Transcript Diapositive 1 - Siapec

Autopsies: still a gold
standard ?
Sarro R.; Bosman F.; Rotman S.
IUP, Bugnon 25, CH-1011 Lausanne
Background




Pathology in Switzerland is confronted with a declining
number of medical autopsies.
This can be explained by advances in medical imaging
which offers clinicians a high level of certainty in
diagnosis. For this reason, clinicians often assume that
autopsies have no added value.
In this study, we compared clinical with autopsy
pathology diagnoses in order to evaluate the contribution
of autopsies to the control of the quality of health care.
Between January 2004 and December 2008, we
reviewed all adult autopsy cases in our Institute. A
retrospective analysis of 717 consecutive autopsy
reports was performed.
Methods




We excluded incomplete autopsy cases and also pediatric
autopsies (< 16 years old).
We confronted the clinical diagnoses listed on the autopsy
request form with pathological diagnoses in the autopsy
report.
Clinically important discrepancies were defined as:
-major :missed diagnoses for which detection and treatment
before death might have resulted in cure or prolongation of
survival or for which detection before death would probably
have led to a change in management.
-minor: missed diagnoses without clinical implication.
Sudden death was defined as death occurring less than 12
hours after the first symptoms.
Autopsy rate
CHUV AUTOPSY RATE
1000
800
600
400
200
0
Number of medical autopsies
2005


Number of patient deceaded in
the CHUV
2006
2007
2008
Number of forensic autopsies
Mean medical autopsy rate for adult patients (>16 years)
death was 10.7 % (2005-2008) and mean total autopsy rate
(medical autopsies and forensic autopsies of patients
deceaded in the hospital) was 15.8 %. Data from 2004 are not
avalaible.
For paediatric patients deceaded during the same period
(2005-2008) the mean autopsy rate in our hospital was 48%.
Study population
 Among 717 autopsy cases:
 35.6% (255) were qualified as sudden death
 59.8 % (429) concerned men
 Provenance (place of death) of the 717 cases of autopsy:
9.9 %
3.6 %
22.7%
32.3 %
31.4 %
Old age home
intensive care unit
(ICU) CHUV
CHUV (other than ICU)
Other hospital
OTHER (home, public
space,…)
Results
Reasons for requesting an autopsy
 63.6 %: cause of death without a specific medical problem
 69.3 %: specific questions:
 confirmation of diagnosis
 elucidation of complications of medical interventions
 elucidation of etiology or pathogenesis
Results
Autopsy for cause of death
Cause of death was established in 526 cases (73.3%):
 In 164 of 255 sudden death cases (64.3 %) a cause of
death was established
 In 362 of 462 non sudden death cases (78.3%) a cause
of death was established
 In 456 of 717 (63,6%) with specific question of the
cause of death, a cause was established in 66.2% (302
cases)
Autopsy with specific questions


103 (14.4%) clinically important diagnoses (major) were
established that were missed by the attending clinicians
166 (23.2%) missed diagnoses without important clinical
implications (minor) were encountered
Distribution of major, minor discrepancies and concording
autopsy diagnoses in the amount of autopsies with specific
question (69.3 %)
15.9 %
21.9 %
60.4 %
major discrepancy
minor discrepancy
concordant
other
Clinical diagnoses
Myocardial infarction
Septic choc
Autopsy diagnoses
Myocarditis
Hemophagocytic syndrome
secondary to lymphoproliferative
disease
Hemorhagic /Hypovolemic choc (2)
Acute pulmonary oedema
Pulmonary fat embolism
Infectious pneumonia
Broncho- pneumonia
Pulmonary thrombo-embolism
Adenocarcinoma of the lung (2)
Pulmonary thrombo-embolism (1)
Tuberculosis (2)
Hepatic insufficiency
Hepatic tuberculosis
Clinical diagnoses
Autopsy diagnoses
High risk coronary syndrome Mesenteric ischemia and necrosis
Stroke
Interstitial pneumopathy
Intra- abdominal hemorhage
Metabolic gliomatosis,
Post epileptic oedema
Bacterial pneumonia
Peritoneal carcinosis
Intra- abdominal lipomatosis
Peritoneal carcinosis
Abscess secondary to
diverticulitis
Invasive adenocarcinoma
Peptic ulcer
Ulcerated gastric metastasis
Major missed diagnoses without clinical diagnoses
 Invasif tumor or lymphoma discovered
24
 CT, XR, US, coronarography, biopsies, endoscopy normal 12
disclosing diagnoses of:
 Pulmonary embolism
 Coronary stenosis,
 Invasive carcinoma,
 Hemorhagic infarctus of hypophysis
 Infectious
8
 Post-operatory complication
4
 Hemochromatosis, prothetic valve thrombus, cardiac
biventricular thrombosis, chronic hydrocephalus with acute
issue, central nervous system vasculitis, Horton disease,…
Minor missed diagnoses: examples













Meningiomas
Gastro-intestinal adenomas
Other tumors
Renal angiomyolipoma
Cirrhosis
Auricular myxoma
Floppy cardiac valve
Ovary mature teratoma
Adrenal ganglioneuroma
Parathyroid adenomas
Schistosomiasis
Vascular amyloidosis
Sarcoidosis
12
23
25
8
17
2
1
1
2
2
2
5
1
Discussion



We have a low autopsy rate in our university teaching
hospital. Often clinicians do not request an autopsy and
when requested frequently family members do not
accept the request . We have no data on the autopsy
request rate.
For autopsies to function as a valid tool in monitoring of
clinical diagnostic performance, a high autopsy rate is
needed in order to avoid selection bias
The important number of minor missed diagnoses might
be partly caused by the high number of incompletely
documented autopsy request forms.
Conclusions



In a high % of cases the autopsy allowed to establish the
cause of death. In sudden death cases this is lower,
which can be explained by acute pathologies not
resulting in a perceptible anatomical substrate
Missed diagnoses with important clinical implications
continue to be found in spite of progress in medical
imaging.
autopsies should be reconsidered as essential for quality
control of health care
Web link :http://www.chuv.ch/iup/