"You’re Telling Me I Was Given Artificial Blood?":
The Controversy over Emergency Research

By Mary Anderlik
Health Law & Policy Institute

An ambulance arrives at the scene of a grisly traffic accident. EMS personnel discover that one of the drivers is still alive, but bleeding heavily from a stomach wound. They rush her to the hospital for surgery, pumping her body full of IV fluids on the way. The next day, the same ambulance arrives at the scene of a shooting. The victim is rushed to the hospital, but she is not given IV fluids until she is wheeled into surgery.

A man injured in a motorcycle accident is brought to the emergency room, unconscious and bleeding. The doctors immediately begin to administer a liquid, ignorant of whether it is an experimental blood substitute or a salt water solution.

In each of these vignettes, a person has been enrolled in a research study, a medical experiment, without his/her knowledge or consent or the knowledge or consent of a family member or other surrogate. Can that be legal? Can it be ethical?

The first vignette is derived from a research study conducted in 1989-1992. The common practice in the U.S. was to administer IV fluids to trauma patients immediately. That practice rested on animal studies conducted in the 1950s and 1960s showing benefits from perfusion of vital organs. Researchers in Houston, reviewing subsequent animal studies, believed that timing might be important, i.e., administration of fluids to trauma patients before bleeding was controlled through surgery might do more harm than good by accelerating blood loss. They designed and carried out a study on trauma patients, relying on those patients’ "implied consent" to be research subjects. On even-numbered days, patients were given IV fluids immediately; on odd-numbered days, IV fluids were delayed. The results, published in the New England Journal of Medicine in 1994, weighed against the common practice: 70% of patients in the delayed group survived to discharge versus 62% in the immediate group. Letters debated the proper interpretation of the results, but no one questioned the common practice of relying on implied consent in emergency research.

Implied consent is well-established in the case of a patient who is unable to communicate, unaccompanied by a surrogate, and in need of immediate medical treatment. But the extension of this concept to research eventually raised questions. In 1996, the federal Food and Drug Administration issued a new rule describing the conditions under which emergency research would be permitted without prior informed consent. The FDA regulates research involving drugs and medical devices. The Department of Health and Human Services, through the Office for Protection from Research Risks, also regulates research. In connection with the issuance of the 1996 rule, the two agencies agreed to harmonize their standards for emergency research.

Under the 1996 rule, researchers requesting a waiver of the requirement of prior informed consent must establish that: (1) potential subjects are in a life-threatening condition; (2) existing treatments are unproven or unsatisfactory, the value of the new intervention can only be determined under emergency circumstances, and the research would not be feasible without a waiver of the prior informed consent requirement; (3) the risks associated with the new intervention are reasonable given the patient’s medical condition and standard treatments; and (4) the prospect of direct benefit to subjects has been established by animal and pre-clinical studies. Local institutional review boards or IRBs are charged with ensuring that these conditions are met, and the waiver must also approved by the FDA.

The rule includes some additional safeguards for subjects, e.g., efforts must be made to contact a surrogate in advance of the experimental intervention and must be continued after the intervention, at any point where patient or surrogate consent becomes possible it must be sought and the option of withdrawal from participation must be presented, and an independent data monitoring committee must exercise oversight over the research. A real innovation in the 1996 rule is a provision that requires public disclosure of the proposed research and consultation with the community from which subjects may be drawn. Finally, the rule leaves in place any state laws or regulations. In May of this year, the Arizona law governing consent was amended to authorize emergency research without prior consent under certain conditions. The Florida Board of Medicine is expected to take action on the issue this month.

In answer to the first question, then, this kind of research is now legal—at least in some states. The second of the two opening vignettes is based on the first study to be approved under the new rule. That study concerned HemAssist, a blood substitute. Artificial blood, if effective, has several advantages. There is no need to type and match—a significant benefit in an emergency. Artificial blood has a longer shelf life than real blood, and it would be the answer to persistent blood shortages. The research involved two group of subjects: trauma patients, for whom consent was waived, and elective surgery patients who were able to consent. Subjects were randomly assigned to either receive HemAssist or a salt water placebo; the study was blinded, meaning that the treatment team did not know if patients were in the experimental or "control" subgroups. Based on preliminary data showing a higher survival rate for the control group, the research with trauma patients was halted in April of 1998. (The study of elective surgeries continued.)

Is this kind of research ethical? Defenders of the FDA rule point to the past and anticipated benefits of emergency research and the protections for subjects built into the regulations. Critics are concerned about potential exploitation of minorities and distrustful of researchers. They argue that community consultation is not a substitute for individual consent, either as a check on researcher self-interest or as an expression of the principle of personal autonomy. Others have a more practical concern. They worry that patients and families will be stuck with the bill, since clauses in insurance policies often exclude anything labeled "experimental" from coverage. (This would not be a concern where a pharmaceutical company or other interested party pays for the intervention.) Principle aside, one might question whether the FDA’s conditions have been met in the research studies approved. Could the value of the intervention only have been determined under emergency circumstances? Was the community consultation process a meaningful attempt to inform the public and elicit feedback?

Currently, the National Bioethics Advisory Commission is considering the possibility of "advance directives" for research participation. Some day we may all carry around a card that states our willingness (or unwillingness) to participate in emergency research. Unfortunately, advance directives in the area of end-of-life care have not lived up to expectations. The solution to the problem of emergency research may not be simple.

10/27/98