Issue 23 / 22 June 2015

HOSPITAL autopsies are the “ultimate medical audit” and their decline could have long-term consequences for student education, research, treatment validation and quality assurance, according to a leading pathology expert. 
Professor Roger Byard, senior specialist forensic pathologist with Forensic Science SA and professor of pathology at the University of Adelaide, told MJA InSight that hospital autopsies were “the only way to verify that new diagnostic and imaging tools are working”.
Professor Byard was commenting on a study published last week in the Journal of Clinical Pathology that found the practice of hospital autopsy in the UK was “on the verge of extinction”. (1)
The research was based on data from 184 National Health Service (NHS) trusts across England, Scotland, Wales and Northern Ireland on the number of autopsies performed in 2013 as a percentage of the total inpatient deaths. 
As previous research indicated that autopsy rates were higher in stillbirths, neonates and young children, the researchers excluded data from these categories. They found a mean autopsy rate of 0.69% of all UK hospital deaths. 
The authors wrote that their results confirmed the continued decline of hospital autopsy, with 23% of all UK NHS trusts reporting that they had not performed any autopsies during 2013, saying future research was needed to focus on the impact of this decline on patient safety, audit, research and teaching. 
“While debate continues over the value of hospital autopsy in medical practice, if action is not taken imminently, the practice may disappear”, the authors wrote. 
Professor Jane Dahlstrom, professor of anatomical pathology at the Australian National University and staff specialist at Canberra Hospital, told MJA InSight that the declining autopsy rate identified in the UK study was applicable to both Australia and the rest of the developed world. 
“One reason is that doctors might think that with current imaging technology, they already know why the patient has died and don’t need an autopsy to confirm that.”
However, Professor Dahlstrom said that despite improvements in technology, “some doctors do get it wrong”. 
Professor Ibrahim Zardawi, conjoint professor at the University of Newcastle’s school of medicine and public health and senior pathologist at Douglass Hanly Moir Pathology, agreed, telling MJA InSight that hospital autopsies were crucial in determining the clinical appropriateness of the treatment the patient received. 
“Sometimes we just don’t know whether a certain management system we applied to a patient was effective, and without a final examination after death, there is no absolute arbiter.”
Professor Dahlstrom said that from the perspective of the patient’s family, the “bad wrap” autopsies received in the mainstream media and popular culture and the delay in being able to arrange a funeral also contributed to the decline in autopsy rates. 
However, she said Australia had a much higher frequency of perinatal autopsy than the UK, with the ACT reporting a rate of 53.5% for the period 2006‒2010. (2)
Professor Dahlstrom said there were two driving factors for performing a perinatal autopsy — the desire of the family and treating doctor to understand why the child died, and the need for guidance in planning for future pregnancies.
She said that apart from determining cause of death, autopsies could also provide national data, identify adverse drug reactions and serve as an education tool. “A hospital autopsy is often the first time a medical student will see a person who is dead, and they get a systemic view of all the organs in a body as a whole.”
Much of the medical community’s breakthroughs in understanding diseases such as HIV, legionellosis and mesothelioma had come from performing autopsies, Professor Dahlstrom said.
Associate Professor David Ranson, deputy director of the Victorian Institute of Forensic Medicine, told MJA InSight that hospital autopsy offered an important “quality assurance measure” for clinical practice. 
However, he acknowledged that the move away from routine hospital autopsy was a “vicious cycle that is hard to break”. 
“With a reduction in the number of hospital autopsies being performed, there’s a lesser need for pathologists-in-training to learn how to do it”, Professor Ranson said. 
Professor Roger Byard agreed, saying that improving hospital autopsy rates had to be a combined effort.
While doctors should be requesting autopsies more regularly, “pathologists also need to stop putting all the blame on clinicians”, Professor Byard said.
“Doctors and pathologists both need to work together more closely and realise the true value of performing hospital autopsies.”
(Photo: Mauro Fermariello / Science Photo Library)

9 thoughts on “Autopsy nears “extinction”

  1. Kevin B. ORR says:

    As a student at the Royal Prince Alfred Hospital at the University of Sydney in trhe late ’40s, I found the learning experience much better than lectures or books. That depended, of course, on the teaching ability of the pathologist. Anatomy and the cause  of human illness was so much easier to understand but, of course ,needs the added help of microbiology and all those other “miracles” that appear regularly in this 21st Century.

    Could autopsies be done on an adhoc basis chosen from those patients not considered to require one; with suitable permission from patient, relatives, etc. of course?  

  2. Michael Nixon says:

    Autopsies are an excellent teaching resource for medical students and also for doctors who would like confirmation of their diagnosis.

    There is no harm in doctors suggesting to relatives (and even the terminally ill patients) that the deceased be offered an autopsy for the above puropses.

    A further suggestion in large teaching hospital areas is to donate their body to medical science – (Or do we already have enough cadarvers?)

  3. Peter Nixon says:

    in the  USA, the 1910 Flexner Report on medical education resulted in closure of many substandard medical schools.  One requirement of the report was that teaching hospitals should have a very high (near 100%) post mortem examination rate as an accreditation requirement.  How would Australian medical schools fare today?

    Post mortem examination rates of 100% are achievable given the appropriate culture.  in 1961-63 when I was a resident at Sydney Hoepital, the Pathologist, the late Eddie Hirst, was a great teacher who cultivated that culture and, with the aid of “extras” coming from the coroner, that thospital usually achieved an autopsy rate of >100%.

    Many factors mitigate against a high autopsy rate in current teaching hospitals, but perhaps the greatest is the cost of an autopsy and current budget measures.  The cost of autopsies needs to be subsumed for accreditation purposes, rather than charged to either the unit of the treating doctors or to the pathology department. 

    Personally, I learned more anatomy from post mortem examinations than from those first anatomy dissections;  perhaps because it was review of something first learned (inadequately) in the overcrowded dissection room.  Of course, post mortem examinations were also a great way to learn pathology; and pathology is basic to an ordered understanding of disease processes and thus of diagnosis. 

  4. Hugh Dillon says:

    The reality is that hospital autopsies are virtually defunct, at least in western societies. In the distant past, many families were not consulted about hospital autopsies — they happened almost automatically.  The demise of the hospital autopsy is due to multiple reasons but one of them is that there is growing respect for the wishes of the families of people who die in hospitals and another is respect for the bodies of those people. Another is that doctors seem largely incapable of persuading the families of the benefits of conducting autopsies.  Their own curiousity is generally insufficient. I do not expect to see hospital autopsies revived to any significant extent.

    While I accept that this means that students and interns as well as more experienced clinicians do not have the benefit of an internal post mortem examination in most cases, at least in some the opportunity may still be there. I am a coroner at the NSW Coroners Court.  Each day approximately 10 bodies arrive at the mortuary.  Some are from hospitals.  Autopsies are not conducted in all cases. Nevertheless a number are carried out. There seems no reason in principle why students, interns and experienced clinicians could not avail themselves of the opportunity to observe autopsies carried out by the forensic pathologists, or, at the very least, discuss the cases with the pathologists. Yet this seems to happen quite rarely.  Why?


  5. Nigel Buxton says:

    There is also a reduction in Coronial autopsies with a tendency to push for a death certificate rather than proceed to either full or limited autopsy. Some certificates are quite inappropriate eg 1a Suspicious heart failure”! The onus is on the police to obtain the certificate and their appreciation of disease is rudimentary. There is little appreciation of dying with a disease and dying FROM a disease. Also, whether the mode of death is appropriate for the disease that the certificate provides as a cause of death. Eg sudden collapse with the cause of death being given as Bowel Cancer with no clinical evidence of deterioration.

    If natural deaths are not examined, that skill will be lost – and the subtle changes seen by trained Autopsy pathologists (hospital or Forensic) will be missed by the inexperienced. If autopsy rates drop – there will be insufficient trained Forensic Pathologists – and then there is the risk that unnatural deaths may be missed, clinical complications may not be appreciated an dgereral standards will drop.

    Rarely do Clicians ring to discuss cases (they used to attend as previous contributers have said). The only folk that ring me are the Intensive care para-medics wanting to find out why one of their transfers succumbed to injuries in spite of their expert care!!

  6. Kevin A Lee says:

    In my years as a forensic pathologist. I often worked in hospitals where coronial and hospital autopsies were carried out simultaneously. The hospital cases were numerous and invariably attended by senior clinicians down to the medical students on the firm. Cross-pollination occurred, as forensic cases often showed untreated and unaltered pathology, and forensic pathologists learned anatomical pathology subtleties.


    Over time, the clinical autopsy numbers and their attendees shrank to almosty nothing. I heard clinicians saying “I don’t want autopsies. I don’t want you finding out where I’ve gone wrong.” or “He can’t have an XYZ  because the CAT scan was normal”. Eventually I had a senior registrar nearly faint because he had never seen the inside of a dead body before.


    There are many causes for this diminution. There is increasing time pressure on clinicians and pathologists (some pathologists avoid notifying the autopsy times as case demonstration slows them down). Our faith in medical technology increases, but still the autopsy is the final arbiter of the pathology. As autopsies become less common, the least experienced members of the team are given the ever more delicate process of the request. An autopsy is not the clean “lifting of the corner of the sheet” seen on TV and many medicos are now un-used to gross pathology, in both senses of the term. I could go on and on.


    Not much has changed since Rudolf Virchow wrote of the pleasure of studying and learning the human body in order to improve our practice of medicine. We cannot return the autopsy to its previous position, but we will do ourselves little service by ignoring its possible contributions.

  7. Judith Alison Fryer says:

    Now retired, I spent more than 30 years as an Anatomical Pathologist and learned, from my very first job as a Teaching Fellow in a university, the value of the autopsy in medical education. Attendance at autopsies in the Teaching Hospitals provided opportunity for students to actually visualise disease processes and come to grips with the secondary effects.

    Correlation of autopsy findings with clinicians was always rewarding and there were sometimes surprises. My major interest in Neuropathology led to many interesting and valuable sessions with radiologist colleagues in the years during the development of CT and MRI.

    Increasing dependence on imaging technology together with economic and cultural factors have contributed over the years to the fall in the autopsy rate. However we cannot abandon this valuable tool entirely. Treating doctors require special skills in obtaining permission for post mortem examination.

  8. Sue Ieraci says:

    There is a fundamental reason why the role of autopsy has changed – diagnostic methods are better. With the advent of sophisticated imaging techniques, we generally no longer need to wait for the person to die to make a diagnosis. By all means investigate suspicious or undiagnosed causes of death, but there is no need to lament the change in role for autopsies – it’s progress, and it’s for the better.

  9. William Slater says:

    Strongly disagree with Sue. How can you be sure?? There is no dispute that autopsy rate is very low…. have these sophisticated imaging techniques actually been throughly tested against the ‘gold standard” of direct anatomical visualisation? I very often hear my surgical colleagues remark that they found much more pathology than was anticipated ….and on the very rare ocassion where an autopsy (oronial or otherwise) is done unexpected and clinically signifcant patholoy is often found. Progress? I don’t think so.

Leave a Reply

Your email address will not be published.