Placebo

3.1 Is Sham Surgery Accepted by Potential Subjects?

Even though the potential subjects’ acceptability of a clinical trial is not a de facto mandatory condition for its initiation, a low acceptability score could render the process of finding/selecting subjects cumbersome. Therefore, one of the first questions that should be asked by any investigator who aims to perform sham surgery should be related to the subjects’ acceptability of the procedure. There have been a few studies dealing specifically with this issue. Frank et al. performed a study on subjects with and without PD to assess their willingness to participate in neurosurgical trials for this disease. The investigators selected three groups of patients: with PD, without PD but with other neurological diseases (dementia excluded), and patients from primary care. They then gave each the option to be either included in an unblinded trial, a blinded trial, or not to participate at all. Most subjects from each group preferred to be included in the unblinded trial. The highest number of subjects selecting nonparticipation was in the PD group (34%, while in the other two the maximum nonparticipation portion accounted for 10.4%). Also, the PD group of subjects was the least willing one to be involved in a blinded study (24.5%, while the other groups favored this option in a percent from 35% to 40%) (Frank et al., 2008). The authors concluded: “patients with PD, when compared with patients with non-PD neurology or primary care, may have adapted to their chronic illness and may not be so desperate that they would be more eager to participate in risky research. In fact, they appear more cautious” (Frank et al., 2008). The conclusion is subjective, and not based on the actual study. Maybe, for example, the PD patients did not want to be included in the blinded trial because they were directly affected by the procedure, and felt the risks were too high. Whenever we would like to analyze the opinions of patients regarding a certain medical procedure, we must take into account all the possible reasons for a certain response, and whether their replies are in line with their actual beliefs. Such patient surveys tend to be performed more frequently and used as objective proofs, suggesting that patients agree with more controversial issues, not taking into account the validity of the questionnaires that are used or the mere fact that, for a certain procedure to be agreed upon, needs to be obtained an individual consent and not a population agreement regarding its usefulness. Moore et al., in the TransEuro project, found that subjects enrolled in PD clinical trials tend to be more educated, younger, with a higher cognitive score, and better motor function compared to patients who were eligible, but not included in the trial, and argued that this could raise problems regarding a parity of access to clinical trials (Moore et al., 2014). However, this could also be a method of protecting vulnerable subjects. Swift, in a qualitative study about the perspective of the patients and their relatives about sham surgery in PD, showed that participation was acceptable for a small majority of interviewers, but the main reasons for accepting it seemed to be the severity of the disease and the lack of good treatment options. Moreover, the surveyed persons preferred real to sham surgery; this comes to support the idea that subjects see themselves primarily as patients, that the acceptance for participating in clinical trials is not based on altruistic reasons (Swift, 2012), and that therapeutic misconception might be significant.

3.2 Sham Surgery as a Form of Mitigated Trolley

Albin believes that sham surgery could be seen as a form of mitigated trolley (2005). The trolley dilemma is often used to test the moral intuition for circumstances in which a few people are put at risk to save more. In this problem, a runaway trolley goes down a track toward five men who will be killed if it is not stopped or diverted. The trolley cannot be stopped, but it can be diverted. However, on the secondary line, there is another person who will be killed by the trolley. Therefore should one save five with the price of one life through action, or save one and have the other five die by not acting? Albin considers that sham surgery can be partially assimilated with a trolley in which the conductor diverts the line but puts a padding on the front, to cushion the impact of the trolley. The decision similar to the one made by the trolley conductor is to perform a clinical trial, and subsequently to put some people at a mitigated risk (the surgery is partially simulated, so cushioned), to aid many (Albin, 2005). What is wrong with this approach? Even if it apparently leads to a maximization of the benefits, and is often used by the supporters of utilitarian ethics in health care, we believe that it contradicts the utilitarian moral theory. Jeremy Bentham, in his book Deontology of the Science of Morality, said that an action is correct or incorrect, deserving or not, receiving approval or disapproval, reported to the tendency in which it causes the increase or decrease in the quantity of public happiness (Bentham, 1834). This means that, when we analyze whether an act is moral or immoral from a utilitarian perspective, we should examine not only the good done to the ones directly affected by our actions but also the good generated by them in the general population. If a physician saves a few lives at the expense of sacrificing one, apparently, he does more good directly; however, his actions might cause a decrease in the trust in physicians in general—Why should I, as a patient, go to a doctor if he might sacrifice me for the good of others? This decrease in trust would cause a decreased addressability of patients toward the health-care system, and a decreased therapeutic compliance, therefore causing more harm overall. Similarly, this could decrease the trust in clinical trials, and subsequently the number of potential subjects willing to participate.

3.3 Risk to Benefit Analysis

One of the major reasons for accepting sham surgeries is represented by the fact that minor risks for a few patients are considered to be less important than the overall potential benefit the results of the study might lead to, for both the subject and the population potentially benefiting from it. For example, Olanow et al. stated, as a conclusion of their study, that “This study did not confirm the clinical benefits reported in open-label trials. Furthermore, unanticipated and potentially disabling off-medication dyskinesias developed in greater than 50% of patients. We cannot therefore recommend fetal nigral transplantation as a therapy for PD at this time. It is possible, however, that enhanced benefits can be obtained in patients with milder disease, with transplantation of higher numbers of cells, and with more prolonged immunosuppression” (Olanow et al., 2003). Therefore even if more than 50% of all subjects developed “unanticipated and potentially disabling off-medication dyskinesias,” the authors concluded that more studies should be performed to search for additional potentially beneficial effects, and not to specifically search for methods of minimizing the risks for subjects, and possibly, patients.

The potential benefit for patients with PD is significant if the therapy will actually have a positive clinical effect. Another question we must ask is how would the subjects from the control group benefit from it? Normally, if a medical intervention is shown to be beneficial during a clinical trial, the subjects from the placebo group would receive the same procedure after the end of the trial; therefore, they would benefit from all the positive results of the trial without risking any unforeseen complication generated by the implantation of fetal stem cells in their brains. However, for this purpose, they would suffer two surgical interventions (one for the trial and one for the therapy), which might be associated with significant risks, especially taking into account the fact that most patients are old, with a severe pathology, and subsequently have a higher surgical risk. Performing sham surgery on patients with PD, puts them at a more than minimal risk, especially regarding anesthesia-related, which should warrant a careful analysis of the benefit differential compared to less invasive measures. Weijer considered that sham surgical procedures should be analyzed as nontherapeutic interventions, and argued that a nonsurgery control is likely a better option. According to him, therapeutic procedures should pass a test of clinical equipoise, and a harm-to-benefit analysis should be performed. Nontherapeutic procedures do not offer the prospect of benefits to the individual subject, and therefore, the harm-to-benefit analysis is not appropriate. Instead, this analysis should be replaced with the minimization of the risks consistent with a sound design and the reasonability of the risks in relation to the knowledge to be gained (Weijer, 2002). By using these principles regarding risk analysis, the use of sham surgery should be forbidden. Minimization of risks, in nontherapeutic interventions, cannot be correlated with the magnitude of the benefit; therefore, any risks that are more than minimal are in contradiction with the principle of nonmaleficence, and should not be allowed in clinical trials using sham controls.

3.4 What Role Should Collateral Benefits Have in the Decision to Allow Sham Surgery?

Beside a direct benefit for the subjects, derived from the actual therapeutic intervention, the possibility for clinical trial subjects to receive collateral benefits generated by the inclusion in the trials is sometimes discussed. In this case, all subjects would benefit from the best possible treatment (Emanuel and Emanuel, 1992; Hostiuc, 2014a,b; King, 2000). Nancy King argued that collateral benefits should not be used as types of benefits included in the risk analysis of clinical trials for two main reasons: (1) by providing a potentially higher standard of care for subjects compared to patients, we might potentially discourage the improvement of the standard treatment, and (2) as collateral benefits are under the control of the investigators/sponsors, they might become a means of manipulating or even coercing vulnerable patients to enter the clinical trial (King, 2000). We agree with her opinion, that these collateral benefits should not be taken into account as an element of the risk to benefit analysis. However, if the study generates such a benefit, the participants should receive it, as a reward for their altruistic participation in the study.

3.5 Are Placebo-Controlled Trials Actually Needed in Surgery Studies?

We presented above a few reasons for which investigators prefer placebo to BAT in randomized clinical trials. These reasons (and others) might cause them to try a slight bend of the strict ethical rules governing the use of placebo in clinical trials. For example, Dekkers and Boer argued that, for Parkinson's trials, an alternate design could consist of a core assessment procedure, in which measurement protocols are applied to the subjects before and after the surgical intervention (2001). Avins argues that unbalanced randomization might be less morally problematic in some instances, with the risk, however, of losing statistical power (1998). Macklin suggested that “Cellular-based surgical therapies have much in common with pharmacologic treatments and lend themselves to evaluation in randomized, double-blind, placebo-controlled trial” (1999). Even if this approach is viable from a scientific point of view, this is not necessarily the case from a moral viewpoint. The risks associated with anesthesia and a surgical procedure are inherently higher than in most pharmacological clinical trials, especially for older individuals with severe associated pathologies. If they are to be accepted by the potential subjects, this is because of an inherent wish to get better and to receive an experimental treatment with significant benefits from a health-care point of view (Swift, 2012; Weijer, 2002). Between these two types of therapies are also substantial differences regarding the way concepts like autonomy, therapeutic misconception, and trust are perceived by the patients. Therefore we should not try to minimize, but rather emphasize, the differences between surgical and clinical trials in order to reveal the particularities of the latter. Only by doing this the ethical issues raised in practice by sham surgery are maximized and the therapeutic misconception can be minimized.

3.6 Autonomy Versus Therapeutic Misconception

According to Lidz and Applebaum, therapeutic misconception occurs “when a research subject fails to appreciate the distinction between the imperatives of clinical research and of ordinary treatment, and therefore inaccurately attributes therapeutic intent to research procedure” (Lidz and Appelbaum, 2002). The therapeutic misconception may be caused by the patient's expectations that the investigation will act in his/hers best interest even during a clinical trial, by the lack of understanding regarding the concept of randomization, by treatment constraints associated with clinical trials, or by a wish that the study will be beneficent to them (Byrne and Thompson, 2006). There are two main responses to therapeutic misconception: to accept it as an inevitable consequence of clinical trials or to implement measures whose purpose is to reduce it, including the use of the “neutral discloser,” rewriting the informed consent forms, changes in the information algorithm used by physicians when trying to enroll a patient in a clinical trial, changes in monetary rewards, and research advertisements (Emanuel, 2008, pp. 633–644). Therapeutic misconception is especially high in fields in which the patients are highly vulnerable like oncology or psychiatry (Hostiuc, 2015). We believe that surgery could also fit in this category as a patient scheduled for a surgical procedure most likely expects a direct benefit resulting from the intervention. Moreover, sometimes even the proper information might not change his/her preconception regarding the clinical utility of the surgery. Therefore, to minimize therapeutic misconception, subjects must be explicitly informed about the sham surgery and the understanding by the subjects of this issue should be tested explicitly before they sign the informed consent.

The above-presented list is by no means exhaustive; it shows, however, the complexity of the problem and the difficulties of its ethical analysis. A series of guidelines have been developed regarding the possibility of using placebo (sham surgeries) in research, of which one of the easiest and most useful for surgical investigators is the one by Tenery et al., who based their ethical analysis on the following elements: (1) Placebo-controlled trials should only be used in surgery if there are no other designs that could lead to the necessary information. (2) Special care should be given to the obtaining of the informed consent. The potential subjects should clearly know the risk and the particularities of each arm of the study, with an emphasis on the interventions that would/would not be performed. It is recommended for a third party (not the investigator) to obtain the informed consent. (3) Placebo controls should not be used when the investigators investigate the usefulness of a slightly modified surgical procedure. (4) Placebo controls should be allowed when a surgical procedure is developed for an affliction for which there is no surgical treatment, or if the efficiency of the standard surgical procedure is questionable, and if it is known that the affliction is potentially influenced by placebo, or if the risks of the placebo intervention are small. (5) If the surgical treatment has high risks, and the standard, nonmedical treatment is efficient and acceptable to the patients, it should be offered in all the arms of the study (2002). Additionally, we believe that a first step should consist of a proper analysis of the acceptability of the sham surgery by potential subjects. Moreover, specific measures should be taken to minimize issues like coercion generated by potential collateral benefits, or therapeutic misconception.