Establishing a confidential Maternal Death Enquiry: the Irish experience
A report of the National Confidential Enquiry into Patient Outcome and Death (2008) Death following...
-
Upload
alexis-sharlene-elliott -
Category
Documents
-
view
230 -
download
1
Transcript of A report of the National Confidential Enquiry into Patient Outcome and Death (2008) Death following...
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Death following a first time, isolated coronary artery bypass graftThe heart of the matter
Study aim
• The aim of the study was to examine whether there are
identifiable changes in care processes, including the
functioning of cardiac teams, that impact on patient
outcome following a first time isolated CABG.
Patient inclusion/identification
• Three year period – 1st April 2004 – 31st March 2007
• All units in England, Wales, Scotland and Northern Ireland– NHS and Independent sector
• Cases - patients that died in hospital following surgery– Direct from units
• Controls - patients discharged alive following surgery– CCAD or direct from the unit
Patient identification
• Office of Population Census and Surveys (OPCS) codes – K40 – Saphenous vein graft replacement of coronary artery– K41 – Other autograft replacement of coronary artery– K42 – Allograft replacement of coronary artery– K43 – Prosthetic replacement of coronary artery– K44 – Other replacement of coronary artery– K45 – Connection of thoracic artery to coronary artery
• Or by defining the operation as CABG only as defined in the minimum data set of the SCTS
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Referral and admission process
Referral and admission
• Referrals via a variety of different routes
• Range of admission processes
• Use of local networks of cardiac care using a ‘hub and spoke’
model
• Referral protocols, treatment protocols and quality improvement
methods should be in place within these networks
NSF Coronary Heart Disease Department of Health 2000
Use of protocols for referral
• There were discrepancies in the use of protocols as described by surgeons and those reported to be present in each unit– In 674/910 (75%) patients a standard written protocol was used
for referral• even though in 349 patients a standard written protocol did not exist
in cardiothoracic unit according to OQ!
• Surgeon completing SQ may have interpreted the question differently than person completing OQ.
• There was a 8% stated increase in use of written protocols over the three years of the study
Case study 1
An elderly patient with triple vessel CAD with ACS and cardiac failure was transferred from another hospital directly to the cardiac ICU.
The referral was made between a cardiology SHO to a senior nurse on the ICU without the knowledge of the CT team. By the time that the consultant CT surgeon was informed the patient had developed renal failure, had an EF of <20% and required the use of an IABP with inotropic support but was conscious.
A decision to undertake CABG was made despite a stated mortality of >50%. The patient died five days postoperatively from multi-organ failure. No autopsy was performed.
Advisors comments:
• Lack of senior level communication between the referring cardiologists and cardiothoracic surgeons for this urgent high risk patient
• Inappropriate for a nurse to accept such a patient on to the ICU
• Was the surgeon put under undue peer pressure to proceed with surgery in view of the likely outcome?
• The advisors wonder whether palliative care would been more appropriate for this patient
Case study 1
Pre-admission assessment
• 49/58 of CTU had a pre-admission clinic
• 44/49 of CTU used a pre-admission proforma
• 34/49 of CTU used an ICP– In only 33 % (272/821) of patients were ICPs used
• non elective patients in the study sample
– Many of the ICPs were not fit for purpose because they did not contribute to the patient care pathway
– Lack of standardisation– No minimum data set of information
• Personnel performing pre-admission assessment – 23/49 (47%) cardiothoracic surgeon and nurse – 9/49 (18%) nurse only– 9/49 (18%) included an anaesthetist
• Variation in personnel involved and function
National Good Practice Guidance on Pre-operative Assessment for Inpatient Surgery,Modernisation Agency. 2003
Pre-admission assessment
Delays to first cardiothoracic review
• 41/910 (5%) delay in opinion of surgeons
• 57/822 (7%) delay in opinion of advisors – unable to assess in 102 (12.5%)
• A delay to the first review affected the diagnosis of the
patient in 8/57 cases (advisor opinion)
• A delay to first review affected the outcome of the patient
in 33/57 cases (advisor opinion)
Case study 2
A written referral, regarding an elderly patient, was made by a consultant cardiologist from a DGH, to a consultant CT surgeon. The surgeon was on annual leave.
Four weeks after referral the cardiologist contacted the surgeon again by letter. The surgeon denied knowledge of the patient. It subsequently transpired that the referral letter had been misfiled awaiting the return of the surgeon. In the mean time the patient’s condition had deteriorated and they were referred to another CT centre.
The patient underwent CABG which was complicated by postoperative cardiac failure and they died after a protracted period on the ICU.
Case study 2
Advisors comments:
• Not possible to determine whether the delay in the referral of this patient affected the clinical course
• However concerned that no formal cross cover arrangements had been arranged for new referrals during the surgeon’s absence
• Wondered if a generic team system for referrals to cardiothoracic units should be considered using a cardiac network approach
Deterioration during transfer
• It was difficult to assess whether patients deteriorated during
transfer
• However of the 405 patients that were inter or intra- hospital
transfers to the CTU 27 (6.7%) were judged by the advisors to
have deteriorated during the transfer
• Majority of these patients had evolving infarcts with recurrent
chest pain
Case study 4
A middle aged patient was transferred as an emergency from another hospital to the cardiology service with ACS. The patient underwent urgent angiography which showed severe stenosis of the right coronary artery.
A verbal message was left by the cardiologists for the cardiothoracic team that this patient required urgent CABG. Although blood for a troponin level had been taken the result was not available prior to surgery.
It transpired postoperatively that the patient had an evolving myocardial infarct. They developed a low cardiac output state following surgery and died.
Case study 4
Advisors comments:
• Communication between the cardiologists and surgeons was very poor
• No formal referral process occurred with insufficient information given regarding the potential evolving infarct
• If the troponin level had been available would this have changed the decision for surgery?
• Should the surgeon have made more effort to check information on the patient’s condition was correct before operating?
Delays to time to operate
• Missing data for 20% of cases for each year of the study
7 patients had intervals ≥ 250 days
Key findings
• Written protocols for referral of patients were available in 28/58 of CTUs. Discrepancies occurred in the use and presence of protocols.
• In 272/821 (33%) ICPs were used; variation in the quality was noted.
• 99% of patients were referred by a cardiologist.
• In 57/822 (7%) of cases there was a delay from referral to the first CT review and in 33 of these patients outcome was adversely affected.
• 27/405 (7%) patients deteriorated during the transfer.
Recommendations
• Cardiothoracic units need to adhere to the requirement of the National Service Framework for Coronary Artery Disease and use protocols for referrals to their unit. These protocols should be standardised nationally for patients who require coronary artery bypass graft surgery. The degree of urgency of referral should be emphasised within these protocols. (Clinical Directors)
Recommendations
• Cardiothoracic units need to ensure that monitoring systems are in place to record nationally agreed audit data on referrals and the decision to operate. These systems need to identify patients who are in danger of breaching national agreed waiting times so that surgery can be expedited. (Clinical Directors)
Recommendations
• Patients who have acute myocardial ischaemia and require CABG need special attention. Many of these patients are intra or inter-hospital transfers. This group of patients should have surgery performed as soon as their clinical condition permits based on appropriate investigation and pre-operative therapeutic optimisation. (Clinical Directors)
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Scheduling of operations
Out of hours
• Out of hours defined as:
between 17:59 and 08:00 and any time at weekends
• 68/760 (9%) operated upon out of hours
• Approximately equal numbers of elective and non-
elective cases out of hours
Out of hours
• Only one out of hours case not operated upon by a
consultant surgeon
• Only one out of hours case not induced by a consultant
anaesthetist
In hours
• Normal hours 112/642 (17%) operated upon by SpR or SAS
• Normal hours 19/604 (3%) anaesthetised by SpR or SAS
Key findings
• Less than 10% out of hours.
• High level of consultant involvement.
• Better care overall out of hours.
• Scheduling did not appear to have a detrimental effect
on quality of care.
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Multidisciplinary case planning
Multidisciplinary case planning
• 4 out of 58 units had protocol for multidisciplinary case planning
• 21 out of 58 units held pre-operative multidisciplinary meetings– 17 of these held weekly meetings
Individual case planning
• Only 225/905 (25%) discussed
• Age <55 – 36%
• No influence by gender
• EuroSCORE
Multidisciplinary case planning
• Effect on treatment?– No difference in global assessment– More likely to have a written pre-operative plan
Key findings
• Only 21 of the 58 units held pre-operative MDT meetings.
• Only one in four patients in this study were discussed at a
pre-operative MDT meeting.
• Patients who were discussed at a pre-operative MDT
meeting were more likely to have a clear written operative
plan.
Recommendations
• Each unit undertaking coronary artery bypass grafting should hold regular preoperative MDT meetings to discuss appropriate cases. Core membership should be agreed and a regular audit of attendance should be performed. (Clinical Directors)
Recommendations
• Each unit should have a clear policy for which cases should be discussed at preoperative MDT meetings. (Clinical Directors)
• Trusts and consultants should identify time within the agreed job plan to allow participation in MDT meetings. (Clinical Directors)
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Patient investigations
Patient investigations
• 72/821 (9%) did not have appropriate pre-operative investigations
• Over half of those judged to have affected outcome
Recurring themes
• Missed abnormal blood films
• Lack of TOE
• Delays between investigations and surgery
• Poor assessment of renal function
Case study 6
An elderly patient was noted to have a raised pre-
operative WCC. Post-op developed haematuria and
further rise in WCC to 26 Antibiotics started, but oliguria
developed. WCC 96.
Film then reviewed and leukaemia diagnosed.
Case study 7
A middle aged diabetic patient had a delay of six
months from stress test to angiography and a further
delay of six months to surgery.
At operation the severity of disease was far greater
than anticipated.
The patient died of a postoperative MI.
Case study 8
An elderly diabetic patient had a raised pre-operative
creatinine. No further investigations of renal function
were undertaken. The patient was returned from ICU to
ward without a catheter. Creatinine was >400 and K+
5.4. The SpR prescribed frusemide which was followed
by abdominal distension and a cardiac arrest.
Key findings
• Almost 1:10 patients did not receive appropriate pre-
operative investigations.
• In half of these patients the outcome was judged to have
been adversely affected.
• Use of a written protocol for investigations was
associated with a better overall standard of care.
Recommendations
• There should be a written protocol available for pre-operative investigations. (Clinical Directors)
• Pre-operative investigations should be contemporaneous; where delay has occurred between assessment and surgery consideration should be given to repeating investigations. (Clinical Directors)
• There must be a system in place to ensure that pre-operative investigations are reviewed by a senior clinician and acted upon. (Clinical Directors)
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Medical management
Medical management
• Advisors were asked to make judgements about appropriateness of medical management in different groups of patients defined by whether their treatment was stopped prior to surgery
Medical management
• Overall picture complex
• The majority of patients on pre-operative beta-blockers
or calcium antagonists continued this therapy and were
judged to have received appropriate management
Key findings
• The majority continued on beta-blockers, potassium channel blockers and calcium antagonists, however a substantial minority stopped these drugs prior to surgery.
• The majority of patients stopped anticoagulant therapy with the exception of LMW heparin.
• A substantial minority continued anticoagulant therapy or clopidogrel and these had a higher rate of postoperative bleeding complications including tamponade.
Recommendations
• Further studies should be undertaken to establish the risks and benefits of continuing pre-operative medication. Guidelines should be produced based upon sound evidence. (SCTSGBI/NICE)
• NCEPOD supports the ACC and AHA guidance that clopidogrel should be stopped prior to surgery wherever practicable.
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Non-elective, urgent, in-hospital cases
Identification and management
• Only 14/57 (25%) of units had a policy for the management of these patients
• Shared care appeared not to be the norm
Review frequency
• 22 patients not reviewed properly– Lack of timeliness– Lack of senior involvement
Medical management
• Almost one in eight patients not managed appropriately– Lack of response to myocardial ischemia
Investigations
• 26 patients not investigated properly– In 15 of these patients outcome was thought to be adversely
affected
Timing of surgery
• 31 patients operated on out of hours
– 256 operated on in hours
– Advisors felt delay contributed to poor outcome in 21/256
cases
– 208 patients waited 3 or more days for surgery
Key findings
• 3/4 hospitals did not have a policy to ensure timely and appropriate review of these patients.
• Medical management of these patients was inappropriate in 36 cases (12%).
• Investigations were not appropriate in 26 cases (9%) and it was felt that outcome was affected by this deficit in 15 cases.
• Peer review identified cases where surgery was inappropriately performed in the presence of an acute MI and also inappropriately not performed when patients were clearly unstable despite medical therapy.
Recommendations
• There should be a protocol to ensure timely and appropriate review of unstable cases that involves both cardiologists and cardiac surgeons. (Clinical Directors)
• The senior surgeon needs to be aware of any change in clinical status in the preoperative period to ensure that surgery is still appropriate. (Consultant Cardiothoracic Surgeons)
Recommendations
• Given the high mortality when operating soon after an acute infarct more use should be made of strategies to optimise clinical condition, provide symptom relief and allow surgery to be performed at a later date (IABP and PCI). (Clinical Directors)
• A “track and trigger” system should be used to provide early recognition of clinical deterioration and early involvement of consultant staff. (Clinical Directors)
Case study 14
A middle-aged patient with CRF underwent an uneventful triple vessel CABG. Despite recognising the likelihood that postoperative filtration would be required, the patient was discharged from ICU to an area that could not provide renal support. The patient was readmitted to ICU two days later for ventilation and haemofiltration due to fluid overload but died 24 hours later.
Key findings
• Neither height and/or weight nor BMI were recorded in almost a quarter of patients.
• More than half of the patients were overweight or obese.
• There was a high level of comorbidity in this group of patients. The majority were managed reasonably, but in a number of cases there was room for improvement, particularly in the management of renal disease.
• There were discrepancies between surgeons and anaesthetists in the grading of LV function.
Recommendations
• All patients should have height, weight and a BMI recorded on admission, unless their clinical condition precludes this. (Medical Directors)
• Where comorbidity exists, there should be a clear written management plan which is followed in order to optimise the physical status of the patient prior to surgery, and identify the need for specific postoperative support to be available. (Clinical Directors)
Recommendations
• There should be clear guidance about how to estimate
LV function, and at what point in the patient journey this
should be ascertained and recorded. Units should audit
discrepancies in recorded LV function from surgeons
and anaesthetists and where there are significant
differences ensure that systems are in place to address
this. (Clinical Directors and Audit Leads)
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Anaesthetic process
Pre-operative assessment
• 901/923 (98%) patients were assessed by an
anaesthetist prior to surgery; 706/896 (79%) of the
anaesthetists were consultants
• Of those patients not assessed by a consultant,
regardless of the category of surgery or the patients
EuroSCORE, the proportions of patients were similar to
those assessed by consultants
Case study 15
An elderly patient presented as an emergency with ACS for CABG. The patient was seen by an anaesthetic SpR an hour prior to surgery and clearly documented in the patient’s notes which included a comprehensive history and examination.
The proposed anaesthetic was explained to the patient along with the associated risks. The entry was dated, timed and signed with a designation. In contrast the consent for surgery was taken by a surgical SHO with limited documentation of the risks of surgery.
Intra-operative care
• In 898/922 (97%) a consultant was the most senior
anaesthetist at induction
• Median number of years in post = 10
• Median number of Clinical Programmed Activity in
cardiothoracic anaesthesia = 6
• In 85% of patients the consultant was a member of ACTA
Advisor commentary
• Overwhelming majority of cases anaesthetised by consultants
• Quality of anaesthetic charts generally good
• Occasional tensions between consultant anaesthetist and surgeons– appropriateness of trainee surgeons for complexity of case– disagreement on post bypass stability
Case study 17
An elderly patient underwent an uneventful CABG. A locum consultant anaesthetist was responsible for the anaesthetic care. After coming off bypass the anaesthetist considered that the patient was not sufficiently stable to be transferred to the ICU. However, the consultant CT surgeon was of a contrary view and the patient was transferred to the ICU against the advice of the anaesthetist.
Shortly after arriving on the ICU the patients clinical condition rapidly deteriorated and had re-grafting of the coronaries with bypass on the ICU. The patient died.
The anaesthetist stated that one of the reasons for the surgeon’s decision was pressure to proceed with next case due to time constraints of the operating list.
Case study 17
Advisors comments:
• Poor in-theatre team working and that the views of the anaesthetist should have been considered more carefully
• The pressure to proceed with the next case indicated poor theatre time management
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Peri-operative complications and postoperative care
Level of postoperative care
• L3 – 744• L2 – 66• No issues
• Operations only proceed if bed available• Lack of beds = cancellations, not inappropriate
postoperative care
Step down of care
• Most patients stepped down appropriately
• Only 10/868 patients transferred early
• Result of bed pressures
Complications
• 722 cases considered by advisors
• 127 complications not managed appropriately
• In 95 cases (of 127) this deficit was felt to have affected
outcome (75%)
Complications
• Ischaemic bowel
– 54 cases
– Frequent concerns over management
– Raised at every advisor meeting
– Many examples of poor care in this study
• Case study 20 is illustrative
• Junior staff led, delays, continuing futile care are some of the
problems
Complications
• Tamponade– 53 cases
– Often delay in diagnosis
– Junior staff
– Haemodynamics often ascribed to poor LV function
– Low use of echocardiography in postoperative period
Key findings
• All patients received an appropriate level of care immediately postoperatively.
• Pre- and postoperative complications were felt to be inadequately managed in 127/821 cases (16%).
• Of these 127 cases it was felt that inadequate management of the complications may have led to death in 95 patients.
• Advisors raised concern over the role of cardiac ICU and general ICU in the management of patients with a complicated postoperative course.
Recommendations
• Patients who have a more complicated postoperative period are difficult to manage. Any interaction between different medical specialities about patient management should be at consultant-to-consultant level, in particular for patients with suspected intra abdominal pathology. (Consultants)
Recommendations
• Cardiac recovery areas/critical care units are best suited to managing the majority of patients who recover uneventfully. Patients who are developing critical illness and additional organ failure should be managed in an environment with sufficient throughput of such patients to have the resources and experience to provide optimum outcomes. (General Critical Care Units).
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Appropriateness of surgery
Appropriateness of surgery
• Advisor opinion stated surgery was:
– Appropriate 687/815 (84%)– No appropriate 66/815 (8%)– Could not assess 62/615 (8%)
• Plan available 759/909 (83%)
• Plan followed 696/736 (95%)
• Written or pictorial
record of disease extent 529/636 (83%)
• Critical incidents were observed more frequently when a clear operative plan was not available
Availability of operative treatment plan
Recurring themes
• Failure to adapt technique to pre- or intra- operative findings e.g. off-pump cases not converted to on-pump in the face of deterioration
• Avoidance of high risk procedures because of fears about effect of surgeons position in league tables
• Elderly patients receiving multiple arterial revascularisations rather than venous grafting
• Failure to accept need for palliation rather than “heroic” surgery
Case study 22
An elderly patient with IHD and poor LVF underwent off-pump CABG. All the diseased vessels could not be revascularised. The patient died in the immediate postoperative period of a VF arrest.
Advisor comment:
• Should this patient have been converted to “on-pump” before the coronaries were deemed ungraftable?
Case study 23
A middle aged patient was operated upon by an unsupervised SpR2. Whilst taking down the internal mammary artery the patient arrested. There was delay in getting the patient on bypass because the pericardium had not been opened. The patient developed a persistent vegetative state and ultimately died.
Advisor comment:
• Should an SpR2 be operating without supervision from an immediately available consultant?
Key findings
• 84% of cases received an appropriate operation.
• A clear written operative plan was available in 83% of cases.
• Consultant anaesthetists were involved in most (97%) cases.
Recommendations
• A clear written operative plan should be available. This should include contingency arrangements where the findings at surgery dictate an alternative approach [back planning]. (Clinical Directors and Surgeons)
• Where unexpected events occur during surgery, surgeons should have an adaptable approach, and modify the operation to suit the circumstances of the case. (Surgeons)
Recommendations
• A clear description of the extent of the disease should be recorded. (Surgeons)
• Where an operation performed deviates from the operation planned, the reason for this should be clearly documented. (Surgeons)
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Communication, continuity of care and consent
Handover documents
• Theatre to critical care
– 18/58 units did not have
• Critical care to ward
– 13/58 units did not have
Consent
• Only 643/821 forms returned to NCEPOD
• Complications noted in 588/643 (91%) forms
• Risk of death stated in 298/643 (46%) forms
• Notes also examined for risk of death– Combined notes and consent forms showed risk of
death only in 437/821 cases (53%)
Grade of clinician and risk of death
• Consultants– 120 cases– ROD given in 70%
• SHOs – 201 cases– ROD given in 18%
Key findings
• Use of protocols for handover and structured handover information was poor.
• 7 out of 58 units (12%) did not provide written information sheets about coronary artery bypass grafting to patients.
• The consenting process for patients undergoing coronary artery bypass grafting was poor. Consultant involvement in the consent process was low, almost one third of patients were consented by SHOs and no risk of death could be found in 384 cases (47%).
Recommendations
• Protocols must exist for handover between clinical teams and patient locations to ensure effective communication and continuity of care. (Clinical Directors)
• A consultant should obtain consent for coronary artery bypass grafting. (Consultant Cardiothoracic Surgeons)
Recommendations
• Potential complications must be recorded on the consent form. This should detail the likely complications and the incidence of these complications based on local data. (Clinical Directors and Consultant Cardiothoracic Surgeons)
• An accurate risk of death must be quoted on the consent form. This should take into account the proposed procedure and clinical status of the patient. (Clinical Directors and Consultant Cardiothoracic Surgeons)
A report of the National Confidential Enquiry into Patient Outcome and Death (2008)
Multidisciplinary review and audit
Multidisciplinary review and audit
• Frequency of morbidity and mortality meetings
• Methods to feedback
• Autopsies
Morbidity and mortality audit meetings
• 43/58 (74%) CTUs held regular M&M audit meetings– Reasons for not holding meetings:
• Insufficient time in week• No audit lead • Numbers too small• Hospital in provisional liquidation
• 38/43 (88%) CTUs held monthly or more frequently M&M audit meetings
• Only 7/52 (16.7%) CTUs graded quality of patient care at M&M audit meetings– Each site had different system
• Procedures for providing feedback from morbidity and mortality audit meetings varied between cardiothoracic units often without clear identifiable systems being in place.
• 822/910 (91%) of cases were reviewed at a morbidity and mortality audit meeting.
• An anaesthetist attended a morbidity and mortality audit meeting for 396/923 (44%) of cases.
Morbidity and mortality audit meetings
Case study 25
An elderly patient developed persistent metabolic acidosis and low cardiac output state post CABG. They returned to theatre for chest re-exploration and a laparotomy. No cause was found for the condition and they died from multi-organ failure.
The case was referred to the coroner but an autopsy was not performed. There was no evidence of the case being reviewed by the surgical team in a morbidity and mortality audit meeting.
Case study 25
Advisors comments:
• What was the cause of death in this patient? • A morbidity and mortality meeting should have been
held• If an autopsy had been performed this may have made
a valuable contribution in determining the cause of death
• It is to difficult to see what the surgical team learnt from this case
Key findings
• 43/58 cardiothoracic units held regular morbidity and mortality audit meetings, of which 38/43 of these held meetings monthly or more frequently.
• Procedures for providing feedback from morbidity and mortality audit meetings varied between cardiothoracic units often without clear identifiable systems being in place.
• An anaesthetist attended a morbidity and mortality audit meeting for 396/910 (44%) of cases.
• The total number of autopsies fell from year 1 (172, 46%) to (87, 34%) in year 3 of the study.
Recommendations
• Morbidity and mortality audit meetings should be held in all cardiothoracic units. The majority of units should hold meetings at least monthly. If the numbers of cases performed in a unit are small, alternative arrangements should be made to incorporate these cases in other surgical audit meetings. (Clinical Directors and Audit Leads)
• A common system for grading of quality of care of patients should be employed for all patients discussed in morbidity and mortality audit meetings. The peer review scale used by NCEPOD provides such a system. (Clinical Directors)