Transcript NCEPOD2002 - National Confidential Enquiry into

FUNCTIONING
AS A TEAM?
The 2002 Report
of the National
Confidential Enquiry
into Perioperative
Deaths
Study method
• April 2000 to March 2001
• Deaths on day of or within 3 days
of surgery
• First occurrence for each surgeon
Sample size
Total deaths reported
21 991
Included
20 736
Died within 3 days
7184 (35%)
Surgical Qs analysed
2114
Excluded
1255
Died between 4 & 30 days
13 552 (65%)
Anaesthetic Qs analysed
1911
Reporting all deaths
• 70/216 trusts/groups> 25% difference
in number of deaths from 99/00
• 46% of cases reported more than 3
months after death
• Approximate 4% of cases
mis-reported
• Unable to trace consultant
anaesthetist in 5% of sample cases
Facilities
• Questions previously in individual
clinical questionnaires
• Information needed by hospital
• 81% return rate
• Data questionable in some cases
Recovery facilities
• Previous anecdotal evidence of
delays caused to operating lists due
to lack of recovery beds
• 237 hospitals had less than 1.5
recovery beds/theatre
Staffing of ICUs
• 32% had less than 7 sessions/week
• 12% of ICUs had no funded
consultant sessions
• ICS guideline - 10 fixed + 5 flexible
• NCSC regulations - patients to be
seen twice daily by consultant
Recommendation
Management should ensure that an
appropriate number of funded
sessions for consultants trained in
critical care are allocated to the ICU
to allow appropriately qualified
medical staff to be available to the
ICU at all times
FUNCTIONING
AS A TEAM?
The 2002 Report
of the National
Confidential Enquiry
into Perioperative
Deaths
Percentage
Age profile
40
35
30
25
20
15
10
5
0
2000/01
1994/95
1999/00
0 to 9 10 to
19
20 to
29
30 to
39
40 to
49
50 to
59
Age (years)
60 to
69
70 to
79
80 to
89
>90
Percentage
ASA profile
50%
40%
30%
20%
10%
0%
2000/01
1994/95
1999/00
1
2
3
ASA
4
5
Referrals to the surgical team
• 295 patients (14%) were
transferred from another hospital
before their final operation
• 402 (19%) were referred by a
medical specialty in the same
hospital
Preoperative care
• 88% (430/487) of hospitals reported
pre-admission assessment clinics
Nurses
266
62%
Nurses & doctors
121
28%
Doctors
36
8%
Not answered
7
2%
Total
430
Patients in this sample
• 234 of 356 day case or elective
admissions were assessed in a
pre-admission clinic
• Only 15 were reported to have had
action taken as a result of the clinic
visit
Health professional who assessed
the patients
Nurse
129
55%
PRHO
77
33%
Surgical SHO
72
31%
Anaesthetic consultant
31
13%
Surgical consultant
9
4%
Physician
8
3%
Pre-admission assessment
• Protocols for assessment and referral
by the clinic need to be explicit
• Anaesthetists should be involved in
the development of the assessment
guidelines
• Findings of morbidity/mortality
reviews should be considered when
reviewing protocols
Delay once admitted
A 71-year-old female with no previous medical problems
was admitted at 03.00 with an acute abdomen. At this
time a HO assessed her and noted that she was shocked.
The results of her serum biochemistry investigations
were creatinine 471 micromol/l and a base excess of
-11.8 mmol/l. At 07.40 she was reviewed by a surgical
SHO who confirmed the admission findings. At 09.30 she
was referred to a surgical registrar and consultant. At
11.30 she was reviewed by a consultant anaesthetist
who agreed to take her to the ICU for resuscitation. A
bed was available there at 14.00. Laparotomy started at
16.50.
Recommendation
National protocols should be
formulated to identify which
inpatients would benefit from a
more detailed preoperative
cardiovascular assessment,
including echocardiography
Recommendation
When a formal preoperative
medical assessment is indicated, an
experienced physician, preferably a
consultant, must make it. It is the
responsibility of that physician to
fully understand the operative risks
of the patient’s medical condition
Anticipated risk of death
2000/01
1994/95
Not expected
12%
13%
Small but significant risk
17%
18%
Definite risk
53%
60%
Expected
15%
9%
Drug prescribing
Recommendation
There need to be national
guidelines for clinical prescribing
in hospitals in order to reduce the
risk of drug error
Monitoring
Direct BP
929
49%
CVP
922
48%
PA pressure
97
5%
Cardiac output
56
3%
Monitoring
• Should pulmonary artery pressure
and cardiac output be measured
more often?
• There were 18 cases where national
guidelines for minimal monitoring
were breached
Recommendation
There are national agreed
standards for anaesthetic
monitoring. The absence of an
essential anaesthetic monitor
constitutes an unacceptable clinical
risk that must be the subject of
audit
FUNCTIONING
AS A TEAM?
The 2002 Report
of the National
Confidential Enquiry
into Perioperative
Deaths
Involvement of the consultant
surgeon in decision making
100
2000/01
Percentage
80
1994/95
60
40
20
0
Involvement
Grade of most senior operating
anaesthetist
Percentage
80%
2000/01
60%
1994/95
1999/00
40%
20%
0%
Consultant
NCCG
Grade of anaesthetist
SHO
Grade of most senior operating
surgeon
Percentage
80%
2000/01
60%
1994/95
1999/00
40%
20%
0%
Consultant
NCCG
Grade of surgeon
SHO
Recommendation
The decision to operate in complex
cases can benefit from the formal
involvement of others apart from
the surgeon. Critical care specialists
should be more directly involved
Decision-making & team working
77 year-old woman under the care of a
physician with nausea, vomiting &
constipation. Four days later - perforated
viscus. Surgeon arranged laparotomy.
SHO anaesthetist called consultant who
asked for a second opinion. Surgeon
declared that he was only a technician
and could not make any decisions.
Surgery done - Hartmann’s procedure for
faecal peritonitis. Patient died in ICU
several hours later.
Problems with diagnosis
• There were 12 deaths due to acute
appendicitis
• The diagnosis needs skill and
experience
Appendicitis
21 year-old man seen in A&E by SHO
with abdominal pain & vomiting.
Tachycardia, pyrexia, urine normal &
high white cell count. Sent home as
UTI. 5 days later readmitted
moribund with peritonitis. Cardiac
arrest. ICU. Laparotomy gangrenous appendix & widespread
peritonitis. Died 24 hours later.
Recommendation
Failure to diagnose acute
appendicitis can still cause death in
fit young adults. It is essential that
experienced clinicians are available
to ensure that cases are not missed
Patients admitted under the care
of physicians
79 year-old woman admitted under
care of a physician with abdominal
pain & vomiting. 4 days later an
abdominal X-ray film showed
intestinal obstruction. Operation for
strangulated femoral hernia & small
bowel resection. Transferred to HDU
but died.
Preoperative preparation &
timing of surgery
• Physicians need to raise their
awareness of surgical conditions
existing or developing inpatients
under their care
– Planning
– Co-operation
– Teamwork
Medical Records
Patient admitted with abdominal pain &
constipation. Had been previously
investigated. Patient unclear about his
condition. Notes & X-rays not available. 3
days later perforated colon & laparotomy
done. Original x-rays still not available. In
fact 2 weeks before, a barium enema &
flexible sigmoidoscopy had diagnosed an
obstructing carcinoma of colon.
Recommendation
Non-availability of a patient’s
previous notes at the time of an
acute admission is a major
administrative failure and should
be exposed as such
FUNCTIONING
AS A TEAM?
The 2002 Report
of the National
Confidential Enquiry
into Perioperative
Deaths
Destination after surgery
2000/01
1994/95
ICU
36%
33%
HDU
7%
3%
Ward
42%
46%
Died in theatre
11%
12%
Died in recovery
4%
4%
Postoperative ward care
• Problems of
– Poor record keeping
– Hypotension
– Oliguria
Postoperative ward care
A 76-year-old ASA 3 female without recognised
co-existing medical disorders had a mastectomy
and axillary clearance. Three days later she was
found collapsed with diarrhoea, hypotension and
hypoxia. There were no entries in the medical
notes between her clerking on admission and this
collapse, at which time the entry was “low BP all
the time since mastectomy”. By this time the
patient was in fast atrial fibrillation, dehydrated
and in renal failure. Despite aggressive
resuscitation she died later that day.
Postoperative ward care
An 87-year-old female had a
cholecystojejunostomy to relieve jaundice caused
by a carcinoma of the head of the pancreas. She
was otherwise fit. At 04.00 on the second
postoperative night the urine output decreased,
but this was not reported to the on-call doctor
until 07.00, by which time it had been 4 ml/hour
for two hours. No action was taken. The SpR
ward round took place at 09.00, at which time
the patient showed clear signs of hypovolaemic
shock. Blood results showed a haemoglobin level
of 3.7 gm/dl.
Postoperative ward care
An 85-year-old man had a gastrectomy. He suffered
from type II diabetes mellitus and mild angina. He was
reviewed on the second postoperative day because of
poor urine output and hypotension. Blood gas
analysis revealed a PaO2 of 5.2 kPa and a base excess
of -7.6 mmol/l. He had a positive fluid balance since
operation of 6 litres. He had a raised JVP, a pleural
effusion, and his cardiac rhythm had changed to atrial
fibrillation. The medical SpR thought that a
cardiorespiratory cause for his deterioration was
unlikely but that he might have suffered an intraabdominal event. A laparotomy was performed later
that day. No new pathology was found.
Recommendation
If a medical team is involved in a
patient’s perioperative care it
should also be involved in any
morbidity/mortality review of the
case and receive a copy of the
discharge summary and, where
available, the autopsy report
FUNCTIONING
AS A TEAM?
The 2002 Report
of the National
Confidential Enquiry
into Perioperative
Deaths
Avoiding or diminishing
postoperative complications
•
•
•
•
Careful patient selection
Preoperative preparation
Anticipation
Early recognition
Some patients are too ill
for anaesthesia & surgery
Unanticipated intraoperative
complications
Percentage
60
50
2000/01
40
1994/95
30
20
10
0
Incidence
Recommendation
Where perioperative complications
contribute to the cause of death,
these should be recorded on the
death certificate
Recommendation
Complications may arise following
endoscopic surgery. Remedial
actions should be rehearsed
Complications after endoscopic
surgery
62 year-old woman. Laparoscopically
assisted vaginal hysterectomy, sacral
colpopexy & colposuspension for urinary
incontinence & prolapse. Problems after
surgery. Admitted 2 weeks later but
discharged. Continued to be unwell. 2
months later readmitted & ureteric
damage diagnosed. Died on table
during corrective surgery.
Autopsy rate
80
Percentage
2000/01
60
1994/95
40
20
0
Yes
Type of autopsy
100
2000/01
Percentage
80
1994/95
60
40
20
0
Hospital
Coroner
Coroner’s autopsy rate following
referral
100
2000/01
Percentage
80
1994/95
60
40
20
0
Autopsy rate
The Autopsy
• Cases in which no autopsy was
performed may not have been fully
investigated
Lack of autopsy
Fit 75 year-old man being treated for
small recurrences of bladder tumour.
Following day had massive
haematemesis and died. Surgeon
anticipated autopsy but Coroner’s
officer refused and pressurised trainee
surgical staff into writing a death
certificate. Consultant surgeon
complained but no autopsy done.
The Autopsy
• Some of the autopsies were
unsatisfactory and did not explain
the death. Problems included:
– Brevity
– Failure to examine operation
site
– Lack of clinicopathological
correlation
– Lack of histology
The Quality of the Autopsy
Report (1)
Percentage
100
2000/01
1994/95
80
60
40
20
0
His
tor
y
His
tol
o
gy
Co
r re
lati
on
Percentage
The Quality of the Autopsy
Report (2)
60
50
40
30
20
10
0
2000/01
1994/95
Ex
cel
le
Go
od
nt
Sa
tisf Poor
act
or y
Un
acc
ep
tab
le
Clinical diagnosis compared with
autopsy findings
2000/01
1994/95
Confirmation of clinical findings
75%
72%
Major Discrepancy
6%
1%
Failure to explain death (good
autopsy)
5%
2%
Failure to explain death (poor
autopsy)
10%
2%
Receipt of autopsy report by
clinical team
100
2000/01
Percentage
80
1994/95
60
40
20
0
Reports received
Problems with communication
Surgeon: ‘Our coroner does not
permit communication between his
pathologist and the surgeon unless
the surgeon has a specific
question’
Problems with communication
Surgeon: ‘I would like permission to
forward a copy of the postmortem
report on my patient to NCEPOD’
Coroner: ‘In my opinion NCEPOD is
not an “interested party”. If they
require a copy they need to apply
directly to my office’
The clinicians perspective of the
Autopsy
• The autopsy is a fundamental part
of the on-going examination of
clinical practice
• Coroners must understand their
role in supporting this requirement
Communication
• Arrangements for communication
between clinicians and pathologist
should be formalised
• Clinicians should provide a case
summary, including contact details
for further discussion
• Pathologists and clinicians should
hold multidisciplinary audit
meetings
Recommendation
Autopsies should be the subject of
a formal external audit process.
Clinicians should be involved in
evaluating the quality of reports
and the basis of conclusions drawn,
including the cause of death
FUNCTIONING
AS A TEAM?
The 2002 Report
of the National
Confidential Enquiry
into Perioperative
Deaths
Issues for consideration
•
•
•
•
•
•
Increased remit
Name
Data collection methods
Case control
Feedback
Dissemination of findings &
recommendations
Future studies
• Who Operates When II? - Publ. Nov
2003
• Gastrointestinal Endoscopy - Publ.
2004
• Critical Care & the Medical Patient Publ. 2004
• Emergency admissions - Publ. 2005
• Ruptured AAAs - Publ. 2005
FUNCTIONING
AS A TEAM?
The 2002 Report
of the National
Confidential Enquiry
into Perioperative
Deaths