October 2011
Feature Story

Anne Paxton

From across a hospital room, the average person should have no problem determining the blood type of a unit of red blood cells, as it’s likely to be displayed prominently in large block letters. But another piece of information is in much smaller print: the unit’s expiration date. Backtrack 42 days from that, and in most cases you’ll arrive at the date when the blood was drawn from a donor, hence how long it has been on the shelf.

That bit of small print is garnering a lot of attention these days. The question is, if the expiration date hasn’t been reached, does it matter how old the unit is? It may—or it may not.

It’s been three years since a study conducted at the Cleveland Clinic was reported in the New England Journal of Medicine, then splashed across the mainstream press, startling many people with the finding that cardiac surgery patients who get fresher blood seemed to have better outcomes. The study put the phenomenon of a potentially harmful red blood cell “storage lesion” squarely on the map, igniting public worries that older blood was not as good.

But as critics of the Cleveland Clinic study weighed in, charging that it was flawed research, a string of questions has been left unanswered. Should clinicians and patients be concerned about how long a unit of red cells has been stored? Should the FDA-imposed storage limit of 42 days be shortened? Should certain categories of patients routinely receive fresher red cells? Or could commercial red cell treatments or storage techniques limit the storage lesion’s potentially negative effects?

“We do know that older red cells have decreased efficacy in oxygen delivery. Laboratory, animal, and human data indicate that a 42-day-old cell has decreased capacity to deliver oxygen and increase oxygen utilization compared with a fresh cell,” says Philip Spinella, MD, pediatric intensivist and director of the Pediatric Critical Care Translational Research program at Washington University School of Medicine, St. Louis. “We don’t know if that’s clinically important for people receiving transfusions, but some of us are concerned.”

According to a survey he conducted, 66 percent of blood bank directors at children’s hospitals in the U.S. and Canada are concerned about giving old blood, but 60 percent also say there is not enough evidence to change practice.

On the other side are experts like James AuBuchon, MD, president and CEO of the Puget Sound Blood Center in Seattle, who says, “The way we store blood today is not optimal, but we don’t have anything we can do other than shorten the storage period, and we don’t know whether the breakpoint between ‘okay’ and ‘not so good’ is seven, 10, 14, 21, or 35 days. Different groups have proposed arbitrary cutpoints, but there is no biochemical reason to think it happens at the stroke of midnight on a particular day, or happens the same way for every unit of blood. And we don’t have good human clinical data.”

At the moment, the debate is truly in a state of equipoise, Dr. AuBuchon says, as everyone awaits the results of several prospective randomized trials and ancillary studies, still some three years off. But many might agree with Paul M. Ness, MD, director of transfusion medicine and professor of pathology, medicine, and oncology at Johns Hopkins University, who writes in a recent editorial that “The question of whether or not older blood is less safe is the most critical issue facing transfusion medicine.”

Under FDA regulations, the storage limit for whole blood is 21 days, but red blood cells are granted a longer shelf life. Thirty-five days was the red cell storage limit mandated by the FDA until the mid-1980s, when additive solutions were developed that could be mixed with red blood cells after they were prepared from whole blood. “The additive allowed the cells sufficient metabolic substrates to continue their biochemical functions for a longer period of time,” Dr. AuBuchon explains. With the use of additive solutions, the FDA allowed red blood cells to be stored for up to 42 days.

But it’s established that red cells change during storage, bringing a series of biochemical, metabolic, structural, inflammatory, and physiologic alterations to the cells and their supernate. These include decreased 2,3-DPG, increased rigidity, a high PA I (plasminogen activator inhibitor) level, high CD40 ligand if not leukoreduced, nitric oxide depletion, an increase in interleukin-10, and a decrease in tumor necrosis factor-α. And, in the view of many scientists, the test of red cell survival and recovery that the FDA uses as a standard is a crude measure of red cells’ effectiveness in performing their key task: oxygenating tissues.

To meet the FDA standard, “The red blood cells are collected from a healthy donor, radio-labeled with an appropriate probe, and then reinfused to the same donor. You can store the red cells under conditions you want to test prior to reinfusion,” explains Zbigniew“Ziggy”M. Szczepiorkowski, MD, PhD, associate professor of pathology and of medicine and medical director of the transfusion medicine service at Dartmouth-Hitchcock Medical Center.

“You have to show that 75 percent of the cells survive after 24 hours.” Any new additive solution has to meet that standard. “But the standard only looks at one facet of the problem,” Dr. Szczepiorkowski says. “Does it mean the cells are functional? No, it doesn’t. Does it mean they are good for the patient? No. It just means they’re circulating and therefore not significantly damaged so as to be taken up by the spleen or liver.”

When the Cleveland Clinic study came out in 2008, its findings landed on the front pages of the Wall Street Journal and the Los Angeles Times. In the study, “Duration of red-cell storage and complications after cardiac surgery”(Koch CG, et al. N Engl J Med. 2008; 358:1229–1239), 2,872 patients received blood that had been stored for no more than 14 days and 3,130 patients received blood stored for more than 14 days. Use of red cells stored for more than two weeks was associated with a significantly increased risk of postoperative complications and a reduction in short- and long-term survival. The higher rates of pulmonary and multiorgan failure are one possible explanation for the increased mortality.

The Cleveland Clinic study had problems in that it was a retrospective analysis of data, says Jerome L. Gottschall, MD, vice chair of the CAP Transfusion Medicine Resource Committee and senior medical director of the BloodCenter of Wisconsin, Milwaukee. “But it was important because it was published in the New England Journal, and therefore the issue caught the attention of a wider range of the medical community.”

“It woke everybody up,” agrees Neil Blumberg, MD, director of the blood bank, Strong Memorial Hospital, Rochester, NY. “This paper served a very singular, useful function, even considering the limitations of this study, in calling attention to an issue we haven’t resolved and we don’t know the answers to.” He has been trying to convince clinicians and other blood bankers for some time that transfusions are considerably more toxic than was generally thought and are a major contributor to morbidity and mortality in patients with life-threatening illnesses.

But blood bankers like Dr. AuBuchon take a different view. “The Cleveland Clinic study was confounded by the fact that the cardiac surgery patients who tended to get older blood also tended to get more units of blood. And that same research group had shown from the same data set two years earlier that the more blood you get, the worse your mortality risk.” So from this study’s data, he says, it’s not possible to say whether stored blood really puts one at increased risk, or whether some categories of patients are more vulnerable.

These flaws and others are so serious that, from Dr. Szczepiorkowski’s perspective, “I’d say the study should never have been published. The data presented in the study is unadjusted for a number of variables; you can identify 10 to 15 different variables that are not comparable in different groups.” Dr. AuBuchon and Dr. Szczepiorkowski are part of the BEST (Biomedical Excellence for Safer Transfusion) Collaborative of manufacturers and blood bankers, which submitted a letter to the New England Journal editor objecting to the Cleveland Clinic study. “The letter was not published, but since then a number of studies have shown the differences between old and new blood were not there,” Dr. Szczepiorkowski says.

Despite the unresolved controversy, the study provoked enough alarm among clinicians that “we heard some people saying ‘I want only fresh blood,’” Dr. Szczepiorkowski says. In fact, the fear at the time was that there would be no funding for prospective trials to settle the issue. “But fortunately the funding happened and we have good studies coming along.”

In the U.S., the National Institutes of Health and the Department of Defense began sponsoring an array of research projects over the past two years. The National Heart, Lung and Blood Institute (NHLBI) is funding nine studies, one of them a large, prospective, randomized trial, that will help answer the questions that have been raised about storage of red blood cells. In addition, several prospective trials are planned in Canada. (See “Red cell storage studies on deck or underway”)

“I think the trials will cover several important areas and the other studies will be laboratory-based,” says Dan Waxman, MD, president of America’s Blood Centers and executive vice president and chief medical officer of Indiana Blood Center, Indianapolis. “The fact that we’re finally getting this NHLBI funding, $3.9 million per year over four years, is a good start to looking at this scientifically.”

Thanks to the research that is going on today, what happens to blood during storage will be better understood, agrees Dr. Gottschall. “And if the age of blood is an important mechanism that may be causing harm to some patients, these studies will help us understand what those mechanisms are on a basic science level and on a clinical level,” he says.

It’s clear that there are many different elements creating the storage lesion, says Jordan A. Weinberg, MD, associate professor of surgery, University of Tennessee Health Science Center in Memphis, whose research on trauma patients has also shown that transfusions of older blood tend to lead to worse outcomes. “If you put a bag of red cells on a shelf for six weeks, you can look at a multitude of changes. The question is, which changes have clinical implications?”

The hypothesis being tested in Dr. Weinberg’s current NHLBI-funded study, one of the few in vivo studies, is that relatively older blood impedes microvascular perfusion. “We’re trying to see if we can actually demonstrate this by capturing measurable changes in patients’ microcirculation. Other arms of the same study will look at nitric oxide signaling as related to red cell storage age, and the effects of transfusion of relatively older red cells on immune function.”

The possibility that some blood donors’ blood may be more susceptible to the storage lesion than others’ blood has not been well explored, Dr. AuBuchon says. “We have several normal donors we’ve encountered over the years whose red cells do not store as well as the average person’s. But we don’t know which donors are more likely to cause these problems, and we don’t know which patient groups are more susceptible.”

Dr. Weinberg says his research team will be looking at this issue as part of its study. “It’s not well understood. But when you donate blood you’re giving a mix of red cells. Some are quite fresh and others are close to expiring.” At least theoretically, he says, “It might be the age of the donor, that older people have a higher proportion of older cells circulating in their bodies.”

Others are skeptical. Merlyn H. Sayers, MBBCH, PhD, chief executive officer of Carter BloodCare in Bedford, Tex., believes no such effects exist or are highly unlikely. “If there were categories of regular donors whose blood did not store properly, then they would have been identified over the years that we’ve been doing blood banking and transfusion.”

While researchers are enthusiastic about the NHLBI’s commitment to funding studies on the issue, there are other issues suggesting that it can be difficult to arrive at a scientific consensus. John Roback, MD, PhD, associate professor and director, Center for Transfusion and Cellular Therapies, Emory University Hospital, for example, is leading a study investigating mechanisms by which older RBCs may be able to perturb nitric oxide signaling, possibly leading to increased morbidity and mortality. From discussions at a recent working group meeting, which brought all the RBC storage researchers together at the NHLBI, it was apparent “there’s not even a clear, generally accepted understanding of how older RBCs might adversely affect nitric oxide signaling. Some say it’s less S-nitrosylated hemoglobin after storage; others believe it’s the fragmentation of RBC membranes that occurs during storage and produces microparticles, while others believe it’s something intrinsic to the red cell. So even a critical part of the analysis like that is still up in the air.”

The basic research is interesting and well-done work, Dr. AuBuchon says. “We’re learning a lot about the basic physiology of red blood cells and how that physiology is perturbed during storage. But the connection between the basic science and the bedside really has to be made.” In sum, he says, “I don’t know how this play is going to end. No one is thinking older red cells are going to be better for patients, but yet we don’t have the data to know whether or not older cells are harmful.”

Whatever the research eventually shows, it’s also going to be enormously complicated to turn it into solid conclusions or use it to set new policies, Dr. Szczepiorkowski believes. “The results of the studies may be completely inconclusive. If you pick a number as a cutoff, someone could always say, if you improve storage conditions, or reverse the process of red cells’ getting older in the bag,” then that number would likely be different. With new technologies for storage, he suggests, “maybe the issue of old blood versus new blood will be a moot point.”

While the studies proceed, there are still plenty of contexts in which transfusion decisions have to be made without conclusive data, often based on anecdotal experience, or what is sometimes called “casual empiricism.” U.S. Army war theatres are a good example.

Dr. Spinella, in fact, became interested in the red cell storage lesion when he was stationed at the largest combat hospital in Iraq in 2004 and 2005. “As a result of the lack of platelets available then, our standard approach was to use fresh whole blood for combat casualties requiring platelets. But we very quickly noted that those patients receiving fresh whole blood were being resuscitated much more efficiently than when older red cells were used. Clearly the addition of platelets could have been making the difference, but we also asked ourselves whether the lack of a storage lesion in red cells, platelets, and plasma was conferring a potential benefit.”

On the flip side, the Army was transfusing stored components to these casualties that were dramatically old relative to what’s given in the U.S.—a median age of 33 days as compared with the median U.S. age of about 17 to 21 days. The most common age of RBCs the Army transfused during this time period was 42 days, the day of expiration. “We thought the fact that these patients in coagulopathy and shock were getting old red cells could be detrimental to them as well.”

After Dr. Spinella and other U.S. Army military medical leaders briefed the Army surgeon general on the available retrospective evidence and the biologic plausibility that red cell age was a problem, Army policy was changed. “Our briefings convinced him to send a second shipment per week of RBCs. That’s a huge, major policy change, and we’ve seen a major reduction in red cell age and, anecdotally, an increase in survival of massive transfusion patients.”

Dr. Spinella, who is now conducting several of the mechanistic NHLBI and DOD studies to determine what features of stored red blood cells may be harmful, is the first to admit there are many confounders to anecdotal evidence, and he is eager to see the results of the prospective randomized trials. “But while there’s no randomized evidence that old red blood cells are harmful, conversely there’s no prospective evidence that older red cells are helpful,” he says. “That assumption is based on red cell survival and recovery studies which do not at all determine efficacy.” For example, “We don’t even know if older red cells immediately upon transfusion can get into the microvasculature. They are non-deformable after 14 days, so how are they going to get into capillaries?” These older cells over time retain their deformability, he says, but critically ill patients with shock and coagulopathy need these conditions reversed immediately. Prolonging the time to resolution of shock and coagulopathy can increase the risk of poor outcomes.

When many current blood banking policies were put in place, he points out, there was less emphasis than there is now on evidence-based medicine. “We used to give whole blood in the ‘70s, then in the 1980s we changed to modified whole blood with platelets taken out, then we went to components, without any evidence that there was at least equal efficacy. We just took it for granted that if it can circulate, then it must be okay.”

Because of the high cost of clinical trials, Dr. Spinella hopes the evidence collected from translational research examining efficacy and safety of RBCs according to storage age can inform future clinical trials. “We need research into blood storage solutions to extend functional shelf life, not just to extend shelf life. Efficacy can be defined in two different ways—by clinical trials which take millions of dollars to do, or by directly looking at red cells’ ability to improve oxygen utilization at the cellular level.” Indirect ways of looking at red cell efficacy, for example by using dynamic Near Infrared Spectroscopy, has the potential to be informative, he says.

In a survey he conducted on policy for age of red cells at children’s hospitals, “we analyzed the transfusions by patient illness and found that the only children who get fresh cells preferentially are neonatal cardiac surgery patients. After that, no matter how sick they are, 85 percent of the hospitals will basically give those kids the oldest red cells in the inventory. They do that because inventory management is very important. But what some of us are concerned about—and what still needs to be studied in a prospective, randomized way—is whether, by not giving those patients fresh red cells, you are not allowing for improved outcomes.”

Of course, keeping enough inventory on the shelves is a worry when there is any discussion of possibly reducing the shelf life of red blood cells. Operating on a first in/first out basis allows blood banks to get maximum use from their inventory, but it means that each time a unit is ordered, the oldest unit is issued first. Some blood centers and hospitals make an exception for neonatal patients undergoing open heart surgery; the protocol in those cases may call for blood that is three to seven days old, ABC’s Dr. Waxman says. “They do that because as blood is stored and ages, potassium is released from the red cells. So for this patient group they would want fresher blood.”

But every time fresher blood goes to one group—say, to ICU patients—everyone else gets older blood. That’s one argument that Dr. Ness has been able to use in addressing clinicians’ concerns about older blood at Johns Hopkins. “We’ve had meetings with the cardiac surgeons and have gone over the data with our anesthesiologists to discuss why changing practices would be a bad idea, right now, because there are some patients who do need fresh blood and if some people get it who don’t need it, there won’t be enough left.”

Most large hospitals carry a combination of newly drawn fresh blood and RBCs that were first sent to smaller hospitals and, when they weren’t used relatively quickly, were rotated. “If a blood center collects 1,000 units a day, probably 60 to 70 percent of that is going to go to large hospitals and 30 to 40 percent to smaller, rural hospitals,” Dr. AuBuchon explains. “Those smaller hospitals need to have adequate inventory to cover not only daily use but also emergencies, but of course they usually don’t use all that blood, so that comes back into the system and is circulated back to the larger hospitals.”

The U.S. tends to have a pretty close match—within a couple of percentage points—between the blood collected and the blood transfused, says Emory’s Dr. Roback. “But you have to understand that it’s a ‘just in time’ delivery system. In September in Atlanta, we’ll probably have enough blood on the shelf. But if you’re in New York City and just got hit by a hurricane, or in Atlanta after getting hit by a snowstorm, you can quickly find yourself with a much smaller buffer zone in terms of blood supply. If you don’t have 150 units on the shelf to tide you over for a couple of days, but instead have only 70 or fewer, there would be a significant issue.”

If the shelf life of RBCs were radically shortened to 28 or 21 days, it would make life difficult for blood centers, Dr. Waxman says. “For years we practiced with 35 days storage of red cell units and most of us could do that; some centers have continued to use CPDA-1 [citrate-phosphate-dextrose-adenine-1] 35-day units, mainly for pediatrics. But I’ve seen projections if we cut storage to 28 or 21 days, and that could have a dramatic effect on the blood supply.”

Research by Dr. Sayers of Carter BloodCare includes a model of likely outcomes if shelf life is shortened from 42 to 28 days, a move he suggests could have serious consequences. In a study of 2010 inventory that he has submitted for publication, “we calculated the ages of units when they were sent to the hospital and found that, if red cell life were reduced from 42 to 28 days, more than 10 percent of O Negative cells we currently send to hospitals would be too old.”

Hospitals’ use of O Negative blood far exceeds the percentage of the donor base who have that blood type, he notes. “We’d have to significantly increase recruitment and collections in that category, a group of individuals who are very, very difficult to recruit.” Without additional donor recruitment, the ability of blood programs to cover hospital needs would be compromised, he says.

“The bottom line is that patients genuinely needing transfusion die if they don’t get the transfusion, so one cannot look at these studies and ignore the risk of jeopardizing availability, because blood is life-saving,” Dr. Sayers says. But he agrees that different policies for some patients might be warranted. “We need to find out if there are categories of transfusion recipients for whom special care should be taken by providing fresher units.”

America’s Blood Centers is addressing this need with a new program called AIM (Appropriate Inventory Management), now being rolled out to member centers. “The first version tracks the age of the blood we are distributing and the percentages of ABO-Rh,” says Dr. Waxman. “Our next version will be looking at blood usage on the hospital side for different types of patients. So we hope with this program we’ll learn much more about inventory management and have an impact on how we are storing, distributing, and utilizing blood.”

Emerging technologies are raising hopes that new additive solutions can further extend RBC shelf life by inhibiting and potentially reversing the red cell storage lesion. “In this country, pretty much since the 1980s, we have had the same storage solution for red cells, and we really haven’t advanced the field in terms of storing,” says Dr. Gottschall. “We obviously need to do that.”

At the recent working group meeting of the NHLBI-funded researchers, two companies presented research on potential products, Dr. AuBuchon says. “One is working on an anaerobic storage technique because they think oxidative damage really can explain a lot of what all these other laboratories are seeing. They deplete the oxygen that would otherwise be present during storage, and this prevents a lot of the oxidative damage that might otherwise occur. Another company offered an alkaline additive solution that would maintain red cells for a longer period of time, allowing them to meet FDA criteria for up to 56 days and thus, at least theoretically, offer a ‘better’ or ‘younger’ red cell at shorter storage times.”

For the blood products industry, Dr. Szczepiorkowski points out, millions of dollars hang on the clinical studies now in progress. He tends to think the studies will not show differences in patient outcomes for older blood, but if they do find, say, that a significant difference occurs in blood after day 14, and that, as a secondary outcome, people die because of older blood, “then there would be no interest on the part of industry in making better bags or solutions, because they would have to repeat the study, and that’s an investment many companies will not make.”

In part, this reluctance stems from U.S. regulatory policy, which has been extremely conservative when it comes to licensing products that go beyond just the basic human component, Dr. Waxman says. “That’s why I think the U.S. has limited access to pathogen reduction technology. So if someone has a way to inject nitric oxide to alleviate the storage lesion, for example, it will have to go through many studies to prove that the method does not create new adverse effects.”

Regulatory issues aside, many experts emphasize that more conservative approaches to transfusion are part and parcel of any measures to address the effect of storage on RBCs. “My hope is that the approach to RBC storage for the transfusion medicine community includes three things,” says Allan Doctor, MD, chief of pediatric clinical care at St. Louis (Mo.) Children’s Hospital and associate professor of pediatrics and biochemistry and molecular biophysics, Washington University School of Medicine. “First, a more thoughtful approach to indications for transfusion, because we are probably over-transfusing in general. Second, an effort toward blood conservation, meaning limiting blood draws for lab tests, reinfusing shed blood in the OR, and doing things to optimize hematopoiesis, such as erythropoietin to stimulate RBC production. And the third strategy is to use fresher blood. But if you don’t do the first two things, No.3 is a problem.”

Dr. Blumberg strongly agrees that transfusion guidelines should continue to be tightened, but he emphasizes the need to weigh competing interests within the health care system. “There are two sides here. One is the blood center and transfusion service trying to maintain a precious and not easily replaceable source of care. And the other is bedside clinicians and patients.” In his view, it’s the latter group that should take priority. “If there’s a way to save lives by screwing up blood bank inventory, then we should do it,” he contends.

“We’ve had much bigger adjustments. We’ve dealt with the AIDS epidemic, with universal leukoreduction, with the need to get cytomegalovirus-negative blood to some patients. This is going to be a piece of cake compared to the real issues we face in transfusion medicine, which are making transfusion safer and getting our clinical colleagues to transfuse more conservatively.”

Since the Cleveland Clinic study, much important research has been initiated that will clarify such policy choices. “It still remains entirely unclear whether the storage lesion is a phenomenon that has real clinical relevance,” Dr. Weinberg says. “But I think the work we are doing will shed a lot of light on what’s going on. It still may not answer the question ultimately as to whether giving relatively older blood is clinically harmful or not, or to which patients. But it will get us closer.”

Related Links

• Red cell storage studies on deck or underway