Postmortem examinations using magnetic resonance imaging: four year review of a working service

SL/AC
26th June 2002

The Editor
British Medical Journal
BMA House
Tavistock Square
London WC1H 9JP

Dear Sir

It is disappointing that the paper by Bisset et al 1 contains
insufficient detail to reassure sceptics that magnetic resonance imaging
(MRI) is an investigation which provides a cause of death of similar
precision to “invasive autopsy” or which may be regarded as providing
similar “value for money”.

Most disturbing is the lack of detail relating to the most prevalent
“cause of death” – cardiac ischaemia or cardiac death”: what is the
pathological substrate of this description? What are the MRI appearances
which allow this “confident diagnosis of cause of death” to be given?
What were the precise findings – both of MRI and “invasive autopsy” - in
that case of “old brain ischaemia… history of ischaemic heart disease”
where the “clinical history and magnetic resonance imaging findings were
inconclusive”? (We acknowledge that this may appear carping criticism,
given the equal lack of precision in “cause of death” elicited at autopsy
– “ischaemic heart disease” does not, in our view, constitute a precise
cause of death, failing to give a precise description of what pathological
substrate causes the ischaemia).

We would welcome detail of precisely what pathology is “unlikely to
cause macroscopic changes in anatomy”, those conditions being “excluded
from examination” in this series: what are the limitations of this
imaging? It wouid be of great interest to know precisely what is meant
by “Knowing the clinical history is important for evaluating images”: that
detail might help to answer the questions raised by the authors’ comment
that “nearly half the cases referred by hospital doctors were accepted by
the coroner without any post-mortem examination” – what was the precise
role of MRI in permitting the coroner to come to that view? Could the
coroner have come to that view without MRI, or, if he had come to a view
without that imaging, would the cause of death have been different?

A similar question is raised by those “cases from general practice”
where “doctors were either too busy or unavailable to visit the mortuary
to review the body after death”: the inference we draw is that review of
the body was necessary to exclude external evidence of injury. If another
doctor were to do that and, in the absence of evidence of injury or other
suspicious circumstance, to give a cause of death based on the medical
history – the Scots “View and Grant” – then is there any need for MRI?
Unhappily, in the absence of sufficient “invasive autopsy” validation of
MRI findings in this series, these questions cannot be answered. It is
not particularly impressive that the authors suggest MRI is ‘at least as
accurate’ as certification, in the absence of an autopsy!

We may be accused of lack of impartiality, as pathologists keen to
resist radiologists ‘poaching’ on our patch. That is not our motivation:
our main concern is that this paper is insufficient to inform important
current issues. For many years this department has called for radical
review of archaic, ambiguous and anomalous law relating to death
certification, coroners and disposal: now, following the Shipman case,
the retained organs ‘scandals’, the Allitt case and other issues there has
been a Home Office Review of death certification, there is a current
Review of Coroner Services and all these issues will be considered by the
Smith Inquiry. We have an opportunity now, as has not been present in 30
years, to design a system of investigation of death, based on informed
public debate, which balances the public interest in robust investigation
of death against the individual human rights of the deceased and bereaved
in a society very different from that in which all this legislation was
framed.

The issue that lies behind the use of MRI in the coroner’s inquiry is
whether – if these cases were, as stated, ‘non-suspicious’ – an “invasive
autopsy” would be in breach of Article 9 of the European Convention on
Human Rights now incorporated into UK law? We would wish to see a system
of death investigation which concentrates on obtaining detailed accounts
of the circumstances of the death, and the medical history to allow a
fully-informed decision by a designated medically-qualified person as to
the cause of death, the need for, and method of, further examination of
the body after death – be it “invasive autopsy”, limited sampling for
(say) toxicology, or “imaging” of whatever type - and the need for an
independent, public judicial inquiry (A Coroner’s Inquiry and Inquest).
Such a system could allow death certification, registration and disposal
certification to be expedited by being performed together, decisions to be
made by those with specialised knowledge of medicine (rather than lay
coroners’ officers, or Registrars), a rational use of detailed post-mortem
examination and a close link with public health medicine, allowing
mortality data to assist in surveillance of clinical effectiveness.

It is unfortunate that this paper does not provide the detailed
information which, we feel, would have assisted the current Review of
Coroner Services: it would be interesting to know whether the extreme
brevity of the article is the authors’ personal choice of presentation or
the result of editorial policy.

1. Bisset, R.A.L., Thomas, N.B., Turnball, I.W. and Lee, S.
Postmortem examinations using magnetic resonance imaging: four year review
of a working service. BMJ 2002;324:1423-1424

Stephen Leadbeatter, senior lecturer

Ryk James, senior lecturer

Andrew Davison, senior lecturer

Deborah Cook, specialist registrar

Wales Institute of Forensic Medicine
Cardiff

It is difficult to understand how a "working service" of using MRI as
an alternative to necropsy can be deemed legally acceptable without any
validation that the "confident diagnosis of the cause of death" made by
the radiologists is correct, particularly as the fact that the cases were
referred to the coroner implied that there was an element of uncertainty
in the cause of death. I am unaware that MRI is a valuable tool for the
diagnosis of ischaemic heart disease, so what is the knowledge base and
expertise in making such a confident diagnoses by MRI? Is there any
evidence that death has no impact on MRI appearances?

It was interesting to note that the clinical history was important
for the evaluation of images. As a pathologist I can frequently predict
the cause of death from history alone; the critical cases are those where
fine macroscopical and even microscopical findings overturn the presumed
diagnosis. Without validation from a standard necropsy, I suggest that
the diagnosis of the cause of death by MRI is no better than that
predicted by history alone - unfortunately the paper does not give any
detail to either support or refute my suggestion.

Michael Jarmulowicz FRCPath