Does Placebo Surgery-Controlled Research Call for New Provisions to Protect Human Research Participants?

Placebo surgery-controlled research (PSCR) occurs at the intersection of two large controversies in research ethics: namely, whether the scientific and ethical standards for surgical research should be as stringent, or more or less stringent, than for other clinical research; and which conditions might justify including placebo controls in research. Uncertainty and disagreement regarding the ethical justification of PSCR can be expected.

According to Franklin G. Miller (2003), the core disagreements concerning PSCR are whether placebo surgery controls can ever be justified; if so, how much risk is permissible; and which benefits might render placebo surgery risks reasonable. Commentators run the gamut on the question of what benefits might justify the risks of PSCR. Some suggest the only morally relevant benefit to consider is direct therapeutic benefit to participants (Rothman and Michels 1994; Macklin 1999). Others maintain that researchers should be permitted to invite prospective participants to enroll in PSCR involving risks of harm because of a broad range of possible benefits; possible future benefits to themselves or to someone they love who suffers from the same condition; contribution to the scientific understanding of a topic of great personal interest; personal curiosity about the condition that ails them; future benefit to strangers who suffer from the same condition; and benefit to the scientific community more generally (Freeman et al. 1999).

Miller takes a middle position. He agrees with many of the critics of PSCR that placebo surgery arms offer no prospect of therapeutic benefit to participants. However, he maintains that PSCR can be justified when the risks to participants in the placebo arm are limited, the risks are reasonable in light of the anticipatedscientific knowledge, and the study design can be expected to yield the knowledge sought. Miller attempts to convince opponents of PSCR that not all placebo surgery arms pose high risk and that some placebo-controlled surgical studies have both reduced risk to acceptable limits and answered important research questions. In addition, he recommends a cap on risk in the placebo arm.

Miller surveys the placebo arms of a number of surgical research studies in order to demonstrate that while they offer no prospect of therapeutic benefit, the risks are not necessarily unacceptable. A more thorough analysis of the risks and benefits associated with these studies challenges the assumption that the placebo arm can offer no prospect of therapeutic benefit and suggests that different conclusions are possible about the relative justifiability of the different studies. First, the cyclosporine used in the placebo surgery-controlled fetal tissue transplantation study by Freeman and colleagues, not only posed risks but also offered the prospect of therapeutic benefit to participants randomized to the placebo arm of the study. Animal studies demonstrate that cyclosporine alone improves parkinsonian behaviors, making the use of cyclosporine an intervention of possible benefit to all participants in the study (Freeman 1999; Olanow 2003). Secondly, antibiotics were provided to participants in the placebo arm not solely to blind the investigators and participants but also to minimize the risks of infection from the skin incision. Thirdly, the burr holes in the placebo surgery arm did not go through the skull but were shallow and did not penetrate the inner cortex of the skull, reducing the small, but real, risk of an epidural hematoma in the placebo arm. Finally, because any participant might require emergent intubation in the course of stereotactic surgery and because the frame can interfere with access to a participant's airway, anesthesiologists and neurosurgeons consider both general anesthesia and sedation to be reasonable options. Without engaging here in a full critique of the relative merits of the two placebo surgery-controlled studies of fetal tissue transplantation, it is clear that...