CENTRAL NEURAXIAL BLOCKADE: MAJOR COMPLICATIONS Third National Audit project of the Royal College of Anaesthetists
2009 Dr Gareth Symons
Major Complications of central neuraxial block: report on the Third National Audit Project of the Royal College of Anaesthetists. Cook, T.M. Counsell ,D. British Journal of Anaesthesia 102 (2): 179-90 (January 2009)
Context - ?Benefits
Masters (2002) Vets Affairs Epidural RCT (2001) Epidural & Survival (wijeysundera 2009)
Aims
To prospectively evaluate the number of CNB
procedures performed each year within the NHS system in the UK To assess the number of serious complications resulting from those procedures
Study Design
Prospective Multi-Centre National Audit within NHS Multi-disciplinary involvement Data collection over 12 month period
Methods - Denominator
Every NHS Hospital performing surgery was
enrolled in the study A local reporter (LR) was appointed at each hospital 2 weeks of data collection Blocks included were epidurals, spinals, Combined spinal-epidurals (CSEs) and caudals
Methods - Denominator
LR classified their datas accuracy Data regarding attempted then abandoned
CNB procedures was not sought 2 weeks of CNB procedures multiplied by 25
Methods Numerator Events
Spinal Infections Spinal bleeding Major nerve damage Wrong route injection errors Death where the anaesthetic/analgesic
procedure is implicated as causal
Methods - Numerator
The LR was responsible for ensuring
reporting of events but reports were accepted from all hospital staff and neurologists, neurosurgeons, radiologists and neuroradiologists were deliberately involved Cases were reviewed by a panel including representatives from the specialties involved
Methods - Interpretation
Diagnosis if this was not clear cases were
included Causation Results were analysed under two sets of assumptions, one optimistic and the other pessimistic. If causation was considered certain, likely, possible or unlikely cases were included in the pessimistic analysis. Cases considered unlikely were excluded from the optimistic analysis
Methods - Interpretation
Follow up was arranged for 6 months post
any event but if this data was not available the worst result was assumed (ie the patients symptoms did not improve)
Methods - validation
Relevant bodies were consulted to ensure
that major complications of CNB were not missed. NHS Litigation Authority National Reporting & Learning Service The Medical Protection Society The Medical Defence Union Medical Journals Google was consulted
Methods-Endpoints
NPSA severity of outcome scale None - no harm Low - Minimal harm Moderate* - Significant but not permanent
harm, or moderate increase in treatment Severe - Permanent harm due to the incident Death - Death due to the incident
* Return to surgery, unplanned re-admission, prolonged episode of care or transfer to ICU
Methods Endpoints
The primary end points of the study were the
incidences of permanent harm due to complications of the various types of CNB performed within the 1 yr audit period in an NHS hospital. The incidence of decompressive laminectomy in adult patients undergoing a perioperative epidural block was also calculated.
Results
All hospitals reported their denominator data 92% of LRs graded their data accurate 707,425 CNB procedures were performed in
the study year (extrapolated)
Results
Breakdown of CNB procedures: 46% Spinals 41% Epidurals 6% CSEs 7% Caudals The majority of CNBs were for obstetric
(45%) or perioperative care (44%)
Results - Events
108 cases reported Review of other databases revealed 1 case
which met inclusion criteria and this was subsequently reported. 56 cases were either minor and irrelevant complications of CNB or did not meet the inclusion criteria Of the 52 eligible event patients, 22 made a complete recovery from their serious complication
Results Recovered patients
7 epidural abscesses 3 meningitis 1 Vertebral Canal Haematoma 7 Nerve injury 3 Cardiovascular collapse* 1 respiratory arrest a/w intrathecal opioid
*Requiring either CPR or ICU Admission
Results - Analysis
The remaining 30 patients were included in
the pessimistic analysis Under further review, 16 patients in this group were thought likely to make a good recovery or the complication was thought to be tenuously linked to the CNB, this left 14 events to be considered in the optimistic analysis
Results Key Endpoints
Death or serious injury :
4.2 per 100,000 CNBs (CI 2.9 6.1) Deaths: 0.8 per 100,000 CNBs (CI 0 1.8)
Results
Decompressive laminectomy: 12.3/100,000 patients (perioperative epidural)
Results Permanent Harm
Perioperative 8 per 100,000 Chronic Pain 2.5 per 100,000 Obstetric 1.2 per 100,000 Paediatric 0 per 100,000 Non-anaesthetists 0 per 100,000
Results Permanent Harm
CSE 9.6 per 100,000 (CI 2.6 24.5) Epidural 6.1 per 100,000 (CI 3.6 9.7) Spinal 2.2 per 100,000 (CI 1.0 2.4) Caudal 2.1 per 100,000 (CI 1.0 11.7)
Results Severe Injury or Death
Epidural Abscess 8 patients Vertebral Canal Haematoma 5 patients Nerve injury 7 patients Spinal Cord Ischaemia 4 patients Wrong route error* 1 patient Cardiovascular collapse 3 patients Miscellaneous 2 patients
Results Deaths
6 deaths total resulting from CNB* 2 abscesses 3 cardiovascular collapses 1 IV bupivicaine injection
3 deaths recorded under optimistic interpretation
Prognosis after Injury
Ischaemia - 5/5 no or minimal improvement Abscess - 7/12 major improvement Nerve injury - 9/13 major improvement Meningitis - 3/3 major improvement Vertebral canal haematoma 6/8 major
improvment Total 25/41 major improvement
Implications
Informed consent ?Incidence of less serious complications Risk/benefit of CNB
