-year mortality was also significantly less in patients given newer blood (7.4% vs. 11%; p<0.001).2

Another retrospective study of more than 400,000 patient records in Scandinavia reported a 5 percent increased risk of death among patients who had received older blood (stored for >30 days). However, the researchers cautioned the “risk pattern was more consistent with weak confounding than with an effect of the momentary exposure to stored RBCs.”3

Confounding was a limitation in both of these observational studies, Dr. Kleinman agreed. “If you get older blood, you are likely to get more units of blood,” he said. “And if you get more units of blood, you are probably sicker to begin with. The people who received the older units had lower survival, but that was most likely a surrogate for their underlying conditions.”

But, in 2013, when researchers from the National Institutes of Health conducted the first randomized, blinded trial, they showed that blood transfused at the end of the storage period increased mortality. This was, though, an animal study: Beagles infected with Staphylococcus aureus pneumonia were randomized to transfusion with either seven- or 42-day-old canine universal donor blood.4

They observed a number of problems associated with older-blood transfusion, including increased mortality (p=0.0005), increased arterial alveolar oxygen gradient at 24-48 hours after infection (p<0.01), and more severe lung damage as evidenced by increased necrosis, hemorrhage, and thrombosis (p=0.03).

Dr. Roback noted that these animal trials were done under fairly extreme conditions. “If you transfuse a big volume of red cells stored for nearly 42 days, and if you simultaneously add another insult – in this case experimental pneumonia – the test subjects that received older red cells did much worse compared with those who received fresher red cells. So, in those extreme conditions, there is an adverse effect caused by older red cells that doesn’t happen when you use fresher red cells.”

When Dr. Roback and co-authors explored this question in a limited number of people, though, they found similar outcomes. In the first study, 43 hospitalized patients with transfusion orders were randomly assigned to receive either fresh (<14 days) or older (>21 days) stored RBC units; investigators noted a significant reduction in NO-mediated vasodilation at 24 hours post-infusion with older RBC units (p=0.045), while fresh RBC transfusions had no effect (p=0.231).5

In a second study, Dr. Roback and colleagues evaluated blood stored up to 42 days, and found “previously unrecognized vaso-inhibitory activity of stored RBCs,” concluding that, “through this novel mechanism, transfusion of small volumes of stored blood may be able to disrupt physiologic vasodilatory responses and thereby possibly cause adverse clinical outcome.”6

“In those very specialized systems in which we specifically looked at one experimental parameter – in this case vascular responsiveness – there was a difference between receiving the freshest blood and the oldest stored blood,” Dr. Roback said.

What Do the Studies Say?

But do these physiologic differences between fresh and stored blood affect patient outcomes? And, if so, how can RBC storage practices be changed to address these adverse effects?

“The [stored] red cells look different in the bag [in storage], but does that affect their usefulness when transfused?” Dr. Kleinman asked. “Some of the biochemical elements in older blood are restored upon transfusion and circulation. For instance, there’s a compound called 2,3 DPG (2,3-diphosphoglycerate), which is part of an energy metabolism circuit with adenosine triphosphate (ATP). During storage, that compound is depleted, but within 24 hours of transfusion, it is replenished – just through the act of being transfused.”

A number of randomized, controlled trials (RCT) have recently made a strong case for the vitality of old blood, or, at the very least, that fresh blood is not necessarily superior.

First, there was the Age of Blood Evaluation (ABLE) pilot trial, which enrolled critically ill adults from tertiary care intensive care units (ICU) at 64 centers in five countries.7

RBC units were leukoreduced (for an explanation of this process, see the SIDEBAR, “What is the Big Deal About Leukoreduction?”) for storage in an additive solution. The ABLE researchers compared outcomes between patients who received fresh RBC (stored <8 days) and those who received the oldest available compatible blood (the standard practice).

More than 5,000 RBC units were transfused in each study group, and “fresh” blood had been stored for an average of 6.1 days compared with 22 days in the “standard” blood group. At 90 days, 37 percent of patients in the fresh-blood group and about 35 percent in the standard group had died, for an absolute risk difference of 1.7 percentage points. In a survival analysis, the hazard ratio for death was 1.1 for fresh- versus standard-blood groups.

“We surmise that the use of fresh red cells is not justified at this time,” the ABLE researchers concluded. “We might also infer that changes to red cells or the storage medium that have been documented in many laboratory studies have limited clinical consequences.”

However, they noted that most patients received transfusions based on a “restrictive transfusion strategy,” specifically a mean pre-transfusion hemoglobin level of 7.7 g/dL, so exposure to red cells was much less than potential exposure levels at centers that opt for a “more liberal” blood transfusion strategy.

Still, based on ABLE’s outcomes “we can say for sure fresh blood [did not yield] an advantage even in the sickest patients,” said Paul Hébert, MD, of the University of Montreal in Canada, one of the study’s investigators. “Can these results be extrapolated to other populations? Yes, to some degree, particularly for acutely ill patients, or perioperative patients that are less sick. Chances are that fresh blood won’t improve their lot either.”

Next up was the Red-Cell Storage Duration Study (RECESS), a multicenter, prospective trial examining 1,481 patients 12 years or older who were undergoing complex cardiac surgery and were likely to need an RBC transfusion.8

The RECESS patient population included 1,481 cardiac surgery patients who were selected to receive blood either housed for ≤10 days or for ≥21 days. To determine the effect of blood storage duration on patient outcomes, the investigators measured the change in each patient’s Multiple Organ Dysfunction Score (MODS), then compared the scores from before surgery with scores obtained seven days after surgery (or until the patient’s time of death or discharge, whichever came first).

The investigators found no significant differences between the two groups, even after comparing the change in MODS for only post-operative scores. The mean change in the MODS score at seven days (the study’s primary outcome was 8.5 (out of a 24-point scale) in the short-term storage group and 8.7 points in the longer-term storage group (p=0.44).

There were also no significant differences between the groups for all-cause mortality, 28-day change in MODS, length of hospital stay, or length of ICU stay, leading the authors to conclude that “a between-group difference of 1 point or less in the change in MODS is unlikely to be clinically significant or to warrant a major change in the practice of blood banking,” according to lead investigator Marie Steiner, MD, of the University of Minnesota in Minneapolis.

Although RECESS looked at fresher blood versus middle-aged blood (not the oldest stored blood), Dr. Roback praised the trial for being one of the largest, most well-designed RCTs published on the age of transfused blood.

The Fresher, the Better?

Of course, RECESS and ABLE were not performed specifically in patients with blood disorders. For hematologists, results from the TOTAL trial hit closer to home, according to Dr. Kleinman.

In TOTAL, the findings of which were presented at the 2015 ASH Annual Meeting, 290 children (age range = 6-60 months) with malaria or sickle cell disease were randomized to receive leukoreduced RBCs that had been stored for one to 10 days versus 25 to 35 days.9 Investigators looked specifically at whether refrigerated blood storage diminished the blood’s ability to transfer oxygen to tissues, as measured by transfused patient’s blood lactate levels, explaining that, when tissue oxygen levels are critically low, lactate levels rise, and when tissues are successfully re-oxygenated, lactate levels may fall again.

Mean lactate levels were not statistically different between the two groups and neither were 30-day recovery rates, they stated, suggesting that longer-storage RBCs are not inferior to shorter-storage RBCs.

In Dr. Kleinman’s estimation, the results from the TOTAL trial have been overlooked. “This study hasn’t been talked about much, but I believe it moves the debate a little more in favor of the argument that stored blood doesn’t affect clinical outcomes,” he said. Additionally, the TOTAL results may have significance for global health policy decisions regarding the acceptable duration of RBC storage, particularly in developing nations where blood supplies are already limited.

Another pediatric study, ARIPI, echoes the TOTAL results, demonstrating that the use of fresh RBCs (≤7 days) compared with standard stored blood did not improve outcomes (including major neonatal morbidities such as intraventricular hemorrhage or nosocomial infection) in premature, very-low-birth-weight infants who required a transfusion.10

Lastly, Dr. Kleinman described a meta-analysis and systematic review evaluating 12 trials with over 5,000 participants that reported the following:11

  • Similar risk of death for fresh versus old transfused blood (relative risk [RR] = 1.04; p=0.45)
  • No difference in adverse event with age of red cells (RR=1.02; p=0.74)
  • A 9% relative increase in the risk of infection with fresh blood (RR=1.09; p=0.04)

An ongoing Canadian multicenter trial, INFORM, may bolster this conclusion, the authors noted. The three-year trial, started in 2012 and completed last year, “aims to determine the effect on in-hospital death rates of transfusing the freshest available blood compared with standard-issue blood.”

The experts who spoke with ASH Clinical News were divided on how INFORM results may, or may not, affect the “fresh versus old” debate – especially in terms of patients who receive large volumes of end-date blood.

“If the investigators are able to do a subgroup analysis of patients who did receive the oldest blood, INFORM may completely change the course of the debate,” Dr. Roback pointed out.

Dr. Hébert expressed more skepticism, though. “We all hope that it will give us new information, but the trial is basically comparing ‘front of the fridge’ versus ‘back of the fridge’ blood policies,” he said. “The chances of the investigators finding something beyond what we found [in ABLE] are not likely.”

The Debate Rages On

Given the similar rates of patient outcomes among ABLE, RECESS, TOTAL, and ARIPI, clinicians may feel that the new-versus-old debate has been settled – but not quite.

“In the scientific, academic community, the topic has seen advancement,” Dr. Kleinman said. “These trials offer fairly definitive data, but have those people who are strong proponents of fresh blood changed their minds? That we don’t quite know.”

One reason for lingering concern is that the trialists set their own limits for what constitutes “fresh” and “old” RBC units – and there may never be a universal definition, Dr. Hébert said.

“The definition of ‘fresh’ is different at different institutions,” he explained. “[In ABLE] we were pragmatic and set a limit of seven days for fresh blood, since it takes two days to get the blood into the hospital in the first place. For us, a week was fresh enough.” Though, he adds, setting a universal definition was not the goal of the trial. “What we really wanted to do was understand the clinical impact and outcomes.”

What about focusing on the other end of the spectrum of blood freshness – deciding how old is too old? In other words, is there a point beyond which stored blood becomes ineffective or even harmful? The jury is still out on whether using blood at the extreme end of the storage cycle (up to or at 42 days) would have detrimental outcomes for transfusions, and it may be a while until the medical community gets those answers.

Running a trial that tests old versus older or oldest blood would be problematic, Dr. Hébert said. “For me, old blood is over 30 days, but we still don’t know at what point blood gets ‘very old’ and potentially dangerous. To push the limit of that question in patients becomes unethical, though.”

Dr. Roback agreed, explaining that patients who require a high-volume transfusion are generally much sicker, and giving them only old blood may not be feasible, even in a controlled, experimental situation.

“To see the same effect in humans as we saw in the NIH’s trial in beagles, we may need to transfuse many more units of near out-of-date blood,” he said. “In humans, the effects of transfusing the oldest stored blood might not be seen unless patients receive only old RBC units, if they receive a high volume of it, and if they have a co-existing insult.”

Dr. Steiner acknowledged that, while RECESS did not address the potential impact of blood stored near or at the 42-day limit, “those patients are infrequent and, therefore, difficult to study.”

Then there is the issue of fresh versus old blood transfusions in patients with chronic conditions, such as sickle cell disease, or in those receiving hematopoietic cell transplantation (HCT) or primary chemotherapy for leukemia. There is scant scientific evidence in these areas, according to Dr. Kleinman, making it difficult to definitively judge whether fresh RBCs are superior to older RBCs.

However, he advocates for a common-sense approach. “You might want to give fresher blood to patients who need chronic RBC transfusions for the simple logistical reason that fresher blood lasts longer; they won’t have to be transfused as frequently, and you’ll avoid the iron overload problems that can occur in patients who undergo multiple transfusions.”

The efficacy of older RBC in these patients isn’t necessarily being called into question, he said, “but you want to minimize the number of patient visits. You want their hemoglobin up for a greater number of weeks so that they don’t reach iron overload quite as quickly.”

Practical Implications

Taken together, do the results from the various trials offer the transfusion community any comfort? In a sense, yes: There’s no need to change the 42-day storage limit or the first-in, first-out approach to blood storage.

At Dr. Steiner’s institution, the first-in, first-out protocol has always been used, Dr. Steiner said, so the RECESS results did not render any changes to that approach. Overall, she said, the RECESS results “do not support preferentially transfusing ‘new’ blood to provide a clinical outcome advantage in adult cardiac surgery.”

“We have no evidence to say people shouldn’t accept the standard inventory control practices that institutions use,” Dr. Kleinman agreed. “If the patient gets older rather than fresher blood, there is convincing evidence that he or she is not being put in jeopardy. There are always going to be exceptions to that rule, but there is much less reason for people to demand a fresher unit of blood.”

This news is particularly reassuring for blood banks or institutions with limited access to transfusion services or that aren’t big enough to sustain a constant influx of fresh blood. After all, in the real world, inventory control is very much influenced by outside forces, such as a drop in blood donations during the holidays, seasonal illnesses (cold and flu), or even just the size of the potential donor pool in a geographic area, Dr. Kleinman noted.

“Some institutions may be comfortable with a 35-day out-date and be able to avoid a supply shortage, while others may have to use the maximum of 42 days,” he said. “I think we’ve shown these centers that they can safely continue their current inventory management strategies.”

Dr. Roback explained that, in the Atlanta region, some smaller hospitals are served by the Emory University blood bank and, when those smaller outfits have blood that is getting close to expiration, they will send it back to the bank in exchange for fresher blood. “Here at Emory, we have so many more transfusion recipients than the smaller centers, so it is easier for us to use it and not let it go to waste,” he said. “I believe that blood supply management methods like this are going to keep operating the way they have been.”

Anecdotally, Dr. Hébert said he has seen a shift in the demand for fresh blood at his institution. “I think that the larger trials like RECESS and ABLE are having a big impact on practice,” he said. “Before, we were getting a demand for fresh blood from cardiac surgical programs, surgical trauma, and neonate programs. It seems that the RECESS and ABLE results have reduced that demand.”

Dr. Hébert also shared that when the ABLE trial was first proposed back in 1988, the head of the country’s blood banking service expressed some jitters about the potential results and their impact on blood-banking policies.

“I think after the trial results, everyone breathed a huge sigh of relief. These results are enormously reassuring to everyone – to our patients, our colleagues, the people who provide us with the blood, and the blood bankers.”—By Shalmali Pal


  1. Zimring JC. Fresh versus old blood: are there differences and do they matter? Hematology Am Soc Hematol Educ Program. 2013;2013:651-55.
  2. Koch CG, Li L, Sessler DI, et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med. 2008;358:1229-3.
  3. Edgren G, Kamper-Jorgensen M, Eloranta S, et al. Duration of red blood cell storage and survival of transfused patients (CME). Transfusion. 2010;50:1185-95.
  4. Solomon SB, Wang D, Sun J, et al. Mortality increases after massive exchange transfusion with older stored blood in canines with experimental pneumonia. Blood. 2013;121:1663-72.
  5. Neuman R, Hayek S, Rahman A, et al. Effects of storage-aged red blood cell transfusions on endothelial function in hospitalized patients. Transfusion. 2014 November 13. [Epub ahead of print]
  6. Alexander JT, El-Ali AM, Newman JL, et al. Red blood cells stored for increasing periods produce progressive impairments in nitric oxide-mediated vasodilation. Transfusion. 2013;53:2619-28.
  7. Lacroix J, Hebert PC, Fergusson DA, et al. Age of transfused blood in critically ill adults. N Engl J Med. 2015;372:1410-18.
  8. Steiner ME, Ness PM, Assmann DJ, et al. Effects of red-cell storage duration on patients undergoing cardiac surgery. N Engl J Med. 2015;372:1419-29.
  9. Dhabangi A, Ainomugisha B, Cserti-Gazdewich C, et al. Tissue oxygenation by transfusion in severe anemia with lactic acidosis (TOTAL): a prospective, randomized, non-inferiority trial of blood storage duration. Abstract #769. Presented at the 2015 ASH Annual Meeting, December 7, 2015; Orlando, Florida.
  10. Fergusson DA, Hébert P, Hogan DL. Effect of fresh red blood cell transfusions on clinical outcomes in premature, very low-birth-weight infants: the ARIPI randomized trial. JAMA. 2012;308:1443-51.
  11. Alexander PE, Barty R, Fei Y, et al. Transfusion of fresher versus older red blood cells to hospitalized patients: a systematic review and meta-analysis. Blood. 2015 December 1. [Epub ahead of print]

What is the Big Deal About Leukoreduction?

Leukoreduction is the process of removing white blood cells (WBCs) from blood before storage.1 Decades’ worth of research has shown that removal of leukocytes is associated with improved clinical outcomes, such as a reduction in the incidence and severity of febrile transfusion reactions, a reduction in the risk of cytomegalovirus transfusion, and a reduced risk of alloimmune platelet refractoriness.2

Leukocytes are considered a contaminant of other cellular blood components, including RBCs, and they have been recognized as a contributor to – if not the cause of – a number of transfusion-related adverse events, including immunologically mediated effects, infectious disease transmission, and reperfusion injury.

In other words, “leukocytes in the red cells have no clinical value,” Dr. Hébert stated. “But the potential harms attributed to them [in banked blood] are substantial. Generally speaking, the white cells go along for the ride, but no good comes of them being in the bag.”

Given the benefits of leukoreduction, 20 countries (including Canada, New Zealand, much of the European Union, the United Kingdom, the United Aram Emirates, and Qatar) have mandated universal leukocyte reduction (ULR) as a matter of public blood safety policy.

When Canada implemented its leukoreduction policy, Dr. Hébert and colleagues evaluated its role in decreasing post-operative mortality and infection. Compared with a control period, they noted that leukoreduction lowered the odds of death, but not serious nosocomial infections. However, it did cut the frequency of post-transfusion fevers.3

The U.S. FDA does not mandate ULR of RBCs, although the agency’s blood product advisory committee has twice called for a ULR protocol to be implemented.

However, the additional cost of leukocyte reduction has delayed implementation of the policy in the United States. “The differences in practice, belief, and opinion on how best to spend money for blood components has formed the basis for the controversy in the United States over ULR,” according to one set of experts. “In fact, the ULR debate has become so politicized that it has become one of the most divisive issues in the history of U.S. transfusion medicine.”4


  1. U.S. Department of Health & Human Services. “What is a blood transfusion?” Accessed January 9, 2015 from
  2. Blajchman MA. The clinical benefits of the leukoreduction of blood products. J Trauma. 2006;60(Suppl 6):S83-90.
  3. Hébert PC, et al. Clinical outcomes following institution of the Canadian universal leukoreduction program for red blood cell transfusions. JAMA. 2003;89:1941-9.
  4. Bassuni M, Blajchman M, Al-Moshary M. Why implement universal leukoreduction? Hematol Oncol Stem Cell Ther. 2008;1:106−123.