Catching, tracking, and tackling TRALI

October 2007
Feature Story

Anne Paxton

Among all the possible risks of transfusion, TRALI isn't a new discovery. It took a long time to become Topic A in the blood banking world.

"If you look back at the history of TRALI [transfusion-related acute lung injury], it was first described by Mark Popovsky in the early 1980s," says Patricia Kopko, MD, executive vice president medical affairs and director of the histocompatibility laboratory, BloodSource, and assistant clinical professor of medical pathology at the University of California at Davis.

"But you remember what else was going on in transfusion medicine at that time. That was before we ever had the first HIV test, and it couldn't compete."

"Talking about fatalities, with TRALI it was maybe one in 50,000 at worst, and at that point in certain cities in this country, one in 100 units had HIV in it."

Now, all of a sudden TRALI has become the riskiest part of transfusion, she says. In 2003 the Food and Drug Administration for the first time reported TRALI as the No. 1 cause of transfusion death, with Canada and Europe reporting the same. "Up until that point, the No. 1 cause was the wrong unit of blood. That's when we realized we have another important risk from transfusion," Dr. Kopko says.

"We're not in the post-HIV era—that's not gone—but it's been reduced along with HCV to such incredibly low levels that we now have enough mental energy to direct to other risks of transfusion," says James P. AuBuchon, MD, E. Elizabeth French professor and chair of pathology at Dartmouth- Hitchcock Medical Center, Lebanon, NH. "Bacterial contamination a couple of years ago—that's not at an end either. But we've made a beginning, and we have more and more progress on mistransfusion. TRALI was the other one of the big three nonviral risks."

Since TRALI unseated hemolytic transfusion reaction for first place, the blood banking community has mobilized to stem the syndrome. An important consensus conference took place in 2004 in Canada, from which came a standard definition adopted in North America and Europe. Measures adopted by the United Kingdom's National Blood Service to reduce TRALI from plasma were already cutting the number of cases in half by 2005. Experimental models and prospective studies have started dotting the research landscape.

Most significant in accelerating the campaign against TRALI were the AABB's two key recommendations issued almost a year ago: that U.S. blood centers initiate measures to reduce the risk of TRALI in plasma by November 2007, and that they initiate measures to reduce the risk of TRALI in apheresis platelets by November 2008.

Mark Popovsky, MD, chief medical officer of Haemonetics, Braintree, Mass., and associate professor of pathology at Harvard Medical School, was a member of the AABB task force that developed the recommendations and time frames for blood centers to address TRALI. "By issuing these guidelines, the AABB is providing an opportunity for blood collectors and hospitals to do more to reduce the impact of the syndrome," he says.

Under the AABB guidelines, however, "it's completely left up to each center to determine the measures to reduce risk," Dr. Kopko says. "So what we are going to do to meet those deadlines is an exceedingly hot topic."

Unlike bacterial contamination and hemolytic reaction, TRALI, which is believed to occur in one in 5,000 transfusions, remains a syndrome with somewhat mysterious etiology.

It appears that patients who have had underlying hematologic disease, patients who are in intensive care settings, and patients undergoing surgery may be at increased risk, but the jury is still out on who fits the high-risk profile, says Dr. Popovsky. "We're still lacking an understanding of who the high-risk patient is, which persons are more likely to develop TRALI."

However, researchers believe the causes of TRALI can be both immunologic and nonimmunologic. "Acute lung injury and pulmonary edema are the end-stage process, and how you get there is thought to occur through two major but different mechanisms," says Steven H. Kleinman, MD, clinical professor of pathology and laboratory medicine at the University of British Columbia, Victoria, and senior medical advisor to the AABB.

"One is the antibody in the donor unit binds with the antigen on patients' white cells. There has to be a match, and if recipients do not have the antigen, there is no match to trigger the immunological cascade." Or, as Dr. AuBuchon puts it: "That antibody has to go into just the right unlucky recipient in order to find an antigen it's targeted against; that explains the relative rarity of this type of reaction."

The second mechanism has nothing to do with antibodies, Dr. Kleinman says. "It's through substances called biological response modifiers, such as breakdown products or cytokines, that are released by cells and accumulate in the blood," he explains. "If they are transfused to a recipient who already has neutrophils primed or activated—for instance, through sepsis or inflammatory responses—and they happen to have a critical concentration of one of these substances, they might trigger TRALI."

It's becoming better known that more than one mechanism can trigger TRALI, Dr. Popovsky believes. "Both the antibody-mediated model as well as the two-hit model are important. We know they both may be operating in the same patient, and there may be other models as well."

But "there's an awful lot about TRALI that we don't know," points out Morris Blajchman, MD, professor of pathology and medicine at McMaster University, Hamilton, Ontario.

"We have all seen cases of TRALI occurring in patients who receive blood products where no antibody is demonstrable in the offending blood product. You can also have a donor with an antibody that can cause severe TRALI in one individual and find that some of the recipients of that donor's plasma will not have any ill effects. A lot of research will be required to get a better handle on our understanding of the pathogenesis of TRALI."

Promising research is underway on several fronts. The National Institutes of Health has funded two major efforts examining TRALI, says Darrell J. Triulzi, MD, medical director of the Institute for Transfusion Medicine, Pittsburgh. One is the SCCOR grant, a specialized center for clinical research award made to the University of California at San Francisco.

Dr. Triulzi is an investigator on a study in the second research project, a network of blood-center-based studies called REDS II. His paper on leukocyte antibody prevalence is to be presented at the AABB conference this month. "It is designed to provide the best data we have on the prevalence of HLA and neutrophil antibodies in donors who have risk."

Unfortunately, a significant sector of the donor population—multiparous women, or those who have had multiple pregnancies—has a stronger likelihood of carrying these antibodies and may be a major source of TRALI. These donors are considered to pose a particularly high risk to recipients.

The most commonly accepted explanation relates to the body's capacity to reject transplants. "Pregnancy exposes the woman to the fetus, and the fetus carries antigens inherited both from the mother and the father," Dr. Kleinman explains. "Those from the father are 'foreign.' So these HLA antigens are present on leukocytes but also present on many other tissue cells."

"When the baby is in utero the pregnant woman is exposed to these, and a proportion of women make antibodies to these antigens. And we have evidence that this increases with the number of previous pregnancies."

The many unanswered questions about TRALI have made diagnosing it a challenge. In the past, there were many cases that were not properly diagnosed. Moreover, diagnosis remains difficult because at present there is no diagnostic test that tells the physician this episode of shortness of breath associated with a severe respiratory illness is caused by the blood that person received.

"Most hospitals have a fairly well-developed system whereby someone thinks something went wrong, and they call the laboratory and begin a transfusion reaction investigation," Dr. AuBuchon says. In the past, "the standard transfusion reaction investigation focused on whether an ABO error occurred, so the usual investigation will be 'negative' in the case of TRALI."

"The important thing for the laboratory to remember," he says, "is if they get a report of somebody having respiratory difficulty shortly after transfusion, they too need to think about TRALI and initiate appropriate consultative assistance to the clinician so the patient can be properly cared for."

"That includes getting a chest x-ray to make sure the problem is not fluid overload, because if it is, then the patient will benefit from diuresis. If it's TRALI, some advocate not giving diuresis. They believe the patient will do better and actually benefit from more fluid."

Ideally, the transfusion medicine director will have read about TRALI and be able to step in, because the clinician probably hasn't encountered a case of TRALI, Dr. AuBuchon adds.

"We just need to make the diagnosis," Dr. Blajchman says, "and be aware of TRALI and support the patient based on the extent of their respiratory embarrassment. If you support such patients for a couple of days, the vast majority of patients—90 to 95 percent—actually recover."

But there is strong evidence that many cases of TRALI go undiagnosed. "It's very difficult sometimes at the bedside to tell the difference between TRALI and transfusion-associated circulatory overload," says Lorna Williamson, MD, medical director of the National Blood Service division of National Health Service Blood and Transplant in the United Kingdom. "We're looking more critically at fluid balance charts from the 48 hours before an episode. In fact, when you look at some of these cases, it's pretty obvious there is a very positive fluid balance."

Dr. AuBuchon, who is chair of the CAP Transfusion Medicine Resource Committee, pointed to a recent paper that reported results when a medical center tracked back to determine how frequently TRALI might be occurring in its ICU patients. Of the cases identified, none had been reported to transfusion services at the time. "They were only found by retrospective review of this group of experts."

Dr. Kopko had similarly startling results from a BloodSource study. "We actually had a local fatality from a unit of plasma and we found antibodies in the donor's blood to human neutrophil antigen 3a [HNA 3a, formerly HNA 5b],"she says.

"So when the FDA came to investigate, they asked if we would perform a lookback. The woman was a frequent plasma donor, and a number of the patients receiving this woman's plasma and who had TRALI reactions had been unreported to either the transfusion service or the blood collection facility. She had a very strong antibody. In a number of cases they found either the patient ended up on a ventilator or there was new pulmonary edema that hadn't been there before." There were also less severe febrile reactions.

The AABB recommendations have quickly brought fundamental policy changes to U.S. blood centers. Says Dr. Kleinman, "We don't have any absolute figures as to how many U.S. blood banks have done this, but there is a general idea that the majority of institutions have moved toward meeting their plasma needs with male plasma, to reduce the number of units with leukocyte antibodies." The American Red Cross blood centers, Dr. AuBuchon reports, are well on their way to using 100 percent all-male plasma. The Community Blood Center of Greater Kansas, which collects about 145,000 units of red cells a year, switched to all-male plasma last spring, as well as to plasma that's frozen in 24 hours rather than fresh frozen, says Jay E. Menitove, MD, a hematologist who is executive and medical director of the blood center.

But it's not as drastic a step as it seems. Though shortages of AB plasma continue to be a problem, Dr. Menitove says, "if you look at how we tried to implement male-only, you have half male/half female whole blood donors, and only about 45 percent or so of plasma derived from whole blood donations is actually used as direct plasma transfusions. So if you have 45 percent needed and half the donors are male, you can meet the needs for plasma transfusions."

The Community Blood Center estimated there would be an additional $5,000 to $10,000 per month in transportation costs, but that has been trimmed, and the center did not increase its charges to hospitals.

This is an evidence-based approach, Dr. Menitove explains. "Again, it's an intervention not to address 100 percent of the risk hypotheses, but it addresses the parts we can address, the part we can change, and for the rest, we'll just have to keep watching it."

But will all-male plasma do the trick?

Some of the data have started to come in. The United Kingdom has already witnessed a striking decline in TRALI cases after implementing its all-male plasma policy (see "All-male plasma makes UK TRALI tumble"). And preliminary data are to be presented at the October AABB annual meeting from one center that implemented a male-only or primarily male-only plasma program, Dr. Popovsky says. "The number of case reports to that blood center dropped dramatically, indicating there is value in using such a policy." Though it will take another year to see what impact the policies have in the U.S., "I would predict we will see a significant decrease."

However, it may be difficult to determine that in the U.S. because of underreporting and underdiagnosis and the lack of central data collection. "If you get a fatality from a transfusion, that has to be reported to the FDA," Dr. Kleinman points out. "But a very sick patient doesn't get reported, so there's no good way to keep track of reactions in a given category like TRALI, although since certain organizations like the American Red Cross require reporting, we can look at part of the pie."

"One of the troubles with the whole field of transfusion medicine is there is not too much time devoted to the subject in medical school," says Walter "Sunny" Dzik, MD, co-director of the blood transfusion service at Massachusetts General Hospital, Boston.

As a result, "Until we recognized that TRALI was pretty common and pretty serious, we kind of took plasma for granted as a relatively safe product." For example, as several articles in the medical literature reported recently, physicians often order plasma to address a mild to moderate elevation of INR on a prothrombin test. "We have the suspicion doctors think they are fixing the lab test results with plasma, but it really does not have much effect on the lab test," Dr. Dzik says.

In fact, Dr. Kopko wonders whether the success of the UK program can be replicated in the United States, because British doctors actually use less plasma. "I think clinicians tend to have a perception that plasma is the safest component to transfuse because you tend not to have the incompatibility you have with red cells," she says.

While some experts believe that pooling plasma dilutes the antibodies that can cause TRALI from any one individual, Dr. Blajchman says there is a flip side to the dilution benefit of pooling. "Theoretically, pooling can reduce the risk of TRALI. But variant CJD can be transmitted by blood transfusion, and if you pool 1,000 units, receiving a pooled product theoretically increases your chance of getting vCJD by 1,000-fold. So pooling can be a double-edged sword."

A study being funded by the National Institutes of Health is actively looking for cases of TRALI in a couple of big U.S. hospitals, Dr. Kleinman says. "They have one year's worth of data prior to these interventions, and they will get a year or more after the interventions, so it might be the most controlled environment available to tell whether the measures made a difference."

Providing male-only plasma or male-only platelets is not a trivial matter to resolve, Dr. Blajchman emphasizes, "because women represent 40 to 50 percent of blood donors, and if they stop donating, we're in trouble."

He's not alone in that concern, especially since addressing TRALI from platelet donations is an even thornier problem. The United Kingdom has already moved forward successfully with all-male or nearly all-male pooled platelets, which in Europe are made from a pool of four donors. But it is not doing anything just yet to address platelets collected by apheresis, says Dr. Williamson of NHS Blood and Transplant. "What we're about to do is to test any new female donor who wants to come onto the apheresis program for HLA antibodies. We could exclude all female donors, but we're concerned we would lose rather a lot of donors at once, and many have given apheresis platelets for years and not caused TRALI in any way."

The United Kingdom is planning to conduct trials of platelet additive solution in early 2008. "That is a method for suspending platelets that combines additive solution with some plasma," Dr. Williamson says. "So instead of having 300 mL of plasma in the platelets, you'd only have maybe 100 mL, which you need because platelets don't store properly in a 100 percent additive solution. We're not really sure whether that would be particularly effective, because that's still quite a lot of plasma and could still cause TRALI if the donor has a strong antibody."

However, Dr. AuBuchon notes, no platelet additive solution is licensed in the United States. "A few companies are interested in pursuing it, but as of yet there is not even one in the application phase before the FDA."

Dr. Menitove says some strategies now under discussion at his blood center include questioning women about their history of pregnancy or conducting testing to determine if there are HLA antibodies. "We could then decide which female donors to accept for platelets. We're not asking that yet on the questionnaire, but I suspect we will over time."

Many blood centers in the United States are worried about how to address the AABB's second recommendation on apheresis platelets. "That's a big deal right now. We haven't quite figured out what to do," Dr. Kopko says. "We are hoping that one of the automated platforms for HLA antibody testing gets licensed in the next couple of months." The technology that's available now is manual and not suited to the high-throughput testing needed to test hundreds to thousands of donors each day.

"With plasma, you can go to predominantly male, because female plasma will be sent to manufacturers of albumin, IVIg, and various drugs. That part was easy. But if we were to say no women can donate platelets, we would have a horrendous platelet shortage in this country."

The implications of that are enormous, she cautions. "There were 30 reported fatalities from TRALI last year, mostly from plasma. So if you reduce fatalities by five percent but you have hundreds of thousands of fatalities because there are no platelets and patients can't get chemotherapy or surgery—that's the reason it's become such a timely topic."

Coinciding with the plans to reduce TRALI through screening are some companies' efforts to try to inactivate the pathogens that may be present, Dr. Blajchman says. "Pathogen inactivation of blood products is coming down the pike. Octapharma is a European company with a pathogen-inactivated plasma product which they claim has not been associated with any TRALI. How carefully that data has been collected is unclear, but it makes sense scientifically." The value of pathogen-inactivated plasma with respect to TRALI reduction comes not from the pathogen inactivation but from the fact that it is a pooled product, Dr. Blajchman emphasizes.

At Massachusetts General Hospital, a study is being conducted of the utility of asking additional questions of the donor about pregnancy history, since the Uniform Donor Questionnaire does not ask about number of pregnancies, says Dr. Dzik.

He considers switching to all-male plasma to be possibly too blunt an instrument. "It's one approach, but at my hospital we feel it might be a little bit drastic on the blood supply and might have unintended consequences, for reasons that might only be known in hindsight," Dr. Dzik says. "After all, there have been times in the past when receiving male-only blood products would not have been to the advantage of the recipient," he adds. Furthermore, many female donors have not had children. "It is a shame to defer them when they can't possibly produce TRALI."

Evidence is emerging, Dr. Dzik adds, that TRALI is even more common than previously realized. "We see cases that do not meet the Canadian consensus conference definition but when investigated are obviously TRALI. It's like the tip of an iceberg. The consensus conference has defined for us the big piece above the water, but there is ice below the water as well, cases of TRALI that are milder but still important. These milder cases prove that clinical severity depends not only on donor factors but also on characteristics of the recipient." Identifying mild cases may be important because if the donor is not excluded, then "the next time it might be a fatal case," Dr. Dzik says.

But Dr. Popovsky tends not to believe the all-male policy will have an impact on the availability of plasma units or platelets. "I know there are blood centers that have that concern because of the number of females they have in their donor base and the number of females they have found to be antibody-positive." Effective recruitment of new donors, and use of technology such as automated collection, he thinks, will make it possible for blood centers to keep donations stable.

The Community Blood Center of Greater Kansas City has not seen an impact as of yet on its donor base, says Dr. Menitove. "I think we'll have to really monitor that when we start making changes related to platelet collection, because there it will be more obvious that we are changing the mix of male to female donors, and probably saying to females that we prefer they donate whole blood," he says.

In the UK, Dr. Williamson says, "we mention it in our leaflet, but we haven't felt the need to make a big splash about it with the donors. There are no implications for their health if they have these antibodies, so we don't particularly want to get people worried they might have something to indicate future illness."

Another important prong of the AABB recommendations is education about appropriate blood use, and specialists in transfusion medicine see this as a boon to their efforts to tame usage.

"We're talking about manipulating components that are transfused, but one of the major ways is to only use blood products in appropriate indications," says Dr. Kleinman. "We have lots of evidence that there is over-transfusing, even when patients really need blood, and it's an issue not just for TRALI but in general, when complications are preventable because the patient should never have gotten a unit. So general education about appropriate use of transfusion remains an important message. That's a message that the transfusion medicine community has been trying to get to clinicians for 10 or 20 years now."

Dr. Kopko hopes that the spotlight being shone on TRALI will raise clinicians' awareness. "I have encountered residents who did not know a case was TRALI until the attending came in the next day because they had not heard of it. That's one of the things we've really worked on locally. We've provided ourselves to anybody who will let us speak about TRALI, whether it's at grand rounds, physician group meetings, or medical staff meetings at the hospitals."

When Dr. AuBuchon gives lectures to residents these days and begins discussing TRALI, he sees heads nodding, "which is good," he says. "It means the information is getting down into the medical school curriculum, and there will be a broader understanding of TRALI."

But pathologists need to do even more than they are in educating the physicians at their hospitals about what TRALI is, because early intervention makes a difference in prognosis, Dr. Popovsky maintains. "It's fair to say there is more attention paid to transfusion medicine now than 10 or 15 years ago, but still, compared to other subjects, there is relatively little attention."

"Often that leadership role has to fall on the shoulders of the blood bank or transfusion service director. But the better armed the physicians are who transfuse blood, whether they're surgeons or medical practitioners, the more likely they will actually recognize TRALI, and do the right thing."

Related Links

• TRALI defined
• All-male plasma makes UK TRALI tumble