Universal leukoreduction-fix
  or folly?

October 2000
Cover Story

Anne Paxton

It could be the single biggest health innovation in transfusion medicine since the HIV and HCV tests. It could be a waste of hundreds of millions of dollars much better spent on other blood safety measures. But whether one sides with its proponents or its foes, universal leukocyte reduction , or ULR, will soon have taken hold at half of the nation’s blood centers-probably long before the controversy over its merits dies down.

Leukocyte reduction as a blood safety precaution has been used on select patients since the 1980s. But it was not until 1998 that some blood organizations began pressing to have all blood products subject to the leukocyte reduction filtering process.

Since then, the Food and Drug Administration has endorsed ULR-though it hasn’t formally mandated it, many European countries have mandated it, and the pressure to adopt it everywhere in the United States has steadily intensified. "The big problem is the ’U’ that goes in front of the ’LR,’" says Herbert Silver, MD, medical director of transfusion medicine at Hartford Hospital, Hartford, Conn.

Last spring, the American Red Cross announced that all of its service regions will be converted to ULR by the end of the year. Blood centers affiliated with America’s Blood Centers, which represent almost half of the blood suppliers, are making the call on the local level to use or not to use leukoreduced blood. In September, Rhode Island Blood Center, which is the only blood center in the state, became the first to provide 100 percent leukoreduced products to all the hospitals in the state. Overall, an estimated 40 percent of the nation’s blood supply is already being leukoreduced.

But the added expense of roughly 20 to 25 percent per unit, or much more by some estimates, makes opting for ULR a high-stakes choice. "What we’re talking about here is a cost increase to the health care system in the neighborhood of $300 million," says James MacPherson, chief executive officer of America’s Blood Centers, Washington, DC.

To date, the College has taken no position on ULR. But many specialists in transfusion medicine fear that adoption of ULR means an important clinical decision is being taken away from physicians and imposed on them by blood centers. "I believe the decision whether or not to provide a cellular component as leukocyte-reduced is the practice of medicine," says James P. AuBuchon, MD, professor of pathology and medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH. "Therefore this choice should be made by physicians responsible for the patient’s care."

On the other hand, Edward L. Snyder, MD, professor of laboratory medicine, Yale University School of Medicine, and director of blood transfusion services at Yale-New Haven Hospital in Connecticut, argues that perhaps physicians need to look at the cost issue differently. "Some physicians, in an attempt to control cost, resort to making the point that there are not enough data" to establish the value of ULR, he says. "There is no doubt it will cost money to universally leukoreduce blood, but cost savings achieved in other areas would make leukoreduction feasible. That is how we funded leukoreduction at Yale."

The FDA’s position on ULR has done little to resolve these opposing views. In September 1998 the FDA’s Blood Products Advisory Committee was asked to decide whether the benefits of ULR outweigh its risks. The committee voted "yes" 13-0, with three abstentions. But the FDA announced June 16 of this year that it will not formally recommend ULR in the guidance document it is preparing. Jay Epstein, MD, director of the FDA’s Office of Blood Research and Review, said, "FDA remains convinced at a scientific level independent of cost that ULR is an improvement in safety and purity." But he stressed the goal of expanding ULR on a voluntary level.

"Their legal advisors have told them they can’t mandate it, and they can’t send out a guidance without changing the Code of Federal Regulations, which is a very cumbersome and complex process and certainly can’t be done quickly," says Lawrence Petz, MD, professor emeritus of pathology and laboratory medicine at the University of California at Los Angeles. "So I think there is no FDA mandate imminent-and there may never be a mandate."

A mandate may not be necessary, in fact, because if ULR becomes clearly recognized as the industry standard, it will influence the FDA’s licensure and review decisions and become a de facto requirement. A December 1999 estimate by AdvaMed, the Advanced Medical Technology Association (formerly the Health Industry Manufacturers Association), predicted that with use of LR growing at six percent per month, the United States will achieve ULR by December 2001.

In the meantime, the FDA is scrutinizing its quality control guidelines for LR. Those guidelines call for QC testing of at least one percent of products or a minimum of four per month, with 100 percent of tested units containing <5.0 x 106 residual donor leukocytes per final product. European guidelines are tighter: They define LR as <1.0 x 106 residual leukocytes/unit and recommend sufficient testing so as to detect a 10 percent failure rate in the LR process.

The FDA is considering the more stringent European specifications as well as a stricter QC standard: a statistical approach to ensure with 95 percent confidence at least 95 percent of the blood units meet specifications, Jong-Hoon Lee, MD, said at a December 1999 workshop on ULR sponsored by the FDA. Dr. Lee is chief of the Blood and Plasma Branch in the FDA’s Center for Biologics Evaluation and Research.

Estimating the potential cost of ULR has become controversial in itself. The American Association of Blood Banks says there are three ways to come up with a figure. Based on the Canadian Coordinating Office for Health Technology Assessment, ULR for the United States can be estimated to cost $400 million per year. Based on figures from the British government, ULR will cost the United States $672 million per year. And based on unit costs for 12 million red blood cell units ($20 for leukocyte reduction plus $7 for labor, QC, and overhead) and unit costs for 6 million platelet units ($40 for leukocyte reduction plus $7 for labor, QC, and overhead), ULR would cost $606 million per year.

There is skepticism in some quarters about those numbers. "I think some real questions need to be raised about the relationship of added charges to leukoreduction to the true costs of doing it," says Laurence A. Sherman, MD, JD, chair of the CAP’s Transfusion-Transmitted Viruses Committee and retired professor of pathology with Northwestern University, Chicago. "The estimates I’ve heard and seen for the costs of LR in the course of blood collection and processing are significantly less than what I’ve seen and heard people are paying."

Whichever estimate is used, says Lawrence Goodnough, MD, professor of medicine and pathology and immunology at Washington University and director of the transfusion service at Barnes-Jewish Hospital, St. Louis, he considers ULR a misallocation of money. He and 30 colleagues recently wrote a letter (Transfusion. 2000;40:751-752) opposing implementation of ULR based on the scientific merits.

The FDA’s BPAC did agree that the benefit-to-risk ratio of LR is sufficiently great to require ULR, he concedes. "However, the minutes stated clearly many committee members agreed there are insufficient studies and everyone wanted more data." But Dr. Goodnough also looks askance at the origins of ULR. "As far as I’m concerned, this whole controversy has nothing to do with blood safety. This is a business plan implemented by the blood centers and commercial leukoreduction companies."

To date, Dr. Goodnough adds, both LR and non-LR blood products remain FDA-approved. After the BPAC vote, "all blood centers, not just the Red Cross, had the clear understanding the FDA would give them two years’ ramp-up time and eventually require ULR. Now it’s not going to happen, so hospital-based transfusion medicine specialists like me feel strongly this is the practice of medicine and needs to be determined by a hospital transfusion committee." Hartford Hospital’s Dr. Silver regards the FDA’s dodging of a formal rule as a softening of its position that may take some of the pressure off blood centers to move quickly toward adopting ULR. "I don’t see it as inevitable," he says.

Dr. Petz concurs that ULR is the practice of medicine, essentially taking what has been a clinical decision and denying physicians the choice of non-leukoreduced products. "The question is, who has responsibility for taking care of patients, and the answer at hospitals is physicians," he says. "If a physician orders a non-leukoreduced product for a patient, is it justifiable for someone in the blood center to say we’ll send you leukoreduced products?"

According to the AABB, there is consensus that LR is appropriate for three indications: prevention of febrile nonhemolytic reactions, prevention of HLA alloimmunization, and prevention of transmission of cytomegalovirus (Association Bulletin No. 99-7. Aug. 2, 1999). These are the main reasons physicians order about 15 to 20 percent of blood products leukoreduced now, and there is good evidence for using LR cellular products for these patients, says Dr. Sherman.

Not conclusively substantiated by clinical trials, the AABB says, are potential benefits including prevention of immunomodulation, prevention of virus transmission other than CMV, and prevention of transmission of variant Creutzfeldt-Jakob disease. There has been clinical evidence both supporting and discrediting an association between leukocyte reduction and a decreased risk of postoperative infection, AABB reports. Age, type of surgery, previous open-heart surgery, and preoperative platelet count are each more significant predictors of postoperative mortality, the AABB says. The possibility of transmission of some viruses less pathogenic than HTLV-I/II could be decreased by LR, but the risk from these viruses is not considered high. The consensus on vCJD is that LR is not likely to have any significant impact on the transmission rate.

"Where there have been some data suggesting LR may decrease or stop infection rates and the like, the evidence is not nearly as clear-cut, and there may be confounding factors," Dr. Sherman maintains. "Because the large bulk of patients fall into these other categories, ULR becomes of questionable value in their case."

Neil Blumberg, MD, professor of pathology and laboratory medicine and director of transfusion medicine for the blood bank at the University of Rochester Medical Center, New York, views the evidence for transfusion immunomodulation as convincing. "Of 10 randomized clinical trials, eight support the position that transfusion immunomodulation is clinically important," he says. (In contrast, only one study, he says, demonstrates LR’s ability to prevent CMV transmission.) He will present a paper at the AABB meeting next month describing an implementation trial of LR transfusions in cardiac surgery. Bearing out previous randomized trials, he reports, mean outcomes in transfused surgical patients after LR was introduced improved for eight out of eight clinical and economic variables, including deaths, ICU hours, ventilator hours, antibiotic days, fever days, lengths of stay, charges, and costs.

In this study, Dr. Blumberg and his colleagues found a net estimated actual cost savings per year of $1,700 per patient after introduction of leukoreduction. With half of the blood in the United States used for surgery, he estimates nationwide savings of ULR could be as high as $6 to $12 billion per year. "We didn’t include savings in other types of patients, such as a reduced need for CMV serotesting," he adds. "We assumed the savings on postoperative morbidity would greatly outweigh CMV and febrile reactions savings."

Dr. Petz finds the claim that LR reduces postoperative infection to be not very well documented, and Dr. AuBuchon, too, believes the research so far is inadequate. "The prospective studies that have been published to date are all from Europe, they all used blood components slightly different from what we use in this country, and they have come to opposing conclusions despite similar designs," Dr. AuBuchon says. He expects that a prospective, randomized study recently closed at Massachusetts General Hospital in Boston will provide data early in 2001 that will determine whether ULR truly decreases length of stay.

Even more controversial is the possible connection between transfusions and immunomodulation over the long term. "These are questions that have been percolating in everyone’s mind," says Ronald A. Sacher, MD, new director of Hoxworth Blood Center at the University of Cincinnati and previously professor of medicine, pathology, and oncology at Georgetown University Medical Center, Washington, DC. "Why there may be certain long-term immunological effects, what is the long-term morbidity of the patients, what is the relationship of leukocyte reduction in that equation. Perhaps long-lived leukocytes can produce microchimerism, and this may play a role in potential immunosuppressive effects of transfusion. The jury is still out on that."

Dr. Snyder, on the other hand, finds the evidence compelling and compares leukoreduction to a tetanus shot. "I think to deny a young male or female patient leukoreduced blood because there is no indication now is to mortgage their future. If such a patient became ill in their 70s, they could present for cancer treatment already alloimmunized because of a transfusion they got when younger. It’s similar, in my perspective, to not giving a person a tetanus shot now because they’re not likely to run into a rusty nail."

Can leukocyte reduction decrease the recurrence of cancer? Dr. Blumberg believes there is strong evidence suggesting it does. "If as little as 10 percent of the association between transfusion and increased death due to cancer is causal (an increase in cancer death of one percent out of a total nine percent observed difference), the mortality from immunomodulation-induced cancer recurrence is estimated at 2,000 deaths per year."

"There are lots and lots of papers on this topic. It’s quite confusing," says Dr. Snyder. "Some show there are benefits of LR in preventing tumor recurrence. Others show there are not. Then we have meta-analyses which say maybe, and that’s my final answer."

On the issue of LR and graft-versus-host disease, he is more definite. "There is a high level of consensus that LR will not prevent post-transfusion graft-versus-host disease," Dr. Snyder said at the FDA workshop. He maintains that gamma radiation is the most prudent measure. LR will, however, probably enhance the safety of blood against bacteria, he added. "But, due to various issues, suffice it to say that removal of bacteria from blood by leukoreduction filters is not an indication which will likely be translated into label copy."

Although adverse reactions to LR blood are rare, critics of ULR suggest that more reactions could result if ULR is adopted. Leukocyte reduction may be done in a blood center shortly after the blood is collected (prestorage LR), at the hospital after the blood has been delivered (in-lab post-storage LR), or near the patient (bedside LR). Under current AABB standards, a leukocyte-reduced component is defined as one with <5 x 106 residual donor leukocytes per final product.

"There are different technologies involved in producing blood components that are nearly devoid of white cell content," and sometimes the technology affects the final component to induce either allergic or idiosyncratic type reactions in transfusion recipients including facial flushing and dyspnea, says Ira A. Shulman, MD, chair of CAP’s Transfusion Medicine Resource Committee and vice chairman of the Department of Pathology, LA County-USC Medical Center, Los Angeles. Furthermore, depending on the filter set used and the storage duration of red blood cells prior to filtration, hemolysis of red blood cells may occur.

In fact, many of the drawbacks seen in leukoreduction have been blamed on bedside filters. In May 1999 the FDA issued a medical alert that bedside LR has been associated with serious hypotension; other problems include complexity of the setup for a busy nursing staff, inappropriate priming of the filters, filter clogging, and return and discard of the blood product because of excessive time outside adequate storage conditions. A formal study commissioned by the Canadian government in 1998 found that prestorage leukocyte reduction of platelets either by filtration or apheresis techniques was an estimated four to five times more effective in preventing febrile reactions than filtration after storage in the laboratory or filtration at the bedside. Similarly, prestorage filtration of red blood cells was 15 to 25 percent more efficacious than hospital or bedside filtration.

The American Red Cross is performing a study, in collaboration with the Army, on three different filters to determine which works best. But Dr. Snyder believes a move away from bedside filtration is desirable. "I think if we’re going to get the maximum benefit of leukoreduction, it has to be prestorage," he said at the FDA workshop.

Joseph D. Sweeney, MD, medical director of three hospital blood banks and associate professor of medicine at Brown University, Providence, RI, says transfusion medicine physicians are clearly correct in stating the evidence for ULR is not conclusive, based on currently published literature. "But if we look at many interventions in medicine, we really don’t find they have been introduced after such trials," he points out. He cites magnetic resonance imaging as an example. "An MRI costs twice as much as a CT scan. It gives better images than a conventional CT scan, and it would be reasonable to conclude that surgeons using MRI images might have better outcomes. But in fact I don’t know of any randomized study that shows this."

"Similarly, with nucleic acid testing for HIV-1, we can’t prove at this point it will make a material difference. Yet I don’t hear the same opposition," he says. He attributes this in part to transfusion medicine physicians having been slighted to some extent, because they haven’t been consulted about their opinions, and leukoreduction is an area that has been under their control. "In other areas like [adding the test for] HIV-1, it was never a clinical decision; it was all done in blood centers."

"We are all very concerned, because of the various diseases, particularly HIV, in the past 15 years, that we don’t miss something we should have jumped on," Dr. Sherman says. "For example, we’ve instituted HTLV-I testing at a time when we had very limited data about its relationship to disease and frequency in the donor population, but the feeling was we didn’t know enough to say it was not consequential, and it would seem to have a very long incubation period, so it might be a very long time before we had enough data."

ULR has a definite additional cost, Dr. AuBuchon notes: a diminishing donor pool. Because African-American donors with sickle-cell trait have blood that tends to clog the filters, it is likely they would be deferred. "Only about 10 percent of donated units come from African-Americans, and only 10 percent of them are carriers, but moving to ULR will reduce the available blood supply by one percent," he cautions. "That seems small, but when we’re losing two percent because of geographic deferrals for vCJD [donors who visited the United Kingdom for six months or more between 1980 and 1996] and anticipating a shortage of blood already, that may be important. Furthermore, explaining why this loss is occurring specifically to one minority population may be difficult, and may result in loss of allegiance of this and other minorities to blood donation."

As for the more traditional costs, Riverview Hospital in Noblesville, Ind., found the savings from ULR made up for the expense. The 156-bed hospital has been 100 percent leukoreduced since 1988, using filters at the patients’ bedside, and began buying all prestorage-leukoreduced blood products from a blood center in 1997. Nancy Chance, MLT(ASCP) RMT, blood bank coordinator, says the hospital has not had any adverse reactions to leukoreduced blood products, even when it was using bedside filters. On the contrary, she says, "pre-medication and nursing time for febrile reactions has been minimized to almost nil," and several studies at Riverview have linked ULR with significant cost reductions because fewer postoperative transfused infections develop.

A Value Analysis Team at Riverview scrutinizes every new product-from bedpans to blood-before it’s used, and leukoreduction underwent the team’s full analysis. "Every department that plays a part in the patient’s recovery has a representative on that team," Chance says, "and their job is to play the devil’s advocate." Leukoreduced blood wouldn’t have made it in the door in the first place and the practice could not be maintained without the data to support it, she says.

But Celso Bianco, MD, vice president of medical affairs at New York Blood Center and immediate past president of America’s Blood Centers, says Chance’s experience is unusual. "[The hospitals] don’t believe in us when we tell them that in-hospital times and admissions and all that will be shortened and that their costs will go down," he said at the FDA workshop. "I think that they really resent that they have not been part of this process. We have been evolving our thinking toward leukoreduction. . . . We have not been able to bring all those participants of the health care system to come with us."

Not surprisingly, since hospitals are expected to bear the brunt of ULR’s cost, they have been unenthusiastic about it. Eyeing the unreimbursed blood safety measures of the last two decades, many argue they can only absorb so many cost blows before those blows start to cripple patient care.

Dr. AuBuchon sees dollars being misdirected in the push for ULR. "If we are going to spend more money to make the blood supply safer, there are already several larger, well-documented risks that patients face that the field is not currently addressing. We already know that bacterial contamination and mistransfusions each cause the death of about two dozen patients annually in the U.S. and have for decades. Yet, for reasons I don’t understand, the field is not focused on that. Instead, it continues to focus on stamping out the last case of viral transmission and potential risks rather than actual ones." Also promising candidates, Dr. Sacher says, are refining donor questionnaires and truthful completion of the questionnaires, perhaps with technological strategies to speed up the process.

However, says Dr. Sweeney, it seems the cost-effectiveness criteria that are being applied to ULR are different from the criteria applied to other blood safety measures. The FDA requirement imposed in August 1999 that donors resident in the UK be declined "resulted in substantial donor loss, as many as three percent," Dr. Sweeney says. "There was really no justification for that request, considerable expense was incurred, we’ve lost some repeat donors, and there was almost no outcry from the transfusion community."

Practically speaking, he says, the additional cost of ULR will relate to filtering red blood cells. He cites several factors, however, that will directly or indirectly reduce the net cost of ULR-for example, savings from much less HLA alloimmunization, eliminating CMV testing, being able to move more patients through clinics because bedside filters aren’t needed, and lack of need for nursing procedures or in-service training on bedside filtration.

At least some hospitals have taken advantage of these in their conversion. When Vanderbilt University Medical Center, Nashville, Tenn., moved to 100 percent prestorage leukoreduction in September 1998, two weeks before the favorable vote by BPAC, "We did it in a cost-neutral way," reports Anne Thomas Neff, MD, director of Vanderbilt’s transfusion service.

Vanderbilt already was transfusing a significant percentage of leukoreduced red blood cells because of its transplant and oncology emphases. The vast majority of its blood products, however, were not leukoreduced. The major impetus for Vanderbilt to move to ULR was a study trial published in December 1997 that found no significant advantage to giving patients single-donor platelets compared with random-donor platelets as long as the platelets are leukoreduced.

"Once technology has caught up so that random-donor platelets are more and more like single-donor platelets, that begs the question associated with increased cost," Dr. Neff says. When the hospital turned to the less expensive platelets product, "the money we saved paid for the additional expense associated with red cells." Other savings derived from not having to have two inventories of blood products or keep filters in stock, and from using leukoreduction as the equivalent of CMV-safe blood, she adds.

The question of how blood safety measures will be reimbursed, of course, goes beyond ULR. MacPherson of America’s Blood Centers points out there is a two- to three-year delay in adjustments to Medicare and Medicaid payments under the DRG system, where most transfusions take place. That means more trouble in the future, he predicts. "Most of the new technology that’s coming down the pike is going to be a whole lot more expensive. Technologies in the future, like viral inactivation, are going to double or triple the price of blood. So you can’t have a two- to three-year delay when you try to talk about that."

Melissa McMillan, associate director of public relations for America’s Blood Centers, says blood centers support a formal FDA rule on ULR rather than a standard-of-care approach, in part because of reimbursement. "When it comes to getting reimbursed by the Health Care Financing Administration, having something just be standard of care is not good enough." Since ULR is not included in the "market basket" used to calculate hospitals’ DRG rates under prospective payment, blood centers have been intensively lobbying Congress to adequately reimburse hospitals for blood products.

Last December, when the Balanced Budget Refinement Act of 1999 was passed, the ABC, the American Association of Blood Banks, and the American Red Cross were successful in getting specified outpatient payment classifications (APCs) to cover blood, with reimbursement amounts pegged to cost data provided by the blood groups. The inpatient setting is not yet taken care of and still requires congressional action. But now, said the FDA’s Dr. Epstein at the December workshop, since HCFA has created a fee schedule for outpatient reimbursement, "it is at least possible for blood centers to argue to hospitals that the real cost of providing an inpatient unit is not less."

Inadequate reimbursement has become even more pressing in the last year, McMillan adds, because up to now there have been small changes in the cost of blood. For that reason, blood centers are also seeking a long-term solution: a system that would set aside money to pay for any new screening procedures blood suppliers must adopt.

Many supporters of ULR concur with opponents that without mad-cow disease (transmissible spongiform encephalopathy) and the human version, variant Creutzfeldt-Jakob Disease (vCJD), the concept of ULR would never have reached critical mass. There has been no known human case of vCJD contracted via blood transmission. ("And I imagine if there were a human case the Internet would go ablaze," Dr. Shulman says.) However, a recent Lancet article on transmission of TSE by transfusion in sheep (Lancet. 2000;356:999-1000) has triggered a flurry of discussion. As there is no evidence that leukocyte reduction would prevent prion transmission in any case, even the Lancet article is not compelling for ULR to prevent prion transmission.

Nevertheless, many western European nations, including Great Britain, Austria, Germany, France, Portugal, Switzerland, and Ireland, have imposed ULR because of vCJD. Canada, too, imposed ULR, and a large factor there, Dr. Silver says, was a rash of hepatitis cases, blamed on the government. "Canada was stung when they didn’t get in early enough on hepatitis. As a result they’re gun-shy, so they’re going in much more quickly" on new blood safety measures, he says.

Although there were scientific concerns, the political force spurring the field has been concern about transmission of vCJD. "And I think most people probably would agree that transmission of vCJD is unaffected by leukoreduction," says Dr. Snyder.

"Nothing much would be happening now except for one thing, which is that new variant CJD in Europe may be transmissible by transfusion," Dr. Blumberg contends. "We don’t think it’s likely, but we can’t prove that, and there was some data that unlike CJD, the infectious protein particle, the prion, was present in lymphocytes, so somebody got the idea leukocyte reduction could reduce the risk. I don’t know even if the FDA would have had the BPAC look at this issue but for Europe. So the irony is we’re doing ULR not largely because of research showing it’s beneficial, but because of a will o’ the wisp concern to make up for the fact we didn’t stop the AIDS epidemic." But whatever the motivations were that brought ULR to the BPAC’s attention, Dr. Blumberg adds, "the decision to go with ULR is the right public health decision."

Dr. Petz would like to see the United States make a decision on ULR not influenced by the Europeans’ policies. "It’s true other countries are doing it, but we in the United States should look at the data ourselves and be a leader" on blood safety issues, he says.

Transfusion experts tend to agree that if leukocyte reduction did not add cost, fair-minded physicians would support it. "If it was free, we’d all do it," is one blood expert’s assessment, because few disagree that taking impurities out-and white cells are an impurity-is a good thing. "If you asked me which type of blood transfusion I want, I’d say leukoreduced, because of all these theoretical benefits," Dr. Sacher admits. "My personal view is that ULR may have significant benefits because of immunomodulatory effects and prevention of long-term morbidity. But that’s not proven, and whether in fact because of theoretical benefits the country can support ULR at $500 million-that’s the issue," he says.

David Pittman, MD, representing Barnes-Jewish Christian group in St. Louis, pointed out at the FDA ULR workshop that the fixed reimbursement hospitals receive means adopting ULR will force tradeoffs and could compromise other hospital services. "What are we going to give up? Are we going to have fewer nurses? Are we going to have less SOPs, less adequate QC, fewer OR techs, fewer custodians? Because in our system what gets cut in general is personnel, the people that actually take care of the patients." Harvey G. Klein, MD, chief of the National Institutes of Health’s Division of Transfusion Medicine Clinical Center, agrees. "The FDA’s own mandate is safety and efficacy, not cost, but cost may be a safety factor if inappropriate expenditures prevent more appropriate public health interventions."

Many say if they had the choice, ULR would not be their first priority, because more solid data exists on other measures that would have an equal, if not greater, impact on patient safety. Dr. Shulman cites, for example, virally inactivated plasma products versus raw plasma, and plateletpheresis transfusion versus pools of random platelets as "two issues which are equally as important to transfusion practice but, for whatever reason, are not being debated at this fever pitch."

Having been buffeted by various blood crises for the last two decades, blood centers are at the center of a war between science and emotion, mixed with politics. As a result, rejecting any promising blood safety measure, even before all the evidence is in, can be risky. But the risks include any adverse impact on health care, says the FDA’s Dr. Lee: "You might produce the highest quality blood, but if it results in lower quality patient care overall, then you have defeated your public health mission."

As a matter of public policy, many experts say, the government and third-party payers must decide how far they are willing to go toward funding a mandate for 100 percent safety, especially as new technologies emerge for removing subsets of cells from blood products and sterilizing blood components of infection. "We have a kind of ’zero-sum’ game in health care" in which wins in one sector are made at the expense of other sectors, Dr. Sherman says. "From an ethical standpoint, we have to recognize that if we get more support or resources for things we think are significant in the blood arena, then it has the probability of reducing expenditures somewhere else."

Dr. Petz does not believe the payoff justifies adopting ULR. "If someone gave me half a billion dollars and I spent it taking white cells out, and they came back a year later and asked how many lives have you saved, I’d have to answer not very many, if any at all." Dr. Snyder, on the other hand, sees ULR as valuable and inevitable. "My belief," he says, "is that non-leukoreduced blood products eventually will go the way of fresh whole blood."

When the Consortium of University Health Centers meets in October to discuss ULR, it will do what the FDA’s Blood Products Advisory Committee was barred from doing when it voted on ULR in 1998: consider the cost as well as the benefits. "When a new technology is absolutely proven to be beneficial, it becomes compelling regardless of the cost. But there still needs to be a way of recapturing the expenditure," says Dr. Shulman. "Otherwise we can be forced out of business-and then we won’t be able to help anybody."

Anne Paxton is a freelance writer in Seattle.