When it comes to administering red-blood-cell transfusions, scientific evidence has shown that more isn’t better, but new study findings suggest that fresher may be, at least for critically ill patients. The question now is whether the storage age of red blood cells presents a potential patient safety risk and, if so, what hospitals and practitioners should do about it.
Research conducted at the Cleveland Clinic not only jump-started that discussion but catapulted it into the public arena. The study found adult cardiac surgery patients who received red blood cells stored for more than 14 days fared significantly worse than those who received younger red blood cells (Koch CG, et al. N Engl J Med. 2008; 358:1229–1239).
“We thought the issue of fresh blood has been dead for 25 years but it’s back,” says Jim MacPherson, CEO of America’s Blood Centers. After giving transfusions for 60 years, he adds, “we still have unanswered questions about whether or not to target blood cells of a certain age to certain kinds of patients.”
In the New England Journal study published March 20, researchers did a retrospective review of 6,002 patients with a median age of 70 years who received red-blood-cell transfusions during coronary-artery bypass grafting, heart-valve surgery, or both at the Cleveland Clinic from June 30, 1998 to Jan. 30, 2006. The researchers drew the line for defining older versus newer blood at 14 days because it happened to cut the historical population into two equal groups.
Cardiac surgery patients administered red blood cells stored for more than 14 days were more likely to die in the hospital, be intubated beyond 72 hours, and develop renal failure, multiorgan failure, and a composite of serious complications including the aforementioned. In addition, one-year mortality was significantly higher for patients who received older blood (11 percent versus 7.4 percent).
Renal failure and pulmonary complications (pneumonia and a long time on ventilator) showed the strongest association with older red blood cells, says Gerald Hoeltge, MD, a study coauthor and pathologist in the section of transfusion medicine at Cleveland Clinic.
The two study groups were composed of cardiac surgery patients who received either all older blood or all newer. Most patients in the study received three or fewer units of red blood cells. The median for the two groups was the same—two units, Dr. Hoeltge says.
The Cleveland Clinic began using all leukoreduced blood in 2002, so the study spans the timeframe in which that occurred. And interestingly, the older-blood group received more leukoreduced red blood cells than the group receiving newer blood.
Anesthesiologist Aryeh Shander, MD, says that while the NEJM article showed that patients receiving blood older than 14 days had higher all-cause mortality, it can be seen in some graphs that the effects started even earlier. “So we are not certain whether 14 days [of storage] should be the cutoff; it could be a lot earlier,” says Dr. Shander, chief of the Department of Anesthesiology, Critical Care and Hyperbaric Medicine at Englewood (NJ) Hospital and Medical Center and clinical professor of anesthesiology, medicine, and surgery at Mount Sinai School of Medicine, New York, NY.
The Food and Drug Administration allows red cells to be stored under refrigeration for 21 days when the anticoagulant preservative system used is ACD, CPD, or CP2D; for 35 days when CPDA-1 is used; and for 42 days when solutions such as AS-1, AS-3, or AS-5 have been added. Irradiated red blood cells have an expiration date of 28 days from the date of irradiation. If a unit of red blood cells has an expiration date less than 28 days prior to irradiation, the shorter expiration date must be used (AABB Standards for Blood Banks and Transfusion Services, ed. 25).
The NEJM-reported research comes with all the usual caveats associated with a retrospective study. The researchers controlled for patients’ comorbidities but, as Dr. Hoeltge says, there can be all sorts of “confounding variables that one hasn’t controlled for in such studies.”
Even so, he adds, “It’s a very important study because it’s the largest retrospective study of its nature and shows a very strong correlation of older red-blood-cell storage with outcomes.”
Critics of the study point to differences between the two groups that could have affected the patients’ outcomes. For example, says pathologist Ira A. Shulman, MD, a greater percentage of patients who received older blood had abnormal left ventricular function, mitral regurgitation, and/or peripheral vascular disease.
Also, transfusion of group O red-cell units was greatly over-represented among patients who received younger blood, says Dr. Shulman, director of laboratories and pathology at Los Angeles County–University of Southern California Medical Center and editor of an online discussion forum on blood sponsored by the California Blood Bank Society (cbbsweb.org).
Pathologist Neil Blumberg, MD, of Strong Memorial Hospital, Rochester, NY, said in an online discussion about the NEJM-reported study that based on his experience (as yet unpublished data), group O patients may be at lower risk of thrombosis and multiorgan failure syndromes.
Another “red flag that uncontrolled confounders are at play” [in the Cleveland Clinic study]: “The recipients of the older blood received more leukocyte-reduced transfusions than did the recipients of younger blood ...,” Dr. Blumberg said in his online comments. He pointed out in the same note that randomized trial data from numerous studies demonstrate that mortality rates of cardiac surgery patients who receive exclusively leukocyte-reduced blood products are lower by 30 to 50 percent.
Dr. Shulman says that based on his reading of the literature, “blood products that are leukocyte-reduced shortly after being collected undergo some ‘storage lesion’ changes at a slower rate than similarly aged non-leukocyte-reduced blood products. Irradiation of cellular products can also accelerate certain blood product storage lesion changes.” Since some of the patients in the Cleveland Clinic study received both non-leukocyte-reduced and leukocyte-reduced red blood cells and since the authors did not comment on the use of irradiated blood products, he suggests that a subgroup analysis of patients who received only younger versus older, nonirradiated leukocyte-reduced blood products might provide additional insight into the clinical importance of the product-storage duration.
The Cleveland Clinic study isn’t the only recently published research linking negative clinical outcomes to older red-cell transfusions. Critically ill combat casualties in shock appeared to do better when they received increased amounts of plasma as opposed to increased amounts of red blood cells with a median age of 33 days, according to a retrospective study (Spinella PC, et al. J Trauma. 2008; 64(2Suppl): S69–77).
Phillip Spinella, MD, and colleagues wrote another paper that reviews and compares the risks of using fresh whole blood with using old red cells in patients with severe traumatic injury. In this review, they describe the biologic plausibility and consistency of the data that support the concept that the transfusion of old red blood cells increases the risk of multiorgan failure and death in critically ill patients who receive old red blood cells (Spinella PC, et al. Crit Care Med. 2007; 35:2576–2581).
“There are some of us who believe that one of the many reasons why fresh whole blood is preferable is that it contains functional red cells compared to the use of stored older red cells,” says Dr. Spinella, the study’s lead author and a pediatric intensivist at Connecticut Children’s Medical Center, Hartford. During his deployment with the U.S. Army to Baghdad in 2004 and 2005, Dr. Spinella helped develop and advance the 31st Combat Support Hospitals Fresh Warm Whole Blood Program in addition to caring for ill children and adults with traumatic injuries and medical illnesses.
The NEJM article on the Cleveland Clinic study says that “although the mechanism linking adverse outcomes with increased duration of red-cell storage remains unclear, several factors may contribute.” On the list of suspects are “decreased [red-cell] deformability, which can impede microvascular flow; depletion of 2,3-diphosphoglycerate, which shifts the oxyhemoglobin dissociation curve to the left and reduces oxygen delivery; increased adhesiveness and aggregability; reduction in the concentrations of nitric oxide and adenosine triphosphate; and accumulation of proinflammatory bioactive substances.”
The article doesn’t list microvesicles as a potential explanation for the impact of the older red cells, Dr. Hoeltge says. But microvesicles, which he defines as “bits of cell membrane that bud off from red cells during storage, “add to the number of preformed elements in the bloodstream and can get in the capillaries. In addition,” he says, “having microvesicles in the blood indicates the red cells have a smaller surface than when first collected.”
For the patient who is bleeding to death, Dr. Spinella says, red-blood-cell transfusion is lifesaving because it at least increases intravascular volume and cardiac output. In his view, however, if the patient survives that initial hemorrhage, the older red blood cells potentially increase the person’s risk of developing multiorgan failure due to inflammatory, immune modulation and vasoregulatory dysfunction. Recent data suggest that via nitric-oxide-related mechanisms, when older red cells reach tissues lower in oxygen they vasoconstrict arteries in those tissue beds. “This may be one of the unintended consequences of transfusing older red blood cells to critically ill patients, which actually increase their risk of morbidity and mortality,” Dr. Spinella says.
“Many people find that concept difficult to believe because infusion of the older red cells initially causes the person’s hemoglobin and blood pressure to go up,” he says. “But the negative effects are invisible and may not occur until two to three days later.”
What about the potential impact of leukoreduced blood on patient outcomes? “You can fix one problem and still have nine others with older red cells,” Dr. Spinella says. The leukoreduction may improve the inflammatory issue, he agrees, but he points out that the NEJM-reported study showed that the cohort that did the worst got more leukoreduced blood.
But “after 14 days of storage, red blood cells become non-deformable where they can’t bend to get through capillaries, which are five microns whereas red cells are eight microns,” Dr. Spinella says. “That’s why sickle cell anemia patients have ischemia problems—the red cells can’t bend to get through the capillaries.” A few studies referred to in the article by Spinella, et al., in Critical Care Medicine show that red cells stored more than 14 days have decreased deformability acutely after being transfused. “Potentially over time they may regain that ability, but when a person is critically ill, the clock is running; you have to increase oxygen delivery now,” Dr. Spinella says.
Duke University researcher and pulmonologist Timothy McMahon, MD, PhD, thinks the deleterious effect of stored blood in some populations is probably due to a “conglomeration of things,” including depletion of nitric oxide, or NO, a gaseous free radical that causes blood vessels to dilate so they can deliver oxygen. NO also affects the ability of red cells to change shape and get through microvessels.
In a study published last October in the Proceedings of the National Academy of Sciences, Dr. McMahon and colleagues showed that red blood cells release a form of NO that relaxes blood vessels, which the researchers believe could account for how the red cells participate in regulating their own blood flow (Bennett-Guerrero E, et al. Proc Natl Acad Sci USA. 2007;104:17063–17068. Epub Oct. 11, 2007).
“We and others have shown that red blood cells have this capacity,” Dr. McMahon says. “What’s new in the October 2007 paper is that we found that when one stores red blood cells, [those cells] lose bioactivity and the active form of NO that serves that activity.” The bioactive form of NO is bound to sulfur groups and referred to as (S)NO.
The researchers found nitric oxide became and remained very low just three hours after blood was collected. Thus, the results of the study may not explain entirely why older red blood cells are worse than fresher ones “unless that early loss of something takes several days or weeks to fully produce a functional deficit,” Dr. McMahon says.
Instead, Dr. McMahon thinks the nitric oxide study may shed light more directly on findings that patients who get transfused more don’t do as well as those who are transfused less—or findings that transfused patients fare worse than those who don’t receive the treatment, “all else being equal.”
The good news about the nitric oxide depletion in stored blood is that it may be easily correctable. Another Duke-based study published in the same issue of the Proceedings of the National Academy of Sciences showed that it’s possible to restore NO to red cells and get the bioactive (S)NO form by adding a simple NO solution made in the laboratory, Dr. McMahon says (Reynolds JD, et al. 2007;104: 17058–17062. Epub Oct. 11, 2007). But more research is required before putting NO into blood because NO can be toxic. For example, in sepsis, which is caused by infection, there is stimulation of the NO-producing enzyme in cells, including white blood cells, he says. And “the hypotension you see in sepsis is related in part to NO overproduction.”
The U.S. Department of Health and Human Services is stepping up to the plate in the blood safety debate that many believe can be resolved only with prospective studies. At press time, the HHS Advisory Committee on Blood Safety and Availability had the issue on its May 29–30 meeting agenda. Jerry Holmberg, PhD, executive secretary of the committee, told CAP TODAY: “The report coming out of Cleveland Clinic but also studies from Duke University have raised concerns regarding adverse outcome of older blood. The Duke study came out first, and the credibility of the 6,000 cases reviewed at the Cleveland Clinic carries a lot of weight. I think it’s prudent for the American people that we ask more questions about the issues of safety and efficacy of older blood, which is currently within the FDA licensed shelf life.”
To promote prospective studies that can answer definitively whether fresher blood is better, the National Institutes of Health has issued a request for applications titled “Immunomodulatory, Inflammatory, and Vasoregulatory Properties of Transfused Red Blood Cell Units as a Function of Preparation and Storage” (RFA-HL-08-005). The agency has $3.5 million committed for fiscal 2009 to support an estimated five to eight research projects.
Simone Glynn, MD, MPH, branch chief of the Transfusion Medicine and Cellular Therapeutics Branch, Division of Blood Diseases and Resources, National Heart, Lung, and Blood Institute, says the goal of the RFA is to “foster novel basic and translational research to identify the molecular and cellular changes that occur during red-blood-cell unit collection, manipulation, and storage”—and to evaluate the impact of those storage lesions on people who receive the transfusion. “We’d like to evaluate the effect of the storage lesion elements on the blood vessel wall, host cells, and tissue oxygenation,” she says.
Basic research is the focus of the RFA, Dr. Glynn says, though the NIH is interested in looking at early physiological work, which could include humans in the clinic. However, the NHLBI’s Transfusion Medicine and Hemostasis Clinical Trial Network is evaluating the possibility of developing a randomized, blinded phase three clinical trial to study the effect of older versus newer blood. A potential design could be to have one arm receiving older red-blood-cell units that represent standard of care in many hospitals, and the other arm, younger red-blood-cell units, Dr. Glynn says.
“Our thinking here at NHLBI is that we essentially need phase three clinical trials in different patient populations to evaluate whether the age of transfused red cells” has an impact. “We don’t know that yet,” says Dr. Glynn, who adds there are multiple instances in public health where a prospective clinical trial disproved findings from observational studies suggesting that a treatment benefited people. “An example is the use of estrogen and progesterone hormone replacement therapy,” she says.
Los Angeles County-USC Medical Center’s Dr. Shulman believes that a randomized, controlled clinical trial in cardiac surgery and other patient groups needs to take into account the various ways in which blood is collected, and whether and when it’s irradiated, leukocyte-reduced, or both. “The trial should be designed to determine, if possible, at what storage age red-cell units become more deleterious than helpful to patient outcomes,” he says.
In response to the NIH’s RFA, Dr. McMahon and colleagues at Duke University are submitting a research proposal to look at vasoactivity in stored red blood cells.
In research already underway, he and others at Duke are looking at how simple changes in the way in which blood is drawn and processed early on might be able to prevent some of the biochemical changes seen in stored blood. He notes that “normal red cells function by constantly going through a cycle of oxygenation and deoxygenation and NO loading and NO unloading, and when we take the red cells out of that environment and put them in a static environment at low oxygen levels, we obviously interfere with that cycle.”
In Dr. McMahon’s view, research is needed to come up with some sort of biomarker or testing that can identify patients who will and won’t benefit from a transfusion. “It might be parameters that tell you about the patient’s blood oxygen delivery capacity and red-cell vasoactivity, and the red blood cells’ deformability.” He thinks, in fact, that the next $3 million in research monies might be best spent on identifying such testing rather than on (or integrated within) clinical trials looking at the impact of fresher versus older red-blood-cell transfusions.
Englewood Hospital’s Dr. Shander questions the need for prospective, randomized trials to settle the question of whether fresher blood is more beneficial to patients. He relays how his friend Jonathan Waters, MD, chief of anesthesia at Magee-Womens Hospital in Pittsburgh, to make a point about the impact of 42 days of refrigeration on stored blood, left a piece of raw steak in the refrigerator for that same length of time.
Stored blood and steak are both protein, Dr. Shander says, but a person would probably “think about 1,700 times before cooking and eating 42-day-old steak.” And while blood is sterile and probably not growing anything, it’s decaying as a protein for 42 days, he points out. “So why spend time arguing about whether the literature is equivocal on whether older blood is more harmful than fresher blood? There’s no question in anyone’s mind that the longer a protein decays, the less functional it is. And there are lots of other things in the blood—it’s basically a bag of immune complexes.”
“In general, the NIH and FDA will only accept data from randomized, controlled prospective trials and may reject observational data,” he says, “even if there are more than 1,500 publications out there that provide a strong signal on the risk of allogeneic RBCs. With our current incomplete knowledge, are patients going to accept being in the control group and receive older blood? One should also remember that avoidance of blood through known and effective blood conservation methods obviates this concern.”
If additional studies show a clear benefit of using fresher blood, or if people agree that existing retrospective study findings justify giving younger blood to critically ill patients, the blood banking community and hospitals will have a challenge on their hands.
In fact, based on the Cleveland Clinic study, Dr. Shulman says, “some hospital transfusion services are now under pressure by clinicians to limit the use of red-cell transfusions to units no older than 14 days.”
MacPherson of America’s Blood Centers reports that “some blood centers have been contacted about the study wanting to know if hospitals and physicians should be concerned, and a few surgeons have said they want younger blood for certain patients.”
“Logistically,” he says, “it would be nearly impossible to provide everyone with blood less than 14 days old without collecting and wasting far more blood than we currently do. However, if we knew which patient groups would benefit the most from fresher red blood cells, we might be able to work that out if the blood centers and hospitals had interactive computer software for better inventory management.”
For the past two decades, however, the nation’s blood inventory management system has been set up for 42 days of shelf life. Fresh blood is shipped to rural hospitals where it’s either used or held until it has about two weeks of shelf life remaining. The unused blood comes back and is sent to inner-city hospitals where it’s typically used before expiring, MacPherson says. “That approach keeps outdating to less than one percent of red cells.”
Dr. Spinella says reversing the pattern of sending older blood to inner-city hospitals would help. “In fact, if less-sick patients in the community hospital setting received older blood, they might not have an adverse effect if they aren’t in a hyper-inflammatory state and at high risk of infection,” he says.
MacPherson says the average shelf life of blood when it is transfused is not now known, but that most blood transfused is probably 20 days or less on average. “If blood is presumed good until 42 days, the feeling has been, why track it to find out when it was transfused?”
Dr. Spinella suggests that it become standard for all blood banks to record the storage age of all red blood cells transfused. “This would allow us to start to determine which patient groups appear to be at risk from the transfusion of old red blood cells, which would support the investment in prospective clinical trials in these populations,” he says. “Right now most blood banks don’t record this, which makes it very difficult for us to study it.”
One way to circumvent the issue of whether older blood may be harmful in many cases is to transfuse conservatively. In an editorial, “New Blood, Old Blood, or No Blood?” accompanying the NEJM article on the Cleveland Clinic study, John W. Adamson, MD, of the VA San Diego Healthcare System, says “blood management makes good economic sense” and, consistent with the evidence presented in the Cleveland Clinic study, “good medical sense.”
Englewood Hospital’s Dr. Shander agrees, noting that most transfusion recipients in the U.S. receive one or two units of blood. And it might be possible, he says, to reduce blood transfusions for that group by using modalities that adequately replenish hemoglobin and have less impact on the immune system. Englewood Hospital has, in fact, had a bloodless program in place since 1994 with very low morbidity and mortality. Patients receiving care there do not receive blood unless “at risk for losing life, organ, or limb,” Dr. Shander says. “Under the latter circumstances, we will give blood but then we do so gingerly in small amounts to try to overcome the risk period.”
Dr. Hoeltge at the Cleveland Clinic recommends avoiding blood transfusions whenever appropriate. He says physicians need to be comfortable managing the patient with a lower hematocrit than in the past. “Patients going to elective surgery need to have their preoperative anemia corrected before surgery.” Hospitals should also make use of intraoperative blood salvaging devices. “There’s no newer blood than the blood salvaged during surgery,” he says.
As Susan Roseff, MD, medical director of transfusion medicine at Virginia Commonwealth University Medical Center, Richmond, sums up: “We know that blood has deleterious effects, and maybe it’s true that older blood is not as good as fresher blood for certain populations....So we should restrict it as much as possible to ensure transfusion is necessary. That’s the lesson we should all learn and embrace.”
Karen Lusky is a writer in Brentwood, Tenn.