Case and Commentary
Aug 2016

Is Consent to Autopsy Necessary? Cartesian Dualism in Medicine and Its Limitations

Megan Lane and Christian J. Vercler, MD, MA
AMA J Ethics. 2016;18(8):771-778. doi: 10.1001/journalofethics.2016.18.8.ecas2-1608.


When a hospitalization ends in death, a request for an autopsy can lead to an emotionally charged encounter between a physician and the deceased patient’s family. A case is presented in which a cardiac surgeon, believing he might have made a mistake, requests an autopsy, but members of the deceased patient’s family believe that she would not have wanted an autopsy performed. A central question discussed in this commentary is whether and when consent for autopsy is necessary. We discuss two theoretical frameworks that support differing views on this question. Beyond engaging this philosophical discussion, we also highlight a practical approach to discussing an autopsy with a grieving family by using the case presented.


Dr. Zinker, a cardiac surgeon, is surprised by the rapid and unexpected death of his patient, Ms. Foster, whose postsurgery recovery had appeared complication-free. Prior to her death, she complained of shortness of breath and was found to have a pleural effusion, an abnormal volume of fluid around one of her lungs. The pleural effusion was drained, but she developed unstable blood pressure, which required increasing administration of vasopressors. Despite these supports, Ms. Foster died. Dr. Zinker wants to order an autopsy to determine the exact cause of Ms. Foster’s death. Specifically, he wants to discover whether something he did during the surgery contributed to her unexpected death.

Dr. Zinker meets with Ms. Foster’s husband, Jim, and adult daughter, Alta, to obtain consent for an autopsy. He explains to them what’s involved in an autopsy and how long it might take. The normal protocol for autopsy includes examining both the head and the body, but Dr. Zinker explains that the family may limit autopsy to the body if so desired. He also explains that the autopsy should take a few hours and that preliminary results would be available shortly after the procedure was complete. He tells the family that final results might not be available for several weeks because academic autopsies often involve microscopic evaluation of tissues, which requires additional histologic processing. Jim and Alta are reluctant to consent to the autopsy of Ms. Foster, stating, “She’s already suffered so much. I can’t imagine putting her body through that.” They are also concerned about delaying her funeral. Finally, they ask, “Why do you want to do an autopsy, anyway? Why does it matter now?”

Dr. Zinker wonders what to say next. Is understanding what happened to Ms. Foster worth further upsetting her grieving family? How should he counsel them about the autopsy process? Should he admit that there might have been a surgical error? He contacts the hospital’s pathologist for advice.


This case illuminates concerns about postmortem care and communication revolving around a central question: For whom do we perform autopsies? Is the act of determining cause of death for the benefit of the deceased, his or her family, a physician, or future patients? The autopsy is a unique medical procedure from which a deceased patient does not benefit, yet the information gleaned from it can provide closure for a grieving family, data for quality improvement, or evidence for criminal prosecution.

In the case, Dr. Zinker wants to ask a pathologist to perform an autopsy to learn from his possible mistakes and to improve his practice. Autopsies continue to play a critical role in improving patient care and diagnostic ability. A 2003 meta-analysis found that the median error rate for clinically undiagnosed conditions “involving a principal underlying disease or primary cause of death” was 23.5 percent [1]. The autopsy is a powerful tool for student learning as well as improvement in care. For example, the Accreditation Council for Graduate Medical Education requires residents in anatomic and clinical pathology to perform at least 50 autopsies and to review autopsy reports with a faculty member [2]. Although the autopsy is a critical component of medical education and clinical development, performing autopsies can conflict with the wishes of stakeholders.

Living individuals—for example, family members, physicians, criminal investigators—have competing interests in the dead. To determine whether consent for autopsy is ethically required in this case, it is necessary to discuss the status of Ms. Foster’s body after death. Should her remains be considered a deceased patient who is entitled to the respect expressed toward the living or an inanimate object that physicians and interested parties can act upon at will? Or, is there a kind of middle ground that should be considered? In our pluralistic society, settling a larger metaphysical question of what happens to the self or identity after death is impossible. Hence we will examine the general Cartesian view of the body after death that pervades contemporary clinical practice and then take a clinical sociocultural approach in an effort to respect multiple interpretations of the quiddity of a corpse. We will argue that a plurality of beliefs about the body after death call for consent to autopsy in most circumstances. We then consider how physicians might approach families of deceased patients to request an autopsy in light of this perspective.

A Cartesian Perspective on the Postmortem Body and Its Limitations

Cartesian dualism. Cartesian dualism, born of sixteenth-century rationalist philosophy, provides a framework in which Ms. Foster’s death would completely separate her body from her mind and past identity. René Descartes’s cogito ergo sum (“I think, therefore I am”) posits the existence of the self in the (immaterial) mind as opposed to in the (material) body [3]. According to this view, in death, thinking stops, and so the body is separated from the mind and the thinking, reflective “I” no longer exists there. In death, the body becomes an object, and the person is no longer embodied: “even if there were no body, the soul [mind] would not cease to be all that it is” [4]. Dualism permeates discussions of death in the medical community; specifically, this principle provides a basic philosophical underpinning of medical definitions of death. Harvard Medical School’s 1968 “Definition of Irreversible Coma” defines as a “new criterion of death” irreversible coma, in which “individuals … have no discernible central nervous system activity” [5]. The 1981 Defining Death report [6] provides two definitions of death pivoting around the function of the brain: the “higher brain” definition, which defines death as a cessation of higher cortical functions that make “consciousness, thought, and feeling possible” [7] and the “whole brain” definition, which posits the brain as the primary regulator or the “integrated functioning of brain, heart and lung” [8]. Concluding that the “higher brain” “may well exist only as a metaphorical concept” [9], the report espoused “whole brain” formulations in its proposed Uniform Determination of Death Act, which the authors claimed “does not appear to conflict with the view that the soul leaves the body at death” [10]. With the death of the brain, does the body become an object, on a Cartesian view?

Whatever the answer to this question, Cartesian dualism influences medical definitions of death and can also influence a practitioner’s view of the postmortem patient. If the dead body is an object, we would not expect this object to have any independent wishes, thoughts, or desires, and, presumably, there would be no autonomy to respect within a consent process. If Ms. Foster’s dead body is an object, one could argue that Dr. Zinker would not need consent for an autopsy of her body.

Limitations to dualism. Although Cartesian dualism is consistent with some rationalist underpinnings of the practice of medicine, such as the medical determination of death, a Cartesian perspective on the dead body as object ignores a complex plurality of cultural and social perspectives on the body after one is deceased. John Drayton, in “Bodies-in-Life/Bodies-in-Death,” suggests that the objective use of the body dictated by dualism is complicated by bereavement, or “our memories of others…based on behaviour associated with their bodies” [11]. D. Gareth Jones, author of Speaking for the Dead: Cadavers in Biology and Medicine, also points to this idea:

When we turn to a cadaver’s instrumental value, we recognize that it serves as a vital source of memories and responses…. As we remember a person who has died, we respect the person who was. All that remains of the person is the cadaver, and yet our respect for that person, and for the memory of that person, leads to respect for the person’s remains, a link that is not readily broken [12].

Both Drayton and Jones suggest that a person’s body holds unique value beyond that of a lifeless object. This worth is constructed through others’ memories of and respect for the deceased person, and it is exercised in a broad range of social, cultural, and spiritual traditions surrounding disposal of the dead. Drayton and Jones’s emphasis on the continued value of the deceased’s identity in relation to the deceased’s body highlights the clinical and ethical importance of consent for autopsy and complicates clinical understandings of death influenced by dualism.

Respect for Sociocultural Traditions and Consent

Respecting sociocultural differences. Given the value of the body after death expressed by thinkers like Drayton and Jones and that a deceased person’s wishes can be preserved through living relatives, it follows that family members’ consent should be taken into consideration to proceed to an autopsy. As an extension of family members’ connections to the patient, the patient’s perceived desires, wishes, and sociocultural beliefs should be incorporated into the family’s decision about whether to perform an autopsy, analogous to surrogate decision making, in which next-of-kin must rely on an incapacitated patient’s past attitudes, actions, and values to make health care decisions. Like clinical decisions, the consent process and autopsy can be adjusted to accommodate the sociocultural beliefs of an individual and his or her family while gaining the information that can prompt a clinician’s request for the procedure.

An example of this conceptual accommodation occurs in the cooperation between New Zealand forensic pathologists and Māori patients and family members. Māori cultural beliefs about death and the body conflict with allopathic forensic practices of organ retention and biomaterial destruction, but accommodations can and have been made to honor Māori beliefs [13]. This approach acknowledges that the value of a dead human body can transcend that of an object, and that the dead person’s identity can be preserved and respected through family members.

The clinician should discuss the patient’s and family’s views and values in a consent process to autopsy, similar to all other aspects of clinical care of the patient. In the case presented, it would fall to Dr. Zinker to obtain consent from Ms. Foster’s husband and daughter and try to understand their sociocultural background to best accommodate Ms. Foster’s perceived wishes. By taking a sociocultural approach, the physician can try to honor the autonomy of the deceased person by allowing the next-of-kin to communicate their perceptions of that person’s wishes and desires and by respecting the deceased’s spiritual and cultural perceptions of death.

An exception to this approach. There are some circumstances in which consent to autopsy is not required, most notably in forensic cases when information from an autopsy might lead to prosecuting or convicting someone who caused harm to the deceased. This exception is understood to be a compromise that expresses respect for individual autonomy and motivates collective justice. Information from an autopsy can be used to gain information about a crime and prevent such crimes from occurring to others in the future. In the case presented, if Dr. Zinker believed that a mistake he made would be systematic and that more of his patients could be at risk of dying under similar circumstances, the protection of others in the future could justifiably override respect for autonomy. If Dr. Zinker’s mistake occurred out of negligence and could be reproduced in the future, the protection of others could also override deference to Ms. Foster or her family’s beliefs. Neither scenario seems to be applicable in this case, however.

An Approach to Consent for Autopsy

The benefits gained by performing an autopsy are not so great as to immediately trump potentially competing values held by a deceased patient’s family members. In a forensic case the need to pursue justice, as expressed in long-standing legal precedent, overrides even family members’ objections to autopsy. Approaching a family for autopsy consent after a loved one has died (regardless of the cause of death) is nearly always difficult. In addition to concerns about suffering (suggested in the case above by the response Ms. Foster’s family gave Dr. Zinker), lack of information about why an autopsy might be needed and the steps of the procedure itself have been identified as leading reasons for autopsy refusal [14]. In our case, Dr. Zinker should clarify why he is suggesting the procedure, communicating to the family that he wants to determine whether something he did during Ms. Foster’s surgery could have contributed to her death. Error disclosure is now a well-established practice in medicine, and the Michigan model disclosure method has been shown to reduce legal claims and liability costs [15]. In following the Michigan model, Dr. Zinker should “reach out to those harmed, be honest, explain where appropriate, disclose [his] mistakes, and learn from [his] experiences” [16]. By explaining that he might have done something during surgery that contributed to Ms. Foster’s death and that an autopsy could bring that to light and prevent it from being repeated, Dr. Zinker could better communicate clearly and honestly with Ms. Foster’s family. He might say, for example:

I was surprised and saddened by Ms. Foster’s death. I have to tell you that I thought her surgery went very well, and it is shocking to me that she died so suddenly and unexpectedly. I know she has been through a lot, and this seems like putting her through even more—but I would like to know whether anything I did contributed to her unexpected death. If so, I want to know, so that I can apologize to you and so that we can work to prevent outcomes like this—if, indeed, they are preventable—in the future.

Dr. Zinker could also employ the aid of a bereavement counselor when communicating this message. A bereavement counselor is a specialist in the grieving process who could potentially aid Dr. Zinker in communicating effectively with her family. An audit of the necropsy rate in one hospital department in the United Kingdom found that the patient affairs officer was 18 percent more successful in securing consent for autopsies than junior physicians [17]. The authors speculate that the patient affairs officer had a higher comfort level in requesting autopsies and better communication skills than the junior physicians [17]. If Ms. Foster’s family continues to refuse the autopsy after a clear explanation of its perceived need, this refusal should be honored, as would be expected were a living patient to refuse a procedure.


The term “autopsy” comes from the Greek term autopsia, meaning “a seeing with one’s own eyes” [18]. We tend to project our own varied perspectives of death and our desires for knowledge onto the acts of requesting and consenting to an autopsy. In medicine, we are drawn to a Cartesian dualism to characterize death. There is the tendency to forget that a patient is still physically in the hospital after CPR has ceased or that a brain-dead patient might still be viewed as living to some family members. We live in a pluralistic society with many views and rituals of death, and imposing one view in which the body is an object that can be used for information without consent can potentially harm patients and families whose beliefs differ. In order to preserve autonomy and ensure beneficence, a discussion about the benefits and risks of autopsy with the family is paramount.


  1. Shojania KJ, Burton EC, McDonald KM, Goldman L. Changes in rates of autopsy-detected diagnostic errors over time. JAMA. 2003;289(21):2850.

  2. Accreditation Council for Graduate Medical Education. ACGME program requirements for graduate medical education in anatomic pathology and clinical pathology. Effective July 1, 2016. Accessed June 9, 2016.

  3. Descartes R. Discourse on Method and Meditation on First Philosophy. Cress DA, trans. Indianapolis, IN: Hackett Publishing; 1980.

  4. Descartes, 18.

  5. A Definition of Irreversible Coma. JAMA. 1968;205(6):85.

  6. President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Defining Death: Medical, Legal and Ethical Issues in the Determination of Death. Washington, DC: US Government Printing Office; 1981. Accessed June 9, 2016.

  7. President’s Commission, 38.

  8. President’s Commission, 6.

  9. President’s Commission, 40.

  10. President’s Commission, 42.

  11. Drayton J. Bodies-in-life/bodies-in-death: social work, coronial autopsies and the bonds of identity. Br J Soc Work. 2013;43(2):267.

  12. Jones DG. Speaking for the Dead: Cadavers in Biology and Medicine. Farnham, England: Ashgate Publishing; 2000:57.

  13. Hudson ML, Allan CA, Bedford KR, Buckleton JS, Stuart K. The impact of Māori cultural values on forensic science practice in New Zealand. J Forensic Sci. 2008;53(2):380-383.
  14. Hull MJ, Nazarian RM, Wheeler AE, Black-Schaffer WS, Mark EJ. Resident physician opinions on autopsy importance and procurement. Hum Pathol. 2007;38(2):342-350.
  15. National Patient Safety Foundation’s Lucian Leape Institute. Shining a light: safer health care through transparency. Boston, MA: National Patient Safety Foundation; 2015. Accessed May 4, 2016.

  16. National Patient Safety Foundation’s Lucian Leape Institute, 29.

  17. Kamal IS, Forsyth DR, Jones JR. Does it matter who requests necropsies? Prospective study of effect of clinical audit on rate of requests. BMJ. 1997;314(7096):1729.

  18. Autopsy. Dictionary.com Accessed May 20, 2016.


AMA J Ethics. 2016;18(8):771-778.



The people and events in this case are fictional. Resemblance to real events or to names of people, living or dead, is entirely coincidental. The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.