Prevention and Pre-Hospital Care
Most brain injuries are preventable. In the paediatric age group, falls and road accidents account for most head injuries. Falls in the elderly, particularly when accompanied by anticoagulant medication, are a major cause of head injuries. Such injuries cause considerable morbidity and have huge costs to society through failure to return to work, fractured family relationships, homelessness, imprisonment and requirement for care & support. The need for guidelines for the prevention, diagnosis and treatment of sports concussion has led to a surge in research activity worldwide that will be helpful to patients with other forms of brain injury.
Major trauma networks are ‘getting the right patient to the right place’. Progress has been made in identifying and treating, in timely fashion, secondary insults to the brain including hypoxia, hypotension, fits and intracranial haematomas. We have developed tools (e.g. www.goodSAMapp.org) that enable people trained in basic airway management, who happen to be within a few hundred metres of such incidents, to be alerted. The early detection of an intracranial haematoma would allow for focussing of the trauma pathway to the needs of the individual patient. For all these reasons, access to better monitoring at the roadside is required.
If patients survive the first few minutes, there is a therapeutic window of opportunity to intervene and prevent further death of brain cells. However, clinical trials of promising neuroprotective agents have been universally unrewarding apart from Nimodipine for aneurysmal subarachnoid haemorrhage. The therapeutic window of opportunity after subarachnoid haemorrhage is of the order of days whereas it may only be minutes or hours after a head injury. Future drug trials for head injury will require a clinical platform capable of the timely stratification of patients in to homogeneous subgroups at the roadside and the ultra-early administration of safe neuroprotective agents capable of rapidly entering the brain in effective concentrations.
Aims and specific short (1 year), medium (2-3 years) and long (4-5 years) objectives
Short term – 1 year:
- The new SBNS Brain Injury Classification System compiled and published
- Publication and dissemination of the Bristol Paediatric and London TBI epidemiology studies (TriBAL) and of the Pre-Hospital NIRS Infrascanner study
- Network meetings of pre-hospital care organisations
- Ethics and grant applications for prospective collaborative data collection relating to head injury management submitted
- Biomarker study grants are already submitted. Community response systems are already in place and being evaluated.
- Grant application for development of pre-hospital multichannel NIRS device for imaging
Medium term – 2-3 years:
- Over this time frame we expect to have good pre-hospital data collaboration to enable the subsequent assessment of interventions
- Multichannel NIRS development and testing of other pre-hospital diagnostic tools underway
- Biomarker studies nearing completion
- Lower body negative pressure for ICP pilot complete and if successful developed for further possibly pre-hospital study
- Completed evaluation of pre-hospital community alerting system and video triage
Long term – 4-5 years:
- The trial of novel neuroprotectants within 30 minutes of injury
- The implementation of on-scene diagnosis of some forms of brain injury (e.g.extra-axial haematomas) and their targeted management
- Manipulation of blood pressures for specific injury types
- A trial of a specific biomarker as a rule out tool conducted in remote areas
All the above can be evaluated more accurately (on an individual patient level) with a brain injury classification system that truly reflects the type of brain injury sustained.
SUBTHEME 1: Epidemiology including mechanism of injury: why has prevention ‘failed?
Epidemiological surveys of TBI In London (Traumatic Injury to Brain Across London, TriBAL) and Bristol are nearing completion. TriBAL is commissioned by NHS England to undertake a 4-month audit of TBI across the four major trauma centres and their referring trauma units. All patients with a positive CT scan are included with data collected on mechanism of injury, pattern of injury, co-morbidities (e.g. on anticoagulation) as well as management (in TUs and at the MTCs). The analysis phase has started. The Bristol survey of all ED attendances with head injuries in the paediatric age group is now complete [Dr Julie Mytton].
The next step is to determine why prevention has ‘failed’ in individual patients and what the scope is for designing new policies around prevention using the principles of the Child Death Overview Panels in combination with the Roadmapping expertise of the MIC.
The MIC is working with professional sports organisations to develop technologies to monitor and assess concussion including sensors, physiology and cognitive measures.
SUBTHEME 2: Emergency Service Assessment and Management
Currently, traumatic brain injury is managed uniformly. The pre-hospital data field does not specify the type of brain injury, if any. Hence all such patients receive the same treatment (usually intubation, and transfer to a neurosurgical unit). On scene diagnosis of injury would permit more directed therapies to be initiated. It is vital to dispatch resources appropriate for the presumed severity of injury. Currently dispatch criteria are often based on “Mechanism of Injury” (e.g. fall > 2 floors, ejected from car) and interrogation of callers.
Here below there are a number of studies looking at improving this aspect:
2.1 Automated prediction of injury in Road Traffic Collisions
(RTCs)TRL (https://trl.co.uk) accumulate comprehensive data from RTCs. This includes on board gyroscope sensors and cameras which many cars now have. A combined Imperial and TRL PhD student is currently modelling the patterns of brain injury from such data. This may, in the future, enable prediction of severity of injuries from such data at the time of the incident and activate the appropriate emergency response automatically.
2.2 Instant-on-Scene Video Triage (London Ambulance Service and GoodSAM)
With mobile phone technology, visualising the scene and patient can help dispatch decisions. We are currently undertaking a pilot study with London Ambulance Service as to the benefit of on scene video from Incident Response Officers (IROs) and the public.
2.3 On Scene Imaging (Imperial, London’s Air Ambulance, Infrascanner)
Knowing the nature of an injury could help decision making – do you bypass the nearest hospital to go an additional hour to the neurosurgical centre? This is especially important in remote regions such as East Anglia where travel times can be long. More importantly, earlier diagnosis would enable condition-targeted treatments. Diagnosis can either be anatomically based (with imaging) or “likelihood” based with biomarkers.
With London’s Air Ambulance and Imperial, we have undertaken a HTC funded pilot study into the use of a single channel Near Infra-Red Spectroscopy (NIRS) machine (Infrascanner) and its ability to reliably detect extra-axial haematoma in the hyperacute setting. Our results demonstrate reasonable sensitivity and specificity in hospital use, but moderate sensitivity and poor specificity in pre-hospital application.
At Imperial, we currently use a 48 channel NIRS machine to study spreading cortical depolarisations. Such a device overcomes many of the limitations of a single channel NIRS device. We are preparing a grant application with Clare Elwell’s group at UCL to miniaturise such a device for pre-hospital use.
SUBTHEME 3: Early stratification of brain injury, outcome prediction and RCT design
3.1 SBNS Brain Injury Classification System
Brain injury is a very heterogeneous condition. Different types of closed injury (e.g. extradural, subdural, diffuse axonal injury) can affect different parts of the brain (e.g. frontal, temporal…) to differing degrees (force / mechanism of injury). Additionally, there are constitutional patient differences (e.g. age, the use of anticoagulant). This makes studying brain injury difficult. The current radiological classification systems (e.g. Marshall and Rotterdam) do not have the resolution to distinguish different types of injury. Tools that rely on them to predict outcome while useful on a cohort level, are unreliable on an individual patient level. The Society of British Neurosurgeons (SBNS) is developing a new system to make classification more robust. This will enable more useful comparative analysis of injuries and their management / outcome.
Neuroprotective agents: Multiple studies have failed to demonstrate any benefit from steroids, progesterone, other pharmacological agents or therapies such as cooling. This is likely to be in part due to the delay in their administration. Studies need to re-evaluate such therapies administered within 30 minutes of injury and analysed specifically for the type of brain injury sustained. The feasibility of such studies is being assessed (Adina Michael-Titus)