Blood Not-So-Simple

A “no-consent” medical study of an experimental blood substitute is creating an uproar among researchers and bioethicists. Controversy over the study, which is under way at 31 hospitals across the nation, is pulling back the curtain on similar studies ready to be launched under President Bush's “war on terrorism” as well as Project Bioshield, an ambitious plan to protect citizens against chemical, biological, radiological, or nuclear attacks through the development of commercial products, such as vaccines for anthrax and smallpox, and the development of blood substitutes.

Bush signed Bioshield into law on July 21, 2004, authorizing $5.6 billion for the project. One Bioshield recipient, Northfield Laboratories, a small biotech company in Illinois, received $4.9 million, administered by the U.S. Army, in the past two fiscal years to develop PolyHeme, the fake blood at the center of the controversy.

The Defense Department has spent decades and millions of dollars in pursuit of the Holy Grail of trauma medicine: a blood substitute that doesn't have to be refrigerated (battlefield conditions don't allow that) and that doesn't have to be matched by blood type (a process that can take more than an hour). Despite initial enthusiasm for several products, most have had to be abandoned when dangerous side effects, such as strokes and heart attacks, emerged.

Despite decades of development, PolyHeme has failed to win Food and Drug Administration (FDA) approval. But the FDA did approve the current study, which was launched in 2003. If this study is positive, Northfield stands a shot at winning the FDA approval necessary to market PolyHeme commercially.

Should this happen, one of Northfield's first customers will likely be the Army, with which Northfield says it has had stepped-up dialogue since the U.S. invasion of Afghanistan. However, PolyHeme isn't being tested on soldiers. It's being administered to civilians under a rarely used provision known as the “emergency research consent waiver.” The provision, codified by the FDA in 1996, allows researchers to enroll patients, without their consent, for approved studies, in life-threatening circumstances when the patient is too ill to consent and a patient proxy or guardian can't be reached. The waiver is necessary, according to Northfield's Web site, since it “is expected that patients enrolled in this trial will be unable to provide informed consent because [of] the nature and extent of their injuries.”

* * *

So what's wrong with this? Faced with bleeding to death, wouldn't most of us rather get experimental blood over nothing at all? And since many patients with severe blood loss are unconscious, doesn't it make sense to allow doctors to enroll us without our consent?

The answers to those questions are not quite straightforward. First, it appears that PolyHeme is being tested in the wrong population. The standard of care for acute blood loss in the United States is human blood. That means that no other product has been found to be as safe and effective as real blood -- something the civilian population has ready access to in emergencies. Dr. Robert Solomon, chairman of the ethics committee of the American College of Emergency Physicians, says, “If you look at trauma units in the U.S., there really isn't a demand for blood substitutes.” The American Red Cross agrees, saying, “We are unaware of any specific areas where trauma care is affected by a shortage of available blood.”

Northfield says its product isn't meant to replace blood in hospitals, but is only intended for use on ambulances when no blood is available. However, most ambulance rides in urban areas, where the study is being carried out, last only minutes and once at the hospital, Type O blood, known as the “universal donor,” can be given immediately. However, the study protocol allows PolyHeme to be given for up to 12 hours after arrival at the hospital in those patients who received the blood substitute during the ambulance ride. Leonard Glantz, associate dean of the School of Public Health at Boston University, says his institution turned down the PolyHeme study because “it breached basic ethical requirements.”

What's more, Northfield chose an opt-out form of consent, meaning that everyone within range of the participating hospitals who is injured severely enough to require blood is automatically enrolled in the PolyHeme study unless they are conscious and decline or they learned about the study in advance and chose to wear a blue, plastic, hospital-type bracelet that states, “I decline the Northfield PolyHeme study.” Opt-out studies are desirable from the perspective of drug companies, since by default they enroll far more patients than opt-in studies. The concept of no-consent studies is also controversial and new -- so new that in early 2005 one of the sites testing PolyHeme, Duke University, had to get the state medical board to change North Carolina's Patient Bill of Rights, requiring patients to give informed consent prior to enrollment in a study. The board changed the rule and the Duke study continues to enroll patients.

Worse than the opt-out system, charges Solomon, is that the public was misled about the safety of PolyHeme during “community consultations.” Before drug companies can launch no-consent emergency research, the FDA requires that they hold public meetings during which the public is informed of the potential risks and benefits of the clinical trial. But not many people ever learn about the study to make a decision to opt out. According to Ross McKinney Jr., vice dean for research at Duke University School of Medicine, community consultations around Duke “reached about 450 people” -- something he acknowledged during a radio interview on March 17, was only a “tiny fraction” of Durham County's more than 1 million residents. And records on file at the FDA show that the few community members who did show up were reassured by study investigators that “in clinical trials to date, PolyHeme has demonstrated no clinically relevant adverse effects.”

What Northfield didn't tell the public is that an earlier, unpublished study of PolyHeme showed that 10 of 81 patients who received the blood substitute suffered heart attacks and two of those patients died. None of the 71 patients treated with real blood had heart attacks. The study, performed on patients undergoing repair of aortic aneurysms, was stopped prematurely in 2001. (The failure to report halted studies is causing serious ethical debate among scientists, who say that such “negative publication bias” causes researchers to reach invalid conclusions.) Northfield, confronted with the unpublished study results after they were reported in The Wall Street Journal in February, said the heart attacks were unrelated to PolyHeme and that the aneurysm patients had cardiovascular risk factors. But patients in the study who received real human blood shared the same risk factors. (Northfield also claimed that the problem in the halted study might have been physician inexperience in administering PolyHeme.) Since then, the product has not been changed. Worse, in the current study, patients with risk factors such as high blood pressure and diabetes are not excluded.

Sofia Twaddell, a Northfield spokeswoman, says that companies could simply enroll all patients in no-consent trials and that Northfield, in providing an opt-out mechanism, is going further in protecting patients than FDA emergency-research regulations require. “Nowhere [in the FDA regulation] does it say you need to have an opt-out mechanism,” Twaddell says. “The FDA's guidance suggests considering an opt-out.”

* * *

For soldiers, faced with the risk of bleeding to death on the battlefield where there is no ready access to real blood, a 5 percent or even 25 percent risk of a heart attack with PolyHeme might look like a good deal. But for civilians, who are likely to get blood within minutes, a 25 percent risk of a heart attack might tilt the risk-benefit analysis in an entirely different direction.

So why isn't PolyHeme being tested on soldiers, for whom the cost-benefit analysis is a better match? Company representatives and the military claim they can't test a blood product under combat conditions. Twaddell said it's too hard to conduct studies in battlefield conditions. “The Army did it once with hemostatic bandages and they couldn't keep track of them -- they got lost. It's very hard to keep data when people are shooting at you.” But some medical experts say that's silly. They point to earlier emergency-research studies conducted in war zones and in chaotic settings, such as steroids for head injuries and defibrillators in shopping malls, that in their details undermine claims about the problems of testing fake blood.

The real reason PolyHeme isn't being tested on soldiers may lie in the broken trust between the military and its recruits. In the wake of the Gulf War and serious concerns about the mandatory anthrax vaccine, soldiers protested and organized. An October 2004 court ruling forced the military to allow soldiers to decline the vaccine. With support for the Iraq War at an all-time low and with the difficulty President Bush faces in finding more recruits, experimental testing on soldiers may be politically risky -- especially in view of the ugly history of military testing on recruits. “Having soldiers walk under the mushroom cloud after atomic detonations -- that's not wholesome,” said Kenneth Kipnis, a bioethicist and professor of philosophy at the University of Hawaii at Manoa.

The PolyHeme study has already enrolled about 600 of the needed 720 patients, and some patients have indeed been given blood not only in the ambulance but also after arrival at the hospital, where real human blood would have been available. Worse, future projects funded under Project Bioshield are eligible to invoke the same “emergency research consent waiver,” meaning that untold numbers of patients in untold numbers of communities could be opting in just by not opting out, when they arrive in hospitals too unwell to actively refuse to participate in the study.

At least one other blood substitute, Hemopure, manufactured by a company called Biopure, appears likely to be tested using the same opt-out consent system. Asked whether the study protocol will use an opt-out or opt-in consent, Biopure's spokesman said the opt-out design is “how they are currently done.” A spokesperson for the Navy, which is running the Hemopure study, said it would be premature to comment on consent waivers because “there could be potential changes in protocol.” Let's hope.

Jeanne Lenzer is a freelance medical investigative journalist living in New York.