Author: Mark Winstanley / Codes: C3AP4, CAP1, CAP12, CAP18, CAP28, CAP3, CAP6, CAP7, HAP18, HAP35, HAP6, HAP7, PAP18 / Published: 01/11/2018
Andy Neill, Dave McCreary
Are fluid boluses in children with DKA dangerous?
Kuppermann, NEJM, 2018
There has been a lot of prior observational data associating fluid boluses in children with DKA with cerebral oedema and bad outcomes. It is unclear whether or not this is correlation or causation. The uncertainty has led to widely differing recommendations on how to resuscitate kids with DKA. Depending on where you work may have been castigated for giving too much or too little fluid.
at baseline cerebral oedema is uncommon in DKA with rates of <1%
4 arm RCT with different rates of resuscitation
1) fast with 0.9% saline
2) slow with 0.9% saline
3) fast with 0.45% saline
4) slow with 0.45% saline
this works out to 10ml/kg bolus in slow group with 5% dehydration over 48 hrs compared with 20ml/kg bolus with 10% rehydration over 24 hrs
primary outcome was 2 consecutive reductions in GCS within 24 hrs
– 1300 across the groups
– no difference in primary outcome
– <3% had the drop in GCS (the primary outcome)
– only 12 episodes (0.9%) of brain injury following adjudication.
– this is a really important trial and suggests that the prior observational association of cerebral oedema and fluid boluses was correlation rather than causation
– this paper does not answer the question as to whether the fluid boluses are helping or not but they certainly don’t seem to be causing harm.
– for now if you think the child needs a bolus then i think you’re OK to give one but I will probably stick to 10ml/kg and reassess and will feel free to give more if I think they need it.
Andy Neill, Dan Horner @RCEMprof
More info on TERN can be found on the RCEM Learning website
“A qualitative study of practitioner perspectives on medical record keeping in sudden onset disasters” by Anisa Jafar (ST4 OOPR, PhD Candidate and Research Associate at HCRI, Manchester)
Global health is an increasing concern to many EM clinicians. Indeed, some will go a step further by volunteering to work for humanitarian non-governmental organisations such as médecins sans frontières. Yet learning lessons from disaster relief missions is difficult as in most cases medical record-keeping is notoriously poor. In order to change this, we need to understand what unique challenges these situations present to record-keeping and what the practical and cultural barriers to improving it are. This study aimed to tackle this through qualitative work with healthcare professionals who have worked in this area.
The author used an inductive approach taken from a base-line of grounded theory. Over a 2-month period they conducted 13 semi-structured interviews with healthcare workers in the areas of general surgery, anaesthesia, emergency medicine, paediatrics, physiotherapy and orthopaedic surgery. The participants were all experienced in working in disaster settings, had between them worked for over 15 different organisations and represented 9 different nationalities. Thematic analysis of the interviews found several key themes emerging, to include: the need to incentivise medical documentation; the fact that both paper and electronic record keeping have limitations and an ideal approach would encompass both; and the acceptance that creating systems directly tailored to the disaster environment would be more likely to succeed than trying to adapt civilian solutions to the problem.
The results of this work will be fed back to the WHO emergency medical team secretariat and other key global organisations delivering emergency response in order that when documentation is redeveloped to meet new WHO minimum reporting standards, repeating past mistakes can be avoided and the standard of documentation and data capture can move forward.
Anisa’s take home message: “To use a qualitative lens on this perennial problem has captured and condensed essential ideas into digestible, practical information for practitioners and policy-makers to use when planning how to document the care provided in some of the most challenging of circumstances.”
“What are the CT scan findings and outcomes for patients taking warfarin with mild head injury? A quantitative analysis of AHEAD data” by Rachel Evans et al. (CT2 doctor, Sheffield Teaching Hospitals)
Head injury is a common presentation to the ED and a frequent cause of CT scan requests. In the UK, NICE recommends that all patients with a head injury who are on warfarin have a CT scan to rule out intracranial injury. However, people have started to question whether this is always necessary. Is that 82 year old man with the innocuous injury 22 hours ago likely to have an intracranial injury and if so, are we likely to intervene? The AHEAD study was a large, multicentre observational study that examined the rate of adverse outcomes in head injured patients taking warfarin. This study looked at the imaging findings for these patients and the outcomes of those who had abnormal neuroimaging.
The AHEAD study recruited 3,534 patients from across 33 different Emergency Departments in England and Scotland. Despite national guidance that all head injured patients on warfarin should receive a CT scan, only 1,897 (53.7%) of those recruited underwent neuroimaging. Of the study patients with a GCS of 14 or 15, 153 patients had an intracranial abnormality likely to be attributable to their injury, of which the most common abnormality was subdural haematoma (37%), followed by mixed types of haemorrhage (21%), subarachnoid haemorrhage (16%) and intracerebral haemorrhage (14%). Of the patients with a demonstrated intracranial injury, 9 went on to have neurosurgery and only 1 patient died. The bottom line is that the risk of intracranial injury in anticoagulated patients with a Glasgow Coma Scale of 14 or above after blunt head injury is low. More work needs to be done to model the potential costs, benefits and harms associated with CT scanning in these low risk patients and we hope to see a decision model in due course.
Rachel’s take home message: “Patients with head injuries on warfarin who are GCS 15 and do not have other reasons for a CT scan (eg mechanism, loss of consciousness) are highly unlikely to suffer any serious sequalae. They may not need a CT scan. Current guidelines promote a generic “one-size-fits-all” approach to highly individual, varied patients. Talk to your patient and make a sensible decision together.”
Cauda equina is rare (1 in 33,000 to 1 in 100,000) but it’s a huge medico legal consideration. In 5 years, the MPS was notified of 63 likely claims, and 20 cases were conculded. 43 of these cases related to general practice. The NHS Litigation Authority had 78 claims, with 24 concluded cases, and damages paid in 12. 21 of these 78 claims related to ED.
Looking at the case reports
Chronic back pain, attended with difficulty passing urine – given antibiotics for presumed infection.
Perineal numbness, progresing to saddle anaesthesia, progressing to urinary incontinence attributed to weak pelvic floor.
Back pain and urinary symptoms, treated as a UTI.
Anatomy wise, the cauda equina, or horse’s tail, is the very end of the spinal cord. The spinal cord ends between L1 and L2, continuing as the filum terminale, and then as a collection of nerve roots called the cauda equina. These nerve roots provide sensory components to the lower limb (anything that begins with an “S” nerve root) and if they’re compressed, causing a lower motor neuron lesion, it’s a true orthopaedic emergency.
Lumbar disc herniations (but only 1-6%)
Spinal epidural haematoma
The cauda equina nerves contain motor and sensory components. This means motor signs are a flaccid paralysis in both legs and areflexia. Import3antly, areflexia also affects the urinary detrusor and rectal smooth muscle – so the residual tone of the internal urethtal and anal sphincter predominates causing urinary and fecal retention. This This ultimately leads to overflow incontinence. This is an important point to remember – it’s not primary incontince, but overflow incontinence associated with cauda equina. Urinary retention is the most consistent finding – sensitivity of 90%.
Damage to the sensory nerves causes reduced perineal sensation, with saddle anaesthesia – so the patient can’t feel when bladder and bowels are found, compounding the risk of retention. This also means that if you catheterise a patient for retention, then pull gently on the catheter – the true cauda equina patient will be unaware. The paint with retention secondary to pain will feel you. There’s also damage to the nerve supply to both legs, and the nocioceptors – causing sensory disturbance in the legs, and neuropathic pain.
Differential Diagnosis (non-compressive causes of spinal cord dysfunction)
Myelopathies (e.g. HIV related)
Spinal cord infarction
medicalprotection.org failure to act on cauda equina
rebelem.com cauda equina syndrome
Dave McCreary, Andy Neill
Does CT head and C-Spine always have to go hand-in-hand?
Title of Paper:
Low Yield of Paired Head and Cervical Spine Computed Tomography in Blunt Trauma Evaluation [link]
Journal and Year:
The Journal of Emergency Medicine. 2018.
* We love a good CT in ED, especially on people who bump their heads
* In the strive for a near-zero miss rate on injury in trauma, when we’re scanning heads we’ve been moving towards more frequently scanning the c-spine as a matter of routine
* Previous studies have shown high rates of concomitant injury in the head and neck
* Its easier for patient flow to just get both scans done in a single trip rather than risk multiple trips to the scanner (for example after you find the subdural)
* What’s the harm in the extra scan?
- Ionising radiation & the thyroid
- Addition time and resources to perform and report unnecessary scans
* Retrospective chart review of trauma imaging in an urban level 1 trauma centre in the USA
* Adult blunt trauma patients who received a CT of head and c-spine, ordered at the same time
* Non-trauma CTs
* Penetrating trauma
* Delayed presentations (>24h post injury)
* Trauma secondary to likely medical cause (CVAs etc)
* Solo CT neck
* CT head and neck performed at different times
* System to determine clinically significant injuries defined a priori by expert panel (emergency physician, neurosurgeon, and neuroradiologist)
* Any requiring operative intervention
* Rotterdam score >2 for nonoperative injuries – validated score, categorises severity of adult traumatic brain injury.
* 6 month mortality increases with score – 1 = 0%, 2 = 7%
* Injury associated with C-Spine instability or vascular injury
* Primary: yield of CT for clinically significant injury (CSI) in both head and neck
* Secondary: Yield of CT for any injury in both the head and neck
Summary of Results:
* 3223 CT neck scans for blunt trauma
* 2888 (89.6%) had head and neck simultaneously
* Falls 38%
* Assault 17%
* MVC 15%
* Head 15% (CSI 9.2%)
* SAH most common CSI
* Neck 4.3% (CSI 2.9%)
* Fracture >1 vertebra most common CSI
* Primary outcome:
* CSI in both head & neck = 0.5% [95% CI 0.3 – 0.8%]
* Yield for CSI head when CSI neck vs no CSI neck = 14 vs 9% (difference 5% [95%CI 0.4 – 13.7%])
* Yield for CSI neck when CSI head vs no CSI head = 5 vs 3% (difference 2% [95%CI 0.3 – 5.8%])
* Secondary outcome:
* Any injury head & neck = 1.4% [95% CI 1.0 – 1.8%]
* Yield for any head when neck injury vs no neck injury = 32 vs 14% (difference 17% [95%CI 9.6 – 26%])
* Yield for any neck when head injury vs no head injury = 9 vs 3% (difference 6% [95%CI 3.0 – 8.6%])
These findings argue against automatically paired ordering of CT head and neck scans and suggest CT should be ordered individually. When injury is found in one region, imaging of the other region should be considered.
Clinical Bottom Line:
Would be helpful if they’d looked at notes / requests to see if there was clinical suspicion of injury / where the clinical suspicion for injury was
Depending on your institutional policies, this could be useful. For patients in where you have a pretty low suspicion for injury (though obviously not low enough to not scan in the first place!) then you could scan the area of concern first and go back to complete the neighbouring scan if you find an injury. It’s probably more labour intensive and might involve a little faffing around, but for some departments and for specific patients it might be a useful approach.
Other #FOAMed Resources / References:
This was also reviewed on Journalfeed.org – Worth signing up to their mailing list for regular spoon feed journal updates
The Rotterdam score on Radiopaedia
Ellen Weber, Mark Winstanley
This interview was recorded at EuSEM18 in Glasgow.
Charlotte Davies, Bethan Nichols
This was recorded at RCEM Spring CPD conference 2018
Rick Body, Mark Winstanley
This interview was recorded at EuSEM18 in Glasgow.