Journal Logo

Ethical Issues in the Design of Randomized Trials: To Sham or Not to Sham

Dowrick, Adam S., PhD1; Bhandari, Mohit, MD, PhD, FRCSC2

doi: 10.2106/JBJS.L.00298
Scientific Articles
Free

Abstract: The placebo effect is based on the expectations of the patient regarding the effectiveness of the treatment. The high levels of stress and rituals involved with surgery can lead to a strong placebo effect. However, the ethical principles of performing sham surgery to measure any placebo effect have been questioned, and sham-controlled surgical trials are rarely conducted. While there are a number of ethical principles that must be considered to justify the implementation of a sham-controlled surgical clinical trial, four areas deserve particular attention: equipoise, risk minimization, informed consent, and deception. Particularly in orthopaedics, where equipoise is common, sham-controlled trials may be important to ensure that inferior or ineffective treatments do not become standard practice.

1Department of Orthopaedic Surgery, The Alfred Hospital, 55 Commercial Road, Melbourne, 3004, Australia. E-mail address: a.dowrick@alfred.org.au

2Division of Orthopaedic Surgery, Department of Surgery, McMaster University, 293 Wellington Street North, Suite 110, Hamilton, Ontario, L8L 2X2, Canada

When investigating the effectiveness of a treatment, well-designed and executed randomized trials are regarded as providing the highest level of evidence. By randomizing patients, the researcher aims to minimize the influence of bias on study results. As part of the thorough assessment of the safety and efficacy of a new drug, a comparison with a placebo control is often performed to determine the “true” effects of the drug over and above any placebo effects. However, the introduction of surgical procedures is rarely held to these standards of scrutiny1. While not all orthopaedic procedures are amenable to placebo-controlled trials, consideration ought to be given to exploring the use of this methodology. The ethics of performing a sham operation with no direct benefit to the patient has created debate as to the appropriateness of such a research design.

The conduct of research involving humans is subject to the principles outlined in the World Medical Association (WMA) Declaration of Helsinki2. This brief document acts as a guide to inform the policies and decisions of institutional review boards (IRBs) and federal agencies. Because the document describes principles rather than rules, it is open to interpretation, which has resulted in queries about the ethics of implementing a sham-controlled arm in surgical trials3. We discuss the potential for a strong placebo effect in surgery and describe principles to consider when evaluating or designing a sham-controlled surgical trial.

Back to Top | Article Outline

The Placebo Effect in Surgery

The placebo effect refers to a positive change in symptoms that are related to a participant’s perceptions of the treatment rather than to the mechanisms of the treatment itself. As such, it is not the pill or surgical procedure that is the placebo, but the expectations of the patient regarding the effectiveness of the treatment4. Therefore, the patient who weighs the risks and benefits of a procedure as well as the postoperative morbidity and who elects to have an operation is likely to have certain expectations of improvement5,6. Furthermore, the patient wants the decision to accept the risks of surgery to be validated by a positive outcome7. In addition, the rituals involved with surgery create an aura of expectation for the patient8. The rituals of the admission process, anesthetic and other team review, preparation for surgery, and the entrance into the surgical suite where staff are in gowns and wearing face masks all factor into heightening expectations of a postsurgical improvement in symptoms8-10. Greater placebo effects also can be expected when there are higher levels of stress11,12. All of these factors suggest that a strong placebo response can be expected with surgery. Subjective outcomes such as pain and function may be particularly susceptible to placebo effects13,14.

Back to Top | Article Outline

Principles to Consider When Deciding on the Appropriateness of a Sham-Controlled Surgical Arm

While there are a number of ethical principles that must be addressed to justify the implementation of a sham-controlled arm as part of a clinical trial2,15-18, there are four areas that deserve particular attention: equipoise, risk minimization, informed consent, and deception.

Back to Top | Article Outline

Equipoise: Is There a Best Proven Treatment?

Equipoise exists when there is doubt about the superiority of one treatment method over another19, or when an informed expert has no preference for one treatment over another for a particular patient with a given condition15. From this definition, it is clear that equipoise can be relevant at both the community and the individual level. To address the question of whether equipoise exists, surgeons must ask themselves, “Is there a best proven treatment for this patient with this injury or disorder?” In orthopaedics, the reality is that many treatments are “known,” “accepted,” or “standard” but may not yet be “proven” in the true sense of the word as it applies to scientific research. That is, given the same orthopaedic condition in the same patient, it is common for two surgeons to manage the patient with different methods or implants, and hence, equipoise at the community level is often present. According to Freedman, this is enough to justify the ethics of enrolling a patient into a clinical trial20. The reason that equipoise is considered necessary for the commencement of a clinical trial is that it seems ethically justifiable to allow a participant’s treatment to be decided by the process of randomization when there is the possibility of disagreement between two informed experts about which treatment would be better for a specific patient with a given orthopaedic condition19.

In implementing a clinical trial protocol, consideration must also be given to the equipoise of an individual surgeon. If all of the ethical principles have been addressed and a protocol is approved for a trial and is ready to be implemented, it may be unethical for a surgeon to enroll patients into the trial if the surgeon has a strong belief that one treatment arm is truly superior to the other. One example is a trial comparing outcomes in patients older than sixty years who present with a subcapital femoral neck fracture where the two arms are hemiarthroplasty and total hip arthroplasty. A surgeon who would not consider implanting a total hip in a patient who is eighty-five years old and who has limited mobility requirements would not have equipoise for this patient and would consider it ethically dubious to change his or her practice purely because the flip of a coin suggested use of a total hip arthroplasty rather than a hemiarthroplasty. For this reason, this surgeon should not participate as an investigator in the trial because a selection bias would be introduced.

Back to Top | Article Outline

Have the Risks Been Minimized?

Perhaps the most contentious issue with regard to sham-controlled surgical trials is that participants in the control arm are exposed to risks, including infection and bleeding, with no direct immediate benefit for the participant. This is in stark contrast to a medical placebo-controlled study involving the consumption of a pill, where the greatest risk comes from withholding an active treatment from the patient13. Some authors contend that the WMA Declaration of Helsinki’s principle to minimize risks means that a participant randomized to a placebo control should have no risk21. However, the counterargument suggests that much useful medical research would be unable to proceed if this was to prevail because control participants in trials frequently receive minimally invasive procedures such as blood draws and biopsies or radiation exposure from imaging that is above standard practice22. Similarly, participants in phase I drug trials of first use in humans are also subjected to an element of risk that is not necessarily lower than that for patients in a sham-controlled surgical trial13. For this reason, some have argued that IRBs should consider a differentiation between the ethical standards of clinical practice and the ethical principles of clinical research13. Furthermore, in considering the ethical principles of a sham-controlled surgery arm of a clinical trial, the reviewer should not dismiss the benefits that come with study participation23,24. Study participants generally receive additional monitoring that includes imaging, clinic visits, and patient interviews above standard practice, all of which provide ongoing attention and validation of the status of the orthopaedic patient24. Finally, if the treatment arm is found to be ineffective, the control group benefits25.

The control arm needs to be as similar to the treatment arm as possible while balancing the need for good scientific design without subjecting the control participant to excessive risk with little or no benefit26. Risk can be said to be minimized when “there is no practicable alternative method of validly testing study hypotheses that poses fewer risks.”13 As a result, groundbreaking sham-controlled surgical studies conducted in the 1950s and 1960s prior to the requirement for research to be conducted under the guidelines for Good Clinical Research Practice would not be allowed to proceed today. Where to draw the line is debatable and depends on the individual study. The final decision of what is acceptable will be made by each individual IRB contingent on how it defines minimal risks and the balance of the risk-benefit ratio.

Back to Top | Article Outline

Has the Patient Given Informed Consent?

As with any research study, the need to acquire informed consent from the patient in a sham-controlled surgical trial is paramount, although it must be emphasized that obtaining informed consent from a patient does not automatically justify the ethics of conducting a sham procedure27. To be truly defined as “informed,” consent must be given voluntarily by someone who is mentally competent15. The potential research participant must be clearly informed and given time to consider the purpose and aims of the study; the likely risks, benefits, and discomforts associated with participation; the right to refuse participation or to later withdraw consent; and the right to have any questions answered by a qualified investigator2,15.

To promote informed consent, one group with a sham-controlled surgical study design required patients to write the following statement in their own medical chart: “On entering this study, I realize that I may receive only placebo surgery. I further realize that this means that I will not have surgery on my [body]. This placebo surgery will not benefit my [condition].”28 Such novel measures may be necessary if sham surgery is to be utilized in future clinical trials.

Back to Top | Article Outline

The Need to Deceive

Sham-controlled surgery has a greater potential for the need to deceive patients than controlled trials that use a pill placebo. Because the surgeon cannot be blinded to which arm the patient is randomized to, the surgeon may need to more deliberately deceive the patient in order to maintain the blind26. This deception may need to be extended throughout the follow-up period until completion of the study. This ethical dilemma can be easily overcome by having another surgeon who is independent of the trial perform all of the necessary follow-up visits. In this way, the operating surgeon will have no ongoing interaction with the patient, thereby ceasing the continued deception of the patient by the operating surgeon26,29. At the end of the study when the blind is broken, patients should be informed as to the group they had been randomized to and debriefed22. When a treatment arm is shown to be better than sham surgery, the patient should be offered the active treatment as soon as practicable.

Back to Top | Article Outline

Examples of Sham-Controlled Surgical Trials

Three trials are most often cited when discussing evidence for and against the ethics of conducting sham-controlled surgical trials. The first involves studies that investigated ligation of the internal mammary artery, which was becoming a widely used procedure for the treatment of angina in the 1950s30,31. The ligation procedure was found to be no better than a sham operation involving a skin incision under local anesthesia without ligation of the internal mammary artery, leading to the abandonment of the procedure. These studies have been criticized for the lack of informed consent, which was typical at the time they were performed. However, they did show the power of the evidence that can be provided by a sham-controlled surgical trial.

The second example of a sham-controlled trial involved the transplant of fetal neural tissue from elective abortion procedures into the brains of patients with severe Parkinson disease32,33. Patients in these studies (including the sham surgery arm) had skin incisions and burr holes drilled into their skull, and in one of the studies, all participants received low doses of cyclosporine. These studies are controversial and have been at the center of the ethical debate about sham-controlled surgical trials. The ethical uncertainty over the use of fetal cells and the risks involved with the extent of the sham procedure have dominated the debate and have perhaps made it easier for opponents to dismiss the use of sham surgery in clinical trials.

The third example of a sham-controlled trial involved patients with knee arthritis who reported knee pain despite maximal medical treatment for six months28. Patients were randomized to three study arms: arthroscopic debridement, arthroscopic lavage, or a sham procedure. The sham arm involved intravenous sedation rather than general anesthesia and three 1-cm incisions to the knee. This study used a unique method to promote informed consent by asking patients to write in their own medical file that they understood that they might be randomized to receive a sham surgery procedure that might not benefit the arthritis28. The study showed that the arthroscopic procedure was no better than a sham procedure, and it was recommended to abandon the arthroscopic procedure. By minimizing risk, obtaining informed consent, and not actively deceiving patients, this study provided an example of how sham surgery can be used in orthopaedic trials.

Back to Top | Article Outline

Quantifying the Placebo Effect

The likely enhanced effect of the placebo response in surgical trials has implications on how sham-controlled surgical trials should be designed34. The key to the placebo effect is the blinding of patients and outcome raters to the treatment arm. If patients know what group they are in, they will have certain expectations that may dictate the outcomes over and above any placebo effect. As such, assessments of blinding integrity can be valuable to inform the validity of the research findings and the ethics of having put subjects at risk to create the blind35. Secondly, in any trial comparing a treatment with a placebo control, it may be necessary to include a third arm of “standard medical treatment” in order to be able to account for the natural course of the disease and regression to the mean, which are often put forward as possible reasons for the placebo effect14,36. The “true” placebo effect can be accounted for by subtracting any benefit of the placebo from usual treatment (i.e., nontreatment or observation) and comparing that to any benefit of the active treatment. There are also implications on the sample sizes required for sham-controlled surgical trials. Because of the large effect size generally shown by surgical studies, a smaller sample size is required in comparison to drug trials37.

Back to Top | Article Outline

Conclusion

The debate over the use of sham surgery in clinical trials has been dominated by a handful of papers in three different medical specialties. Although some of these papers have ethical queries that would likely not be approved by contemporary IRBs, they provide evidence of how patients can be put at unnecessary risk and society can be burdened with a needless financial liability when surgical procedures are not scrutinized in the same manner as new medications. Particularly in orthopaedics, where equipoise between surgeons is common, such research may be important to ensure that inferior or ineffective treatments do not become standard practice. If researchers plan their protocols to address the four principles of equipoise, risk minimization, informed consent, and deception, there seems to be no rationale to insist on a blanket ban of sham-controlled surgical trials in orthopaedics. The final decision of whether the ethical issues of a trial have been resolved will be made by the individual IRB and, ultimately, the potential participant.

Investigation performed at McMaster University, Hamilton, Ontario, Canada

Back to Top | Article Outline

References

1. Reitsma AM, Moreno JD . Surgical research, an elusive entity. Am J Bioeth. 2003 Autumn;3(4):49-50.
2. World Medical Association. WMA declaration of Helsinki - ethical principles for medical research involving human subjects. 2008.
3. La Vaque TJ, Rossiter T . The ethical use of placebo controls in clinical research: the Declaration of Helsinki. Appl Psychophysiol Biofeedback. 2001 Mar;26(1):23-37; discussion 61-5.
4. de la Fuente-Fernández R, Schulzer M, Stoessl AJ . The placebo effect in neurological disorders. Lancet Neurol. 2002 Jun;1(2):85-91.
5. Flum DR. Interpreting surgical trials with subjective outcomes: avoiding UnSPORTsmanlike conduct. JAMA. 2006 Nov 22;296(20):2483-5.
6. Sussman MD. Ethical requirements that must be met before the introduction of new procedures. Clin Orthop Relat Res. 2000 Sep;(378):15-22.
7. Mehta S, Myers TG, Lonner JH, Huffman GR, Sennett BJ . The ethics of sham surgery in clinical orthopaedic research. J Bone Joint Surg Am. 2007 Jul;89(7):1650-3.
8. Johnson AG. Surgery as a placebo. Lancet. 1994 Oct 22;344(8930):1140-2.
9. Liberman R. An analysis of the placebo phenomenon. J Chronic Dis. 1962 Aug;15:761-83.
10. Lasagna L. Placebos. Sci Am. 1955;193:68-71.
11. Beecher HK. Surgery as placebo. A quantitative study of bias. JAMA. 1961 Jul 1;176:1102-7.
12. Beecher HK. Evidence for increased effectiveness of placebos with increased stress. Am J Physiol. 1956 Sep;187(1):163-9.
13. Miller FG. Sham surgery: an ethical analysis. Am J Bioeth. 2003 Autumn;3(4):41-8.
14. Hróbjartsson A, Gøtzsche PC . Is the placebo powerless? An analysis of clinical trials comparing placebo with no treatment. N Engl J Med. 2001 May 24;344(21):1594-602.
15. Striefel S. Ethical research issues: going beyond the Declaration of Helsinki. Appl Psychophysiol Biofeedback. 2001 Mar;26(1):39-59; discussion 67-71.
16. Horng S, Miller FG . Is placebo surgery unethical? N Engl J Med. 2002 Jul 11;347(2):137-9.
17. Albin RL. Sham surgery controls: intracerebral grafting of fetal tissue for Parkinson’s disease and proposed criteria for use of sham surgery controls. J Med Ethics. 2002 Oct;28(5):322-5.
    18. Wilcox CM. Exploring the use of the sham design for interventional trials: implications for endoscopic research. Gastrointest Endosc. 2008 Jan;67(1):123-7.
    19. London AJ, Kadane JB . Placebos that harm: sham surgery controls in clinical trials. Stat Methods Med Res. 2002 Oct;11(5):413-27.
    20. Freedman B. Equipoise and the ethics of clinical research. N Engl J Med. 1987 Jul 16;317(3):141-5.
    21. Macklin R. The ethical problems with sham surgery in clinical research. N Engl J Med. 1999 Sep 23;341(13):992-6.
    22. Horng S, Miller FG . Ethical framework for the use of sham procedures in clinical trials. Crit Care Med. 2003 Mar;31(3 Suppl):S126-30.
    23. Vawter DE, Gervais KG, Freeman TB . Does placebo surgery-controlled research call for new provisions to protect human research participants? Am J Bioeth. 2003 Autumn;3(4):50-3.
    24. Stock G. If the goal is relief, what’s wrong with a placebo? Am J Bioeth. 2003 Autumn;3(4):53-4.
    25. Jones JW, McCullough LB, Richman BW . The ethics of sham surgery in research. J Vasc Surg. 2003 Feb;37(2):482-3.
    26. Angelos P. Sham surgery in research: a surgeon’s view. Am J Bioeth. 2003 Autumn;3(4):65-6.
    27. Albin RL. Sham surgery controls are mitigated trolleys. J Med Ethics. 2005 Mar;31(3):149-52.
    28. Moseley JB, O’Malley K, Petersen NJ, Menke TJ, Brody BA, Kuykendall DH, Hollingsworth JC, Ashton CM, Wray NP . A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med. 2002 Jul 11;347(2):81-8.
    29. Angelos P. Sham surgery in clinical trials. JAMA. 2007 Apr 11;297(14):1545-6; author reply 1546.
    30. Dimond EG, Kittle CF, Crockett JE . Comparison of internal mammary artery ligation and sham operation for angina pectoris. Am J Cardiol. 1960 Apr;5:483-6.
    31. Cobb LA, Thomas GI, Dillard DH, Merendino KA, Bruce RA . An evaluation of internal-mammary-artery ligation by a double-blind technic. N Engl J Med. 1959 May 28;260(22):1115-8.
    32. Olanow CW, Goetz CG, Kordower JH, Stoessl AJ, Sossi V, Brin MF, Shannon KM, Nauert GM, Perl DP, Godbold J, Freeman TB . A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol. 2003 Sep;54(3):403-14.
    33. Freed CR, Greene PE, Breeze RE, Tsai WY, DuMouchel W, Kao R, Dillon S, Winfield H, Culver S, Trojanowski JQ, Eidelberg D, Fahn S . Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N Engl J Med. 2001 Mar 8;344(10):710-9.
    34. Kaptchuk TJ, Goldman P, Stone DA, Stason WB . Do medical devices have enhanced placebo effects? J Clin Epidemiol. 2000 Aug;53(8):786-92.
    35. Heckerling PS. Placebo surgery research: a blinding imperative. J Clin Epidemiol. 2006 Sep;59(9):876-80. Epub 2006 Jun 23.
    36. Ernst E, Resch KL . Concept of true and perceived placebo effects. BMJ. 1995 Aug 26;311(7004):551-3.
    37. Olanow CW. Double-blind, placebo-controlled trials for surgical interventions in Parkinson disease. Arch Neurol. 2005 Sep;62(9):1343-4.

    Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.

    Copyright © 2012 by The Journal of Bone and Joint Surgery, Incorporated